tag' [this should be unbuffered.]

campbeld2 at my-deja.com writes:
> I'm resubmitting this question on the off chance someone can help me,
> please. I suspect it is to do with nested stdout results, but I'm at a
> loss here.
> Thanks
>   Dyan
> 
> 
> In article <7rtin6$9oa$1 at nnrp1.deja.com>,
>   campbeld2 at my-deja.com wrote:
> > This sample program illustrates my dilema.
> >
> > I am trying to execute a command ( could be
> > any command ) within Python and capture the
> > output.  It works fine when I run it at command
> > level, however when I invoke the script in the
> > browser the code does not return any values from
> > the os.popen(...)readlines.  I tried try: and
> > except: but nothing is trapped.
> >
> > Any ideas?
> >
> > ------
> >
> > """
> >  test.py :
> >  returns correct results at command level (great)
> >  returns nothing through Browser !!!
> > """
> > import os
> >
> > print 'Content-Type: text/html\n\n',
> > for line in os.popen( 'cd','r' ).readlines():
> > 	print '...within for loop \n\n'
> > 	cd = line
> >
> > print '
cd = ', cd, '\n\n'
> >
> > #end of test.py
> >
> > Sent via Deja.com http://www.deja.com/
> > Share what you know. Learn what you don't.
> >
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

-- 
Cliff Daniel
Level 3 Communications




From cgw at fnal.gov  Wed Sep  1 00:10:02 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Wed, 1 Sep 1999 04:10:02 GMT
Subject: running python on a floppy?
In-Reply-To: <37cc7e3b.1941695@nntp.best.ix.netcom.com>
References: 
	<37CA50EC.4FC03560@home.com>
	
	<37cc7e3b.1941695@nntp.best.ix.netcom.com>
Message-ID: <14284.42778.543818.343109@buffalo.fnal.gov>

Dennis Lee Bieber writes:
 > 
 > 	I have two SCSI externals, and my DELL came with an IDE
 > internal. The closest I have come to the "CoD" is sometimes having the
 > SCSI drive on my Amiga thunk when it first tries to register a disk,
 > with the disk being "unreadable". In all instances, ejecting the disk,
 > waiting a moment for everything to reset, and reinserting the disk has
 > worked.

It's fortunate for you that you've never experienced the "Click Of
Death".  But of course this kind of N=1 evidence doesn't prove much.
It's like saying "My grandfather smoked 10 cigars a day and he lived
to be 99".  That's great, but it doesn't disprove that smoking causes
cancer.  The Click Of Death is very real,  I've witnessed it on
several different machines, as have several other people of my
acquantance.

Isn't it funny that Iomega's new drive is called the "Clik"?  Couldn't 
have picked a worse name, if you ask me.

In response to the original poster - trying to squeeze software onto
floppies is still very much worth doing.  ZIP drives have nowhere near 
the ubiquity of floppies.  It's a rare machine that doesn't have a
floppy drive.  ZIPs seemed, a few years back, to be growing in
popularity, but that trend seems to me to have slowed substantially -
partly, in my opinion, because CD writers have become so cheap, and
the cost per megabyte of ZIP media is too high, compared with CD
media.  And CD drives are much more prevalent than ZIPs.  Most
computers made in the last 5 years or so have a CD-Rom drive as part
of the "baseline" configuration, with ZIP as a extra-cost option.
Just my opinion, but I think ZIP disks are on the way out.  Iomega
stock certainly isn't worth what it used to be (last time I checked,
anyhow).




From doughellmann at home.com  Thu Sep  9 20:18:24 1999
From: doughellmann at home.com (Doug Hellmann)
Date: Fri, 10 Sep 1999 00:18:24 GMT
Subject: equivalent of enum?
References:  <14296.14104.456361.61180@dolphin.mojam.com>
Message-ID: <37D84F85.A37FAA9C@home.com>


Skip Montanaro wrote:
> 
> If you expect to have properties associated with particular pieces
> (position, strength, etc), you might want to consider representing them by
> instances of a specific ChessPiece class instead:

This is probably the best advice for this situation.

> In more direct answer to your question, there is no enum type in Python.
> It's simple enough to declare objects that you treat as constants, however:

There are a lot of times when having an enumeration is nice, and a class
would be overkill.  I forget where I've seen the idiom, and I certainly
didn't make it up, but a really nice way to make an enum is to do
something like:

	one, two, three = range(3)

	print one, two, three

Works every time.  If you forget to increase the range size when you add
a new value, you will get a runtime error because the tuple unpack won't work.

Doug



From mal at lemburg.com  Wed Sep 15 10:41:51 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 15 Sep 1999 16:41:51 +0200
Subject: Python speed and `pcre'
References:    <00af01bef543$0affe8f0$f29b12c2@secret.pythonware.com>  <37CFB6C4.35578D33@lemburg.com> 
Message-ID: <37DFB02F.D097FC4@lemburg.com>

Fran?ois Pinard wrote:
> 
> "M.-A. Lemburg"  writes:
> 
> > [substrings]
> 
> > Note that Python 1.5.2 already support such beasts... albeit rather well
> > hidden (for obvious reasons ;-) and not fully interoperatable yet:
> 
> > >>> s = '12345678901234567890'
> > >>> snippet1 = buffer(s,8,3)
> [...]
> > You can always turn a buffer back into a string using the str()
> > function.
> 
> > Perhaps the string module should be extended to use "s#"
> > more often...
> 
> Hello.  Thanks for the pointer, yet I'm just starting to understand your
> reply, and it may take some while, still.  The manual is very terse about
> the `buffer' function, relating it to `buffer' methods, without explaining
> any further what these are...  Also, I do not understand the "s#" quote.
> Sorry for being a bit opaque, here.  Would someone suggest some magnetic
> reference to me about these, which a beginner could afford to read and
> understand?  I'm not ready yet to dive into the interpreter sources :-).

Sorry, buffer objects are still somewhat fuzzy and undocumented.
All I can refer you to is the source code in Include/bufferobject.h
and Objects/bufferobject.c.

The "s#" quote was meant for experts: it refers to a format
string passed to PyArg_ParserTuple() for converting Python
objects to C data. "s#" tells the function to give back a
char* Pointer and a length integer and to try to use the buffer
interface of the object to fetch these from the object. By
using "s#" you can make your function compatible to all objects
implementing the buffer interface, ie. not only strings but
also memory mapped files, buffers, etc.

Hope that helps at least a bit,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   107 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From jam at quark.emich.edu  Fri Sep 17 14:36:50 1999
From: jam at quark.emich.edu (Jeff)
Date: Fri, 17 Sep 1999 14:36:50 -0400
Subject: Newbie: Where can I find this module?
In-Reply-To: <37E2712E.3EF8A9F5@alliedsignal.com>; from H. P. Friedrichs on Fri, Sep 17, 1999 at 09:49:51AM -0700
References: <37E2712E.3EF8A9F5@alliedsignal.com>
Message-ID: <19990917143650.D8674@quark.emich.edu>

On Fri, Sep 17, 1999 at 09:49:51AM -0700, H. P. Friedrichs wrote:
> I have a redhat 5.0 linux box on which I've written some python scripts.
> I also have mysql up and running on the same box. I'd like to be able to
> access the mysql tables from python.
> 
> I checked the web and found a site with some sample code to let me do
> this. The first line of the sample imports a module called "mysqldb".
> 
> Where can I find that? I presumed it might be on the MySQL website, but
> it's not readily apparent to me if it is. I have also conducted other
> net searches without luck. Can anyone clarify?
> 
> Thanks
> Pete
> 

try 

regards,
J

-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From aahz at netcom.com  Wed Sep  8 17:12:29 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Sep 1999 21:12:29 GMT
Subject: binary mail with python
References: <036101bef9e7$b907a480$b4bb32c1@clara.ircom.unilim.fr> <19990908211233.A3782@optiplex.palga.uucp>
Message-ID: <7r6jft$lq8@dfw-ixnews3.ix.netcom.com>

In article <19990908211233.A3782 at optiplex.palga.uucp>,
Gerrit Holl   wrote:
>On Wed, Sep 08, 1999 at 12:48:44PM +0200, Daniel ARGOLLO wrote:
>> 
>> I am searching for any python package that can pack different binary
>> files into a MIMI file to be sent with the mail suport of python....
>
>Maybe you're looking for mimimy?  I've never used it, but the name
>means a lot (did I say it right?).

First of all, it's "MIME".  The two Python libraries are mimetools and
MimeWriter.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From stephan at pcrm.win.tue.nl  Mon Sep 13 04:39:00 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 13 Sep 1999 08:39:00 GMT
Subject: mmap interface in Python
References: <877llyzvld.fsf@pc-hrvoje.srce.hr>
Message-ID: 

On 10 Sep 1999 16:31:42 +0200, Hrvoje Niksic  wrote:
>When discussing Python speed, many people advised me to "slurp the
>whole file in".  The only reason why I didn't like that is the obvious 
>one -- it burns a lot of memory for large files.
>
>But then I remembered: in C under Unix, I would use mmap() to map the
>file to memory, and operate on that memory.  I don't see why the same
>thing couldn't be available in Python -- under Unix at least.

It sounds cool, but I'm not sure if it buys you all that much
performance-wise. If you're afraid to "burn too much memory", well,
the memory gets swapped out, so who cares? If you want to do a
regular expression match, the match() function will scan the entire
file anyhow and cause it to be loaded into core. So in that case,
you don't win anything either.

Finally, by ussing mmap(), you loose compatibility with lesser OSes.

So I'm not sure if you really want this, although it might be cool
to mmap() some of the special files in Linux... 

Greetings,

Stephan



From dboddin at uumail.de  Mon Sep  6 13:54:17 1999
From: dboddin at uumail.de (E-14)
Date: Mon, 6 Sep 1999 19:54:17 +0200
Subject: accessing parallel interface
Message-ID: <7r0v0k$o98$1@newsread.do.de.uu.net>

How can access the parallel interface, to set and read bits.
I wonna connect an A/D converter, speed isn't important.

Detlef





From mcg at halcyon.com  Sat Sep 11 17:12:38 1999
From: mcg at halcyon.com (Michael C. Gilbert)
Date: Sat, 11 Sep 1999 14:12:38 -0700
Subject: Sparc Linux?
Message-ID: 

Hi --

Thinking about buying an old Sparc Station with SparcLinux. Any trouble
running installing 1.5.2 on that?

-- Michael





From lynn at siprelle.com  Tue Sep 14 19:53:29 1999
From: lynn at siprelle.com (Lynn Siprelle)
Date: Tue, 14 Sep 1999 16:53:29 -0700
Subject: [Newbie] test_signal hangs
Message-ID: <37DEDFF7.1DC7C4B3@siprelle.com>

Hi there--

I'm totally new to Python, haven't even written a Hello World yet,
because I'm still trying to get it up and running on my fershlugginer
system. I'm wanting to experiment with Python and Zope on my iServer
websites. OK, so I figured out how to compile Python (the iServer
environment tricks the configure script into thinking it's on a NeXT
box, so you have to go in and manually change configure.h to comment out
the dyld stuff). And now I'm trying to run the test set, but it hangs on
test_signal--I let it run for more than an hour, and it just sat there.

Should I even care? Or is this a real problem?






From tim_one at email.msn.com  Fri Sep 17 01:42:59 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 17 Sep 1999 01:42:59 -0400
Subject: Marking translatable strings
In-Reply-To: 
Message-ID: <000101bf00cf$801bd080$d12d153f@tim>

[Fran?ois Pinard, on internationalization of strings in program text]

I know less than nothing about this -- always figured that if American
English was good enough for God, it's good enough for everyone else .

So I'll just touch on some Python issues:

> ...
> [Python] has eight type of strings: ', ", ''', """, r', r", r''' and
> r"""; and I thought that maybe we could just discipline ourselves to
> give more meaning to all these differences ...

Indeed, the number of ways to spell strings in Python is quite unPythonic!
The language generally avoids offering gratuitous choices.  Too late now,
though -- all ways of spelling strings are already used in all conceivable
contexts, and no rational discipline can be imposed a posteriori.

For the purpose of finding Python strings, note that under the covers it was
deliberately designed so that there's really just one basic
string-recognizing state machine.  ' vs ", and ''' vs """, make no
difference at all to lexing, and r- vs not-r makes no difference either.
The only difference in lexing is that the 4 single-quoted varieties (', r',
", r") refuse to span lines except via backslash+newline, while the 4
triple-quoted varieties don't treat newline specially.

> ...
> Surely, since doc strings use """ exclusively, ...

Really sorry, but you can't rely on that.  I've seen all 8 flavors of
strings used as docstrings -- what makes a string a docstring is not how
it's spelled, but where it appears (if the first executable stmt in a
module, class or def is a string, that's a docstring; and nothing else is).

> ...
> The most comfortable (the less intrusive) way would be to call:
>
>         _(TEXT)
>
> to get the translation of text.

I agree with everyone else  that the special meaning of "_" in
interactive mode is unlikely to create a problem for you.

> ...
> Experience shows that horizontal space is often a valuable resource while
> writing code, so we should try to be economical there for overall
> legibility, much more than typability.

I like the way you think .  Really, Python takes pride in its
legibility, so worrying about that is very welcome here.  Note one odd
thing:  since Python doesn't support assignment expressions,
horizontally-challenged 8-line "if" stmts simply don't occur.  So there's
less horizontal pressure here than you may be used to.  OTOH, Python doesn't
support augmented assignments (+= & friends) yet either, and that adds
horizontal pressure (but less, I think, than the lack of assigment
expressions relieves).

> ...
> and we could use''"TEXT" or ""'TEXT' for strings needing to be marked
> only, and translated only later.

I don't understand the distinction here, so just noting that there's nothing
in the compiled code that can distinguish an instance of

    ''"TEXT"

in the source from an instance of

    "TEXT"

or even of

    ''   "TEXT"""""""r''''''

Catenation of adjacent string literals is a compile-time "optimization", or
more accuractely slavish aping of C rules -- as if someone thought C were a
pleasant string language .

> ...
> One reason to modify the interpreter would be for speed, maybe,
> by ensuring that a given string is translated only once even in a loop,
> like for what _() currently does in C when compiled with GNU C.  It might
> depend on how exactly strings are intern'ed within Python, which I do not
> know.

You can force interning via calling the intern() builtin.  By default, the
only strings that get interned are those literals that "look like" they
*may* be attribute names.  For example, "Pinard" is interned by magic, but
"F. Pinard" is not.

In any case, automatic interning is purely an internal optimization, and the
rules can-- and do --change from release to release.  So don't even think
about relying on their specific incarnation today.

Within a code block (def, class or module body), but not across code blocks,
all string literals are accumulated and stored uniquely in a vector of
constants attached to the code object.  So, e.g., within a single def, 112
instances of "F. Pinard" will all resolve to a single string object.  If you
use a runtime translation function, it can exploit this via a cache, as in

def _(s, cache={}}:
    translated = cache.get(s, None)
    if translated is None:
        translated = cache[s] = translateit(s)
    return translated

Caching on the address of s (id(s)) would be quicker, but while that's safe
for string literals, it can create false hits if _ is ever applied to
computed strings (two instances of which may have the same address if their
lifetimes are disjoint).  Explicit interning can worm around that, but
interned strings are effectively immortal.

So there are puzzles here.  It's also possible to write a little bytecode
optimizer to traverse the code object's constant vector and translate the
strings it finds once and for all; the puzzle there is which strings to
leave alone.

> ...
> it would be nice if internationalization could be achieved
> portably to all flavours of Python, just doing nothing on older
> implementations, anyway, or doing it slower.  That is, that the string
> syntax should not be modified.

Python favors explicit approaches.  Another advantage is that a fully
explicit scheme should work unchanged for JPython too (a variant of Python
that compiles directly to Java bytecode).

> ...
> Problems arise when languages have purely dynamic scoping, because textual
> domains are lexically scoped by intent.  I did internationalize a rather
> big Scheme program already, but it was all within a single textual domain,
> so I did not hit any wall there.  But I would guess that Scheme is much
> simpler than Emacs LISP, in actual practice -- I'm not sure.

Scheme, unlike elisp, is lexically scoped.  So, yes, Scheme is much
friendlier than elisp in this respect.  elisp also has interesting
complications due to, e.g., buffer-local vrbls vs global vrbls of the same
name, which is a novel kind of dynamic scoping (novel because it can't be
resolved either by reference to the static source text or by staring at the
runtime call stack!).

> Python should be simple as well.  Yet, after the XEmacs
> internationalisation failure, one good lesson, that we would be foolish
> to not retain, is that we should take time to think things out, before
> starting to think too big.

"Do the simplest thing that could possibly work" is good advice in such
situations.  We'll fix it later .

> ...
> It is not clear to me how doc strings are used in Python.

It's not yet clear to anyone.  They were inspired by elisp, but because our
Benevolent Dictator didn't decree every detail of how they were to be used,
people have thrashed for years wondering what to do with them.  The saving
grace is that they're available for runtime introspection & manipulation via
__doc__ attributes, so any explicit scheme that can be applied to runtime
strings can be applied to them as well.

If you register a module's translation domain via binding a
conventionally-named module attribute (let's pick one at random, say
"_domain"), any tool mucking with docstrings should be able to find that.
If you can get at a code body at runtime, you can get a handle to the code's
global namespace, which is an ordinary Python dict that almost always
corresponds to the module namespace in which the code body was defined.
Then you can simply look up "_domain" in that dict.  Python maintains a lot
of info for runtime introspection so you don't have to invent elaborate &
delicate compile-time schemes for retaining it yourself.

so-easy-it-will-probably-write-itself-ly y'rs  - tim





From greg.ewing at compaq.com  Wed Sep  1 00:11:08 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 01 Sep 1999 16:11:08 +1200
Subject: creating a list of class instances from file / stdin (AGAIN)
References: <37C7D49B.CA7DEAE1@pcc.net>
Message-ID: <37CCA75C.71D6590F@compaq.com>

Stephen wrote:
> 
>   What is the syntax to do:
> for line in mb_list:
>     create an instance of this class and append it to a new list????
> Can I combine the mb_list.append statement above to create instances of
> the class while reading the input?

Sure, you can:

  data = []
  for line in mb_program.readlines():
    f1, f2, f3 = string.split(line)
    data.append(MyClass(f1, f2, f3))
    
Hope that helps,
Greg



From guido at cnri.reston.va.us  Fri Sep 24 15:06:56 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 24 Sep 1999 15:06:56 -0400
Subject: decomposing an intersecting polygon?
References: <240919991023082871%joe@strout.net>
Message-ID: <5lso4486y7.fsf@eric.cnri.reston.va.us>

Joe Strout  writes:

> I'm looking for an algorithm to decompose a self-intersecting 2D
> polygon into a set of non-intersecting polygons, according to the
> "odd-even" winding rule.  That is, if the polygon defines a
> five-pointed star by five lines, the center of the star should not be
> filled.

I'm not sure if it uses the odd-even rule, nor how easily it can be
converted to what you're looking for, but there's code in Grail that
checks whether a point is inside or outside a polygon.  I'll copy it
here:

    def poly_pointin(self, x, y):
        """Is point (x,y) inside polygon with vertices coords?

        From C code by Paul Bourke at
        
        Determining whether or not a point lies on the interior of a polygon.

        The algorithm is a little pesky because a point on an edge of the
        polygon doesn't appear to be treated as *in* the polygon. Not doing
        anything about this at the moment.
        """

        counter = 0
        p1 = self.coords[0]
        for i in range(1,len(self.coords)):
            p2 = self.coords[i]
            if y > min(p1[1], p2[1]):
                if y <= max(p1[1], p2[1]):
                    if x <= max (p1[0], p2[0]):
                        if p1[1] != p2[1]:
                            xintersect = \
                              (y-p1[1])*(p2[0]-p1[0])/(p2[1]-p1[1])+p1[0]
                            if p1[0] == p2[0] or x <= xintersect:
                                counter = counter + 1
            p1 = p2
    
        return counter % 2

--Guido van Rossum (home page: http://www.python.org/~guido/)



From cgw at fnal.gov  Sat Sep 18 14:52:20 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Sat, 18 Sep 1999 18:52:20 GMT
Subject: Any good debuggers for Python? (was Re: using python debugger
 (pdb) inside emacs debugger mode ...)
In-Reply-To: <199909172318.QAA17505@hpcll563.cup.hp.com>
References: <199909172318.QAA17505@hpcll563.cup.hp.com>
Message-ID: <14307.57188.583343.140730@gargle.gargle.HOWL>

Ovidiu Predescu writes:

 > I would still like to have a pdb-mode for source files, similar
 > with gdb-mode

Could you elaborate a little bit on what you would like this mode to
do?  Do you mean being able to click on expressions in the source
buffer and print out their values, etc?  This might not be too hard to
add to pdb.el without needing any big modifications to pdb.py, in
particular I'm not sure that gdb annotations are required.  Or are you 
talking about something else?

 >
 > A better thing might be to have gdb support debugging Python programs, just
 > like it supports the other languages, but this is a different topic.

It's probably safe to say that this will never happen, support for
Python debugging would be just *too* different than the languages GDB
already supports.  But what would be good would be to have a Python
debugger that presents the identical interface as GDB, so that it
could be used in contexts where GDB is used (like XEmacs, and the
other various GUI front-ends to GDB that are out there).... This is
something I'd like to work on but quite honestly I haven't had the
time recently, possibly in a few months when I finish up some current
projects, I'll be able to get back to my "PyGDB" project.



From stephan at pcrm.win.tue.nl  Mon Sep 27 10:16:00 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 27 Sep 1999 14:16:00 GMT
Subject: Python compilers?
References: <87aeq86am8.fsf@ethereal.dhis.org>
Message-ID: 

On 27 Sep 1999 03:07:27 -0500, Nolan Darilek  wrote:
>I've been programming with C++ for nearly a year, and after seeing how
>other languages are structured, and their benefits, I'm becoming
>disenchanted with it.
>
>So, I've been searching for another language. One item which I would
>like to have, though, is the ability to compile to native code. While
>I completely agree with open source philosophies, I'd like to have the
>option to not release source if, for example, I was working on a game
>and wanted to dissuade the lazy from peeking at the script source for
>the solution. :) I know about freeze, but the created binaries are
>extremely large, which makes sense considering how they are made. :)
>
Well, you could of course code in Perl. 
That will teach them trying to read your code!
<1.5 wink>

>So, are there any native Python compilers? Someone on IRC mentioned a project
>called Viper, and my searches seem to indicate that it is under
>development, but is there any information on the web about its status?

I don't think it's ready for prime time...
and that's probably an understatement.

>One final question. How compatible/cross-platform are Python
>bytecodes? If I wanted to distribute applications in Python while
>hiding the sources from the user, would it be possible to distribute
>.pyc files, or the bytecodes specific to a specific version/platform?
>
Python's .pyc files are completely cross-platform, so you can just
distribute them. However, the bytecode format might change between
versions.

IMHO, you have basically 4 options, in order of diminishing
ease-of-implementation.
1. Don't bother about people reading the source.
   Remember, they are just spoiling their own gaming fun.
   The popular "nethack" game has source available, and yes,
   some people do indeed read it. That's up to them
2. Use CPython and distribute the .pyc files. Make sure that
   you have .pyc files available for all recent Python versions.
   Also note that Python bytecode is relatively easy to disassemble
   and "reverse engineer". Again, people who do that are spoiling
   their own game.
3. Use JPython and distribute the Java .class files. I think there
   are event .class -> C converters. This doesn't buy you anything
   speedwise compared to plain CPython, but it gives you more
   obfuscation. 
4. Have the "meat" of the program run on a server under your control,
   and only distribute a client that connects to the server program.
   This is the only setup that is *really* safe, but it's going to
   be a big hassle both for you and your players. Only do this if
   you offer a monetary prize when winning the game.


Greetings,

Stephan



From news at dorb.com  Mon Sep 27 11:40:50 1999
From: news at dorb.com (Darrell)
Date: Mon, 27 Sep 1999 11:40:50 -0400
Subject: Circular refs
References: <000501bf08f9$f25e79e0$3acbd9c2@peridot.optichrome.com>
Message-ID: 

Instead of letting Noisy hold the offending references keep them in a
dictionary and let Noisy hold a key to find it.

OR:
Untested code that I half expect to be beaten up for.
Add a destroy method to class Noisy that cleans out it's data members.
def destroy(self):
    for k in self.__dict__.keys():
         setattr(self, k, None)

Also check out tools for finding this sort of thing such as cyclops which is
a variation on plumbo.

--
--Darrell
Adrian Eyre  wrote in message
news:000501bf08f9$f25e79e0$3acbd9c2 at peridot.optichrome.com...
> Is there a reasonable fix for this?
>
> >>> class Noisy:
> ...     def __init__(self, name):
> ...         self._name = name
> ...         print "Created " + self._name
> ...     def __del__(self):
> ...         print "Destroyed " + self._name
> ...
> >>> a = Noisy("a")
> Created a
> >>> b = Noisy("b")
> Created b
> >>> del a
> Destroyed a
> >>> del b
> Destroyed b
>
> Fair enough, but...
>
> >>> a = Noisy("a")
> Created a
> >>> b = Noisy("b")
> Created b
> >>> a.b = b
> >>> b.a = a
> >>> del a
> >>> del b
> [Sound of memory dripping onto the floor]
>
> --------------------------------------------
> Adrian Eyre 
> Optichrome Computer Solutions Ltd
> Maybury Road, Woking, Surrey, GU21 5HX, UK
> Tel: +44 1483 740 233  Fax: +44 1483 760 644
> http://www.optichrome.com
> --------------------------------------------
>





From khalsa at acay.com.au  Mon Sep  6 05:51:12 1999
From: khalsa at acay.com.au (Gurvijay Singh Bhatti)
Date: Mon, 06 Sep 1999 19:51:12 +1000
Subject: Class in class
Message-ID: <37D38E90.2E5713A9@acay.com.au>

Dear everyone,

Like C++ and Java, is it possible to declare a class in a class? If it
is then could you tell how?

regards

GSB




From yopen at my-Deja.com  Fri Sep  3 18:02:44 1999
From: yopen at my-Deja.com (Bill Wilkinson)
Date: Fri, 3 Sep 1999 17:02:44 -0500
Subject: CONTEST: PyWin crashes with pure Python code!
References: <816010E2456BD111A48700805FBBE2EEFDF5A3@ex-quebec-u1.baan.com>
Message-ID: <#rQuwel9#GA.224@cpmsnbbsa03>

I am on a PIII 450 with 128 ram. and am unable to reproduce a crash, even
with greatly increasing
the size of Christian's file.

I am starting to suspect some sort of dll mismatch.
Can you tell me:
a. What service pack you are running.
b. Whether you have VC++ installed.
and if so, do you have any of the service updates installed?

Gaetan Corneau  wrote in message
news:816010E2456BD111A48700805FBBE2EEFDF5A3 at ex-quebec-u1.baan.com...
> It crashes on my IntelliStation (twin PIII 500Mhz, 256Mb RAM, 5Gb free
disk
> space).
>
> The last trace I get is "818800". I tries it on a 50Mb file I generated,
and
> it worked fine.
>
> I don't understand...
> ______________________________________________________
>    Gaetan Corneau
>    Software Developer (System Integration Team)
>    BaaN  Supply Chain Solutions
>    E-mail: Gaetan_Corneau at baan.com
>    Compuserve: Gaetan_Corneau at compuserve.com
>    ICQ Number: 7395494
>    Tel: (418) 266-8252
> ______________________________________________________
> "Profanity is the one language all programmers know best"
>





From mkrisch at radiata.com  Wed Sep 22 18:48:01 1999
From: mkrisch at radiata.com (Mark Krischer)
Date: Thu, 23 Sep 1999 08:48:01 +1000
Subject: lists and append, and loop iteration
References: <37E8860B.C024C4F8@radiata.com> <14312.40896.797841.421283@gargle.gargle.HOWL>
Message-ID: <37E95CA1.7EAB9685@radiata.com>

Charles G Waldman wrote:
> I would write
> 
>  dict['new_key'] = dict.get('new_key',[]) + [new_value]

thanks, that's just the kind of elegant trick i was looking for!

--mk



From jasbahr at NOSPAM.origin.ea.com  Mon Sep 27 20:00:52 1999
From: jasbahr at NOSPAM.origin.ea.com (Jason Asbahr)
Date: 27 Sep 1999 19:00:52 -0500
Subject: More Positions Available for Python Game Programmers
Message-ID: 

Electronic Arts (and its subsidiary Origin Systems) is looking to hire
talented Python programmers who are interested in working on massively
multiplayer online games.  Top salaries, benefits, and relocation (to
Austin, Texas) are included.  Please contact Jeff Anderson of Origin 
Systems, Inc. at 512-434-6726.  Resumes by email are encouraged, 
please send them to: janderson at origin.ea.com


-- 
Jason L. Asbahr
Origin Systems, Inc.
jasbahr(@)origin.ea.com
< Remove anti-spam text to reply. >



From doughellmann at home.com  Thu Sep  9 20:09:07 1999
From: doughellmann at home.com (Doug Hellmann)
Date: Fri, 10 Sep 1999 00:09:07 GMT
Subject: Python example: possible speedup?
References: <87puztv2gt.fsf@pc-hrvoje.srce.hr>  <87aeqw2vrc.fsf@pc-hrvoje.srce.hr>
Message-ID: <37D84D5A.B2562946@home.com>


Hrvoje Niksic wrote:
> 
> Timothy R Evans  writes:
> 
> > I had a very similar problem, I have this University project to
> > visualise the Debian package file, so I needed to read in into
> > useful language (C was discard as fast but too much work).  The read
> > tiem was really slow, so I only do it once.  After loading the file
> > and creating a data structure that stores the information, I pickle
> > it.  Read time drops by an order of magnitude.  Not so applicable to
> > your situation possibly but you could use it if you don't install
> > new packages often.
> 
> Thanks for the suggestion.  However, I do install new packages often,
> so that's not really an option.

Probably not every time you run your app, though.  You could parse the
data, pickle the results, and keep up with the mod times on the two
files.  If the source data is newer than the pickle database you have to
update the pickle data.  That would slow down some of your processing,
but in most cases would still let you take advantage of the faster
pickle version.

Doug



From gmcm at hypernet.com  Fri Sep  3 10:19:14 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 3 Sep 1999 09:19:14 -0500
Subject: NAN
In-Reply-To: <37CEFF67.B487F3FF@kla-tencor.com>
Message-ID: <1275758007-72447266@hypernet.com>

Ken Koller wrote:

> Is there a constant defined as NAN for floats?
> 
> I noticed that 1.0E500 returns the string '1.#INF' and -1.0E500
> returns '-1.#INF' I was just wondering if you could do something
> like:
> 
> f = math.NAN
> f = math.INF
> f = -math.INF
> 
> etc...
> 
> I realize that NAN is implementation dependent and that Python
> shouldn't care about the IEEE type implementations.
> 
> Seems like it could be useful to have a constant defined for these.

Some months ago the Timbot (aka Tim Peters) posted some truly 
esoteric code that tries to discover what these are for your platform 
and C library. Deja news or the search at python.org should find it.

- Gordon



From bowman at montana.com  Fri Sep 10 23:04:12 1999
From: bowman at montana.com (bowman)
Date: Fri, 10 Sep 1999 21:04:12 -0600
Subject: high school programming & python
References: 
Message-ID: <9DjC3.639$JG2.4104@newsfeed.slurp.net>

Les Schaffer  wrote in message
news:m2hfl2y2tt.fsf at netmeg.net...
> 2.) the AP tests which kids take, according to her, require for
> advnace placement in a comp sci class that they know the basics of C++
> up through objects.

Among my co-workers are several recent graduates and a former college
instructor. According to them, at the local university, a language other
than C or C++ is a prerequisite for the C courses. I can't speak for the AP
tests, but it would seem to me a good grounding in an excellent didactic
language like Python would be more valuable than a half baked exposure to
C++. My first exposure was Fortran IV thirty five years ago, and some days I
don't think I'm even ready for C++ yet.








From davidopp at megsinet.net  Sun Sep  5 14:22:54 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Sun, 05 Sep 1999 14:22:54 -0400
Subject: V - GUIDO !!!  was Re: Pirate -> RAD
References: <7qno38$6uk$1@news.is.co.za> <37D17998.5CF4@warande.uu.nl>
	 <37D1F135.6F242B93@megsinet.net> <936540416snz@vision25.demon.co.uk>
Message-ID: <37D2B4FD.9F6D9098@megsinet.net>

Dear Phil,

I don't know quite how such a program could utilize Parrot.  I am not averse to
making such a program as V - GUIDO compatible with the other major GUI tool kits such
as Tkinter and PyGtk except for the fact that this would involve a tremendous amount
of additional work for each GUI tool kit (Unless Parrot does some kind of "magic" so
that I wouldn't even have to concern myself with that).  There are different looks
and feels to all of the GUI tool kits out there.  How would Parrot account for these
differences?  Would a user of V- GUIDO choose which kind of code they wanted the
program to output,  and then a Paxo file with an argument for that type of output be
sent to Parrot and then Parrot would do this on its own?

I don't know how much additional work would be involved to maintain compatibility  as
the tool kits themselves change.  It would help to understand exactly how Parrot
works and how Paxo works.  I am not averse to incorporating useful existing code into
this program just as long as it makes things easier not just more complicated.

The advent of a WYSIWYG GUI tool such as V - GUIDO will be a great advance and may
help to give Python a bigger following.  The trick will be to make it easy enough for
a beginner to use and at the same time powerful enough for an advanced programmer to
want to use it too.   I have been looking at several GUI makers out there to figure
out which features would be good to include.

Would it be appropriate to bring the the GUI SIG back from the dead and make this a
formal ongoing project?  Instead of arguing about which GUI tool kit is better, we
could focus our energies on creating V - GUIDO and making it compatible with all of
the major GUI tool kits.

Sincerely,

David Oppenheimer

"Any Sufficiently Advanced Technology Is Indistinguishable From Magic."
- Arthur C. Clarke

P.S.  Lets make some magic happen !


Phil Hunt wrote:

> In article <37D1F135.6F242B93 at megsinet.net>
>            davidopp at megsinet.net "David Oppenheimer" writes:
> > I have an idea for creating a  WYSIWYG  GUI builder for Python that
> > will allow you
> > to create slick looking GUI's just by dragging and dropping elements into a
> >  visual
> > design space that will output code into wxPython.
>
> Why just wxPython? Python has lots of GUI toolkits, e.g. Tkinter,
> PyGtk, etc. Why not work with all of them?
>
> > After a user was satisfied
> >  with
> > the appearance of the GUI, the program would output the code into
> > wxPython code
> > which would be seamlessly integrated with the code that had already
> > been written inside of a design space (make it so that it would
> > work flawlessly as either a
> > component of the major IDE's including PythonWin, IDLE, Python in
> > a DOS box or
> > have it callable from a command line).  It will also be able to
> > take written
> >  code
> > and show a preview of what your GUI would look like without having
> > to first save the program and run it thru the Python interpreter.
> > This would allow you to
> > instantly see the effect of changes made to your code. It would be
> > awesome to
> > integrate it seamlessly with the PythonWin IDE and also be callable
> > from the
> > command line in IDLE or Python in a DOS box.
> >
> > If you would like to join with me in this creative endeavor
> > (code name V - GUIDO which stands for Visual Graphical User
> > Interface Development Optimizer), I would be happy to share the
> > details with you.
>
> I like the name!
>
> > Sincerely,
> >
> > David Oppenheimer
> >
> > P.S. This is still just in the brainstorming stage right now.  I am doing
> > conceptual planning right now; features, how it should look, how it should act,
> > etc.  I would ask anyone who is serious about doing such a project to join
> >  forces
> > with me and see if we can make something happen.
>
> I think there is a lot of potential for V-GUIDO to work with Parrot.
>
> For example, V-GUIDO could use Parrot's XML output format
> (i.e. ``Paxo'' format) as its file format.
>
> This would mean that V-GUIDO wouldn't have to include a code generator;
> it could use Parrot's. Also, V-GUIDO would be able to produce code
> for any of Parrot's backends; these will include Tkinter and wxWindows,
> and other Python and non-Python backends.
>
> e.g:
>
>  [V-Guido] --> (Paxo file) --> [Parrot] --> (Python/wxWindow code)
>                                         or
>                                         --> (Python/Tkinter code)
>                                         or
>                                         --> other formats
>
> Alternately V-GUIDO and Parrot could both use the Glade XML format.
>
> --
> Phil Hunt - - - - - - - - -  philh at vision25.demon.co.uk
>  - Linux will be 8 years old on 17th September! See: -
> http://www.vision25.demon.co.uk/prog/linuxbirthday.html




From phd at emerald.netskate.ru  Wed Sep  1 05:13:31 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Wed, 1 Sep 1999 13:13:31 +0400 (MSD)
Subject: CGI speedup strategies
In-Reply-To: <37CCE7C9.D340BBF5@strs.co.uk>
Message-ID: 

On Wed, 1 Sep 1999, Ian Clarke wrote:
> I have always been dissatisfied with the way that CGI scripts work,

   Many people feel simillary...

> loading the interpreter and then the script, not to mention making SQL
> connections, loading databases etc - and this for *each and every*
> access.  Ok, software like modpython (or is is pyApache?) allieviates

   It is PyApache, and there is hing called httpdapy.

> some of this overhead, but it is still far from ideal.
> 
> Has anyone tried actually implementing a Python webserver which just
> starts a new thread and runs a piece of Python code in it every time
> someone makes a HTTP request?  This server could then cache any objects
> (such as files, or SQL connections) in memory to be used by any Python
> scripts (this would obviously require all code to be thread-safe).

   Have you heared of Medusa? Zope? Seems Zope is THE answer for ll your
questions. www.zope.org.

> Can anyone think of any problems with this technique?
> 
> Ian.
> 
> -- 
> Ian Clarke -                    http://www.gnu.demon.co.uk/
> "A subversive is anyone who can out-argue their government"

Oleg.
---- 
    Oleg Broytmann     http://members.xoom.com/phd2/     phd2 at earthling.net
           Programmers don't die, they just GOSUB without RETURN.




From skip at mojam.com  Mon Sep 13 18:02:35 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 13 Sep 1999 17:02:35 -0500 (CDT)
Subject: List to string?  Docstrings?
In-Reply-To: 
References: 
Message-ID: <14301.29819.582168.558337@dolphin.mojam.com>

    Fran?ois> 1) For converting a string into a list, I used:

    Fran?ois>         chars = list(text)

    Fran?ois>    For getting back the string from the list, I tried:

    Fran?ois>         import operator
    Fran?ois>         text = reduce(operator.concat, chars)

The normal idiom for smashing a list into a string is

    import string
    text = string.join(chars, "")

but your method is just as effective.  The operator module and the reduce
builtin function are new additions to Python relative to the string module.
I think that would explain why most people use string.join instead of your
solution.

    Fran?ois> 2) I still miss the purpose of __doc__ strings.  It's true
    Fran?ois>    that I can write:

    Fran?ois>         print module.__doc__
    Fran?ois>         print function.__doc__

    Fran?ois>    and sometimes get useful information from these.  But what
    Fran?ois>    would I be loosing, for my own things, if I was using mere
    Fran?ois>    `#' comments instead?

Only that you deny other programs the opportunity to present you with useful
information.  For example, IDLE could presumably import a module and display
an object's doc string in a popup window if you asked for it.  It couldn't
do that with comments.  Similarly, batch-oriented documentation tools can
assemble a bunch of doc strings together into a more-or-less cohesive whole
based upon traversing a module's dict recursively, emitting the whole mess
as HTML, LaTex, nroff or what-have-you.  There is a tool floating around
called gendoc that does this sort of thing I believe.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From aahz at netcom.com  Wed Sep  8 21:39:50 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Sep 1999 01:39:50 GMT
Subject: Inspecting long running programs
References: <19990909111034.12856@nms.otc.telstra.com.au>
Message-ID: <7r7356$3b4@dfw-ixnews8.ix.netcom.com>

In article <19990909111034.12856 at nms.otc.telstra.com.au>,
Greg McFarlane   wrote:
>
>If the program is not running Tkinter, but using select.select say,
>you could create a "debug socket" which would listen for connections
>and eval() any input received on that socket.  This would be a little
>more work that the Tkinter method, but would give the same
>functionality.

asyncore (which Medusa is built on) includes this functionality.  It can
be kind of trippy to just telnet to a blind socket and get a Python
interpreter session.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From bwarsaw at cnri.reston.va.us  Fri Sep 17 15:16:26 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 17 Sep 1999 15:16:26 -0400 (EDT)
Subject: JPython and groupdict?
References: <7rtn97$rju$1@news1.news.iol.ie>
Message-ID: <14306.37770.230274.373916@anthem.cnri.reston.va.us>

>>>>> "w" == war   writes:

    w> I wonder can anyone let me know if I can get the following
    w> Python to run in JPython:

groupdict() is in the current CVS snapshot of JPython, and will be
included in the next beta release (which will probably be out sometime 
next week).

Also, JPython questions are best asked on the mailing list
jpython-interest at python.org.  See for details:

    http://www.python.org/mailman/listinfo/jpython-interest

-Barry



From menion at mindless.com  Wed Sep 29 20:26:10 1999
From: menion at mindless.com (joshua M. Schmidlkofer)
Date: Wed, 29 Sep 1999 18:26:10 -0600
Subject: Python, MySQL, and MySQLmodule.c 1.4 - Python Version.
Message-ID: <37F2AE22.AC3505FE@mindless.com>

We are trying - mind you with no success - to get Python to speak
properly with MySQL, under the Redmond curse [WinNT 4.0].  Can anyone
help me?  If windows were a knife I would be badly injured and
bleeding because the lack of any form of proper handle.


thanks,
  Joshua

We are using:
    MySQL - the registed version 3.22.xx?  The latest as of Sept 1, 99
    Python - 1.52
    MySQLmodule.c - 1.4, the one from the Mysql.[com||org] website on
the contrib page.  The author says that under NT, 'good luck'. 
However, I am a glutton for punishment.   

    Nt4.0, with SP5. [that's probably it, it's probably a stupid ploy
by M$ to get people to use VB].



From tim_one at email.msn.com  Sat Sep 11 16:42:57 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 11 Sep 1999 16:42:57 -0400
Subject: Python or Java or "X" from me and my little job
In-Reply-To: <37DAA2AB.1ABFA8E9@megsinet.net>
Message-ID: <000a01befc96$3a9834a0$6aa2143f@tim>

>>> 2)Should I go with JAVA instead?

[Gordon McMillan, communicating Stiff Truth with hirsute precision]
>> Not unless you shave with a belt sander and use a gasoline-based
>> aftershave.

[David Oppenheimer]
> Q:  Are trying to point out that Java isn't the right tool for the job
> or are you implying that Java is for insane people or both?

Writing "lightweight sysadmin" tools in a language desiged to prevent you
from making any changes to the system would be certifiably insane.  I happen
to know that Gordon does shave with a belt sander (see his Java book), but
his aftershave is extract of cayenne pepper in a base of cinnamon oil.

but-only-because-it's-cheaper-than-gasoline-ly y'rs  - tim





From bob at horvath.com  Tue Sep 28 23:03:54 1999
From: bob at horvath.com (Bob Horvath)
Date: Wed, 29 Sep 1999 03:03:54 +0000
Subject: re vs. sgmllib (was: Moving from Perl to Python)
References: <000001bf088c$2708e040$ec2d153f@tim>
Message-ID: <37F1819A.DD5CF81@horvath.com>

Tim Peters wrote:

> [Jon Fernquest]
> > Since regular expressions are just a short-hand way of specifying
> > (practically only *some*) regular languages whereas finite state
> > machines can specify any regular language the next logical step
> > would be a set of finite state tools like those that Xerox sells
> > (for several thousands of dollars I might add).
>
> Then I hope we can skip the next logical step and leapfrog illogically to a
> real parser .  Part of the problem here is that what people want to
> parse these days-- from programming language fragments to SGML --isn't
> regular.  That doesn't stop them from trying to do it with regexps, and
> input-sensitive bug-ridden code is the result.  Heck, most people find it a
> challenge to write a correct regexp to match a Python string -- or even a C
> /**/ comment.  Not that regular languages aren't useful, but I expect their
> appropriate non-trivial applications will always be a wizard art.

I have not had the time to check out the Python parsers as much as I would
like, but I have played around with JavaCC and liked it.  If there were
something in Python that was similar to JavaCC, this would be very nice.  None
of the Python parsers seem to jump out as being "the one to use". Perhaps
I need to spend some more time - a nudge in the right direction would be
appreciated?

>
> Note that Perl is in the process of adding recursive "regexps".
>
> the-mind-boggles-ly y'rs  - tim

Recursive regexps?  Yikes!




From robin at jessikat.demon.co.uk  Tue Sep 14 06:48:25 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 14 Sep 1999 11:48:25 +0100
Subject: Different Shaped Windows?
References: <37DE0F7E.714FD6F7@megsinet.net>
Message-ID: 

In article <37DE0F7E.714FD6F7 at megsinet.net>, David Oppenheimer
 writes
>Dear Group,
>
>From a posting in another  newsgroup, I am told that you can:
>
>=====================================
>"create windows that have different shapes than the
>ordinary "square" ones. You can do than thanks to the "regions" : What
>does
>that mean?
>In Win32 you can define regions and than assign them to already created
>windows. For example:
>
>SetWindowRgn(handle, CreateEllipticRgn(0,0,100,200), true)
>
>will create an elliptic window. MSDN has more info about regions. You
>can apply all sorts of operations to your regions like intersection,
>addition, etc. to obtain different shapes.
>=================================================
>
>Question is:  Can this same task (creating a non rectangular window) be
>accomplished with any of the GUI toolkits available to Python such as
>Tkinter or wxPython?
>
>David O.
>
>
>
there's a Tk extension called TkTrans which works with 8.0.5 at least to
create shaped toplevels.
-- 
Robin Becker



From gmcm at hypernet.com  Tue Sep 21 17:01:05 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 21 Sep 1999 17:01:05 -0400
Subject: Difficulty with `cmp', still
In-Reply-To: 
Message-ID: <1274174979-62990182@hypernet.com>

Fran?ois Pinard wrote:

> Hi, people.  I feel a bit opaque, as I cannot quickly get this
> code to work:

Oh no, don't feel opaque.  
 
> # Return the comparison value of two versions, of two file names
> # with versions.
> def compare_versions(a, b):
>     a0, a1, a2, a3, a4 = _weights(a)
>     b0, b1, b2, b3, b4 = _weights(b)
>     print dir(__builtins__)
>     cmp = __builtins__.cmp              # because of `import cmp'
>     above return cmp(a0, b0) or a1 - b1 or a2 - b2 or a3 - b3 or
>     a4 - b4
> 
> 
> When used in batch, I get "AttributeError: cmp".  When used
> interactively, everything seems to work fine, and
> `compare_versions' does what it should. The `print' statement is
> a mere try to understanding the problem, and it has the effect of
> adding this line to the output, just before the traceback:
> 
> 
> ['clear', 'copy', 'get', 'has_key', 'items', 'keys', 'update',
> 'values']
> 
> 
> If I do the same `print' interactively, I get (refilled in this
> message):
> 
> 
> ['ArithmeticError', 'AssertionError', 'AttributeError',
> 'EOFError', 'Ellipsis', 'EnvironmentError', 'Exception',
> 'FloatingPointError', 'IOError', 'ImportError', 'IndexError',
> 'KeyError', 'KeyboardInterrupt', 'LookupError', 'MemoryError',
> 'NameError', 'None', 'NotImplementedError', 'OSError',
> 'OverflowError', 'RuntimeError', 'StandardError', 'SyntaxError',
> 'SystemError', 'SystemExit', 'TypeError', 'ValueError',
> 'ZeroDivisionError', '_', '__debug__', '__doc__', '__import__',
> '__name__', 'abs', 'apply', 'buffer', 'callable', 'chr', 'cmp',
> 'coerce', 'compile', 'complex', 'delattr', 'dir', 'divmod',
> 'eval', 'execfile', 'exit', 'filter', 'float', 'getattr',
> 'globals', 'hasattr', 'hash', 'hex', 'id', 'input', 'int',
> 'intern', 'isinstance', 'issubclass', 'len', 'list', 'locals',
> 'long', 'map', 'max', 'min', 'oct', 'open', 'ord', 'pow', 'quit',
> 'range', 'raw_input', 'reduce', 'reload', 'repr', 'round',
> 'setattr', 'slice', 'str', 'tuple', 'type', 'vars', 'xrange']
> 
> 
> So far that I know, I'm not playing tricks nor trying to more
> clever than Python, I would think my code is rather straight. 
> I'm surely not trying to play with the builtin lists, the most I
> do is trying to get the `cmp' from it.  I do not see how or why
> the builtin list get reduced when my code is executed in batch,
> and why `cmp' is disappearing.  Would someone have a suggestion
> or an idea of what I should study to grasp what is happening?

It's a wrinkle in the space-time continuum that interactive is 
smart enough to navigate, but batch mode falls right into.

You need to:
 
import __builtin__

and use __builtin__.cmp

This is the real source of the builtins (it's a module). 
__builtins__ is a dictionary. Recall things like eval and exec, 
where you can pass a globals dict. If this dict is empty, 
Python will stuff an entry "__builtins__" into it for the eval'ed / 
exec'ed code to use. I'm not really sure of the sequence of 
events here, but apparently the shortcut hasn't been properly 
prepared when you try to use it!

BTW, I fell for it, and voluteered to carry out the BDFL's 
"suggestion".

- Gordon



From pj at sgi.com  Thu Sep  9 18:12:59 1999
From: pj at sgi.com (Paul Jackson)
Date: 9 Sep 1999 22:12:59 GMT
Subject: question about the safety of os.popen()
References: <37D823B9.9D4337EE@dnr.state.ak.us>
Message-ID: <7r9bdb$3116b@fido.engr.sgi.com>

|> mail = os.popen("/usr/sbin/sendmail -t", "w")
|> 
|> I believe that this method of using os.popen() is relatively
|> safe since it does not pass any user input directly to the

This is not safe, if it might be run by a setuid or setgid
program.

>From my internal SGI mail archives, discussing a quite
unrelated piece of code, comes the following ancient (circa
1994) paragraph, that happens to explain the problem:

|> In the original bug report, and this is popularly known,
|> you set IFS='/:' and run vi, then do a ``preserve'' command
|> which causes vi to run expreserve.  expreserve in turn does
|> a popen() to send mail to the user.  Becauses popen does
|> something like ``/bin/sh ...''  the IFS goes into effect and
|> turns ``/usr/bin/mail'' into something morally equivalent to
|> ``usr bin mail''.  The bad guy has a shell script called
|> ``usr'' in the local directory that then gets run with the
|> permissions of expreserve (setgid sys, big deal!).

As I recall, you end up wanting to purge the environment of more
than just IFS before safely invoking a subshell (os.popen or
os.system)

See also os.fork(), os.execl(), and the Python source popen2.py,
for safer alternatives and implementation ideas (mainly, avoid
the intermediate shell).

Now if you not concerned with safe setuid/gid usage, then I
believe its quite safe - the user can't get anymore done by
confusing your code than they could do directly anyway.
-- 

=======================================================================
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj



From jam at quark.emich.edu  Thu Sep 16 15:46:20 1999
From: jam at quark.emich.edu (Jeff)
Date: Thu, 16 Sep 1999 15:46:20 -0400
Subject: cgi post question
In-Reply-To: ; from Felicia Neff on Thu, Sep 16, 1999 at 12:08:23PM -0400
References: 
Message-ID: <19990916154620.B8285@quark.emich.edu>

On Thu, Sep 16, 1999 at 12:08:23PM -0400, Felicia Neff wrote:
> Yes, this is yet another newbie question:
> 
> I am trying to write a CGI script that takes input from a POST, modifies
> some of the input and then POSTs that info to another CGI script.  I know
> that perl has a function that will do this.  Is there such a function in
> python (and if it is do_POST() from the CGIHTTPServer module, is there any
> more documentation besides that at
> http://www.python.org/doc/current/lib/module-CGIHTTPServer.html?)?
> 
> Thanks -- Felicia
> 

hmm.. well, I can't offer any code that I've written to do this, but my
suggestion would be to use the built-in libraries as much as possible. if
you are using cgi scripts via apache or something, most of the work is
already done for you-- iirc, you get the POST data on 'sys.stdin'.. you can
simply call 'sys.stdin.read()' and pass it into the FieldStorage class or
something.

my advise would be to get something working in it's simplest form before
adding features like posting the results of one form into another.

questions: what are you trying to accomplish? a poll system? some kind of
order system? why do you need to modify the data and pass it to another
module? can you do it all inline to simplify the process a little?

other than the newsgroups, the documentation on python.org is indeed 'all'
of the documentation.. what specifically about using that module do you not
understand? is the problem with the module itself or with how it's
documented?

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From bthomas at trey-industries.com  Tue Sep 14 13:50:38 1999
From: bthomas at trey-industries.com (Ben Thomas)
Date: Tue, 14 Sep 1999 17:50:38 GMT
Subject: Python or Java or "X" from me and my little job
References: <37DA8D01.3D6616A@trey-industries.com>  <37DD356F.86D56591@trey-industries.com> <7rlsvn$ct0$1@ites.inria.fr>
Message-ID: <37DE8A37.9406E040@trey-industries.com>

Wow! I better get some knowledge on shell scripts too. Good grief, starting out at
square one again. Well, I almost have the script written in Python, after that I
will write a shell script and compare it. Thanks for the help.

Ben



Cedric Adjih wrote:

> Ben Thomas  wrote:
> > Actually I just want to
> > 1)get current date and come up with its string like: 01251999
> > 2)zip some files and name the archive 01251999
> > 3)ftp the archive off site
> >
> > this is my first task but I know there will be others. It isn't hard to code
> > I am just new to Python.
>
>   Well if all you need to do is that simple, maybe sticking
> with shell scripts would do. For instance:
>
> #! /bin/sh
> FILENAME=archive-`date +"%Y-%m-%d"`.zip
> FILELIST=`find /home/mydir -type f -mtime 1`
> zip -r $FILENAME $FILELIST || exit 1
> ncftpput -u thomas -p MYPASSWRD ftp.myserver.net /home/thomas $FILENAME
>
> You need to read the man pages for each of the programs, and are better
> using the standard "tar" + "gzip" if you compress files for Unix only
> (using something like "tar -czvpf $FILENAME $FILELIST")
> You can do more complicated change checking using 'md5sum', 'sort',
> 'comm' etc...
> Only after that you might want to switch to Python :-)
>
> -- Cedric
> Not speaking for INRIA.




From fredrik at pythonware.com  Wed Sep 29 05:19:02 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 29 Sep 1999 09:19:02 GMT
Subject: What GUI-lib to use
References: <37F1D710.47A82D23@bibsyst.no>
Message-ID: <00a301bf0a5b$b0acce80$f29b12c2@secret.pythonware.com>

Thomas Weholt  wrote:
> Could Tk get a more modern look perhaps and solve all our problems?

maybe, but that depends on what "more
modern look" means.






From skip at mojam.com  Thu Sep 16 00:31:23 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 15 Sep 1999 23:31:23 -0500 (CDT)
Subject: TreeCtrl Object (win32ui)
In-Reply-To: <5llna74j55.fsf@eric.cnri.reston.va.us>
References: <7rm9bb$1rh$1@nnrp1.deja.com>
	<7ro8ql$eao$1@nnrp1.deja.com>
	<5llna74j55.fsf@eric.cnri.reston.va.us>
Message-ID: <14304.29339.755635.920928@dolphin.mojam.com>

    Guido> Yes.  IDLE 0.5 has a TreeWidget module that works with Tkinter.
    Guido> It is not in any distribution yet, you need to get it via the CVS
    Guido> archives, see http://www.python.org/download/cvs.html.  IDLE is
    Guido> in Tools/idle/.

Just in case you have already linked up to the CVS repository and it doesn't 
say anything in the cvs.html page (I'm on a slowish modem at the moment, so
can't be bothered to check and will forget by tomorrow morning), be sure you 
use the -d flag to insure you get any recently created directories, e.g.:

    cvs update -d .

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From greg.ewing at compaq.com  Tue Sep 28 17:06:45 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 29 Sep 1999 09:06:45 +1200
Subject: Tkinter "scrollregion" verbosity
References: <19990918073158.A171346@vislab.epa.gov> <19990920150017.42437@nms.otc.telstra.com.au>
Message-ID: <37F12DE5.B4E35B1B@compaq.com>

Greg McFarlane wrote:
> 
> Maintaining this list across versions of Tk (and Blt
> and Tix and ...) would become a nightmare, so "always a string" is
> probably as good as you'll get :)

Methods could be provided for commonly used non-string 
attributes. Should be, I would say.

Also, how about variants of cget() which return different
types, e.g. cgeti() for getting integer attributes, to fill
any gaps in the set of methods.

Greg



From ovidiu at cup.hp.com  Fri Sep 17 19:18:46 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Fri, 17 Sep 1999 16:18:46 -0700
Subject: Any good debuggers for Python? (was Re: using python debugger 
 (pdb) inside emacs debugger mode ...)
In-Reply-To: Your message of "Fri, 17 Sep 1999 16:51:06 EDT."
             <19990917165106.A97011@vislab.epa.gov> 
Message-ID: <199909172318.QAA17505@hpcll563.cup.hp.com>

On Fri, 17 Sep 1999 16:51:06 -0400, Randall Hopper  wrote:

>  |This is in emacs 20.3, but emacs 20.4 has the same gud.el 
> 
> I'd like one for Emacs myself (ala gdb/dbx with source tracking).  GNU if
> possible but I'll do XEmacs if needed.  However, no luck with the brief try
> I gave PDB/pdb.el/XEmacs.  User friendly PDB is not (see attached for your
> amusement).

I'm using pdb with Xemacs here with no problem. Is a little different from the
gdb interface but is definitely more usable that pdb in command line. I'm using
Xemacs 21.1.6 with pdb.el that comes with it. I would still like to have a
pdb-mode for source files, similar with gdb-mode, but I guess the gdb-mode uses
the gdb annotations to be able to do that, which pdb doesn't have.

A better thing might be to have gdb support debugging Python programs, just
like it supports the other languages, but this is a different topic.

Greetings,
Ovidiu

-- 
Ovidiu Predescu 
http://andromeda.cup.hp.com/  (inside HP's firewall only)
http://www.geocities.com/SiliconValley/Monitor/7464/





From AKiss at GEOMETRIA.hu  Wed Sep  8 04:11:54 1999
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Wed, 8 Sep 1999 10:11:54 +0200
Subject: 1.5.2 NT install question(s)
Message-ID: <01BEF9E2.947FA130@exchange1.geometria.hu>

Hi,
Try to set the TK_LIBRARY and TCL_LIBRARY. 
Arpad

Mark Jackson  wrote in message
news:<7r3hib$24f$1 at news.wrc.xerox.com>...
> I've done this twice recently and encountered problems both times.
> 
> On a fresh system (never had Python or Tcl/Tk) things (for example,
> PySol) seem to work, but whenever I use Tkinter (specifically, when
> _tkinter is imported) I get a warning box about not being able to find
> tcl80.dll in the specified path.  (tcl80.dll is present, in Program
> Files\Tcl\bin.) What's wrong, and how do I put it right?
> 
> On a system where 1.5.1 (with Tcl/Tk) had previously been installed, I
> carefully uninstalled both before installing 1.5.2.  Here both "import
> _tkinter" and "inport Tkinter" are OK, but Tkinter._test() yields:
> 
>     TclError: Can't find a usable init.tcl in the following directories:
> 	{} ./lib/tcl8.0 D:/tc18.0/library {D:/Program Files/library}
>     
>     This probably means that Tcl wasn't installed properly.
> 
> (init.tcl is present, in Program Files\Tcl\lib\tcl8.0.)  What's wrong,
> and how do I put it right?
> 
> Thanks in advance.  Sorry to be asking such trivial-seeming questions,
> but I found nothing helpful through python.org.  And I haven't dealt
> with the innards of NT enough for its design coherence to become
> manifest (yeah, right).
> 
> -- 
> Mark Jackson - http://www.alumni.caltech.edu/~mjackson
> 	Spend the optimal amount of time on each decision
> 	and pretty soon you run out of life.
> 				- Steven Waldman
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From warren at nightwares.com  Fri Sep 10 12:40:37 1999
From: warren at nightwares.com (Warren 'The Howdy Man' Ockrassa)
Date: Fri, 10 Sep 1999 11:40:37 -0500
Subject: And now for something completely boneheaded...
References: <37D623A2.DC09F0CC@nightwares.com> <37D9200A.19E5277B@home.com>
Message-ID: <37D93485.50B4893C@nightwares.com>

"Eric B." wrote:

> try:
> >>> import sys      #or import 
> >>> dir(sys)          # or dir()

Got 'er, thanks. :) When I tried (as suggested by others, in addition to
your posting)

  module.__dict__

lo and behold, those documentation strings which follow functions made a
*lot* more sense...!

Using a language which *encourages* internal documentation like that is
dern near enough to bring a tear to my eye... Beautiful...

--WthmO



From wtanksle at dolphin.openprojects.net  Thu Sep  2 13:39:40 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 02 Sep 1999 17:39:40 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com><37CCE10D.D07D023B@strs.co.uk><7qjgq8$k1i$1@newshost.accu.uu.nl> <14285.38262.884171.416195@dolphin.mojam.com>  <7qm25l$ig9$1@nnrp1.deja.com>
Message-ID: 

On Thu, 02 Sep 1999 14:38:48 GMT, Mr Kelvin Lee wrote:

>btw... are 'scripting languages' the same things as 'interpreted
>lanaguages??

No.  Scripting is the act of directing something's actions; thus calling
something a scripting language implies that it's used for directing
something outside of the language.

In the old days, the only way to script an application was for the
application to include its own scripting language.  Nowadays, any language
can be used to script, and we refer to languages which are designed for
that purpose as scripting languages.  Python will not lose any of its
ability to script when someone makes a compiler for it.

And BTW, the definition "if you can't change a running program then it's
compiled..." is faulty -- there are many interpreted languages you can't
change at runtime, and there are many Lisp compilers.

>KL.

-- 
-William "Billy" Tanksley



From bthomas at trey-industries.com  Tue Sep 14 11:31:04 1999
From: bthomas at trey-industries.com (Ben Thomas)
Date: Tue, 14 Sep 1999 15:31:04 GMT
Subject: python call external program (WHAT THE?!)
References: <37DE6156.FD4384C6@trey-industries.com> <37DE68B7.49D0C5A5@mach.uni-karlsruhe.de>
Message-ID: <37DE6980.2FC85A6F@trey-industries.com>

I just tried but putting the batch file in three places on a drive. The I tried
them all.

system('d:\mybat.bat') = ok
system('d:\data\mybat.bat') = ok
system('d:\data\vacman\mybat.bat') = fails, it works fine from DOS prompt!


what gives?
Ben



Andreas Cardeneo wrote:

> Ben Thomas wrote:
> >
> > very simple question:
> >
> > What would I do to have Python execute something from the DOS command
> > line. I am very new to python but I bet it is in a moudule but don't
> > know the command or the modules to look in. I want to
> >
> > dir >%myvar.txt
> >
> > #this code would run dir from the command line and pass the var "myvar"
> > , dir would dump a listing of the files into a text file called xxxx.txt
> >
> > pointers?
> >
> > Ben
>
> Hi Ben,
>
> you should use the system() function from the "os" module.
>
> os.system("dir *.txt") shows all files in the current directory matchin
> *.txt.
> The function returns the exit code.
>
> Bye,
> Andreas
>
> --
> Andreas Cardeneo                Universit?t Karlsruhe (TH)
> Tel: +49 721 608-4576           Institut f?r F?rdertechnik und Logistiksysteme
> Fax: +49 721 75 83 78           Hertzstra?e 16 - 76187 Karlsruhe
> eMail: Andreas.Cardeneo at mach.uni-karlsruhe.de




From george.russell at clara.net  Mon Sep 27 21:40:23 1999
From: george.russell at clara.net (George  Russell)
Date: Tue, 28 Sep 1999 01:40:23 GMT
Subject: NewBie - GetOpts
Message-ID: <37f0192c.6314994@news.clara.net>

I'm trying to get a filename passed in as an argument - 
ie myprog -f myfile

So far, I have imported getopt

optlist=getopt.getopt(sys.argv[1:],'f:')

which reads the switch and the argument into a data structure from
which I can't actually access (since I don't even know what kind of
structure - a sequence? a list? a tuple? an array / vector?)(what is a
tuple?)

I'm new to python, and things have been going very well so far.

print optlist gives
([('-f','filename')],[])

I think this has two lists, one empty. Inside what data structure?

Can anyone show me howto access the data? Or point me to the right
place in the tutorial / library reference, or some example code.

Thanks.
George Russell



From pinard at iro.umontreal.ca  Thu Sep 23 12:50:17 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 23 Sep 1999 12:50:17 -0400
Subject: Sequence traversal
In-Reply-To: Sven Havemann's message of "Thu, 23 Sep 1999 17:44:21 +0200"
References: <37EA4AD5.1B54A185@tu-bs.de>
Message-ID: 

Sven Havemann  writes:

> List traversal is elegant in python! - But what if I want to step
> through two sequences in parallel?

Tee hee!  A question I know how to answer! :-) :-)

> I assume that indexing in a list l is not too efficient (O(log len(l))?),
> while stepping through l using 'for i in l' will certainly be O(1).

Indexing is efficient.  A list, despite its name, is allocated in a memory
array, so any item is accessed in constant time.  When the list grows,
the array might sometimes be reallocated -- some extra-space is included
at the end to foresee possible further grows.  A tuple is also allocated
as a memory array, but with no extra-space, because a tuple cannot grow.

> l1 = list("Names ")
> l2 = [4,2,1,5,3,9]

You might like to replace:

> for i in range(len(l1)):        #   not so cool
>   if l1[i]=='a' and l2[i]==2: print "a and 2 found!"

with:

   if ('a', 2) in map(None, l1, l2):
        print "a and 2 found!"

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From mcfletch at vrtelecom.com  Mon Sep 27 11:34:04 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Mon, 27 Sep 1999 15:34:04 GMT
Subject: PythonWin,Slider control....
In-Reply-To: <7slfak$300@journal.concentric.net>
References: <7slfak$300@journal.concentric.net>
Message-ID: 

Well, here's an example, doesn't do anything (I just ripped it out of one of
my projects, replacing an OLE control with the slider and calling the
CreateWindow method instead of CreateControl.

*Mark* The docs for win32ui.CreateSliderCtrl for some reason say to use
PySliderCtrl::Create which doesn't exist.  The docs for PySliderCtrl
correctly name the creation function as CreateWindow (though why it's
different from the OLE control I do not know).  Feel free to include the
(rather trivial) demo below if you would like to do so.

Cheers, and hope this helps,
Mike

(There are wrapping problems in the code, but you should be able to figure
those out :) )...
8<_____________________________ exampleslider.py ______________
import win32con, win32ui
from pywin.mfc import dialog

class MyDialog(dialog.Dialog):
	'''
	Example using simple controls
	'''
	_dialogstyle = (win32con.WS_MINIMIZEBOX | win32con.WS_DLGFRAME |
win32con.DS_MODALFRAME | win32con.WS_POPUP | win32con.WS_VISIBLE |
win32con.WS_CAPTION | win32con.WS_SYSMENU | win32con.DS_SETFONT )
	_buttonstyle = (win32con.BS_PUSHBUTTON | win32con.WS_TABSTOP |
win32con.WS_CHILD | win32con.WS_VISIBLE)
	### The static template, contains all "normal" dialog items
	DIALOGTEMPLATE = [
		# the dialog itself is the first element in the template
		["Example slider", (0, 0, 50, 43), _dialogstyle, None, (8, "MS Sans
Serif")],
		# rest of elements are the controls within the dialog
		# standard "Close" button
		[128, "Close", win32con.IDCANCEL, (0, 30, 50, 13), _buttonstyle],
	]
	### ID of the control to be created during dialog initialisation
	IDC_SLIDER = 9500
	def __init__(self ):
		dialog.Dialog.__init__(self, self.DIALOGTEMPLATE)
	def OnInitDialog(self):
		rc = dialog.Dialog.OnInitDialog(self)
		# now initialise your controls that you want to create
		# programmatically, including those which are OLE controls
		# those created directly by win32ui.Create*
		# and your "custom controls" which are subclasses/whatever
		win32ui.EnableControlContainer()
		self.slider = win32ui.CreateSliderCtrl( )
		self.slider.CreateWindow( win32con.WS_TABSTOP | win32con.WS_VISIBLE,
			(0,0,100,30),
			self._obj_,
			self.IDC_SLIDER
		)
		return rc
###
if __name__ == "__main__":
	dia = MyDialog()
	dia.DoModal()

-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of BGV
Sent: September 26, 1999 11:50 AM
To: python-list at cwi.nl
Subject: PythonWin,Slider control....


Has anyone managed to generate a window with a slider control in PythonWin
using
CreateSliderCtrl in win32ui? I can't seem to get it to work.

TIA

BGV









From jam at quark.emich.edu  Mon Sep 20 10:22:36 1999
From: jam at quark.emich.edu (Jeff)
Date: Mon, 20 Sep 1999 10:22:36 -0400
Subject: CygWin version
In-Reply-To: ; from Hirsch, John on Sun, Sep 19, 1999 at 09:44:42PM +1000
References: 
Message-ID: <19990920102236.A12848@quark.emich.edu>

On Sun, Sep 19, 1999 at 09:44:42PM +1000, Hirsch, John wrote:
> Sorry to appear so stupid, But if I open an MS-DOS window then install
> Doskey it installs OK and works. But if I then start Python in the same DOS
> window, Doskey does not seem to work within Python what is my problem.?
> 

I believe what the documentation is referring to is being able to use the
cygwin compiled version of 'readline' in the win32 environment, which you
ought to be able to find binaries of on the net. readline does what doskey
does for DOS-- unlike doskey, however, it allows programmers to hook into it
as a generic library, and with the cygwin compiler (a 'gcc' that runs under
win32) it can run cross-platform in UNIX and win32 environments.

in short, you (probably) need readline ;)

hope that helps.

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From vdcappel at imec.be  Wed Sep 22 10:01:48 1999
From: vdcappel at imec.be (vdcappel at imec.be)
Date: Wed, 22 Sep 1999 14:01:48 GMT
Subject: Dynamic loading on HP-UX 10 (yet again)
Message-ID: <7sang4$98j$1@nnrp1.deja.com>

   I had a problem with dynamically loading a module compiled with
 HP's ANSI C++ (aCC) compiler into python.  The dynamic loader couldn't
 find a symbol called __head, and indeed this symbol isn't defined in
 any library --- I expect a bug in the HP software.  However, __head is
 not even used anywhere, and therefore this patch to Python 1.5.2 works:

diff -u Python-1.5.2/Python/importdl.c Python-1.5.2_n/Python/importdl.c

--- Python-1.5.2/Python/importdl.c      Wed Jan 27 18:53:10 1999
+++ Python-1.5.2_new/Python/importdl.c  Wed Sep 22 14:05:03 1999
@@ -725,9 +725,9 @@
            shl_t lib;
            int flags;

-           flags = BIND_FIRST | BIND_DEFERRED;
+           flags = /* BIND_FIRST | */ BIND_DEFERRED;
            if (Py_VerboseFlag) {
-             flags = DYNAMIC_PATH | BIND_FIRST | BIND_IMMEDIATE |
+             flags = DYNAMIC_PATH | /* BIND_FIRST | */ BIND_IMMEDIATE |
                      BIND_NONFATAL | BIND_VERBOSE;
              printf("shl_load %s\n",pathname);
            }


   Has anyone else encountered the same problem?  Does this patch work
 for you?

   For the sake of completeness:
$ python
Python 1.5.2 (#6, Sep 22 1999, 14:05:17) [C] on hp-uxB
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>
$ uname -a
HP-UX yeti B.10.20 E 9000/879 unknown


   Happy,
   arnie.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gmcm at hypernet.com  Wed Sep  8 17:57:58 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 8 Sep 1999 16:57:58 -0500
Subject: Infinity and you (was Re: Turing Compliant?)
In-Reply-To: 
Message-ID: <1275298483-100087660@hypernet.com>

William Tanksley wrote:

> On 06 Sep 1999 22:20:33 -0400, Alex wrote:
> 
> >Poor William.  First manically flamed on slashdot,
> 
> You noticed!  :-)
> 
> That was a pathetic flame, though -- it appeared to be more of a
> protest against The System than against myself or my post.  What a
> rube.

My already low opinion of /. went thru the floor when I saw the flame 
was boosted because it was "humorous". I'd call /. a sports-bar for 
geeks, but that would be an insult to sports-bars.

- Gordon



From greg.ewing at compaq.com  Mon Sep  6 17:21:28 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 07 Sep 1999 09:21:28 +1200
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <935457373snz@vision25.demon.co.uk> <7qh7b6$g9g$1@wagner.wagner.home> <37CEFA88.430F28EA@compaq.com>  <37D2F842.2E55763@compaq.com> <37d35d1d.28540328@news.erols.com>
Message-ID: <37D43058.F6FA63A3@compaq.com>

Robert Kern wrote:
> 
> Umm...  Add a self argument to f.

Yes, of course, sorry.

I must never post untested code, however trivial.
I must never post untested code, however trivial.
I must never post untested code, however trivial.
...

Greg



From greg.ewing at compaq.com  Thu Sep 30 18:08:51 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 01 Oct 1999 10:08:51 +1200
Subject: Circular refs
References: <000001bf0a47$95c5fec0$ce2d153f@tim>
Message-ID: <37F3DF73.423A54E@compaq.com>

Tim Peters wrote:
> 
> The prime thing that killed adopting lexical
> closures was the cyclic trash problem then, and Greg Ewing just bumped into
> that again recently in his implementation.

Er, my implementation is actually a *solution* to the
cyclic trash problem.

Greg



From nascheme at enme.ucalgary.ca  Wed Sep 22 20:33:52 1999
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Thu, 23 Sep 1999 00:33:52 GMT
Subject: Garbage Collection in Python - current status of issue?
References: <37E8ECAD.54CEA947@strs.co.uk>
Message-ID: 

Ian Clarke  wrote:
>The FAQ (and related pages), however, while pointing out what the
>various arguments and counter-arguments are, do not indicate whether
>there is any ongoing work in this area - so is there any ongoing work in
>this area?  What is the current status of this issue?

I have a patch for Linux (it probably works on other Unixes too):

    http://www.acs.ucalgary.ca/~nascheme/python/gc.html

I've been using GC with Python for a while now.  It works fine
for me.  I haven't had the time (or the interest) to do a Windows
port yet.  Maybe if someone requests it.


    Neil



From ivnowa at hvision.nl  Wed Sep 22 04:44:28 1999
From: ivnowa at hvision.nl (ivnowa at hvision.nl)
Date: Wed, 22 Sep 1999 08:44:28 GMT
Subject: lists and append, and loop iteration
In-Reply-To: <37E8860B.C024C4F8@radiata.com>
References: <37E8860B.C024C4F8@radiata.com>
Message-ID: <199909220841.KAA05325@axil.hvision.nl>

On 22 Sep 99, at 17:32, Mark Krischer wrote:

> this might be a newbie a question, but is there a reason why append
> doesn't create a list if the list doesn't exist?
> 
> it seems to add a bit of complexity to have to do something like:
> 
> dict = {}
> 
> if 'new_key' not in dict.keys():
> 	dict['new_key'] = [new_value]
> else:
> 	dict['new_key'].append(new_value)
> 
> it seems like i should just be able to say "list.append(new_value)" and
> if list doesn't exist, 
> i get "list[0] = [new_value]

I don't think the dict can "see" the append... what you're doing is 
trying to get a value from a dict, then doing something with that 
value. So essentially it's two steps:

a = dict['new_key']
a.append(new_value)

even though you can write them as one step with 
dict['new_key'].append(new_value). If 'new_key' doesn't exist, 
nothing is returned, so there's nothing to append to.

> am i missing something fundamental that this would completely break?
> 
> 
> and while i'm bending your ear, let me ask a second question.
> 
> if i do:
> 	for element in list:
> 		do something with element
> 
> is there any to find out what my current index is? i've basically been
> doing something like:
> 	list.index(element)
> 
> this happens to work fine for what i'm doing now, but only because i
> don't care about doubles in the list, but should that ever be a problem,
> how will i know which element i'm dealing with.
> 
> i figure i can always do:
> 	for i in range(len(list)):
> 		do something with list[i]
> 
> but it just feels like i should be able to things the other way as well.

No other way that I know of... you can work around this, by keeping a 
counter, but using range() is better. 


--Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Snippets: http://tor.dhs.org/~zephyrfalcon/snippets/



From mal at lemburg.com  Wed Sep 29 16:29:04 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 29 Sep 1999 22:29:04 +0200
Subject: keys and values lining up
References:   <5l905p9a4c.fsf@eric.cnri.reston.va.us>
Message-ID: <37F27690.70A0279B@lemburg.com>

Guido van Rossum wrote:
> 
> Michael Haggerty  writes:
> 
> > There's always
> >
> >     >>> d = {'key1' : 'val1', 'key2' : 'val2', 'key3' : 'val3'}
> >     >>> apply(map, [None] + d.items())
> >     [('key1', 'key2', 'key3'), ('val1', 'val2', 'val3')]
> >
> > if you have to go the items() route.
> 
> I don't know how well that will work for a really large dictionary,
> because all arguments get pushed on the stack (not the C stack, but
> still...).
> 
> But it's a really cool trick (by which I meant I hadn't tought of it
> :-).  Might be worth noting in the Python book of idioms.
> 
> Of course, if you have NumPy, there's always
> 
>         keys, values = transpose(d.items())
> 
> which probably beats everything else.

Or maybe the lists() and tuples() function in mxTools (available from my
Python Pages):

>>> d = {'key1' : 'val1', 'key2' : 'val2', 'key3' : 'val3'}
>>> import NewBuiltins # provided by mxTools
>>> lists(d.items())
(['key1', 'key2', 'key3'], ['val1', 'val2', 'val3'])
>>> tuples(d.items())
[('key1', 'key2', 'key3'), ('val1', 'val2', 'val3')]

depends on what you want to do with the result... (and yes
tuples() and lists() is written in C too).

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                    93 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From yopen at my-Deja.com  Fri Sep  3 21:44:42 1999
From: yopen at my-Deja.com (Bill Wilkinson)
Date: Fri, 3 Sep 1999 20:44:42 -0500
Subject: Using Python in ASP scripts
References: 
Message-ID: 

Have you read this bit?
http://starship.python.net/crew/aaron_watters/pws.html

It might help.




Nikolai Kirsebom  wrote in message
news:dBYz3.682$XR4.5942 at news1.online.no...
> I have two NT 4.0 PC's running IIS 4.0 where I'm using Python in ASP
> scripts.  On the one machine I have no problems running the scripts -
> however on the other machine I can't get the scripts running.  If I change
> the LANGUAGE token to something else (VBScript) it works - so obviously it
> must have something to do with the 'registration' of Python as the
scripting
> language.  I've executed the pyscript.py and it reported all ok.
>
> When I select the 'asp' page containing the python-code, the browser seems
> to 'hang' - waiting for response from the server, and it seems to wait
> endlessly.
>
> Does anyone have any hints on where to look (variables, paths, IIS-config,
> access ????).
>
> Thanks for any help.
>
> Nikolai
>
>
>





From MHammond at skippinet.com.au  Fri Sep 10 03:58:04 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 10 Sep 1999 17:58:04 +1000
Subject: ASP with Python question
References: <37d87ed7@news.interq.or.jp>
Message-ID: <7radmm$euq$1@m2.c2.telstra-mm.net.au>

Glen Malley wrote in message <37d87ed7 at news.interq.or.jp>...

>It would really suck if the Session was the only object
>not working, because it's the only object I need...;-P

Im afraid it really sucks :-)  This is the only known bug in the Active
Scripting support :-(

Mark.






From Elie.Naulleau at westgroup.com  Thu Sep 23 14:02:22 1999
From: Elie.Naulleau at westgroup.com (Elie Naulleau)
Date: Thu, 23 Sep 1999 13:02:22 -0500
Subject: Python and Swig
Message-ID: <37EA6B2E.DE7A09AA@westgroup.com>

I am experiencing some linking problems.
I am using Python. I need to incorporate C functions
accessible from Python. To do this I am using swig (www.swig.org).
If I follow the steps advised by the documentation, everything is fine
till
the linking stage:

swig -python stemmer.i
OK
gcc -c stemmer.c stemmer_wrap.c -I/csr0/gnu/lib/python1.5/config
-I/csr0/gnu/include/python1.5/
OK
ld -shared stemmer.o stemmer_wrap.o -o stemmermodule.so
error

-shared in not recognized by Solaris linker, it is likely to be -d y
So I do it again:

ld -d y stemmer.o stemmer_wrap.o -o stemmermodule.so
Undefined                       first referenced
 symbol                             in file
Py_InitModule4                      stemmer_wrap.o
PyExc_TypeError                     stemmer_wrap.o
exit                                stemmer_wrap.o
tolower                             stemmer.o
malloc                              stemmer_wrap.o
sprintf                             stemmer_wrap.o
PyString_FromString                 stemmer_wrap.o
PyType_Type                         stemmer_wrap.o
.umul                               stemmer.o
.urem                               stemmer.o
.rem                                stemmer.o
Py_BuildValue                       stemmer_wrap.o
PyErr_Occurred                      stemmer_wrap.o
PyString_AsString                   stemmer_wrap.o
sscanf                              stemmer.o
PyModule_GetDict                    stemmer_wrap.o
qsort                               stemmer_wrap.o
fclose                              stemmer.o
_Py_NoneStruct                      stemmer_wrap.o
strncat                             stemmer_wrap.o
PyArg_ParseTuple                    stemmer_wrap.o
PyErr_SetString                     stemmer_wrap.o
strncmp                             stemmer_wrap.o
printf                              stemmer.o
__iob                               stemmer_wrap.o
PyDict_SetItemString                stemmer_wrap.o
realloc                             stemmer_wrap.o
free                                stemmer_wrap.o
strcmp                              stemmer.o
fgets                               stemmer.o
fopen                               stemmer.o
strcpy                              stemmer.o
PyInt_FromLong                      stemmer_wrap.o
strlen                              stemmer.o
PyExc_NameError                     stemmer_wrap.o
PyInt_AsLong                        stemmer_wrap.o
fprintf                             stemmer_wrap.o
ld: fatal: Symbol referencing errors. No output written to
stemmermodule.so

I running Solaris 2.5, gnu compiler, Sun solaris ld
Reading ld man page did not help. Neither compiling object files with
gcc -shared
Any help ?

Thanks

Elie Naulleau
Elie.Naulleau at westgroup.com





From hniksic at srce.hr  Tue Sep 14 12:07:47 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 14 Sep 1999 18:07:47 +0200
Subject: More random python observations from a perl programmer
References: <000601befe4c$3e585ae0$54a2143f@tim>
Message-ID: <87u2oxeass.fsf@pc-hrvoje.srce.hr>

"Tim Peters"  writes:

> [Hrvoje Niksic]
> > ...
> > I don't see the purpose of raw_input() and input() when
> > sys.stdin.readline() and eval(sys.stdin.readline()) are available.
> 
> Convenience;

My point exactly.  How often do you really need input()?  Using it to
input numbers is wrong and potentially very dangerous, as shown
before.

I think the additional verbosity of eval(sys.stdin.readline()) is a
very good indication of what is really going on.

> similarly there's no "purpose" to print given that
> sys.stdout.write() is also available.

I find `print' very convenient.  I would find it more convenient if
there were a way for it to print to stderr, though.



From ivanlan at callware.com  Fri Sep  3 23:55:21 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Sat, 4 Sep 1999 03:55:21 GMT
Subject: calling python interpreter
Message-ID: <37D09829.FF030FC2@callware.com>

Hi All--

Greg Ewing wrote:
> 

[bobbit]

>   cvx my_script
> 
> is fast enough not to be a bother, considering the
> rate at which I can write executable scripts!
> 
> (If you're wondering, cvx is named after a PostScript
> operator and stands for "convert to executable".)
> 

Um, shouldn't that be

        my_script cvx
???-)

-ly y'rs,
Ivan 
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From phd at emerald.netskate.ru  Thu Sep  9 10:38:51 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Thu, 9 Sep 1999 18:38:51 +0400 (MSD)
Subject: Query system date
In-Reply-To: <7r8fjh$dno$1@nnrp1.deja.com>
Message-ID: 

On Thu, 9 Sep 1999 mdperry_26 at my-deja.com wrote:
> I am new to python and I am trying to figure out how to return the
> computers system date and time and then format it as MM-DD-YYYY.  I have
> searched the documentation up and down and have not found anything about
> how to accomplish this.  Any help would be greatly appeciated.

   Reread docs on builtin "time" module. You need time.time(), localtime()
and so on...

> Thanks,
> 
> Mark.

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From aahz at netcom.com  Tue Sep 14 19:51:42 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Sep 1999 23:51:42 GMT
Subject: More Pythonic?
References: <7rkh5u$1qq@dfw-ixnews13.ix.netcom.com> <14302.34633.583713.527695@bitdiddle.cnri.reston.va.us>
Message-ID: <7rmn2e$jrq@dfw-ixnews7.ix.netcom.com>

In article <14302.34633.583713.527695 at bitdiddle.cnri.reston.va.us>,
Jeremy Hylton   wrote:
>
>The way I dealt with this problem in my urllib replacement (see
>http://the-tech.mit.edu/~jeremy/python/urllib2.py) was to raise an
>exception where the exception instance is actually a valid HTTP
>response object.  
>
>A typical usage would be something like this:
>
>try:
>    resp = urllib2.urlopen('http://frob.nitz/')
>except HTTPError:
>    print resp.code, resp.msg

Aside from the obvious glitch that you need to use urllib2.HTTPError,
your code generates a NameError on resp.  I think I know how to fix it,
but I'd like you to tell me in case I miss something obvious.

Overall, your code looks similar to the code I was planning to write,
but the big feature your code is missing is the stack of redirects.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From donn at u.washington.edu  Fri Sep  3 16:32:30 1999
From: donn at u.washington.edu (Donn Cave)
Date: 3 Sep 1999 20:32:30 GMT
Subject: smtplib: subject? to?
References: <19990903202313.A23408@optiplex.palga.uucp> <19990903142456.A2124@optiplex.palga.uucp> 
Message-ID: <7qpb8u$ej8$1@nntp6.u.washington.edu>

Quoth Gerrit Holl :
| On Fri, Sep 03, 1999 at 02:07:34PM -0400, Moshe Zadka wrote:
| > SMTP expects an RFC-822 message, including a To: and Subject:. Python has
| > tools to parse rfc822, but not to write, so you will have to  read
| > the docs, and use .write() like regular folk 
|
| But if I have to put a To: in the message, why do I have to give recipients?

Basically, one is for RFC822 and the other is for SMTP.  The SMTP service
actually delivers the mail, possibly to the same address as show in "To"
but at your discretion.  You may also add the address(es) shown in "Cc",
or addresses that appear nowhere in the header (``blind CC'').  It's
kind of a creepy system, but to my knowledge the SMTP service is not
expected to understand anything about the RFC822 data.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From gmcm at hypernet.com  Wed Sep  8 18:40:56 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 8 Sep 1999 17:40:56 -0500
Subject: Security considerations of execfile()
In-Reply-To: <7r6ev2$j$1@nnrp1.deja.com>
Message-ID: <1275295904-100242755@hypernet.com>

Preston Landers wrote:
> 
> I'm thinking of using the execfile() built-in function in one my
> programs (Pagecast) and I'm wondering about some security
> considerations.
> 
> Basically, the program will execute all files with a certain
> extension in a certain directory as Python code. 
...
> My question is what kind of security, if any, does execfile()
> provide (my guess: none) and should I consider working with the
> rexec module?
...
Can't really say much about rexec (except I understand it's quite 
restrictive) 'cause I haven't used it.

I've used plain execfile without worry in a sort of similar 
situation, but only I (and root) could put things in the magic 
directory.

You can pass execfile a dictionary to use instead of globals. The 
trick is, though, that if the dict does NOT have something called 
"__builtins__" in it, Python will stick the real __builtins__ in 
there.

I'm not sure what kind of attack you're worried about. Personally, 
not being the paranoid type, I'd be inclined to use just OS security 
if I could. 

- Gordon



From a.eyre at optichrome.com  Thu Sep 16 11:12:53 1999
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Thu, 16 Sep 1999 15:12:53 GMT
Subject: bsddb module
In-Reply-To: <5liu5b4htd.fsf@eric.cnri.reston.va.us>
References: <5liu5b4htd.fsf@eric.cnri.reston.va.us>
Message-ID: <001101bf0055$f2fe5e80$3acbd9c2@peridot.optichrome.com>

> You are probably confused by the BSDDB versioning.  The bsddb module
> that comes with the Python source code requires Berkeley-db *version
> 1.85*.  I don't know where to find that, but I expect that there are
> plenty of ftp archives that have it (do a websearch).  Anyway, the
> copy of bsddb 1.85 that I have has a plain vanilla BSD style license
> allowing commercial use etc. -- so I'm confident that the installer is 
> fine.

I managed to find it, but it seems to need a bit of work to port to
win32. Is the version that you use available via CVS?

--------------------------------------------
Adrian Eyre 
Optichrome Computer Solutions Ltd
Maybury Road, Woking, Surrey, GU21 5HX, UK
Tel: +44 1483 740 233  Fax: +44 1483 760 644
http://www.optichrome.com 
--------------------------------------------




From henk_p at my-deja.com  Tue Sep 28 10:38:45 1999
From: henk_p at my-deja.com (henk_p at my-deja.com)
Date: Tue, 28 Sep 1999 14:38:45 GMT
Subject: How does a CGI script get complete request header text ?
References: <7sju0v$pc2$1@nnrp1.deja.com> <7sntje$b32$1@nnrp1.deja.com> 
Message-ID: <7sqjti$7tu$1@nnrp1.deja.com>


	interesting.  I'm curious, though: how does mod_psi compare to
	mod_fastcgi and mod_pcgi?


Well, maybe you can give it a try, download at:

http://huizen.dds.nl/~d50/psi_0.01.tar.gz

I created it because I could not get mod_fastcgi to work on my
freebsd3.2 box. The fastcgi module keeps crashing the python interpreter
It is different from pcgi in that it doesn't need a 'wrapper' executable
to
connect apache to the script. So it doesn't have the overhead of
forking the wrapper for each request.


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



From JohnH at PHM.GOV.AU  Thu Sep 30 20:25:37 1999
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Fri, 1 Oct 1999 10:25:37 +1000 
Subject: Python Win32 book
Message-ID: 

Hi
Just to let you know that the 'Python programming on Win32' book will not be
released until early next year according to ORA customer service not
September this year.



From carvdawg at patriot.net  Sat Sep 11 17:03:43 1999
From: carvdawg at patriot.net (Harlan Carvey, CISSP)
Date: Sat, 11 Sep 1999 17:03:43 -0400
Subject: new to python
Message-ID: <37DAC3AF.7A6FE617@patriot.net>

I'm just starting Python programming, on NT.  I was wondering if anyone
has example scripts used for NT sysadmin'ing...

C




From warren at nightwares.com  Tue Sep  7 20:47:13 1999
From: warren at nightwares.com (Warren 'The Howdy Man' Ockrassa)
Date: Tue, 07 Sep 1999 19:47:13 -0500
Subject: Python interface to Groupwise mailoxes?
References: 
Message-ID: <37D5B211.E9C7C9F3@nightwares.com>

"S. Blakey" wrote:

> Even some good documentation of Novell's proprietary crap would
> appreciated.

That's the problem. Novell's way of doing things is, well, Novell's.
I've tried to hack into my GroupWise box at work using non-Novell tools,
in non-MS IDEs, and had nothing but grief.

However note that Novell *does* market a POP server which *in theory*
makes their G'Wise mailboxen available to any POP3 client. It's spendy
-- about US$1300 for a 50-seat license as I recall -- but it opens the
gateway, so to speak, and in large companies or companies with many
users on many OSen or many users travelling amortizes well.

I am still pushing hard at work for that implementation, as the current
scheme is intolerable (G'Wise client hoses NT4's networking, surprise.)

Don't know if they have a gateway for SMTP for sending stuff. Sigh.
Their site used to have lots of info like that available, but recently
they seem to have buried the relevant docs under a lot of noise, so it's
darn near impossible to find even the POP server, which in fact they
released under the aegis of being something like "Netscape mail server
for Novell", when you search their site. I found the POP server before
AOL sucked up Mozilla though, and that might be a factor too.

Short answer: If anyone has figured out Novell's server scheme for mail
I'd sure love to know about it, but I fear it has not been done yet. (Or
did I see a Red Hat install go by a while back which included a G'Wise
client?)

--
    warren ockrassa | nightwares | mailto:warren at nightwares.com
                    http://www.nightwares.com/



From skip at mojam.com  Thu Sep  2 17:20:05 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu,  2 Sep 1999 16:20:05 -0500 (CDT)
Subject: CGI speedup strategies
In-Reply-To: 
References: <37CCE7C9.D340BBF5@strs.co.uk>
	<036801bef458$d262c3e0$f29b12c2@secret.pythonware.com>
	
Message-ID: <14286.59784.375719.89442@dolphin.mojam.com>


    Phillip> Fredrik Lundh wrote:
    >> http://www.nightmare.com/medusa (non-threaded, ultra-fast)
    >> http://www.zope.org (threaded, based on medusa)

    Phillip> I looked at both of these a while back, and decided I didn't
    Phillip> like them because they didn't have the same sort of interface
    Phillip> that SocketServer does. So I wrote one that (sort of) has the
    Phillip> same interface, but it is unfortunately not complete:

I could have sworn I saw something posted within the past month that was
some sort of melding of the SocketServer asyncore modules, but when I
searched for it at Deja News, nothing turned up.  Was I dreaming?  Pointers
appreciated.
 
Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From rainer2207 at my-deja.com  Sun Sep 12 23:43:18 1999
From: rainer2207 at my-deja.com (rainer2207 at my-deja.com)
Date: Mon, 13 Sep 1999 03:43:18 GMT
Subject: SAX Parser Installation
Message-ID: <7rhrsk$s3u$1@nnrp1.deja.com>

I'm having trouble getting the SAX Parser to work on a Win98/IDLE
machine. Can anyone tell me how to install it? Thanks.

Rainer


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From wade at lightlink.com  Mon Sep  6 15:20:08 1999
From: wade at lightlink.com (wade at lightlink.com)
Date: Mon, 06 Sep 1999 19:20:08 GMT
Subject: how often does re.compile compile?
Message-ID: <7r1451$3n7$1@nnrp1.deja.com>

Hello,
    I'm a recovering Perl user who's using the re module to parse a
bunch of text. (Miscellaneously tagged.)

Looking at other people's code, I notice folks tend to do their
re.compile()'s at the global level.  I've got one pattern that is used
only within one function. If I say re.compile() inside the function
definition, does the pattern get compiled afresh every time the function
is called?

Thanks --

-- Wade Leftwich





Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From MHammond at skippinet.com.au  Tue Sep  7 20:29:27 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 8 Sep 1999 10:29:27 +1000
Subject: How to get the python sourcecode line, when debuging the Python C sources?
References: <7r3bqk$hbv$1@newshost.germany.net>
Message-ID: <7r4alh$7gl$1@m2.c2.telstra-mm.net.au>

Stefan Knappmann <101.139915 at germanynet.de> wrote in message

>Does anybody know in which C variable I can find the Python Stack?
>Or how to get the activ Python source code line from within the C
>debugger?

Move down the stack to eval2() in ceval.c.  f->f_lineno will tell you the
line number.  f->f_code->co_filename + 20 bytes is the start of the ASCII
string containing the filename.  Adding something like "(char *)(((char
*)f->f_code->co_filename))+20)" to the watch window should do the trick.

Mark.






From jblake at stamp-coin.com  Thu Sep  9 04:52:22 1999
From: jblake at stamp-coin.com (jonathon)
Date: Thu, 9 Sep 1999 08:52:22 +0000 (UTC)
Subject: Python Error Messages
Message-ID: 


	Why are PythonError messages so annoying?

	I've been trying to debug a script for the last four days,
	to no avail.  With two of the four databases I run it on,
	the script does just what it is supposed to.  On the other
	two databases, it doesn't.

	The current exasperating error message:

	Traceback (innermost last)
	  File "mass.py", line 21102 in ?
            print changed_field, " is record_list[28] "
          File "mass.py", line 20610 in print_web_page
            table_prefix = " "
	  TypeError:  illegal argument type for built in operation.

	**********************************************************

	Line 21100 et al are:

       if temp_field == changed_field :
           good_datafile.write( temp_line + '\n' )
           test_datafile.write( test_line + '\n' )  #  line 21102
           check_point = print_country(record_list, temp_line)
           check_point = create_country_table( record_list, country_line )
           check_point = print_web_page ( record_list )
       else:
           error_file.write( temp_line + '\n' )

	***

	Line 20607 et al are:

   make_table = '
 '
   table_prefix = " "  # line 20610
   title_name = (title_prefix + country_name + bang_two + scott_data + bang_two + michel_data )
   add_keywords = ( pre_keywords + country_name + country_section_name + scott_data + michel_data + y_t_data + s_g_data + minkus_data + end_of_line )
   file_description = ( pre_describe + description_prefix + country_name + bang_two + country_section_name + bang_two + scott_data + bang_two + michel_data + end_of_line )

	***************

	Now what is infuriating, is that the script has run over this
	particular code sequence at least 4000 times, before giving me
	this error.  

	For those that want to look at the script in its entirety, it
	is at http://www.stamp-coin.com/python/python_errors/index.html.
	[  22000 lines for the script.  The approxomate file size of
	the data files is lsited there.  ] 

	******************************

	I don't have the error message for the other frustration, but
	the python script was truncating a record, for no apparant 
	reason.  Worse still, none of the previous records had been
	truncated.  

	One of them is a simple assignment, the other is a simple
	writeline to file.

	Questions:

	What would trigger these error messages, when there were no
	apparant problems earlier?

	How do I do exception handling that maintains something sane,
	when the script obviously isn't?

	One thing i have noticed, is that the print statements I put
	in, to try to follow what was happening, tend to indicate
	that the actual problem is more than a couple of lines away
	from the line(s) that are quoted on the screen.

	xan

	jonathon




From andy at robanal.demon.co.uk  Sat Sep 18 18:54:25 1999
From: andy at robanal.demon.co.uk (Andy Robinson)
Date: Sat, 18 Sep 1999 22:54:25 GMT
Subject: Marking translatable strings
References: 
Message-ID: <37e708ce.4562650@news.demon.co.uk>

Fran?ois Pinard  wrote:

>Hello, people.
>
>So, I want the big picture right now.  That is: a technique for marking
>strings for automatic extraction and building of PO files, and a technique
>for using PO files from within Python scripts.  I foresee that Python
>introduces an usual difficulty in that the textual domain for translations
>may vary quite unexpectedly, when the control dynamically flies between
>independent packages under different textual domains.

Big picture:  are you looking for run-time translation, or
'build-time'?  i.e. should a Python application be able to switch
languages in mid-run when a user hits a menu item?  The latter can
actually be easier - each string needs to be replaced with a lookup
into a mapping from  (language, message_id) -> local message, and it
is really quick to test.

Also do you want an automatic and theoretically perfect tool to
understand every context in which quotes occur and do the right thing,
or a practical one to reduce the cost of internationalizing apps by
90%.

I made the mistake of hard-coding messages into a Delphi app I later
had to painstakingly internationalise.  I considered doing something
semi-automatic so I could use the English messages as the key, but
then hit a situation where a simple English message was used twice in
two different contexts.  Now, I have a function 'tr' - short for
translate - and an explicit constant, and do stuff like 
   ShowMessage(tr(USER_FORGOT_THE_FLOPPY));
But this involves adding USER_FORGOT_THE_FLOPPY to a separate file
every time.  

In Python, I'd be tempted to have a special tag to use at the
beginning of the string I wanted internationalized, say
"@@@USER_FORGOT_THE_FLOPPY".  Then I'd write a script to walk through
the project finding these, and on each occurrence 
(a) replace it with a function to do a run-time lookup into a
constants database, or a constant in a module, and 
(b) add the string to my database or the source of my constants
module.   We end up with a database in no particular language e.g.
{ 
...
("LINGO", "USER_FORGOT_THE_FLOPPY") : "USER_FORGOT_THE_FLOPPY"
...
}
Of course someone else needs to sit down later and add entries for all
the languages you wish to support (including your own native one), but
it does not have to be the programmer.  And your app can at least run
in development at each stage before running the database-builder
script.

Tagging the strings which really are messages and using 'constants'
seems a lot safer than messing with the language.

Come to think of it, this could be quite a useful Python package - we
could kick it off with a few dozen common error messages in several
languages to get people going!

Python-parsing-Python-beats-Delphi-parsing-Delphi'ly yours,

Andy Robinson







From da at ski.org  Mon Sep 27 11:16:58 1999
From: da at ski.org (David Ascher)
Date: 27 Sep 1999 10:16:58 -0500
Subject: README: Python-URL! - the weekly guide to Python resources (Sep 27)
Message-ID: <97BA604195CAFA9B.690B8AB0E422F93C.097AABEA8F5976F1@lp.airnews.net>

Python grows up in several ways this week:

    The standard Python documentation is available in book form by
    toExcel.com, and the royalties go to the PSA.  According to Guido
    they're nicely done.
      http://www.toexcel.com/bookstore/book_detail.asp?isbn=1583483756
      http://www.toexcel.com/bookstore/book_detail.asp?isbn=158348373X
      http://www.toexcel.com/bookstore/book_detail.asp?isbn=1583483748

    Barry Warsaw releases JPython 1.1b3, which once again allows
    optional access to non-public Java class members.
      http://www.jpython.org/

    Comp.lang.python.announce is flooded by announcements of new
    software. 
      http://www.egroups.com/listsaver/python-announce-list/

    Greg Ewing implements the extended call syntax that Guido had
    agreed to consider for 1.6 (he wins the grammar-hacker award of
    the month!).
      http://www.deja.com/getdoc.xp?AN=528523696

    In a thread on Donn Cave, master of fork & friends, points us to
    his tool to invoke shell commands on Unix and get the output.
      http://www.deja.com/getdoc.xp?AN=529129001

    The IDLE part of the CVS tree is now mirrored nightly, for those
    who want IDLE but don't care for the rest of the Python CVS tree.
      http://starship.python.net/crew/da/idledists/

    Dr. Dobb's Journal's Python Resource CD-ROM is a nice collection
    of Python distributions, tools, software, documentation and articles
      http://www.digitalriver.com/dr/v2/ec_MAIN.Entry10?SP=10023&PN=1&V1=164146&xid=2823

    The Python-URL! and Tcl-URL! archives are back, thanks to Jean-Claude
    Wippler
      http://mini.net/cgi-bin/url/

    Reminder: The Python Conference is accepting paper submissions
    until the end of the month.
      http://www.python.org/workshops/2000-01/cfp.html
      http://www.deja.com/getdoc.xp?AN=525600398


=========================================================================
Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

    Archive probing trick of the trade:
        http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
  http://www.egroups.com/list/python-url-leads/

To receive a new issue of this posting in e-mail each Monday morning,
ask  to subscribe.


-- The Python-URL! Team--
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From fredrik at pythonware.com  Sat Sep  4 05:22:07 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 4 Sep 1999 09:22:07 GMT
Subject: termios disappears during object deletion
References: 
Message-ID: <00a601bef6b7$901a3fc0$f29b12c2@secret.pythonware.com>

Brien Barton  wrote:
> Can anyone explain this behavior??  For some reason, the termios and TERMIOS
> modules get wiped out when python is running the __del__ method during
> program termination.  I've tried this with Python 1.5.1 on Linux and with
> Python 1.5.2 on SGI IRIX.  If you explicitly delete the object, you don't
> have this problem.
> 
> ~> cat test.py
> import math, termios, TERMIOS
> 
> print 'math=', math
> print 'termios=', termios
> print 'TERMIOS=', TERMIOS
> 
> class Test:
>     def __del__(self):
>         print 'deleting Test object ...'
>         print 'math=', math
>         print 'termios=', termios
>         print 'TERMIOS=', TERMIOS

starting with 1.5.1, Python performs a relatively aggressive
cleanup when the interpreter is shut down:

    http://www.python.org/doc/essays/cleanup.html

among other things, this means that destructors must hang
on to all global resources they are using, e.g.:

        def __del__(self, math=math, termios=termios, TERMIOS=TERMIOS):
            ...






From nolanl at my-deja.com  Sun Sep 12 13:43:03 1999
From: nolanl at my-deja.com (nolanl at my-deja.com)
Date: Sun, 12 Sep 1999 17:43:03 GMT
Subject: Examples using urllib, htmllib
Message-ID: <7rgon4$4a0$1@nnrp1.deja.com>

Can someone point me to some example Python
scripts that I can download that show accessing
the web using the urllib, and htmllib.  I'm
interested in automating the login, downloading,
uploading and logoff tasks that I have to do
manually now.  One problem I may have is that the
logon and other operations at the web site in
question (Freediskspace.com) appear to be Java
applets so if you have pointers on that too,
they'd be appreciated.

 thanks


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From garry at research.att.com  Wed Sep 22 15:26:28 1999
From: garry at research.att.com (Garrett G. Hodgson)
Date: Wed, 22 Sep 1999 19:26:28 GMT
Subject: high school programming & python
References: 
		 
		  <937300939snz@vision25.demon.co.uk>
		 <$JL+lCAFvt33Ewr8@jessikat.demon.co.uk>
		   <37DF9080.21A8D285@calibre-dd.com>
Message-ID: <37E92D64.6C2C3288@research.att.com>

Stephen J King wrote:
> 
> Robin Becker wrote:
> > >> You were lucky to have plugboards. We had to key things in with a bunch
> > >> of switches.
> > >
> > >Hard little metal toggle switches, too -- none of these posh molded
> > >plastic rocker switches!
> 
> Luxury! When we was kids all we 'ad was old wires which we 'ad to strip
> with our teeth and wrap around dirty binding posts with our bare
> fingers.

you had *fingers*?

-- 
Garry Hodgson			"Hey, mister, can ya tell me,
garry at sage.att.com		where a man might find a bed?"
Software Innovation Services	He just grinned and shook my hand,
AT&T Labs			"No", was all he said.



From a.eyre at optichrome.com  Mon Sep 27 11:30:37 1999
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 27 Sep 1999 15:30:37 GMT
Subject: May I embed IDLE into my Windows application?
In-Reply-To: <7qmq5v$cvv$1@newsource.ihug.co.nz>
References: <7qmq5v$cvv$1@newsource.ihug.co.nz>
Message-ID: <000601bf08fd$40104760$3acbd9c2@peridot.optichrome.com>

> When I use _tkinter_d.pyd, it doesn't crash, but IDLE still
> doesn't work.  I want Python as an extension  language embedded
> in my application, I have to embed an editor. Do you have any
> suggestion?

I found that if sys.argv is empty, then IDLE will not run. Use:

PyObject* tmp = Py_BuildValue("[s]", "dummy");
PySys_SetObject("argv", tmp);
Py_DECREF(tmp);

You also need to trap stderr to see why it doesn't work. Set
sys.stderr to be something else.

--------------------------------------------
Adrian Eyre 
Optichrome Computer Solutions Ltd
Maybury Road, Woking, Surrey, GU21 5HX, UK
Tel: +44 1483 740 233  Fax: +44 1483 760 644
http://www.optichrome.com 
--------------------------------------------




From mal at lemburg.com  Thu Sep  9 05:31:21 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 09 Sep 1999 11:31:21 +0200
Subject: Creating a python object from C
References: <7r7hs3$qn8$1@news.is.co.za>
Message-ID: <37D77E69.1BC529E0@lemburg.com>

Jacques Oosthuizen wrote:
> 
> Is it possible to instantiate a python object from a C module.!!!!

Yes, all you have to do is call the Python class object with
the constructor arguments, e.g. use PyObject_Call*(). It will
then return an instance.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   113 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From skip at mojam.com  Tue Sep  7 14:51:46 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue,  7 Sep 1999 13:51:46 -0500 (CDT)
Subject: CGIHTTPServer howto?
In-Reply-To: <37d54d12.1058989@read.news.global.net.uk>
References: <7qt9cn$rp7$1@bunyip.cc.uq.edu.au>
	<37D44115.EA834C@ix.netcom.com>
	<7r2ndf$b01$1@bunyip.cc.uq.edu.au>
	<37d54d12.1058989@read.news.global.net.uk>
Message-ID: <14293.24207.709110.125953@dolphin.mojam.com>

    Dave> If you want a full blown HTTP server written in python then take a
    Dave> look at medusa (I dont have the URL to hand, but it is listed in
    Dave> the third-party modules on the python site).

One of the easier URLs to remember:

    http://www.nightmare.com/medusa/

!

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From egelando at online.no  Fri Sep 24 04:44:20 1999
From: egelando at online.no (Egeland)
Date: Fri, 24 Sep 1999 10:44:20 +0200
Subject: Keybindings: Recognizing Control, Alt, etc on different platforms
Message-ID: <%WGG3.573$pn4.3098@news1.online.no>

Hi all. My app uses the "" binding, and I must know in the callback
procedure if thats a
Control-k or plain k coming in. On Linux, say, I can use the clause

if event.keysym=="k" and event.state==4:    # Control-k, not just 'k'
    # do something

but that wont work on win32. Is there some platform independent way of
recognizing modifications like
Control, Alt, and Shift ?


Thanks in advance,


Anders Moe





From philwise at geocities.com  Wed Sep  1 15:22:13 1999
From: philwise at geocities.com (Phil Wise)
Date: Wed, 01 Sep 1999 19:22:13 GMT
Subject: Doing POST opperations from Python
References: 
Message-ID: <37cd7cc4.8655980@news.freeserve.net>

On Wed, 01 Sep 1999 09:53:09 -0500, "Jeffrey Kunce"
 Anounced to the
internet:

:->>I am trying to produce POST requests within python
:->
:->Skip gave you a good answer.
:->Let me provide some example code.
:->Look at urlpost.py at http://starship.python.net/~jjkunce/
:->
:->  --Jeff
:->
:->
:->
Thanks for all the replys.


"640K ought to be enough for anybody"
- Bill Gates 1981



From m.faassen at vet.uu.nl  Wed Sep 29 09:45:34 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Sep 1999 13:45:34 GMT
Subject: python as a first language?
References: 
Message-ID: <7st55u$dca$1@newshost.accu.uu.nl>

Craig Hagerman  wrote:
> Hi,
> I want to get some feedback on how good Python is to learn for a beginner.
> I am interested in learning programming (OOP especially). I studied (bit
> didn't really learn) Fortran ages ago, studied formal logic (math) in
> university and have been playing around with web technologies (Javascript
> et al) and now want to try out bigger and better things.

> What are people's
> thinking regarding Perl vs Python for a beginner,

Not to start a flame war, but Python may be easier to learn for a beginner,
as Python has less special cases (in its syntax) than Perl. Python also
has a well-integrated object oriented system, while Perl's is rather 
unconventional..  Perl tends to produce programs that are a bit faster, 
and offers more extension modules (though Python already offers plenty).

For the rest Perl and Python are about equal. 

> or how about Python
> compared to Eiffel or Sather for this purpose? Any good suggestions for
> what road I should head up first?

It's hard to say. I haven't done more than glance at the Eiffel and Sather
references in the past, but they both seem to be well-designed languages.

Eiffel and Sather are less dynamic than Python. The advantage of this is
that they can compile to faster code, and that, once you have a well-designed
system, you can set it in stone and build on it without further worries.

Python is very flexible and dynamic; your objects can (under usual 
circumstances) always be changed during the running of the program. The
disadvantage of this is that Python code doesn't run as quickly as
code written in some other languages (but still more than quick enough
for most tasks), and that you set some interfaces in stone (other
programmers could always mess around with them, if they were 
malicious or stupid). But the advantage the flexibility is that you can
very rapidly code something that works, without getting trapped in
design paralysis (something that often happens to me when I write C++
code).

The quick result nature of Python, combined with the clean design, may make
it especially suited towards beginners. Beginners expect quick results to
remain motivated enough to continue learning. At the same time it's
bad to foist something onto beginners that teaches them bad programming
practices, but Python is just fine in this respect as well. Though of course
you can write messy code in any language.

Regards and good luck,

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



From da at ski.org  Mon Sep 13 03:54:18 1999
From: da at ski.org (David Ascher)
Date: 13 Sep 1999 02:54:18 -0500
Subject: README: Python-URL! - the weekly guide to Python resources (Sep 13)
Message-ID: 

This week's dispatch from the world of spam, spam and spam 'n eggs:

    Frank Willison, O'Reilly's editor in chief, muses about the
    open source communities in his column, and wears a subversive
    t-shirt when he talks to TechNetCast:
      http://www.oreilly.com/frank/
      http://www.technetcast.com/tnc_program.html?program_id=38

    Mike Olson gives a one-page (well, one HTML page!) introduction to
    CORBA, ORBs, etc. in LinuxWorld:
      http://www.linuxworld.com/linuxworld/lw-1999-09/lw-09-corba_1.htmlc

    Remco Gerlich reveals how to use Python's underappreciated
    line-completion tool automatically:
      http://www.deja.com/getdoc.xp?AN=522511625

    Jeffrey Kunce gives some sample code to interface to Novell's
    Groupwise mailboxes:
      http://www.deja.com/getdoc.xp?AN=522571735

    In a continuing heroic effort to answer as many questions as the
    bots, Robert Kern points out the best resources for folks wishing
    to learn how to program using Python:
      http://www.deja.com/getdoc.xp?AN=523189453

    The regularly expressed problem of finding strings is match()'ed 
    by, naturally speaking, Tim Peters:
      http://www.deja.com/getdoc.xp?AN=523934915

    Once again, Greg Ewing discreetly throws a grenade into the jello
    bowl.  This time, it's an implementation of a perennial favorite
    topic from the functional subspace, list comprehensions:
      http://www.deja.com/getdoc.xp?AN=523672423

=========================================================================
Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

    Archive probing trick of the trade:
        http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
  http://www.egroups.com/list/python-url-leads/

To receive a new issue of this posting in e-mail each Monday morning,
ask  to subscribe.


-- The Python-URL! Team--
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From moshez at math.huji.ac.il  Tue Sep  7 03:17:48 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 7 Sep 1999 09:17:48 +0200 (GMT+0200)
Subject: Infinite Cardinals and Pseudo-Mathematics
Message-ID: 

[A Parody of TomC's Post about Quoting Strategies]

The following message will be posted periodically until observed
clue-levels in these parts improve, or until the heat death of the
Universe arrives.

EXECUTIVE SUMMARY: 
    To send more accurate messages about infinities, please try to 
    make sure you do not make a mess out of the fairly simple concepts
    of mathematical infinities. This isn't History. People expect accurate
    statements, backed by proofs not wild assumptions.

LONG STORY:

Wouldn't you like to make your messages corrrect? If so, I have some
posting tips for you. If not, just ignore this. (Of course, if you don't
want your messages to be accurate, it's not clear why to bother to send
them in the first place. :-) I'm going to take a bit of time to explain
this, because newcomers to comp.lang.python often lack the necessary 
mathematical skills were I to send a superbrief message.

Here's the issue: you appear to have sent a mish-mash of mathematical
facts. Worst of all, you have done so by not having a firm grasp of
the definitions of the terms you used. That is what mathematics is
about, you know. :-)

If all you want to do is appear mathemtically knowledgable, that's one thing,
but here you seem to be spouting nonsense just to fill up a message. This
is neither an on-topic post, nor one with any useful information. Here's
why.

First of all, this spreads mathematical disinformation about infinities --
you're supposed to know the basic definitions of which you're talking.
Otherwise you'll probably seem like a clueless newbie to the rest of us.
This isn't really an issue of exclusivity (I know that more then a handful
of people know this stuff nowadays), so much as it is an issue of keeping
things accurate.

Second, putting forth statements like "there might be only two infinities"
does little good. It misleads others to think that this was what Goedel
or Cohen proved, when what they proved that there might be no infinities
between aleph null and the continuum.

For example, here's how an explanation about infinities should look if
you'd like them to be more effective.

    A cardinality of a set is usually taken to be the minimal ordinal
    which is equi-potent with the set. This, of course, implies the Axiom
    of choice. In particular, we have that any set is equi-potent with its
    cardinality, and thus when a set is infinite, the cardinality is
    infinite.
    
    Of course, with no axiom of choice the mechanisms to prove the axiom
    of cardinality are so ugly, we wouldn't want to think about it.
    However, a short look at the relevant texts will show that at least
    the last statement of the previous paragraph is still true.

    On the other hand, this has nothing to do with whether a Turing
    Machine needs infinite memory: it does not. After $n$ steps, a Turing
    Machine (say, for simplicity, with one ended strip), will only need
    the first $n$ places of the strip. This implies that every Turing
    Machine that stops needs only a finite amount of memory. Of course, it
    will need a different amount of memory for every input, and in
    general, that amount is infinite.

Notice how in the text above, sound mathematical arguments are presented,
with a reference to relevant texts where it would be to encumbered to 
present the argument. This is the way it should be.

If you are receiving this message in response to a news posting, please
understand that all universities provide courses about this subjects, 
so it is seldom necessary to make wild guesses. Some high schools also
offer classes on these subjects.

Here's a section about the basic mathematical infinities. Perhaps you were
unaware of these facts before you got swallowed by an arguement. It's not
only a good read; it's critical to understanding the terms you are talking
about.

  Two mathematical sets, A and B, are said to be equi-potent if there is a
  one-to-one and onto function from A to B. Equi-potency is easily seen to
  be reflexive (A is equi-potent to A), symmetric (if A is equi-potent to
  B, B is equi-potent to A) and transitive (if A is equi-potent to B, and
  B is equi-potent to C, then A is equi-potent to C). A is said to be no
  more potent then B is there is a one-to-one function from A to B. A
  clever argument, which we will not repeat here, shows that is A is no
  more potent then B and B is no more potent then A, then A and B are
  equi-potent. B is said to be more potent then A if A is no more potent
  then B, but they are not equi-potent.

  Cantor's diagonal argument proves that A is less potent then the set of
  all functions from A to the set {0, 1}. Aleph null is the set of natural
  numbers. A set is said to be of cardinality aleph null is it is
  equi-potent with the natural numbers. The continuum is the set of
  functions from the natural numbers to {0, 1}. A set is said to be of the
  power of the continuum if it is equi-potent to that set. The continuum
  hypothesis is that there are no sets which are less potent the the
  continuum and more potent then aleph null. This has been proved to be
  independant of the other axioms of set theory.

It's even more annoying when people say obviously incorrect things, or
agree with previous such things. Please don't do that.

I'm honestly not trying to annoy you!  I'm just trying to give tips
about what works well in mathematics, and what doesn't. This
used to be standard fare before one said anything said anything about
infinities, but now something seems to be lost.  
-- 
This is a parody, and should not be taken seriously. However, the gist of
it is true: please try to have a mathematical clue before saying anything
about infinities. There are traps for the uninitiated!




From warren at baker.ie  Fri Sep 17 11:38:52 1999
From: warren at baker.ie (war])
Date: Fri, 17 Sep 1999 16:38:52 +0100
Subject: JPython and groupdict?
Message-ID: <7rtn97$rju$1@news1.news.iol.ie>

I wonder can anyone let me know if I can get the following Python to run in
JPython:


[In Python]
>>> import re
>>> re.compile('a(?P.*)c').match('abc').groupdict()
{'mid': 'b'}

[JPython]
>>> import re
>>>  re.compile('a(?P.*)c').match('abc').groupdict()
Traceback (innermost last):
  File "", line 1, in ?
AttributeError: instance of 'org.python.modules.MatchObject' has no
attribute 'g
roupdict'

Thanks in advance,
Warren.





From gmcm at hypernet.com  Sat Sep 25 09:18:25 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 25 Sep 1999 09:18:25 -0400
Subject: string.split and re.split inconsistency
In-Reply-To: <7sif27$guu$1@dokkum.cs.uu.nl>
Message-ID: <1273857204-9580684@hypernet.com>

Han-Wen Nienhuys writes:
 
> Is there any particular reason why string.split and re.split tke
> arguments in different order?  I think it is confusing, and would
> prefer to have it fixed (although I realise that a backwards
> compatible fix is hard)
> 
> libre has split(pattern, string) and libstring has split(string,
> sep)

In string split, the string is the "object" and sep is a qualifier. 
In re, the "object" is the pattern. If you compile the pattern, it 
already has a split method (that takes a string). In 1.6, split 
will be a method of string objects. So the confusion will go 
away, just not the way you want. (Your solution would make 
split inconsistent with the other functions in whichever module 
you changed).

- Gordon



From nosleep at atl.mediaone.net  Sat Sep 11 12:37:28 1999
From: nosleep at atl.mediaone.net (Kristopher Johnson)
Date: Sat, 11 Sep 1999 16:37:28 GMT
Subject: yacc/lex for Python
Message-ID: <37DA861F.57212BC3@atl.mediaone.net>

Are there any lex- or yacc-like tools available for Python?

- Kris



From jfisher at are.berkeley.edu  Fri Sep  3 15:51:01 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Fri, 03 Sep 1999 19:51:01 GMT
Subject: Printing data members of a class
References: <7qk6f6$8cj$1@nnrp1.deja.com> 
Message-ID: <7qp8qv$u2j$1@nnrp1.deja.com>

In article ,
  Michael Hudson  wrote:
> John Fisher  writes:

> >
> > >>> class x:
> > ... 	data = array([1, 2])
>
> Honk! Honk! You don't actually write it like that do you? You're
> setting yourself up for the famous mutable class attribute problem.

Oh, no!  The utility of that class would be pretty limited anyway.  I
was just trying to fabricate the simplest example that illustrated my
problem.

> A point to bear in mind: __repr__ must return a string. Don't do this:
>
> class x:
>     ...
>     def __repr__(self):
>         print self.data
>
> do this:
>
> class x:
>     ...
>     def __repr__(self):
>         return `self.data`
>
> This trips me up every now and again.

Heh heh.  How about:
class x:
    ...
    def __repr__(self):
        print self.data,
        return ''

But seriously, thanks for the replies, guys.

> HTH,
> Michael
>

John


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jeremy at cnri.reston.va.us  Tue Sep  7 10:04:21 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Tue, 7 Sep 1999 10:04:21 -0400 (EDT)
Subject: "if" syntax error
In-Reply-To: <7r1i82$7s2@dfw-ixnews8.ix.netcom.com>
References: <37D138C1.3EABC811@dialnet.net>
	<7qrkg4$6g1@dfw-ixnews21.ix.netcom.com>
	<37D42D1D.8A10016A@hotmail.com>
	<7r1i82$7s2@dfw-ixnews8.ix.netcom.com>
Message-ID: <14293.7013.621596.946861@bitdiddle.cnri.reston.va.us>

>>>>> "AM" == Aahz Maruch  writes:

  AM> In article <37D42D1D.8A10016A at hotmail.com>, Andrew Kesterson
  AM>  wrote:
  >>  Ugh... I'm a C programmer by nature..  Python's indentation
  >> rules are going to take some DEFINITE getting used to....

  AM> Yup.  I still hate it.  OTOH, it does lead to more readable
  AM> code, IMO.  -- --- Aahz (@netcom.com)

Does that mean you hate readable code? 

Jeremy
 



From gmcm at hypernet.com  Tue Sep 21 14:29:42 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 21 Sep 1999 14:29:42 -0400
Subject: Python at work
In-Reply-To: <37E7B894.4CE9849F@EarthLink.Net>
Message-ID: <1274184062-62443636@hypernet.com>

Greg Copeland wrote:

> I'm a consultant and currently wanting to use Python and wxPython
> on a Win32 platform (Win95 more than likely).  The plan is to use
> this as a client interface to an Access97 database (thus the need
> for ODBC).  I have created a simple Access database and tried a
> couple of SQL queries against it from Python.  Things do seem to
> work.  Now, having said that, are there any glaring issues that
> I'm missing.  I certainly don't want to offer this solution to my
> client only to have it blow up in my face because of some obtuse
> issue; especially if it's currently well known. In short, does
> the above sound like reliable combination?  Are there any issues
> that should raise a flag or cause alarm?

If you are planning on having the DB open by multiple users, 
get off of Access and onto a real DB (Access is not a server - 
Sybase's SQLAnywhere will kick it's butt). Also use Marc-
Andre Lemburg's mxODBC module for ODBC (take the 
Starship link off python.org and look for him in the Crew 
Quarters), because it's a good deal more sophisticated.

If it's all on one desktop, use whatever feels best! (Most of 
mxODBC's advanced features will be lost on Access).


- Gordon



From tim_one at email.msn.com  Wed Sep 22 00:54:29 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 22 Sep 1999 00:54:29 -0400
Subject: YA-Newbie: baffled by deepcopy/pickle issues...
In-Reply-To: <8E48C2767PaCmAnRDLM@161.48.81.148>
Message-ID: <000101bf04b6$8d8ab520$342d153f@tim>

[Fred Pacquier]
> I'm working on my first sizable app in Python/wxPython, and I was making
> good progress until I hit this snag.
>
> ... [much background but no code and no tracebacks] ...
>
> Any hints on how to track this one down ? Thanks,

Easy:  create a *small* app that displays the same symptoms.  The answer
will likely become obvious in the process of doing that.  If it doesn't,
you'll have a small piece of executable code to post, and the answer will be
obvious to someone else.

code-fixes-itself-if-you-let-it-ly y'rs  - tim





From MHammond at skippinet.com.au  Tue Sep  7 20:49:30 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 8 Sep 1999 10:49:30 +1000
Subject: Initial thrashing about question...embedding Python
References: <37d7ad42.1540192@news.newsguy.com>
Message-ID: <7r4br5$8va$1@m2.c2.telstra-mm.net.au>

Owen F. Ransen wrote in message <37d7ad42.1540192 at news.newsguy.com>...
>If I embed Python in a Windows app which does not
>have a text output screen, where does all the printed
>output go?

As for normal Python, it goes to whereever sys.stdout or sys.stderr is
pointing.  In most GUI apps, this goes nowhere until you explicitely change
it.  Pythonwin for example simply defines an instance with a "write" method,
and sets sys.stderr and sys.stdout to that.  This class instance then
manages writing it to the GUI window.

In the sort term (ie, while developing) you may find the win32 module
"win32traceutil" useful - add this as the first line of Python code
executed, and all output will be redirected to another process (you can see
the output by running win32traceutil as a script in another process).
Again, this is what I use for Pythonwin when I manage to break its startup
code during development - at least I can see the tracebacks that prevent my
real object from being instantiated.

Mark.






From " HPeter.Friedrichs" at alliedsignal.com  Mon Sep 13 18:11:21 1999
From: " HPeter.Friedrichs" at alliedsignal.com (H. P. Friedrichs)
Date: Mon, 13 Sep 1999 15:11:21 -0700
Subject: How do you exit python during execution of script?
References: <37DD64AE.2CCA341@alliedsignal.com> <7rjq2g$oe1@dfw-ixnews14.ix.netcom.com>
Message-ID: <37DD7688.B43910B4@alliedsignal.com>

Why does this generate an exception?

import sys
try:
    print "normal operation"
    sys.exit()        # I want to leave the script here
except:
    print "something went wrong"
    sys.exit()        # instead, I dump out here...why?

#...script continues here....
#.....
#....
Aahz Maruch wrote:

> In article <37DD64AE.2CCA341 at alliedsignal.com>,
> H. P. Friedrichs  wrote:
> >
> >Is 'exit' an option? If so, how do you use it?
>
> import sys
> sys.exit()
> --
>                       --- Aahz (@netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)

--

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------





From jesse at student.usyd.edu.au  Tue Sep 14 10:37:43 1999
From: jesse at student.usyd.edu.au (Jesse Sweeney)
Date: Tue, 14 Sep 1999 14:37:43 GMT
Subject: equivalent of enum?
References:  <14296.14104.456361.61180@dolphin.mojam.com> <37DE4A92.D4F50CE9@strs.co.uk>
Message-ID: <37de5928.24771401@news.usyd.edu.au>

On Tue, 14 Sep 1999 14:16:03 +0100, Ian Clarke  wrote:

>
>> In more direct answer to your question, there is no enum type in Python.
>> It's simple enough to declare objects that you treat as constants, however:
>
>I have always been a little dissatisfied with the way enums are normally
>implemented, namely with an integer.  Why not make an "enum collection"
>type (I am sure there must be a better name).  In essence this would be
>an object which had several enums, say:
>
>Sex: (male | female)
>employed?: (yes | no)
>os: (linux | windows | macos)
>
>etc.  Normally these would be treated as 3 enums and would be stored as
>three integers - 3 X 32 bits, or 96 bits, yet you can actually store
>this data using just 4 bits (and that would contain a little bit of
>redundancy).  The first bit would determine male or female, the second
>bit employed or unemployed, and the third and fourth bits would store
>the os.
>
>In practice this could be implemented by creating an enum class.  We
>could then create a function which takes an object containing enums (and
>possibly allow these to be mixed with other data types) and compresses
>the data bit by bit.  In the end we would get really efficient data
>storage.  

This is the equivalent of a C struct with bit fields (and I assume you could
implement it that way for a Python extension type). However, the general
consensus[1] over in C-land is that using bit fields is usually not a great
idea, or at least usually a misguided one. The reason being that although the
data space is reduced, the size of the code generally increases. Also, since
integers conveniently fit so nice and snugly into CPU registers, accessing
fields in C is typically a fair bit slower than accessing the separate ints or
chars.

	Cheers, Jesse.


[1] Offtopic trivia of the day: I recently learned that 'consensus' is the most
frequently misspelled word in the English language. 'Brunet' comes in second.



From quentin at att.com  Thu Sep 30 04:00:31 1999
From: quentin at att.com (Quentin Stafford-Fraser)
Date: Thu, 30 Sep 1999 09:00:31 +0100
Subject: ANN: PyGarmin - Python classes for talking to Garmins
Message-ID: <37F3189F.79D8413E@att.com>

PyGarmin - Python classes for talking to Garmins
================================================

I've written some Python classes which implement much of the protocol
used by Garmin GPS receivers to talk to each other and to PCs.  It's
based on the official Garmin spec.

It's certainly not complete yet, but it allowed me to write a small app
which transfers waypoints and track information to and from my GPS 12XL.

You can find out more at http://www.uk.research.att.com/~qsf/garmin/ .
Let me know if it's useful!

Quentin
-- 
----------------------------
Dr Quentin Stafford-Fraser
AT&T Laboratories Cambridge
http://www.uk.research.att.com/~qsf



From 71351.356 at compuserve.com  Sat Sep 25 10:27:06 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Sat, 25 Sep 1999 10:27:06 -0400
Subject: Source line number for log purpose (newbie)
References: <7shiqh$d96$1@ssauraaa-i-1.production.compuserve.com> <19990925130431.B11328@optiplex.palga.uucp> <7sig2t$oar$1@ssauraaa-i-1.production.compuserve.com>
Message-ID: <7simbr$7s3$1@ssauraac-i-1.production.compuserve.com>

Ok.  I looked in Exception library and don't see the relevance.   I am
logging normal code, not exceptions.   For example,
log.prtl( SERIOUS, "About to create dest file %s" % newfilename)

In this case, to help maintainers of the code, I want to also log the line
number where this statement is found in the source.

I'm now thinking this has to done by digging into the Python extension
mechanisms and system variables deep in the system?


T. C. Mits <71351.356 at compuserve.com> wrote in message
news:7sig2t$oar$1 at ssauraaa-i-1.production.compuserve.com...
> Thanks, I'll look at the Exception module again.
>
> Gerrit Holl  wrote in message
> news:19990925130431.B11328 at optiplex.palga.uucp...
> > T. C. Mits:
> > > In c there was a way to get the current source line number via a
macro,
> > > __LINE__.
> > >
> > > I need to be able to do the same thing.  I created a logger class that
> uses
> > > a level setting.  At high levels I want to
> > > log the source line number where the inline log.prtl( level, msg)
method
> > > call appears.
> > >
> > > I'm sure this is easy but am getting lost digging through the Python
> source
> > > and docs.
> > >
> >
> > That's possible via the 'Exception' module.
> >
> > regards,
> > Gerrit.
> > --
> > RTFM: Read That FreshMeat
> >
> >
>
>





From bwarsaw at cnri.reston.va.us  Mon Sep 20 20:47:04 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 20 Sep 1999 20:47:04 -0400 (EDT)
Subject: Marking translatable strings
References: 
	<7s339v$d78@dfw-ixnews3.ix.netcom.com>
Message-ID: <14310.54664.634772.150567@anthem.cnri.reston.va.us>

>>>>> "AM" == Aahz Maruch  writes:

    AM> Y'know, it might help those of us with an interest in I18N but
    AM> little actual experience if you posted some resources that
    AM> gave information about what the heck a "PO file" is and how
    AM> one is suposed to generate and access it in other languages.
    AM> (Yes, I could search the web, but it'd be difficult for me to
    AM> determine which sites are the most useful.)

I thought I had included a link to the GNU gettext package.  Between
it, the Solaris manpages, and Martin von Loewis' paper from IPC6,
you've about exhausted my knowledge of internationalization issues.
Hopefully some other folks will have more useful information, but here 
are two links:

GNU gettext (follow the online manual link)

    http://www.gnu.org/software/gettext/gettext.html

Martin's paper (search down towards the end of the page)

    http://www.python.org/workshops/1997-10/proceedings/

-Barry



From doughellmann at home.com  Wed Sep  8 21:14:31 1999
From: doughellmann at home.com (Doug Hellmann)
Date: Thu, 09 Sep 1999 01:14:31 GMT
Subject: carriage return / Tkinter
References: <7r5s4q$91n$1@nntp3.uunet.ca>
Message-ID: <37D70B2E.B0DED81D@home.com>


Greg Wilson wrote:
> 
> Hi.  I've created an Entry widget using Tkinter, and would like a command
> run when the user presses carriage return --- how do I do this?
> Thanks,
> Greg

Bind your callback to the event '':

entry.bind('', my_handler)

def my_handler(event):
  print 'Got a "" event.'

Doug



From mwh21 at cam.ac.uk  Fri Sep  3 09:48:19 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Fri, 3 Sep 1999 14:48:19 +0100
Subject: UK Python Programming Job Opportunity
In-Reply-To: <7qof8u$mh2$1@newsreader1.core.theplanet.net>
References: <7qof8u$mh2$1@newsreader1.core.theplanet.net>
Message-ID: 

On Fri, 3 Sep 1999, John Parkey wrote:

> If there is anybody who is interested in a 6-9 month contract using Python,
> in Bucks UK, please drop me a line.

Not yet... 
 
> jparkey at rapidbs.com
> 
> (I hope nobody is offended by my posting this to the newsgroup?)

Well, I'm certainly not, and I'll be even less offended if you are still
doing it in twenty months or so when I graduate...

Regards,
Michael




From fredrik at pythonware.com  Tue Sep 28 04:58:25 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 28 Sep 1999 08:58:25 GMT
Subject: Python Berkley DB and speed issues
References: <37F06752.7D98775@bibsyst.no>
Message-ID: <009001bf098f$a1599840$f29b12c2@secret.pythonware.com>

Thomas Weholt  wrote:
> Was just wondering about how Python is when it comes to speed using it`s
> built-in dictionaries. I`m converting a perl-project where a Berkley DB
> hash-table stored several million entries about files stored on cd-roms.
> The speed in the perl-project was ok, and it seems like Python has a
> very similar approach to this method, using pickle- and shelve-modules (
> or Berkley DB). My question is : will the speed drop as the data gets
> larger, using pickle and shelve, or do I have to use Berkely DB?

note that shelve runs on top of a hashed database engine,
such as Berkeley DB (via dbhash/bsddb) or gdbm.  if no such
engine is installed, it falls back to a "slow-and-simple" pure
Python implementation:

http://www.python.org/doc/current/lib/module-anydbm.html






From fdrake at acm.org  Tue Sep  7 15:28:10 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Tue, 7 Sep 1999 19:28:10 GMT
Subject: CGIHTTPServer howto?
In-Reply-To: <37d54d12.1058989@read.news.global.net.uk>
References: <7qt9cn$rp7$1@bunyip.cc.uq.edu.au>
	<37D44115.EA834C@ix.netcom.com>
	<7r2ndf$b01$1@bunyip.cc.uq.edu.au>
	<37d54d12.1058989@read.news.global.net.uk>
Message-ID: <14293.26442.739967.306563@weyr.cnri.reston.va.us>

Dave writes:
 > python library as it stands (or at least the fact that it is UNIX
 > specific should be documented). It would be better off included with

  I've adjusted the documentation source to reflect this; thanks for
pointing it out.


  -Fred

--
Fred L. Drake, Jr.	     
Corporation for National Research Initiatives



From davidopp at megsinet.net  Fri Sep  3 08:48:12 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Fri, 03 Sep 1999 08:48:12 -0400
Subject: Spanish Inquisition?!
Message-ID: <37CFC38C.20A9C267@megsinet.net>

In a previous thread (about languages interpreted versus languages
compiled) there was talk of the Spanish Inquisition.  I am guessing that
this is 1) a joke or 2) the code name for a "secret" project to create
compiled code from python code and thereby permanently blur the line for
Python.  Anyone know anything about this?

I know no one ever expects the Spanish Inquisition, but if its coming
I'd like to know about it.

David O.

"Any Sufficiently Advanced Technology Is Indistinguishable From Magic"
- Arthur C. Clarke

P.S. Creating compiled code from Python  that didn't require the
interpreter to be distributed with it would be high magic indeed.




From guido at cnri.reston.va.us  Thu Sep  2 08:44:40 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 02 Sep 1999 08:44:40 -0400
Subject: And Now For Something Completely Similar !!!
References: <37CCDAF8.E79FE189@megsinet.net> <028f01bef450$9d976ba0$f29b12c2@secret.pythonware.com> <37CD22A5.4DD4B332@megsinet.net> <006f01bef47c$c0679b10$f29b12c2@secret.pythonware.com> <37CD9117.DD7E5100@megsinet.net> <37cd9f19.13139313@news.erols.com>
Message-ID: <5l906p5vpj.fsf@eric.cnri.reston.va.us>

kernr at mail.ncifcrf.gov (Robert Kern) writes:

> On Wed, 01 Sep 1999 16:48:24 -0400, David Oppenheimer
>  wrote:
[...]
> >2) Why isn't a module which has broad appeal like the Python Imaging Library
> >included with the core Python 1.5.2 package?  
> 
> Release schedules, lean&mean-ness, not enough people needing to
> manipulate images in Python, etc.

You said it.  The key issues here are independent developers and
independent development schedules; I can't afford to repackage Python
each time a new PIL comes out and I certainly don't want to have to
wait until PIL is stable before releasing Python.

This is a fact of life, and it's how most software distributions are
done.  I *do* package a Tcl/Tk version with Python 1.5.2 for Windows
(as you noetd latest in your post); this was done because (1) Tcl/Tk
8.0.5 was pretty stable at the time, and (2) I needed Tcl/Tk in order
to make IDLE usable.  It turns out that Tcl/Tk 8.1 and 8.2 are not
compatible with standard Python 1.5.2, so this ended up being a good
decision; nevertheless I may decide to drop this in the future (or I
may decide to keep it) depending on the pace of development in the
Tcl/Tk world.

Note that this opens up a market for "packaging companies" (like Red
Hat does for Linux) which bundle stable versions of lots of software
together for ease of installation.  More power to them...!

> >For that matter why not include
> >NumPy also.  
> 
> 
> That one, I'd have to agree with.  I think that NumPy arrays and
> ufuncs should become core datatypes.  They're just too useful.
> 

Yeah...  This is actually in my plans for Python 1.6 (although I
haven't started working on it honestly).

--Guido van Rossum (home page: http://www.python.org/~guido/)



From MHammond at skippinet.com.au  Thu Sep  2 01:32:25 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 2 Sep 1999 15:32:25 +1000
Subject: Adding manual access to IDLE
References: <37CB245C.3AF@loop.com>  <37CDB31B.EFF@loop.com>
Message-ID: <7ql25d$si5$1@m2.c2.telstra-mm.net.au>

Jesse Weinstein wrote in message <37CDB31B.EFF at loop.com>...
>I realise that my code is rather unportable ;) but I don't know how to
>get more portable code, such as portable browser or doc location.  If
>anyone can help, it would be gratefully apreacieated.

I can help with _less_ portable code 

For Windows, it may make sense to attempt an import of "win32api" - if that
succeeds, you have access to WinExec() which will invoke the correct program
registered for HTML, and also to the registry functions that you can use to
get the location of the help files (assuming they have not been moved since
installation.

Mark.






From jsouto at dnet.br  Sat Sep 11 09:24:27 1999
From: jsouto at dnet.br (Jairo Souto)
Date: Sat, 11 Sep 1999 10:24:27 -0300 (EST)
Subject: Text UI
Message-ID: 

dwalker at iximd.com wrote:
> [snip]
>     It's not perfect, could stand to be looked over by a real Pythoneer, but
> it works for my use right now.  I have webspace so I probably should put it
> up for anyone to come and get it.  In the meantime, if someone wants to play
> with it, send me email and I'll send it to ya right away.
> 
>                         Damond

yes. I want look at it as I need to go on ttys for an
internal mysql application.

--Jairo Souto  jsouto at dnet.br  (+55) (38) 9971-0014





From kernr at mail.ncifcrf.gov  Wed Sep  1 18:28:19 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Wed, 01 Sep 1999 22:28:19 GMT
Subject: Printing data members of a class
References: <7qk6f6$8cj$1@nnrp1.deja.com>
Message-ID: <37cda459.14483153@news.erols.com>

On Wed, 01 Sep 1999 21:39:53 GMT, John Fisher
 wrote:

>Hey folks,
>
>I'm trying to set up a class so that when instances of it are printed, a
>Python prints just a certain data member of the instance.  For example:
>
>>>> class x:
>... 	data = array([1, 2])
>
>>>> y = x()
>>>> y
>[1 2]
>
>or at least
>>>> print y
>[1 2]
>
>I've tried playing around with repr, and defining __repr__() in the
>class, but haven't had any luck yet.  I'd appreciate any suggestions on
>how to do this.

from Numeric import *
# That is what you're using, right?

class x:
    data = array([1, 2])

    def __str__(self):
        return str(self.data)

    __repr__ = __str__  
    # If you must; it would be better to have __repr__ return a string
    # that could be meaningfully eval'ed, but if you need the
    # behaviour in your first example, this is how to do it.

>Thanks!
>
>John

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From moshez at server.python.net  Sat Sep  4 02:06:29 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Sat, 4 Sep 1999 02:06:29 -0400 (EDT)
Subject: raw_input() and readline
Message-ID: 

The documentation of raw_input() says it trys to use readline whenever
available, but there seems to be a problem:

file t.py:
print raw_input()

If I use python

>>> execfile("t.py")

Then I have readline functionality, but if I use 

python t.py

I don't have any readline functionality.
I would appreciate any help.




From bob at horvath.com  Sat Sep 25 08:53:20 1999
From: bob at horvath.com (Bob Horvath)
Date: Sat, 25 Sep 1999 12:53:20 +0000
Subject: wxPython on Debian Linux
References:  <37E8C20A.F4051E56@horvath.com> <87u2omg1yc.fsf@baal.winnegan.fake>  <87hfkkrth1.fsf@baal.winnegan.fake>
Message-ID: <37ECC5C0.F64195A2@horvath.com>

Siggy Brentrup wrote:

> Lars Marius Garshol  writes:
>
> > * Bob Horvath
> > |
> > | I haven't tried it myself, but itsn't it at
> > |
> > | http://www.debian.org/Packages/unstable/interpreters/python-wxwin.html
> >
> > * Siggy Brentrup
>
> [...]
>
> > | But you can always get the source package and build you own local
> > | .deb.
>
> > If you do, please let me know.
>
> No chance, I have python installed locally updating the sources
> regularly from the CVS tree and am using Tkinter and Pmw 0.8.1.
>
> Taking too much space, Bob?

Bope, I just istalled it.  Very nice!

>
>
> CU
>   Siggy
>
> --
> noch nichts Aufregendes:
>
> Siggy Brentrup - bsb at baal.infodrom.north.de - voice: +49-441-6990134




From bowman at montana.com  Tue Sep 14 09:58:54 1999
From: bowman at montana.com (bowman)
Date: Tue, 14 Sep 1999 07:58:54 -0600
Subject: French pun on Python! :-)
References: 
Message-ID: <37DE549E.2FBDB88A@montana.com>

Fran?ois Pinard wrote:
> 
> for the word `piton'.  A `piton' is a familiar word to design any mechanical
> device on which someone pushes, like an electrical button, or often, one

And a 'piton' is a small mechanical aid one hammers into obscure cracks
to allow climbing an otherwise impossible pitch. Maybe the IS a parallel
here.

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From khowe at performance-net.com  Fri Sep 17 14:07:18 1999
From: khowe at performance-net.com (Kevin Howe)
Date: Fri, 17 Sep 1999 15:07:18 -0300
Subject: lopping over Dictionaries
Message-ID: <3uvE3.157430$5r2.248606@tor-nn1.netcom.ca>

Is it possible to iterate through list of keys in a Dictionary?

something like this:

tel = {'jack': 4098, 'sape': 4139}

for (name, value) in tel.keys:
 print name, value


Thanks





From MHammond at skippinet.com.au  Tue Sep 14 19:29:01 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 15 Sep 1999 09:29:01 +1000
Subject: TreeCtrl Object (win32ui)
References: <7rm9bb$1rh$1@nnrp1.deja.com>
Message-ID: <7rmlor$jk4$1@m2.c2.telstra-mm.net.au>

Ian wrote in message <7rm9bb$1rh$1 at nnrp1.deja.com>...

>If I were to be doing this in C++ I would use the HitTest function
>and pass to it the point at which the right mouse button was depressed.
>This function doesn't appear to be available through win32ui.pyd. Is
>there another way around this?

Nope - I struggled with the same thing - the half-baked registry editor
stopped where it is for exactly that reason :-(

Please feel free to either add the missing methods yourself, or tell me
exactly what you need added to make this work.

>I'm new to using this library (1 week) and this has been the first
>real roadblock I have faced. My prototype application has really
>advanced very quickly and I think it would have taken me at least
>3-5 times as long if I were to be using C++. Great work Mark and
>all those who have (and continue to) help you.

Thanks :-)

Mark.






From aahz at netcom.com  Sun Sep 26 11:50:42 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Sep 1999 15:50:42 GMT
Subject: Intelligent datastructure?
References: 
Message-ID: <7slfci$snu@dfw-ixnews8.ix.netcom.com>

In article ,
Magnus L. Hetland  wrote:
>
>I only have two datatypes - integers and sequences. The thing I have
>been thinking about lately is the implementation of the sequences. I
>want them to be usable both as simple lists or arrays, and as general
>hash-tables (with numeric keys) without the user having to care about
>the difference... Thus, you could make yourself a sequence like this:
>
>  s = [1,2,3]
>
>and then do something like
>
>  s[100] = 4
>
>without getting any complaints. (I think I will only allow positive
>integers as indices, but I'm not sure...)
>
>If I now reference s[50], for instance, or s[400], I would like to get
>0, which is the default value for all variables (or elements).
>
>I would like feedback on that strategy (of basically allowing anything
>to be done without throwing index-exceptions or stuff like that), but
>what I am mainly interested in is the task of making these sequences
>"intelligent".

Being generally interested in simple code, if I were in your shoes I'd
probably just use a dict-based class.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From aa8vb at yahoo.com  Thu Sep  9 09:32:06 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 9 Sep 1999 09:32:06 -0400
Subject: tkinter: identifying callback button?
In-Reply-To: <7r7anm$21a7$1@hub.org>; from Lewis on Thu, Sep 09, 1999 at 03:49:10AM +0000
References: <7r7anm$21a7$1@hub.org>
Message-ID: <19990909093206.A1428575@vislab.epa.gov>

Lewis:
 |I have a bunch of (named) buttons which call the same callback.
 |I've been trying to figure out how (or if) I can get the callback
 |function to be able to identify which one.
 |
 |    for x in range(0,5):
 |        b = Button(frame,text='#%s' % x,command=self.xxx, name='b%s' % x)
 |        b.pack()

I asked this several months ago.  Here are the answer highlights:

  1) Use a generic callable object as a wrapper (__call__)
  2) Use a specific callback class object
  3) Use a lambda or function (written in-line or generated by a function)
  4) Write python call code as a string and "exec" it in a lambda or function

They sound tricky, but they're really not.  The attached simple code shows
how to use these techniques for both full and partial resolution of
callback arguments at registration time.

(Note: "partial" means specifying some of the arguments at registration
time, while leaving others until invocation time -- e.g. for cases where
your Tk callback takes arguments).

My personal favorite for full resolution is #1 (callable object).  Clean
appearance, and no custom code to write per-callback to make it fly.

Randall



-------------- next part --------------
#!/usr/bin/env python
#
#  closure.py  -  Demonstrate various ways to generate and use
#                 parameterized callback functions.
#
#                 Examples are shown for both total and partial resolution
#                 of callback arguments at registration time.
#
#                 Techniques:  lambdas, callable objects, deferred execution.
#

class Call:
  """Instances of this class store a function as well as a list of arguments.
     When they are called, the function will be called together with the
     arguments used for creating the instance.
     Slightly different than lambda, but nicer syntax."""
  def __init__ (self, func, *args):
    self.func = func             # save the function (or bound method, or ...)
    self.args = args             # save the arguments to use 
  def __call__ (self):
    apply (self.func, self.args) # call function, using args as arguments.

class CallbackHookFull:
  """
  Demonstrates various ways to do full resolution of callback arguments
  at registration time, and return/invoke the callback.
  """
  
  def ExecCommand(self,command):
    exec command

  def DoIt(self,color):
    """ The callback we eventually want to end up calling, but with no args. """
    print color

  def Invoke(self):
    self.doit_cb()

  #---------------------------------------------------------------------
    
  def Register1( self, color ):
    """ Generate a callable class that'll invoke the callback w/ args """
    self.doit_cb = Call( self.DoIt, color )

  #---------------------------------------------------------------------
    
  class ColorCallback:
    def __init__( self, obj, color ):
        self.obj   = obj
        self.color = color
    def callback( self ):
        self.obj.DoIt( self.color )

  def Register2( self, color ):
    """ Wrap the args in a class, and have a class method with the appropriate
        signature do the calling.  """
    self.doit_cb = CallbackHookFull.ColorCallback( self, color ).callback

  #---------------------------------------------------------------------
    
  def Register3a( self, color ):
    """ Generate a closure which is a wrapper around the invocation w/ args """
    self.doit_cb = lambda callback=self.DoIt, color=color: \
      callback(color)

  #---------------------------------------------------------------------
    
  def Register4a( self, color ):
    self.doit_cb = lambda callback=self.DoIt, color=color: \
      callback(color)

  #---------------------------------------------------------------------
    
  def Register5a( self, color ):
    """ Build a string with the invocation we want, and then build a closure
        that will interpret that string when invoked. """
    self.doit_cb = lambda self=self, command="self.DoIt('"+color+"')": \
      self.ExecCommand(command)

  #---------------------------------------------------------------------

  def BuildCallback3b( self, callback, color ):
    return lambda callback=callback, color=color: callback(color)

  def Register3b(self,color):
    """ Nicer-looking 3a with lambda generated by a function """
    self.doit_cb = self.BuildCallback3b( self.DoIt, color );
  
  #---------------------------------------------------------------------

  def BuildCallback4b( self, callback, color ):
    def _( callback=callback, color=color ): callback(color)
    return _

  def Register4b(self,color):
    """ Same as the 3b, but use def instead of lambda; exactly the same
        because functions are just named lambdas. """
    self.doit_cb = self.BuildCallback3b( self.DoIt, color );
  
  #---------------------------------------------------------------------

  def BuildCallback5b( self, callback, color ):
    return lambda self=self, command=callback+"('"+color+"')" : \
      self.ExecCommand(command)
      
  def Register5b(self,color):
    """ Nicer-looking 5a with lambda generated by a function """
    self.doit_cb = self.BuildCallback5b( "self.DoIt", color )

#===============================================================================

class CallbackHookPartial:
  """
  Demonstrates various ways to do partial resolution of callback arguments
  at registration time (i.e. for cases where the callback takes argument(s) at
  invocation time as well as registration time), and return/invoke the callback.
  """

  def ExecCommand(self,command):
    exec command

  def DoIt(self, color, what_to_paint):
    """ The callback we eventually want to end up calling, but with one arg
        (what_to_paint) instead of two. """
    print what_to_paint + " = " + color

  def Invoke(self):
    self.doit_cb( "walls" )

  #---------------------------------------------------------------------
    
  class ColorCallback:
    def __init__( self, callback, color ):
        self.cb    = callback
        self.color = color
    def callback( self, what_to_paint ):
        self.cb( self.color, what_to_paint )

  def Register2( self, color ):
    """ Wrap the args in a class, and have a class method with the appropriate
        signature be the callback.  """
    self.doit_cb = CallbackHookPartial.ColorCallback(self.DoIt, color).callback


  #---------------------------------------------------------------------
    
  def Register3a( self, color ):
    """ Generate a closure which is a wrapper around the invocation w/ args """
    self.doit_cb = lambda what_to_paint, callback=self.DoIt, color=color: \
      callback(color, what_to_paint)

  #---------------------------------------------------------------------

  def BuildCallback3b( self, callback, color ):
    return lambda what_to_paint, callback=self.DoIt, color=color: \
      callback(color, what_to_paint)

  def Register3b(self,color):
    """ Nicer-looking 3a with lambda generated by a function """
    self.doit_cb = self.BuildCallback3b( self.DoIt, color );
  
  #---------------------------------------------------------------------

  def BuildCallback4b( self, callback, color ):
    def _( what_to_paint, callback=self.DoIt, color=color ): \
      callback(color, what_to_paint)
    return _

  def Register4b(self,color):
    """ Same as the 3b, but use def instead of lambda; exactly the same
        because functions are just named lambdas. """
    self.doit_cb = self.BuildCallback3b( self.DoIt, color );


if __name__ == "__main__":
  cb = CallbackHookFull()
  cb.Register1('orange')
  cb.Invoke()
  cb.Register2('white')
  cb.Invoke()
  cb.Register3a('blue')
  cb.Invoke()
  cb.Register5a('red')
  cb.Invoke()
  cb.Register3b('green')
  cb.Invoke()
  cb.Register4b('purple')
  cb.Invoke()
  cb.Register5b('yellow')
  cb.Invoke()

  print "--------"

  cb = CallbackHookPartial()
  cb.Register2('white')
  cb.Invoke()
  cb.Register3a('blue')
  cb.Invoke()
  cb.Register3b('green')
  cb.Invoke()
  cb.Register4b('purple')
  cb.Invoke()

From dhellmann at zcentral.com  Thu Sep 23 21:06:55 1999
From: dhellmann at zcentral.com (Doug Hellmann)
Date: Fri, 24 Sep 99 01:06:55 GMT
Subject: pybox utility library
Message-ID: 

pybox

This library consists of lots of Python utility classes and scripts
including a base class for creating command line applications with
formatted help output and argument handling, a user preference file
management class, and a class for accessing the win32 registry with a
higher level of abstraction than the win32api module provides.

See http://members.home.net/doughellmann/pybox/index.html for more
details.

(Previous users of the pybox library please notice that the web site has
moved.)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

pybox
utility library - Python utility classes and scripts; such as a base
class for creating command line applications with formatted help output
and argument handling, user preferences management, and a high-level
Win32 registry access class.  (23-Aug-99)

-- 
----------- comp.lang.python.announce (moderated) ----------
Article Submission Address:  python-announce at python.org
Python Language Home Page:   http://www.python.org/
Python Quick Help Index:     http://www.python.org/Help.html
------------------------------------------------------------



From billtut at microsoft.com  Fri Sep 17 01:42:30 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Fri, 17 Sep 1999 05:42:30 GMT
Subject: CygWin version
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB91215@RED-MSG-50>

If your on 95 running doskey will give you command history.
NT has the equivalent of doskey always on.

Bill

> -----Original Message-----
> From: Hirsch, John [mailto:JohnH at PHM.GOV.AU]
> Sent: Thursday, September 16, 1999 6:07 AM
> To: python-list at cwi.nl
> Subject: CygWin version
> Importance: High
> 
> 
> Hi all
> In the Python tutorial it says "The interactive editing and history
> described here are optionally available in the Unix and 
> CygWin versions of
> the interpreter."
> Is there a CygWin version of the interpreter and if so where 
> can I get it.
> If not is there another way to get interactive editing and 
> history in the
> command line interpreter.
> Thanks in advance
> 
> Jonathan Hirsch
> johnh at phm.gov.au
> 
> 
> 



From herzog at online.de  Sat Sep  4 05:01:19 1999
From: herzog at online.de (Bernhard Herzog)
Date: 04 Sep 1999 11:01:19 +0200
Subject: Returning none
References: <1275758005-72447341@hypernet.com> <5ORz3.14$JG1.2704@news.shore.net> 
Message-ID: 

clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves) writes:

> def new_apply( f, a ):
>     try:
>         result = old_apply( f, a )
>     except NoResultError:
>         return
>     else:
>         return result

There's one problem with this approach. If the NoResultError is raised
whithin f, new_apply should reraise the exception and not just silently
return nothing.

You could, of course, examine the traceback object to find out where the
exception was raised, but that's very ugly and probably implementation
dependent.

-- 
Bernhard Herzog	  | Sketch, a drawing program for Unix
herzog at online.de  | http://www.online.de/home/sketch/



From davidopp at megsinet.net  Wed Sep 29 12:19:48 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 29 Sep 1999 12:19:48 -0400
Subject: More Python books coming...
References: <5lpuz483p4.fsf@eric.cnri.reston.va.us>
Message-ID: <37F23C23.1DD36623@megsinet.net>

I found a free e-book that has recently been published on paper and sells
for $39.95.  I will e-mail this book (about a 1 meg e-mail) to anyone that
writes me and requests it.

Sincerely,

David Oppenheimer

Guido van Rossum wrote:

> I did a search for Python on Amazon.com and found several new Python
> books that are likely to hit the shelves soon.  Here's the list:
>
> The Quick Python Book
>         Daryl Harms, Kenneth McDonald. Published by Manning
>         Publications. Expected September 1999.
>
> Python Essential Reference
>         David Beazley. With a foreword by Guido van Rossum. Published
>         by New Riders. Expected October 1999.
>
> Programming with Python
>         Tim Altom. Published by Prima Publishing. Expected October 20,
>         1999.
>
> Python and Tkinter Programming
>         John E. Grayson. Published by Manning Publications. Expected
>         November 1999.
>
> Python Annotated Archives
>         Martin Brown. Published by McGraw-Hill. Expected September 1999.
>
> Python Programming on Win 32
>         Mark Hammond, Andy Robinson. Published by O'Reilly. Expected
>         September 1999.
>
> Sams Teach Yourself Python in 24 Hours
>         Ivan Van Laningham. Published by Sams. Expected November 1999.
>
> We expect to have a bookstand at the Python conference in January
> where you can buy these and other Python books.  I've also added this
> to http://www.python.org/doc/Books.html.
>
> --Guido van Rossum (home page: http://www.python.org/~guido/)




From Doug at news.vex.net  Sun Sep 12 13:03:28 1999
From: Doug at news.vex.net (Lewis)
Date: 12 Sep 1999 17:03:28 GMT
Subject: ANN: Tk Slashdot Headline Viewer
Message-ID: <7rgmd0$2htg$1@hub.org>

Inspired by the Slashdot headline viewer demo in the wxPython
package, i decided to implement my own in Tkinter.

My implementation, while simpler, is much cuter. (-:

Should anyone care to try it out, you can snag it from....

http://www.vex.net/~x/python_stuff.html#slashdot

...





From mkrisch at radiata.com  Thu Sep 23 22:44:09 1999
From: mkrisch at radiata.com (Mark Krischer)
Date: Fri, 24 Sep 1999 12:44:09 +1000
Subject: Becoming root within a Python script
References:  <87zoydrw4u.fsf@baal.winnegan.fake>
Message-ID: <37EAE579.2499F85C@radiata.com>

why don't you call the python script using sudo, or make a wrapper
script that calls sudo .py

(sudo lets a user run an executable as root--i know it exists on linux,
i assume it can be compiled for other unices....)

--mk

Siggy Brentrup wrote:
> 
> Fran?ois Pinard  writes:
> 
> > Hi, people.  I have a small problem, for which I see no elegant solution yet.
> > Is there a common idiom I could use?
> >
> > I would like, within a executing Python started as non-root, ask for the root
> > password, and then, continue executing part of the script while being root,
> > becoming the previous again after some sub-job is done.
> >
> > Of course, I can kludge something in, but I would prefer something nice. :-)
> 
> What you are asking for requires suid root scripts, a really bad thing
> modern Unices don't allow any more.
> 
> You have to hack up a wrapper written a compiled language, which
> starts the interpreter with root priveledge.
> 
> A while ago I did some work on such a beast using an md5 registry for
> trusted scripts, but after rethinking I threw it away - just to big a
> security risk.
> 
> CU
>   Siggy
> 
> --
> noch nichts Aufregendes:
> 
> Siggy Brentrup - bsb at baal.infodrom.north.de - voice: +49-441-6990134



From davidopp at megsinet.net  Tue Sep  7 20:23:59 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Tue, 07 Sep 1999 20:23:59 -0400
Subject: 1.5.2 NT install question(s)
References: <7r3hib$24f$1@news.wrc.xerox.com>
Message-ID: <37D5AC9E.44A69F13@megsinet.net>

I would strongly recommend installing Python in a pathway without spaces, as
recommended by Robert Kern.  Additionally the following may help:

It can't find the module because it's not on the PYTHONPATH.  Execute this
python code to see what directories are in your PYTHONPATH:

At a python command line:

>>> import sys
>>> print sys.path

To fix it add this to your autoexec.bat:

set PYTHONPATH=c:\Tcl\bin

(This is after you go back and install Tcl to the pathway C:\)

or add the directory to the value in the Registry at

HKEY_LOCAL_MACHINE\SOFTWARE\Python\PythonCore\1.5\PythonPath

or as a subkey.

Hope that helps,

David Oppenheimer

Mark Jackson wrote:

> I've done this twice recently and encountered problems both times.
>
> On a fresh system (never had Python or Tcl/Tk) things (for example,
> PySol) seem to work, but whenever I use Tkinter (specifically, when
> _tkinter is imported) I get a warning box about not being able to find
> tcl80.dll in the specified path.  (tcl80.dll is present, in Program
> Files\Tcl\bin.) What's wrong, and how do I put it right?
>
> On a system where 1.5.1 (with Tcl/Tk) had previously been installed, I
> carefully uninstalled both before installing 1.5.2.  Here both "import
> _tkinter" and "inport Tkinter" are OK, but Tkinter._test() yields:
>
>     TclError: Can't find a usable init.tcl in the following directories:
>         {} ./lib/tcl8.0 D:/tc18.0/library {D:/Program Files/library}
>
>     This probably means that Tcl wasn't installed properly.
>
> (init.tcl is present, in Program Files\Tcl\lib\tcl8.0.)  What's wrong,
> and how do I put it right?
>
> Thanks in advance.  Sorry to be asking such trivial-seeming questions,
> but I found nothing helpful through python.org.  And I haven't dealt
> with the innards of NT enough for its design coherence to become
> manifest (yeah, right).
>
> --
> Mark Jackson - http://www.alumni.caltech.edu/~mjackson
>         Spend the optimal amount of time on each decision
>         and pretty soon you run out of life.
>                                 - Steven Waldman




From madq968 at djeksta.comNOSPAM  Thu Sep 30 16:53:44 1999
From: madq968 at djeksta.comNOSPAM (madQ)
Date: 30 Sep 1999 20:53:44 GMT
Subject: Download Ia.n.i.!!! It's free!
Message-ID: <7t0iko$kdr$17@nslave1.tin.it>

Download Ia.n.i. RemoteControlSystem 1.2 beta. It's free!!!
New site: http://jump.to/IaniProject




From zessin at my-deja.com  Sat Sep  4 05:14:35 1999
From: zessin at my-deja.com (Uwe Zessin)
Date: Sat, 04 Sep 1999 09:14:35 GMT
Subject: PYTHON on ALPHA VMS 7.2?
References: <37D005D9.38C27D05@msx.upmc.edu>
Message-ID: <7qqnto$u71$1@nnrp1.deja.com>

In article <37D005D9.38C27D05 at msx.upmc.edu>,
  Jack Lewis  wrote:
> Has anyone successfully built PYTHON on ALPHA VMS 7.2 ?
>
> As well as been able to compile and link 'C' programs with Embedded
> PYTHON calls?
> I can't seem to get past the link errors:
>
> %LINK-W-NUDFSYMS, 2 undefined symbols:
> %LINK-I-UDFSYM,         CTL$GL_IMGHDRBF
> %LINK-I-UDFSYM,         _PYIMPORT_INITTAB
> %LINK-W-USEUNDEF, undefined symbol _PYIMPORT_INITTAB referenced
>         in psect PYIMPORT_INITTAB offset %X00000000
>         in module IMPORT file
> DSA900:[PYTHON.PYTHON-1_5_2.VMS.O_ALPHA]PYTHON.OLB;10
> %LINK-W-USEUNDEF, undefined symbol CTL$GL_IMGHDRBF referenced
>         in psect $LINK$ offset %X00000010
>         in module VMS__GETBUILDINFO file
> DSA900:[PYTHON.PYTHON-1_5_2.VMS.O_ALPHA]VMS.OLB;10

I didn't got to test BUILD_DEMO.COM whether it works on OpenVMS Alpha.


Please take the original file, modify it as show below and try again:

    '_OLBDIR'PYTHON.OLB /library, - ! again for strdup() lookup
    PYTHON_VMS:CONFIG.OPT /options - ! created by CONFIG_INITTAB2MAR.COM
  'SYSEXE_LINK'
$!

(This change will make the procedure stop working on VAX - consider
it as a quick hack!)

CTL$GL_IMGHDRBF is resolved by using /SYSEXE and _PYIMPORT_INITTAB is
resolved by CONFIG_INITTAB.OBJ - the General Manual for Python on
OpenVMS explains the last part.

                             ---

I will be out of town for the next 2 weeks - if it does not work,
free feel to contact me after 20-SEP-1999.

--
Uwe Zessin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From warren at nightwares.com  Fri Sep 10 23:17:19 1999
From: warren at nightwares.com (Warren 'The Howdy Man' Ockrassa)
Date: Fri, 10 Sep 1999 22:17:19 -0500
Subject: high school programming & python
References: 
Message-ID: <37D9C9BF.A5184784@nightwares.com>

Les Schaffer wrote:

> I am tutoring a high school kid in programming this semester -- he is
> doing an independent study.

Heh, I've written instructional software for grade/middle schools, and
have a friend whose son is doing Pascal right now in Grade 9 (15 yrs.
old for non-US system reference).

> they \quote use \unquote C++, having recently changed from pascal. but
> they dont teach any class/object stuff. when i asked her how far they
> get in the class, she said "up through arrays".

That makes no sense. Might as well teach the monkeys assembler. To teach
C++ and ignore OOP is like teaching English grammar and ignoring
literary classics. Worse.

Python is a *superb* interface language for both coding in general and
OOP as well; that is how Guido designed it. (In fact I have direct
experience with languages written by corp folks deriving lots of profit,
and have seen nothing quite so elegant nor addressable as Python; Guido
van Rossum is, to my eyes, a genius on par with Linus Torvalds.)

Python can begin as a simple CLI code emulating "cal" on non-*nix
systems and work up to massive GUI applications, such as, gamewise,
PySol or, webwise, grail. (Many, many other completed apps in between,
as well as more modules than I think *anyone* can enumerate to describe
its full functionality, including full 3-D rendering engines.)

This is beyond doubt simultaneously the most accessible OS-addressing
language I have ever encountered, the most GUI-capable, the most
coder-friendly, and the most powerful/flexible/extensible. When you
count its release platforms as well... well.

I have only rippled its surface, mind you. But I know my intiuition. I
think this is the best language going *anywhere*.

> they dont do any graphical/GUI coding at all.

That is a massive error. Console window apps are useful, and knowing how
to do them is good, but such "instruction" has no place in 1999 soon to
be 2000. The result is a class of kids disenfranchised with coding
because they can't even do what you can do with HTML, for God's sake.

> 1.) very few kids in the high school are interested in taking a
> programming class.

I think the fault there is in presentation. Instead of offering a
"programming" class, how about a "Web class"? How can you do this great
Web stuff? Take this class and find out!

> 2.) the AP tests which kids take, according to her, require for
> advnace placement in a comp sci class that they know the basics of C++
> up through objects. the question then is whether something like
> learning/using python would enable a student still to place in the AP
> tests if they desired.

Oh god. The process and the mechanism, once it is understood, maps very
well. With knowledge of OOP -- regardless of its derivation -- all OOP
langs are accessible, provided you have a lang ref to thumb through, and
a reasonable amount of time. (Of course C and C++ hurt, badly and
desperately, to try to understand.)

I can tell you this. We would never hire someone with pure academic
knowledge of one language. Our employability base would have to include
portability of knowledge -- a generalization of technique lost in the
intricacies of C++. (In fact strict C++ knowledge would probably be a
deficit.)

Javaesque and CGI coders we want. C++ academics are too specialized to
suit our needs.

I don't think I'm the only one to feel that way, and I know for certain
that I am not the only potential recommender of hirees in the US
corporate world to believe this.

Note I am a complete newbie to Python, but I am not new to coding in
general, and so while I know little now of Python techniques I can
assure you professionally that this is a language which will be used in
US$100,000 bids for clients.

It is faster than C++ can ever be at prototyping and it is possible for
a .py prototype to be so successful that no C++ engine needs to be
built.

My young friend's teacher is being gradually seduced to this language,
and for good reason. After all four lines to a GUI hello world is vastly
superior to what it takes in any other HLL... or VHLL...

Your HS teacher is badly out of date, methinks.

FWIW when I took a HS "programming" course in 1984 -- 1984! -- it was
using Commodore machines with tape loads, pure BASIC as in 10 print
"hello" 20 goto 10, and our teacher referred to ASCII as
a-ess-cee-double-i; he also had a cache of slide rules. No kidding. We
wrote our code on quadrille paper for the line spacing; damn near
Fortran in most ways.

Point is that some HS programming teachers, especially if they are
older, might not be in touch with current tech. This is not, after all,
porting, polishing and specing a 351 hemi engine as in auto shop, where
tech proceeds slowly. This is 2000 and beyond, and some languages are
just not viable any more, just as some perspectives are not.

For what it is worth, professionally, I would not hire this teacher of
which you speak to code any kind of program I ever needed for a
multi-kilobuck client. I would only hire students s/he had that didn't
fit the mold.

--
  Warren Ockrassa | Becker Communications | http://www.beckerinc.com/



From framiere at yahoo.fr  Tue Sep 28 11:11:26 1999
From: framiere at yahoo.fr (Florent Rami�re)
Date: Tue, 28 Sep 1999 17:11:26 +0200
Subject: Non blocking print
Message-ID: <7sqlu2$mt8$1@oceanite.cybercable.fr>

Hello,

    I would like to know how to use a non bloking IO, like print.
    In perl I would have used $|=0, what should I use in python ?

Thanks for you answers...
Florent Rami?re





From Andreas.Cardeneo at mach.uni-karlsruhe.de  Tue Sep 14 11:24:39 1999
From: Andreas.Cardeneo at mach.uni-karlsruhe.de (Andreas Cardeneo)
Date: Tue, 14 Sep 1999 17:24:39 +0200
Subject: python call external program
References: <37DE6156.FD4384C6@trey-industries.com>
Message-ID: <37DE68B7.49D0C5A5@mach.uni-karlsruhe.de>


Ben Thomas wrote:
> 
> very simple question:
> 
> What would I do to have Python execute something from the DOS command
> line. I am very new to python but I bet it is in a moudule but don't
> know the command or the modules to look in. I want to
> 
> dir >%myvar.txt
> 
> #this code would run dir from the command line and pass the var "myvar"
> , dir would dump a listing of the files into a text file called xxxx.txt
> 
> pointers?
> 
> Ben

Hi Ben,

you should use the system() function from the "os" module.

os.system("dir *.txt") shows all files in the current directory matchin
*.txt.
The function returns the exit code.

Bye,
Andreas

-- 
Andreas Cardeneo 		Universit?t Karlsruhe (TH)
Tel: +49 721 608-4576		Institut f?r F?rdertechnik und Logistiksysteme
Fax: +49 721 75 83 78		Hertzstra?e 16 - 76187 Karlsruhe 
eMail: Andreas.Cardeneo at mach.uni-karlsruhe.de



From guido at CNRI.Reston.VA.US  Wed Sep 22 19:36:10 1999
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Wed, 22 Sep 1999 19:36:10 -0400
Subject: Standard Python documentation in book form!
Message-ID: <199909222336.TAA16411@eric.cnri.reston.va.us>

I am pleased to announce that toExcel.com is selling nicely printed
and bound copies of the key Python 1.5.2 documentation: library
reference, language reference, and tutorial.

The books are printed on demand.  I have received review copies and
the print quality is very good.  Especially the library reference (496
pages) is well worth your money if you want hardcopy!

Royalties go to the PSA.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From mlh at idt.ntnu.no  Fri Sep  3 09:01:32 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 03 Sep 1999 15:01:32 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <19990819174800.D5227@xs4all.nl> <37bc313b@cs.colorado.edu> 
Message-ID: 

"Rich Salz"  writes:

> >No, I'm sorry, the expressivity of "break LABEL" and "continue LABEL"
> >are much clearer than the cicumlocutions necessary for exceptions.
> 
> You sure?  I wasn't aware that I could break and return a value, such as
> the name of the file that wasn't found.  At least not without a lot of
> circuml^H^H^H^H^H global variables.
> 

Acutally - in the line of considering labelled breaks expressive...
The most expressive construct in this venue would be GOTO, no? 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From skip at mojam.com  Thu Sep 23 14:19:52 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 23 Sep 1999 13:19:52 -0500 (CDT)
Subject: The DirtyExeT proposal
In-Reply-To: <7sdof1$414@mail.psy.uva.nl>
References: <7sdof1$414@mail.psy.uva.nl>
Message-ID: <14314.28488.503761.107726@dolphin.mojam.com>

    Ionel> Objective:
    Ionel> ----------
    Ionel> Package a Python application that (comprises | depends upon)
    Ionel> several Python source files into a single EXE file. This process
    Ionel> should be automatic, customizable and largely independent of the
    Ionel> concrete Python application at hand.

    Ionel> Comments:
    Ionel> ---------
    Ionel> 1.  I hope the wizards of the Python community will find an easy
    Ionel> way to implement this idea and give us the tool to create bulky
    Ionel> standalone applications with a click of the mouse.

Ionel,

Check out any of a variety of freeze/xfreeze/squeeze references in the
archives of the list.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098   | Python: Programming the way Guido indented...



From ionel at psy.uva.nl  Thu Sep 23 13:34:45 1999
From: ionel at psy.uva.nl (Ionel Simionescu)
Date: Thu, 23 Sep 1999 19:34:45 +0200
Subject: The DirtyExeT proposal
Message-ID: <7sdof1$414@mail.psy.uva.nl>

DirtyExeT is a proposal I hereby submit to your attention.


Objective:
----------
Package a Python application that (comprises | depends upon) several Python
source files into a single EXE file. This process should be automatic,
customizable and largely independent of the concrete Python application at
hand.


Context:
--------
One has a working Python application.

He wants to demonstrate it to a potential client, or in some other context.
In order to do that he should make sure that Python is installed on the
client's machine, as well as many of the libraries he use, and he should
also check the environment settings.

Because it is rarely the case that everything is setup on the target
machine, one must often undertake the effort of packing up his working
environment, dress it up with smart install scripts and hope it will work
when unfolded elsewhere.

An automated and elegant solution is needed. Ideally, a certain command
should be available in Python that transforms an application into a
selfcontained, standalone exe file. Size is not important in many
situations, but functionality and time to market/client.


Solution:
---------
A Python application, APP, is basically defined by its main module, app.py.

The task of identifying the dependencies of APP, can be automated by
recursively looking at the import statements.

APP is thus constituted by a set of named files; more specifically, we shall
consider the set of p-code files.

-------------------------------------------------------------------
The basic idea of the DirtyExe method is to (attach|append) these files to
the body of a special Python.exe interpreter, and then fix the result to
remain a valid exe.
-------------------------------------------------------------------

This approach avoids a true compilation step, is fast, can be quite
effective and it has been already proved in practice [1].

I will hereby suggest what modifications to the Python interpreter are
necessary and how it might work.

On Win32, an EXE may contain resources and there is a mechanism to use them.
In order to allow the interpreter to accept and use p-code patches one needs
to augment its implementation for the  routine.

The modified import will look for the files not just on the disk but firstly
in its own table of resources.

Moreover, the interpreter should always start by executing a p-code resource
string called, for example, "PYTHONSTARTUP".

By now, the EXE is already dirty as it contains not only host specific
object code but also Python specific object code (the content of the
PYTHONSTARTUP string).

Such an EXE may be changed to contain arbitrary Python applications in its
resources table. It is left to the patching phase to actually manipulate
these resources and thus create custom applications.

---

On other platforms than Win32, the p-code files pending to an apllication
could, in principle, be appended to PYTHON.EXE as new global variables.
Being able to actually use this code requires that the EXE uses a global
PYTHONSTARTUP string containing p-code, and this string can be replaced by
the code of the new, custom application.

When there is no distinction between object code and some form of resources,
it might be a bit more difficult to handle the symbol table of the exe in
order to add new global variables. Yet, it should be doable because it's
just appending new records and shifting around some addresses for some of
the existing ones.

(Python.exe is thus turned into a willing host for python applications. :-)

---

Conclusion:
-----------
A method is suggested by which an EXE file that contains a Python
interpreter can be augmented with pseudo code files representing a custom
application.

The net effect is that developers will be able to get standalone
executables.



Comments:
---------
1.
I hope the wizards of the Python community will find an easy way to
implement this idea and give us the tool to create bulky standalone
applications with a click of the mouse.
2.
This method really fits Python: just think about python(.exe) eating up your
p-code, :-)
3.
For the sophisticated, several kinds of python.exe builds may be designed,
as some libraries are included or not.
4.
For the casual user, and, attention, for the students preparing their term
projects, such a tool would be quite a compelling reason to use Python,
because it hides its actual source code. For example, they might claim to
have coded it in C++... They should not, of course, do such things. :-)
5.
For the professionals, they would have an easier time with clients locked in
their respect for the "compiled" languages. The clients will not be able to
tell a Pythonish composit from a "genuine" exe.
6.
There is some other work to be done that I did not even mention: checking
for the distinction btw. console and windowed applications, accompanying the
resulting exe with binary files that might be required, etc.

...


ionel



----------------------------------------------------------------------------
-----

[1] For a supposed implementation of the DirtyExe idea, see the enviroment
"Dialect Standard Edition", at http://www.aristar.com/DeveloperTools.html.

The DirtyExe method is nowhere explicitly pointed to, but if you made an exe
with their tool you'll see that there is no other way to do it.

I was annoyed of that little bastard, Dialect, not mentioning its parents,
i.e. Python, but I was impressed of its ease in creating EXEs. This posting
is the resolution of this conflicting thoughts.









From yopen at my-Deja.com  Sat Sep  4 15:10:53 1999
From: yopen at my-Deja.com (Bill Wilkinson)
Date: Sat, 4 Sep 1999 14:10:53 -0500
Subject: New to Python...looking for scripting language
References: <37D12E9D.EDB193B8@patriot.net>  <37D168BB.D4C90858@patriot.net>
Message-ID: <#A8jdjw9#GA.278@cpmsnbbsa02>

sure.....

I can put some on my site as examples.  I don't know if you would find the
ones I have very useful.  But, I will post the ones that I can when I go
back to the office tuesday.

Harlan Carvey, CISSP  wrote in message
news:37D168BB.D4C90858 at patriot.net...
> Bill,
>
> Thanks for your input.  Do you have any examples of your admin scripts
> available?
>
> Carv
>
> Bill Wilkinson wrote:
>
> > Python is one of the languages that works with the Windows Scripting
Host.
> > It exposes several objects that are quite handy for doing the things you
> > want.  There is a File System Object and such like.  WSH will be built
in to
> > Win2000, but for now it is a seperate download....  (I am not sure, but
it
> > might be included with the Pythonwin files)
> >
> > Also, using Pythonwin you can do some of the things you mention easily
> > without the WSH.
> > In my opinion, Python is ideal for NT scripting.  I use it for all my
admin
> > scripts.
> >
> > Bill Wilkinson
> > Harlan Carvey, CISSP  wrote in message
> > news:37D12E9D.EDB193B8 at patriot.net...
> > > Hey all,
> > >
> > > I am looking for a scripting language for NT environments.  I am NOT
> > > trying
> > > to start a language war.
> > >
> > > I have been to the Python.org site quite a bit, and haven't found what
I
> > > am looking
> > > for, so I will post here.
> > >
> > > With Python, can I:
> > > - access the Registry (get, and set values, create keys, etc)?
> > > - access the services (get a list of running services)?
> > > - access the audit policy (get and set that policy)?
> > > - search for files on the remote hard drive?
> > > - access permissions on the remote system?
> > >
> > > I haven't seen modules that allow this...nor have I seen example
> > > scripts.  I am not
> > > a MS/C++ programmer, and I want to be able to leave scripts for
> > > sysadmins who
> > > aren't either.  They need to be able to maintain the scripts.
> > >
> > > Thanks,
> > >
> > > H. Carvey
> > >
>





From rvollmert at gmx.net  Fri Sep 24 10:51:14 1999
From: rvollmert at gmx.net (Robert Vollmert)
Date: Fri, 24 Sep 1999 16:51:14 +0200
Subject: Becoming root within a Python script
In-Reply-To: ; from Will Ware on Fri, Sep 24, 1999 at 04:14:15AM +0000
References:  
Message-ID: <19990924165114.A753@krikkit.vollmert.home>

On Fri, Sep 24, 1999 at 04:14:15AM +0000, Will Ware wrote:

[snip]
> if os.popen('whoami').readline()[:-1] == 'root':
[snip]

> Dumb question for those wiser than myself:
> The os.popen().readline()[:-1] construction is pretty ugly.
> Is there a better way to do it?

How about 'if os.getuid() == 0:' ?

-- 
Robert Vollmert                                      rvollmert at gmx.net



From guido at cnri.reston.va.us  Thu Sep  2 08:28:53 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 02 Sep 1999 08:28:53 -0400
Subject: tkinter & threads
References: 
Message-ID: <5lemgh5wfu.fsf@eric.cnri.reston.va.us>

Robin Becker  writes:

> Is Tkinter thread safe? I looked at Guido's brownian demo and there are
> no special precautions about locking the canvas. Also I believe there is
> some restriction on which thread Tk can live in.

This is a 1.5.2 feature; I added enough locks to _tkinter to make it
thread-safe.  However on Windows there are still problems (you can
only call update() and friends from the main thread) and I think I've
seen one report from non-Windows where things broke under certain
circumstances.  The Brownian demo is there specifically to show that
Tkinter is now thread-safe.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From cors at recercai.com  Tue Sep  7 16:39:45 1999
From: cors at recercai.com (Carlos Ors)
Date: Tue, 7 Sep 1999 22:39:45 +0200
Subject: Problem with python embedded
Message-ID: <37d523e1.0@sahara.cablecat.com>

I try running the SimpleEmbedded.java but I get next error:

     'import exceptions' failed; using string-based exceptions

How can I solve this problem?

                                               Thanks in advance.


Please (if you can) answer me to karlos at softhome.net





From ivanlan at callware.com  Fri Sep 17 11:30:49 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 17 Sep 1999 15:30:49 GMT
Subject: Splitting c.l.py (was Re: python call external program (WHAT THE?!))
References: 
		<7rr088$74b@dfw-ixnews8.ix.netcom.com> <7rs3oo$5jt$1@news.udel.edu>
Message-ID: <37E25EA9.A0BE7BF6@callware.com>

Hi All--

Terry Reedy wrote:
> 
> Rather than splitting the group, what might be helpful are agreed-upon
> 'designators' such as WIN, W95, WNT, UNX, TK consistently placed at the
> head of subject lines when appropriate.
> 

Splitting the group sucks.  The community is not a paramecium; there's
no point to fission without syzygy.

It might be helpful with the WIN, W95 etc. markers, but I belong to
another mailing list in which such marking is enforced, and what it does
as far as I can tell is restrict what you think you can say to a set of
pre-ordained categories.  Then somebody comes in with something that
doesn't fit in a box, so they dump the category markers, and get flamed.

So I guess I'm against that too.

-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From MSteed at altiris.com  Fri Sep 10 12:05:08 1999
From: MSteed at altiris.com (Mike Steed)
Date: Fri, 10 Sep 1999 10:05:08 -0600
Subject: Text UI
Message-ID: <65118AEEFF5AD3118E8300508B124877073BEE@webmail.altiris.com>

One possibility worth investigating is wxWindows, which includes a curses
port.  I have not used the curses port, nor I do not know whether the Python
interface will work with it....

wxWindows: http://www.wxwindows.org/
wxPython: http://alldunn.com/wxPython/

-Mike

> -----Original Message-----
> From: smithy at promail.pl [mailto:smithy at promail.pl]
> Sent: Friday, September 10, 1999 9:54 AM
> To: python-list at python.org
> Subject: Text UI
> 
> 
> As in subject - I'm seeking a console-based user interface toolkit for
> Python. All I found by now is snack.py module, an interface to newt.
> Unfortunately, this is a very simple toolkit and almost the 
> same functionality
> is achievable with dialog or whiptail in bash scripts.
> 
> What I'm looking for is an event-driven user interface 
> library. Something
> resembling the Borland's TurboVision from DOS (I need this 
> under Linux).
> 
> Has anyone written something like this? A search on 
> python.org, yahoo and
> a few other search engines gave no results.
> 
> -- 
> ?ukasz Kowalczyk
> lukowtempac.fuw.edu.pl              
> http://tempac.fuw.edu.pl/~lukow/



From hirsch at noreaster.cit.cornell.edu  Tue Sep 21 13:11:45 1999
From: hirsch at noreaster.cit.cornell.edu (Matthew Hirsch)
Date: Tue, 21 Sep 1999 13:11:45 -0400
Subject: Thanks
References: <37E7809E.A871480C@noreaster.cit.cornell.edu>
Message-ID: <37E7BC4F.C5DE4F1D@noreaster.cit.cornell.edu>

Thanks guys.  I really appreciate your help.

Matt




From m.faassen at vet.uu.nl  Thu Sep  2 06:21:31 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Sep 1999 10:21:31 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com> <37CCE10D.D07D023B@strs.co.uk> <7qjgq8$k1i$1@newshost.accu.uu.nl> 
Message-ID: <7qlj3b$p7u$1@newshost.accu.uu.nl>

William Tanksley  wrote:
[snip]
> Java's not interpreted, though -- I've never seen ANY Java interpreter.
> Java is always compiled, usually to bytecode, and then the bytecode
> machine is emulated.

Um, I was under the impression that this is what a modern interpreter
actually does? Parse the source to bytecode and execute the bytecode.
After all, Python does the same, right?

Regards,

Martijn




From celkaim at microform.fr  Mon Sep 27 11:36:15 1999
From: celkaim at microform.fr (Cyril Elkaim)
Date: Mon, 27 Sep 1999 17:36:15 +0200
Subject: Locking some methods from certain clients
Message-ID: <37EF8EEF.F8AF3D31@microform.fr>

	Hi,

	I've the following question:

	I want to lock the accesse of some method of an object from other
objects using it. But on a class by class basis. What I want is the
possibility to know the class-name of a caller and if this caller is on
my access list let it use my method. Something like this:

class callee:
	def method(self):
		if __caller__ == "agreed":
			execute...
		else:
			raise callError, "caller no authorized to execute"

this class will be able to call the method
class aggreed:
	def run(self):
		x = callee()
		x.method()

this class not
class other:
	def ren(self):
	x = callee()
	c.method()

It's something like the export clause in Eiffel.

	Thanks in advance
	Cyril Elkaim



From steinme at norway.eu.net  Tue Sep 21 15:58:03 1999
From: steinme at norway.eu.net (Stein M. Eliassen)
Date: Tue, 21 Sep 1999 19:58:03 GMT
Subject: make test failed in time_test
References: <37E7DB67.5DE00C61@norway.eu.net> <00b601bf0469$0e96e330$f29b12c2@secret.pythonware.com>
Message-ID: <37E7E34B.CABD4689@norway.eu.net>

Fredrik Lundh wrote:
> hope this helps!

It's working now.
Thanks!



From g.gruber at xsoft.co.at  Mon Sep  6 09:17:58 1999
From: g.gruber at xsoft.co.at (Gerhard W. Gruber)
Date: Mon, 06 Sep 1999 15:17:58 +0200
Subject: Dictionaries with invalid key?
Message-ID: <37D3BF06.4A49572@xsoft.co.at>

Usually a dictionary conists of a key and a value. I thought that keys
are strings. Now I wrote some code that managed to create a key that is
a number instead of a string. Is this a bug or is this ok?
-- 
bye,
   Gerhard
email: sparhawk at eunet.at
       g.gruber at xsoft.co.at



From nessus at mit.edu  Wed Sep 29 12:59:53 1999
From: nessus at mit.edu (Douglas Alan)
Date: 29 Sep 1999 12:59:53 -0400
Subject: When Good Regular Expressions Go Bad
References: <000101bf0a47$98084da0$ce2d153f@tim>
Message-ID: 

"Tim Peters"  writes:

> > It seems to me that even when a regular expression fails to match a
> > string, you might want to know just how far it was able to get before
> > getting stuck.  (And indeed I do!)

> I confess I'm at a loss to think of a compelling use for it, though.

Here's what I'd like to use it for: I'm writing a function that does a
directory walk.  The client of the function can provide a regular
expression, indicating the pathnames on which they would like to
match.  I would like to optimize the function by pruning any
directories for which the contents could not possibly match the
regular expression.

Sure, the function could instead accept a filtering function as an
argument, instead of the regular expression, and the client could then
provide their own pruning, or I could just require that the regular
expression also match directories to be descended into.  But these
interfaces are not nearly so nice.

|>oug



From robin at alldunn.com  Sun Sep 26 18:57:56 1999
From: robin at alldunn.com (Robin Dunn)
Date: Sun, 26 Sep 1999 15:57:56 -0700
Subject: wxPython problems
References: <7sjl5n$4gc$1@sunsite.icm.edu.pl>
Message-ID: <6yxH3.1168$Ir.64618670@news.randori.com>

>     I got strange problem with wxPython on RH6.0. I've written the small
> application and each time I run it the some texts ( eg. buttons' captions
> and labels )  look  in other way. I mean that the font's size and,style
> changes each time and I have no idea why it happens.
>

I've noticed this too, (some labels are bold but should not be) but I've
only seen it when running in a VMWare virtual machine and just thought that
it was something to do with a beta X-Server.  Other GTK apps were doing it
too, not just wxPython.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!






From pinard at iro.umontreal.ca  Thu Sep  2 09:33:14 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: Thu, 2 Sep 1999 13:33:14 GMT
Subject: Python speed and `pcre'
In-Reply-To: "Fredrik Lundh"'s message of "Thu, 2 Sep 1999 14:59:49 +0200"
References:    <00af01bef543$0affe8f0$f29b12c2@secret.pythonware.com>
Message-ID: 

"Fredrik Lundh"  ?crit:

> Fran?ois Pinard  wrote:
> > I do not fully understand the cost of splitting.  If strings are immutable,
> > there is really no need to copy a slice contents.

> except that 1) each string needs an extra reference pointer (as long
> as we're using refcounting, at least), and 2) that people can easily
> end up in situations where huge strings are clogging up all available
> memory, just because some part of their program is hanging on on to a
> 2-character substring...

Exactly my say.  The inefficiency is related to the chosen implementation
of the garbage collector.  It is not "modern" to push such considerations
into the programming style of everybody, when the language syntax much
invites for something cleaner and simpler.

> (fwiw, the unicode string type implements substring support, using it
> for the "split" and "slice" methods.  still not sure if that's really
> a great idea...)

In general, simplicity is a great idea.  The language implementor gets
all the trouble! :-)

> > Somewhere, it is said that some extra-space is allocated at end of
> > sequences, to allow for some peaceful growth.  Does `str=str+str1'
> > at least attempt to use of this extra-space?  It should, in my opinion.

> how?  before you tell me, consider that strings are immutable,
> and variable names are references.

Yes, of course.  Given strings are immutable, an implementation could have:

        ref=str
        str=str+str1

and have `ref' points to a substring of `str', without copy most of
the time, when `str1' is small.  Once again, the chosen implementation
induces inefficiencies, which are then translated into stylistic issues for
all users.  This is not ideal.  One might well explain at length how the
gargabe collection occurs, but if the language is really healthy, users
should not have to fly at such down a level.  It is very interesting to
know for the general culture of the community, of course, but it should
not influence programming style.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From jlj at cfdrc.com  Thu Sep  2 10:08:41 1999
From: jlj at cfdrc.com (Lyle Johnson)
Date: Thu, 02 Sep 1999 14:08:41 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <37c2a462.349531009@192.77.87.74> <935523275snz@vision25.demon.co.uk> <7qh7jk$gaa$1@wagner.wagner.home> <936222198snz@vision25.demon.co.uk>
Message-ID: <37ce82e6.85623249@192.77.87.74>

On Wed, 01 Sep 99 21:43:18 GMT, philh at vision25.demon.co.uk (Phil Hunt)
wrote:

>In article <7qh7jk$gaa$1 at wagner.wagner.home>
>           vitus at wagner.rinet.ru "Victor Wagner" writes:
>> Phil Hunt  wrote:
>> : The Parrot tool has other potential advantages, e.g: 
>> : 1. you only have to learn one syntax, that can then be used for
>> : lots of languages and GUI libraries.
>> 
>> Which is completely useless,
>
>Fine, I'm not forcing you to use it.
>
>
>-- 
>Phil Hunt....philh at vision25.demon.co.uk
>

You know, it just warms my heart to see people criticize newly
announced projects just because that project is not useful for the
criticizer. Maybe that's the norm on some other newsgroups but we're
generally a little more cordial on comp.lang.python.

Phil, good luck with "Parrot". It sounds like it will fill a
previously unaddressed (or at least under-addressed) need to migrate
between different user interface toolkits easily, a pretty cool idea.



From thatcher at bmd.clis.com  Sun Sep 12 22:18:03 1999
From: thatcher at bmd.clis.com (Robert Thatcher)
Date: Mon, 13 Sep 1999 02:18:03 GMT
Subject: Free Wallpaper, Clip Art and Backgrounds by Robert Thatcher
Message-ID: <37dc5457.1@newport>

Free Wallpaper, Clip Art and Backgrounds by Robert Thatcher at http://www.clis.com/thatcher/index.htm



From wtanksle at dolphin.openprojects.net  Tue Sep 14 21:03:04 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 15 Sep 1999 01:03:04 GMT
Subject: high school programming & python
References:   <937300939snz@vision25.demon.co.uk> <$JL+lCAFvt33Ewr8@jessikat.demon.co.uk> <7rmmjr$qv6@dfw-ixnews11.ix.netcom.com>
Message-ID: 

On 14 Sep 1999 23:43:55 GMT, Aahz Maruch wrote:
>Robin Becker   wrote:
>> writes

>>>Machine language! You young people today don't know what hardship
>>>is... when I were a lad, we had to manually patch the ALU using
>>>plug boards to change the logic.

>>You were lucky to have plugboards. We had to key things in with a bunch
>>of switches.

>You were lucky to have switches.  We had to rewire the traces.

Traces indeed.  They didn't let us touch the traces; we had to hand-orient
a jar of magnetic filings using a hammer and the earth's magnetic field.

And then hand it over to the DP professionals, and hope they didn't decide
to shake the jar.

>                      --- Aahz (@netcom.com)

-- 
-William "Billy" Tanksley



From cdaniel at xcom.net  Wed Sep 15 02:58:57 1999
From: cdaniel at xcom.net (Cliff Daniel)
Date: Wed, 15 Sep 1999 06:58:57 GMT
Subject: Threads + Re(gex) Baffled!?!?!?
Message-ID: <37df3dfb.8684516@news.sjo.level3.net>

I'm totally baffled right now, wrt threads.  I have built a test
program using sockets to pull down just a few lines of text,
parse it and junk the result.  

For practical purposes I have spawned 15 worker threads who
each will connect to a different host (480 Hosts total).  They would
then execute code similar to this:

---------------------------------------
Sock.send('some-command')
Reply = ReadUntil('> ', Sock, Timeout)

if Reply == '':
    return(-1)
else:
    tmp = re.split('\r\n', Reply)
    for line in tmp:
	m = pat.match(line)
	.etc.

Upon running the script you see nice and speedy results at first with
very little cpu usage.  However, the cpu usage grows continously until
the program ends.  I have noticed that when the bloating occurs the
a lot of the cpu time is spent in the kernel which would lead me to
believe this is some sort of locking issue hidden within the 're'
module?  When I put a 'continue' in front of the pat.match() every
thing works just fine.  

With re:
CPU states: 12.5% idle, 18.8% user, 46.4% kernel
26.54% test.py

Without:
CPU states: 19.1% idle, 35.8% user, 10.6% kernel,
0.76% test.py

The pattern I'm matching is as follows:
pat = re.compile('\s+(\w+)\s+\(.+?\):\s+(\d+)\s+(\d+)')

Is there any known issues with locking and 're'?

The system is an UE5000 with 8 cpus, Solaris 2.6, Python 1.5.2

If anyone has any ideas about this I desperately need some help.  I
hate to have to dump this entire project for performance reasons.
I've stripped out EVERYTHING in the program just to reproduct this to
rule out my code.

Regards,
Cliff




From davidopp at megsinet.net  Thu Sep 16 02:45:14 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 16 Sep 1999 02:45:14 -0400
Subject: high school programming & python
References:  <37DF9080.21A8D285@calibre-dd.com>  <37DFED78.6B258366@bioreason.com> 
Message-ID: <37E091FA.10EFE3D8@megsinet.net>


James Whitley Moore wrote:

> Why not end this thread in the usual way?
>
> I never wanted to be a high-level language programmer.
>
> I always wanted to be. . .
>

A World Class Skydiver and All Around Daredevil.

 (Being a beach bum wouldn't be so bad either.  I was watching a baseball
game with a friend when a commercial for a local tech school came on the
air and it went something like this "....So you went and got a four year
degree and what did it get you?  ...  A job serving drinks?  "  I looked at my friend and said to him ...
"Man, I would love a job like that... wear a swimsuit to work, get a tan,
hang out on the beach, and get paid for it.  That would be the life!"

David O.




From billjoeckel at hotmail.com  Sat Sep 11 13:58:56 1999
From: billjoeckel at hotmail.com (billjoeckel at hotmail.com)
Date: Sat, 11 Sep 1999 17:58:56 GMT
Subject: hiring python-solaris for web project
Message-ID: <7re58s$fhl$1@nnrp1.deja.com>

Hi! My name is Bill Joeckel with Tri-Com Technical Services, I have a
client that has a contract opening for a python-solaris developer for
a web project. The skills specifically looking for python cgi style
web development, mail handling software to provide confirmations
of actions on the web site and announcements to users, ODBC connectivity
to Db2 and software to process data files exchanged via FTP.Call
work 18002806919 or at home call anytime 19132561985.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bowman at montana.com  Sun Sep  5 00:32:23 1999
From: bowman at montana.com (bowman)
Date: Sat, 04 Sep 1999 22:32:23 -0600
Subject: A scrollable Frame or Canvas widget
References: <37C73097.8165F42C@uniserve.com> <37D1EA28.DF1A0EEF@alum.mit.edu>
Message-ID: <37D1F257.2F015097@montana.com>

pehr anderson wrote:
> 
> I'm trying to throw a phalnx of textboxes for a form that is bigger
> than the available window size. I need to be able to scroll the whole
> mess, but I'm trapped with the lack of docs and no understanding of tcl.

For a good example, look in your distro under Demo/tkinter/matt  for the
oddly
named:

canvas-with-scrollbars.py 


-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From Howard at quercus.demon.co.uk  Sat Sep 18 11:46:23 1999
From: Howard at quercus.demon.co.uk (Howard Oakley)
Date: Sat, 18 Sep 1999 16:46:23 +0100
Subject: [Mac] Python mode for Alpha editor
References:  <01HW.B4090E63001D90290D2F1590@news1.radix.net>
Message-ID: 

In article <01HW.B4090E63001D90290D2F1590 at news1.radix.net>,
"T. Bridgman"  wrote:

>There is in fact a Python mode for Alpha but it doesn't seem to work in release 
>7.0 or later.  Does this version work with 7.+?

It has been developed with 7.2, and I have tried to follow the
compatibility guidelines for future releases using Tcl 8.x. So it should
work fine - if not, please let me know.

Howard.






From gmcm at hypernet.com  Fri Sep 24 07:32:05 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 24 Sep 1999 07:32:05 -0400
Subject: Dialect, A Python derived environment (and language ?)
In-Reply-To: <000001bf064f$ba05f980$cc2d153f@tim>
References: <7sciqs$n2e@mail.psy.uva.nl>
Message-ID: <1273949950-4000037@hypernet.com>

[Tim, on http://www.aristar.com/DeveloperTools.html] 

> > (Nice feature: makes an exe right away.)
> 
> Very easily, too!  Just click "make an .exe" in the IDE, and it
> makes one. Just what I've been waiting for:  instead of a
> burdensome 20-byte cross-platform
> 
> print "Hello world"
> 
> file, you can get the same effect now with a self-contained 685Kb
> Windows-only .exe:

Gee, just ran your app thru Installer. For the exe that's 
Run+CArchive([HelloWorld.py]) it comes out to 24K. For the 
one that's got python15.dll etc. built-in, it's still only 277K.

- Gordon



From fredp at multimania.com.nospam  Wed Sep 22 08:57:16 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 22 Sep 1999 12:57:16 GMT
Subject: YA-Newbie: baffled by deepcopy/pickle issues...
References: <000101bf04b6$8d8ab520$342d153f@tim>
Message-ID: <8E499CFF7PaCmAnRDLM@161.48.81.148>

tim_one at email.msn.com (Tim Peters) said :

>> Any hints on how to track this one down ? Thanks,
>
>Easy:  create a *small* app that displays the same symptoms.  The answer
>will likely become obvious in the process of doing that.  If it doesn't,
>you'll have a small piece of executable code to post, and the answer
>will be obvious to someone else.
>code-fixes-itself-if-you-let-it-ly y'rs  - tim

It's what I usually do, but this time I half-hoped I could avoid it, because 
there's a lot of dumbing-down work implied at this stage of the app. But I 
guess I'm just going to have to bite the bullet...

thanks-for-the-grow-up-and-face-reality-ly yrs - fp :-))

-- 
YAFAP : http://www.multimania.com/fredp/



From Markus_Ewald at gmx.net  Wed Sep  8 07:05:21 1999
From: Markus_Ewald at gmx.net (Markus Ewald)
Date: Wed, 08 Sep 1999 13:05:21 +0200
Subject: Calling Python functions from C
Message-ID: <37D642F1.9FAE941C@gmx.net>

I'm trying to use Python for various things in a game. One of these
things is controlling particles.

I had used SeeR (http://home.elka.pw.edu.pl/~ppodsiad/seer/) in the past
which could easily compile its scripts into bytecode and keep them in
memory, you could then easily call functions in the script.
Only bad thing was it had some funny floating point bugs which
...well... caused some undesired effects on my particles ;)

So, two questions:

My game kept pointers to its particles in a simple linked list.
Each particle type (rain, smoke...) had a script which contained
functions for creating, moving, rendering and destroying the particle.
To add a particle my game just called the creation function in the
script for that particle-type, which just malloced as much memory as the
particle would need, set some initial values and return the pointer to
my game.
This, of course, wouldn't be possible in that way using Python, so what
is the best (quickest) solution ?
Write a Python module for storing particles and let the script keep them
?

Other question:
How can I keep a compiled Python script in memory and call functions
(not executing the whole thing) in it ?
I didn't find any example on the net until now and have already tried
alot...
like getting a callable object from a script from a _node to a
CodeObject to a PyObject with
  PyParser_SimpleParseFile();
  PyNode_Compile();
  PyEval_EvalCode();
and then calling functions to execute something in it...
  PyImport_ExecCodeModule();
  PyObject_CallFunction();
well, didn't work ;..(

Thanks,
and if anyone knows an url with something about embedding python, i'd be
happy :)

-Markus-






From news at dorb.com  Mon Sep 27 11:10:38 1999
From: news at dorb.com (Darrell)
Date: Mon, 27 Sep 1999 11:10:38 -0400
Subject: module location
References: <7sj6ii$1jc$1@essle.valdyas.org>
Message-ID: 

Two approaches I know of:

Define a module to hold an interface to some component.

abstractDB.py

It imports the specific version of db and provides any glue. All access to
DB is via abstractDB which can be imported by a module deeply buried in some
package.

Remindes me of this sort of thing:
        import myPackage.myMod
        myMod=myPackage.myMod

Or:

pass an instance that provides the desired interface to the lower level
packages. This is the most common and probably most flexible.

--
--Darrell
Boudewijn Rempt  wrote in message
news:7sj6ii$1jc$1 at essle.valdyas.org...
> I'm working on my first large(ish) python project - a linguistics database
> application - and I'm trying to keep my sources ordered, and my layers
> separated, and having some problems doing it right.
>
> So I've got a database access module, a middleware module, an abstract
> gui module, and finally the actual application, which binds the gui to
> the middleware. Currently, I've laid my project out as follows:
>
> lng (contains start script, readme, on so on)
>   /kuralib (contains the actual app
>    /db   (database layer)
> /gui  (gui widgets)
> /lng  (middleware - table and record objects)
>
> But now my problem is that the middleware also needs to access the db
> layer - as does the application itself (for connecting and so on). What's
> the received wisdom in this case? Do I make the db layer a separate
> module which I install in /usr/lib/python/site-packages, or do I make
> db a submodule of lng, and pass the connection on to the main application?
>
> I really want to keep everything as separate as possible, to ease the
> task of people who want to port the application to windows - something
> I won't be doing myself.
>
> --
>
> Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt





From evan at tokenexchange.com  Fri Sep  3 11:58:13 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 3 Sep 1999 10:58:13 -0500
Subject: Newbies: Re: Returning none
References: <000001bef265$fe64f340$192d153f@tim>  <37CF5499.BA24C33E@prescod.net>  
Message-ID: 

Michael Hudson  wrote in message
news:Pine.LNX.4.10.9909031629210.21765-100000 at atrus.jesus.cam.ac.uk...
> Got bored.
[and implemented the whole damn thing!]

Wow.  That looks ... remarkably painless.  Why don't you think it's a good
idea?





From eddthompsonNOzlSPAM at ssi.parlorcity.com  Fri Sep  3 07:16:36 1999
From: eddthompsonNOzlSPAM at ssi.parlorcity.com (EuphoriaDJ)
Date: Fri, 03 Sep 1999 04:16:36 -0700
Subject: multiOS install
Message-ID: <26f15c5e.d9be7115@usw-ex0102-015.remarq.com>

I am working on a prayer book parser,
it will either remind a person that it is time to pray and 
what prayer it is time for or it will just tell the user 
what prayer should be said at the time. Most of the code is 
pretty streight foreward but I want to be able to give this 
to people who can run it on other OS's
I think on the Mac there is a program that turns your python 
into native code so that you do not need python libs is this 
true  and is it availible for Windoze also?
next is there anything special I should do for linux user to 
help with install 

TKs

remove N

* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!




From carvdawg at patriot.net  Sat Sep  4 10:37:17 1999
From: carvdawg at patriot.net (Harlan Carvey, CISSP)
Date: Sat, 04 Sep 1999 10:37:17 -0400
Subject: New to Python...looking for scripting language
Message-ID: <37D12E9D.EDB193B8@patriot.net>

Hey all,

I am looking for a scripting language for NT environments.  I am NOT
trying
to start a language war.

I have been to the Python.org site quite a bit, and haven't found what I
am looking
for, so I will post here.

With Python, can I:
- access the Registry (get, and set values, create keys, etc)?
- access the services (get a list of running services)?
- access the audit policy (get and set that policy)?
- search for files on the remote hard drive?
- access permissions on the remote system?

I haven't seen modules that allow this...nor have I seen example
scripts.  I am not
a MS/C++ programmer, and I want to be able to leave scripts for
sysadmins who
aren't either.  They need to be able to maintain the scripts.

Thanks,

H. Carvey




From init__self at my-deja.com  Tue Sep  7 16:30:50 1999
From: init__self at my-deja.com (Python Addict)
Date: Tue, 07 Sep 1999 20:30:50 GMT
Subject: Feast or famine!
Message-ID: 

One moment the book by Mark Lutz is the only Python book on the shelf
and before you know it, you've got tons of them! Well, you will in a few
weeks.

A casual stroll around the usual places has found these
soon-to-be-released titles:

 - Python Annotated Archives 
   Martin Brown 
   ISBN: 0072121041
   (If it's anything like his excellent Perl book, this is going to
   fill a void. Let's hope it sells well and we get a Python: The
   Complete Reference too!)

 - Python and Tkinter Programming
   ISBN: John E. Grayson
   1884777813
 
 - Python Programming on Win 32 
   Mark Hammond, Andy Robinson, Robert Denn
   ISBN: 1565926218
   (Again, much needed. If it's like the Unix-ish one, it'll be 
   wonderful)
 
 - Python Reference Manual; February 19, 1999, Release 1.5.2
   Guido Van Rossum, Fred L., Jr. Drake
   ISBN: 1583483748
   (What's all this then? No piccie on Amazon.com or description)
 
 - Quick Python Book
   Daryl D. Harms, Kenneth McDonald 
   ISBN: 1884777740

I-wonder-if-Amazon-will-do-a-bulk-discount-ly-yr's,
Addicted.




From zeevkotzer-withouth-this at hotmail.com  Sat Sep  4 14:22:47 1999
From: zeevkotzer-withouth-this at hotmail.com (Zeev kotzer)
Date: Sat, 4 Sep 1999 20:22:47 +0200
Subject: Changing the current thread priority under Win32
Message-ID: <7qro4h$611$1@news.netvision.net.il>

Anybody has the magic python lines that can set/reduce the priority of a
single-threaded python script, running on WinNT?

Thx in advance!





From bobklin at idt.net  Sat Sep  4 12:50:51 1999
From: bobklin at idt.net (Bob Kline)
Date: Sat, 4 Sep 1999 12:50:51 -0400
Subject: Failure building MySQLdb
Message-ID: <7qrimc$bkn@nnrp3.farm.idt.net>

I'm trying to build MySQLdb version 0.0.3.  The README is a bit sparse.  It
says:

"To build:       python build.py
To install:     follow additional instructions after build.

Uh, that's it."

OK, worked fine on Linux (2.2.5).  On NT the following diagnostics are
spewed out:

cp: f:\python/lib/python1.5/config/Makefile.pre.in: No such file or
directory
make: Makefile.pre.in: No such file or directory
make: *** No rule to make target `Makefile.pre.in'.  Stop.
make: *** No targets specified and no makefile found.  Stop.
Traceback (innermost last):
  File "build.py", line 14, in ?
    import MySQLdb
  File "MySQLdb.py", line 19, in ?
    from _mysql import *
ImportError: No module named _mysql

The version of Python:
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Installed from py152.exe and python-win32all-125.exe, downloaded from
http://www.python.org/download/download_windows.html.  Obviously the
NT installation is missing some pieces.  Can I get a tip on where to look
for the missing pieces?  If it's in the FAQ (I looked) just whack me over
the head and give me a string to search for.

Thanks.

Bob Kline
bkline at rksystems.com






From salhir at earthlink.net  Thu Sep  2 23:28:47 1999
From: salhir at earthlink.net (Sinan Si Alhir)
Date: Thu, 02 Sep 1999 22:28:47 -0500
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B52F@HQOBGEX1>
Message-ID: <37CF406F.4EAB7454@earthlink.net>

ScherBi at BAM.com and comp.lang.python -

Various papers at http://home.earthlink.net/~salhir may be of interest
to you:

*  What is the Unified Modeling Language (UML)?
   (http://home.earthlink.net/~salhir/#whatistheuml)

*  Applying the Unified Modeling Language (UML)
   (http://home.earthlink.net/~salhir/#applyingtheuml)

*  The Foundation of the Unified Modeling Language (UML)
   (http://home.earthlink.net/~salhir/#thefoundationoftheuml)

*  Succeeding with the Unified Modeling Language (UML)
   (http://home.earthlink.net/~salhir/#succeedingwiththeuml)

*  The Object-Oriented Paradigm
   (http://home.earthlink.net/~salhir/#theobjectorientedparadigm)

*  Extending the Unified Modeling Language (UML)
   (http://home.earthlink.net/~salhir/#extendingtheuml)
   This paper elaborates on extending the UML and explores the
necessity, sufficiency, and consistency of the UML?s extension
mechanisms to conclude that there are inconsistencies with the UML?s
extension mechanisms..

*  "The True Value of the Unified Modeling Language (UML)"
   [Distributed Computing magazine, July 1998]
   (http://home.earthlink.net/~salhir/#thetruevalueoftheuml)

*  "Unified Modeling Language Extension Mechanisms"
   [Distributed Computing magazine, December 1998]
   (http://home.earthlink.net/~salhir/#umlextensionmechanisms)
   This paper also very briefly introduces the Core Modeling Language
(CML).

*  "Reuse and the Unified Modeling Language (UML)"
   (http://home.earthlink.net/~salhir/#reuseandtheuml)

*  "The Unified Modeling Language (UML) - One Year After Adoption of the
Standard"
  
(http://home.earthlink.net/~salhir/#theumloneyearafteradoptionofthestandard)

*  "Understanding the Unified Modeling Language"
   [Methods & Tools, April 1999]
   (http://home.earthlink.net/~salhir/#understandingtheuml)

*  "From the Unified Modeling Language (UML) to C++"
   [Visual C++ Developers Journal (Online), June 1999]
   (http://home.earthlink.net/~salhir/#fromtheumltocpp)

I welcome your comments and questions.

Thank you.

Sincerely,
Sinan Si Alhir
Email: mailto:salhir at earthlink.net
WWW:   http://home.earthlink.net/~salhir

-- --- --- 
Read what people are saying about "UML in a Nutshell" (Comments and
Feedback) [http://home.earthlink.net/~salhir/#umlinanutshell]


ScherBi at BAM.com wrote:
> 
> >Thanks for your reply Bill.
> You're quite welcome.
> 
> >I'll try it the other way around. Is the UML something like a OO
> >Database or something like XML? Or anything in between?
> 
> For starters, the 'L' stands for language in both.  Let me quote from
> _UML_In_A_Nutshell_ here:
> 
> "The UML is a modeling language for specifying, visualizing, constructing,
> and documenting the artifacts of a system-intensive process"
> 
> Note the 'system-intensive process'; it's not directed solely at software
> design, but that seems to be where it has had it's greatest impact.  Other
> applications may be business process modeling, factory process modeling,
> etc.
> 
> >1.) Object Primer(Scott W. Ambler SIGS Books/
> >     Cambridge University Press, 1995)
> >2.) UML in a Nutshell (OReilly)
> 
> >Which - or would you recommend both?
> 
> I've not seen _Object_Primer_, but I'm quite happy with the Nutshell book.
> I've seen others I didn't like as much.
> 
> >I started documenting my work on AI in a simple word processor, then
> >thinking of creating a database (to get hold of the code and my
> >thoughts) and recently I heard something about CASE tools and UML.
> >I'm really in great need of a tool which helps me to keep overview and
> >organize my work.
> 
> A UML tool, I think, could only be a part of the solution here.  It's not
> going to provide a database to hold code.  There are mechnisms for handling
> documentation, however these bits of documentation are typically attached to
> specific 'objects' within the tool.
> 
> >>I use UML, via ObjectDomain (www.objectdomain.com).
> >>I am involved in developing ....
> >>.. In other words, create classes and class diagrams in
> >>UML, design the class interfaces, and play around with the design at this
> >>level.  When you think you have it right, generate the stub code and fill
> in
> >>the methods.  You can fill in the methods without getting caught up in the
> >>interface issues, which makes things much easier.
> 
> >Feels like UML is a sort of IDE??
> No, but a CASE tool which uses UML as it's modeling language is a bit like
> an IDE.  Again uml is just a language, like Python, C, or Latin.
> 
> >>..where you understand the hierarchy and the operations, then
> >>reverse engineer it into objectdomain.  You can then automatically
> generate
> >>a class diagram.
> 
> >Something like a class browser used in C/C++ IDE's like Watcom?
> Yes, I suppose.  But it shows you the class hierarchies in UML.
> 
> >>And no, I don't work for ObjectDomain.  It's the only UML tool I know of
> the
> >>generates and reverse engineers Python.  OD is written in Java.  I don't
> 
> >Whats so wrong about Java? When you use different OS's - Java should a
> >logical way to go.
> 
> It's a bias born of experience ;-). The idea, the promise, the goal of Java
> is noble and worthy. The reality of Java falls short.  I would prefer a
> native code version.  It would stand alone( no jvm version dependencies),
> and it would execute faster.  I suspect it would be more stable, also.  That
> said, I'm quite happy with OD, but if they came out with a native code
> version for a suitable platform, I would drop the Java version immediatly.
> 
> UML is a big thing.  I'm still a UML neophyte, but I'm getting use out of
> it.  It took time, reading and experimenting, to get to this point, but it
> has been worth it.  Not only have I begun to learn UML, but my OO
> programming skills and understanding have improved too.  My company has
> found enough value in it to send me out for training.  If you think UML
> might work for you, I can't stress enough the value in downloading a tool,
> like OD, and giving it a spin.
> 
> >Thanks for your patience
> Your welcome,
> 
> Bill Scherer




From lewisjd at msx.upmc.edu  Fri Sep  3 13:31:06 1999
From: lewisjd at msx.upmc.edu (Jack Lewis)
Date: Fri, 03 Sep 1999 13:31:06 -0400
Subject: PYTHON on ALPHA VMS 7.2?
Message-ID: <37D005D9.38C27D05@msx.upmc.edu>

Has anyone successfully built PYTHON on ALPHA VMS 7.2 ?

As well as been able to compile and link 'C' programs with Embedded
PYTHON calls?
I can't seem to get past the link errors:

%LINK-W-NUDFSYMS, 2 undefined symbols:
%LINK-I-UDFSYM,         CTL$GL_IMGHDRBF
%LINK-I-UDFSYM,         _PYIMPORT_INITTAB
%LINK-W-USEUNDEF, undefined symbol _PYIMPORT_INITTAB referenced
        in psect PYIMPORT_INITTAB offset %X00000000
        in module IMPORT file
DSA900:[PYTHON.PYTHON-1_5_2.VMS.O_ALPHA]PYTHON.OLB
;10
%LINK-W-USEUNDEF, undefined symbol CTL$GL_IMGHDRBF referenced
        in psect $LINK$ offset %X00000010
        in module VMS__GETBUILDINFO file
DSA900:[PYTHON.PYTHON-1_5_2.VMS.O_ALPHA
]VMS.OLB;10





From philippe.boite at pacwan.fr  Sat Sep  4 07:45:36 1999
From: philippe.boite at pacwan.fr (Philippe Boite)
Date: Sat, 4 Sep 1999 13:45:36 +0200
Subject: Spanish Inquisition?!
References: <37CFC38C.20A9C267@megsinet.net>
Message-ID: <7qr293$1v7t$1@news5.isdnet.net>

The Spanish Inquisition is a famous sketch by the Monty Pythons.
Somebody says :
- I wouldn't expect the Spanish Inquisition...
And then enter 3 red Spanish inquisitors, one of theme saying :
- Nooobody expects the Spanish Inquisition...

Python was named after the Monty Pythons, so there are a lot of references
to them, for example in the tutorial :
- The Parrot sketch
- And nice red uniforms...

There are also often references in this ng (for a recent example, the
lumberjack song).

David Oppenheimer  a ?crit dans le message :
37CFC38C.20A9C267 at megsinet.net...
> In a previous thread (about languages interpreted versus languages
> compiled) there was talk of the Spanish Inquisition.  I am guessing that
> this is 1) a joke or 2) the code name for a "secret" project to create
> compiled code from python code and thereby permanently blur the line for
> Python.  Anyone know anything about this?
>
> I know no one ever expects the Spanish Inquisition, but if its coming
> I'd like to know about it.
>
> David O.
>
> "Any Sufficiently Advanced Technology Is Indistinguishable From Magic"
> - Arthur C. Clarke
>
> P.S. Creating compiled code from Python  that didn't require the
> interpreter to be distributed with it would be high magic indeed.
>





From gmcm at hypernet.com  Thu Sep  9 14:01:40 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 9 Sep 1999 13:01:40 -0500
Subject: Embedding question (doing Python from C)
In-Reply-To: <37D7CE73.43898B33@ncsa.uiuc.edu>
Message-ID: <1275226148-104438625@hypernet.com>

Randy Heiland writes:

> I have a C++ server executing Python scripts using just
> PyRun_SimpleString and it works great -- mostly.  However, it
> doesn't handle more complicated Python commands/constructs, e.g., a
> 'for' loop. Can someone offer advice?

What seems to be the problem? PyRun_SimpleString can't hack Windows 
line endings (should just be '\n'), and requires a null-terminated 
string, but said SimpleString's contents can be arbitrarily complex 
Python code.

- Gordon



From philh at vision25.demon.co.uk  Sat Sep 11 20:27:58 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 12 Sep 99 00:27:58 GMT
Subject: yacc/lex for Python
References: <37DA861F.57212BC3@atl.mediaone.net>
Message-ID: <937096078snz@vision25.demon.co.uk>

In article <37DA861F.57212BC3 at atl.mediaone.net>
           nosleep at atl.mediaone.net "Kristopher Johnson" writes:
> Are there any lex- or yacc-like tools available for Python?

Oodles of 'em.

My Parrot program uses John Aycock's parsing framework.

See http://www.vision25.demon.co.uk/prog/parrot.html for details.

-- 
Phil Hunt - - - - - - - - -  philh at vision25.demon.co.uk
 - Linux will be 8 years old on 17th September! See: -
http://www.vision25.demon.co.uk/prog/linuxbirthday.html




From jayseye at my-deja.com  Wed Sep 15 10:01:01 1999
From: jayseye at my-deja.com (jayseye at my-deja.com)
Date: Wed, 15 Sep 1999 14:01:01 GMT
Subject: TreeCtrl Object (win32ui)
References: <7rm9bb$1rh$1@nnrp1.deja.com>
Message-ID: <7ro8ql$eao$1@nnrp1.deja.com>

Does Python offer a cross-platform Tree control,
or does it only provide access to the Win32 control
when running under Windows?

Thanks.



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Sun Sep 19 16:08:06 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 19 Sep 1999 16:08:06 -0400
Subject: Force anything to be a string.
In-Reply-To: <99091911524208.00550@quadra.teleo.net>
Message-ID: <000201bf02da$affbbce0$102d153f@tim>

[Patrick Phalen, on inadvertently stomping builtin function names]
> ...
> How *hard* would it be to implement a warning message for this and is
> it worth it? What marks the difference between a warning and
> an error message?

It's irritatingly difficult, but only because Python has no warning msgs
today.  So it means inventing a whole bunch of machinery and conventions
where none currently exist.  I think it should be done anyway, but since I'm
not volunteering my own time to do it can't whine if nobody else does.

Note that Macs don't have a concept of stdout or stderr, and at least under
Windows the "no DOS box" pythonw variant doesn't have a visible stdout or
stderr.  So there's currently no good place to direct warnings even if there
were some.  That's not a new problem (there's no good place to direct fatal
error msgs either!), but I think Guido would rather see machinery get
invented to address that first.

don't-everyone-rush-in-all-at-once-now-ly y'rs  - tim





From philh at vision25.demon.co.uk  Tue Sep 14 19:37:27 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 14 Sep 99 23:37:27 GMT
Subject: python call external program (WHAT THE?!)
References: <65118AEEFF5AD3118E8300508B124877073BF5@webmail.altiris.com>
Message-ID: <937352247snz@vision25.demon.co.uk>

In article <65118AEEFF5AD3118E8300508B124877073BF5 at webmail.altiris.com>
           MSteed at altiris.com "Mike Steed" writes:
> > From: Ben Thomas [mailto:bthomas at trey-industries.com]
> > Sent: Tuesday, September 14, 1999 9:31 AM
> > To: python-list at python.org
> > Subject: Re: python call external program (WHAT THE?!)
> > 
> > 
> > I just tried but putting the batch file in three places on a 
> > drive. The I tried
> > them all.
> > 
> > system('d:\mybat.bat') = ok
> > system('d:\data\mybat.bat') = ok
> > system('d:\data\vacman\mybat.bat') = fails, it works fine 
> > from DOS prompt!
> > 
> > 
> > what gives?
> > Ben
> 
> What gives is that \v is an escape code (representing, I believe, the ASCII
> VT character).  There are a few ways around this.
> 
> # Use raw strings
> system(r'd:\data\vacman\mybat.bat')
> 
> # Escape your backslashes
> system('d:\\data\\vacman\\mybat.bat')
> 
> # Use forward slashes (at least this works on WinNT)
> system('d:/data/vacman/mybat.bat')

You missed out:

# use Linux instead of MS rubbish
system('/usr/local/data/myshell.sh')

-- 
Phil Hunt - - - - - - - - -  philh at vision25.demon.co.uk
 - Linux will be 8 years old on 17th September! See: -
http://www.vision25.demon.co.uk/prog/linuxbirthday.html




From arnold at dstc.edu.au  Wed Sep  1 02:12:30 1999
From: arnold at dstc.edu.au (David Arnold)
Date: Wed, 1 Sep 1999 06:12:30 GMT
Subject: threads on Solaris 
In-Reply-To: Your message of "Mon, 30 Aug 1999 18:38:26 EST."
             <199908302338.AA04307@ch1d2833nwk> 
References: <199908302338.AA04307@ch1d2833nwk>
Message-ID: <199909010612.QAA06055@piglet.dstc.edu.au>

-->"Ted" == Ted Horst  writes:

  Ted> Does anyone know why I get 4 threads when I start the
  Ted> interpreter under Solaris (as reported by /bin/ps -aL) ?

  Ted> This is python 1.5.2 built with a SUNPro compiler (sorry, I
  Ted> don't have the config options).

i see the same behaviour, also SUNPro, but i *know* mine is configured
with the thread module enabled, which i imagine has something to do
with it?

  Ted> SunOS ch1d2964syb 5.6 Generic_105181-12 sun4u sparc

SunOS fig.dstc.edu.au 5.6 Generic_105181-11 sun4u sparc


using dbx, i see that there are 3 threads and 4 LWPs active:

  (dbx) threads
   >    t at 1  a l at 1  ?()   running                 in _libc_read()
        t at 2  b l at 2  ?()   running                 in __signotifywait()
        t at 3         ?()   sleep on (unknown)      in _swtch()
  (dbx) lwps
   >l at 1 running          in _libc_read()
    l at 2 running          in __signotifywait()
    l at 3 running          in ___lwp_cond_wait()
    l at 4 running          in _door_return()
  (dbx) 

thread t at 1 (lwp l at 1) is the interpreter's interactive prompt waiting
for me to type something.  idly speculating, t at 2 could handle signal
delivery?  l at 3 (in the cond_wait, actually a cond_timedwait) is called
by a function _age ... no real clues on what that might be doing.

what t at 3 and l at 4 are doing i've no idea (no useful call stack from
dbx).

but at least your observed behaviour isn't unique ;-)




d



From wtanksle at dolphin.openprojects.net  Sat Sep 25 21:20:41 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sun, 26 Sep 1999 01:20:41 GMT
Subject: Moving from Perl to Python
References: 
Message-ID: 

On Sun, 26 Sep 1999 00:35:25 +0000 (UTC), jonathon wrote:
>On 25 Sep 1999, Harry George wrote:

>>I also have the perl books and like cookbook.  I'm in the process of
>	[ Text deleted ]
>>Except for copyright issues, I'd say just redo all the Perl 

>	A request to translate all the Perl Books into Python?  

I think he just wanted the Perl Cookbook, which would indeed be great
(except, of course, for the parts which are trivial in Python -- we don't
need no hashes-and-list man page ;-).

>	Other interesting/useful language translations might be

>	_Numerical Fortran_/_Numerical C_/_Numerical Pascal_  	
>			<< Or has this been done?  >>

That's facinating indeed --  but I suspect that NumPy already has all of
those implemented (in C and linked to Python).

>	Donald Knuth's _The Art Of Programming_  Python instead of MX.
>	<< At one time, Bill Gates automatically hired anybody who had
>	worked through 75% or more of the questions in that book. >>

No, all they had to do was read through the whole series and be able to
prove it.  75% solution rate would be amazing.  However, he didn't say
he'd hire you; he said, "send your resume to me."  I can't see any
employer turning anyone who's done that away, though.

Impossible -- the purpose of MIX is to provide an assembly language which
allows the algorithms to be demonstrated at a low level.  If Knuth used
Python for everything else, he'd still have to invent an assembly language
for the chapters on code generation.  (See his homepage for more details.)

>	Sedgewick's book on algorithms.  Don't remember the title 
>	offhand.  :-(     

"Algorithms".  A very good idea.

>	xan
>	jonathon

-- 
-William "Billy" Tanksley



From gregm at iname.com  Wed Sep  1 07:44:13 1999
From: gregm at iname.com (Greg McFarlane)
Date: Wed, 1 Sep 1999 21:44:13 +1000
Subject: position on the screen
In-Reply-To: <37cc6b98.873539@news.telepac.pt>; from Xana on 31 Aug 1999 at 11:59:54PM
References: <37cc6b98.873539@news.telepac.pt>
Message-ID: <19990901214413.34597@nms.otc.telstra.com.au>

On 31 Aug, Xana wrote:
> I'm working with PythonMegaWidgets and i have a problem using
> Pwm.dialog boxes.
> Can anyone tell me if  (and how) I can define the position in wich the
> window will appear in the screen ?

You can use the same methods as any Tkinter.Toplevel. For example:

    import Pmw 
    d = Pmw.Dialog(buttons = ('Yes', 'No', 'Dunno'))
    d.geometry('+400+400')

will position the dialog window 400 pixels down and across from the
top left corner of the screen.

If you want to display the dialog modally (ie:  disallow any user
inputs except to the dialog), use the activate method with the
geometry option.  For example,

    d.activate(geometry = '+400+400')

See the Pmw.MegaToplevel manual page for the full story.

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com



From hschroeder at gmx.net  Sat Sep 25 15:19:31 1999
From: hschroeder at gmx.net (Henning Schroeder)
Date: Sat, 25 Sep 1999 21:19:31 +0200
Subject: GUI-builder for PyQt/PyKDE
Message-ID: <387js7.klc.ln@odin.henning-schroeder.de>

Hi there,

the above subject is no inquiry on such a thing. 

Actually I am working on a visual IDE based on the Qt and KDE bindings 
for Python. Therefore I implemented a small component library which fits 
the needs for visual development. I call the library ViPyl. It
consists mainly of some wrapper classes around Qt widgets and adds property
support to them. Unique and easy event handling is under
development and currently not available. So it is not ready for real
development yet.

If you are interested int ViPyl and the GUI builder look at the following page
for some screenshots and for an alpha version to download:
  http://www.henning-schroeder.de/en/python/vipyl/


Suggestions are welcome!

  Henning



From fleck at informatik.uni-bonn.de  Tue Sep 21 02:43:55 1999
From: fleck at informatik.uni-bonn.de (Markus Fleck)
Date: 21 Sep 1999 06:43:55 GMT
Subject: ANN: Python Announcements...
Message-ID: <7s79fb$nfg@news.rhrz.uni-bonn.de>

Dear fellow Pythoneers,

the articles queue for comp.lang.python.announce has been stalled since
August 22, 1999 - mostly due to a temporary lack of time on my part.
There is currently a backlog of about 80 messages.

You can help to speed up the moderation process significantly by adhering
to the posting guidelines that are posted bi-weekly to comp.lang.python and
comp.lang.python.announce. In particular, life would become a lot easier
for me if all submitted announcements already included the requested HTML
mark-up footer (which is used to give a one-sentence summary of each
announcement at the www.python.org home page).

You can always find the latest version of the comp.lang.python.announce
posting guidelines at the following URL:

  .

Currently pending announcements will not be sent out all at once, but will
be distributed over the next few days, so as to avoid overloading the
inboxes of those people who have subscribed to comp.lang.python.announce via
the "python-announce-list" mailing list. If you want to convert your mailing
list subscription to digest format, or temporarily suspend your mailing list
subscription, you should do so now by visiting the list's Mailman web page
at .

I appreciate your patience, and the offers for help that I have received.
Of course, I am truly delighted to see so many announcements being posted
to comp.lang.python.announce - and all the time we're getting closer to
having the 1000th posting to comp.lang.python.announce by Y2K... -:^)


    Happy hacking, --
    Markus Fleck (Moderator)


from sys import*;from string import*;a=argv;[s,p,q]=filter(lambda x:x[:1]!=
'-',a);d='-d'in a;e,n=atol(p,16),atol(q,16);l=(len(q)+1)/2;o,inb=l-d,l-1+d
while s:s=stdin.read(inb);s and map(stdout.write,map(lambda i,b=pow(reduce(
lambda x,y:(x<<8L)+y,map(ord,s)),e,n):chr(b>>8*i&255),range(o-1,-1,-1)))
                             -- RSA in Python (or so I've been told )
-- 
----------- comp.lang.python.announce (moderated) ----------
Article Submission Address:  python-announce at python.org
Python Language Home Page:   http://www.python.org/
Python Quick Help Index:     http://www.python.org/Help.html
------------------------------------------------------------



From druid at primenet.com  Tue Sep 28 20:29:04 1999
From: druid at primenet.com (Michael A. Davis)
Date: 29 Sep 1999 00:29:04 GMT
Subject: Problem installing Python on AIX
References: <7sr2i2$hto$1@fir.prod.itd.earthlink.net>
Message-ID: <7srmgg$rtl$1@nnrp03.primenet.com>

In article <7sr2i2$hto$1 at fir.prod.itd.earthlink.net>,
Jamey Cribbs  wrote:
>I am attempting to install Python 1.5.2 on AIX version 4.2.1.  When I run
>./configure, I am getting errors in the config.log stating:
>
>ld: 0706-006 Cannot find or open library file: -l m
>              ld:open(): A file or directory in the path name does not
>exist.
>
>After this, the make also bombs out.  Does anyone have an idea what is
>wrong?  I can email the log file if that will help.

I am no AIX guru, but managed to reproduce this error message on with a
tiny C program:

-------------------------------------------------------------------
$ cat quiz.c
#include 
#include 
int main(int argc, char ** argv) { printf("%f\n", pow(2.3,1.5)); }
$ xlc -o quiz quiz.c -la
ld: 0706-006 Cannot find or open library file: -l a
        ld:open(): A file or directory in the path name does not exist.
$ echo $LIBPATH

-------------------------------------------------------------------

I'd say you don't have libm.a on your LIBPATH.

Mike

druid at primenet.com


>
>Any help would be greatly appreciated.
>
>Jamey Cribbs
>cribbsjg at earthlink.net
>
>
>


--
-------------------------------------------------------------------------------
--mad
druid at primenet.com



From philh at vision25.demon.co.uk  Fri Sep  3 16:28:55 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Fri, 03 Sep 99 20:28:55 GMT
Subject: Pirate -> RAD
References: <7qno38$6uk$1@news.is.co.za>
Message-ID: <936390535snz@vision25.demon.co.uk>

In article <7qno38$6uk$1 at news.is.co.za>
           jacques at integra.co.za "Jacques Oosthuizen" writes:
> Hi All
> 
> I am busy working on a tool I named Pirate ("Python RAD") it has a component
> library similar to Delphi and shall in the future work with win32,gtk,qt. It
> will have a IDE.
> 
> Before I do to much and work to hard on it, will anyone be interested in
> using a
> tool like this.

Possibly. It depends what features it has, and how easy it is to fit
in with other software.

-- 
Phil Hunt - - - - - - - - -  philh at vision25.demon.co.uk
 - Linux will be 8 years old on 17th September! See: -
http://www.vision25.demon.co.uk/prog/linuxbirthday.html




From hnowak at cuci.nl  Sat Sep  4 03:58:54 1999
From: hnowak at cuci.nl (Hans Nowak)
Date: Sat, 4 Sep 1999 07:58:54 GMT
Subject: Snippets site temporarily (?) down
Message-ID: <199909040800.KAA18341@dionysus.fw.cuci.nl>

OK, the title says it all... Well, in fact it's not just my snippets 
page, it's the whole damn hvision.nl site that's down. Again.

I have been thinking about moving the site elsewhere, since this 
isn't the first time that this happens. I do not want to have 
annoying popup ads and such, though (geocities/xoom/etc). Anyone has 
some free webspace I can use? 

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/



From guido at cnri.reston.va.us  Thu Sep  2 08:27:00 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 02 Sep 1999 08:27:00 -0400
Subject: Queue class deadlocking application
References: 
Message-ID: <5lhfld5wiz.fsf@eric.cnri.reston.va.us>

No real help here, but (1) this is an issue for the thread-sig, and
(2) you need to provide details about your platform and Python
configuration; threads are (unfortunately) very platform dependent and 
not all platforms have quality thread implementations.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From davidopp at megsinet.net  Fri Sep  3 02:31:55 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Fri, 03 Sep 1999 02:31:55 -0400
Subject: OH, I'm A Lumberjack And I'm OK...
References: <37CE2648.D4254FCC@megsinet.net> <37ceeef4.5698550@news.erols.com>  <37CF55DC.816F38F5@megsinet.net> 
Message-ID: <37CF6B5A.DC88A5E5@megsinet.net>

OK, I lied, here's yet another post in my continuing struggle to run painter.py

I took Robin's PYTHONPATH suggestion and now my traceback is as follows:

>>> Traceback (innermost last):
  File "C:\Python\Pythonwin\pywin\framework\scriptutils.py", line 237, in
RunScript
    exec codeObject in __main__.__dict__
  File "C:\Python\scripts\painter.py", line 65, in ?
    im = Image.open(sys.argv[1])
IndexError: list index out of range

I have now gotten to line 65 before I get an error (this is a tremendous
improvement of 53 lines more of code processed before an error kills the
program).

All Hail Robin Dunn, Master of wxPython and all around troubleshooter.  Got any
ideas what the heck this last error means Robin?

David O.


Robin Dunn wrote:

> >
> > PATHS     (yes, I took the suggestion to place my Python & related files
> in
> > pathways that have no spaces)
> >
> > Path to Python executable: C:\Python
> > Path to Tcl executable: C:\Tcl\bin
> > Path to PIL: C:\Python\PIL
> > Path to Tkinter: C:\Python\Lib\lib-tk
> > Path to Painter (a program that I am more than curious to see work, I am
> almost
> > obsessed now): C:\Python\scripts\painter.py
> >
> > My Traceback (error message):
> >
> > >>> Traceback (innermost last):
> >   File "C:\Python\Pythonwin\pywin\framework\scriptutils.py", line 237, in
> > RunScript
> >     exec codeObject in __main__.__dict__
> >   File "C:\Python\scripts\painter.py", line 12, in ?
> >     import Image, ImageTk
> > ImportError: No module named Image
> >
> > My Autoexec.bat settings:
> >
> > PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\WINDOWS;
> > set PATH=%PATH%C:\mingw32\bin;
> > set PATH=%PATH%C:\PYTHON;C:\PYTHON\PIL;C:\TCL\BIN;C:\PYTHON\LIB\LIB-TK;
> >
>
> It can't find the module because it's not on the PYTHONPATH.  Execute this
> python code to see what directories are in your PYTHONPATH:
>
> import sys
> print sys.path
>
> To fix it add this to your autoexec.bat:
>
> set PYTHONPATH=c:\Python\PIL
>
> or add the directory to the value in the Registry at
>
> HKEY_LOCAL_MACHINE\SOFTWARE\Python\PythonCore\1.5\PythonPath
>
> or as a subkey.
>
> --
> Robin Dunn
> Software Craftsman
> robin at AllDunn.com
> http://AllDunn.com/robin/
> http://AllDunn.com/wxPython/  Check it out!




From aa8vb at yahoo.com  Mon Sep 20 08:18:16 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 20 Sep 1999 08:18:16 -0400
Subject: Any good debuggers for Python? (was Re: using python debugger (pdb) inside emacs debugger mode ...)
In-Reply-To: <199909182333.QAA12325@hercules.cup.hp.com>; from Ovidiu Predescu on Sat, Sep 18, 1999 at 04:33:01PM -0700
References: <14307.57188.583343.140730@gargle.gargle.HOWL> <199909182333.QAA12325@hercules.cup.hp.com>
Message-ID: <19990920081816.A350855@vislab.epa.gov>

Ovidiu Predescu:
 |I didn't figure out exactly all the details, but the basic idea is to use
 |GDB to handle all the interaction with the user instead of pdb. pdb
 |becomes only a debugging engine that's loaded in a Python program and
 |knows how to setup breakpoints, continue the program etc. All these
 |operations are published in a simple Python API that could be called from
 |GDB.

I believe this is how DDD/PyDB works as well (providing the user with a
graphical Python debugger).

 |What I'm working on right now is adding scripting language support to GDB
 |so that this kind of functionality could be implemented in GDB using an
 |interpreted language and not C. The language I'm embedding is surprise,
 |Python ;-). This work is part of the WDB project here, at HP, but all the
 |work we're doing is going back in the FSF main tree. For more information
 |on WDB, take a look at:
 |
 |http://www.hp.com/esy/lang/tools/Debuggers/index.html

Sounds interesting.  Hope to hear more about this later.

-- 
Randall Hopper
aa8vb at yahoo.com



From fredrik at pythonware.com  Thu Sep 23 03:17:56 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 23 Sep 1999 07:17:56 GMT
Subject: Newbie Question
References: <37E773E7.B3A0CEA0@ecs.soton.ac.uk> <37E8AF24.1CDE9916@isogmbh.de> <002f01bf04e6$7d6a4950$f29b12c2@secret.pythonware.com> 
Message-ID: <00a001bf0593$c37a8060$f29b12c2@secret.pythonware.com>

Tom Culliton  wrote:
> string.atoi(), string.atof(), ... are also substantially faster (about
> twice as fast by my tests) than int(), float() and the like.

that was true in earlier versions, but as of 1.5.2,
int/float is a little bit faster than atoi/atof.

(if you write the latter as "string.atoi" etc,
the extra lookup time makes the builtin
versions about twice as fast on my box).

> Generally I would recommednd using int() when you need to
> coerce a "number" of arbitrary type to an integer (like you
> would a C/C++ cast), and using string.atoi() to interpret strings
> as integers.

good advice!






From claird at starbase.neosoft.com  Tue Sep 14 20:46:54 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Sep 1999 00:46:54 GMT
Subject: high school programming & python
References:  <937300939snz@vision25.demon.co.uk> <$JL+lCAFvt33Ewr8@jessikat.demon.co.uk> <7rmmjr$qv6@dfw-ixnews11.ix.netcom.com>
Message-ID: <2BAF9178787E2D18.F80B717C3A0115B2.701123DC5A66FDBE@lp.airnews.net>

In article <7rmmjr$qv6 at dfw-ixnews11.ix.netcom.com>,
Aahz Maruch  wrote:
>In article <$JL+lCAFvt33Ewr8 at jessikat.demon.co.uk>,
>Robin Becker   wrote:
>>In article <937300939snz at vision25.demon.co.uk>, Phil Hunt
>> writes
>>>In article 
>>>           nospam at bitbucket.com "Phil Mayes" writes:
>>>>
>>>> This is, as you imply, a fallacious argument; if carried to its logical
>>>> extreme, we shouldn't teach C because students will be spoiled and not
>>>> want to use assembler.  In fact, assembler spoils people for machine
>>>> language :-)
>>>
>>>Machine language! You young people today don't know what hardship
>>>is... when I were a lad, we had to manually patch the ALU using
>>>plug boards to change the logic.
>>
>>You were lucky to have plugboards. We had to key things in with a bunch
>>of switches.
>
>You were lucky to have switches.  We had to rewire the traces.
			.
			.
			.
If a mercury delay line won't pass it, it's not worth computing.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From greg.ewing at compaq.com  Mon Sep  6 18:00:26 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 07 Sep 1999 10:00:26 +1200
Subject: Class in class
References: <37D38E90.2E5713A9@acay.com.au>
Message-ID: <37D43979.E7A7B4AE@compaq.com>

Gurvijay Singh Bhatti wrote:
> 
> Dear everyone,
> 
> Like C++ and Java, is it possible to declare a class in a class? If it
> is then could you tell how?

class Foo:

  ...

  class Blarg:

    ...

The class Blarg is now a class attribute of class Foo,
so you can refer to it as Foo.Blarg, or as self.Blarg
within any method of class Foo.

Greg



From paul at prescod.net  Sat Sep  4 07:35:29 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 4 Sep 1999 11:35:29 GMT
Subject: Returning none
References: <000601bef688$f83412e0$ce2d153f@tim>
Message-ID: <37D10401.B6360540@prescod.net>

Tim Peters wrote:
> 


> Let's shift the ground of the debate:  I'd bet Guido's life that this won't
> go into 1.6, and I think most long-timers would agree with that regardless
> of their opinion of the technical merits pro or con (new keyword?!  more
> likely Tom Christiansen posts 10 energetic msgs bemoaning the lack of Python
> docs in PowerPoint format ).  But what about Python2?  If there were a
> clean way to plug the hole in that, would you object?

I want to point out that in the first message bearing this now disgraced
subject line I did say that this should be a version 2, not 1, change.
Forseeing the current flamage I said:

> This is obviously an incompatible change so I propose it for Python 2.

 Paul Prescod



From ashwood at usc.edu  Fri Sep  3 19:37:23 1999
From: ashwood at usc.edu (Joseph Ashwood)
Date: Fri, 3 Sep 1999 16:37:23 -0700
Subject: Win32 Specific: SimpleMAPI
References: 
Message-ID: 

I think the phrase DUH comes to mind, I found the function I wrote called
InitSimpleMAPI, so I obviously don't need that one. Sorry about that.
                Joseph
Joseph Ashwood  wrote in message
news:eK104el9#GA.277 at cpmsnbbsa03...
> Someone on here has to know thet answer, I just can't seem to find it. I'm
> looking to call the SimpleMAPI functions: InitSimpleMAPI, MAPILogon, and
> MAPIAddress. Could someone please shed some light on how to do this?
>                 Joseph
>
>





From mlh at idt.ntnu.no  Sun Sep 19 12:19:19 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 19 Sep 1999 18:19:19 +0200
Subject: My own pyhton in CGI...
Message-ID: 

I'm having a problem - I want to use the newest version of Python for
CGI, and the one installed by the techies around here is a bit old...

So - I installed my own in my home directory (which works just fine)
but when I try to write something like

  #!/my/home/directory/foo/bar/python

it doesn't work.

It there a standard problem/solution here, or must I dig more deeply
into it? (I think I have set the permissions for the python and its
super-directories properly...)

--

  Magnus
  Lie                                     Please forward all spam to
  Hetland                                                uce at ftc.org



From tim_one at email.msn.com  Sat Sep 25 23:01:02 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 25 Sep 1999 23:01:02 -0400
Subject: Moving from Perl to Python
In-Reply-To: 
Message-ID: <000601bf07cb$5e1ede20$602d153f@tim>

[Harry George, on converting Perl to Python]
> ...
> 2. Work out the regular expression tricks early -- I'm currently
>    looking for a python idiom for
>       s/@@([a-z]+)@@/$lookup{$1}/g
>
> Python needs a Cookbook.  Except for copyright issues, I'd say just
> redo all the Perl Cookbook examples in python -- or maybe O'Rielly has
> one in the works already.

Translation of that Perl into Python is straightforward but unsatisfying:

    import re
    lookup = { ... }
    pattern = re.compile(r"@@([a-z]+)@@")
    # s is input string, t is output string
    t = pattern.sub(lambda m: lookup.get(m.group(1), ""), s)

The point to using lookup.get with an empty-string default is to slavishly
mimic Perl's behavior of "erasing" the entire match string if lookup doesn't
contain the key.  I would much rather see an exception raised in that case,
and only partly because Python would prefer to raise one .

Over the long term, it's more fruitful to write Python from the start than
to mimic dubious Perl features.  For example, the data you're starting with
probably didn't come with @@xyz@@ thingies naturally, but that notation was
invented because someone knew it would be easy to substitute using Perl
s///g.  If the project started in Python, you'd be more likely to use
%(xyz)s notation in the data, so that substitutions could be done simply and
much more quickly via Python's

    t = s % lookup

idiom.  Regexps are not Python's forte, and all of speed, clarity and
maintainability will increase in proportion to your zeal in purging them.

so-ends-chapter-1-of-the-what-not-to-cook-book-ly y'rs  - tim





From pinard at iro.umontreal.ca  Mon Sep 13 17:37:53 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 13 Sep 1999 17:37:53 -0400
Subject: List to string?  Docstrings?
Message-ID: 

Hello, people.  Still learning Python.  I wrote around 6000 Python lines
in the last 10 days! :-)  I have two quick questions for you.

1) For converting a string into a list, I used:

        chars = list(text)

   For getting back the string from the list, I tried:

        import operator
        text = reduce(operator.concat, chars)

   but still, I wonder if you would not have a better suggestion...

2) I still miss the purpose of __doc__ strings.  It's true that I can write:

        print module.__doc__
        print function.__doc__

   and sometimes get useful information from these.  But what would I be
   loosing, for my own things, if I was using mere `#' comments instead?

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From donn at u.washington.edu  Fri Sep 24 14:28:19 1999
From: donn at u.washington.edu (Donn Cave)
Date: 24 Sep 1999 18:28:19 GMT
Subject: SIGCHLD, fork, listen
References: <37EC1ACA.8963554B@hk.rfa.org> <7sd6q4$gqq$1@salome.rfa.org> <7sdqfn$m0o$1@nntp6.u.washington.edu>
Message-ID: <7sgfs3$ejg$1@nntp6.u.washington.edu>

Quoth Bill Eldridge :
| Another option would be to use inetd, but I need to
| do some host/address verification, and couldn't figure
| out how to latch onto that kind of socket info from
| a new spawned inetd handler using Python.  Ideas?

Would this work?

    sock = socket.fromfd(0)
    print sock.getpeername()

inetd should give you the socket descriptor on unit 0.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From srenner at lycosmail.com  Tue Sep  7 11:11:53 1999
From: srenner at lycosmail.com (srenner at lycosmail.com)
Date: Tue, 07 Sep 1999 15:11:53 GMT
Subject: Suitability for games?
References: <37D5086C.69204F4C@jonjh.freeserve.co.uk> <7r2vlu$68m$1@newshost.accu.uu.nl>
Message-ID: <7r39ur$kdt$1@nnrp1.deja.com>

 This is the general
> principle -- Python can still perform quite well as a higher level
> controlling system (which can include a lot -- GUI, loading/saving,
> scripting of in-game objects and events, etc), even when it's not
> fast enough to do low level graphics manipulation. You use a C or C++
> library for that, but you can wrap those with Python.


"A C or C++ library" meaning Mesa or another OpenGL implementation, if
you want to use hardware accelerated graphics. PyOpenGL is available and
being maintained actively by David Ascher. I help maintain the demo
directory. If anyone has trouble installing or running the module, I'll
try to help -- mail me.

>
> For object scripting, there are some interesting developments. Python
does
> not have 'simulated threads' by default. [ Discussion of threads]
> So I'd recommend Python for game scripting. . .

Well, threads are interesting, but the rendering has to all happen in a
single process --  one can't very well have multiple streams of OpenGL
calls being shuffled together. The new nvidia chip has 4 pixel
pipelines. Presumably each pipeline has a separate context (state
machine) and so the chip could be driven by 4 renderering threads. But
you would want to balance the loads so that each pipeline was doing
about the same amount of work per frame. It's not like Guido's Brownian
motion demo with one thread per 'object', because the objects will have
to be in heirarchies for any reasonably sized and detailed scene. Now,
one might have two phases, rendering and 'world update', with objects
updating themselves in threads, but the rendering can't be split. As
long as the update doesn't step on the renderer . . .

Actually that sounds good. Watch the PyOpenGL demo directory for an
implementation.

sr


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From olipt at mayo.edu  Thu Sep 30 16:04:08 1999
From: olipt at mayo.edu (Travis Oliphant)
Date: Thu, 30 Sep 1999 15:04:08 -0500
Subject: Memory Error
In-Reply-To: <37F385AC.7057E983@v1.wustl.edu>
References: <37F385AC.7057E983@v1.wustl.edu>
Message-ID: 

> Hi,
> 
> I'm passing fairly large arrays around (using Numeric Py)
> and I'm getting the error:
> 
>         MemoryError: can't allocate memory for array
> 
> although I have plenty of memory left on my SGI (Indigo2
> with 448 megs). Why would I be seeing this message although
> the machine hasn't even starting swapping?   Is there a default
> stack size I can adjust somehow?
> 

I don't know exactly what's happening for you here.  But, let's
investigate.

I'll assume you are using NumPy 11.

A grep through the source shows that this error is used three times in
arrayobject.c:

All three times it is used is in PyArrayFromDimsAndData when malloc fails
while trying to allocate memory for the dimensions, or the strides, or the
data from the heap.  So, I don't think it is a stack-size question unless
malloc works in an unfamiliar way on an Indigo2.

Most likely it is failing when allocating data memory based on the
dimensions variable based to the C-routine PyArrayFromDimsAndData.

To further analyze your problem, I will need to know what command you are
using to get the failure you describe since PyArrayFromDimsAndData is used
several times in the source.

Sorry I can't be of more help.

Travis
 






From pinard at iro.umontreal.ca  Tue Sep 14 21:05:52 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 14 Sep 1999 21:05:52 -0400
Subject: timesync with python?
In-Reply-To: "Fredrik Lundh"'s message of "Thu, 9 Sep 1999 12:01:50 GMT"
References: <37D66AF3.6A3E96F7@phoenix-edv.netzservice.de> <002701befabb$1a884880$f29b12c2@secret.pythonware.com>
Message-ID: 

Holger Jannsen  wrote:

> But I also thought that someone has an idea about that before and did
> it likely better than me. So, do you know a weblink to a script which
> already nearly does what I need? Perhaps directly with interrogation at
> port 37 NTP?

"Fredrik Lundh"  writes:

> from the eff-bot archives:

> import socket, time
> HOST = "my.server"
> PORT = 37
> TIME1970 = 2208988800L # 1970-01-01 00:00:00
> s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> s.connect(HOST, PORT)
> b = s.recv(4)
> t0 = ord(b[3]) + (ord(b[2])<<8) + (ord(b[1])<<16) + (ord(b[0])<<24L)
> t1 = int(time.time()) + TIME1970
> dt = int(t0 - t1)
> print "delta is", dt, "seconds"

Isn't that fun?  I just coded this, a few days ago.  My formulation is
almost identical, yet a tiny bit cleaner, maybe:


import socket, struct, time

def time_delta(machine):
    client = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    client.connect(machine, socket.getservbyname('time', 'tcp'))
    data = client.recv(4)
    client.close()
    remote_secs = int(struct.unpack('!I', data)[0] - 2208988800L)

    local_secs = int(time.time())
    return remote_secs - local_secs

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From hnowak at cuci.nl  Tue Sep  7 17:00:44 1999
From: hnowak at cuci.nl (Hans Nowak)
Date: Tue, 7 Sep 1999 21:00:44 GMT
Subject: Snippets site has moved!
Message-ID: <199909072102.XAA20264@dionysus.fw.cuci.nl>

Due to problems with my previous host (hvision.nl), which has proven 
to be more than a little unstable, I deciced to move my snippets 
site. The new URL is 

http://tor.dhs.org/~zephyrfalcon/snippets/ 

Thanks to Blake Winton for offering me the account and the web space. 
(Lots of people suggested the Python Starship as the new snippet 
home, but I am not a PSA member. :-/ )

I will remove the HTML from the old address as soon as possible. By 
the way, I haven't been able to add any new snippets yet (counter is 
currently at 130). I do have a lot of them on my harddisk, 
unprocessed, some of which I hope to add soon. New snippets are 
always welcome, by the way. 

One more thing: The site still mentions hey_psst at usa.net as the email 
address to contact me about snippets, and while this address is still 
valid, I do not check it too often (due to my slow Internet 
connection; usa.net isn't that fast). You will get an answer more 
quickly if you contact me at zephyrfalcon at hvision.nl. I will get rid 
of the usa.net address in the near future.


--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/



From aahz at netcom.com  Sun Sep 19 12:34:07 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Sep 1999 16:34:07 GMT
Subject: Marking translatable strings
References: 
Message-ID: <7s339v$d78@dfw-ixnews3.ix.netcom.com>

In article ,
=?ISO-8859-1?Q?Fran=E7ois_Pinard?=   wrote:
>
>So, I want the big picture right now.  That is: a technique for marking
>strings for automatic extraction and building of PO files, and a technique
>for using PO files from within Python scripts.  I foresee that Python
>introduces an usual difficulty in that the textual domain for translations
>may vary quite unexpectedly, when the control dynamically flies between
>independent packages under different textual domains.

Y'know, it might help those of us with an interest in I18N but little
actual experience if you posted some resources that gave information
about what the heck a "PO file" is and how one is suposed to generate
and access it in other languages.  (Yes, I could search the web, but
it'd be difficult for me to determine which sites are the most useful.)

Personally, if you want to accomplish anything useful in the 1.6
timeframe, I think you'll be best off creating some sort of I18N module
or package and constructing an automatic approach that relies on strings
being used through this package if they are intended to be
internationalized.

This does ignore the big problem of doc strings, but IMO doc strings
should be used more by programmers than end-users -- I *don't* like them
for printing output.  In any event, I don't think you can solve this
problem in the pre-2.0 timeframe, though it's certainly worth laying the
groundwork now.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From aahz at netcom.com  Tue Sep  7 20:49:22 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Sep 1999 00:49:22 GMT
Subject: How to get the python sourcecode line, when debuging the Python C sources?
References: <7r3bqk$hbv$1@newshost.germany.net> <7r4alh$7gl$1@m2.c2.telstra-mm.net.au>
Message-ID: <7r4bqi$jds@dfw-ixnews11.ix.netcom.com>

In article <7r4alh$7gl$1 at m2.c2.telstra-mm.net.au>,
Mark Hammond  wrote:
>Stefan Knappmann <101.139915 at germanynet.de> wrote in message
>>
>>Does anybody know in which C variable I can find the Python Stack?
>>Or how to get the activ Python source code line from within the C
>>debugger?
>
>Move down the stack to eval2() in ceval.c.  f->f_lineno will tell you the
>line number.  f->f_code->co_filename + 20 bytes is the start of the ASCII
>string containing the filename.  Adding something like "(char *)(((char
>*)f->f_code->co_filename))+20)" to the watch window should do the trick.

And people wonder why I think C is a dead language.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From cjc26 at nospam.cornell.edu  Wed Sep  1 22:10:32 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 2 Sep 1999 02:10:32 GMT
Subject: All rise for judge Timothy Peters...
References: <000201bef43e$b8af52c0$df2d153f@tim>
Message-ID: 

Note how the Timbot tries to distract us from the fact that no human
records of him exist.

but-the-mobile-exoskeleton-should-be-ready-real-soon-now-ly yr's, Cliff


Pada Wed, 1 Sep 1999 01:56:23 -0400, Tim Peters bilang:
| You be the judge!  Here are some pictures of me from last year:
| 
|     http://www.puer.to/bae/p/txtipe01.htm
| 
| I turned to singing after the embarrassment of having been picked in the 26th
| round of the 1995 June draft by a Minnesota Twins farm team:
| 
|     http://www.miraclebaseball.com/pete19.html
| 
| All in all, becoming a judge was the best career move I made since I got fired
| for that "New Coke" disaster.
| 
| thanks-for-caring-and-they're-guilty-as-charged-ly y'rs  - tim


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From fredrik at pythonware.com  Fri Sep 17 09:37:40 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 17 Sep 1999 13:37:40 GMT
Subject: cgi post question
References: 
Message-ID: <006401bf0111$f2487590$f29b12c2@secret.pythonware.com>

Felicia Neff  wrote:
> I am trying to write a CGI script that takes input from a POST, modifies
> some of the input and then POSTs that info to another CGI script.  I know
> that perl has a function that will do this.  Is there such a function in
> python (and if it is do_POST() from the CGIHTTPServer module, is there any
> more documentation besides that at
> http://www.python.org/doc/current/lib/module-CGIHTTPServer.html?)?

CGIHTTPServer is a web server implementation.  you're
probably more helped by the urllib client library:
http://www.python.org/doc/current/lib/module-urllib.html

import urllib
...
result = urllib.urlopen(myurl, mydata)

(currently limited to application/x-www-form-urlencoded
posts, by some strange reason).

hope this helps!






From bernhard at alpha1.csd.uwm.edu  Wed Sep  8 14:52:00 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 8 Sep 1999 18:52:00 GMT
Subject: Programmers Editor
References: <37D5916B.606E@zeusedit.com>  <00da01bef9e0$6e1b5c70$f29b12c2@secret.pythonware.com>
Message-ID: 

On Wed, 8 Sep 1999 09:52:14 GMT, Fredrik Lundh  wrote:
>Bernhard Reiter  wrote:

>    The gist of it is that Python is absolutely free, even
>    for commercial use (including resale).
>and the newsgroup has always reflected that, as
>long as the emphasis has been on Python.
That's fine. IMO the emphasis was the last post was not entirely on python.
(see my other post.)

>but to help them, it's probably better to put up
>a site somewhere comparing different tools (or
>just pointing to them)
That would be nice.
As an FAQ: Which (free software?) editors or development tools are
known to be used by a lot of python developers.
Or are suitable/integrated with python on Unix/Mac/Win.

It is just good to know which editors can do python syntax highlighting.
It might be too much work, too. Next time I shall save my bytes for
more hacking. ;-)

	Bernhard
-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.    (www.uwm.edu/~bernhard)
Association for a Free Informational Infrastructure              (ffii.org)



From philh at vision25.demon.co.uk  Sat Sep  4 21:30:32 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 05 Sep 99 01:30:32 GMT
Subject: Pirate -> RAD
References: <7qno38$6uk$1@news.is.co.za> <37D17998.5CF4@warande.uu.nl>
Message-ID: <936495032snz@vision25.demon.co.uk>

In article <37D17998.5CF4 at warande.uu.nl> M.Oti at warande.uu.nl "Martin Oti" writes:
> 
> PS3: Pirate and Parrot. How ironic. ;) Is this a coincidence? 

Yes, AFAIK. Parrot is named after the Parrot Sketch.

-- 
Phil Hunt - - - - - - - - -  philh at vision25.demon.co.uk
 - Linux will be 8 years old on 17th September! See: -
http://www.vision25.demon.co.uk/prog/linuxbirthday.html




From aahz at netcom.com  Sat Sep  4 13:22:12 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 4 Sep 1999 17:22:12 GMT
Subject: "if" syntax error
References: <37D138C1.3EABC811@dialnet.net>
Message-ID: <7qrkg4$6g1@dfw-ixnews21.ix.netcom.com>

In article <37D138C1.3EABC811 at dialnet.net>,
Andrew Kesterson   wrote:
>
>CGIHeaders = cgi.FieldStorage()
>if CGIHeaders["Listing_Type"].value == "Residential":
>
>when I compile it, I get this error:
>
>  File "C:\Program Files\Python\Source\Skydev\addlisting.py", line 82
>     if CGIHeaders["Listing_Type"].value == "Residential":
>      ^
> SyntaxError: invalid syntax

Is this the actual indentation within the program?  I ran into a similar
problem yesterday (not in *my* code ;-) that essentially looked like
this:

class foo():
  def a():
    pass
def b():
    pass
  def c():
    pass

Note that "def b()" is outdented an extra level.  You can also run into
similar problems if you mix tabs and spaces.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From bernhard at alpha1.csd.uwm.edu  Thu Sep  9 14:42:51 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 9 Sep 1999 18:42:51 GMT
Subject: Programmers Editor
References: <37D5916B.606E@zeusedit.com>  <00da01bef9e0$6e1b5c70$f29b12c2@secret.pythonware.com> <3dg10p1b5j.fsf@amarok.cnri.reston.va.us>
Message-ID: 

On 08 Sep 1999 15:00:24 -0400, Andrew M. Kuchling  wrote:
>Just follow up to this posting with notes on
>configuring your editor of choice.

vim after version 5.x comes with python syntax highlighting rules,
just open a file with ".py" ending and vim will know that it is python.

You might need to give vim your background complextion and turn syntax
on though.

The commands to turn it on usually go in the ~/.vimrc
note that I have the background set to dark, too. (light is the other option.)

" Switch syntax highlighting on, when the terminal has colors
" Also switch on highlighting the last used search pattern.
if &t_Co > 2 || has("gui_running")
  set background=dark
  syntax on
  set hlsearch
endif

If you use the terminal program version of vim (called vim) your terminal
has to do color and the TERM entry and terminfo/termcap information have
to be correct of course. The GUI versions (gvim) can do color themselfs.

If you compile vim you can disable syntax highlighting.


Someone should elaborate on the tab use in vim a bit.
	Bernhard

-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.    (www.uwm.edu/~bernhard)
Association for a Free Informational Infrastructure              (ffii.org)



From lmariusg at ifi.uio.no  Wed Sep 22 12:47:28 1999
From: lmariusg at ifi.uio.no (Lars Marius Garshol)
Date: 22 Sep 1999 18:47:28 +0200
Subject: wxPython on Debian Linux
Message-ID: 

Does anyone know how to find this? 

Searches on deja.com, debian.org and the wxPython site have so far
been unsuccessful. (Yeah, I've tried what it says on the site, and
also searched in dselect, all to no avail.)

Any help would be most gratefully received.

--Lars M.



From eugene.leitl at lrz.uni-muenchen.de  Fri Sep 10 14:20:23 1999
From: eugene.leitl at lrz.uni-muenchen.de (eugene.leitl at lrz.uni-muenchen.de)
Date: Fri, 10 Sep 1999 11:20:23 -0700 (PDT)
Subject: [MMTK] ANN: Biopython.org (Call for Participation)
Message-ID: 

From: Jeffrey Chang 

The Biopython Project (http://www.biopython.org/) is a new open
collaborative effort to develop freely available Python libraries and
applications that address the needs of current and future work in
bioinformatics, including sequence analysis, structural biology, pathways,
expression data, etc.  When available, the source code will be released
as open source (http://www.biopython.org/License.shtml) under terms
similar to Python.

This is a Call for Participation for interested people to join the
project.  We are hoping to attract people from a diverse set of
backgrounds to help with code development, site maintenance, scientific
discussion, etc.  This project is open to everyone.  If you're interested,
please visit the web page, join the biopython mailing list, and let us
know what you think!

Jeffrey Chang 
Andrew Dalke 




_______________________________________________
mmtk maillist  -  mmtk at starship.python.net
http://starship.python.net/mailman/listinfo/mmtk




From MHammond at skippinet.com.au  Fri Sep 10 20:56:44 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 11 Sep 1999 10:56:44 +1000
Subject: ASP with Python question
References: <37d87ed7@news.interq.or.jp> <7radmm$euq$1@m2.c2.telstra-mm.net.au> <37d8c687@news.interq.or.jp>
Message-ID: <7rc9d2$ch6$1@m2.c2.telstra-mm.net.au>

Glen Malley wrote in message <37d8c687 at news.interq.or.jp>...
>Mark Hammond  wrote:
>>
>>>It would really suck if the Session was the only object
>>>not working, because it's the only object I need...;-P
>>
>>Im afraid it really sucks :-)  This is the only known bug in the Active
>>Scripting support :-(
>
>Hrm, color me an unhappy camper..;-P
>
>Guess I'll have to figure a better way out to solve my
>problem. I'm trying to incorporate an interface to Mailman
>into our existing ASP-based management system.

Well, you could help me to fix it :-)  I havent even investigated this
problem.  Im happy to fix it, but just need some prodding to get it to the
top of my list.  Someone making an effort to help track this down always
prompts me into action.

Also, helping an ASP interface to mailman get running sounds worthwhile.
[Next we need mailman running as a true NT service :-]

Mark.






From MHammond at skippinet.com.au  Sat Sep 11 23:10:27 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sun, 12 Sep 1999 13:10:27 +1000
Subject: ASP with Python question
References: <37d87ed7@news.interq.or.jp>
Message-ID: <7rf5ke$3c1$1@m2.c2.telstra-mm.net.au>

Glen Malley wrote in message <37d87ed7 at news.interq.or.jp>...
>
>Hi, I've got a question regarding ActiveX Scripting
>with Python. I'm having no trouble using the Request
>object or the Response object, but for some reason
>I cannot figure out how to properly use the Session
>object.

Ive had a look at this.  Unfortunately the only IIS server I have handy is
IIS5 that comes with Windows 2000.

I can make session objects work OK by using an undocumented method:
-- some asp file.
<%
Session.SetValue('Key', 4)
%>

My custom session variable is <%= Session("Key") %>
-- end of asp file

Works for me.  Note that "SetValue" is needed to set a value, and
"Session(Key")" is a short-cut to "Session.Value("Key")".  Note that both of
these are methods - hence you need to pass the value as an arg.

There are special hacks in the COM package to recognize "Value", and allow
it to be used using call syntax (ie, "Session("Key")").  However, as Python
does not support assignment to a function call, it is not possible to make
the syntax "Session("Key") = value" work - so you are stuck with "SetValue"

I would appreciate it is someone can tell me if this also works with IIS4
under NT4.

Mark.






From aa8vb at yahoo.com  Fri Sep  3 09:57:42 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 3 Sep 1999 09:57:42 -0400
Subject: Programming the serial ports in unix
In-Reply-To: <7qmlev$1hf2$1@nnrp01.ops.uunet.co.za>; from Alan Langman on Thu, Sep 02, 1999 at 10:14:25PM +0100
References: <7qmlev$1hf2$1@nnrp01.ops.uunet.co.za>
Message-ID: <19990903095741.A924529@vislab.epa.gov>

Alan Langman:
 |Hi
 |
 |I wanting to write some python code to chat to my serial port. Does anyone
 |know of some unix examples, modules etc.
 |Anyhelp would be great appreciated.

You might check out:

   http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=515106896
   http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=509452346

Randall



From pinard at iro.umontreal.ca  Sat Sep 18 20:31:08 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 18 Sep 1999 20:31:08 -0400
Subject: Marking translatable strings
In-Reply-To: andy@robanal.demon.co.uk's message of "Sat, 18 Sep 1999 22:54:25 GMT"
References:  <37e708ce.4562650@news.demon.co.uk>
Message-ID: 

andy at robanal.demon.co.uk (Andy Robinson) ?crit:

> >So, I want the big picture right now.  That is: a technique for
> >marking strings for automatic extraction and building of PO files,
> >and a technique for using PO files from within Python scripts.

> Big picture: are you looking for run-time translation, or 'build-time'?

Run time, of course.  This is only at run time that a program knows which
language the user wants.

However, before this happens, translatable strings must be extracted,
distributed to national teams which orchestrate the work with translators,
and the resulting translation files made available (through the usual
channels), automatically validated, canonicalised, cleared with the team,
and checked for legalese (some packages requiring some, especially when
the FSF is involved).  The big picture is to get all this into movement.
It goes beyond "just" adding internationalisation mechanics in programs.

> i.e. should a Python application be able to switch languages in mid-run
> when a user hits a menu item?

Yes.  This almost never occurs in the cases I observed until now, but at
least in theory, all the mechanics is already in place within `gettext'.

> Also do you want an automatic and theoretically perfect tool to understand
> every context in which quotes occur and do the right thing, or a practical
> one to reduce the cost of internationalizing apps by 90%.

I want something not far from perfect, but it cannot be fully automatic.
Which strings are translatable or not are ultimately decided by a human,
helped by heuristics so string marking is done faster.  Translations are
fully done by humans (automatic translation is not quite successful).

> [...] so I could use the English messages as the key, but then hit a
> situation where a simple English message was used twice in two different
> contexts.

This is the usual problem with this approach, yet the problem occurs
rarely in practice.  It is often (but not always) resolved easily by
using longer strings, instead of building strings with translated inserts
(and moreover, this is more painful for translators, languages have
widely varying syntaxes).  When this happens, the simplest is to fake
two different strings for the speciality cases, and provide an English
translation file containing only the problematic strings to recover the
ambiguous English writing.  Ulrich has other plans for making this more
transparent, but we are not fully decided yet, this is not implemented.

The other approach is using numbers, or identifiers for each string.
It does not have the problem you mention, but it has other problems, and
in my own opinion, adding everything, this is worse overall, on average.
But some people are proponent of this one, and religious fights quite
easily pop up between the two approaches.  People are extremely touchy when
their own language is involved.  Some English speaking people, strangely,
are prone to think they understand all about others' linguistic problems :-).

> In Python, I'd be tempted to have a special tag to use at the
> beginning of the string I wanted internationalized, say
> "@@@USER_FORGOT_THE_FLOPPY".

Having the tag within the string has the disadvantage that even English
requires processing, and that the overall program looses in legibility, which
is a serious drawback.  That's why I'm looking for tagging methods leaving
the original string intact, and if possible, the language intact as well.

> Then I'd write a script to walk through the project finding these,
> and on each occurrence (a) replace it with a function to do a run-time
> lookup into a constants database, or a constant in a module, and (b)
> add the string to my database or the source of my constants module.

It is better to aim managing translations on a wider scale, for a lot
of packages and languages, with common tools and specialised teams.
This is the purpose of the PO file format, and the Translation Project.
If each language, or even each package, was seeking its own implementation,
it makes things more complex overall, in the big picture.  On the other
hand, many years in that area taught me that it is fairly difficult to
get everybody to agree and collaborate.  Nowadays, with most maintainers,
everything goes rather easily, while only a few are ready to argue endlessly.

At the beginning, it was really, really exhausting: nobody was agreeing to
nothing, and we had strong interference with negative effects.  We had to
start somewhere, we could just not aim everything at once.  This is easier
on me now, because the project is well started, and is slowly growing.
Some initial choices are set and most people collaborate at following them.
As things will get more solid and universal, we will dare further steps.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From pinard at iro.umontreal.ca  Fri Sep 24 10:40:56 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 24 Sep 1999 10:40:56 -0400
Subject: Becoming root within a Python script
In-Reply-To: Paul Foley's message of "25 Sep 1999 01:01:08 +1200"
References:  <871zbptasz.fsf@baal.winnegan.fake>  
Message-ID: 

Paul Foley, as many others, replied to my initial question.  I would like
to thank you all, without discrimination.  Your replies were thoughtful
and gave me many interesting insights on this matter.  Keep happy, all!

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From madq968 at djeksta.comNOSPAM  Wed Sep 29 12:57:01 1999
From: madq968 at djeksta.comNOSPAM (madQ)
Date: 29 Sep 1999 16:57:01 GMT
Subject: Download Ia.n.i.!!! It's free!
Message-ID: <7stgct$d9k$160@nslave1.tin.it>

Download Ia.n.i. RemoteControlSystem 1.2 beta. It's free!!!
New site: http://jump.to/IaniProject




From fdrake at acm.org  Thu Sep 23 11:13:11 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 23 Sep 1999 15:13:11 GMT
Subject: Postscript printing in Python
In-Reply-To: <37E98032.FA3B77F5@ans.com.au>
References: <37E98032.FA3B77F5@ans.com.au>
Message-ID: <14314.17287.318075.418034@weyr.cnri.reston.va.us>

Clement writes:
 > I need to do a mail merge to a postscript printer.  Does anyone know of
 > any tools that is available to the Python community?   
 > 
 > I checked the libraries and other Python documentation and failed to
 > find any help on printing to a postscript printer!

  It's not clear what you want.  Can't you just open a connection to
the printer and write to it?  The API will differ by platform; on Unix 
you'd do something like:

      import os
      printer = os.popen("lpr", "w")

      printer.write("%!PS-Adobe-2.0\n")
      ...
      printer.close()

  I imagine you can get some sort of handle to a printer on Windows
that allows similar operations.
  On the other hand, you may want something a little higher level.
Grail (http://grail.python.org/) includes support for printing to a
PostScript printer (Unix-style) via an extended "writer" object; see
the formatter module in the standard library for a description of the
interface.  It's pretty easy to use, but may not be what you're
looking for.  Adapting it to work on Window shouldn't be too hard if
you can get a way to send the PostScript to the printer yourself.


  -Fred

--
Fred L. Drake, Jr.	     
Corporation for National Research Initiatives



From fredrik at pythonware.com  Thu Sep 23 10:44:03 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 23 Sep 1999 14:44:03 GMT
Subject: VC++ grrhhh
References: 
Message-ID: <00c801bf05d2$15a81030$f29b12c2@secret.pythonware.com>

Robin Becker  wrote:
> anyone know how to get round nonconstant initialiser errors from VC++
> with the construction
> 
> statichere PyTypeObject Arraytype = {
>         PyObject_HEAD_INIT(&PyType_Type) // error here
>         0,
>         "array",

http://www.python.org/doc/FAQ.html#3.24






From hniksic at srce.hr  Wed Sep 15 20:11:44 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 16 Sep 1999 02:11:44 +0200
Subject: More random python observations from a perl programmer
References: <000401beff39$efb9c2e0$612d153f@tim>
Message-ID: <87k8prr9z3.fsf@pc-hrvoje.srce.hr>

"Tim Peters"  writes:

> [Hrvoje]
> > My point exactly.
> 
> What, you're opposed to convenience ?

When it provides easy access to something potentially dangerous and
rarely useful, I am.  Obviously, you disagree that this is the case
with input().

> Python is used by all sorts of people for all sorts of things, and
> not every app (not even most Python apps, I'd wager) need to worry
> about inventively malicious users.

It's not about worrying about inventively malicious users, but about
being aware of what's going on.



From steffen at cyberus.ca  Mon Sep 13 07:51:04 1999
From: steffen at cyberus.ca (Steffen Ries)
Date: 13 Sep 1999 07:51:04 -0400
Subject: [NEWBIE]:  Tk icon
References: <7rhsk2$gbi$1@msunews.cl.msu.edu>
Message-ID: 

"Shannon Watters"  writes:

> How can I get rid of the Tk logo in the top left corner of a Tk(inter)
> window?  It's 'tacky'.  I don't use that word much but it's the best one
> that came to mind when I was staring at it up there.  It's making my highly
> complex "Hello world" button look unprofessional (<-spelled?).

don't know, if this works on WINx, but to set the icon in Tkinter, you
do something like:
----8<----
from Tkinter import *
root=Tk()
icon_image = PhotoImage('icon.gif')
icon = Toplevel()
Label(icon, image=icon_image).pack()
root.iconwindow(icon)
----8<----

hth,
/steffen
-- 
steffen at cyberus.ca		<> Gravity is a myth -- the Earth sucks!



From ransen at nemo.it  Tue Sep  7 21:30:53 1999
From: ransen at nemo.it (Owen F. Ransen)
Date: Wed, 08 Sep 1999 01:30:53 GMT
Subject: Embedding Python into an MFC app - thread problems.
References: <7pv1jd$gmd@newsops.execpc.com> <37d78aef.6768198@news.newsguy.com> <003101bef93d$18876400$f29b12c2@secret.pythonware.com>
Message-ID: <37d6b14e.2576188@news.newsguy.com>

On Tue, 7 Sep 1999 14:27:19 GMT, "Fredrik Lundh"
 wrote:

>Owen F. Ransen  wrote:
>> Er, where did you get the Pythonwin embedding documentation?
>> I've got the DrDobbs Python Resource CD-ROM and can find no
>> real help in embedding Python in Windows apps. Any pointers
>> much appreciated...
>
>here's one:
>    http://www.python.org/doc/current/
>check the "Extending and Embedding" and
>"Python/C API" documents.

Thanks. This one has some Windows specific hints, which
the ones I saw before did not have.

Owen


--------------------------------------------------
Owen F Ransen                       ransen at nemo.it
http://www.ransen.com       the home of Repligator 
Winner SIAF awards - Best Graphics Program of 1999



From aahz at netcom.com  Mon Sep 27 20:57:17 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 28 Sep 1999 00:57:17 GMT
Subject: keys and values lining up
References: 
Message-ID: <7sp3pd$r0p@dfw-ixnews5.ix.netcom.com>

In article ,
Nathan Clegg   wrote:
>
>If d is a dict, am I guaranteed that d.keys() and d.values() will line up,
>in proper order, if called successively with no modification in between? 
>That is, is the following safe?
>
>k,v = d.keys(), d.values()

How about d.items()?
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From pinard at iro.umontreal.ca  Sat Sep  4 10:26:59 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 04 Sep 1999 10:26:59 -0400
Subject: mailman and spam? :-) [was: Re: 18 AND OVER 59713]
In-Reply-To: actgal8@aol.com's message of "Fri, 03 Sep 1999 17:09:53 -0600"
References: <030999170953@aol.com>
Message-ID: 

actgal8 at aol.com writes:

> free xxx pics:
> http://207.240.225.250
> 5LwxM\qH

I've seen some SPAM on python-list recently.  Mailman is said to include
some anti-spam features, and python-list is said to be an example showing
how well Mailman behaves.  So, spammers are challenging Mailman, here? :-)

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From davidopp at megsinet.net  Fri Sep  3 01:00:15 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Fri, 03 Sep 1999 01:00:15 -0400
Subject: OH, I'm A Lumberjack And I'm OK...
References: <37CE2648.D4254FCC@megsinet.net> <37ceeef4.5698550@news.erols.com> 
Message-ID: <37CF55DC.816F38F5@megsinet.net>

... I play with Python all night and I sleep all day... lol

I received an e-mail message which told me that WIN ZIP sometimes deletes part
of the extension and to try adding ".tar" to the file or placing a "." before
the tar extension if it was missing.  This allowed me to extract the files
Robert Kern was recommending, but still the PIL doesn't work.

Again I will relate some hopefully relevant info in the hope (getting less
hopeful about this all the time) that someone will know what the $%@# is going
on and how to fix it so I can run files from the PIL.

Computer: Packard Hell P-75
OS: Windows 98
Python Version: Python for Windows 1.5.2 with all of the win32all - 125
extensions

PATHS     (yes, I took the suggestion to place my Python & related files in
pathways that have no spaces)

Path to Python executable: C:\Python
Path to Tcl executable: C:\Tcl\bin
Path to PIL: C:\Python\PIL
Path to Tkinter: C:\Python\Lib\lib-tk
Path to Painter (a program that I am more than curious to see work, I am almost
obsessed now): C:\Python\scripts\painter.py

My Traceback (error message):

>>> Traceback (innermost last):
  File "C:\Python\Pythonwin\pywin\framework\scriptutils.py", line 237, in
RunScript
    exec codeObject in __main__.__dict__
  File "C:\Python\scripts\painter.py", line 12, in ?
    import Image, ImageTk
ImportError: No module named Image

My Autoexec.bat settings:

PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\WINDOWS;
set PATH=%PATH%C:\mingw32\bin;
set PATH=%PATH%C:\PYTHON;C:\PYTHON\PIL;C:\TCL\BIN;C:\PYTHON\LIB\LIB-TK;

I promise this will be my last posting regarding this (I am sure that some of
you are as sick of seeing me post about this as I am frustrated by not being
able to get the PIL to work on my machine)  Thanks for all of the suggestions
and help.  I am certain that the level of support and helpful suggestions I have
gotten here are much greater than I would have received in the onion (mine)
fields.

Thanks,

David O.

Darrell wrote:

> As a check.
> Down loaded PIL_win32.tar.gz from this link and extracted it with WinZip 7.0
>
> --
> --Darrell
> Robert Kern  wrote in message
> news:37ceeef4.5698550 at news.erols.com...
> > On Thu, 02 Sep 1999 03:24:59 -0400, David Oppenheimer
> >  wrote:
> >
> > >I am going to give up on trying to get the PIL to work with Pythonwin
> > >1.5.2 unless someone who has it working with their setup can give me
> > >explicit instructions on how to do this.  I am unable to implement
> > >Robert Kern's instructions due to the fact that I cannot get my Win Zip
> > >program to unzip the tar.gz file format.
> >
> > This is probably a browser problem.  Clear your cache first.  Be sure
> > that the .gz and .tar extensions are associated with Winzip in your
> > browser.  Their MIME types should be "application/x-gzip" and
> > "application/x-tar", respectively.  At least, that's how I have them.
> >
> > Then try downloading again (use a shift-click from
> > http://starship.python.net:9673/crew/da/Code/PIL rather than a
> > straight left-click).
> >
> > >So ... time for me to put PIL for Pythonwin 1.5.2 on the pile of
> > >software corpses and set it ablaze.  Can you feel the heat?!  Anyone
> > >have any marshmallows?
> >
> > You'll be fine.  BTW, _Pythonwin_ is the name of Mark Hammond's IDE
> > for Python on Windows, not the general name for Python on Windows.
> >
> > >David O.
> >
> > Robert Kern           |
> > ----------------------|"In the fields of Hell where the grass grows high
> > This space            | Are the graves of dreams allowed to die."
> > intentionally         |           - Richard Harter
> > left blank.           |




From jon at rdt.monash.edu.au  Thu Sep 23 01:48:38 1999
From: jon at rdt.monash.edu.au (Jonathan Giddy)
Date: 23 Sep 1999 05:48:38 GMT
Subject: lists and append, and loop iteration
References: <37E8860B.C024C4F8@radiata.com>
Message-ID: <7scevm$hv0$1@towncrier.cc.monash.edu.au>

These must be some of the cool new control statements from a very early 
1.6 alpha release ;~)

Mark Krischer  writes:

>if i do:
>	for element in list:
>		do something with element


>i figure i can always do:
>	for i in range(len(list)):
>		do something with list[i]




From moshez at math.huji.ac.il  Thu Sep 16 02:14:48 1999
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 16 Sep 1999 08:14:48 +0200 (GMT+0200)
Subject: python call external program (WHAT THE?!)
In-Reply-To: <004b01beffc1$a4158f80$f29b12c2@secret.pythonware.com>
Message-ID: 

On Wed, 15 Sep 1999, Fredrik Lundh wrote:

> yet another Linux zealot wrote:
> > # use Linux instead of MS rubbish
> 
> looks like someone hasn't read the Linux Advocacy
> mini-HOWTO in quite a while.  

> and this news-
> group is open for all Python users.  if you have
> a problem with that, go start your own news-
> group.

While Fredrik has a point (too many Linux advocates, including sometimes
myself, get too heated about it), he has another point in his sarcastic
comment too -- many Python users are old-time UNIXers, who don't know and
don't care about such things as COM, OLE, Win32 API....

Should we consider forking the group, creating 

comp.lang.python.windows
comp.lang.python.misc
comp.lang.python.numeric (Now that the Matrix-SIG is more about using
                          NumPy rather then developing for it)
(And if anyone volunteers, 
comp.lang.python.moderated)

I've been thinking about this quite a while, but never got the nerve to
suggest it in public until I saw Fredrik's post.
--
Moshe Zadka . 
INTERNET: Learn what you know.
Share what you don't.




From vascog at uol.com.br  Thu Sep  2 20:29:20 1999
From: vascog at uol.com.br (Joao Neto)
Date: Thu, 2 Sep 1999 21:29:20 -0300
Subject: help about JPython
Message-ID: <7qn4fm$qas$1@fu-berlin.de>

Hi!

I download JPython but I'm unable to run the Python
Interpreter even on the simplest examples.

I'm using Win98 and IE4.0, I already insert the JP directory
into the PATH and CLASSPATH sets.

Let me give you and example:

=============== The program (that compiles ok)

import org.python.util.PythonInterpreter;
import org.python.core.*;
import java.applet.*;

class hello extends Applet {

public void init() {

PythonInterpreter interp = new PythonInterpreter();


PyObject x = interp.eval("2+2");
System.out.println("x: "+x);

}

}

================== when I run it, the java Console gives me this

java.lang.NullPointerException
 at org/python/core/PySystemState.
 at org/python/util/PythonInterpreter.
 at org/python/util/PythonInterpreter.
 at hello.init
 at com/ms/applet/AppletPanel.securedCall0
 at com/ms/applet/AppletPanel.securedCall
 at com/ms/applet/AppletPanel.processSentEvent
 at com/ms/applet/AppletPanel.processSentEvent
 at com/ms/applet/AppletPanel.run
 at java/lang/Thread.run
Microsoft (R) VM for Java (tm), 4.0 Release 4.79.0.2424

=============================

can anyone help me?

thanks in advance!!!

joao neto





From chad at vision.arc.nasa.gov  Sun Sep 19 18:35:19 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Sun, 19 Sep 1999 22:35:19 +0000
Subject: Redhat Python RPM
References: <19990916191315.A8674@quark.emich.edu> <99091617110409.02826@quadra.teleo.net> <19990916203830.C8674@quark.emich.edu> <37E1F783.2B7BB10A@vision.arc.nasa.gov> 
Message-ID: <37E56527.A260A6A2@vision.arc.nasa.gov>

Robin Dunn wrote:

> > First of all, I have sort of taken over the Python4Linux project
> > from Oliver while he is busy working on an open source Python
> > project.  I'm currently negotiating for some server space (my
> > ISP gives me 50 megs, which I'm rapidly using up) and more
> > bandwidth (my ISP serves its pages dead slow :)
>
> Why not use the Starship?  That's what it's there for, to serve the Python
> community.
>

Well, I signed up for an account, over a month ago, and still haven't
gotten one or any feedback about it.     I've been travelling and busy w/
other things so hadn't followed through on it since then.  Believe me,
I'd like to set it up there...  Christian, or anyone else who adminsters
these accounts, can you contact me?  I'll submit my application again,
perhaps it didn't make it.

 PSA membership renewal time is here as well...

Chad
chad at vision.arc.nasa.gov





From MHammond at skippinet.com.au  Mon Sep 13 18:23:59 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 14 Sep 1999 08:23:59 +1000
Subject: Python & ISAPI
References: <7rjs6r$ia5$1@gxsn.com>
Message-ID: <7rjtk5$63h$1@m2.c2.telstra-mm.net.au>

Not as far as I know.  You can use Python in ASP, which is generally a
better answer...

Mark.

Phil Harris wrote in message <7rjs6r$ia5$1 at gxsn.com>...
>Hi all
>
>Has anyone done anything with Python and ISAPI?






From loki at enteract.com  Sat Sep  4 21:30:44 1999
From: loki at enteract.com (loki)
Date: Sat, 04 Sep 1999 20:30:44 -0500
Subject: 1st issue boot mag
Message-ID: <37D1C7C4.23C7@enteract.com>

I have a 1st issue boot magazine for sale at:
http://www.enteract.com/~loki/boot.htm



From pinard at iro.umontreal.ca  Fri Sep 17 18:22:55 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Sep 1999 18:22:55 -0400
Subject: Reaching subtrees
Message-ID: 

Wow!  Re-wow!  Look:

>>> for (a, (b, c)) in [(1, ((2, 3), (4, 5))), ((11, 12), (13, 14))]: 
...	print a, b, c
... 
1 (2, 3) (4, 5)
(11, 12) 13 14
>>> def x(a, (b, c)):
...	print a, b, c
... 
>>> x(1, ((2, 3), (4, 5)))
1 (2, 3) (4, 5)
>>> x((11, 12), (13, 14))
(11, 12) 13 14

So, if I understand and believe what I see, it might be quite easy in
Python to reach precise subtrees of trees, given that skeletons match?
This, combined with `try:' statements, I guess one could fastly see is some
tree matches a fixed skeleton, and then rebuild it differently if it does.
Is this fully documented, and meant to be a dependable part of the language?

This feature pushes Python forward for elegant writing for simple, modest,
fixed transformational grammars, which is something I happen to often need.
This is worth more investigation.  Moreover, who knows, for non fixed
grammars, maybe there is still some way to reach the underlying mechanics in
the application?  (Short of using `exec', of course!)  For bigger grammars,
optimisation and preconditioning is usually necessary, but maybe Python
could be a nice vehicle for such needed preparation, itself, given the
underlying mechanics is later accessible on the result of the preparation.

Am I making sense to anybody?  Would someone know any Python work in that
direction which I could admire, then shamelessly use for inspiration? :-)

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From emile at fenx.com  Sat Sep  4 11:09:32 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 4 Sep 1999 08:09:32 -0700
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <935457373snz@vision25.demon.co.uk> <7qh7b6$g9g$1@wagner.wagner.home> <37CEFA88.430F28EA@compaq.com>
Message-ID: 

Greg Ewing  wrote in message
news:37CEFA88.430F28EA at compaq.com...
> Victor Wagner wrote:
> >
> > Phil Hunt  wrote:
> > : Use subclassing.
> > No, thanks. I've large experience of programming with toolkits where
you
> > need to create a subclass for each of hundreds dialogs in your
program.
>
> Note that, in Python, instead of subclassing you can
> plug functions into an instance of an existing class.
> Code calling it will never know the difference.
>
>

Greg,

Off topic and probably obvious, but, how would you plug a function into
an instance?

Hoping-to-say-WOW-but-anticipating-a-slapped-forehead-ly yrs

--

Emile van Sebille
emile at fenx.com
-------------------





From pinard at iro.umontreal.ca  Sun Sep 19 15:06:53 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 19 Sep 1999 15:06:53 -0400
Subject: Marking translatable strings
In-Reply-To: Skip Montanaro's message of "Fri, 17 Sep 1999 08:45:32 -0500 (CDT)"
References:  <000101bf00cf$801bd080$d12d153f@tim> <14306.17916.191557.237458@dolphin.mojam.com>
Message-ID: 

Skip Montanaro  writes:

>     Tim> I agree with everyone else  that the special meaning of
>     Tim> "_" in interactive mode is unlikely to create a problem for you.

> Am I the only person who occasionally pastes code from a Python file
> I'm editing into an interactive session to try it out in isolation?

After you email, I tried, hoping to raise your count from 1 to 1.5 (:-),
but pasting multi-lines did not work well for me.

Are you much using the interactive `_' when you paste?  Besides making
Python demonstrations to friends, I did not use `_' much myself so far...

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From tjreedy at udel.edu  Thu Sep 16 20:59:04 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 17 Sep 1999 00:59:04 GMT
Subject: Splitting c.l.py (was Re: python call external program (WHAT THE?!))
References:  <7rr088$74b@dfw-ixnews8.ix.netcom.com>
Message-ID: <7rs3oo$5jt$1@news.udel.edu>

Rather than splitting the group, what might be helpful are agreed-upon 
'designators' such as WIN, W95, WNT, UNX, TK consistently placed at the
head of subject lines when appropriate.

T J. Reedy




From news at dorb.com  Tue Sep 21 21:44:33 1999
From: news at dorb.com (Darrell)
Date: Tue, 21 Sep 1999 21:44:33 -0400
Subject: Marking translatable strings
References:  <7s339v$d78@dfw-ixnews3.ix.netcom.com>  <1aUF3.223$46.6340@typhoon.nyroc.rr.com> <7s94hn$rjj@dfw-ixnews3.ix.netcom.com>
Message-ID: 

Tweaked PyString_Format to do it. Posting can be motivational.

>>> d1={'AA':'aa'}
>>> d2={'BB':'bb'}
>>> d3={'CC':'cc'}
>>>
>>> s= fmt("%(AA)s %(BB)s %(CC)s",d1)
>>> print s
aa %(BB)s %(CC)s
>>> s= fmt(s,d2)
>>> print s
aa bb %(CC)s
>>> s= fmt(s,d3)
>>> print s
aa bb cc
--
--Darrell
Aahz Maruch  wrote in message
news:7s94hn$rjj at dfw-ixnews3.ix.netcom.com...
> In article <1aUF3.223$46.6340 at typhoon.nyroc.rr.com>,
> Darrell  wrote:
> >
> >Wish there was a dictionary formatter that wouldn't throw a keyError when
it
> >was missing a value. Then I could avoid the $& kluge. Better yet accept
an
> >error handler function. If the function isn't provided then throw the
error.
>
> How about using try/except?
> --
>                       --- Aahz (@netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6  (if you want to know, do some
research)





From fredrik at pythonware.com  Tue Sep 21 16:40:20 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 21 Sep 1999 20:40:20 GMT
Subject: what happened to PyAstro?
References: <7s86g7$ega$1@nnrp1.deja.com>
Message-ID: <011e01bf0471$868dd1c0$f29b12c2@secret.pythonware.com>

Marco  wrote:
> I'm quite a newbie in the python world, but with an interest in
> astronomy I was looking for some useful modules.
> there's the python & astronomy site, which speaks of a PyAstro module,
> but the link to that one is dead.
> Searching the python.org site, there is nothing to be found.

http://starship.python.net/crew/pavlos/

(in theory, the python.org website search should
cover the starship site, but it seem to have missed
this one...)

hope this helps!






From robin at jessikat.demon.co.uk  Thu Sep 23 10:17:18 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 23 Sep 1999 15:17:18 +0100
Subject: VC++ grrhhh
Message-ID: 

anyone know how to get round nonconstant initialiser errors from VC++
with the construction

statichere PyTypeObject Arraytype = {
        PyObject_HEAD_INIT(&PyType_Type) // error here
        0,
        "array",

when trying to compile as a separate module. Certainly PyType_Type is
defined in another dll, but it's just an extern right or has M$ started
fiddling with C semantics again? I have tried declaring as an import,
but that doesn't work either.
-- 
Robin Becker



From tim_one at email.msn.com  Thu Sep 30 22:32:06 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 30 Sep 1999 22:32:06 -0400
Subject: How Can I do vi-like substitutions?
In-Reply-To: <7t10nc$97k$1@thoth.cts.com>
Message-ID: <000601bf0bb5$277f3f20$322d153f@tim>

[Michel Huster]
> I want to do some complex replace inside python. I am used to vi
> power like:
>
> s/\([0-9]*\) at \([0-9]*\)/\2 AT \1/
>
> which will change
> 1234 at 5678
> into
> 5678 AT 1234
>
> I can only find re.sub(expr, repl, count) but this does not 'remember' the
> saves matches between arguments.

?

>>> import re
>>> re.sub(r"(\d*) at (\d*)", r"\2 AT \1", "1234 at 5678")
'5678 AT 1234'
>>>

read-the-docs?-ly y'rs  - tim





From da at ski.org  Thu Sep 23 01:33:07 1999
From: da at ski.org (David Ascher)
Date: Wed, 22 Sep 1999 22:33:07 -0700 (Pacific Daylight Time)
Subject: CVS tarballs available? Anyone want some?
In-Reply-To: <37E799E1.CFB9862@strs.co.uk>
Message-ID: 

On Tue, 21 Sep 1999, Ian Clarke wrote:

> Oops, found them on starship.  Still, a separate IDLE snapshot may be of
> interest to people...

I can add that to the list of tarball scripts.  Good idea. 

--david




From jam at quark.emich.edu  Wed Sep 29 00:07:05 1999
From: jam at quark.emich.edu (Jeff)
Date: Wed, 29 Sep 1999 00:07:05 -0400
Subject: Python TimeServer (RFC 868)
In-Reply-To: 
References: 
Message-ID: <19990929000705.A21607@quark.emich.edu>

On Wed, Sep 29, 1999 at 02:40:30AM +0000, Benjamin Schollnick wrote:
> 	I believe that I need to send it as a binary value (a 4 byte 
> integer), instead of a 
> 	string... But can't figure out a way to do this...?
> 

hmm.. well, if you are already opening the socket yourself, and using
'write' to send data to it, can't you just run 'long(foo)' and write that to
the socket instead of the text version? perhaps something with the 'struct'
module is in order.. you'll need to know the byte order (which the RFC
should give you) before mucking around with writing binary data.

just a suggestion.

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From fredrik at pythonware.com  Wed Sep  1 04:04:25 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 1 Sep 1999 08:04:25 GMT
Subject: And Now For Something Completely Similar !!!
References: <37CCDAF8.E79FE189@megsinet.net>
Message-ID: <028f01bef450$9d976ba0$f29b12c2@secret.pythonware.com>

David Oppenheimer  wrote:
> 1) The Image module is somewhere else and must
> be so defined in my path statement

the Image and ImageTk modules (as well as
the scripts/painter demo) are part of the
Python Imaging Library.

more info here:
http://www.pythonware.com/products/pil/






From greg.ewing at compaq.com  Thu Sep 30 23:26:18 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 01 Oct 1999 15:26:18 +1200
Subject: Python object persistency using pickle...
References: <37F06A5C.52ACEA97@vx.cit.alcatel.fr>
Message-ID: <37F429DA.667DA628@compaq.com>

David CROSSON wrote:
> 
>     pickle.dump(b,f3)
>     pickle.dump(a1,f1)
>     pickle.dump(a2,f2)

That doesn't work because pickle.dump creates a new Pickler 
instance every time you call it. To preserve references among 
objects, all the objects involve have to be pickled using a 
*single* Pickler instance.

One way to achieve that is to arrange for all your objects
to be reachable from a single "root" object and pickle that
object using a single call, e.g.

   pickle.dump((a1, a2, b), f1)

and to restore it,

   a1, a2, b = pickle.load(f1)

Another is to explicitly create your own Pickler instance:

   p = pickle.Pickler(f1)
   p.dump(a1)
   p.dump(a2)
   p.dump(b)

and similarly restore using a singe Unpickler:

  u = pickle.Unpickler(f1)
  a1 = u.load()
  a2 = u.load()
  b = u.load()

Both those methods will store all three objects together in
one file, which is probably desirable because they are so
closely coupled.

If you really want them in separate files, you might be able
to pull some trick like switching the file attached to the
Pickler instance in between calls, but I'm not sure what that
would gain you. You would still have to make sure you unpickled
them in exactly the same order that you pickled them, so you
might as well just put them in the same file.

If you want random access to pickled objects, the shelve module
might help -- although I don't know what it does with references
between separately shelved objects. Anyone else know more?

Greg



From aahz at netcom.com  Fri Sep  3 14:21:55 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Sep 1999 18:21:55 GMT
Subject: Returning none
References: <1275758005-72447341@hypernet.com> <5ORz3.14$JG1.2704@news.shore.net>
Message-ID: <7qp3k3$128@dfw-ixnews8.ix.netcom.com>

In article <5ORz3.14$JG1.2704 at news.shore.net>,
Michael P. Reilly  wrote:
>
>One powerful, often used construct is treating callables as first
>class: does the callable return a value? That doesn't matter, the
>interface needs to be consistant to work properly, and currently it
>does.  Do we now have new functions for apply, map, filter to handle
>"procedures"?  (Are these now procedures?)  You then have to know ahead
>of time if the callable is a function or a "procedure".  You can't mix
>at a non-predetermined point, functions and "procedures".  This just
>adds to the complexity of the language and implementations needlessly,
>removing important functionality.

One could just as easily make the same argument about using string
functions on numbers -- you have to know in advance that something is a
string.  Down this path lies the Perl approach....
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From philippe.boite at pacwan.fr  Sat Sep 18 08:46:24 1999
From: philippe.boite at pacwan.fr (Philippe Boite)
Date: Sat, 18 Sep 1999 14:46:24 +0200
Subject: Python or Java or "X" from me and my little job
References: <1275048881-10426114@hypernet.com> 
Message-ID: <7s01ki$37i$1@news4.isdnet.net>

Yes. Portability.

Les Schaffer  a ?crit dans le message :
m2n1utqnrm.fsf at netmeg.net...
> GMcM said:
>
> > Build in zlib (a matter of uncommenting some lines in Modules/Setup
> > on Linux) and it does.
>
> i'd prefer in these cases using calls to system functions 'tar' and
> 'gzip'. any advantage to using uiltin-in zip?
>
> les
>
> --
> ____        Les Schaffer              ___| --->> Engineering R&D <<---
> Theoretical & Applied Mechanics          |  Designspring, Inc.
> Center for Radiophysics & Space Research |  http://www.designspring.com/
> Cornell Univ.  schaffer at tam.cornell.edu  |  les at designspring.com





From m.faassen at vet.uu.nl  Sun Sep 26 04:25:15 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 26 Sep 1999 08:25:15 GMT
Subject: Dialect, A Python derived environment (and language ?)
References: <000001bf064f$ba05f980$cc2d153f@tim>
Message-ID: <7skl9b$fhg$2@newshost.accu.uu.nl>

Tim Peters  wrote:
> [Martijn Faassen]
>> ...
>> It's too close to Python in syntax for this to be an independent
>> development..

> What's so Pythonic about it?  They use double-underscore names for some
> system vrbls, but then C also reserves names of that form (where do you
> think Guido got the idea ?).  The specific names "__init__" and
> "__del__" are probably not coincidence, but they appear to be the only
> overlap.  The only other overt Pythonism I saw was using "and" and "or"
> instead of "&&" and "||".  Python may not have been the first language to
> use "if", "class" and "break" ...

I thought I saw them use { } syntax for dictionaries, and [] for lists, and
so on. That looked quite Pythonic to me. I didn't study it in that much
detail though, so I don't have much right to say anything. :)

[snip]
> I'm afraid Python's lawyers are far too busy suing Perl for borrowing
> Python's object model to hassle these poor folks for borrowing three pieces
> of Python syntax and ignoring all its semantics .  It may be more
> interesting to study the manual for choices they made Python *didn't*.

Whoops, I didn't intend to bash on it (that badly).

I just was rather surprised they didn't mention Python anywhere; to me it
looks far more like Python than like Basic, but then I mostly used Basic
in the line-number era (and some QuickBasic).

[snip]
> Take Dialect on its own terms, and it's a very interesting design point:
> the language and its libraries are both much smaller than Python, but with a
> little bit of GUI and a little bit of sockets and little bits of other
> things built in.  I find a lot to like in it, and rather than gripe about
> its overlap with Python, I'll be looking for the best ideas to steal.

This is of course the best idea. :)

Regards,

Martijn

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



From davidopp at megsinet.net  Wed Sep  1 15:21:32 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 01 Sep 1999 15:21:32 -0400
Subject: Turing Compliant?
Message-ID: <37CD7CBC.54AD295F@megsinet.net>

What the heck does Turing Compliant mean?  I've heard discussion that
Python is not Turing Compliant.  Is this true and why would this be an
important consideration for someone who is programming in Python?

David O.




From robin at jessikat.demon.co.uk  Tue Sep 21 14:01:22 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 21 Sep 1999 19:01:22 +0100
Subject: Python at work
References: <37E7B894.4CE9849F@EarthLink.Net>
Message-ID: 

In article <37E7B894.4CE9849F at EarthLink.Net>, Greg Copeland
 writes
>I'm a consultant and currently wanting to use Python and wxPython on a
>Win32 platform (Win95 more than likely).  The plan is to use this as a
>client interface to an Access97 database (thus the need for ODBC).  I
>have created a simple Access database and tried a couple of SQL queries
>against it from Python.  Things do seem to work.  Now, having said that,
>are there any glaring issues that I'm missing.  I certainly don't want
>to offer this solution to my client only to have it blow up in my face
>because of some obtuse issue; especially if it's currently well known. 
>In short, does the above sound like reliable combination?  Are there any
>issues that should raise a flag or cause alarm?
>
>Thanks in advance,
>       Greg
>
Is there an alternate way to get what you need directly with COM from
Access? The python com interface is very good.
-- 
Robin Becker



From pj at sgi.com  Wed Sep 15 18:36:41 1999
From: pj at sgi.com (Paul Jackson)
Date: 15 Sep 1999 22:36:41 GMT
Subject: high school programming & python
References: <2BAF9178787E2D18.F80B717C3A0115B2.701123DC5A66FDBE@lp.airnews.net> <7rmto7$umu1@eGroups.com>
Message-ID: <7rp71p$5nfc1@fido.engr.sgi.com>

|>  resplendent with its mercury delay line.

You had electrical computers ??

I'm jealous.  All they had at my college was gears and levers.

Really - the mechanical adding machine in the psych dept was our
"computer" my first couple years at Reed College, before they
sprang for an IBM 1130 - we skipped the tube and delay line era.

-- 

=======================================================================
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj



From gmcm at hypernet.com  Wed Sep  1 18:13:42 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 1 Sep 1999 17:13:42 -0500
Subject: Queue class deadlocking application
In-Reply-To: 
References: <1275908198-63413490@hypernet.com>
Message-ID: <1275902340-63765874@hypernet.com>

Nathan Clegg wrote: 
> Okay, quick update.  This not deadlock so much as it is actually
> freezing.
>  I have auxiliary threads that do not work with Queue at all that
>  also
> stop operating when the others do.  But as soon as I remove the one
> line that reads or writes to the queue, everything works smoothly. 
> Any ideas?

My test is still running without problems. And if it failed, I 
wouldn't expect to see unrelated threads freezing. So it's time to 
get to details of your platform and trying to create a minimal test 
case.

- Gordon



From wware-nospam at world.std.com  Sun Sep  5 15:50:09 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Sun, 5 Sep 1999 19:50:09 GMT
Subject: Turing Compliant?
References: <000201bef6a0$5323adc0$842d153f@tim> <936494840snz@vision25.demon.co.uk>
Message-ID: 

Phil Hunt (philh at vision25.demon.co.uk) wrote:
: ... Python should output the 
: design of a special-purpose chip to solve the problem :-)

I remember reading a web page by a guy who wrote a chip in Scheme
and simulated it, and when the design and simulation were all done,
the program cranked out either VHDL or Verilog.
Maybe 1.6 should have a VHDLgen module. It would probably be
better if it worked from parse trees rather than bytecodes. More
opportunities for parallelism and other such optimizations.
Now that FPGA place&route software is willing to accept VHDL input,
maybe there should be a standard PCI-bus Python accelerator board.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From pehr at alum.mit.edu  Sat Sep  4 23:57:28 1999
From: pehr at alum.mit.edu (pehr anderson)
Date: Sat, 04 Sep 1999 23:57:28 -0400
Subject: A scrollable Frame or Canvas widget
References: <37C73097.8165F42C@uniserve.com>
Message-ID: <37D1EA28.DF1A0EEF@alum.mit.edu>

Me Too!
I have the exact same problem!!!

I'm trying to throw a phalnx of textboxes for a form that is bigger
than the available window size. I need to be able to scroll the whole
mess, but I'm trapped with the lack of docs and no understanding of tcl.

If our experiences aren't coincidence, this seems to be a common pitfall.
I'm ready to switch to a different GUI toolkit like wxPython + wxGTK
and wxWindows but I'd rather get this functionality in tk/tcl.

Please help!

    -pehr anderson


Justin Muir wrote:

> Hello!  I've  been experiencing total embuggerance in attempting to
> make a scrollable Frame or Canvas widget.




From ovidiu at cup.hp.com  Tue Sep 21 17:11:24 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Tue, 21 Sep 1999 14:11:24 -0700
Subject: Any good debuggers for Python? (was Re: using python debugger 
In-Reply-To: Your message of "Tue, 21 Sep 1999 19:50:53 GMT."
              
Message-ID: <199909212111.OAA03868@hpcll563.cup.hp.com>

On Tue, 21 Sep 1999 19:50:53 GMT, wware-nospam at world.std.com (Will Ware) wrote:

> Ovidiu Predescu (ovidiu at cup.hp.com) wrote:
> : Actually PyDB works as a front-end for GDB....
> 
> One thing that would be great would be to replace GDB's rather
> inexpressive control language with something more like Python
> or Scheme. I once tried to cajole the Cygnus folks into doing
> this but they weren't interested. It sounds like PyDB might be
> doing something similar, but with the assumption that the C
> program under study is always the Python interpreter. But a
> Pythonized GDB that could be applied to any C program would be
> a very cool thing.

As I said in a previous posting, this is exactly what we're working on here at
HP, in the WDB project. You will be able to actually extend GDB by writing your
own scripts in Python to help you debug your program. I hope this code will be
eventually be accepted for inclusion on the main FSF tree of GDB (now
maintained by the Cygnus folks).

Greetings,
Ovidiu

-- 
Ovidiu Predescu 
http://andromeda.cup.hp.com/  (inside HP's firewall only)
http://www.geocities.com/SiliconValley/Monitor/7464/





From indiana at ev1.net  Tue Sep 14 22:45:46 1999
From: indiana at ev1.net (James Whitley Moore)
Date: Tue, 14 Sep 1999 21:45:46 -0500
Subject: high school programming & python
References:  <37DF9080.21A8D285@calibre-dd.com>  <37DFED78.6B258366@bioreason.com>
Message-ID: 

Why not end this thread in the usual way?

I never wanted to be a high-level language programmer.

I always wanted to be. . .

Andrew Dalke wrote in message <37DFED78.6B258366 at bioreason.com>...
>wware-nospam at world.std.com (Will Ware):
>> All we had was neutrons. They were slow and heavy, and you couldn't
>> just impose an ambient electric field and expect them to move. Back
>> then, computing took more skill and finesse.
>
>And all jobs had to finish in 15 minutes before 1/2 the neutrons
>decayed!
>
> Andrew
> dalke at bioreason.com





From s311199 at student.uq.edu.au  Mon Sep 27 04:46:43 1999
From: s311199 at student.uq.edu.au (Charles Thornhill-Cole)
Date: Mon, 27 Sep 1999 18:46:43 +1000
Subject: Regular Expressions
In-Reply-To: <37EF1A94.660587B6@carinthia.com>; from Karl Kreiner on Mon, Sep 27, 1999 at 09:19:48AM +0200
References: <37EF1A94.660587B6@carinthia.com>
Message-ID: <19990927184643.A15582@serpent.@>

On Mon, Sep 27, 1999 at 09:19:48AM +0200, Karl Kreiner wrote:
> Hi,
>         I have following problem . I have a multi-line string containing
> several patterns I have to match.
> The problem is, that my patterns may span several lines . The question
> is : How can I express a
> regular expression , that ignores line-feeds and carriage returns ?
> I'm using the module "RE" and the method "search".
> 
> thx in adv
> Charly
> 

The DOTALL option may be what you're after:

mysearch = re.compile(pattern, re.DOTALL)
result = mysearch.search(string)

It lets '.' match any character at all, including a newline.

HTH
-- 
Charles Thornhill-Cole
cat at uq.net.au




From pinard at iro.umontreal.ca  Sun Sep 19 14:14:52 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 19 Sep 1999 14:14:52 -0400
Subject: Marking translatable strings
In-Reply-To: "Emile van Sebille"'s message of "Thu, 16 Sep 1999 07:11:52 -0700"
References:  
Message-ID: 

"Emile van Sebille"  ?crit:

> Where can I find out more about the project?

Hello, Emile.  You mean, the Translation Project?  Someone else asked me
that recently.  Start looking at:

        http://www.iro.umontreal.ca/contrib/po/HTML

You ought to understand that Python is not supported yet, but I would
like that it soon is.  Moreover, I finished translating the robot of the
Translation Project to Python this morning :-).  The translation is still
very Perlish, but I merely want to get Python going for now; I'll refine
my style with time.  I need to do some more checking that everything is
right, but as soon as I am confident enough, I'll put it in production.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From max_slimmer at my-deja.com  Tue Sep 14 02:11:40 1999
From: max_slimmer at my-deja.com (max_slimmer at my-deja.com)
Date: Tue, 14 Sep 1999 06:11:40 GMT
Subject: Python & ISAPI
References: <7rjs6r$ia5$1@gxsn.com> <7rjtk5$63h$1@m2.c2.telstra-mm.net.au>
Message-ID: <7rkoum$v0v$1@nnrp1.deja.com>

In article <7rjtk5$63h$1 at m2.c2.telstra-mm.net.au>,
  "Mark Hammond" 
wrote:
> Not as far as I know.  You can use Python in
ASP, which is generally a
> better answer...
>
> Mark.
>
> Phil Harris wrote in message
<7rjs6r$ia5$1 at gxsn.com>...
> >Hi all
> >
> >Has anyone done anything with Python and ISAPI?
>
We have writen a c++ isapi to dynamically spawn
python servers, the isapi dll keeps track of all
the active python instances and dispatches to
them. If this is of interest let me know I would
be happy to share with you.
max


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fredrik at pythonware.com  Thu Sep 23 03:03:04 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 23 Sep 1999 07:03:04 GMT
Subject: decode jpeg not available
References: 
Message-ID: <007a01bf0591$e3158b60$f29b12c2@secret.pythonware.com>

Victor S. Miller  wrote:
> I have Python 1.5.2 build on Solaris 2.6 with gcc 2.7.2.2.  I also
> have tcl8.1 and tk8.1, IJG jpeg-6b and zlib-1.1.3.  I built PIL 1.0
> with all these, and uncommented and built Python with the WITH_PIL
> stuff for tkinter.  However, when I try to use the viewer.py script
> that came with PIL with a jpeg file, I get the following error:
>
> IOError: decoder jpeg not available

this usually means that the autoconfiguration stuff
didn't find your jpeg library.  the easiest way to fix
this is to tweak libImaging/ImConfig.h by hand, and
rebuild.






From mlinnen at usa.net  Sat Sep 25 15:29:49 1999
From: mlinnen at usa.net (Mike Linnen)
Date: Sat, 25 Sep 1999 19:29:49 GMT
Subject: python and I/O
References: <%4YG3.3867$UD4.20078@news.rdc1.az.home.com> <19990925130345.A11328@optiplex.palga.uucp> <%W7H3.1233$Pv.61638@news.uswest.net> <87u2oibyig.fsf@baal.winnegan.fake>
Message-ID: 

I want to access both serial and parallel I/O ports.  Looks like the serial
port is faily straight forward using termios but how would I write/read from
the parallel port?

Siggy Brentrup  wrote in message
news:87u2oibyig.fsf at baal.winnegan.fake...
> "Bill Rodgers"  writes:
>
> > I don't see how reading and writing to a file descriptor
> > provides access to serial ports, parallel ports or memory
> > mapped I/O.  Usually serial ports require more specific
> > setup criteria such as baud rate, character format (N-8-1, etc).
>
> If you are allowed to access I/O ports directly isn't a language, but
> an operating system issue. Any decent OS, in particular a multitasking
> one, must protect itself and its users from misbehaved programs.
>
> From your answer to Gerrit I presume you want to control a serial
> port. If you're running on *nix, use /dev/ttyS* and look into the
> termios(+TERMIOS) module for setting up serial parameters. I don't
> care enough about Win?? to know their way.
>
> CU
>   Siggy
>
> btw Gerrit: even with your own interpretation RTFM stays bad style.
>
> --
> >>>> In a world without fences, who needs gates?
>
> Siggy Brentrup - bsb at baal.infodrom.north.de - voice: +49-441-6990134
>
>
>





From guido at cnri.reston.va.us  Thu Sep 16 00:10:54 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 16 Sep 1999 00:10:54 -0400
Subject: bsddb module
References: <000901beff72$100796c0$3acbd9c2@peridot.optichrome.com>
Message-ID: <5liu5b4htd.fsf@eric.cnri.reston.va.us>

"Adrian Eyre"  writes:

> I was trying to build the Python source the other day, and noticed
> that the bsddb module needed some other source in order to
> build. When I finally located the source, I found that the licensing
> agreement was such that it could not be used commercially. Does this
> imply that the Python 1.5 installer's version of bsddb.pyd also
> falls under the same agreement, or is it built with a different
> version of the source code?

> The version I used was at: http://www.sleepycat.com/db/

You are probably confused by the BSDDB versioning.  The bsddb module
that comes with the Python source code requires Berkeley-db *version
1.85*.  I don't know where to find that, but I expect that there are
plenty of ftp archives that have it (do a websearch).  Anyway, the
copy of bsddb 1.85 that I have has a plain vanilla BSD style license
allowing commercial use etc. -- so I'm confident that the installer is 
fine.

The sleepycat website has Berkeley DB 2.x, which is indeed a
commercial product.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From gregm at iname.com  Wed Sep  8 21:10:34 1999
From: gregm at iname.com (Greg McFarlane)
Date: Thu, 9 Sep 1999 11:10:34 +1000
Subject: Inspecting long running programs
Message-ID: <19990909111034.12856@nms.otc.telstra.com.au>

Has anyone had experience with doing interactive debugging of long
running programs?

For example, a gui is displaying some non-reproducible and unexpected
behaviour and you would like to inspect the state of the program to
help in tracking down what went wrong.  You need to do this without
restarting the program.

One way is to put hooks into the program which would enable it to be
debugged while running.  For a Tkinter program, you can register a Tcl
command which calls a python function which calls eval (or exec) on
its arguments.  Once this is registered, you can invoke it from
another program using the Tk "send" command.  For example, say this is
the program you want to debug.

==============================
import Tkinter
root = Tkinter.Tk()

# Set up tcl command hook to be used for debugging.
def debugEval(text):
    return eval(text)
root.tk.createcommand('debugEval', eval)

... Do useful stuff ...

root.mainloop()
==============================

Then from another python script, you can call into the other program
using something like this:

# Send "dir(Tkinter)" to the running python program.
root.tk.eval('send tk debugEval dir(Tkinter)')


If the program is not running Tkinter, but using select.select say,
you could create a "debug socket" which would listen for connections
and eval() any input received on that socket.  This would be a little
more work that the Tkinter method, but would give the same
functionality.

Has anyone done this?  Are there standard ways to inspect running
python programs?


-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
                   of precipitation in some areas.



From fdrake at acm.org  Wed Sep  8 17:28:16 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Wed, 8 Sep 1999 21:28:16 GMT
Subject: Pretty-printing Python interactive output?
In-Reply-To: 
References: 
Message-ID: <14294.54512.767663.666574@weyr.cnri.reston.va.us>

Fran?ois Pinard writes:
 > Is there an easy way or device by which `pprint.pprint' could be called
 > automatically on the expression returned by the interactive interpreter?
 > Best would be able to easily switch this on and off depending on the needs.

  Wow!  I never even thought of this when I wrote that module, but it
seems like a really neat application.  I can certainly see the utility 
of it; in my $PYTHONSTARTUP I do a "from pprint import pprint" to keep 
the function convenient, but being able to make it automatic would be
neat.
  (Thomas Heller probably has the right approach, but I've not looked
at code.InteractiveConsole enough to write the subclass offhand.)
  I think Python should define the interactive loop entirely in Python 
to make it easier to do just this sort of thing.  Once initialization
is complete, if sys.interactive_loop is defined, use that as the
InteractiveConsole object, otherwise create one and use it.  That
would allow easy extension of various aspects of the loop within the
$PYTHONSTARTUP file.


  -Fred

--
Fred L. Drake, Jr.	     
Corporation for National Research Initiatives



From ferni at loxinfo.co.th  Thu Sep 23 00:36:05 1999
From: ferni at loxinfo.co.th (Jon Fernquest)
Date: Thu, 23 Sep 1999 11:36:05 +0700
Subject: Matching templates against a tree - any ideas?
Message-ID: <006901bf057d$2c478a00$022292cb@akha>

>...traverse the tree attempting to
>match each subtree in the tree against the template, however this would
>be inefficient (there could be many trees, each if which may be large,
>and many templates).

Seems like overkill for parsing equations, ...but if you were 
parsing natural language...Tree Adjoining Grammars (TAG's)
are very reminiscent of what you're describing.
They consist of large collections of tree template patterns
(possible parses) the individual constituents of which are 
made more complicated/refined through an operation called
tree adjunction, which is basically splitting a tree apart 
at a certain point and inserting another tree.
(Perl hierarchical hashes are great for this!)

They belong to the "mildly context sensitive" class of grammars.
They're  just a little bit more powerful than CFG's, just powerful
enough to deal with natural languages according to Aravind
Joshi who has a research project devoted to them at the University 
of Pennsylvania. TAG's can be parsed in O(n**4) time.
They've been used to parse newspaper text and for translating 
across languages (Synchronous TAG's,Prof Shieber at Harvard).
They can also be implemented as finite state machines
where tree adjunction is defined as a union of finite state
transducers (Roche and Schabes, 1997, 273)
Most TAG code seems to be written in LISP.
I'm writing basic tree adjunction operators now in Perl. 
(Eventually, I'd like to 
move to Python because I really don't believe in the philosophy
of writing obfuscated code that almost all seem to have in 
the Perl community. I feel embarassed to even show Perl code that
hasn't been refined down to some 
head-scratching-arabesque-of-a-one-liner)

TAG Links:
http://www.cis.upenn.edu/~xtag/
http://www.cis.upenn.edu/~joshi/

A paper describing tree templates for natural language:
ftp://ftp.cis.upenn.edu/pub/ircs/tr/95-03.ps.Z

 

Reference:
Roche and Schabes (1997). Finite State Language Processing. MIT Press. 


Jon Fernquest
ferni at loxinfo.co.th
bayinnaung at hotmail.com
http://www.geocities.com/SoHo/Square/3472/index.html

 

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From aa8vb at yahoo.com  Fri Sep 17 16:51:06 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 17 Sep 1999 16:51:06 -0400
Subject: Any good debuggers for Python? (was Re: using python debugger (pdb) inside emacs debugger mode ...)
In-Reply-To: ; from Philip Lijnzaad on Tue, Sep 14, 1999 at 04:37:09PM +0100
References: 
Message-ID: <19990917165106.A97011@vislab.epa.gov>

Philip Lijnzaad:
 |Does anyone have experience with using M-x pdb inside emacs? This part of
...
 |then it gets stuck when stepping into a module that it can't find because
 |os.getcwd() is wrong.
 |
 |Aaarggh! what am I doing wrong? Any comments appreciated!

I feel your pain.  I spent several hours this week trying to find a
source-tracking debugger for Python that works well.  No luck.

[ Is there a "Python Debugger HOW-TO" lurking out there? ]

Of what I tried, PyDB with DDD is very cool:

     http://www.ipass.net/~dbhopper/aa8vb/TMP/DDD-PyDb.tif

This is the kind of interface I'm looking for.  Full source browse and
tracking, click to set a breakpoint, click to examine/watch a variable,
etc.  However it terminates abnormally a good bit, particularly with
multi-module programs.

 |This is in emacs 20.3, but emacs 20.4 has the same gud.el 

I'd like one for Emacs myself (ala gdb/dbx with source tracking).  GNU if
possible but I'll do XEmacs if needed.  However, no luck with the brief try
I gave PDB/pdb.el/XEmacs.  User friendly PDB is not (see attached for your
amusement).

No time to continue this search now.  Possibly the best route is to dig in
and help beef up PyDB.

Randall




-------------------------------------------------------------------------------
NEWBIE PDB SESSION (for your amusement)
  - Python's terse error msgs don't help
  - "(0)?()" ?
-------------------------------------------------------------------------------

> python
Python 1.5.2 (#3, Jul  8 1999, 11:01:48) [C] on irix646-n32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import pdb
>>> import MyModule   
>>> pdb.run( 'MyModule.run()' )
> (0)?()
(Pdb) where
> (0)?()
(Pdb) list
[EOF]
(Pdb) run
*** NameError: run
(Pdb) r
> (1)?()
(Pdb) cont
en> (0)?()
(Pdb) where
> (0)?()
(Pdb) KeyboardInterrupt: 
> (1)?()
(Pdb) where
> (1)?()
  MyModule.py(615)run()
-> import pdb; pdb.run( 'root.mainloop()' )
  /home/rhh/software/python-1.5.2/lib/python1.5/pdb.py(855)run()
-> Pdb().run(statement, globals, locals)
  /home/rhh/software/python-1.5.2/lib/python1.5/bdb.py(343)run()
-> exec cmd in globals, locals
  (0)?()
  (0)?()
  /home/rhh/software/python-1.5.2/lib/python1.5/bdb.py(39)trace_dispatch()
-> return self.dispatch_line(frame)
  /home/rhh/software/python-1.5.2/lib/python1.5/bdb.py(51)dispatch_line()
-> self.user_line(frame)
  /home/rhh/software/python-1.5.2/lib/python1.5/pdb.py(90)user_line()
-> self.interaction(frame, None)
  /home/rhh/software/python-1.5.2/lib/python1.5/pdb.py(113)interaction()
-> self.cmdloop()
  /home/rhh/software/python-1.5.2/lib/python1.5/cmd.py(73)cmdloop()
-> line = raw_input(self.prompt)
(Pdb) print MyModule.run

(Pdb) print MyModule.run.root
*** AttributeError: root
(Pdb) quit
>>> 



From guido at cnri.reston.va.us  Thu Sep  9 08:09:48 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 09 Sep 1999 08:09:48 -0400
Subject: Inspecting long running programs
References: <19990909111034.12856@nms.otc.telstra.com.au> <7r7356$3b4@dfw-ixnews8.ix.netcom.com>
Message-ID: <5lr9k84777.fsf@eric.cnri.reston.va.us>

aahz at netcom.com (Aahz Maruch) writes:

> asyncore (which Medusa is built on) includes this functionality.  It can
> be kind of trippy to just telnet to a blind socket and get a Python
> interpreter session.

Really?  I don't see a trace of this in the version of asyncore I have
in the standard Python library!  I'd love to have this feature!

--Guido van Rossum (home page: http://www.python.org/~guido/)



From tim_one at email.msn.com  Sun Sep 19 18:58:50 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 19 Sep 1999 18:58:50 -0400
Subject: Marking translatable strings
In-Reply-To: <14306.17916.191557.237458@dolphin.mojam.com>
Message-ID: <000801bf02f2$89ae35a0$102d153f@tim>

[Tim]
> I agree with everyone else  that the special meaning of "_" in
> interactive mode is unlikely to create a problem for you.

[Skip Montanaro]
> Am I the only person who occasionally pastes code from a Python file I'm
> editing into an interactive session to try it out in isolation?

Yes -- or at least the only one who pastes in code fragments without also
pasting in the functions they call.

D:\Python>python
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> 1 + 3
4
>>> _
4
>>> def _(): print "hi!"
...
>>> 1 + 3
4
>>> _()
hi!
>>> __builtins__._
4
>>>

That is, the definition of "_" gets installed in the global (== local, here)
namespace, shadowing the "_" in the builtin namespace.  The eval loop
overwrites the builtin "_" specifically (it doesn't follow the normal
scoping rules).  So the pasted code will work fine; you'll simply have a
hard time getting at the builtin interactive "_" shortcut.

i-tell-you-that-guido-thought-of-everything-ly y'rs  - tim





From bwarsaw at cnri.reston.va.us  Thu Sep 16 17:07:18 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 16 Sep 1999 17:07:18 -0400 (EDT)
Subject: Marking translatable strings
References: 
	
Message-ID: <14305.23558.259342.373802@anthem.cnri.reston.va.us>

>>>>> "BH" == Bernhard Herzog  writes:

    BH> If you're careful to only use double-quoted strings inside of
    BH> _(), you can even use xgettext to build the initial
    BH> po-files. xgettext will likely complain about lots of invalid
    BH> character constants and strings if you have single- and
    BH> triple-quoted strings in your source, but otherwise it works
    BH> quite well.

This is exactly what my pygettext.py script is for.  It understands
Python syntax and all the ways to spell a string.  You'd use it
instead of xgettext to generate the .po files, then stuff like the GNU
gettext tools and Martin's intl module from there.

    http://www.python.org/~bwarsaw/software/Code/pytgettext.py

-Barry



From jhauser at ifm.uni-kiel.de  Wed Sep 22 05:55:30 1999
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 22 Sep 1999 11:55:30 +0200
Subject: lists and append, and loop iteration
References: <37E8860B.C024C4F8@radiata.com> <14312.40896.797841.421283@gargle.gargle.HOWL>
Message-ID: <87ln9ztgml.fsf@ifm.uni-kiel.de>

Charles G Waldman  writes:

> I would write 
> 
>  dict['new_key'] = dict.get('new_key',[]) + [new_value]
> 
This is a trick, because it does only supress the exception. But has
the same effect :-)

I could not write
>>> dict={}
>>> dict['new_key'] = dict.get('new_key',[]).append(['new_value'])
>>> dict
{'new_key': None}

To this refered my last note about the reference of a. Although I do
not understand this. If I get not an empty list, why can I do an
append?

Questioned,

__Janko



-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany



From stadt at cs.utwente.nl  Wed Sep 22 04:15:43 1999
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Wed, 22 Sep 1999 10:15:43 +0200
Subject: cgi authentication
References:  
Message-ID: <37E8902F.45998A4D@cs.utwente.nl>

"Magnus L. Hetland" wrote:
> 
> "Dan Grassi"  writes:
> 
> > I need to do authentication from Python, for various reasons using Apache
> > and .htaccess is not a workable solution.  I can get the request to show up
> > with the following code but I can not figure out how to access the returned
> > name/password pair.  Yes, I have looked hard for the info on the web and in
> > the books. "-)
> 
> Well - I have had the same problem, and AFAIK, Apache refuses to give
> you the password (ostensibly for security reasons...) There is a
> patched version available somewhere...
> 
> >
> > Can anyone help?
> >
> 
> --
> 
>   Magnus
>   Lie                                     Please forward all spam to
>   Hetland                                                uce at ftc.org

If all else fails, then you'll have to do it yourself, by simply using:




Login: 
Password: 





Then in SomeScript you can work with the variables 'login' and
'password'. I guess you must have thought of this. But one
never knows :-) 

But you'll have to make your own (encrypted?) password file
for this to work (or get the source of htpasswd to find out
how passwords are encrypted for use with .htaccess).

Richard.



From ovidiu at cup.hp.com  Tue Sep 21 11:41:33 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Tue, 21 Sep 1999 08:41:33 -0700
Subject: Any good debuggers for Python? (was Re: using python debugger 
 (pdb) inside emacs debugger mode ...)
In-Reply-To: Your message of "Mon, 20 Sep 1999 08:18:16 EDT."
             <19990920081816.A350855@vislab.epa.gov> 
Message-ID: <199909211541.IAA00633@hpcll563.cup.hp.com>

On Mon, 20 Sep 1999 08:18:16 -0400, Randall Hopper  wrote:

> Ovidiu Predescu:
>  |I didn't figure out exactly all the details, but the basic idea is to use
>  |GDB to handle all the interaction with the user instead of pdb. pdb
>  |becomes only a debugging engine that's loaded in a Python program and
>  |knows how to setup breakpoints, continue the program etc. All these
>  |operations are published in a simple Python API that could be called from
>  |GDB.
> 
> I believe this is how DDD/PyDB works as well (providing the user with a
> graphical Python debugger).

Actually PyDB works as a front-end for GDB. It sets up breakpoints inside the
interpreter and analyzes the internal C data structures to discover the state
of the Python program.

The thing I was thinking of is a much tighter integration between GDB and the
Python program, as GDB would make calls into the Python program to find out
about its state.

>  |What I'm working on right now is adding scripting language support to GDB
>  |so that this kind of functionality could be implemented in GDB using an
>  |interpreted language and not C. The language I'm embedding is surprise,
>  |Python ;-). This work is part of the WDB project here, at HP, but all the
>  |work we're doing is going back in the FSF main tree. For more information
>  |on WDB, take a look at:
>  |
>  |http://www.hp.com/esy/lang/tools/Debuggers/index.html
> 
> Sounds interesting.  Hope to hear more about this later.

Sure, I'll let you all know when I have something you can play with.


Greetings,
Ovidiu

-- 
Ovidiu Predescu 
http://andromeda.cup.hp.com/  (inside HP's firewall only)
http://www.geocities.com/SiliconValley/Monitor/7464/





From aahz at netcom.com  Wed Sep  8 13:46:14 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Sep 1999 17:46:14 GMT
Subject: And now for something completely boneheaded...
References: <37D623A2.DC09F0CC@nightwares.com>
Message-ID: <7r67d6$nor@dfw-ixnews15.ix.netcom.com>

In article <37D623A2.DC09F0CC at nightwares.com>,
Warren 'The Howdy Man' Ockrassa   wrote:
>
>It's... the obvious question of the day. Or perhaps the hour.
>So I have
>
>  import sys
>
>From that how can I find out what sys can do? 

Here's a different answer:

Go to http://www.python.org/doc/current/lib/lib.html and select the link
for the sys module.  (Note that I strongly suggest downloading the docs
to your local drive to make browsing faster.)
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From mwh21 at cam.ac.uk  Fri Sep  3 09:45:55 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Fri, 3 Sep 1999 14:45:55 +0100
Subject: Spanish Inquisition?!
In-Reply-To: <37CFC38C.20A9C267@megsinet.net>
References: <37CFC38C.20A9C267@megsinet.net>
Message-ID: 

On Fri, 3 Sep 1999, David Oppenheimer wrote:

> In a previous thread (about languages interpreted versus languages
> compiled) there was talk of the Spanish Inquisition.  I am guessing that
> this is 1) a joke or 2) the code name for a "secret" project to create
> compiled code from python code and thereby permanently blur the line for
> Python.  Anyone know anything about this?

I'm presuming it was a joke. There are various bits of vaguely relavent
vapour - Swallow & Viper are two of them. dejanews is your friend...

> P.S. Creating compiled code from Python  that didn't require the
> interpreter to be distributed with it would be high magic indeed.

No kidding.

Regards,
Michael




From tim_one at email.msn.com  Wed Sep 22 01:34:35 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 22 Sep 1999 01:34:35 -0400
Subject: Matching templates against a tree - any ideas?
In-Reply-To: <37E7A272.51E5C4E1@strs.co.uk>
Message-ID: <000401bf04bc$282158a0$342d153f@tim>

[Ian Clarke, wants to do tree pattern-matching]

[John Skaller, suggests a parsing approach]

Note that John Aycock's "Compiling Little Languages" Python framework (see
DejaNews) is built on an Earley parser, constructing a parse tree from input
that matches an arbitrary context-free grammar.  In its latest version it
carries that one more step, exactly as John Skaller suggests, using the
Earley parser again to support pattern-based covering of the parse tree,
where "Patterns are just trees linearized into prefix form, which use
parentheses to denote subtrees".  Elegant!  Very easy to use once you get
the hang of it.  Slow.

Note that algebraic simplification (which appears to be your real thrust) is
a field unto itself; general pattern-based approaches aren't going to beat
focused domain-specific knowledge.

factoring-by-patterns-is-like-doing-arithmetic-in-base-1-ly y'rs  - tim





From gmcm at hypernet.com  Thu Sep  2 10:10:21 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 2 Sep 1999 09:10:21 -0500
Subject: language interpreters/ interpreted languages weaknesses?
In-Reply-To: <37CE69C0.6AC977E0@strs.co.uk>
Message-ID: <1275844938-67218635@hypernet.com>

/F wrote:
> > that's what *you* think.  lots of people are running
> > Python programs without ever noticing, and without
> > ever seeing the source code.  you might have done
> > that, too.
Ian Clarke wrote:
> I am not talking about what the user notices, I am talking about
> what the developer does - and I would venture that most Python code
> is distributed with its source, where as this is rare with Java
> code.  Whether the end-user notices this is not relevant.

You can use a number of tools to package up Python apps in compiled 
form, (including freeze in the Tools directory). Both Fredrik and I 
are authors of utilities of this kind, and I can certainly assure you 
that many developers are using our tools. 

Arguing that there's a "social" component to the distinction between 
interpreted and compiled, and then arguing that it's what the 
'developer does', not what the 'user notices', indicates a fondness 
for disputation in excess of any regard for relevance.

- Gordon



From tim_one at email.msn.com  Wed Sep 22 00:54:24 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 22 Sep 1999 00:54:24 -0400
Subject: Difficulty with `cmp', still
In-Reply-To: 
Message-ID: <000001bf04b6$8b379d60$342d153f@tim>

[Fran?ois Pinard]
> Hi, people.  I feel a bit opaque, as I cannot quickly get this
> code to work:
>
>
> # Return the comparison value of two versions, of two file names
> with versions.
> def compare_versions(a, b):
>     a0, a1, a2, a3, a4 = _weights(a)
>     b0, b1, b2, b3, b4 = _weights(b)
>     print dir(__builtins__)
>     cmp = __builtins__.cmp         # because of `import cmp' above
>     return cmp(a0, b0) or a1 - b1 or a2 - b2 or a3 - b3 or a4 - b4

Answers to two questions you didn't ask :

1. Mucking with double-underscore names is a disease.  There are a few names
like that meant for routine use, primarily special class methods (like
__len__ and __add__).  But most double-underscore names expose obscure
pieces of the internals, and should almost never be used unless you're, say,
writing a Python debugger in Python.  Everyday programs don't need them and
are better off avoiding them.

In this case, I suggest avoiding __builtins__ by changing the way you import
the conflicting name:

    import cmp
    file_cmp = cmp.cmp  # assign a local, non-conflicting name
    del cmp # get the jerk out of our namespace, unshadowing the builtin

or by writing a trivial wrapper with a non-conflicting name:

    def file_cmp(f1, f2, shallow=1):
        import cmp  # the import only affects this function's locals
        return cmp.cmp(f1, f2, shallow)

2. You're familiar with lexicographic comparison of strings (keep going
"left to right" until finding unequal chars).  In Python, strings are
sequences of characters, and lexicographic comparison is used for *all*
builtin sequence types, not just strings.  I don't know what type your
_weights function returns, but it looks like sequences, so barring something
subtle the above is likely equivalent to the simpler:

    return cmp(_weights(a), _weights(b))

> When used in batch, I get "AttributeError: cmp".  When used interactively,
> everything seems to work fine, and `compare_versions' does what it should.
> The `print' statement is a mere try to understanding the problem, and it
> has the effect of adding this line to the output, just before the
> traceback:
> ...
> If I do the same `print' interactively, I get (refilled in this message):
> ...
> So far that I know, I'm not playing tricks nor trying to more clever than
> Python, I would think my code is rather straight.  I'm surely not trying
> to play with the builtin lists, the most I do is trying to get the `cmp'
> from it.  I do not see how or why the builtin list get reduced when my
> code is executed in batch, and why `cmp' is disappearing.  Would someone
> have a suggestion or an idea of what I should study to grasp what is
> happening?

The module __builtin__ contains all the builtin functions compiled into
Python.  But there's a layer of indirection between that and what a module
*sees* as being the builtin namespace, primarily to support a bulletproof
restricted execution mode (e.g., some apps can't afford to let Python code
run the builtin "open" function).  So rather than reference __builtin__
directly, the implementation looks up the related name __builtins__, which
the runtime always arranges to bind to an appropriate mapping object (either
a module or a dict).  It may or may not resolve to the module __builtin__,
or to __builtin__'s global dict.  It will *usually* resolve to one of those
two, but you can't count on that.

The point is that it's a detail of the implementation, and is there for
Python's convenience more than yours .  Don't touch it, and you won't
get surprised.

The language reference manual's section 4.1 ("Code blocks, execution frames,
and namespaces") is the key to much of Python's semantics (including the
__builtin__ vs __builtins__ distinction), as it defines "what names mean".
So that's the section to study if you want a deep understanding.

if-you-want-a-shallow-one-read-"learning-perl"-ly y'rs  - tim





From philh at vision25.demon.co.uk  Thu Sep  2 18:15:22 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 02 Sep 99 22:15:22 GMT
Subject: And Now For Something Completely Similar !!!
References: <5l906p5vpj.fsf@eric.cnri.reston.va.us>
Message-ID: <936310522snz@vision25.demon.co.uk>

In article <5l906p5vpj.fsf at eric.cnri.reston.va.us>
           guido at cnri.reston.va.us "Guido van Rossum" writes:
> kernr at mail.ncifcrf.gov (Robert Kern) writes:
> > On Wed, 01 Sep 1999 16:48:24 -0400, David Oppenheimer
> >  wrote:
> [...]
> > >2) Why isn't a module which has broad appeal like the Python Imaging Library
> > >included with the core Python 1.5.2 package?  
> > 
> > Release schedules, lean&mean-ness, not enough people needing to
> > manipulate images in Python, etc.
> 
> You said it.  The key issues here are independent developers and
> independent development schedules; I can't afford to repackage Python
> each time a new PIL comes out and I certainly don't want to have to
> wait until PIL is stable before releasing Python.

Indeed. And if you did the same thing for all libraries which use 
Python, the whole release process would rapidly become unmanageable.

How about a compromise? At the time of each Python release, take
a snapshot the latest versions of some Python packages, such as 
PIL, PySol, etc, and release them as a ``Python Goodies'' tarball.

This would:

1. make a convenient method where people could install lots of 
useful Python software in one go

2. make it obvious that the Goodies aren't part of the Python distro 
itself

3. allow people to see major programs written in Python -- this helps
to dispell the notion that Python is just a scripting language, and 
also makes available lots of source code, which might be useful to
look at if you are learning Python.

> Yeah...  This is actually in my plans for Python 1.6 (although I
> haven't started working on it honestly).

Any ideas on release date?

-- 
Phil Hunt - - - - - - - - -  philh at vision25.demon.co.uk
 - Linux will be 8 years old on 17th September! See: -
http://www.vision25.demon.co.uk/prog/linuxbirthday.html




From fredrik at pythonware.com  Tue Sep  7 04:46:17 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 7 Sep 1999 08:46:17 GMT
Subject: Tkinter-related problem
References: <19990906215844.A353@krikkit.vollmert.home>
Message-ID: <01c101bef90d$742ed570$f29b12c2@secret.pythonware.com>

Robert Vollmert  wrote:
> Xlib: unexpected async reply (sequence 0x18d)!
> Xlib: unexpected async reply (sequence 0x18e)!
> X Error of failed request:  BadGC (invalid GC parameter)
>   Major opcode of failed request:  60 (X_FreeGC)
>   Resource id in failed request:  0x2c00009
>   Serial number of failed request:  404
>   Current serial number in output stream:  407

just guessing, but are you using threads?

iirc, the "unexpected async reply" indicates that
multiple threads (or processes) are using the same
X server connection.






From godzilla at netmeg.net  Thu Sep  9 21:22:39 1999
From: godzilla at netmeg.net (Les Schaffer)
Date: Fri, 10 Sep 1999 01:22:39 GMT
Subject: Python example: possible speedup?
References:  <7r98lk$kug@dfw-ixnews6.ix.netcom.com>
Message-ID: 

> What if you use range(100000) or range(1000000)?
                            elapsed time
                               (sec)
trying NUM = 10
                  a=a+'h' --> 0.000987
                cStringIO --> 0.000954
            a.append('h') --> 0.001573
trying NUM = 100
                  a=a+'h' --> 0.004399
                cStringIO --> 0.005185
            a.append('h') --> 0.007262
trying NUM = 1000
                  a=a+'h' --> 0.056781
                cStringIO --> 0.054581
            a.append('h') --> 0.072653
trying NUM = 10000
                  a=a+'h' --> 3.217044
                cStringIO --> 0.535160
            a.append('h') --> 0.907285
trying NUM = 100000
                  a=a+'h' --> 324.303910
                cStringIO --> 5.138107
            a.append('h') --> 7.676265
trying NUM = 1000000
 
   still running  :-)  [ i want skip m's computer  !!! :-) ]

-- 
____        Les Schaffer              ___| --->> Engineering R&D <<---
Theoretical & Applied Mechanics          |  Designspring, Inc. 
Center for Radiophysics & Space Research |  http://www.designspring.com/
Cornell Univ.  schaffer at tam.cornell.edu  |  les at designspring.com



From digitome at iol.ie  Sat Sep  4 05:01:27 1999
From: digitome at iol.ie (Sean Mc Grath)
Date: Sat, 04 Sep 1999 09:01:27 GMT
Subject: Turing Compliant?
References: <000201bef6a0$5323adc0$842d153f@tim>
Message-ID: <37d0de12.934673@news.iol.ie>

>[Kristopher Johnson]
>> But can Python efficently solve the Traveling Salesman Problem,
>> or other NP-complete problems?
>
>If it were thought possible to solve such things efficiently, they wouldn't
>be NP-complete <0.5 wink>.
>
[Tim Peters]
>TSP in particular has had person-centuries of intellectual effort tossed at
>it, and the best algorithms known will certainly run faster in almost
>anything other than Python.
>

I have had a situation where an NP complete problem was being
tackled in which the speed of Python was raised as an issue.
Suggestions were made that the algorithm be rewritten in
something faster like C.

The danger in this line of thinking is not realizing that the
computational effort involved in big NP complete problems
is *so* huge that even in optimized micro-code, the
algorithm might take a million years to run.

Tweezers or shovel -- it makes little difference when you
are trying to move a universe...

regards,





From malcolmt at smart.net.au  Wed Sep 29 04:43:28 1999
From: malcolmt at smart.net.au (Malcolm Tredinnick)
Date: Wed, 29 Sep 1999 18:43:28 +1000
Subject: When Good Regular Expressions Go Bad
In-Reply-To: <000101bf0a47$98084da0$ce2d153f@tim>; from Tim Peters on Wed, Sep 29, 1999 at 02:55:19AM -0400
References:  <000101bf0a47$98084da0$ce2d153f@tim>
Message-ID: <19990929184328.A833@Ridcully.home>

On Wed, Sep 29, 1999 at 02:55:19AM -0400, Tim Peters wrote:
> [Douglas Alan]
> > ...
> > It seems to me that even when a regular expression fails to match a
> > string, you might want to know just how far it was able to get before
> > getting stuck.  (And indeed I do!) For instance, let's say that I have
> > the regular expression "^foo.bar", and I try to match it on the
> > string "fooxbaz".  It might be useful to be able to find out that
> > the regular expression was able to get all the way through "fooxb"
> > before giving up the ghost.
> 
> I believe this would be easy to add to any regexp engine I've ever seen, and
> also believe I've never seen one that keeps track of it.  I confess I'm at a
> loss to think of a compelling use for it, though.

Isn't there a problem here with which "failing match" you are going to
report on? Do you take the longest failing match (which will probably
be horrendously expensive in a complexity sense, since you will have
to scan the entire candidate string), or the leftmost failing match
(which could be uninteresting a lot of the time) or what?

While I agree it's sort of a useful thing to have sometimes (error
recovery during parsing - something like detecting possible spelling
errors), it needs to be defined a little better before being
implemented.

Cheers,
Malcolm Tredinnick.




From mal at lemburg.com  Mon Sep 27 17:36:03 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 27 Sep 1999 23:36:03 +0200
Subject: error in compiling with gcc 2.95.1
References: 
Message-ID: <37EFE343.4A590E1E@lemburg.com>

Szilveszter Juhos wrote:
> 
> PYTHONPATH= ./python ./Lib/test/regrtest.py
> Fatal Python error: unlock_import: not holding the import lock
> Abort - core dumped
> 
> Have anybody seen something similar like this? Irix 6.2 Python 1.5.2, gcc
> 2.95.1 . With gcc 2.8.1/ SGI cc 7.1 compiles fine, but there are other
> bugs (as mentioned in the source distr. README).

gcc 2.95.1 seems to have a few optimization bugs. I've had
reports about my mxDateTime package dumping core when compiled
with -On (n>2). Also, shared libs compiled with gcc 2.95.1
are rather unstable when used with code compiled using earlier
gcc versions.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                    98 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From cjc26 at nospam.cornell.edu  Mon Sep 13 20:48:39 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 14 Sep 1999 00:48:39 GMT
Subject: formatting plain text to html
References: <87btb6bmqb.fsf@iname.com>
Message-ID: 

Pada 13 Sep 1999 21:06:04 +0100, david.maslen at iname.com bilang:
| I have a bunch of plain text notes. I want to format them so that any
| URL references ie. 'http://www.python.org/' are replaced by 
| 'http://www.python.org/' 
| 
| There are probably a few other niceties such as replacing linefeeds
| with 
, but that isn't so important.
| 
| Should I just get started with the strings and re modules? Or is there 
| some code in existance that will make this task easier?

I don't know, but the following seems to work:

>>> teststr='look at my webpage http://www.python.org/ I am so l33t'
>>> re.sub('(http://[\w\-\?%~./]+)', r'\1', teststr)
'look at my webpage http://www.python.org/" I am so l33t'

Are there any other characters permitted in URL's?


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From olipt at mayo.edu  Mon Sep  6 21:05:12 1999
From: olipt at mayo.edu (Travis Oliphant)
Date: Mon, 6 Sep 1999 20:05:12 -0500
Subject: Turing Compliant?
In-Reply-To: 
References: <37CD7CBC.54AD295F@megsinet.net>  <7r0q9d$sp6$1@nnrp1.deja.com> <14291.65246.743544.70969@buffalo.fnal.gov> <7r1ibf$eb1@dfw-ixnews14.ix.netcom.com> 
Message-ID: 


> On 6 Sep 1999 23:22:23 GMT, Aahz Maruch wrote:
> >Charles G Waldman   wrote:
> >>jhefferon at my-deja.com writes:
> 
> >>> Not infinite.  Unbounded.
> 
> >>I'd be interested to hear what the difference is, according to you,
> >>between "infinite" and "unbounded".  According to me, they are the
> >>same thing.  "finite"=="bounded" => !finite==!bounded. 
> 
> >Consider the real numbers in the range 0.0 through 1.0.  They are
> >bounded but infinite.
> 
> ...but he's speaking of memory, an item which has to be enumerated in
> order to be measured.  Thus, unbounded memory is infinite memory.
> 
> Oh, and for an equally technical nit-pick: contrary to your claim, none of
> the numbers in the range between 0.0 and 1.0 are infinite.  The
> cardinality of the set, OTOH, is infinite and unbounded.

We won't even bring up that the cardinality of this set is even larger
than the simply unbounded enumerable sets.

I suppose I shouldn't go there but the infinitity of infinities is a
concept that has always enchanted me.

Travis





From dwalker at iximd.com  Mon Sep 13 08:11:08 1999
From: dwalker at iximd.com (Damond Walker)
Date: Mon, 13 Sep 1999 12:11:08 GMT
Subject: Text UI
References: <65118AEEFF5AD3118E8300508B124877073BEE@webmail.altiris.com>    
Message-ID: 

William Tanksley wrote in message ...
>
>This last is a problem for me -- I'd like to have color available.  I need
>it if I'm going to use this as part of my game, Omega.
>


    I tried the keep the curses stuff limited to one class so that it would
act as a wrapper.  it would be a trivial matter (I think) to convert this
one class to ncurses or slang, or even a "real" GUI using tkinter.
Additional code can be put in place to say: try the "standard" ncurses
module, if the import fails, step down to the curses module.

>
>It sounds great.
>

    Go to http://iximd.com/~dwalker and hit the tinter.py link.  It's a 29k
text file.

>
>That works for me.  I like LGPL.  I like the Python license better, but
>LGPL is certainly useful.
>


    I'm not married to the LGPL either...  But I'll probably be changing
that in the near future anyway.  Because I really don't care how people use
it.

                        Damond





From jfisher at are.berkeley.edu  Wed Sep  1 17:39:53 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Wed, 01 Sep 1999 21:39:53 GMT
Subject: Printing data members of a class
Message-ID: <7qk6f6$8cj$1@nnrp1.deja.com>

Hey folks,

I'm trying to set up a class so that when instances of it are printed, a
Python prints just a certain data member of the instance.  For example:

>>> class x:
... 	data = array([1, 2])

>>> y = x()
>>> y
[1 2]

or at least
>>> print y
[1 2]

I've tried playing around with repr, and defining __repr__() in the
class, but haven't had any luck yet.  I'd appreciate any suggestions on
how to do this.

Thanks!

John


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From ken.koller at kla-tencor.com  Mon Sep 13 18:58:59 1999
From: ken.koller at kla-tencor.com (Ken Koller)
Date: Mon, 13 Sep 1999 15:58:59 -0700
Subject: error in module re?
References: <937225378.3989@mhs0096x.mchh.siemens.de>
Message-ID: <37DD81B3.59A983F8@kla-tencor.com>

I think you need to use a raw string such as r'\2-\1' or escape the slash
'\\2-\\1'

 re.sub('(a+)\s(b+)', r'\2-\1', 'aaa bbb')

or

 re.sub('(a+)\s(b+)', '\\2-\\1', 'aaa bbb')

"Chris..." wrote:

> Dear Pythoners...
>
>   Is this an implementation or documentation error?
> >>> import re
> >>> re.sub('(a+)\s(b+)', '\2-\1', 'aaa bbb')
> '\002-\001'
> instead of
> 'bbb-aaa'
> The documentation says, that you can use backreferences in replacement
> string, but only \g works.
>
>   My version is:
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
>
> bye
>   Chris...




From davidopp at megsinet.net  Wed Sep 22 20:05:58 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 22 Sep 1999 20:05:58 -0400
Subject: Please Help Convert This Code
Message-ID: <37E96EE5.95270EA9@megsinet.net>

Dear Group,

I would really appreciate it if someone could help convert this Tkinter
code into equivalent wxPython code.  Need this to be done to create a
drag and drop GUI maker that will have Tkinter AND wxPython as output
choices.  Thanks in advance.

David O.


from Tkinter import *

def write_defaults(Widget):
        widget = Widget()
        widget.place(x=0, y=0)
        filename = "%s.txt" % Widget.__name__
        file = open(filename, "w")
        for key in widget.keys():
                if widget[key] == "":
                        widget[key] = None
                        print widget[key]
                file.write("%s = %s\n" % (key, widget[key]) )
        file.close()



if __name__ == "__main__":
        root = Tk()
        widgets = [ Button,
                        Canvas,
                        Checkbutton,
                        Entry,
                        Frame,
                        Label,
                        Listbox,
                        #Menu,
                        Menubutton,
                        Message,
                        Radiobutton,
                        Scale,
                        Scrollbar]
        for widget in widgets:
                write_defaults(widget)
        root.mainloop()








From randall_burns_hp at my-deja.com  Thu Sep  2 21:01:40 1999
From: randall_burns_hp at my-deja.com (randall_burns_hp at my-deja.com)
Date: Fri, 03 Sep 1999 01:01:40 GMT
Subject: How to sell Python to your boss?
References: 
Message-ID: <7qn6lf$ef0$1@nnrp1.deja.com>

In article ,
  Petri Mikael Kuittinen  wrote:
>
I'm fortunate to have a boss who is very enthusiastic about
Python. (We'd really like to find more developers who share his
enthusiasm form Python-I'd be happy to talk to folks about the
opportunities doing e-commerce here at HP).

> I can understand my bosses opinion. He wants to protect the interest
> of this firm. If I would leave the firm, then someone else would have
> tp spend time learning Python in order to modify my program code.
>
> For several months I have tried to get my colleagues excited to learn
> Python, but so far nobody has been interested to learn new languages,
> except for some mild interest towards Java (which is IMHO an
> overly-hyped language).
We use Java also-but we've found Python to ease both development and
maintenance.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From stephan at pcrm.win.tue.nl  Fri Sep 24 03:19:38 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 24 Sep 1999 07:19:38 GMT
Subject: Becoming root within a Python script
References:  <87zoydrw4u.fsf@baal.winnegan.fake> <37EAE579.2499F85C@radiata.com>  
Message-ID: 

On 24 Sep 1999 00:29:02 -0400, Fran?ois Pinard  wrote:
>Mark Krischer  ?crit:
>
>> why don't you call the python script using sudo, or make a wrapper
>> script that calls sudo .py
>
>I would prefer not run the script as root, but just become root, here and
>there, once in a while, in precise and unusual circumstances.  And give
>the root password when those times come, and only then.  I'm not found on

One thing you can do -- although I'm not sure if this is what you want --
is to have the script started with a setuid wrapper anyway, and then swap the
real and the effective user id, until the time comes when you want
to do something as "root", and then you swap them back.

This is not really Python-specific, and I guess you need to write a
Python wrapper for the system call setreuid(). But if you do
that (supposing you put that wrapper in a module uid), code
might look like this (obviously untested):

import os
import uid # implementing uid is left as an exercise for the reader

def swapreuid():
    """Swap real and effective user id.
    """
    ruid = os.getuid() # get real user id
    euid = os.geteuid() # get effective user id
    uid.setreuid(euid, ruid)

Then make sure you get started with the effective uder id as "root",
and then do a swapreuid() the first thing in your script. This will
set the effective user id back to that of the user.

Then, when you have to do root magic, swap them back so you  can
do the magic.

IIRC, this won't work with sudo, since sudo also sets the real uid to
"root" . So you really have to use a wrappper. By the way, I think
that it is even more secure to have the wrapper swap the real and
effective user id, instead of doing this in the first line of the
script. Whether this is *really* secure, I can't tell. Obviously,
it also depends on the contents of your script.

OK, I hope this is somewhat helpful.
If it's not, please delete this post. ;-)

Greetings,

Stephan



From towen at I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN  Thu Sep 30 19:33:59 1999
From: towen at I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN (Todd Owen)
Date: 30 Sep 1999 15:33:59 -0800
Subject: where is panelmodule?
Message-ID: <37f31267.0@news.highway1.com.au>

g'day.  I found a mention of something called panelmodule (the README
for cursesmodule 1.4 says "is fully functional with the panelmodule
1.1").  But I can't find this ellusive module anywhere!  Does anybody
know where I can find it?

TIA

- Todd




From jam at quark.emich.edu  Thu Sep 30 10:04:23 1999
From: jam at quark.emich.edu (Jeff)
Date: Thu, 30 Sep 1999 10:04:23 -0400
Subject: 'ipfw' IOCTLS available as python module?
Message-ID: <19990930100423.D22550@quark.emich.edu>

greetings all,

I have need to write a firewall configuration script in python, and one of
my design goals is that python not act as just another /bin/bash
replacement. what I'd like is a module that knows how to directly make the
ioctl calls described in the ipfw(4) man page (so I don't have to resort to
'os.system("/sbin/ipchains -F")', which looks ugly). I am willing to work on
the project, but I thought I saw *something* whiz by on the mailing list
regarding this same issue. however my searcg via python.org didn't turn up
anything.

does anyone have any pointers they'd be willing to share?

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From thomas at bibsyst.no  Tue Sep 28 07:27:46 1999
From: thomas at bibsyst.no (Thomas Weholt)
Date: Tue, 28 Sep 1999 13:27:46 +0200
Subject: Python Berkley DB and speed issues
References: <37F06752.7D98775@bibsyst.no> <37F08369.746D535@lemburg.com>
Message-ID: <37F0A632.895A9271@bibsyst.no>

> 
> Hmm, millions of records... you must have a *large* CD collection ;-)
> 
> --
> Marc-Andre Lemburg

Well, yes, but the fact is that each file is scanned and stored. When
you have lots of clipart, images and source-code, that soon becomes alot
of data. I`ve implemented a "pattern-scheme" so that users can define
patterns to look for, ignore or warn about, so that the level of details
can be modified. It works pretty well right now.

Thanks for your input.

Thomas



From justin at ukans.edu  Fri Sep  3 13:49:43 1999
From: justin at ukans.edu (Justin Bradford)
Date: Fri, 3 Sep 1999 17:49:43 GMT
Subject: ORBit bindings/interface?
In-Reply-To: 
References: 
Message-ID: 

> ILU supports this already and is fairly widely used in the Python
> community.  See ftp://ftp.parc.xerox.com/pub/ilu/ilu.html for details.

I want to use these bindings in Gnome applications, so I need ORBit
support specifically (ie. I don't want the user to have to run ILU to
communicate to ORBit to talk to the rest of the Gnome applications).

Or is there something I'm missing?

Justin






From init__self at my-deja.com  Thu Sep 16 13:11:49 1999
From: init__self at my-deja.com (Python Addict)
Date: Thu, 16 Sep 1999 17:11:49 GMT
Subject: high school programming & python
References:    <937300939snz@vision25.demon.co.uk> <37E0F23E.B0199331@strs.co.uk>
Message-ID: 

Rats! You had rats? You lucky buggers. We didn't even have ones, we just had
zeros. And sometimes They forced us to use the letter O instead. Rats ==
luxury.





From bsb at baal.infodrom.north.de  Sun Sep 19 06:03:20 1999
From: bsb at baal.infodrom.north.de (Siggy Brentrup)
Date: 19 Sep 1999 12:03:20 +0200
Subject: NCDaudio aka nas
Message-ID: <87iu57p6af.fsf@baal.infodrom.north.de>

Hi Pythoneers,

after searching the archives (and the NIST registry) revealed no
results, I'm asking here before starting my own effort with SWIG.

Did anybody package an python interface to NCDaudio aka Network Audio
System ?

>From Debian's nas/copyright:
> This is the Debian GNU/Linux prepackaged version of NAS, the Network
> Audio System.  The Network Audio System was written by Jim Fulton,
> Greg Renda, and Dave Lemke at Network Computing Devices, Inc.

> This package was put together by Ian Murdock ,
> and then updated by Michael Nonweiler , using
> sources obtained from: ftp.x.org:/contrib/audio/nas/nas-1.2p4.tar.gz

CU
  Siggy

-- 
noch nichts Aufregendes:

Siggy Brentrup - bsb at baal.infodrom.north.de - voice: +49-441-6990134




From steinme at norway.eu.net  Tue Sep 21 15:24:23 1999
From: steinme at norway.eu.net (Stein M. Eliassen)
Date: Tue, 21 Sep 1999 21:24:23 +0200
Subject: make test failed in time_test
Message-ID: <37E7DB67.5DE00C61@norway.eu.net>

hi,

I'm running Slackware 4.0 and trying a 'make test' with python v1.5.2.
when I got this:

test test_time crashed -- exceptions.AttributeError : altzone

what's this and how do I fix it?


Stein



From scarblac-spamtrap at pino.selwerd.cx  Mon Sep 13 05:05:53 1999
From: scarblac-spamtrap at pino.selwerd.cx (Remco Gerlich)
Date: 13 Sep 1999 09:05:53 GMT
Subject: equivalent of enum?
References:  <14296.14104.456361.61180@dolphin.mojam.com>
Message-ID: 

Skip Montanaro  wrote:
> If you expect to have properties associated with particular pieces
> (position, strength, etc), you might want to consider representing them by
> instances of a specific ChessPiece class instead:

I'm not going to make a chess playing engine, mostly store games,
put comments into them, etc. But a property that they do have is
the way they move.

I'm new to Python, this is the first thing of more than a few lines
that I want to program, and I'm still searching for the right mindset :-).

Maybe an empty ChessPiece superclass, and then subclasses for each
piece type, so it can return a list of squares it can move to?

Hmm. I think I'll just make some functions for that, OO is overkill
there, in my opinion.
>     whiterook = Rook(5, "h", "8", "white")
>     whitenight = Knight(3, "h", "7", "white")
>     ...

This is probably over-killing a bit. The base object that I'll use
is the Position; most move possibilities depend on the rest of the
position anyway. It'll just hold a list of squares, with piece
numbers. This is also easier because you can give a rook that hasn't 
moved yet a different number from one that has moved, because of
castling rules; to let an object change itself like that feels
icky to me :).

This may give rise to a lot of if/elsif/elsif/elsif/if/... type
of code, but then, the rules of chess are unlikely to change
soon so it's not a big problem :).

> In more direct answer to your question, there is no enum type in Python.
> It's simple enough to declare objects that you treat as constants, however:
> 
>     whiterook = 1
>     whiteknight = 2

Yes, that is what I'll do. I like the 
emptysquare, whiterook, whiteknight = range(3)
idiom from the other post :).

-- 
Remco Gerlich,  scarblac at pino.selwerd.cx
  Murphy's Rules, "Life in the fast lane":
   In the fifth edition of FBI's Tunnels & Trolls, an average man may
   make 10-second sprints topping 60 miles per hour. An exceptional
   runner could sprint at 118 mph.



From jeremy at cnri.reston.va.us  Tue Sep 21 18:51:29 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Tue, 21 Sep 1999 18:51:29 -0400 (EDT)
Subject: global in nested functions (viper)
In-Reply-To: <9XRF3.2$Ae3.15469@news-dal.corridex.com>
References: <37E7AF7B.8DEC7CC8@maxtal.com.au>
	<5l67149gwc.fsf@eric.cnri.reston.va.us>
	<9XRF3.2$Ae3.15469@news-dal.corridex.com>
Message-ID: <14312.3057.617522.176041@bitdiddle.cnri.reston.va.us>

>>>>> "ES" == Evan Simpson  writes:

  ES> This sounds relevent to one of my many proposals to make work
  ES> for other folks .  Perhaps instead of causing a reference
  ES> to the (grand-)*parent's locals, this could simply copy the
  ES> object into a local variable at definition time?

This wouldn't help for code that uses assignment.  The expected
behavior of 'x = 2' is that all code that can access x will see the
new value.  Copying into a local variable would make changes visible
only in the local scope.

The problem with the earlier proposal -- that global means look in the 
locals of the containing function unless it also contains a global
decl -- is that you need to add a global decl to every scope between
the one where the variable is created and the one where it is used.
It seems less tedious to define global to mean the nearest enclosing
scope. 

Jeremy



From aahz at netcom.com  Mon Sep  6 10:27:10 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 6 Sep 1999 14:27:10 GMT
Subject: Python speed and `pcre'
References: <000301bef83b$9d7fa620$da2d153f@tim>
Message-ID: <7r0ivu$95j@dfw-ixnews21.ix.netcom.com>

In article <000301bef83b$9d7fa620$da2d153f at tim>,
Tim Peters  wrote:
>
>The design issues are more complicated than that, alas.  Here's one you
>weren't thinking of :  for easy chatting with C, under the covers a
>Python string is always terminated with a null byte.  You can't see that
>from Python code (it's not counted in the string length, and you can't index
>to it), but your C code can rely on it.

Hmmm....  From what I've seen, Python strings work perfectly well with
null bytes in the middle.  Are there any circumstances under which that
does not hold true?
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From daniels at mindspring.com  Sun Sep 12 22:20:46 1999
From: daniels at mindspring.com (Alan Daniels)
Date: 13 Sep 1999 02:20:46 GMT
Subject: equivalent of enum?
References:  <14296.14104.456361.61180@dolphin.mojam.com> <37D84F85.A37FAA9C@home.com>
Message-ID: 

On Fri, 10 Sep 1999 00:18:24 GMT, the infinitely wise Doug Hellmann
(doughellmann at home.com) spoke forth to us, saying...

[snip...]
>	one, two, three = range(3)
>	print one, two, three

This is a handy little trick (thanks for posting it!), and not to be
picky, but the indexing will start at zero instead of one, which could
trip up people new to Python. So this would be better as:

	zero, one, two = range(3)

Or, maybe:

	one, two, three = range(1, 4)

Also, just as a matter of style, I use all CAPS when the number I'm
defining is meant to be a constant. Using ranges is sure going to beat
naming all the enum values by hand, though. :=)

-- 
=======================
Alan Daniels
daniels at mindspring.com
daniels at cc.gatech.edu



From evan at tokenexchange.com  Thu Sep  2 12:52:30 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Thu, 2 Sep 1999 11:52:30 -0500
Subject: break LABEL vs. exceptions + PROPOSAL
References: <37bc1339@cs.colorado.edu> <19990819174800.D5227@xs4all.nl> <37bc313b@cs.colorado.edu> 
Message-ID: 

Here's a (shamelessly unmotivated) example which I find more expressive than
it would be with "break LABEL":

Found, Panic = 'Found', 'Panic'
try:
   for thingy in whatzits:
      for bit in thingy:
         if bit.is_wanted():
            raise Found, bit
         if hasattr(bit, 'panic'):
            raise Panic, bit.panic
      if thingy.other_bit.is_wanted()
         raise Found, thingy.other_bit
      if hasattr(thingy, 'panic'):
         raise Panic, thingy.panic
   else:
      print "Not found"
except Found, it:
   print "Found", it
except Panic, why
   print "Panic!", why

In particular, "break LABEL" doesn't allow you to provide information about
*why* you're leaving the loop, and *what* caused that condition without
explicitly setting up variables to hold that information and then checking
the reason, post-loop.  Exceptions are more general, too; there doesn't have
to be a loop.

The only things I don't like about this are the unnecessary involvement of
the fairly heavy exception machinery, the need to "declare" my break reasons
as variables to ensure an "is" match, and the possibility of a subtle typo
turning my local "break" into an uncaught exception.  Rather than "break
LABEL", I would like to see a construction like the following: (in 2.0, of
course )

try:
  ...
  break , 
  ...
continue , :
  
continue , :
  

The first argument to "break" and "continue" is required to be a string, and
each "break" string is required to match a string in a containing try's
"continue".  This makes it clear that we're breaking out of the suite rather
than raising an exception, and could be implemented efficiently and
trivially with jump, load, and store bytecodes.

Rich Salz  wrote in message
news:rst6om9lj7141 at corp.supernews.com...
> >No, I'm sorry, the expressivity of "break LABEL" and "continue LABEL"
> >are much clearer than the cicumlocutions necessary for exceptions.
>
> You sure?  I wasn't aware that I could break and return a value, such as
> the name of the file that wasn't found.  At least not without a lot of
> circuml^H^H^H^H^H global variables.






From linux_12345 at yahoo.com  Thu Sep  2 07:24:20 1999
From: linux_12345 at yahoo.com (linux_12345 at yahoo.com)
Date: Thu, 02 Sep 1999 05:24:20 -0600
Subject: BRIGHTGATE.COM 54041
Message-ID: <020999052420@yahoo.com>

Brightgate Metasearch is the newest meta-search site on the net.  It
searches over 18 (most used) search engines on the Internet, well above
the average of all other metasearch sites.  There is also an option to
submit a url to 30+ search engines.  Of course, do not forget the MP3 and
Files search options, if you are looking for that special MP3 or file then
chances are you will find it with Brightgate Metasearch.
http://www.brightgate.com



From davis at space.mit.edu  Sat Sep 11 05:58:33 1999
From: davis at space.mit.edu (John E. Davis)
Date: 11 Sep 1999 09:58:33 GMT
Subject: How to get the python sourcecode line, when debuging the Python C sources?
References: <7r3bqk$hbv$1@newshost.germany.net> <7r4alh$7gl$1@m2.c2.telstra-mm.net.au> <7r4bqi$jds@dfw-ixnews11.ix.netcom.com>
Message-ID: 

On 8 Sep 1999 00:49:22 GMT, Aahz Maruch 
wrote:
>>string containing the filename.  Adding something like "(char *)(((char
>>*)f->f_code->co_filename))+20)" to the watch window should do the trick.

It is not necessary to use the initial cast since 

     (char *)f->f_code->co_filename + 20

is already a `char *'.

>And people wonder why I think C is a dead language.

Not really.  What this expression does it actually somewhat
complicated.  For example, suppose X is some object in the python
language, e.g.,

      X = 3.0

How would you form a string from the last 3 bytes of the 8 byte object
represented by X in pure python?  In C you would use `(char *)&X + 5'.     
Why you would want to do this another issue.

--John



From gerrit.holl at pobox.com  Wed Sep  8 15:12:33 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 8 Sep 1999 21:12:33 +0200
Subject: binary mail with python
In-Reply-To: <036101bef9e7$b907a480$b4bb32c1@clara.ircom.unilim.fr>; from Daniel ARGOLLO on Wed, Sep 08, 1999 at 12:48:44PM +0200
References: <036101bef9e7$b907a480$b4bb32c1@clara.ircom.unilim.fr>
Message-ID: <19990908211233.A3782@optiplex.palga.uucp>

On Wed, Sep 08, 1999 at 12:48:44PM +0200, Daniel ARGOLLO wrote:
> Hi There!!!!
> 
> 
> I am searching for any python package that can pack different binary files into a
> MIMI file to be sent with the mail suport of python....
> 

Maybe you're looking for mimimy?
I've never used it, but the name means a lot (did I say it right?).

regards,
Gerrit.
-- 
Unix is easy: type 'man man' and the rest comes of itself.



From jam at quark.emich.edu  Thu Sep 16 15:52:52 1999
From: jam at quark.emich.edu (Jeff)
Date: Thu, 16 Sep 1999 15:52:52 -0400
Subject: newbie - CGI.pm equivalent HTML Forms generator?
In-Reply-To: ; from Kevin Howe on Thu, Sep 16, 1999 at 02:46:52PM -0300
References: 
Message-ID: <19990916155252.C8285@quark.emich.edu>

On Thu, Sep 16, 1999 at 02:46:52PM -0300, Kevin Howe wrote:
> Hi,
> 
> I am a Perl programmer who is learning Python, and was wonder if their is a
> Python package/object/module which allows easy creation of HTML Form
> Elements.
> 
> Thanks,
> Kevin
> 

there are *many* ways to implement python-based dynamic webpages. rather
than listing 'everything', check out  for pointers
to modules that do this. I like medusa a lot, which you can find at
, and there are lots of people that like
Zope, which you can find at . there are also nice
clean simple modules in the standard library (like 'cgi.py') that will let
you use 'print' statements to accomplish the HTML and still parse query
strings and POST data. if you want to try medusa, I posted some code
recently to the medusa mailing list that might help you get started. I can
send you the code via email if you'd like.

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From srenner at lycosmail.com  Tue Sep  7 22:16:58 1999
From: srenner at lycosmail.com (srenner at lycosmail.com)
Date: Wed, 08 Sep 1999 02:16:58 GMT
Subject: idle 0.5: correction
References: <7r42s9$88m$1@nnrp1.deja.com>
Message-ID: <7r4guh$i70$1@nnrp1.deja.com>

In article <7r42s9$88m$1 at nnrp1.deja.com>,
  srenner at lycosmail.com wrote:
> Thanks to CVS I don't know when the class browser first appeared in
idle
>

Thanks to ...Tools/idle/cvs/Entries I know exactly when:

/ClassBrowser.py/1.10/Fri Jun 25 17:08:19 1999//

sr


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From garry at research.att.com  Thu Sep  2 10:31:56 1999
From: garry at research.att.com (Garrett G. Hodgson)
Date: Thu, 2 Sep 1999 14:31:56 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com>
		<37CCE10D.D07D023B@strs.co.uk>
		<7qjgq8$k1i$1@newshost.accu.uu.nl>
		 <14285.38262.884171.416195@dolphin.mojam.com>
Message-ID: <37CE8A5C.586F56F3@research.att.com>

Skip Montanaro wrote:

> The only thing that makes Java "more compiled" than Python or Perl is
> that you have to explicitly run a separate compiler over the source to
> create a .class file.  

that's the real beauty of java.  all the convenience of having to go
through a separate compilation step, combined with the blistering
performance of an interpreted language.

what a deal.

-- 
Garry Hodgson			"Hey, mister, can ya tell me,
garry at sage.att.com		where a man might find a bed?"
Software Innovation Services	He just grinned and shook my hand,
AT&T Labs			"No", was all he said.



From tim_one at email.msn.com  Sun Sep 19 13:35:27 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 19 Sep 1999 13:35:27 -0400
Subject: Force anything to be a string.
In-Reply-To: 
Message-ID: <000001bf02c5$5d191140$102d153f@tim>

[jonathon, finds that "str" has been rebound]
> ...
> 	A little poking around, and I found an object called
> 	str in two modules that the script calls.   After changing
> 	the name of those variables, that error disappeared.  :-)

This sounds like you fixed the symptom, but not the real cause:  it
shouldn't make any difference to *your* module what other modules happen to
name their variables.  Are you, by any chance, doing

from OtherModule import *

?  Then you're simply paying the price for being lazy <0.6 wink> -- that's
*inviting* OtherModule to stomp all over your names.  "import *" is a
convenience for interactive mode, and should almost never be used in a
script (there are a very few modules that were designed to be used that way,
but until you're a Python expert better to pretend it's illegal).

[Tim]
>> ... builtin function names are not reserved, which is both a feature
>> and a bug <0.7 [wink]>.


> 	A bug, IMNSHO.   A really bad bug, since I wasn't even using a
> 	module I wrote.

If you used "import *", you dug your own grave and so get no sympathy.  If
Python did reserve the builtin function names, then either (a) no builtin
function could ever be added to the language again; or, (b) adding a new
builtin function would break existing code that happened to use the same
name.  Neither is tolerable.  I'd like to see a warning msg, though (it
can't be made an error, at least not by default).

>> >>> str = 'oops'
>> >>> str(42)
>> Traceback (innermost last):
>>  File "", line 1, in ?
>> TypeError: call of non-function (type string)
>> >>>
>>
>> The error msg there should look familiar.  It's telling you that the
>> syntax looks like a call, but the object in the caller's position *can't*
>> be called ("call of non-function"); and it's also giving the type of the
>> object that can't be called (str is of "type string").

> 	Another piece of wisdom to add to the list of things I
> 	couldn't find in _The Python Library Reference Manual_.

There are about 900 distinct error msgs in the Python implementation, and
while some of them could certainly be wordier, this one looks pretty darned
good to me!  I didn't explain it because I thought it was unclear, but to
nudge you into really *reading* the error msg and think about what it's
saying.  "call of non-function" doesn't have a heck of a lot of *possible*
meanings, after all .

The "(type string)" part may be obscure (a disconnected sentence fragment)
the first time you see it, but playing around in interactive mode provoking
this msg quickly makes its meaning obvious.  It won't confuse you the 2nd
time you see it.

If the Python doc set contained a 450-page manual detailing every error msg
(figuring a half page per msg), would you have read it <0.9 wink>?

rhetorical-question-cuz-nobody's-gonna-write-that-anyway-ly y'rs  - tim





From robin at jessikat.demon.co.uk  Thu Sep 16 11:01:54 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 16 Sep 1999 16:01:54 +0100
Subject: Marking translatable strings
References: 
 
Message-ID: 

In article , Emile van Sebille
 writes
>Hi Francois,
>
....
>If a change to the language is in the offing, why not introduce
>something like a t"TEXT" structure that behaves as r"..." but also
>serves as a marker.  Then we'd have 12 ways to express a quoted string.
>;-)
>
More generally wouldn't one need to allow for quoted strings prefixed by
either nothing, r or t or rt or tr and if that's the case why stop at
only three prefix symbols (counting empty).
>Otherwise, It sounds interesting.  Where can I find out more about the
>project?
>
>Regards,
>--
>
>Emile van Sebille
>emile at fenx.com
>-------------------
>
>
>Fran?ois Pinard  wrote in message
>news:oq4sgvf0m1.fsf at titan.progiciels-bpi.ca...
>> Hello, people.
>>
>> I had a real strange idea :-).  I first quickly dismiss it, but it is
>so
>> simple that I prefer to ponder it again, and share and debate it a
>little
>> first, maybe.
>>
>> My friends know that I've been working at software
>internationalisation for
>> many years now, with a stress on program messages.  Of course, I want
>to
>> include Python in the realm of my possibilities, and myself start
>writing
>> internationalised scripts soon, in such a way that everything links
>nicely
>> with the Translation Project.
>>
>> So, I want the big picture right now.  That is: a technique for
>marking
>> strings for automatic extraction and building of PO files, and a
>technique
>> for using PO files from within Python scripts.  I foresee that Python
>> introduces an usual difficulty in that the textual domain for
>translations
>> may vary quite unexpectedly, when the control dynamically flies
>between
>> independent packages under different textual domains.
>>
>> I started a discussion with Guido about this, but I'm a slow thinker,
>and
>> would not like to rush things before feeling rather solid, as Guido's
>time
>> is precious.  But on this forum, I thought I could dare exploring
>ideas,
>> asking for your forgiveness for any blunder I could make while
>thinking. :-)
>>
>> 1) Marking strings
>>    ---------------
>>
>> There are many circumstances where strings translation could be
>delayed
>> from where they textually occur, and that language syntactic
>considerations
>> could make the marking difficult in a few cases.  In C, there is a
>> pre-processor between the sources and the compiler, so it is easy to
>introduce
>> an identity macro which special name is recognised by the string
>extractor,
>> and which vanishes before the compiler sees it.  We use:
>>
>>    #define N_(Text) Text
>>
>> for that purpose.  In languages where such preprocessing is difficult,
>like
>> for `ksh' or `bash', strings are especially marked in the syntax, like
>in:
>>
>>    $"translatable text"
>>
>> but this requires a modification to the interpreter.  Emacs, with
>`defsubst',
>> also allows for macro expansion, and we then use tricks as in C
>(anywhere
>> except for doc strings).  Some other flavours of LISP are also open to
>> such tricks.
>>
>> Python has no preprocessing, no special string syntax for markability,
>> and moreover, it has doc strings!  So, at first glance, it looks
>difficult.
>> However, and this is where my strange idea comes to play, it has eight
>> type of strings: ', ", ''', """, r', r", r''' and r"""; and I thought
>> that maybe we could just discipline ourselves to give more meaning to
>all
>> these differences, since after all, if we except some ending backslash
>> considerations, all eight types are equally capable of representing
>> any string.
>>
>> I'm a strange, anal man, who needs a reason behind the slightest
>choices,
>> and believe me, eight types of strings gave me a lot of food for this
>> mania, all along while writing.  I'm still exploring! :-) Yet, after
>having
>> played with Python for almost 10 days, now, I came to realise that I'm
>more
>> naturally tempted to stick to the 'TEXT' notation for computer strings
>and
>> "TEXT" notation for human strings, the reasons being that there might
>> be a lot of apostrophes in human text, and that traditionally, we more
>> usually quote sentences with "TEXT", while we quote words with `TEXT'
>> (note the grave accent at the left).
>>
>> So, the bizarre idea I got is that one could be to formalize this into
>> a rule: strings of type ", """, r" and r""" could be all markable as
>> translatable, while strings of type ', ''', r' and r''' would not be.
>> On the other hand, this might be overkill, as maybe people are used to
>> freely mix types ' and ", and this change could be seen as stressful.
>> Could we choose better?
>>
>> Surely, since doc strings use """ exclusively, there is no choice as
>to retain
>> type """ for translability, wherever it appears.  However, forcing the
>use
>> of """ everywhere we want translatibility is an overhead of four
>characters
>> (just compare "TEXT" with """TEXT"""), while C use three or four
>characters
>> (compare "TEXT" with _("TEXT") or N_("TEXT"), and bash uses only one
>(compare
>> "TEXT" with $"TEXT").  I would like Python to be as comfortable as
>possible.
>> If I could plainly use "TEXT" instead of 'TEXT' to mark
>translatability,
>> I would have an overhead of zero characters, which would be better
>than
>> everything, but I'm not sure if this constraint would be acceptable to
>> Python writers.
>>
>> Another possibility is to use ''"TEXT" instead of "TEXT", making an
>overhead
>> of two characters: that is the compile time concatenation of '' with
>"TEXT".
>> This combination is quite unlikely to me, and a bit uglier.
>>
>> 2) Translating strings
>>    -------------------
>>
>> (Oops, I just received a phone call forcing me to leave fairly soon,
>so I
>> have to be very concise for the remainder of this message.  Let's
>rather
>> develop these in the possible thread that might follow from this
>message.)
>>
>> What would be the most comfortable for me, short of having the Python
>> interpreter modified, is to merely use a function to force the actual
>> translation of a string.  The most comfortable (the less intrusive)
>way
>> would be to call:
>>
>>         _(TEXT)
>>
>> to get the translation of text.  It resembles C, but it overloads `_',
>> which already has a preset meaning, interactively.  If I could push
>the
>> preset `_' somewhere else, maybe on `__', I would do it and reserve
>`_'
>> for translation, which would be much, much more common in the long
>run.
>>
>> Using a function would allow us to build the whole translation chain
>> (administrating the translations with teams, etc.), yet if the syntax
>> could be relieved with the help of Guido, I guess this would be
>welcome.
>> We might need to experiment first.
>>
>> 3) Setting the textual domain
>>    --------------------------
>>
>> In a quick word, I guess that this problem could be fairly easily
>solved
>> through the handy scope rules for resolution of names in Python.  Each
>module
>> could have a standard global variable name setting which textual
>domain to
>> use within it.  So, even with the control flying like hell between
>modules,
>> it would not be a problem on average.  But there are problematic
>cases,
>> like for when untranslated strings are transmitted to other modules,
>for
>> being translated there, or even maybe for plain doc strings.  This
>requires
>> good thought.  This problem is more difficult that many might thing at
>first.
>>
>> OK, I have to rush away now.  Thanks for listening! :-)
>>
>> --
>> Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard
>>
>>
>>
>
>

-- 
Robin Becker



From piller at jlab.org  Wed Sep  8 14:51:23 1999
From: piller at jlab.org (Chip Piller)
Date: Wed, 08 Sep 1999 14:51:23 -0400
Subject: help with serial port I/O
Message-ID: <37D6B02B.493ED51@jlab.org>

Hi,
I am running RedHat Linux 6.0.
I have an instrumentation amplifier that can be controlled by the serial
port.
I can successfully setup the serial line and read and write to the
device using the minicom serial communications program.

I want to write a simple program that will write a command to the
device, wait for one second, and then read the response from the device.

I tried the following code:
#!/usr/bin/python
udl = open('/dev/ttyS1','r+')
udl.write('sen 22\r')              #change instrument sensitivity
setting
udl.write('mag\r')                  #query instrument for signal
magnitude
val = udl.readline()
print 'magnitude readback = '
print val
udl.close()

Here is the output:
[root at piller1 python]# ./test3
magnitude readback =
sen 22

[root at piller1 python]#

The code does write the sen 22 command to the instrument and changes the
value, but not on the first try, I usually have to run the program
twice.
I have not had success with getting any readback from the device using
Python.
Help or suggestions please!
Chip








From cfelling at iae.nl  Fri Sep 10 21:49:20 1999
From: cfelling at iae.nl (Carel Fellinger)
Date: 11 Sep 1999 03:49:20 +0200
Subject: Python example: possible speedup?
References: <000b01befb24$05ca6fe0$f0a0143f@tim> <87g10my6v2.fsf@pc-hrvoje.srce.hr>
Message-ID: <7rccf0$g5$1@vvs.superst.iae.nl>

Hrvoje Niksic  wrote:
> "Tim Peters"  writes:

>> Was the Perl code to which you're comparing it as general as your
>> Python code?

why is it that all of a sudden Tim's postings don't make it to the newsgroup?
I don't recall having seen the above before, and i got the same impression in
other treads as well. And as you all know, whithout Tim's postings there is
hardly any point in reading the newsgroup at all:)

> Robert's solution is ideologically similar to the `grep' solution
> proposed below.

I seem to have missed this grep solution too. or was it part of Tim's posting?

-- 
groetjes, carel



From delgado at Mathematik.Uni-Bielefeld.DE  Wed Sep 22 08:54:21 1999
From: delgado at Mathematik.Uni-Bielefeld.DE (Olaf Delgado)
Date: Wed, 22 Sep 1999 14:54:21 +0200
Subject: May I loop over a changing list?
Message-ID: 

Hi folks!

I am whondering whether the following is considered good (and,
more important: save) programming style in python:

>>> def countdown(n):
...   a = [ n ]
...   for x in a:
...     if x > 0:
...       a.append(x-1)
...   return a
... 
>>> countdown(10)
[10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0]

If my understanding of what 'for x in y' is meant to do is correct, the
for loop should be equivalent to something like

i = 0
while i < len(a):
  x = a[i]
  if x > 0:
    a.append(x-1)
  i = i+1

I think appending to a list while looping over it is a neat way to emulate
a simple queue. The actual example is not quite clever, of course. I
actually use similar pieces of code for doing breadth-first traversals of
graphs.

Cheers,
Olaf





From pinard at IRO.UMontreal.CA  Thu Sep 16 19:13:48 1999
From: pinard at IRO.UMontreal.CA (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 16 Sep 1999 19:13:48 -0400
Subject: Marking translatable strings
References:  
Message-ID: 

Bernhard Herzog  ?crit:

> > 1) Marking strings
> > eight type of strings: ', ", ''', """, r', r", r''' and r"""

> Note, that all these different types are only different at the lexical
> level.  The tokenizer treats them all as STRING-tokens, so could end up
> with a lot of changes to Pythons internals to distinguish them at any
> higher level.

Python internal changes are unwanted merely for marking.  Marking strings
is mainly so tools like `xgettext' could extract them.  Of course, it
does not know Python currently, put this could be improved.  I had my
own version for making experiments, with Emacs LISP, and other languages,
waiting for Ulrich to get back to `gettext' maintainance, if he ever takes
a little break from GNU `libc' :-).

> triple quotes are easier to type than the C-versions because you just
> have to hit the same key several times instead of different keys scattered
> all over the keyboard.

Surely.  The problem is not really the typing, but the legibility of sources.
Translatability marks should really be as little obtrusive as possible,
because in the long run, they are meant to clutter sources all over.
Experience shows that horizontal space is often a valuable resource while
writing code, so we should try to be economical there for overall legibility,
much more than typability.

> With your suggestion about """ you'd expect 'print """red"""' to print
> rouge if your program was properly localized for french, don't you.

Yes.  I am assuming that people normally use """ for long strings, and that
it is very unlikely that people resort to 'print """red"""' in practice.
But even then, this once consumes too much horizontal space.  I think we
need something shorter.  I'm still away from home and did not scrutinise
all this email thread yet, but so far, I guess we could use _(STRING), with
any type of STRING, as both a marker and a gettext caller, and we could use
''"TEXT" or ""'TEXT' for strings needing to be marked only, and translated
only later.  And doc strings.  This would be acceptable overall, I guess.

> A special translated string literal means that the Python interpreter
> would have to generate special bytecode to translate it everytime it's
> executed.

It is not impossible that we could go without any need to modify the
interpreter, at least for prototypes of internationalized packages.
One reason to modify the interpreter would be for speed, maybe, by ensuring
that a given string is translated only once even in a loop, like for what
_() currently does in C when compiled with GNU C.  It might depend on how
exactly strings are intern'ed within Python, which I do not know.  In any
case, it would be nice if internationalization could be achieved portably to
all flavours of Python, just doing nothing on older implementations, anyway,
or doing it slower.  That is, that the string syntax should not be modified.

> > 2) Translating strings
> I dont't think the potential collision with _ in an interactive Python
> session is a problem.

Nice!  Thanks for alleviating this one :-).

> I've used Martin von L?wis' intl module (a wrapper for GNU gettext) [...]

I should at least study what's available.  My main goal is to reach a
global picture soon, because I would like to start internationalizing
all this Python code I'm currently writing.  Besides many small and some
bigger programs here, I surely intend to internationalise the Translation
Project robot suite, after I will have finished to convert it to Python.
The Mailman author (or one contributor? I do not know all people yet! :-)
wrote to me that he wants to get Mailman internationalised, and that
would be a major exercise that would be worth tackling soon, to uncover
difficulties and problems.  I guess I would like to dive in this, or at
least, to participate to the effort of those who are already swimming! :-)

> If you're careful to only use double-quoted strings inside of _(),
> you can even use xgettext to build the initial po-files.

This would not be neat enough to my taste.  I'm quite ready to modify
xgettext, or alter my own extractor in the meantime, to do things properly.

> The only problem I can see with this is that you couldn't get xgettext to
> recognize doc-strings, because you can't mark them with _() or rather N_().
> Actually translating the doc-strings isn't a problem, because tools that
> access the doc-strings could just pass them through _().

Doc strings are recognizable by their location: at the start of a
module or after a def, when sextuple-quoted (hi hi! ' is single-quote,
" is double-quote, ''' is triple-quote, so we could say that """ is
sextuple-quote, or maybe just sex-quote for short :-).  If we retain the
suggestion that we have ''"TEXT" and ""'TEXT', then we coud say that ''"
is quadruple-quote and ""' is quintuple-quote, so giving a meaning to the
missing multipliers...  If I understand things correctly, `xgettext'-type
tools for Python might extract all 4-quoted and 5-quoted strings, as well
as 6-quoted strings when in doc string position, as well as strings given
to some keyword functions (like `_'), exactly as in `xgettext' currently.

> > 3) Setting the textual domain
> With intl, you'd could just define _ locally (i.e. on module level) as
>     def _(text):
>         return dgettext("domain", text)

I hope it might be that simple.  In Emacs LISP, it is a real difficulty,
and notable internationalisation efforts failed because of this oversight,
if I correctly get the stories told to me.  When Richard and I discussed
the matter a few times, I realised that it is difficult to grasp even for
people who know Emacs LISP rather well :-).  And he keeps forgetting :-(.

> > But there are problematic cases, like for when untranslated strings
> > are transmitted to other modules, for being translated there, or even
> > maybe for plain doc strings.

> But this isn't a python-specific problem, is it?  You'd have similar
> problems with C-libraries, or with passing such strings from say a
> third-party plugin to the main program.

I was not thinking of Python specifically, but I was not thinking C either,
where the above problem does not exist in practice.  In bash, say, it is
sufficient to have a stack of textual domains, when getting in or out of
nested scripts, but this is already implemented through forking.  Sourcing
may cause some uncertainty, especially in the context of shell functions.
Problems arise when languages have purely dynamic scoping, because textual
domains are lexically scoped by intent.  I did internationalize a rather
big Scheme program already, but it was all within a single textual domain,
so I did not hit any wall there.  But I would guess that Scheme is much
simpler than Emacs LISP, in actual practice -- I'm not sure.

Python should be simple as well.  Yet, after the XEmacs internationalisation
failure, one good lesson, that we would be foolish to not retain, is that
we should take time to think things out, before starting to think too big.

> For doc-strings you could probably get away with a global variable.

It is not clear to me how doc strings are used in Python.  Some people were
kind enough to reply to one question in a previous thread of this mailing
list, but I still have some study to do before I understand all the replies.
As this area is a bit fuzzy to me, I would fear some oversight if we were
ignoring them too soon...

Thanks a lot, Bernhard, for having taken the time to share your thoughts!

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard



From python-list at teleo.net  Fri Sep 24 13:20:49 1999
From: python-list at teleo.net (Patrick Phalen)
Date: Fri, 24 Sep 1999 10:20:49 -0700
Subject: SIGCHLD, fork, listen
References: <37EC1ACA.8963554B@hk.rfa.org>
Message-ID: <99092410263106.00548@quadra.teleo.net>

On Fri, 24 Sep 1999, Bill Eldridge wrote:

> Another option would be to use inetd, but I need to
> do some host/address verification, and couldn't figure
> out how to latch onto that kind of socket info from
> a new spawned inetd handler using Python.  Ideas?

Consider using Dan Bernstein's open source tcpserver/tcpclient suite, a
reliable replacement for inetd. It also has tcpwrapper-style access
control features.

ftp://koobera.math.uic.edu/www/ucspi-tcp.html



From balance_is_healthy at my-deja.com  Tue Sep 28 06:03:39 1999
From: balance_is_healthy at my-deja.com (balance_is_healthy at my-deja.com)
Date: Tue, 28 Sep 1999 10:03:39 GMT
Subject: where did www.python.org go ?
Message-ID: <7sq3pn$sbj$1@nnrp1.deja.com>

Hi where did... go ?
have they moved ?

other advices on recources are welcomed.
No files please.

thanks

Henrik



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



From peter.stoehr at sdm.de  Mon Sep 13 07:12:05 1999
From: peter.stoehr at sdm.de (Peter Stoehr)
Date: Mon, 13 Sep 1999 13:12:05 +0200
Subject: Announcing: A Python Curses-Based Telnet Client
References: 
Message-ID: <37DCDC05.8E25CCBE@sdm.de>

Hi Moshe,

it would be great if you can send me a copy of your "Python Cursed-Based
Telent Client".

Thanks in advance
    Peter

Moshe Zadka schrieb:
> 
> Is now available.
> 
> I don't have any good public place to put it yet, so
> mail me if you're interested.
> 
> --
> Moshe Zadka .
> INTERNET: Learn what you know.
> Share what you don't.

-- 
Dr. Peter Stoehr 		mailto:peter.stoehr at sdm.de
sd&m AG                 http://www.sdm.de
software design & management 
Thomas-Dehler-Strasse 27, D-81737 Muenchen, Germany
Tel ++49 89 63812-783, Fax -444



From banderson at boi.hp.com  Wed Sep 15 13:50:58 1999
From: banderson at boi.hp.com (Bill Anderson)
Date: Wed, 15 Sep 1999 11:50:58 -0600
Subject: How to sell Python to your boss?
References:  <7qn667$e30$1@nnrp1.deja.com>
Message-ID: <37DFDC82.5D263234@boi.hp.com>

randall_burns_hp at my-deja.com wrote:
> 
> In article ,
>   Petri Mikael Kuittinen 
> wrote:
> > I have lots of programming experience in C, and
> >much less in Python. Despite this I am confident
> >that I would be able to implement
> > the prototype using Python many times faster
> >than using C.
> Well, I agree that you are representative of a
> big constituency out there. We've been making
> substantial use of Python at HP/Cupertino and our
> experience so far indicates that Python can be
> used to develop programs rapidly that can be
> maintained easily.

Some of us, a growing number in fact, are using Python extensively at
teh Boise site as well. :-)

> So far, one of our biggest issues has been finding
> developers enthusiastic about developing in
> Python.  I've seen lots of folks who are rather
> pessimistic about Python's future(my current boss
> is not among them).
> 
> If anyone would like to talk to me about Python
> opportunities at HP/Cupertino, I'd be happy
> to speak with them(we are doing some substantial
> e-commerce applications in Python).

Are you using Zope for any of it?

-- 
Bill Anderson 					Linux/Unix Administrator, Security Analyst
ESBU (ARC)					banderson at boi.hp.com
My opinions are just that; _my_ opinions.



From mgr at propack-data.com  Fri Sep  3 03:48:38 1999
From: mgr at propack-data.com (Matthias Grutzeck)
Date: Fri, 03 Sep 1999 09:48:38 +0200
Subject: getpass.getpass() bugged?
References: <7qm5q0$lct$1@nnrp1.deja.com>
Message-ID: <37CF7D56.C8E2175F@propack-data.com>

super_banane at yahoo.com wrote:
> >>> import getpass
> >>> getpass.getpass()
> Password: hello
> 'hello'
> >>>

Without assigning the returnvalue to a variable the interpreter will
show you the result of the call. If you have a look at the my example,
you will see the behaviour...

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import getpass
>>> getpass.getpass()
Password:
'hello'
>>> s=getpass.getpass()
Password:
>>> print s
hello
>>>

Hope that helps

Matthias

-- 
Matthias Grutzeck              Fon.   +49 (721) 9650 6
Propack Data GmbH              Fax.   +49 (721) 9650 888
Vincenz-Prie?nitz-Str. 1       eMail: mgr at propack-data.com
D-76131 Karlsruhe / Germany    info:  http://www.propack-data.com/



From thomas at bibsyst.no  Wed Sep 29 06:05:23 1999
From: thomas at bibsyst.no (Thomas Weholt)
Date: Wed, 29 Sep 1999 10:05:23 GMT
Subject: What GUI-lib to use
References: <37F1D710.47A82D23@bibsyst.no> <00a301bf0a5b$b0acce80$f29b12c2@secret.pythonware.com>
Message-ID: <37F1E463.6ABD1DE6@bibsyst.no>

Fredrik Lundh wrote:
> 
> Thomas Weholt  wrote:
> > Could Tk get a more modern look perhaps and solve all our problems?
> 
> maybe, but that depends on what "more
> modern look" means.
> 

More like Gtk and KDE. Hotbuttons, themes and all the other crap stolen
from Windows. One of my points is that it`s easier to "sell"
programs/scripts with a modern look, even if they`re free. If anybody
can point me to a page showing off Tk-based apps. that don`t look
aged/out-dated, I`d be more than happy to check it out. 

I don`t mean to offend any die-hard Tk-users, I`d love to use a library
that`s well supported in Python and as a whole, as Tk seem to be, but I
still want to make nice looking apps too. 

Thomas Weholt



From culliton at clark.net  Fri Sep 10 10:02:37 1999
From: culliton at clark.net (Tom Culliton)
Date: Fri, 10 Sep 1999 14:02:37 GMT
Subject: Programmers Editor
References: <37D5916B.606E@zeusedit.com> <00da01bef9e0$6e1b5c70$f29b12c2@secret.pythonware.com> <3dg10p1b5j.fsf@amarok.cnri.reston.va.us> 
Message-ID: <1c8C3.1288$Ym5.38497@iad-read.news.verio.net>

In article ,
Stephan Houben  wrote:
>On 08 Sep 1999 15:00:24 -0400, Andrew M. Kuchling 
> wrote:
>>Anyway, if someone wanted to
>>contribute info on editing Python with Zeus, KDevelop, or whatever, it
>>would be very helpful.  Just follow up to this posting with notes on
>>configuring your editor of choice.
>
>JED is a very nice editor. It comes with Python support right out
>of the box. I actually didn't need anything to configure to work
>comfortably with it. (OK I chose another colour scheme in the menu.)
>
>But make sure you get the latest version (0.98).
>0.97 comes with some older Linux distributions (in my case, RH 6.0),
>but is much less nice.

The latest stable version is actually 0.98-7 available from:

	ftp://space.mit.edu/pub/davis/jed/v0.98/

The LATEST version is B0.99-8 which despite being called a "beta"
version is at least as stable as 0.98-7 and has even nicer Python
support.  It is available from:

	ftp://space.mit.edu/pub/davis/jed/v0.99/

>It is also much more light-weight then Emacs. But then again, there's
>no kitchen-sink built-in.

If Emacs is the editor equivalent of T-Rex, then Jed is Velacoraptor.
Smaller, quicker, more agile, and just as deadly.  I'd hate to have to
live without an emacs like editor, but living with full blown emacs is
another nightmare altogether.

I'm actually using it (GNU Emacs) to compose this message since my ISP
doesn't have a decent version of Jed installed.  When there's any load
on the machine, it burps, and farts, and garbage collects, and auto
saves, and freezes for long stretches.  Not due to code bugs, but just
because it's so bloated that it can't get enough cpu or memory.  Jed
by contrast is about 1/4 the size in memory, uses about 1/2 to 1/4 the
amount of CPU for an equivalent editting task, and in places where
Emacs is painfully sluggish, Jed is peppy and fast.

I guess you could call me a fan... ;-)



From conversy at lri.fr  Thu Sep  2 05:49:22 1999
From: conversy at lri.fr (Stephane Conversy)
Date: Thu, 02 Sep 1999 11:49:22 +0200
Subject: Tkinter and window manager
References: <37CA844C.AE3F285@lri.fr> <038801bef459$434bcb60$f29b12c2@secret.pythonware.com> <37CCEF04.72D60C8@lri.fr> 
Message-ID: <37CE4822.56E93593@lri.fr>

there are two things:

First:

I want to be notified when a document window is raised on top of the
windows stack.
Looking the Xlib and tk doc, we can see that a  event is
triggered when
XRaiseWindow() is called (which I think is called by the window manager
when one
clicks on the bar).

The problem is that when I bind a toplevel to  I don't receive
any events.

Second:

The palette must stay on top of the other window.
I think I can do that by calling tkraise on the palette window each time a
document window is raised.

So the problem is:
How to catch that a document window has been raised ?


Thinking about it a little bit more, and after I saw that this little
script in tk doesn't work:
toplevel .t
proc toto {} { puts "yep" }

bind .t  toto

that is, it soesn't write "yep" when the toplevel is raised, I guess it's a
reparent problem,
because it's the window created by the wm that receives ...

    stef


> True, but one a FocusIn event, you could make palette rise above the
> other windows.
>   def coconut_dropped(self, event):
>     event.widget.tkraise()
>
> But if you are asking "how can I have one window affect the widgets
> in another window?"  That's is much more of a design issue.  I would
> suggest having an "application" object which abstracts what you need
> done and calling methods on that object from the palette.
>
>   class DrawApp:
>     def __init__(self, *args):
>       self.process_arguments(args)
>       self.current_operation = 'pointer'
>     def change_to_pen(self):
>       self.current_operation = 'drawWithPen'
>     def change_to_brush(self):
>       self.current_operation = 'drawWithBrush'
>     def start_polygon(self):
>       self.current_operation = 'drawPolygon'
>       self.operation_data = []
>       self.op_state = 'waitingForFirstPoint'
>     ...
>
> This makes the palette definition seperate from the implimentation of
> the canvas operations.
>
> This approach can also work with multiwindow focus: the palette affects
> the last/current window (different "DrawApp" objects) with the focus.
>
>   -Arcege

--
St?phane Conversy
http://www-ihm.lri.fr/~conversy/
mailto:conversy at lri.fr






From holger at phoenix-edv.netzservice.de  Thu Sep 16 07:04:48 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Thu, 16 Sep 1999 11:04:48 GMT
Subject: timesync with python?
References: <37D66AF3.6A3E96F7@phoenix-edv.netzservice.de> <002701befabb$1a884880$f29b12c2@secret.pythonware.com>
Message-ID: <37E0CED0.8F00D422@phoenix-edv.netzservice.de>

High,

another question I couldn't solve:
I'm searching for a function to set local time, but
couldn't find it in description of module 'time' 
neither other documentations.

Any hints?

Thanx,
Holger



From marshall.taylorNOcpSPAM at beenz.com  Thu Sep  9 18:26:15 1999
From: marshall.taylorNOcpSPAM at beenz.com (Marshall T.)
Date: Thu, 09 Sep 1999 15:26:15 -0700
Subject: Module to open URLs that set cookies?
Message-ID: <15d84bec.887c3642@usw-ex0107-051.remarq.com>

Hello,

I would like to know how to write an http client that will 
accept cookies when opening webpages.  I'm attempting to 
write a module that open a page and note the latency in 
doing so.  I've been successful in doing this with the 
sockets and urllib library modules, but have been foiled by 
pages that set cookies.  Is there a canned library for 
doing this in Python?

Thanks for your time,

Marshall T.

* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!




From davidopp at megsinet.net  Thu Sep  2 03:24:59 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 02 Sep 1999 03:24:59 -0400
Subject: Bring Out Yer Dead
Message-ID: <37CE2648.D4254FCC@megsinet.net>

I am going to give up on trying to get the PIL to work with Pythonwin
1.5.2 unless someone who has it working with their setup can give me
explicit instructions on how to do this.  I am unable to implement
Robert Kern's instructions due to the fact that I cannot get my Win Zip
program to unzip the tar.gz file format.

So ... time for me to put PIL for Pythonwin 1.5.2 on the pile of
software corpses and set it ablaze.  Can you feel the heat?!  Anyone
have any marshmallows?

David O.




From phd at emerald.netskate.ru  Sun Sep 12 07:07:36 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Sun, 12 Sep 1999 15:07:36 +0400 (MSD)
Subject: Problem installing PyGresql in Python 1.5.2-2 on RH 5.0
In-Reply-To: <1103_937113988@ns1.dicksonstreet.com>
Message-ID: 

On Sun, 12 Sep 1999, Felix Thibault wrote:
> I copied pgmodule.c into my config directory (that is:/usr/local/lib/python1.5/config/) and added a line to Setup, but when I did 'make Makefile.pre.in', judging from 
> the output and 'ps' in another virtual terminal 'make do-it-again' (from Makefile.pre.in) just kept making itself over and over. I tried just make'ing the Makefile after I 
> killed it, but that looped, too. I tried a couple of other things but I always had to kill it while it was making 'do-it-again'? 
> I think I installed Tkinter the same way (except I only had to uncomment the lines in Setup, they were already there) without a similiar problem.
> 
> Does this sound familiar to anyone? 
> (the line I added to Setup is : _pg pgmodule.c -I/usr/include/pgsql -L/usr/lib -lpq )

   After adding the line do "cd ..; make" - just make, not make Makefile.pre.in.

> Thanks in advance-
> Felix

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From mal at lemburg.com  Fri Sep  3 10:46:36 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 03 Sep 1999 16:46:36 +0200
Subject: UK Python Programming Job Opportunity
References: <7qof8u$mh2$1@newsreader1.core.theplanet.net>
Message-ID: <37CFDF4C.1D96A89A@lemburg.com>

John Parkey wrote:
> 
> If there is anybody who is interested in a 6-9 month contract using Python,
> in Bucks UK, please drop me a line.
> 
> jparkey at rapidbs.com
> 
> (I hope nobody is offended by my posting this to the newsgroup?)

I hope we get more and more of these postings :-)

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   123 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From gtnorton at my-deja.com  Fri Sep  3 15:47:10 1999
From: gtnorton at my-deja.com (gtnorton at my-deja.com)
Date: Fri, 03 Sep 1999 19:47:10 GMT
Subject: Any trouble with Tcl/Tk 8.2 ?
References: <7qn33a$c2m$1@nnrp1.deja.com> <7qo8q5$oqmb@eGroups.com>
Message-ID: <7qp8jp$tvf$1@nnrp1.deja.com>

In article <7qo8q5$oqmb at eGroups.com>,
  john.grayson at gsc.gte.com wrote:
> john.grayso- at gsc.gte.com wrote:
> original article:http://www.egroups.com/group/python-list/?start=66500
> >                 I was looking to download the new Tcl/Tk 8.2
> > Anyone aware of any bugs or problems with it when used with Python?
> > Is it worth it?, or should I just stick with the Tcl/Tk 8.0 that
came
> > with Python.
> >                                            Thanks,
> >                                                 gtnorton
> Stick with 8.0.5 for the moment. I've have problems trying to use
> 8.1 and 8.2 and I have not had time to investigate further. I believe
> that Guido has seen these problems already...
>
>     john
>
>
                        ********THANKS, JOHN********


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From akuchlin at mems-exchange.org  Wed Sep  8 14:20:38 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 08 Sep 1999 14:20:38 -0400
Subject: Feast or famine!
References: 
Message-ID: <3dk8q11czt.fsf@amarok.cnri.reston.va.us>

init__self at my-deja.com (Python Addict) writes:
>  - Python Reference Manual; February 19, 1999, Release 1.5.2
>    Guido Van Rossum, Fred L., Jr. Drake
>    ISBN: 1583483748
>    (What's all this then? No piccie on Amazon.com or description)

Looks like the online manual, typeset and bound by the company behind
www.toexcel.com.  Other titles available, like the reference manual
and tutorial, are listed at
http://www.toexcel.com/bookstore/author.asp?author%5Fid=1436 .  (I
assume you arranged this, Guido?)

Time to start a Python shelf, I guess.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Dreams are tricky buggers. You can't trust them.
    -- Hob Gadling, in SANDMAN #59: "The Kindly Ones:3"




From fredrik at pythonware.com  Tue Sep  7 07:12:07 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 7 Sep 1999 11:12:07 GMT
Subject: [Newbie] Win95 Python 1.5.2 with Tk,  not recognised
References: <7r2o30$nb1$1@bunyip.cc.uq.edu.au>
Message-ID: <01ff01bef921$d3a4ea30$f29b12c2@secret.pythonware.com>

> I had to comment out the reference to mousewheel event in listbox.tcl and
> textbox.tcl to get Tkinter() to work. Why? The download came from 1.5.2 from
> the python.org site with Tcl8.0. The error message raised was that
> mousewheel event was not recognised.

version mismatch; for more info, see:
http://www.pythonware.com/people/fredrik/fyi/fyi02.htm






From nharlow at my-deja.com  Thu Sep  2 14:02:25 1999
From: nharlow at my-deja.com (nharlow at my-deja.com)
Date: Thu, 02 Sep 1999 18:02:25 GMT
Subject: Generating an event in Tkinter
References: <7qfc5l$mll$1@nnrp1.deja.com> <011701bef37e$bcd9ece0$f29b12c2@secret.pythonware.com> <7qhl27$cgp$1@nnrp1.deja.com> <022101bef44f$8e98b9c0$f29b12c2@secret.pythonware.com> <7qk0ci$3m9$1@nnrp1.deja.com> <014801bef523$808828a0$f29b12c2@secret.pythonware.com>
Message-ID: <7qme39$s4a$1@nnrp1.deja.com>

Thank you so much for all your help.  It was the widget.update() that
did it for me.  Thanks.  I look forward to seeing your Tkinter book
from O'Reilly.

Nick Harlow


In article <014801bef523$808828a0$f29b12c2 at secret.pythonware.com>,
  "Fredrik Lundh"  wrote:
>  wrote:
> > import time
> > name="Analyze"
> > while i< 100:
> > print "%s: This is %s time through the loop"%(name,i+1)
> > widget.event_generate("<>")
> > widget.bind("<>", lambda e, y=1, x=time.sleep: x(y))
> > i=i+1
>
> > It just locks up; I don't even get any errors.
>
> no errors at all?  not even some-
> thing like:
>
> Traceback (innermost last):
>   File "example.py", line 3, in ?
>     while i< 100:
> NameError: i
>
> ???
>
> if I set "i" to 0, and create a widget,
> your program works just fine.  or in
> other words, the loop will now send
> 100 events to the widget, which,
> whenever it gets around to check
> the event queue, will block for
> one second per event.
>
> the following version increases the
> chance that the widget actually gets
> around to process your event:
>
> from Tkinter import *
> import time
>
> widget = Tk()
> widget.bind("<>", lambda e, y=1, x=time.sleep: x(y))
>
> name="Analyze"
>
> for i in range(100):
>         print "%s: This is %s time through the loop"%(name,i+1)
>         widget.event_generate("<>")
>         widget.update() # process events
>
> mainloop()
>
> btw, I think you might be better off using a
> timer [1], but that's me...
>
> 
>
> 1) http://www.pythonware.com/library/tkinter/introduction/basic-
widget-methods.htm
> => alarm handlers and other non-event callbacks
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gerrit.holl at pobox.com  Sat Sep 11 10:35:30 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 11 Sep 1999 16:35:30 +0200
Subject: whrandom.randint(): no LONGs?
Message-ID: <19990911163530.A11476@optiplex.palga.uucp>

Hello,

whrandom.randint() always try to convert the given arguments to a integer. Why?
I'm not able to use Longs now. I don't require it directly, but it might be a
useful thing to know the "why". I can learn from it...

regards,
Gerrit.
-- 
Unix is easy: type 'man man' and the rest comes of itself.



From ransen at nemo.it  Tue Sep  7 18:13:43 1999
From: ransen at nemo.it (Owen F. Ransen)
Date: Tue, 07 Sep 1999 22:13:43 GMT
Subject: Embedding Python into an MFC app - thread problems.
References: <7pv1jd$gmd@newsops.execpc.com>
Message-ID: <37d78aef.6768198@news.newsguy.com>

On Tue, 24 Aug 1999 16:16:44 -0500, "John DeFord"
 wrote:

>I've embedded Python + win32 extensions into my MFC app with the following
>code scrap in my app's InitInstance member (taken from Pythonwin embedding
>documentation):
>John

Er, where did you get the Pythonwin embedding documentation?
I've got the DrDobbs Python Resource CD-ROM and can find no
real help in embedding Python in Windows apps. Any pointers
much appreciated...

TIA

Owen



--------------------------------------------------
Owen F Ransen                       ransen at nemo.it
http://www.ransen.com       the home of Repligator 
Winner SIAF awards - Best Graphics Program of 1999



From tim_one at email.msn.com  Sun Sep 19 18:58:44 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 19 Sep 1999 18:58:44 -0400
Subject: Lazy file.readlines()?
In-Reply-To: <87emfzmzor.fsf@slip-32-101-160-116.ma.us.ibm.net>
Message-ID: <000601bf02f2$85f30260$102d153f@tim>

[meow, about his or her qio module at
 http://members.xoom.com/meowing/python/
]
> ...
> Also added some convenience features like user-specified line
> terminators, automatic chopping, and a way to read SMTP-like sockets
> wicked fast.

Yes, qio.readline() is about the most frightening function in the Python
universe now .

> After adding the requisite bloat, vanilla qio.readline() is still
> about 5-8 times as fast as native Python looping through big text
> files (where stdio is glibc's).

I compiled it for Win32 (btw, "mode_t" was not defined, & there are about a
half dozen warnings about mixing signed and unsigned ints), with
optimization cranked up.  Under Win95 it was between 2-3 times as fast as
1.5.2 Python doing a native readline() loop over an 11Mb text file.  It's
hard to believe that MS's grudging implementation of stdio is that much
better than glibc's, so not sure what to make of this -- possibly that
*nothing* can run really fast under Windows <0.9 wink>.

> ...
> Details at 

Nice work!

if-only-windows-had-any-text-files<0.7-wink>-ly y'rs  - tim





From englim at pc.jaring.my  Wed Sep  8 00:13:20 1999
From: englim at pc.jaring.my (lec)
Date: Wed, 08 Sep 1999 12:13:20 +0800
Subject: Bug in time module?
Message-ID: <37D5E25F.1F7F0A74@pc.jaring.my>

Is this considered a bug since there is no Feb 31, 1999 or Feb 29, 1999?

>>> import time
>>> time.strptime('31/02/1999', "%d/%m/%Y")
(1999, 2, 31, 0, 0, 0, 2, 62, 0)
>>> time.asctime(time.strptime('31/02/1999', "%d/%m/%Y"))
'Wed Feb 31 00:00:00 1999'
>>> time.strptime('29/02/1999', "%d/%m/%Y")
(1999, 2, 29, 0, 0, 0, 0, 60, 0)
>>> time.asctime(time.strptime('29/02/1999', "%d/%m/%Y"))
'Mon Feb 29 00:00:00 1999'




From da at ski.org  Wed Sep 29 13:01:00 1999
From: da at ski.org (David Ascher)
Date: Wed, 29 Sep 1999 10:01:00 -0700 (Pacific Daylight Time)
Subject: Is togl missing from PyOpenGL binary?
In-Reply-To: 
Message-ID: 

On Wed, 29 Sep 1999, Olaf Delgado wrote:

> Hi folks!
> 
> I have installed Python 1.5.2 and PyOpenGL 1.5.5 (I think) in binary form
> on my Windows box. Now many of the example programs require togl (and I
> would like to use it too), but tcl (I have installed the tcl version that
> came with python) complains it can't find it. Searching for togl.* didn't
> reveal anything, so I'm pretty sure there is no dll anywhere. I suppose
> none of the binaries (which I got from the standard places pointed to at
> www.python.org) have not been compiled with togl either. Can the people
> who have kindly provided the binaries confirm this? Does anyone have a
> togl.dll handy? I think it should be in the binary distribution anyway,
> unless, of course, there are important reasons to exclude it.

It is in the installer, but the installer fails sometimes for unexplained
reasons.  I've just put a page describing how to fix this manually at:

   http://starship.python.net:9673/crew/da/Code/PyOpenGL/Togl

Good luck.

can't-wait-for-distutils-to-do-that-job'ly yours, david




From Marcus.Bergmann at isst.berlin.fhg.de  Wed Sep 15 06:07:49 1999
From: Marcus.Bergmann at isst.berlin.fhg.de (Marcus Bergmann; isst.fhg.de)
Date: Wed, 15 Sep 1999 12:07:49 +0200
Subject: make test fails
Message-ID: <37DF6FF5.625B8F96@isst.berlin.fhg.de>

Hi,

i've got trouble during the installation of Python 1.5.2. I configured
with "./configure --prefix=/opt/PYTHON-1.5.2 --with-gcc --with-threads".
Then i did run "make". Both worked fine. But "make test" fails with
"test_popen2". The OS ist Solaris 2.6 on a Sun Ultra2.

Who can help me out?

Thanks
-- 
Mit freundlichen Gr??en
Marcus Bergmann

ISST		| Telefon  : 030/24306-463
Bereich-	| Telefax  : 030/24306-599
Mollstr. 1	| E-Mail   : Marcus.Bergmann at isst.fhg.de
10178 Berlin	| DFN-Mail : G=Marcus;S=Bergmann;O=ISST;P=FHG;A=D400;C=de



From mwh21 at cam.ac.uk  Thu Sep  2 05:55:38 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 02 Sep 1999 10:55:38 +0100
Subject: adding to dictionaries
References: <37CE4626.8FCEA15B@xsoft.co.at>
Message-ID: 

"Gerhard W. Gruber"  writes:

> Is it possible to add an entry to an already existing dictioniary?

Umm... yes?

d={1:2}
d[2]=3

d[2] => 3

although I can't help thinking that's not what you're asking...

confused-ly y'rs,
Michael



From algaba at my-deja.com  Sat Sep 18 04:58:00 1999
From: algaba at my-deja.com (Manuel Gutierrez Algaba)
Date: Sat, 18 Sep 1999 08:58:00 GMT
Subject: TCL code producing python code
Message-ID: <7rvk6m$nt6$1@nnrp1.deja.com>

Usually there're flame wars between TCL and python.

Today I've found a nice way of working together to get the
easiest.

I use my consdiag to generate lots of classes. The problem is
that I have to write a lot of python code. But using TCL is much
easier, for example this code:
#!/usr/bin/tcl

proc escribe {t} {
    puts $t
}

proc escribenonl {t} {
    puts -nonewline $t
}

proc escribecoma {t} {
    puts "$t,"
}

proc gensimple {t modulo clase atributos} {
    escribe "def genera_$t\(con_atributos =None, con_funciones=None):"
    escribe "\tc = a_simple_class()"
    escribe "\tc.do_name_clase($modulo.$clase.__name__)"
    escribenonl "\tc.do_attributes(\["

    foreach i $atributos  {
        escribecoma "\'$i\'"
    }
    escribe "\])"
    escribe "\tif not con_funciones is None:"
    escribe "\t\tc.do_functions\($modulo.$clase\)"
    escribe "\tc.do_lines\(\)"
    escribe "\treturn c"
}

escribe "from consdiag import *
import dibcla
import nucleo
import serv_sok"

gensimple nucleo nucleo nucleo [list puerto_clips puerto_visualizador \
        temporizador mapa_X mapaY]


proc genera_acceso { t n} {
    escribe "\tv=genera_$t\(\)"
    escribe "\tv.generate_latex_code\(\'result$n.tex\'\)"
}

escribe "def genera_fic():"

genera_acceso nucleo 1


escribe "if __name__=='__main__':
\tgenera_fic()
"

Produces a bit of :

from consdiag import *
import dibcla
import nucleo
import serv_sok

def genera_nucleo(con_atributos = None, con_funciones = None):
    c = a_simple_class()
    c.do_name_clase(nucleo.nucleo.__name__)
    c.do_attributes(['puerto_clips', 'puerto_visualizador',
                     'temporizador', 'mapa_X',
                     'mapa_Y', 'mapa_Z',
                     'periodo_actualizacion_mapa', 'dimensiones',
                     'turnador', 'interprete',
                     'maximo_ciclos', 'dicc_jugadores'])
    if not con_funciones is None:
        c.do_functions(nucleo.nucleo)
    c.do_lines()
    return c

def genera_servidor_de_socket(con_atributos = None, con_funciones =
None):
    c = a_simple_class()
    c.do_name_clase(serv_sok.servidor_de_socket.__name__)
    c.do_attributes(['funcion_a_ejecutar', 'tiempo',
                     'conn'])
    if not con_funciones is None:
        c.do_functions(serv_sok.servidor_de_socket)
    c.do_lines()
    return c

def genera_asociacion_nucleo_servidor_de_sockets():
    n = genera_nucleo()
    g = genera_servidor_de_socket()
    e = union_association()
    e.add_union(n, g, "1:1", "1:1")

    return e


def genera_fic():
    t = genera_nucleo()
    t.generate_latex_code('result1.tex')
    t = genera_asociacion_nucleo_servidor_de_sockets()
    t.generate_latex_code('result2.tex')

if __name__=='__main__':
    genera_fic()


I think that this way of producing python code by TCL will
save me some hundreds lines.

I hope this may be useful for anyone else.

--
-------------
Manolo


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From clement at ans.com.au  Thu Sep 23 09:10:23 1999
From: clement at ans.com.au (Clement)
Date: Thu, 23 Sep 1999 23:10:23 +1000
Subject: Python and Postscript
Message-ID: <37EA26BF.63766B3C@ans.com.au>

Hi,

Does anyone know of tools to print forms on a postscript printer in
Python?
-- 

Regards

Clement



From adjih at crepuscule.com  Wed Sep 22 09:35:55 1999
From: adjih at crepuscule.com (Cedric Adjih)
Date: 22 Sep 1999 13:35:55 GMT
Subject: Linux Dependencies! Grrr...What now?
References: <37E8066F.1C13BF33@alliedsignal.com>
Message-ID: <7salvr$6du$1@ites.inria.fr>

H. P. Friedrichs <" HPeter.Friedrichs"@alliedsignal.com> wrote:
> While I have having successfully written some scripts for python 1.4 (on
> my rh5.0 linux box), and successfully installing mysql, I have been
> trying for a week to figure out how to get the mysqldb module to work
> (to allow python to talk to mysql).

> First, I found out that mysqldb will only work with python 1.5.2...which
> necessitated an upgrade. Dependencies! Dependencies! I ended up
> installing python, python-devel, blt, tix _tkinter and tkinter. (I've
> been warned that as the result of my upgrade, I have now "broken"
> netcfg, but I'll deal with that on another day.)

> Now I go to install mysqldb, and it complains, among other things, of
> missing files and directories. Looks like I need mysql-devel.
> Dependencies! Dependencies!

> I go to the web and find it. I try to install the rpm, and it won't. It
> seems I need glibc6 (2.0 and 2.1). But what about everything compiled
> with the existing library? What happens if I "upgrade" to glibc6? What
> do I risk breaking by doing that?

  I don't know. Installing newer libraries on a older system seems
prone to lead to upgrade escalation. The only solution I found
pratical is to have a /usr/local or /home//External, and to
recompile every development tool/library that doesn't work out of the box
with --prefix=/home//External (gtk, glib, swig, wxWindows, PIL,
wxgtk...), and some proper LD_LIBRARY_PATH/PATH settings.

  If it is possible, just take the source rpms, extract them, and
try to compile manually.
Sorry I can't help more.

-- Cedric




From cors at recercai.com  Wed Sep  8 16:42:45 1999
From: cors at recercai.com (Carlos Ors)
Date: Wed, 8 Sep 1999 22:42:45 +0200
Subject: how can I implement a java interface with a jpython class?
Message-ID: <37d67617.0@sahara.cablecat.com>

I'm working with python embedded in java (I'm just working with
PythonInterpreter java class). I have a  java interface and I want to
implement it with a python class. How can I do it?

Where can I find documentation/exemples about all classes that are there in
org.python.core?

                                                     Thank you in advance!


Please, answer me (if you can) to karlos at softhome.net






From bthomas at trey-industries.com  Mon Sep 13 13:36:39 1999
From: bthomas at trey-industries.com (Ben Thomas)
Date: Mon, 13 Sep 1999 17:36:39 GMT
Subject: Python or Java or "X" from me and my little job
References: <37DA8D01.3D6616A@trey-industries.com> 
Message-ID: <37DD356F.86D56591@trey-industries.com>

Actually I just want to
1)get current date and come up with its string like: 01251999
2)zip some files and name the archive 01251999
3)ftp the archive off site

this is my first task but I know there will be others. It isn't hard to code
I am just new to Python.

Ben

"Magnus L. Hetland" wrote:

> Ben  writes:
>
> > I am new to Linux. I have Python 1.5.2 up and working. I want to have
> > Python grab some files, ZIP them, and FTP them offsite. I will let CRON
> > start the job every night. Since I am just getting to look at Python, I
> > wanted to ask to make sure I don't pick the wrong tool for the job.
> >
> >
> > 1)Am I selelcting a good tool to get light weight sys admin jobs done?
>
> Well... If it is a light weight sysadmin-job, maybe a simple shell
> script would do it? (Then you could move it to Python if it grew...)
>
> I mean - if the set of files is fixed (or given as a command-line
> argument) I guess this thing you are talking about could be done in
> two lines in any standard shell (assuming that you have a good ftp
> client like ncftp.)
>
> --
>
>   Magnus
>   Lie                                     Please forward all spam to
>   Hetland                                                uce at ftc.org




From bwarsaw at python.org  Thu Sep 16 13:14:05 1999
From: bwarsaw at python.org (bwarsaw at python.org)
Date: Thu, 16 Sep 1999 13:14:05 -0400 (EDT)
Subject: Marking translatable strings
References: 
Message-ID: <14305.9565.35913.349267@anthem.cnri.reston.va.us>

>>>>>  writes:

    > I had a real strange idea :-).  I first quickly dismiss it, but
    > it is so simple that I prefer to ponder it again, and share and
    > debate it a little first, maybe.

Not so strange!  There's a big push to internationalize Mailman
 and I've promised it as an important goal for the next
release.  Unfortunately I haven't had much time to work on Mailman
lately so my few tentative steps haven't even made it into the CVS
tree yet.  There's been discussion on the
mailman-developers at python.org list and you can peruse the archives for 
much of the current thinking.  Sorry I don't have time to give exact
URLs but the archives are available at:

    http://www.python.org/pipermail/mailman-developers

Just a few quick comments, and then a pointer to some code...

    > Python has no preprocessing, no special string syntax for
    > markability, and moreover, it has doc strings!  So, at first
    > glance, it looks difficult.

The first attempts in the Mailman community for marking strings
followed the reasoning you did.  I don't like it because I want to be
able to use any string spelling as translatable strings.  Anything
else would be way to error prone in a project that has so many hands
on it.

I considered lobbying Guido for another string prefix, say i"..." to
mark translatable strings, but decided against that for a number of
reasons.  The approach I think I will take is one that you also
suggested, using _("text") as the marker for translatable strings.
This has the advantage that `_' is a perfectly good name for a
function, which could be the identity function if, say the end user
doesn't have the I18N stuff installed on their system.  But it could
also easily be the name of the function that does the string lookup,
returning the translated string.  Yes _ has the disadvantage of
already having meaning in interactive mode, but I wasn't too concerned 
about that (maybe I'm being naive tho', as I mentioned I haven't
really gotten /that/ far yet).  If it's a problem __("text") would be
fine (double underscore).

What I /have/ done is start work on a pygettext.py script that is in
the spirit of GNU gettext.  Instead it knows a lot about Python
syntax, including the 9 billion ways to spell `string'.  pygettext
uses Python's standard tokenize module to scan through a file of
Python code, search for _(text) matches, where text is any of 'text',
"text", '''text''', etc.  The output is pretty close to .po style,
with strings normalized to .po format.  The output formatter may still 
have some bugs in it.

Since I don't envision much time to hack on this in the near future,
I'll just go ahead and post the code on my pyware page:

    http://www.python.org/~bwarsaw/software/pyware.html

Give me about 10 minutes and then feel free to grab it.  I'd love to
get any comments or patches that you might come up with.

-Barry



From warren at nightwares.com  Wed Sep  8 04:51:46 1999
From: warren at nightwares.com (Warren 'The Howdy Man' Ockrassa)
Date: Wed, 08 Sep 1999 03:51:46 -0500
Subject: And now for something completely boneheaded...
Message-ID: <37D623A2.DC09F0CC@nightwares.com>

It's... the obvious question of the day. Or perhaps the hour.

So I have

  import sys

>From that how can I find out what sys can do? I have tried things like

  sys.methods

and

  sys.functions

and

  sys.*

and even, in a moment of extreme desperation

  sys.incest

and we won't talk any more about that.

I have received no joy in response. So with a given module, once
imported, what is the way by which I can find out what's defined in the
module and, ideally, how to access it?

What is the obvious which I am missing?

TIA!

--
    warren ockrassa | nightwares | mailto:warren at nightwares.com
                    http://www.nightwares.com/



From kaine_the_bloody at my-deja.com  Thu Sep  9 19:33:36 1999
From: kaine_the_bloody at my-deja.com (Kaine)
Date: Thu, 09 Sep 1999 23:33:36 GMT
Subject: WYSIWYG GUI maker V - Guido
References: <37D5AA97.2C4BF97D@megsinet.net>
Message-ID: <7r9g4g$803$1@nnrp1.deja.com>

In article <37D5AA97.2C4BF97D at megsinet.net>,
  David Oppenheimer  wrote:
> Dear Group,
>
> Looking for input here.  What kinds of features and functionalities
> would you want to have in a WYSIWYG GUI maker?
>
> Thanks,
>
> David O.
>
>
Component based architecture. This could be corba objects. Drag and
drop. The class library should be extensible. A good editor would be
nice or integration with EMACS. A class browser. A project manager.
Basically just copy Delphi's IDE. :)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From aahz at netcom.com  Thu Sep 16 00:44:24 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Sep 1999 04:44:24 GMT
Subject: high school programming & python
References:   <37DFED78.6B258366@bioreason.com> 
Message-ID: <7rpsj8$sqt@dfw-ixnews6.ix.netcom.com>

In article ,
James Whitley Moore  wrote:
>
>Why not end this thread in the usual way?
>I never wanted to be a high-level language programmer.
>I always wanted to be. . .

Oh, I'm a programmer and I'm OK
I code all night and I sleep all day
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From kern at caltech.edu  Sat Sep 25 03:27:10 1999
From: kern at caltech.edu (Robert Kern)
Date: Sat, 25 Sep 1999 00:27:10 -0700
Subject: CygWin version
References:  <37EC7864.73882599@caltech.edu>
Message-ID: <37EC794E.AEB55676@caltech.edu>

Pardon the last post.
It was from me (Robert Kern).  

Serves me right for using the computer lab to read USENET.

Duf wrote:

No, I wrote.

> Sorry for piggybacking.
> Posted and e-mailed.

Umm... I don't seem to have things configured right on my end.  The
e-mail
didn't go through.  :-(

[etc.]

> --
> Robert Kern
> kern at caltech.edu

Ditto.



From amk at mira.erols.com  Mon Sep  6 16:38:30 1999
From: amk at mira.erols.com (A.M. Kuchling)
Date: 6 Sep 1999 20:38:30 GMT
Subject: how often does re.compile compile?
References: <7r1451$3n7$1@nnrp1.deja.com>
Message-ID: 

On Mon, 06 Sep 1999 19:20:08 GMT, 
	wade at lightlink.com  wrote:
>only within one function. If I say re.compile() inside the function
>definition, does the pattern get compiled afresh every time the function
>is called?

	Maybe.  There's a cache containing the most recent 20-odd
patterns, so if you're using fewer patterns, the re.compile() will do
a dictionary lookup, find the precompiled cached pattern, and just
return it without doing any more work, but the common idiom is to
precompile them yourself.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Actors... I don't like actors. It's a profession based on lies and disrespect.
Pretending to be what you are not.
    -- The emperor Augustus in SANDMAN #30: "August"




From pointal at lure.u-psud.fr  Fri Sep  3 06:40:03 1999
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Fri, 03 Sep 1999 10:40:03 GMT
Subject: Pirate -> RAD
References: <7qno38$6uk$1@news.is.co.za>
Message-ID: <37cfa533.359049255@news.u-psud.fr>

On Fri, 3 Sep 1999 08:00:45 +0200, "Jacques Oosthuizen"
 wrote:

>: Hi All
>: 
>: I am busy working on a tool I named Pirate ("Python RAD") it has a component
>: library similar to Delphi and shall in the future work with win32,gtk,qt. It
>: will have a IDE.
>: 
>: Before I do to much and work to hard on it, will anyone be interested in
>: using a tool like this.

Yes (and I think many people are interested).

Maybe you can work with the Parrot author...

A+

Laurent.




From fredrik at pythonware.com  Wed Sep  1 05:03:11 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 1 Sep 1999 09:03:11 GMT
Subject: CGI speedup strategies
References: <37CCE7C9.D340BBF5@strs.co.uk>
Message-ID: <036801bef458$d262c3e0$f29b12c2@secret.pythonware.com>

Ian Clarke  wrote:
> Has anyone tried actually implementing a Python webserver which just
> starts a new thread and runs a piece of Python code in it every time
> someone makes a HTTP request?  This server could then cache any objects
> (such as files, or SQL connections) in memory to be used by any Python
> scripts (this would obviously require all code to be thread-safe).

http://www.nightmare.com/medusa (non-threaded, ultra-fast)
http://www.zope.org (threaded, based on medusa)






From cjc26 at nospam.cornell.edu  Sun Sep 12 22:03:07 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 13 Sep 1999 02:03:07 GMT
Subject: Python example: possible speedup?
References: <000801befc94$69181d60$6aa2143f@tim>
Message-ID: 

Pada Sat, 11 Sep 1999 16:29:56 -0400, Tim Peters bilang:
| I'm sure your absence from the list of approved readers is a
| temporary glitch due to inferior operating systems (like Windows, MacOS,
| EPOC32, Unix, Linux and BeOS)

You just ruled out a whole bunch of OS's there...so what does the timbot
run on then?  ITS?  A Lisp Machine?  AmigaOS?


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From tim_one at email.msn.com  Mon Sep 27 21:35:12 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 27 Sep 1999 21:35:12 -0400
Subject: Circular refs
In-Reply-To: <000701bf0909$46bcb0b0$3acbd9c2@peridot.optichrome.com>
Message-ID: <000801bf0951$b5b67da0$18a0143f@tim>

[Adrian Eyre, on immortal cyclic trash]
> I can dream up a million ways to kludge around this bug, but I
> was referring to a 'fix' to the Python source code. How hard can
> it be?

It's unclear, but I can give you a lower bound:  in all the years people
have been whining about how easy this is to "fix", nobody has managed to fix
it.

your-turn-ly y'rs  - tim





From fredp at multimania.com.nospam  Wed Sep 29 04:56:37 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 29 Sep 1999 08:56:37 GMT
Subject: YA-Newbie: Progress (but still baffled by deepcopy/pickle issues...:)
References: <8E48C2767PaCmAnRDLM@161.48.81.148>
Message-ID: <8E5062134PaCmAnRDLM@161.48.81.148>

Following-up on myself:

Thanks to hints by Tim Peters (ly-yrs) and Jeremy Hylton, I have now 
identified the specific point of failure in my code, and a workaround, so I 
can now resume development -- thanks !

The uncomfortable part of the sitiation is that while I've found what 
doesn't work and what works as a replacement, I still don't understand why 
one solution fails and not the other. This bothers me, so I'll try to 
explain here what's going on in case someone gets interested and can shed 
some light on the matter...

As I said it's a card game, with a simple object structure. The problem 
happens with this combination:

First I have 3 module-global variables that hold wxWindows bitmap handles:

Faces1 = wxBitmap(...) # card faces held in 2 separate bitmaps
Faces2 = wxBitmap(...)
CardBack = wxBitmap(...)

Then I have the Card class :

class Card :
    	def __init__(self, index) :
    	    	self.color = ...
    	    	self.rank = ...
    	    	if ... : self.src = Faces1
    	    	else : self.src = Faces2
    	    	self.Xoffset = ...
    	    	self.Yoffset = ...
    	    	...

The point in this pseudocode is that each Card instance holds a reference 
to the Bitmap the card image is to be fetched from (and its offsets in it).

Next is the Stack class. It is essentially a list of Card instances, with a 
few simple data attributes (integers) and methods for displaying, moving 
around, etc, as in :

class Stack :
    	def showrow(self, row) :
            srcDC.SelectObject(self.Cards[row].src)
            dstDC.Blit(destx, desty, self.Cw, self.Ch, 
            srcDC,self.Cards[row].Xoffset,self.Cards[row].Yoffset,xCOPY)
            srcDC.SelectObject(wxNullBitmap)

The game state is almost entirely contained in a list of a dozen Stack 
instances (Cols), so saving and restoring it should be as easy as:

   def OnGameSave(self, event):
        SaveGame.Cols = deepcopy(self.Cols)

   def OnGameRestore(self, event):
        self.Cols = deepcopy(SaveGame.Cols)
        self.Refresh(1,None)

Unfortunately, as I mentioned initially, it's only that easy ONCE. If the 
state is saved once, and restored once, all is well. If it is restored 
twice, or saved twice, all is wrong. I could not be more precise in my 
first message but now I know exactly _what_ is wrong : the Faces1 and 
Faces2 bitmaps are "gone" (but, strangely, not CardBack...). All the object 
instances and references (addresses, like Card.src) are correct, but there 
is nothing to display except card backs, and it's time to bail out or 
wxPython will crash...

The difference between Faces1/Faces2 and CardBack being that the latter is 
not referenced by Card instances, on a hunch I tried the following changes:

# new global
CardFaces = (Faces1, Faces2)

# in Card __init__:
    	    	if ... : self.src = 0  # store an index instead of bitmap ref.
    	    	else : self.src = 1

# in Stack showrow():
            srcDC.SelectObject(CardFaces[self.Cards[row].src])

This seems to work flawlessly, although I don't understand why. Does anyone 
? If anybody is interested in debugging, I have a trimmed-down module of 
about 300 lines that exhibits both behaviours -- just mail me, but BEWARE 
the .nospam suffix in my return address (sorry Jeremy :)

Thanks for your attention,
fp

-- 
YAFAP : http://www.multimania.com/fredp/



From fredrik at pythonware.com  Fri Sep 17 03:09:16 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 17 Sep 1999 07:09:16 GMT
Subject: Problem compiling Python on Linux
References: <7rrep5$n2e$1@newssrv.otenet.gr>
Message-ID: <008001bf00db$8f857830$f29b12c2@secret.pythonware.com>

Seby Carta  wrote:
> Hi , i have a problem with my Slackware linux and python 1.5.2, exactily
> with thread and python.
> When i run the command:
> 
> >>>import threading
> 
> Python tell me that he doesn't find thread module, i searched in my hdd..and
> module thread.py isn.t!!  why??,

python's probably built without thread support.
see the README file provided with the source
for more details.

(look for --with-thread)

> not the same problem with python for window...  he find above module and the
> thread.py is present on my hdd.

thread is a C module, so I find the latter a bit
unlikely.  maybe you meant "threading.py" (the
higher-level interface to threads)?






From tim_one at email.msn.com  Thu Sep 30 23:05:51 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 30 Sep 1999 23:05:51 -0400
Subject: When Good Regular Expressions Go Bad
In-Reply-To: <14323.30395.394477.753318@dolphin.mojam.com>
Message-ID: <000701bf0bb9$def95240$322d153f@tim>

[Douglas Alan]
> It seems to me that even when a regular expression fails to match a
> string, you might want to know just how far it was able to
> get before getting stuck.

[Tim sez probably easy to add, but seems of dubious utility]

[Skip Montanaro]
> Take a look at
>
>     http://www.musi-cal.com/fast-itineraries.shtml

OK, I have.

> I want to let people set up schedule submissions and give them some help
> when they format their schedules incorrectly.  Currently I use three
> regular rexpressions per selected format, the one that will match a
> correctly formatted line and two shorter ones.  When the real thing fails
> I use the shorter ones to try and give the user some idea of where they
> might have gone astray.  Of course, users aren't told that regular
> expressions underly the pattern matcher.

I don't see how this relates to Douglas's idea, unless perhaps the two
"shorter ones" can match only proper prefixes of strings matched by "the
real regexp", and also one of the other.

Of course people will enter highly structured data incorrectly, and of
course you want to help them get it straight.  Those are two reasons not to
use regexps at all <0.1 wink>.  0.9 seriously!  I entered:

    3/40 @ 7:30pm, Knitting Factory, New York, NY (with Doctor
    Nerve/Meridian Arts Ensemble

and it got rejected, without a clue as to why.  Error detection and recovery
is a Real Pain even with a Real Parser; a mechanical trick like "report
rightmost progress" isn't going to turn the infinitely feebler regexp
gimmick into a solution.

What you have here is a classical "frame" problem:  a number of information
"slots" that need to be filled in.  You picked a rigid format to ease your
own implementation, perhaps because you've been hoodwinked into believing
that "a regexp" *should* be "the solution".  Well, it isn't.  It may be
*part* of a solution, though:  for each slot that needs to be filled in, use
a tiny regexp to search over the input, removing the data that matches then
moving on to the next tiny regexp.  In this way, e.g., you would first
search for "a date", and in my example above would properly complain back to
me that you couldn't find a properly formed date.  It's a little more code
for you to write, but much better for your users, and much more flexible
over the long term.

> compellingness-is-in-the-eyes-of-the-programmer-ly y'rs,

usability-is-in-the-frustration-of-the-user-ly y'rs  - tim





From dubious at jonjh.freeserve.co.uk  Tue Sep  7 08:43:25 1999
From: dubious at jonjh.freeserve.co.uk (Rev. Jon Harford)
Date: Tue, 07 Sep 1999 12:43:25 +0000
Subject: Suitability for games?
Message-ID: <37D5086C.69204F4C@jonjh.freeserve.co.uk>

How suitable would Python be for games that move faster than Solitaire?




From m.faassen at vet.uu.nl  Thu Sep 16 10:06:13 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 16 Sep 1999 14:06:13 GMT
Subject: high school programming & python
References:   <37DF9080.21A8D285@calibre-dd.com>  <37DFE857.2784E45@callware.com>
Message-ID: <7rqtgl$brk$1@newshost.accu.uu.nl>

Ivan Van Laningham  wrote:
> Whippersnappers.  Arrogant young punks.

> All *I* had to work with was an undifferentiated sea of quarks!

You mean you didn't need to pull them out of their virtual states? 
They were already there?

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



From argollo at ircom.unilim.fr  Fri Sep  3 06:30:54 1999
From: argollo at ircom.unilim.fr (Daniel ARGOLLO)
Date: Fri, 3 Sep 1999 12:30:54 +0200
Subject: Freeze and SWIG
Message-ID: <028001bef5f7$670b17a0$b4bb32c1@clara.ircom.unilim.fr>

Hi there,

I've got a problem....

I'm trying to freese a python project that uses some external Cpp code
(in DLLs). The C code was wrapped with SWIG, it means that, for a 
my_class class, there is a my_class.py and incide an

import my_classc statement.

When i freeze my python code, it gives me an error message, because
freeze trys to import a my_classc module that does not exist.

Does anybody have a solution for this????


-------------------------------------------------------
Daniel ARGOLLO
IRCOM - Universit? de Limoges
argollo at ircom.unilim.fr

-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 4638 bytes
Desc: not available
URL: 

From tre17 at student.canterbury.ac.nz  Sun Sep 12 19:08:55 1999
From: tre17 at student.canterbury.ac.nz (tre17 at student.canterbury.ac.nz)
Date: 13 Sep 1999 11:08:55 +1200
Subject: Tkinter & refreshing windows
References: <199909121353.PAA08746@dionysus.fw.cuci.nl>
Message-ID: 

"Hans Nowak"  writes:

> Hello,
> 
> When working with Tkinter I noticed that on certain occasions, Tk 
> doesn't refresh the screen until it's done with a certain task. Hmm, 
> that doesn't sound very clear. :)  An example: I'm currently working 
> on a program that accepts user input, then filters a lot of data 
> based on this input, then displays the results. Since the filtering 
> took a few seconds, I thought it would be a good idea to display the 
> current search results in a listbox while filtering, i.e. for every 
> "record" that passed the test, a string is added to the listbox. You 
> would be able to see some results while the filtering was still going 
> on.
> 
> However, that doesn't seem to work at all. The program starts 
> executing the filtering, and doesn't display anything until 
> everything is done. In fact, trying to change the text of a label, 
> directly before the filtering starts, doesn't work either. 
> 
> I don't know if this is Windows-related (I'm using Win95 :( ), but 
> this is a well-known problem in Delphi, and there you have the 
> Refresh() method. Is there a similar method for Tkinter, to force an 
> update of the screen display?
> 
> TIA,
> 
> --Hans Nowak (zephyrfalcon at hvision.nl)
> Homepage: http://fly.to/zephyrfalcon
> Python Snippets: http://tor.dhs.org/~zephyrfalcon/snippets/
> The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/

The `update' command is what you're looking for.  For an example,
consider the following code:

=============================
#!/usr/bin/python

from Tkinter import *
import time

def loop():
    for x in range(5):
        print x
        l.insert('end', 'Foo %d' % x)
        root.update()
        time.sleep(1)

root = Tk()

l = Listbox(root)
l.pack(fill='both', expand=1)

b = Button(root, text='Start', command=loop)
b.pack(fill='x', expand=1)

root.mainloop()
=============================

The alternative if you want your interface to remain fully responsize
rather then just updating every now and then, is to use
multi-threading.  I seem to remember seeing something about Tkinter
being thread safe now, but I'vbe never used threads and Tkinter.

--
Tim Evans
tre17 at student.canterbury.ac.nz



From dwalker at iximd.com  Sat Sep 11 08:25:54 1999
From: dwalker at iximd.com (Damond Walker)
Date: Sat, 11 Sep 1999 12:25:54 GMT
Subject: Text UI
References: <65118AEEFF5AD3118E8300508B124877073BEE@webmail.altiris.com>  
Message-ID: 

William Tanksley wrote in message ...

[snip]

>
>That's cool.  It would be nice to have a standard Python text mode
>interface (wxPython would be the best imaginable ;-).
>


    Acouple weeks ago I wrapped up an initial version of a Text based UI for
Python and Curses.  It uses plain curses (not ncurses).  It's event based,
supports the notion of a container (either full screen, or dialog based),
controls (has labels, command buttons, text boxes, a progress bar), and is
100% Python.  Before a dialog is painted to the screen, it will gather the
text that is about to be covered by the dialog.  When the dialog goes away,
it automagically repaints the screen.
    It's not perfect, could stand to be looked over by a real Pythoneer, but
it works for my use right now.  I have webspace so I probably should put it
up for anyone to come and get it.  In the meantime, if someone wants to play
with it, send me email and I'll send it to ya right away.
    I was thinking of putting it out under the LGPL but that remains to be
seen.

                        Damond





From swgeorge at my-deja.com  Thu Sep 23 12:46:58 1999
From: swgeorge at my-deja.com (swgeorge at my-deja.com)
Date: Thu, 23 Sep 1999 16:46:58 GMT
Subject: Freezing script gives error message:
Message-ID: <7sdlht$e0f$1@nnrp1.deja.com>

Folks,
  Does anybody know why I am getting the following error message:

Error:  needed directory D:\Program Files\Python\lib\python1.5\config
not found.
Used '.\tools\freeze\freeze.py -h for help.

What am I missing that freeze needs from the source directories?

I have Python 1.5.2 downloaded from the python website and have
downloaded the sources too.  There does not seem to be a python1.5
directory in either.  If you can help me it would be greatly
apprecitated.

Thank you,
  Spencer George

swgeorge at us.ibm.com


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fdrake at acm.org  Wed Sep 29 10:37:58 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Wed, 29 Sep 1999 14:37:58 GMT
Subject: What GUI-lib to use
In-Reply-To: <37F1D710.47A82D23@bibsyst.no>
References: <37F1D710.47A82D23@bibsyst.no>
Message-ID: <14322.9286.610984.493427@weyr.cnri.reston.va.us>

Thomas Weholt writes:
 > I want to develop GUIs to simple scripts fast, writing as little code as
 > possible, but still making it understandable. Looked at GLADE. How does
 > this work with Python? 

  We've had good results with glade so far, but can't say that we've
built any large applications that way.

 > Could Tk get a more modern look perhaps and solve all our problems?

  I'm afraid you'd have to talk to the Tcl/Tk people about that one.
I would not expect large changes.


  -Fred

--
Fred L. Drake, Jr.	     
Corporation for National Research Initiatives



From fredrik at pythonware.com  Thu Sep  9 08:01:50 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 9 Sep 1999 12:01:50 GMT
Subject: timesync with python?
References: <37D66AF3.6A3E96F7@phoenix-edv.netzservice.de>
Message-ID: <002701befabb$1a884880$f29b12c2@secret.pythonware.com>

Holger Jannsen  wrote:
> But I also thought that someone has an idea about that before
> and did it likely better than me. So, do you know a weblink
> to a script which already nearly does what I need? Perhaps 
> directly with interrogation at port 37 NTP?

from the eff-bot archives:

import socket, time

HOST = "my.server"
PORT = 37

TIME1970 = 2208988800L # 1970-01-01 00:00:00

s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.connect(HOST, PORT)

b = s.recv(4)

t0 = ord(b[3]) + (ord(b[2])<<8) + (ord(b[1])<<16) + (ord(b[0])<<24L)
t1 = int(time.time()) + TIME1970

dt = int(t0 - t1)

print "delta is", dt, "seconds"




From greg.ewing at compaq.com  Thu Sep  2 18:36:31 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 03 Sep 1999 10:36:31 +1200
Subject: Parrot: Why the clunky syntax?
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk> <37C2B9CE.95AA5FEC@lri.fr> <37C9FF53.DD466783@compaq.com> <19990830150321.C6858@optiplex.palga.uucp> <7qg7ek$k5e$1@newshost.accu.uu.nl>
Message-ID: <37CEFBEF.58EAB14@compaq.com>

Martijn Faassen wrote:
> 
> So what Greg was asking, why not use a more Pythonesque syntax for parrot,
> where indentation has actual meaning?

That's one part of it. I was also wondering about
the '@'. What meaning does Parrot assign to it?

Greg



From gmcm at hypernet.com  Wed Sep  8 08:37:57 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 8 Sep 1999 07:37:57 -0500
Subject: Bug in time module?
In-Reply-To: <37D5E25F.1F7F0A74@pc.jaring.my>
Message-ID: <1275332084-98066501@hypernet.com>

lec wrote:

> Is this considered a bug since there is no Feb 31, 1999 or Feb 29,
> 1999?
> 
> >>> import time
> >>> time.strptime('31/02/1999', "%d/%m/%Y")

to which Berthold added:

> More very strange things happen for me on linux:

[snip everyday is Sunday]

The only editing Python's time module does is to ensure that you've 
passed the right number and kind of args. Everything else is the 
result of your platform's implementation of strptime. If you want 
something _good_, you should probably use MA Lemburg's mxDateTime.


- Gordon



From dozier at abs.net  Fri Sep 17 15:17:29 1999
From: dozier at abs.net (Bill Dozier)
Date: Fri, 17 Sep 1999 15:17:29 -0400
Subject: [Bug,Mac] Bug in Python 1.5.x ?time origin error
References: 
Message-ID: 

This was brought up in the MacPython SIG mailing list. If I am remembering
correctly, you are correct in your suspicion as to the cause.

In article ,
Howard at quercus.demon.co.uk (Howard Oakley) wrote:

>Gentlepersons,
>
>I have a bug report on Python for Mac versions 1.5.1 and 1.5.2b1. The date
>functions seem to have systematic errors of 1 day short of 4 years (is this
>because the Unix and Mac OS origins of time differ by this amount?).



From gmcm at hypernet.com  Thu Sep 16 13:38:04 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 16 Sep 1999 13:38:04 -0400
Subject: python call external program (WHAT THE?!)
In-Reply-To: <19990916100419.A7870@quark.emich.edu>
References: <37E0F448.695EF4D0@trey-industries.com>; from Ben Thomas on Thu, Sep 16, 1999 at 01:47:44PM +0000
Message-ID: <1274619158-36273305@hypernet.com>

Jeff wrote:

> if you are trying to write portable code, I would go with the
> suggestion below of using forward slashes in path names... if it
> doesn't work by default under windows, you can write yourself a
> simple function that *can* do it for you, and filter all your
> pathnames through it-- the '/' is more natural anyway. have a
> look at the 'os.path' module
> .

Forward slashes work fine on Windows when the run through 
the C library. When passed to os.system, however, they fail 
messily, because that goes nowhere near the C library, and 
the Windows API breaks up the string at each slash, thinking 
they are options.

- Gordon



From clgonsal at keeshah.penguinpowered.com  Thu Sep  2 20:51:35 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Fri, 03 Sep 1999 00:51:35 GMT
Subject: Returning none
References: <000101bef12c$19566520$ad2d153f@tim>  
Message-ID: 

On Mon, 30 Aug 1999 20:30:11 GMT, Michael P. Reilly  wrote:
> 
> I'm guessing that most everyone has forgotten what "None" _is_.  From
> Python Reference Manual: 3.2 "The standard type hierarchy":
> 
>    None
>           This type has a single value. There is a single object with
>           this value. This object is accessed through the built-in name
>           None. It is used to signify the absence of a value in many
>           situations, e.g., it is returned from functions that don't
>           explicitly return anything. Its truth value is false.
> 
> The key here folks is "the absense of a value".  None means "nothing at
> all" that the "procedure" proponents are all for.  It's doing exactly
> what you want, you may just be thinking of it incorrectly (we all are
> probably thinking of None incorrectly at times).
> 
> If you want to not include "return" in your functions, or to require
> that your code must "return None" that's fine.  I think that the
> argument is that the language is not going to improve by requiring
> procedures.

But None *isn't* nothing at all. If it was, [None] would be the same as
[], and calling str(None) would be the same as str(). They're different
though.

None is *something*. You can copy it around, put it into
lists/tuples/maps/attributes, and pass it to functions. That can cause
problems if you meant to pass the result of some computation, and your
function was incorrectly not returning the result, but instead the
default None.

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From arcege at shore.net  Fri Sep  3 20:41:15 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 04 Sep 1999 00:41:15 GMT
Subject: getpass.getpass() bugged?
References: <7qm5q0$lct$1@nnrp1.deja.com>
Message-ID: 

super_banane at yahoo.com wrote:
: The getpass.getpass() function should
: be requesting a password without echoing it...
: but unfortunately it does:

:>>> import getpass
:>>> getpass.getpass()
: Password: hello
: 'hello'
:>>>

: Python I use is v1.5.2
: Same behaviour on Solaris 2.6 and HPUX-10.20.
: Works fine with Python v1.5.1 on Linux, though.

I don't see a problem on my like system.

Python 1.5.2 (#6, Apr 20 1999, 10:35:35)  [GCC 2.7.2.2] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os, getpass
>>> getpass.getpass()
Password:
'hello'
>>> os.uname()
('SunOS', 'golem', '5.6', 'Generic_105182-05', 'i86pc')
>>>

This sounds more like a problem with the terminal emulation (for example
emacs' bad shell modes).

  -Arcege




From aahz at netcom.com  Wed Sep 22 13:02:23 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Sep 1999 17:02:23 GMT
Subject: Help with RegExps
References: <37E8B7B3.F4A9143@isogmbh.de>
Message-ID: <7sb22v$kmi@dfw-ixnews12.ix.netcom.com>

In article <37E8B7B3.F4A9143 at isogmbh.de>, Chris...  wrote:
>
>  I am looking for an regular expression that changes text, only if it
>is not anywhere between '<' and '>'.
>For example, add syntax highlighting to HTML-file:
>    'some text ref # this is a comment'
>should be changed to
>    'some text ref # this is a
>comment'

Try using htmllib instead; already packages all the regexes for you.
(Make sure to use 1.5.2; there are a couple of bugs in 1.5.1)
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From abkout at my-deja.com  Thu Sep 30 19:07:14 1999
From: abkout at my-deja.com (abkout at my-deja.com)
Date: Thu, 30 Sep 1999 23:07:14 GMT
Subject: checking for end of file?
Message-ID: <7t0qeo$q9f$1@nnrp1.deja.com>

Hello,

I've probably missed something obvious, but how
can I test for an end of file in a Python script?
(without attempting to read something from it)

Bob Koutsky


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



From NOSPAMcraig at osa.att.ne.jp  Wed Sep 29 06:32:51 1999
From: NOSPAMcraig at osa.att.ne.jp (Craig Hagerman)
Date: 29 Sep 1999 10:32:51 GMT
Subject: python as a first language?
Message-ID: 

Hi,
I want to get some feedback on how good Python is to learn for a beginner.
I am interested in learning programming (OOP especially). I studied (bit
didn't really learn) Fortran ages ago, studied formal logic (math) in
university and have been playing around with web technologies (Javascript
et al) and now want to try out bigger and better things. What are people's
thinking regarding Perl vs Python for a beginner, or how about Python
compared to Eiffel or Sather for this purpose? Any good suggestions for
what road I should head up first?
   Cheers,
   Craig



From darcy at vex.net  Thu Sep 23 09:14:06 1999
From: darcy at vex.net (D'Arcy J.M. Cain)
Date: 23 Sep 1999 13:14:06 GMT
Subject: SIGCHLD, fork, listen
References: <7sd6q4$gqq$1@salome.rfa.org>
Message-ID: <7sd92t$19fn$1@hub.org>

bill  wrote:
> I'm trying to set up a fairly "simple"
> server that listens for a socket connection,
> forks off a process and goes back to listening.
> This works fine for multiple clients, except
> I get zombies.  I try redefining SIGCHLD as
> SIG_IGN, but this doesn't work (as apparently
> is what's expected under POSIX).  I then
> try redefining it with my own function that
> includes a WAIT.  Fine, except "listen"
> doesn't get signalled when the child process
> dies, only when the next socket connection
> occurs.  (which could be hours later, not
> good).

How are you creating your server?  Are you using SocketServer?  This
supposedly does what you want.  Here is a simple echo server using it.

---------------------------- echod ----------------------------
#! /usr/bin/env python

HOST = ''
PORT = 51000

from SocketServer import *

class echohandler(ForkingMixIn, StreamRequestHandler):
    def handle(self):
        self.wfile.write(self.rfile.readline(16384))

if __name__=='__main__':
    from socket import *
    server = TCPServer((HOST, PORT), echohandler)
    server.serve_forever()
---------------------------------------------------------------

However, on a simple test, I find that this doesn't actually fork.
If someone can suggest why this doesn't fork, you should be able
to use this instead.  Note I tried reversing the order of the inherited
classes with no difference.

The simple test, btw, is to telnet to the port from two windows and type
a line into the second one then the first.  The second one doesn't echo
until I type in something into the first session.

-- 
D'Arcy J.M. Cain       |  Democracy is three wolves
http://www.druid.net/darcy/                |  and a sheep voting on         
+1 416 425 1212     (DoD#0082)    (eNTP)   |  what's for dinner.



From chapman at bioreason.com  Wed Sep  1 10:58:14 1999
From: chapman at bioreason.com (Mitch Chapman)
Date: Wed, 01 Sep 1999 08:58:14 -0600
Subject: threads on Solaris
References: <199908302338.AA04307@ch1d2833nwk> <199909010612.QAA06055@piglet.dstc.edu.au>
Message-ID: <37CD3F06.6010E28A@bioreason.com>

David Arnold wrote:
> 
> -->"Ted" == Ted Horst  writes:
> 
>   Ted> Does anyone know why I get 4 threads when I start the
>   Ted> interpreter under Solaris (as reported by /bin/ps -aL) ?
> 
>   Ted> This is python 1.5.2 built with a SUNPro compiler (sorry, I
>   Ted> don't have the config options).
> 
> i see the same behaviour, also SUNPro, but i *know* mine is configured
> with the thread module enabled, which i imagine has something to do
> with it?

Yes.  Multithreaded C apps also display this behavior.  I don't
develop on Solaris any more, so take this with a grain of salt, but:
There's a main thread, a thread for catching signals, and (I think) 
a reaper thread which finalizes exited threads.  And of course
any threads your application explicitly creates.

I can't remember the conditions under which you get the reaper thread.
I think it's either when you *do* have daemon threads (threads on 
which you don't intend to wait) or when you don't :)

-- 
Mitch Chapman
chapman at bioreason.com



From Gareth.McCaughan at pobox.com  Tue Sep  7 18:42:57 1999
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: 07 Sep 1999 23:42:57 +0100
Subject: Turing Compliant?
References: <37CD7CBC.54AD295F@megsinet.net>
	 <7r0q9d$sp6$1@nnrp1.deja.com>
	<14291.65246.743544.70969@buffalo.fnal.gov>
	<7r1ibf$eb1@dfw-ixnews14.ix.netcom.com>
	
	
	
Message-ID: <86puzul4we.fsf@g.local>

William Tanksley wrote:

> Not to mention the number of infinities -- are there more than two of them
> (countable and uncountable)?  Cohen proved that the question isn't
> answerable -- the first proof-by-example of Godel's Law.

I'm afraid this is very badly garbled.

1. It's been known ever since Cantor that there are infinitely many
   different infinities.

2. Cohen proved that ZF (or indeed ZFC) can't prove that there
   aren't any infinities between the size of Z and the size of R.

3. Goedel had already proved that ZF/ZFC also can't prove that
   there *are* any infinities between the size of Z and the size of R.

4. I've never heard the name "Goedel's Law" before. It's usually
   called "Goedel's Incompleteness Theorem" or, better, "Goedel's
   First Incompleteness Theorem".

5. I don't believe that the continuum hypothesis (the statement
   that there are no infinite cardinals bigger than Z and smaller
   than R) was the first concrete statement to be shown independent
   of the axioms of a serious foundational axiom system. In
   particular, Goedel's "Second Incompleteness Theorem" -- published
   in the same paper as the First, in 1931 -- implies that e.g.
   "ZFC is consistent" can't be proved in ZFC unless ZFC is
   inconsistent. (And, clearly, if ZFC is consistent then it
   also can't prove "ZFC is not consistent".)

("ZFC" is the name given to the usual axiomatisation of set theory.
There are others, with slightly different properties. By "Z" and "R"
I mean the integers and the reals, respectively. There's also an
axiomatic set theory called "Z", which I'm not talking about.)

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



From m.faassen at vet.uu.nl  Sun Sep  5 13:48:53 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 1999 17:48:53 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com> <37CCE10D.D07D023B@strs.co.uk> <7qjgq8$k1i$1@newshost.accu.uu.nl>  <7qlj3b$p7u$1@newshost.accu.uu.nl> 
Message-ID: <7quae5$q02$2@newshost.accu.uu.nl>

William Tanksley  wrote:
> On 2 Sep 1999 10:21:31 GMT, Martijn Faassen wrote:
>>William Tanksley  wrote:

>>> Java's not interpreted, though -- I've never seen ANY Java interpreter.
>>> Java is always compiled, usually to bytecode, and then the bytecode
>>> machine is emulated.

>>Um, I was under the impression that this is what a modern interpreter
>>actually does? Parse the source to bytecode and execute the bytecode.
>>After all, Python does the same, right?

> If you assume Python is an interpreter, then yes, that's a counterexample.
> My point is that Python isn't an interpreter -- or rather, isn't one by
> any meaningful definition of the word.

But I thought a meaningful definition is:

an interpreter is something that executes a program without a compilation to
native machine code; instead some intermediate format is used (or even the
source is executed 'directly').

Of course emulated processors muddle the picture but it's not that useless
a definition either, as you seem to claim.

Regards,

Martijn




From alex at somewhere.round.here  Thu Sep 30 19:07:12 1999
From: alex at somewhere.round.here (Alex)
Date: 30 Sep 1999 19:07:12 -0400
Subject: Large Numeric arrays
References: <37F385AC.7057E983@v1.wustl.edu>
Message-ID: 

> I'm passing fairly large arrays around (using Numeric Py)
> and I'm getting the error:
> 
> 	  MemoryError: can't allocate memory for array

I have run into this problem before when writing C.  I would have
something like 

unsigned long int big_array[20][20][20][20][20];

and my program would segfault before it even started executing the
main() function, even though I had plenty of memory.  The problem went
away when I made big_array an

(unsigned long int)*****

and assigned the array elements in slices at execution time.  My theory
was that the program wanted to lay the whole array down contiguously,
and there was no contiguous chunk of memory available for that, even
though the total memory available was more than enough.  

I know very little of the Numeric internals, but the manual says "Arrays
created from scratch are always contiguous; non-contiguous arrays are
the result of indexing and other structural operations."  Perhaps you
could make your arrays in slices, and then glom the slices together?

Alex.




From skip at mojam.com  Thu Sep  9 17:53:23 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 9 Sep 1999 16:53:23 -0500 (CDT)
Subject: Python example: possible speedup?
In-Reply-To: <7r98lk$kug@dfw-ixnews6.ix.netcom.com>
References: 
	<7r98lk$kug@dfw-ixnews6.ix.netcom.com>
Message-ID: <14296.11347.433292.287485@dolphin.mojam.com>

    Aahz> I'll test this if I have to, but do you know off-hand how this
    Aahz> compares to [list append]
    ....
    Aahz> What if you use range(100000) or range(1000000)?

Executing

    import cStringIO, time, string

    def doit(n, r):
	print "doing", n, "iterations - range function ==", r

	c = time.clock()
	for i in r(n):
	    pass
	print "loop overhead:", time.clock() - c

	c = time.clock()
	a=cStringIO.StringIO()
	for i in r(n):
	    a.write('h')
	a=a.getvalue()
	print "cStringIO:", time.clock() - c

	c = time.clock()
	a = []
	for i in r(n):
	    a.append('h')
	a = string.join(a,'')
	print "list append:", time.clock() - c

	print

    for i in [1000, 5000, 10000, 50000, 100000, 1000000]:
	doit(i, range)
	doit(i, xrange)

I got

    doing 1000 iterations - range function == 
    loop overhead: 0.0
    cStringIO: 0.01
    list append: 0.01

    doing 1000 iterations - range function == 
    loop overhead: 0.0
    cStringIO: 0.01
    list append: 0.01

    doing 5000 iterations - range function == 
    loop overhead: 0.01
    cStringIO: 0.03
    list append: 0.04

    doing 5000 iterations - range function == 
    loop overhead: 0.01
    cStringIO: 0.03
    list append: 0.04

    doing 10000 iterations - range function == 
    loop overhead: 0.02
    cStringIO: 0.06
    list append: 0.08

    doing 10000 iterations - range function == 
    loop overhead: 0.02
    cStringIO: 0.07
    list append: 0.08

    doing 50000 iterations - range function == 
    loop overhead: 0.11
    cStringIO: 0.32
    list append: 0.42

    doing 50000 iterations - range function == 
    loop overhead: 0.1
    cStringIO: 0.32
    list append: 0.39

    doing 100000 iterations - range function == 
    loop overhead: 0.22
    cStringIO: 0.65
    list append: 0.92

    doing 100000 iterations - range function == 
    loop overhead: 0.18
    cStringIO: 0.63
    list append: 0.88

    doing 1000000 iterations - range function == 
    loop overhead: 2.22
    cStringIO: 6.55
    list append: 10.27

    doing 1000000 iterations - range function == 
    loop overhead: 1.77
    cStringIO: 6.35
    list append: 10.04

which suggests there is a performance boost for cStringIO vs list append and
xrange vs range for large numbers, but that for loops with less than about
10,000 iterations, it probably doesn't matter what you use.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From bellings at badlands.NoDak.edu  Tue Sep  7 20:39:48 1999
From: bellings at badlands.NoDak.edu (Brent A Ellingson)
Date: 8 Sep 1999 00:39:48 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com>
Message-ID: <7r4b8k$brh$1@news.ndsu.nodak.edu>

Mr Kelvin Lee (kelvin at scientist.com) wrote:

: What are the most popular interpreted languages around?

I was under the impression that the most popular interpreted language,
by a /very/ large margin, is Intel 80386 machine code.  From the small
amount I've read, I was led to believe that all modern PCs basically
have specialized RISC chips which do nothing but run a small, thin, very 
fast x86 interpreter all day, every day.

That may be simplifying things a lot, but it might give you some idea
how silly the distinction between "interpreted" vs. "compiled" is.

-- 
Brent Ellingson (bellings at badlands.NoDak.edu)
"It is amazing how complete is the delusion that beauty is goodness." 
                                                 -- Leo Tolstoy



From m.faassen at vet.uu.nl  Tue Sep  7 08:15:58 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Sep 1999 12:15:58 GMT
Subject: Suitability for games?
References: <37D5086C.69204F4C@jonjh.freeserve.co.uk>
Message-ID: <7r2vlu$68m$1@newshost.accu.uu.nl>

Rev. Jon Harford  wrote:
> How suitable would Python be for games that move faster than Solitaire?

I've done some thinking about this, though I have little practical
experience.

It depends on what kind of game you're talking about, to start with. A whole
lot depends on that.

Python probably isn't fast enough to do a vast amount of 3d calculations,
but it's fast enough to *drive* these calculations. This is the general
principle -- Python can still perform quite well as a higher level
controlling system (which can include a lot -- GUI, loading/saving,
scripting of in-game objects and events, etc), even when it's not
fast enough to do low level graphics manipulation. You use a C or C++
library for that, but you can wrap those with Python.

For object scripting, there are some interesting developments. Python does
not have 'simulated threads' by default. These are non-OS threads where
the Python interpreter itself does the switching, first executing for
instance 100 bytecodes of 1 thread, then 100 of another, and so on. But,
Will Ware has released MicroThreads which does accomplish this. So look
into this if you're interested in driving a lot of objects in your game with
Python scripts. Microthreads seem to be better suited for this purpose
than the usual threads, which usually don't assume that large an amount of 
threads.

Microthreads can be found at:

http://world.std.com/~wware/uthread.html

There was talk about integrating MicroThreads into Christian Tismer's 
'stackless Python', and there is also talk about integrating *that* with
Python proper, but that's of course a lot of assumptions. :)

So I'd recommend Python for game scripting, and for the code that 
structures and drives lower level speedy code. For the lower level code
Python isn't very suited though.

Regards,

Martijn




From aaron_watters at my-deja.com  Sat Sep 25 16:44:47 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Sat, 25 Sep 1999 20:44:47 GMT
Subject: Gadfly and Python on the Mac
References: 
Message-ID: <7sjc7t$dqj$1@nnrp1.deja.com>

In article ,
  "JMartin"  wrote:
> Does Gadfly run on a Mac with Python? I get errors when I try to
install it.

Please report the problems you have to gadfly-rdbms at egroups.com
(the gadfly mailing list).  Also make sure you are using
gadfly 1.0.  I *thought* gadfly 1.0 ran on the mac, but maybe
the people who reported this made some minor changes to make it
work.  Anyway, if there is a problem, I'd like to fix it.  I know
it can be *made* to work without difficulty.
   -- Aaron Watters

===
I pity the foo'.


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



From aahz at netcom.com  Tue Sep 14 00:04:00 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Sep 1999 04:04:00 GMT
Subject: Query: Pass variable to DOS bat file
References: <37ddbf00.23343510@nntp.ix.netcom.com>
Message-ID: <7rkhfg$5dg@dfw-ixnews7.ix.netcom.com>

In article <37ddbf00.23343510 at nntp.ix.netcom.com>,
Milas Omdalen  wrote:
>
>Seeking to implement a rather long menu of zipped file choices. The
>basic dos bat file does not handle large menu choices very well. But I
>don't see a way to use the zip program in Python.
>
>Can I call dos programs from within Python script?

import os
os.system('unzip foo')

>Can I use a Python unzip program?

That might be better; take a look at the gzip module.  I think there's
also a tar module somewhere.  Unfortunately, this is not compatible with
pkzip.

>Can I pass the menu choice chosen in the Python script to the dos bat
>file?

Python:
os.system ( 'batch foo bar' )

Batch:
echo %1
echo %2

>I see that I can call the Python script file from my bat file. Maybe
>use Global variables? I thought of saving the variable in a file, an
>awfully roundabout implementation.

This is probably not a good idea.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From bob at horvath.com  Wed Sep 22 07:48:26 1999
From: bob at horvath.com (Bob Horvath)
Date: Wed, 22 Sep 1999 11:48:26 +0000
Subject: wxPython on Debian Linux
References: 
Message-ID: <37E8C20A.F4051E56@horvath.com>

Lars Marius Garshol wrote:

> Does anyone know how to find this?
>
> Searches on deja.com, debian.org and the wxPython site have so far
> been unsuccessful. (Yeah, I've tried what it says on the site, and
> also searched in dselect, all to no avail.)
>
> Any help would be most gratefully received.

I haven't tried it myself, but itsn't it at

http://www.debian.org/Packages/unstable/interpreters/python-wxwin.html




From arcege at shore.net  Wed Sep 15 17:05:20 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 15 Sep 1999 21:05:20 GMT
Subject: screen FLY by, where is the pause? Debug needed
References: <37DF9AAC.7541897A@trey-industries.com>
Message-ID: 

Ben Thomas  wrote:
: I am woirking on my first python script. I am trying to find out why I
: am not geting the results I expect. Is there a way to pause these
: 'system' calls so the screens don't fly by so fast? I can't see the dos
: error messages.  I checked my book and deja but I don't see the answer.

You can either call input() or raw_input(), or possibly even call
os.system('pause'); I would still to raw_input() tho.

  -Arcege




From aardvark_three at yahoo.com  Wed Sep  8 08:59:06 1999
From: aardvark_three at yahoo.com (jims)
Date: Wed, 8 Sep 1999 08:59:06 -0400
Subject: Choosing a MySQL interface
Message-ID: 

Hello all

I'm a bit of a Python newbie, and am building an app that uses a MySQL database (the latest version) as the back end. 

I wonder if anyone can tell me whuich is the currently "correct" way to connect to a MySQL database using Python. There appear to be several methods available (like here and here) but I can't tell which one is the right one to use at this time.

It would be nice if someone could post up-to-date instructions on using MySQL as a backend for Python programs. I can't seem to find any, and I've looked for a few days now.

thanks

jim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From davidopp at megsinet.net  Mon Sep 20 00:36:56 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Mon, 20 Sep 1999 00:36:56 -0400
Subject: PythonWin & Windows MFC
Message-ID: <37E5B9E7.93034953@megsinet.net>

Is there any good information anywhere about how to use PythonWin with
MFC to create an event driven program with an attractive GUI?

Thanks,

David O.




From davidopp at megsinet.net  Sun Sep 26 01:44:20 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Sun, 26 Sep 1999 01:44:20 -0400
Subject: Moving from Perl to Python
References:  
Message-ID: <37EDB2B4.2563A5DB@megsinet.net>


William Tanksley wrote:

>
>
> I think he just wanted the Perl Cookbook, which would indeed be great
> (except, of course, for the parts which are trivial in Python -- we don't
> need no hashes-and-list man page ;-).
>

Hashes?!  We don't need no stinking hashes!  

David O.




From bjorn at roguewave.com  Wed Sep  1 02:30:51 1999
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 1 Sep 1999 00:30:51 -0600 
Subject: lex / yacc for Python
Message-ID: <51C9F3C392D0D111A15600A0C99E64A9AF54F3@localhost>

There is of course also possible to do the lexing and parsing in flex/bison
;-)  The nodes in C++, SWIG to wrap them, and implement compiler passes in
python.  If you know lex/yacc and don't feel like learning error tracking
for a new system (in addition to extra syntax etc.)  You might want to give
it a try.  It actually works remarkably well (couple of rough spots, but
what'd you expect ;-)

If you'd like to take a quick look at a very premature project (the parts
you are interested in are working), let me know and I can send you a copy.

Quick question to other language implementors: I'm almost done writing the
core language evaluator (using a rather interesting version of the visitor
pattern that is much simpler, and can't be modeled in less dynamic
language).  The question is what do I do next?

Type inference has been an intest of mine for some years now, so I was
wondering if it should go first, probably followed by the pretty printer
(does anyone know of any current reasearch on minimizing the generated
interface descriptions? -- where I should go to find it?)

If going to a third language, the options are Python (I'm even starting to
understand the internals *sigh*),  i.e. something similar to P-code, wher I
either write the runtime envionment or use someone elses.  I could also use
C, C++,... if I wanted speed rather than quality at this point.

I'm planning to table optimizations for the time being, but I will still do
some fairly complete records of how the core language operates on the python
evaluator....

  => Bool  :-)

> -----Original Message-----
> From:	Mike Fletcher [SMTP:mcfletch at vrtelecom.com]
> Sent:	Tuesday, August 31, 1999 7:50 PM
> To:	python-list at python.org
> Subject:	RE: lex / yacc for Python
> 
> Normally I point new users not to mcf.pars (which I haven't touched in
> ages
> (though I still use it daily at work)) and point them to simpleparse,
> which
> generates tuples for use with the mxTextTools engine.
> 
> Neither system is a lexx/yacc style system.  They use very similar general
> pattern matching machinery which does not use lexxing machinery at all
> (i.e.
> no tokenisation pass).  Both systems are targeted at (moderately) fast
> parsing of a very limited class of grammars.
> 
> http://members.home.com/mcfletch/programming/simpleparse/simpleparse.html
> 
> As noted on your page, mcf.pars is still available as part of mcf.vrml,
> but
> the version of mcf.vrml on starship.python.net is actually an old version
> of
> that package.  I would have to check to see if the mcf.pars package
> changed
> between that version and later versions of mcf.vrml.
> 
> Cheers,
> Mike
> 
> -----Original Message-----
> From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
> Behalf Of Andrew M. Kuchling
> Sent: August 31, 1999 5:50 PM
> To: python-list at cwi.nl
> Subject: Re: lex / yacc for Python
> 
> 
> "Jacques Oosthuizen"  writes:
> > Any one know if there is such a beast
> 
> Quite a few, actually; many Python users seem to be language geeks.
> A list of parsing systems is at:
> 
> http://starship.python.net/crew/amk/python/string.html#parsing
> 
> I've never used any of them, and don't know which ones are still
> maintained, which are toys, and which are the most powerful.  Good luck!
> 
> --
> A.M. Kuchling			http://starship.python.net/crew/amk/
> This isn't life in the fast lane, it's life in the oncoming traffic.
>     -- Terry Pratchett, in alt.fan.pratchett
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From wware-nospam at world.std.com  Fri Sep 24 00:14:15 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Fri, 24 Sep 1999 04:14:15 GMT
Subject: Becoming root within a Python script
References: 
Message-ID: 

=?ISO-8859-1?Q?Fran=E7ois_Pinard?= (pinard at iro.umontreal.ca) wrote:
: I would like, within a executing Python started as non-root, ask for the root
: password, and then, continue executing part of the script while being root,
: becoming the previous again after some sub-job is done.

In another post you said you hoped to avoid using 'su'. Can't
help you there, but I did find a pretty simple version that
seems to work well, at least on my Linux box:

---- begin ----
import os, sys
if os.popen('whoami').readline()[:-1] == 'root':
	print 'Hey, look, I am logged in as root!'
else:
	print 'I am the normal user, about to do some root stuff.'
	os.system('su -c "python %s"' % sys.argv[0])
	print 'Normal user again, done doing root stuff.'
---- end ----

Dumb question for those wiser than myself:
The os.popen().readline()[:-1] construction is pretty ugly.
Is there a better way to do it? Aside from the typing and the
esthetics, I'm not altogether certain that the pipe will be
correctly refcounted away and dealloced. I find myself doing
similar things often enough, I'd prefer to hear a better way.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From MHammond at skippinet.com.au  Tue Sep  7 20:26:12 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 8 Sep 1999 10:26:12 +1000
Subject: CONTEST: PyWin crashes with pure Python code!
References: <816010E2456BD111A48700805FBBE2EEFDF5A4@ex-quebec-u1.baan.com>
Message-ID: <7r4aff$79u$1@m2.c2.telstra-mm.net.au>

Gaetan Corneau wrote in message
<816010E2456BD111A48700805FBBE2EEFDF5A4 at ex-quebec-u1.baan.com>...

Just to let people know what is going on here - we have been discussing this
in email.

It _appears_ we have a DLL version conflict or some other external influence
at work.  People who have MSVC 6 installed seem unable to recreate the
problem at all.  Adding fuel to this is that one person in particular could
create the crash, then installed VC6 and could no longer create it.

At this stage we are still waiting for some more information from people on
this - as there are now only a couple of people who can create this problem
it is taking a little longer than we hoped.

Mark.





From aiki108 at netscape.net  Thu Sep 30 12:59:31 1999
From: aiki108 at netscape.net (David Fisher)
Date: Thu, 30 Sep 1999 11:59:31 -0500
Subject: python UDPServer
References: <7srhvi$v6a$1@nnrp1.deja.com>
Message-ID: <7t04tg$1bu$1@newsreader.wustl.edu>

>That is, the script comes to a screeching halt
>when, in the course of performing the
>initialization for a UDPServer, it defaults to the
>initialization behaviour (for the most part) of a
>TCPServer, which does a listen().  This is a bad
>thing(tm) for a UDP socket.


You're right.  UDPServer should override server_activate().  And in version
1.5.2 is does.
Like this from Lib/SocketServer.py:

class UDPServer(TCPServer):

    """UDP server class."""

    socket_type = socket.SOCK_DGRAM

    max_packet_size = 8192

    def get_request(self):
        data, client_addr = self.socket.recvfrom(self.max_packet_size)
        return (data, self.socket), client_addr

    def server_activate(self):
        # No need to call listen() for UDP.
        pass

Your code runs fine on my win98/python1.5.2, except that the method is
called serve_forever() not server_forever().  You can edit the UDPServer
class in Lib/SocketServer.py to the above.  or upgrade to 1.5.2.  On Linux
the file is probably in /usr/lib/Python1.5.
dnf





From jam at quark.emich.edu  Thu Sep 16 16:11:15 1999
From: jam at quark.emich.edu (Jeff)
Date: Thu, 16 Sep 1999 16:11:15 -0400
Subject: Problem compiling Python on Linux
In-Reply-To: <7rrep5$n2e$1@newssrv.otenet.gr>; from Seby Carta on Thu, Sep 16, 1999 at 10:04:33PM +0300
References: <7rrep5$n2e$1@newssrv.otenet.gr>
Message-ID: <19990916161115.E8285@quark.emich.edu>

On Thu, Sep 16, 1999 at 10:04:33PM +0300, Seby Carta wrote:
> Hi , i have a problem with my Slackware linux and python 1.5.2, exactily
> with thread and python.
> When i run the command:
> 
> >>>import threading
> 
> Python tell me that he doesn't find thread module, i searched in my hdd..and
> module thread.py isn.t!!  why??,
> not the same problem with python for window...  he find above module and the
> thread.py is present on my hdd.
> 
> Please...help me .. if you can....
> 
> Thank you anywere.
> 
> Seby
> 

it sounds like you need to tell python to compile with the threading module
support enabled-- for whatever reason the configure script isn't able to
find it. are you sure your system has the right thread support built in?
what version of libc do you have installed? I don't have any slackware boxen
left (they have been changed to redhat based machines), so I can test these
things myself. does the README file included with the source explain what
you need to do? it may be as simple as modifying the Setup file to tell it
what you want. if you try all this and still have trouble, post to the list
again.

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From gmcm at hypernet.com  Sun Sep  5 22:12:04 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 5 Sep 1999 21:12:04 -0500
Subject: Using Python in ASP scripts
In-Reply-To: 
Message-ID: <1275542437-85413535@hypernet.com>

Nikolai Kirsebom wrote:

> I have two NT 4.0 PC's running IIS 4.0 where I'm using Python in ASP
> scripts.  On the one machine I have no problems running the scripts
> - however on the other machine I can't get the scripts running.  If
> I change the LANGUAGE token to something else (VBScript) it works -
> so obviously it must have something to do with the 'registration' of
> Python as the scripting language.  I've executed the pyscript.py and
> it reported all ok.

Sounds like this is related to a problem I had running any of the 
axscript stuff. I had at one time installed MS's script debugger. Not 
only did the axscript stuff not work, but I couldn't update any of 
the related MS components. While I didn't have to re-install NT, I 
did have to remove and then reinstall everything having to do with 
networking.

- Gordon



From tjreedy at udel.edu  Thu Sep 30 22:35:56 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 1 Oct 1999 02:35:56 GMT
Subject: newbee
References: 
Message-ID: <7t16mc$1il$1@news.udel.edu>

In article , 
ericholmes108 at hotmail.com says...
>
>hi
>
>i want to know how to get started with python the language.

Visit www.python.org and look for tutorial and stuff specific to your 
platform.




From cnetzer at mail.arc.nasa.gov  Thu Sep 30 20:29:16 1999
From: cnetzer at mail.arc.nasa.gov (Chad Netzer)
Date: Fri, 01 Oct 1999 00:29:16 +0000
Subject: lzw.py
References: <37F384C4.9486B017@darwin.in-berlin.de>
Message-ID: <37F4005C.385FFC8E@mail.arc.nasa.gov>

"Dinu C. Gherman" wrote:


Here is what the FSF says that Unisys is saying about their patent (ahem):

    No license or license fees are required for non-commercial, not-for-profit

    GIF-based applications or for non-commercial, not-for-profit GIF-freeware,

    so long as the LZW capability provided is only for GIF. However, a license
is
    required if freeware is incorporated into, or sold or distributed with a
commercial
    or for-profit product, introduced in 1995 [or later], or enhancements of
products
    that were introduced prior to 1995.

I got this at:

     http://www.gnu.org/philosophy/gif.html

There is more to it (IBM and Unisys both are involved, IBM licensing position
is
unclear).

The uncompressing part of the LZW algorithm is apparently NOT patented, BTW.

In any case, it is a bummer, but you can modify your algorithm and release it;
it
is mainly for learning anyway, right?  I wrote some nice coding algorithms in
school (arithmetic, huffman, lz77, each with various adaptability
enhancements)
and learned a lot.  The LZW algorithm is really no longer top-notch; computing

power is such that you can easily write algorithms that make assumptions of
bigger
resources that are probably not covered by the LZW patent (which dealt w/
tree-pruning and pointer rescaling, etc.)

I really should go back and play with these ideas w/ Python.  Dealing w/ C
made many
implementation details VERY cumbersome (especially given that I was exploring
adaptive models), but the Python abstractions may make things very elegant (if
a bit
slower :)

Chad Netzer
chad at vision.arc.nasa.gov





From guido at cnri.reston.va.us  Wed Sep 15 23:42:14 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 15 Sep 1999 23:42:14 -0400
Subject: TreeCtrl Object (win32ui)
References: <7rm9bb$1rh$1@nnrp1.deja.com> <7ro8ql$eao$1@nnrp1.deja.com>
Message-ID: <5llna74j55.fsf@eric.cnri.reston.va.us>

jayseye at my-deja.com writes:

> Does Python offer a cross-platform Tree control,
> or does it only provide access to the Win32 control
> when running under Windows?

Yes.  IDLE 0.5 has a TreeWidget module that works with Tkinter.  It is
not in any distribution yet, you need to get it via the CVS archives,
see http://www.python.org/download/cvs.html.  IDLE is in Tools/idle/.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From skip at mojam.com  Wed Sep 22 12:44:20 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 22 Sep 1999 11:44:20 -0500 (CDT)
Subject: Linux Dependencies! Grrr...What now?
In-Reply-To: <19990922123456.B15540@quark.emich.edu>
References: <37E8066F.1C13BF33@alliedsignal.com>
	<19990922123456.B15540@quark.emich.edu>
Message-ID: <14313.1892.521554.125304@dolphin.mojam.com>

    >> (I've been warned that as the result of my upgrade, I have now
    >> "broken" netcfg, but I'll deal with that on another day.)

The fix is pretty trivial.  There are two functions in
/usr/lib/rhs/netcfg/netcfg.py that have broken defs.  I appended the changes
I made.  Haven't had a problem with netcfg since then.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...


*** netcfg.py~  Mon Aug 17 12:31:42 1998
--- netcfg.py   Fri Jul 16 12:33:32 1999
***************
*** 138,142 ****
        return ''
  
! def LinespeedMenu(master=None, variable, width='20'):
        speedlist = []
        for speed in ['1200', '2400', '4800', '9600', '19200', '38400',
--- 138,142 ----
        return ''
  
! def LinespeedMenu(master=None, variable=None, width='20'):
        speedlist = []
        for speed in ['1200', '2400', '4800', '9600', '19200', '38400',
***************
*** 146,150 ****
        return LabelledMenu(master, 'Line speed:', variable, width, speedlist)
  
! def ModemportMenu(master=None, variable, width='20'):
        portlist = []
        try:
--- 146,150 ----
        return LabelledMenu(master, 'Line speed:', variable, width, speedlist)
  
! def ModemportMenu(master=None, variable=None, width='20'):
        portlist = []
        try:



From ferni at loxinfo.co.th  Sun Sep 26 23:42:22 1999
From: ferni at loxinfo.co.th (Jon Fernquest)
Date: Mon, 27 Sep 1999 10:42:22 +0700
Subject: Matching templates against a tree - any idea?
References: 
Message-ID: <005301bf089a$5ba21d40$1a2292cb@akha>

Fernando Pereira wrote:

>[Matching templates against a tree]...has been the subject 
>of much research. For a recent textbook, check out 
>Franz Baader and Tobia Nipkow's "Term Rewriting and All That", 

Thanks for making the posting on term rewriting systems.

1. Do you think it's possible to make a rewriting system extension 
to an existing language like Python or Perl?

Everything seems to be in ML and ML extensions such as
Isabelle right now. It's so easy to cook up (or integrate 
components into) whole systems in Python 
(or Perl) and since so many people use them in their
day to day work modules have a greater chance of being used.

2. Are these term rewriting systems a broader category 
that would allow someone to implement for example different
linguistic tools?

What about non-roman script text processing?
The composition of the Burmese *orthographic* syllable 
from underlying glyphs cannot be described by a context free
grammar. There are even cross syllable 
dependencies/non-linearities, syllable n sometimes has 
glyphs in syllable n-1...and I assume this holds for all other
Indic scripts.

There are so many problems associated with these scripts
ranging from sorting to word-wrap to what will be chosen 
as an item in the dictionary, it would be nice to have a 
generalized framework in which to think about them.
Do you think rewriting systems (or discrimination nets/tries)
might be a viable approach?

There are so many different logical sets of tokens,
display tokens, ordering/sorting/spelling tokens (a,b,c,d,e..),
phonetic tokens,
transliteration into roman script, 
transliteration of English words into Burmese...
that have to be translated between in any application 
that features Burmese text (also applicable to Indian scripts in general)

Meanwhile, people type fonts into documents. 
For one character/glyph there are about 8 
different keys on the keyboard chosen based on context.
Everyone thinks that unicode is going to solve all their 
problems. Unicode is only an encoding.

I wrote sets of cascaded finite state transducers
to do these translations but maybe term rewriting 
rules are the solution?

Thanks again for the interesting posting.

Jon Fernquest
bayinnaung at hotmail.com
(on the road right now)

 






From " HPeter.Friedrichs" at alliedsignal.com  Thu Sep 16 14:02:16 1999
From: " HPeter.Friedrichs" at alliedsignal.com (H. P. Friedrichs)
Date: Thu, 16 Sep 1999 11:02:16 -0700
Subject: Newbie:Wants to access MySQL with Python
Message-ID: <37E130A8.C5FC1B7F@alliedsignal.com>

Hi:
I have a Redhat linux box running mysql. I'd like to access a database
using python (for cgi purposes). Supposedly, MySQLdb should help me do
this.

I downloaded MySQLdb, and attempted to build it per the instructions.
The script build.py errors out when it can't find:
/usr/lib/python1.4/config/Makefile.pre.in

What is Makefile.pre.in? "Makefile" exists in the above directory, but
Makefile.pre.in does not. What do I do next?

Is there a different python module available to access MySQL?

Thanks in advance
Pete

--

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------





From tim_one at email.msn.com  Sat Sep  4 02:40:06 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 4 Sep 1999 02:40:06 -0400
Subject: Turing Compliant?
In-Reply-To: <37D09D5B.D31636BF@atl.mediaone.net>
Message-ID: <000201bef6a0$5323adc0$842d153f@tim>

[Kristopher Johnson]
> But can Python efficently solve the Traveling Salesman Problem,
> or other NP-complete problems?

If it were thought possible to solve such things efficiently, they wouldn't
be NP-complete <0.5 wink>.

TSP in particular has had person-centuries of intellectual effort tossed at
it, and the best algorithms known will certainly run faster in almost
anything other than Python.

OTOH,

1) If you're looking to write a nice framework in which to enter, visualize,
and report TSP problems & results, Python can do everything including
dynamically generate optimized native machine code tailored to each problem
instance (if you're looking for peak efficiency, anything less is wimpy
compromise ).

2) For problems that haven't yet been studied to distraction, a language
like Python (other very high level languages fit here too, but Python is
particularly well suited to this) lets you try out dozens of approaches with
little effort.  Over the years I've found hard-problem heuristics that
allowed my Python program to run 1000x faster than a colleague's
micro-optimized C, because I can code up & test 20 approaches in the time it
takes them to remember how to declare a vector of functions returning a
vector of pointers to pointers to structs <0.9 wink>.

> That's the true measure of a language

It's one measure of many, and Python does fine here.

> (other than being able to compile its own compiler, of course).

Which is another measure of many.  It's too easy for Python to bother to
compile itself, so it usually compiles Perl instead .

ok-ok-perl-can't-even-be-parsed-ly y'rs  - tim





From lutz at rmi.net  Tue Sep  7 15:01:09 1999
From: lutz at rmi.net (Mark Lutz)
Date: Tue, 7 Sep 1999 19:01:09 GMT
Subject: O'Reilly conference writeup
Message-ID: <199909071901.NAA27541@shell.rmi.net>

Just noticed: the latest installment of Frank
Willison's column includes a few words about 
the open source movement, the recent O'Reilly
Python conference, zope, and more.  See: 

    http://www.oreilly.com/frank/

And while I'm at it, a belated thanks to the 
conference organizers, especially David Ascher,
for putting together a great event.

--Mark Lutz



From jam at quark.emich.edu  Thu Sep 16 16:08:26 1999
From: jam at quark.emich.edu (Jeff)
Date: Thu, 16 Sep 1999 16:08:26 -0400
Subject: python call external program (WHAT THE?!)
In-Reply-To: <65118AEEFF5AD3118E8300508B124877073BF7@webmail.altiris.com>; from Mike Steed on Thu, Sep 16, 1999 at 12:04:16PM -0600
References: <65118AEEFF5AD3118E8300508B124877073BF7@webmail.altiris.com>
Message-ID: <19990916160826.D8285@quark.emich.edu>

On Thu, Sep 16, 1999 at 12:04:16PM -0600, Mike Steed wrote:
> 
> Just to be nit-picky, forward slashes work fine in argv[0] when passed to
> os.system (at least on WinNT, don't know about Win9x).  So
> os.system('c:/temp/junk.bat') works, but os.system('del c:/temp/*.*') and
> the like will probably fail, depending on the executable.
> 
> Of course, it is safest and most portable to use os.path.
> 
> M.
> 

ok, so the problem is actually in the effort to execute external
(command.com) batch files, instead of implementing them in python directly.
to the original poster: what is in the batch files? can you just make simple
os.system() calls to get the job done? this will eliminate the path problem
entirely.

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From davidopp at megsinet.net  Tue Sep 14 05:03:59 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Tue, 14 Sep 1999 05:03:59 -0400
Subject: Different Shaped Windows?
Message-ID: <37DE0F7E.714FD6F7@megsinet.net>

Dear Group,

>From a posting in another  newsgroup, I am told that you can:

=====================================
"create windows that have different shapes than the
ordinary "square" ones. You can do than thanks to the "regions" : What
does
that mean?
In Win32 you can define regions and than assign them to already created
windows. For example:

SetWindowRgn(handle, CreateEllipticRgn(0,0,100,200), true)

will create an elliptic window. MSDN has more info about regions. You
can apply all sorts of operations to your regions like intersection,
addition, etc. to obtain different shapes.
=================================================

Question is:  Can this same task (creating a non rectangular window) be
accomplished with any of the GUI toolkits available to Python such as
Tkinter or wxPython?

David O.






From bwinton at tor.dhs.org  Wed Sep  1 19:16:05 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Wed, 01 Sep 1999 23:16:05 GMT
Subject: Adding manual access to IDLE
References: <37CB245C.3AF@loop.com>
Message-ID: 

On Mon, 30 Aug 1999 17:39:56 -0700, JW wrote:
>I am a very new user of Python, and although IDLE is a godsend(Ooh,
>colors!), it was curiously missing any access to the Python
>Documentation.  I added a simple command to open the docs, but it is
>pretty ugly, and I wonder if anyone else has noticed this problem or has
>any sugestions about dealing with it?
>
>These are my alterations in the IDLE source:
>     	os.system("C:\INTERNET\NETSCAPE\NAVIGA~1\PROGRAM\NETSCAPE.EXE
>C:\python\doc\index.html") #JW

Looks fine to me, up until this point...

I don't have Netscape installed, and if I did, it wouldn't be on C:\...
(It would probably be at /opt/netscape/bin/netscape)

I like the idea, though, and am now wishing for a decent HTML Text
Widget in Tkinter.  (I know that Gnome has one, but I'ld like to
maintain the cross-platformness of IDLE.)  Oh well, I've got some
vacation coming up, perhaps I'll build one.  ;)

Oh, yeah, and the docs are on a different place on my machine as well.

Later,
Blake.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From hniksic at srce.hr  Wed Sep  8 13:31:11 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 08 Sep 1999 19:31:11 +0200
Subject: Python example: possible speedup?
In-Reply-To: Skip Montanaro's message of "Wed,  8 Sep 1999 11:53:20 -0500 (CDT)"
References: <87puztv2gt.fsf@pc-hrvoje.srce.hr> <14294.37495.703177.461853@dolphin.mojam.com>
Message-ID: <87906h5mzk.fsf@pc-hrvoje.srce.hr>

Skip Montanaro  writes:
[...]

Thanks for your comments.  The package files aren't *humongous*, but
they aren't exactly small either.  One file is 700K, and the other is
3M large.  It would be really nice if it were possible to process them
in a stream.

But I'll try it, for the sake of argument, and see if the speedup is
worth it.

> Also, instead of calling next_header on a line-by-line basis, how
> about calling it once per package and returning a dict containing
> the header names as keys?

I'll try this too, and get back to you.  That change will actually
make the code more correct because it will no longer assume that
`Package' header comes before `Status' and `Installed-Size' headers.



From gherman at darwin.in-berlin.de  Wed Sep 15 07:23:14 1999
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Wed, 15 Sep 1999 13:23:14 +0200
Subject: Printing strings with null-bytes in Pythonwin?
Message-ID: <37DF81A2.9B3A4793@darwin.in-berlin.de>

Is is possible that Pythonwin has a problem with strings 
containing null-bytes? Or is the behaviour below to be 
expected as one cannot display null bytes in any meaning-
ful way, so to say? 

If I run the following code from a script:

str = ['\000'] * 4
# str = ['a'] * 4
print `str`
from string import join
print join(str, '')

I get this traceback, pointing to a file winout.py which 
does not seem to mention any problems with null-bytes:

"""
Traceback (innermost last):
  File "C:\Program
Files\Python\Pythonwin\pywin\framework\scriptutils.py", line 237, in
RunScript
    exec codeObject in __main__.__dict__
  File "C:\Usr\gherman\Misc\nullbytes.py", line 5, in ?
    print join(str, '')
  File "C:\Program Files\Python\Pythonwin\pywin\framework\winout.py",
line 207, in write
    return self.template.write(msg)
  File "C:\Program Files\Python\Pythonwin\pywin\framework\winout.py",
line 396, in write
    self.HandleOutput(message)
  File "C:\Program Files\Python\Pythonwin\pywin\framework\winout.py",
line 381, in HandleOutput
    return self.QueueFlush()
  File "C:\Program Files\Python\Pythonwin\pywin\framework\winout.py",
line 364, in QueueFlush
    self.currentView.ReplaceSel(self.outputQueue.get_nowait())
TypeError: ReplaceSel, argument 1: expected string without null bytes,
string found
"""

If you write the same in an interactive Pythonwin window 
you'll get the following, indicating that the problem only 
occurs when you actually try to print the string containing 
null-bytes:

>>> str = ['\000'] * 4
>>> print `str`
['\000', '\000', '\000', '\000']
>>> from string import join
>>> join(str, '')
'\000\000\000\000'
>>> print join(str, '')


Traceback (innermost last):
  File "C:\Program Files\Python\Pythonwin\pywin\framework\interact.py",
line 372, in keyhandler_enter
    self.DrawPrompt()
  File "C:\Program Files\Python\Pythonwin\pywin\framework\interact.py",
line 120, in DrawPrompt
    self.flush()
  File "C:\Program Files\Python\Pythonwin\pywin\framework\winout.py",
line 212, in flush
    return self.template.flush()
  File "C:\Program Files\Python\Pythonwin\pywin\framework\winout.py",
line 399, in flush
    self.QueueFlush()
  File "C:\Program Files\Python\Pythonwin\pywin\framework\winout.py",
line 364, in QueueFlush
    self.currentView.ReplaceSel(self.outputQueue.get_nowait())
TypeError: ReplaceSel, argument 1: expected string without null bytes,
string found
>>>

If the reasoning at the beginning is the official one (null-
bytes cannot be displayed anyway), it's interesting to note 
the difference to a Python interpreter running on  Linux, 
where you get as boring a result as can be expected, but at 
least it does not fail:

>>> str = ['\000'] * 4
>>> print `str`
['\000', '\000', '\000', '\000']
>>> from string import join
>>> join(str, '')
'\000\000\000\000'
>>> print join(str, '')

>>>

Just curious,

Dinu

-- 
Dinu C. Gherman

................................................................
"An average of more than 15 % of adults in 12 industrialized 
countries are functionally illiterate; in Ireland, the United 
Kingdom and the United States, the rates are over 20 %."

  (The State of the World's Children 1999,
   UNICEF, http://www.unicef.org/sowc99)



From nascheme at ucalgary.ca  Wed Sep 15 19:21:02 1999
From: nascheme at ucalgary.ca (Neil Schemenauer)
Date: Wed, 15 Sep 1999 23:21:02 GMT
Subject: Lazy file.readlines()?
References:  <87hfl546rx.fsf@pc-hrvoje.srce.hr> <014801befa94$26f225e0$f29b12c2@secret.pythonware.com>
Message-ID: 

Fredrik Lundh  wrote:
>Hrvoje Niksic  wrote:
>> > - reading a file one line at a time (self.__fp.readline())
>> 
>> I don't see an alternative to this, except to read the whole file at
>> once, which I am trying to avoid, as the files are large.
>
>note that:
>
>    lines = fp.readlines(16384)
>    if not lines:
>        break
>    for line in lines:
>        ...
>
>is usually much faster than
>
>    line = fp.readline()
>    if not line:
>        break
>    ...

What would be cool is if readlines() returned a lazy sequence
object (ie. only read as much is needed using a certain block
size).  This should give the advantages of readlines() without
the concern about sucking up a huge file all at once.

I implemented this idea (probably badly) in pure Python
and got about a 2x speedup verses readline().  It is a small
module so I will post it here.


    being-lazy-has-its-advantages'ly Neil


import string

class BlockFile:
    def __init__(self, file, blocksize=1024*40, sep='\n'):
        self.file = file
        self.blocksize = blocksize
        self.sep = sep
        self.line = -1
        self.lines = []
        self.end = ''

    def __getitem__(self, i):
        try:
            self.line = self.line+1
            return self.lines[self.line]
        except IndexError:
            self.line = 0
            self._get_block()
            return self.lines[0]

    def _get_block(self):
        data = self.file.read(self.blocksize)
        if len(data) == 0:
            raise IndexError
        self.lines = string.split(data, self.sep)
        self.lines[0] = self.lines[0] + self.end
        if len(data) == self.blocksize:
            self.end = self.lines[-1]
            del self.lines[-1] # this _should_ be fast
        else:
            self.end = ''

def test_block(input):
    for l in BlockFile(open(input)):
        pass

def test_normal(input):
    f = open(input)
    while 1:
        l = f.readline()
        if not l:
            break
    

def measure(function, *args):
    import time
    t1 = time.time()
    apply(function, args)
    t2 = time.time()
    apply(function, args)
    return t2-t1

if __name__ == '__main__':
    import sys
    print 'block time', measure(test_block, sys.argv[1])
    print 'normal time', measure(test_normal, sys.argv[1])



From kuncej at mail.conservation.state.mo.us  Mon Sep 13 15:04:28 1999
From: kuncej at mail.conservation.state.mo.us (Jeffrey Kunce)
Date: Mon, 13 Sep 1999 14:04:28 -0500
Subject: Examples using urllib, htmllib
Message-ID: 

>Can someone point me to some example Python
>scripts that I can download that show accessing
>the web using the urllib, and htmllib.  I'm

a simple example: urlpost.py at http://starship.python.net/~jjkunce/ 

>One problem I may have is that the ... operations 
>at the web site i...appear to be Java applets 

uh oh. That may be a problem. I've been using python's
COM capabilities to access https sites, and to automate
browsing through some pretty tricky javascript pages.
I don't know if it would help with java applets, though.
See msiecom.py at the same site.

  --Jeff




From aa8vb at yahoo.com  Wed Sep 15 11:55:42 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 15 Sep 1999 11:55:42 -0400
Subject: Python, FastCGI and FreeBSD
In-Reply-To: <7ro4d9$apg$1@nnrp1.deja.com>; from henk_p@my-deja.com on Wed, Sep 15, 1999 at 12:45:32PM +0000
References: <7ro4d9$apg$1@nnrp1.deja.com>
Message-ID: <19990915115542.A1922501@vislab.epa.gov>

henk_p at my-deja.com:
 |I am experiencing this strange problem with a Python script running as a
 |fastCGI application under FreeBSD 3.2.  For some reason when the fastcgi
 |process manager in Apache starts the script, it (or the Python
 |interpreter) immediately crashes. It does not matter what the script
 |contains, it always crashes upon initialization.  ('Uncaught signal 11,
 |Segmentation Fault' as it says in the apache error log)

Hard to say.  I've not seen this with the Python 1.5.2 package running on
FreeBSD 3.2, though I've not run Python underneath a web server -- only in
the background and interactive.

Try adding "-v" to the Python invocation for the script.  For example, if
the first line is of the form '#!.....python', change it to
'#!/usr/local/bin/python -v'.  This may give you an idea what it's in the
process of doing when it SEGVs (loading a corrupted shared library, etc.).

Randall




From clgonsal at keeshah.penguinpowered.com  Thu Sep  2 19:36:27 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Thu, 02 Sep 1999 23:36:27 GMT
Subject: Returning none
References: <000001bef265$fe64f340$192d153f@tim> <37CC79C5.1BFDC710@compaq.com> <7qidi6$nbb$1@towncrier.cc.monash.edu.au>
Message-ID: 

On 1 Sep 1999 05:28:38 GMT, Jonathan Giddy  wrote:
> Greg Ewing  writes:
> >The compiler knows whether the result of any given
> >call gets used or not. Suppose it generates CALL_FUNCTION
> >when the result is used, and a new opcode CALL_PROCEDURE
> >when it doesn't.
> >
> >CALL_PROCEDURE checks that the result is NotARealObject 
> >(I'll call it Void from now on) and raises an exception 
> >otherwise. CALL_FUNCTION does the opposite.
> 
> This requires that we always use the result of a call to a real function,
> even if we just assign it to a dummy variable.  And that means someone
> will one day spend time trying to find where the dummy variable is used,
> instead of instantly being aware that the function is only called for its
> side effects.
> 
> For example, FileObject.close needs to be a function to return the exit
> status of a popen process.  Hence, every file close will need to be
> assigned (or used in some way).

That's why I suggested the addition of a new type of statement, similar
to 'EVAL' statements in Modula-3. In Modula 3, a function call is not a
legal statement (but a procedure call is). If you reall want to call a
function though without getting the result you would say:

    EVAL MyFunctionWithAUselessResult();

Since 'eval' is already used for something in Python, I'd suggested that
the Python equivalent could be called 'discard'.

    discard inFile.close()

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From skip at mojam.com  Fri Sep  3 14:01:38 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri,  3 Sep 1999 13:01:38 -0500 (CDT)
Subject: Returning none
In-Reply-To: 
References: <1275758005-72447341@hypernet.com>
	
Message-ID: <14288.2906.802899.222390@dolphin.mojam.com>

    G> So put me firmly in the camp that favors a stronger PyLint (I just
    G> recently got bit by a rarely used bare return in some released code).

    C> This is akin to saying array-bounds checking should be done by PyLint
    C> rather than the runtime. The problem is, a lot of these checks simply
    C> can't be done statically.

I'll assert here without proof that the primary mistake programmers are
likely to make in this regard is to include both "return " and
"return" (or the terminating byte code) in the same function.  This sort of
error a program like PyLint could indeed check.  There's be no need to add
the overhead of checking for incorrect function/procedure return semantics
at run-time.

    C> In fact, while I was originally proposing some compile-time checks as
    C> well, I've basically given up on those. They were mostly incomplete
    C> (Python is too dynamic to be able to do much static checking in this
    C> situation) and Tim found a real problem with them.

Perhaps I missed something (this thread is getting rather long and drawn
out).  What sort of situation would PyLint miss that my proposed static
check wouldn't catch?

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From tim_one at email.msn.com  Mon Sep 13 22:37:42 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 13 Sep 1999 22:37:42 -0400
Subject: Queue not releasing memory
In-Reply-To: 
Message-ID: <000601befe5a$1ebbf9e0$e7a0143f@tim>

[Nathan Clegg]
> ...
> I do need the thread-safety of the Queue class.  The large size
> requirement is only an extreme situation, which is why it is important
> that it backs down afterward.

Still unclear why "backing down" is important.  The VM high-water mark is
really of no practical interest on most systems.  Python isn't using the
memory any more, so it's not costing anything except a hole in the address
space (which your libc malloc may or may not release to the OS someday, or
may or may not reuse for its own purposes again -- why do you care?).

> ...
> Is there no way to remove an element from a list without making a copy?

Python lists are contiguous vectors of C pointers.  If you remove an element
from "the front" (as Queue does), all the remaining elements have to "slide
left" one position.  If it's not the queueness but the thread safety that
attracts you, subclass Queue and remove elements from "the rear" (i.e., use
x.pop() instead of del x[0] -- then it acts like a stack instead of a
queue).  Then copying will rarely be needed.  If you want both a queue and
no-copying, implement a two-pointer circular buffer.

> append() doesn't need a copy.

It usually doesn't; it certainly may.

> I was under the impression that lists were internally represented by a
> more dynamic form.

I believe you .

purge-your-soul-of-falsehoods-and-fear-of-artificial-integers-ly y'rs  - tim





From jam at quark.emich.edu  Thu Sep 16 18:23:54 1999
From: jam at quark.emich.edu (Jeff)
Date: Thu, 16 Sep 1999 18:23:54 -0400
Subject: Redhat Python RPM
In-Reply-To: <37E149D4.B7168314@callware.com>; from Ivan Van Laningham on Thu, Sep 16, 1999 at 07:49:40PM +0000
References: <37E149D4.B7168314@callware.com>
Message-ID: <19990916182354.B8531@quark.emich.edu>

On Thu, Sep 16, 1999 at 07:49:40PM +0000, Ivan Van Laningham wrote:
> Hi All--
> Who has the RPM for Redhat 6.0?  The one that doesn't break RHs stuff?
> 
> -ly y'rs,
> Ivan

I use the versions from andrich, and I haven't had any troubles with the
redhat tools.. anything in particular I should test to see? what exactly is
getting broken?

regards,
J

-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From fn at panix.com  Thu Sep 16 12:08:23 1999
From: fn at panix.com (Felicia Neff)
Date: Thu, 16 Sep 1999 12:08:23 -0400
Subject: cgi post question
Message-ID: 

Yes, this is yet another newbie question:

I am trying to write a CGI script that takes input from a POST, modifies
some of the input and then POSTs that info to another CGI script.  I know
that perl has a function that will do this.  Is there such a function in
python (and if it is do_POST() from the CGIHTTPServer module, is there any
more documentation besides that at
http://www.python.org/doc/current/lib/module-CGIHTTPServer.html?)?

Thanks -- Felicia


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
fn at panix.com                              Felicia Neff




From davidopp at megsinet.net  Thu Sep 16 03:01:06 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 16 Sep 1999 03:01:06 -0400
Subject: [Fwd: Article Submissions Wanted at Dr. Dobbs]
Message-ID: <37E095A6.FF9B2267@megsinet.net>

Couldn't resist the urge to post this.

David O.
-------------- next part --------------
An embedded message was scrubbed...
From: "Antony Chesworth" 
Subject: Article Submissions Wanted at Dr. Dobbs
Date: Thu, 16 Sep 1999 00:25:38 +0100
Size: 1310
URL: 

From aboukirev at iname.com  Sun Sep  5 19:57:36 1999
From: aboukirev at iname.com (Alexei Boukirev)
Date: Sun, 5 Sep 1999 18:57:36 -0500
Subject: Python speed and `pcre'
References:    <00af01bef543$0affe8f0$f29b12c2@secret.pythonware.com>
Message-ID: <5FDA3.8705$6k7.2096@nntp0.chicago.il.ameritech.net>

Fredrik Lundh  wrote in message
news:00af01bef543$0affe8f0$f29b12c2 at secret.pythonware.com...
> Fran?ois Pinard  wrote:
> > I do not fully understand the cost of splitting.  If strings are
immutable,
> > there is really no need to copy a slice contents.
>
> except that 1) each string needs an extra reference pointer
> (as long as we're using refcounting, at least), and 2) that people
> can easily end up in situations where huge strings are clogging
> up all available memory, just because some part of their program
> is hanging on on to a 2-character substring...
>
> (fwiw, the unicode string type implements substring
> support, using it for the "split" and "slice" methods.
> still not sure if that's really a great idea...)
>
> > Somewhere, it is said that some extra-space is allocated at end of
sequences,
> > to allow for some peaceful growth.  Does `str=str+str1' at least attempt
> > to use of this extra-space?  It should, in my opinion.
>
> how?  before you tell me, consider that strings are immutable,
> and variable names are references.

Ah-h!  Here goes the difference from Perl.  Perl strings (to my knowledge)
are mutable.  They consist of series of chunks - greatly speeds up all
splits/joins/substring replacement (same things with Perl arrays).  A little
slower on iterating through string contents.
Also, Perl DOES optimize compiled regexps, there's a great deal of analisys
done by regexp compiler to do that (anchors, long constant substrings,
branching prediction).

Alexei





From cfelling at iae.nl  Sun Sep 19 12:33:07 1999
From: cfelling at iae.nl (Carel Fellinger)
Date: 19 Sep 1999 18:33:07 +0200
Subject: Force anything to be a string.
References: 
Message-ID: <7s3383$gn9$1@vvs.superst.iae.nl>

jonathon  wrote:
> On Sat, 18 Sep 1999, Tim Peters wrote:
...
>>Using the builtin str function was the correct approach from the start.
>>From the error msgs you're getting, it looks most likely that you
>>accidentally rebound the name "str" to a string, so "str" no longer refers
...
> 	A little poking around, and I found an object called
> 	str in two modules that the script calls.   After changing
> 	the name of those variables, that error disappeared.  :-)

>>to the builtin function (builtin function names are not reserved, which is
>>both a feature and a bug <0.7>).

> 	A bug, IMNSHO.   A really bad bug, since I wasn't even using a
	module I wrote.   

You probably used `from other_module import *', you naughty boy:)
>From the docs I learned not to do that unless the imported module
was specifically designed for that construct, like Tkinter.

-- 
groetjes, carel



From 71351.356 at compuserve.com  Mon Sep 27 21:12:39 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Mon, 27 Sep 1999 21:12:39 -0400
Subject: Python guide Brain posted
Message-ID: <7sp529$ipg$1@ssauraab-i-1.production.compuserve.com>

FYI:

I created an initial version of a guide to the Python language using
Natrificial's TheBrain technology.
Its posted to their ThinkTank (a database of user created Brains) at:


I won't quote what the Brain is and all that (see www.theBrain.com); it may
appear I'm directly
advocating something or spamming.

Anyway, I think its pretty usable, especially for Python Newbies, like me.
Any suggestions on its layout would be appreciated.

Unfortunately, a 'guide' Brain in contrast to a web hosted 'Web Site' Brain,
requires Windows and a two meg download for a trial version of the client
 which doesn't overwrite DLLs!).   They state that other platforms, like
Unix and palm computers, are under development.

Me.











From john.grayson at gsc.gte.com  Tue Sep 14 21:45:43 1999
From: john.grayson at gsc.gte.com (john.grayson at gsc.gte.com)
Date: Wed, 15 Sep 1999 01:45:43 GMT
Subject: high school programming & python
In-Reply-To: <2BAF9178787E2D18.F80B717C3A0115B2.701123DC5A66FDBE@lp.airnews.net>
References: <2BAF9178787E2D18.F80B717C3A0115B2.701123DC5A66FDBE@lp.airnews.net>
Message-ID: <7rmto7$umu1@eGroups.com>

> If a mercury delay line won't pass it, it's not worth computing.

Ooh! This gives me goosebumps!

When I first started at Thames Polytechnic in London, they still
were keeping their Elliot system running, resplendent with its mercury
delay line.

Plus! (MS reference?) they changed multiple triodes every day.

You havn't lived until you have compiled Algol with paper tape for the
intermediate... 




From arcege at shore.net  Sun Sep 26 08:34:39 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sun, 26 Sep 1999 12:34:39 GMT
Subject: Read stdout from shell command, was: Becoming root
References:    <19990924165114.A753@krikkit.vollmert.home> <7sgbor$13vk$1@nntp6.u.washington.edu>
Message-ID: 

Donn Cave  wrote:
: Quoth wware-nospam at world.std.com (Will Ware):
: ...
: | The thing I'd really like to know is how to do the Python
: | equivalent of a shell command of this form:
: |
: | variable = `some shell command`
: |
: | where "some shell command" prints to standard output.
: | I find I do this quite frequently, usually using something like
: | os.popen('some shell command').readline()[:-1] and sure, it works,
: | but it's really ugly, and I'm always left with the nagging suspicion
: | that since I didn't explicitly close the os.popen(), it might still
: | be floating around.

: Empirically, the pipe file descriptor does close when the object goes
: out of scope, from what I see with other file descriptor unit numbers
: allocated subsequently by pipe() or open().

: I rolled my own pipe/fork/exec system a ways back and use it a lot,
: and you're welcome to it -
: ftp://ftp.u.washington.edu/pub/user-supported/donn/cmdproc.py

: I use it mostly the way you describe, to store the output of a
: command in a variable.  The long way to spell it is

:  variable = cmdproc.RaiseCommand('/bin/sh',
:                 ('sh', '-c', 'some shell command')).expand()

: (Actually I often invoke the command directly rather than through the
: shell, but the above example serves to illustrate both.)

: The RaiseCommand class also raises an exception when the command exits
: with an error, with the exception value set to the diagnostic (unit 2)
: output.  I think that's the feature that really sets this apart from
: just another way to spell popen().

There is also the standard library module "commands", which does all
but remove the trailing newline (if present).  It uses os.popen (and
implicit Bourne syntax), but does quote the arguments for you.

(Although, Donn's would probably be more efficient. :)

  -Arcege




From bernhard at alpha1.csd.uwm.edu  Wed Sep  1 16:13:50 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 1 Sep 1999 20:13:50 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com> <37CCE10D.D07D023B@strs.co.uk>
Message-ID: 

On Wed, 01 Sep 1999 09:17:17 +0100, Ian Clarke  wrote:
>> Are there any other major weaknesses of interpreted languages over
>> conventional languages such as C, C++, Ada etc.?

>Speed is the only one I can think of 

It is harder to do type-checking for the whole program, because most 
interpreted languages also have syntax allowing more dynamic constructs
and require less declaration.

>> What are the most popular interpreted languages around?

Beside from the other suggestions.

Basic, shell and lisp/scheme 
probably deserve mentioning.

	Bernhard
-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.    (www.uwm.edu/~bernhard)
Association for a Free Informational Infrastructure              (ffii.org)



From arcege at shore.net  Fri Sep 10 17:23:00 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 10 Sep 1999 21:23:00 GMT
Subject: Python-based IRC bots..and a #python channel
References: <37D879A1.EE4D10DE@op.net> <19990910223230.A849@optiplex.palga.uucp>
Message-ID: 

Gerrit Holl  wrote:

:> Also, is there an
:> "official" #python channel somewhere? If there isn't, we should
:> establish one. Any info would be greatly appreciated.
:> 

: I agree that we should establish one. But why should it be "official"? Can't we
: all just go to #python? It grows to the 'official' one of itself then.

I seem to remember there being a #python channel on EFnet some time
ago.  I'm not sure what happened to it tho (I don't see it there now).

  -Arcege




From robin at jessikat.demon.co.uk  Wed Sep  1 14:12:40 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 1 Sep 1999 19:12:40 +0100
Subject: tkinter & threads
Message-ID: 

Is Tkinter thread safe? I looked at Guido's brownian demo and there are
no special precautions about locking the canvas. Also I believe there is
some restriction on which thread Tk can live in.
-- 
Robin Becker



From legion at dialnet.net  Sat Sep  4 11:20:33 1999
From: legion at dialnet.net (Andrew Kesterson)
Date: Sat, 04 Sep 1999 10:20:33 -0500
Subject: "if" syntax error
Message-ID: <37D138C1.3EABC811@dialnet.net>

I've got a CGI script (python) set to check the value of form element
"Listing_Type", and it does it as thus:

CGIHeaders = cgi.FieldStorage()
if CGIHeaders["Listing_Type"].value == "Residential":

when I compile it, I get this error:

  File "C:\Program Files\Python\Source\Skydev\addlisting.py", line 82
     if CGIHeaders["Listing_Type"].value == "Residential":
      ^
 SyntaxError: invalid syntax

	There is NO WAY that this is invalid syntax! I've used if comparisons
like that one hundreds of times and compiled fine!
	What's going on? Or is it just that my CGI *really* doesn't want to run
from command line?

-- 
"I never came across anyone in whom the moral sense
was dominant who was not heartless, cruel, vindictive,
log-stupid and entirely lacking in the smallest sense
of humanity. Moral people, as they are termed, are
simply beasts. I would sooner have fifty unnatural
vices than one unnatural virtue."
--Oscar Wilde



From fdrake at acm.org  Fri Sep 10 12:13:58 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 10 Sep 1999 16:13:58 GMT
Subject: And now for something completely boneheaded...
In-Reply-To: <37D9136B.9298EFD1@nightwares.com>
References: <37D623A2.DC09F0CC@nightwares.com>
	<37D7E456.728C0BC1@nightwares.com>
	<37d7f333.949751@news.erols.com>
	<37D9136B.9298EFD1@nightwares.com>
Message-ID: <14297.11846.480435.373371@weyr.cnri.reston.va.us>

Robert Kern wrote:
 > Hmmm.  I think it may be time to reconsider the
 > distribution/documentation split.

Warren 'The Howdy Man' Ockrassa writes:
 > Sigh, no, it's not; all that information is included, in HTML form, in
 > the "Doc" directory in the Win32 install.

  The split Robert is referring to doesn't exist for Windows users
working from the binary installer.  The source package (used by many
Unix users) does not include any documentation, but used to include
the LaTeX source and pre-generated PostScript for all the standard
manuals.
  I still think the split is the right thing.  Most people don't
*want* the LaTeX source or even the PostScript.  Using a separate
distribution for the documentation allows users to select the formats
they're interested in without having to download a lot of other
material.


  -Fred

--
Fred L. Drake, Jr.	     
Corporation for National Research Initiatives



From super_banane at yahoo.com  Thu Sep  2 11:40:50 1999
From: super_banane at yahoo.com (super_banane at yahoo.com)
Date: Thu, 02 Sep 1999 15:40:50 GMT
Subject: getpass.getpass() bugged?
Message-ID: <7qm5q0$lct$1@nnrp1.deja.com>

The getpass.getpass() function should
be requesting a password without echoing it...
but unfortunately it does:

>>> import getpass
>>> getpass.getpass()
Password: hello
'hello'
>>>

Python I use is v1.5.2
Same behaviour on Solaris 2.6 and HPUX-10.20.
Works fine with Python v1.5.1 on Linux, though.

--
Nicolas


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From akuchlin at mems-exchange.org  Tue Sep 21 15:26:58 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 21 Sep 1999 15:26:58 -0400
Subject: Python for younger children
References: <37E70E89.55132AEA@bosware.com.au>
   <37E728B1.8ACA53FB@webone.com.au>
Message-ID: <3d1zbsvzel.fsf@amarok.cnri.reston.va.us>

Stuart Hungerford  writes:
>    I'd be interested to hear your experiences, my 10 and 8 year old are
>    keen to learn as well.  I've downloaded Alice (see www.alice.org)
> which
>    uses Python to manipulate 3D graphics objects, but I'm sure there
> must
>    be a simpler approach.

There's a turtle.py module in Python 1.5.2 that requires Tk.  Try this:

amarok lib-tk>python
Python 1.5.2 (#80, May 25 1999, 18:06:07)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from turtle import *
>>> reset()
>>> forward(30)
>>> right(90)

For some reason it only shows the turtle when it's actually moving;
someone should implement showturtle()/hideturtle().

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
The NSA response was, "Well, that was interesting, but there aren't any
ciphers like that."
    -- Gus Simmons, "The History of Subliminal Channels"




From bsb at baal.infodrom.north.de  Sat Sep 25 15:07:51 1999
From: bsb at baal.infodrom.north.de (Siggy Brentrup)
Date: 25 Sep 1999 21:07:51 +0200
Subject: python and I/O
In-Reply-To: "Bill Rodgers"'s message of "Sat, 25 Sep 1999 11:57:03 -0600"
References: <%4YG3.3867$UD4.20078@news.rdc1.az.home.com> <19990925130345.A11328@optiplex.palga.uucp> <%W7H3.1233$Pv.61638@news.uswest.net>
Message-ID: <87u2oibyig.fsf@baal.winnegan.fake>

"Bill Rodgers"  writes:

> I don't see how reading and writing to a file descriptor
> provides access to serial ports, parallel ports or memory
> mapped I/O.  Usually serial ports require more specific
> setup criteria such as baud rate, character format (N-8-1, etc).

If you are allowed to access I/O ports directly isn't a language, but
an operating system issue. Any decent OS, in particular a multitasking
one, must protect itself and its users from misbehaved programs.

>From your answer to Gerrit I presume you want to control a serial
port. If you're running on *nix, use /dev/ttyS* and look into the
termios(+TERMIOS) module for setting up serial parameters. I don't
care enough about Win?? to know their way.

CU
  Siggy

btw Gerrit: even with your own interpretation RTFM stays bad style.

-- 
>>>> In a world without fences, who needs gates?

Siggy Brentrup - bsb at baal.infodrom.north.de - voice: +49-441-6990134




From tim_one at email.msn.com  Fri Sep 10 02:16:27 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 10 Sep 1999 02:16:27 -0400
Subject: Python Error Messages
In-Reply-To: 
Message-ID: <000201befb54$03f58800$a52d153f@tim>

[jonathon]
> Why are PythonError messages so annoying?

Because the tracebacks you're getting aren't showing the lines of code that
are actually failing.  I can explain why that's happening, but it's obscure:
when Python goes to print a traceback, it uses the platform libc fgets
function to read one line at a time, until the target line number is
reached.  It passes fgets a buffer big enough to hold 1000 characters.
fgets won't actually read a full line if it sees 1000 characters go by
without seeing a newline, so any line of code in your .py file that's longer
than 1000 characters will screw up the line count hopelessly.

There are 7 lines of code in the script more than 1000 characters long, and
that's the source of your traceback woes.  The long lines are at:

744
745
746
20628
20629
20632
20633

Break those lines into non-insane pieces, and the traceback problem will go
away.

> I've been trying to debug a script for the last four days,
> to no avail.

This is a 21,000+ line script, without a class or module of its own, no
testing framework or even hooks, and hard-coded little integers on nearly
every line.  In all honesty, you may well spend the next four months trying
to debug it.

> With two of the four databases I run it on, the script does just what it
> is supposed to.  On the other two databases, it doesn't.

The most likely cause is that the latter databases contain data that the
code doesn't handle correctly, data in a form or combination that simply
never appeared in the former two databases.  That's almost certainly not a
Python problem, but logical errors in the code.

> The current exasperating error message:
>
> Traceback (innermost last)
>   File "mass.py", line 21102 in ?
>          print changed_field, " is record_list[28] "
>   File "mass.py", line 20610 in print_web_page
>          table_prefix = "'
   caption = '"Stamp Data"  "
   table_break = "  "
   table_suffix = "
 "
>  TypeError:  illegal argument type for built in operation.

The 3 long lines at 744, 745 & 746 are each getting counted as two lines by
fgets, so the traceback is actually showing you line 20610-3 = 20607.  The
true line 20610 is

   title_name = (title_prefix + country_name + bang_two + scott_data +
                 bang_two + michel_data )

One of title_prefix etc probably isn't a string (mostly likely is an
integer, or None).  IOW, the code is assuming that earlier code bound these
names to strings, but earlier code didn't.

> Now what is infuriating, is that the script has run over this
> particular code sequence at least 4000 times, before giving me
> this error.

That's normal for a data-dependent error.

> Questions:
>
> What would trigger these error messages, when there were no
> apparant problems earlier?

The overwhelmingly most likely cause is errors in the code that didn't show
up before simply because the pattern of data that triggers them didn't show
up before.

> How do I do exception handling that maintains something sane,
> when the script obviously isn't?

I don't mean to be abrasive, but the plain truth is that the script is
already a maintenance nightmare.  Without a rational design you're not going
to patch in robust behavior after the fact.  This script should be 10x
shorter than it is, and 10x faster too -- whoever wrote it was clearly not
an experienced Python programmer, and didn't do any of the things an
experienced one would have done to make recovering from errors pleasant.
Consider cutting your losses -- if you're the author, sorry, but you bit off
more than you're ready to chew.  Strongly consider hiring a Python
consultant to get you on the right track with this (not a plug -- I don't
work for money or love ).

> One thing i have noticed, is that the print statements I put
> in, to try to follow what was happening, tend to indicate
> that the actual problem is more than a couple of lines away
> from the line(s) that are quoted on the screen.

Right; see above.

> ...
> For the curious the other obnoxious error is:
> "IndexError: string index out of range"
>
> Obnoxious, because:
> A) It is supposed to be looking at a list, not a string.

The msg is almost certainly telling you the truth:  you may *expect* it to
be looking at a list, but it's telling you as plain as day that you actually
gave it a string.  That's a great clue, not a plague .

> B) The actual error is several hundred lines from where
>    the message occurs.

?  Sorry, don't know what that means.

if-you-thought-this-was-fun-try-rewriting-in-perl-ly y'rs  - tim





From jblake at stamp-coin.com  Sun Sep 19 01:39:33 1999
From: jblake at stamp-coin.com (jonathon)
Date: Sun, 19 Sep 1999 05:39:33 +0000 (UTC)
Subject: Force anything to be a string.
In-Reply-To: <000101bf0232$7792e840$072d153f@tim>
Message-ID: 

On Sat, 18 Sep 1999, Tim Peters wrote:

>[jonathon]
>> 	Question:
>>
>> 	1:	How can I ensure that string_check becomes a string,
>> 	and not remain an integer?  [  Other than error trapping at
>> 	the source. An option not available, since this script reads
>> 	files with bad data in them, to correct them. ]
>
>Using the builtin str function was the correct approach from the start.
>>From the error msgs you're getting, it looks most likely that you
>accidentally rebound the name "str" to a string, so "str" no longer refers

	Aaarrgh.  I ran the script with the statement
	print str, " is the trapped string."
	I totally expected the script to fail.

	Instead, it printed out a string "MMMMM"	

	A little poking around, and I found an object called
	str in two modules that the script calls.   After changing
	the name of those variables, that error disappeared.  :-)

>to the builtin function (builtin function names are not reserved, which is
>both a feature and a bug <0.7>).

	A bug, IMNSHO.   A really bad bug, since I wasn't even using a
	module I wrote.   

>The error msg there should look familiar.  It's telling you that the syntax
>looks like a call, but the object in the caller's position *can't* be called
>("call of non-function"); and it's also giving the type of the object that
>can't be called (str is of "type string").

	Another piece of wisdom to add to the list of things I
	couldn't find in _The Python Library Reference Manual_.

>diagnosis-may-be-easier-than-cure-ly y'rs  - tim

	I found a neat little script xref.py, that produces a cross
	reference for a file.  Threw it at my program, and then the
	modules, and found the very_badly_named_variables, that were
	causing those error messages.  
	
	xan

	jonathon




From actgal8 at aol.com  Fri Sep  3 19:09:53 1999
From: actgal8 at aol.com (actgal8 at aol.com)
Date: Fri, 03 Sep 1999 17:09:53 -0600
Subject: 18 AND OVER 59713
Message-ID: <030999170953@aol.com>

free xxx pics:

http://207.240.225.250


5LwxM\qH



From stadt at cs.utwente.nl  Wed Sep 22 09:23:25 1999
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Wed, 22 Sep 1999 15:23:25 +0200
Subject: cgi authentication
References:   <37E8902F.45998A4D@cs.utwente.nl> <00a401bf04d4$ad71d6c0$f29b12c2@secret.pythonware.com>
Message-ID: <37E8D84D.A94F8906@cs.utwente.nl>

Fredrik Lundh wrote:
> 
> Richard van de Stadt  wrote:
> > Then in SomeScript you can work with the variables 'login' and
> > 'password'. I guess you must have thought of this. But one
> > never knows :-)
> >
> > But you'll have to make your own (encrypted?) password file
> > for this to work (or get the source of htpasswd to find out
> > how passwords are encrypted for use with .htaccess).
> 
> iirc, they use crypt (just like /etc/password).  given
> that you've enabled the crypt module, you can use
> something like this:
> 
> def login(user, password, encrypted_password):
>     "Check if user would be able to login using password"
>     pw1 = encrypted_password
>     pw2 = crypt.crypt(password, pw1[:2])
>     return pw1 == pw2
> 
> (derived from an example in "(the eff-bot guide to)
> The Standard Python Library" -- out soon!).
> 
> finding the password file is left as an exercise...
> 
> 

1) How does one enable the crypt module if it is not
available in the Python library? Do I have to ask our
Python maintainer to rebuild it?

2) The crypt-statement looks strange to me.
Why are the first 2 characters of the encrypted
password (apparently retrieved from the password
file) feeded into crypt()?

Or is this just the standard way password encryption is done?

Richard.



From tjreedy at udel.edu  Sat Sep 25 01:20:55 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 25 Sep 1999 05:20:55 GMT
Subject: A more straightforward solution wanted
References: 
Message-ID: <7shm3n$hbi$1@news.udel.edu>

jblake at stamp-coin.com says...
        
>        I'm bashing my head trying to think of a more straightforward
>        way to do the following:
...
>        Is there a more straightforward method of doing this, other
>        than using yet another database, for the lookup functions, etc?

One possibility: use Python dictionaries
        
For instance, after 

>           if scott_number = 1:
>               return_list[44] = " "

mentally replace 

>               if scott_suffix = " ":
>                   return_list[45] = 1
>                   return_list[46] = "Type A"
>                   return_list[51] = " "
>                   return_list[52] = 1
>                   return_list[53] = " "
>               elif scott_suffix = "a":
>                   return_list[45] = 1
>                   return_list[46] = "Type B"
>                   return_list[51] = " "
>                   return_list[52] = 1
>                   return_list[53] = "b"
>               else:
>                  return_list[45] = 1
>                  return_list[46] = "Type A/Italic Z"
>                  return_list[51] = " "
>                  return_list[52] = 1
>                  return_list[53] = "a"

with
                 (return_list[45],
                  return_list[46], 
                  return_list[51],
                  return_list[52],
                  return_list[53]) = {" ":(1,"Type A"," ",1," "),
                                      "a":(1,"Type B"," ",1,"b")}.get(
                  scott_suffix, (1,"TypeA/Italic Z"," ",1,"a"))


Now, the dictionary and default each depend on the scott_number 1 to 5, 
so each can be replaced by a list of choices indexed by scott_number.

IE, (untested)

#before loop, so execute only once
list_of_dicts = [
  "dummy list item for unused index 0",
  {" ":(1,"Type A"," ",1," "),"a":(1,"Type B"," ",1,"b")},
  {#dict for scott_number 2},
  #etc for 3, 4, and 5
]
list_of_default_tuples = [
  "dummy for 0",
  (1,"TypeA/Italic Z"," ",1,"a),
  #etc for 2,3,4,5
]
...

Then, within loop (or func called for each line),
possible after checking scott_number in range 1...5,

  (return_list[45], return_list[46], return_list[51],
   return_list[52], return_list[53]) =\
  list_of_dicts[scott_number].get(
    scott_suffix, list_of_default_tuples[scott_number])

This compresses about 100 lines to less than 20.

Terry J. Reedy




From tom22 at verio.com  Wed Sep  1 15:18:35 1999
From: tom22 at verio.com (Tom)
Date: Wed, 01 Sep 1999 12:18:35 -0700
Subject: Newby:Looking for an article describing C++ vs. Python
Message-ID: <37CD7C0B.351D5E8D@verio.com>

Hi,

This is from a real newby to Python and should not be considered flame
bait.

I have programmed in C for years and have recently begun using C++ and
classes, etc. I am interested in Python/Perl/Tlc as a scripting language
but without having studied either of them I would like to know, for
example, what kinds of things that you can do in C++ that you can not do
in Python. I know that the Python language is written in C and I know
that a compiled language will tend to run faster that an interperted
language.

Is there an article where someone has done a comparison of C++ to
Python?

Thanks
Tom





From fredrik at pythonware.com  Thu Sep  2 08:39:32 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 2 Sep 1999 12:39:32 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References:  <37CE562F.1A368670@strs.co.uk> <002701bef535$68a88920$f29b12c2@secret.pythonware.com> <37CE69C0.6AC977E0@strs.co.uk>
Message-ID: <009701bef540$356cfe00$f29b12c2@secret.pythonware.com>

> I am not talking about what the user notices, I am talking about
> what the developer does

hmm.   I'm pretty sure you used the word "user" in
your last post.   sorry for not understanding that
you really meant something else.

over and out /F




From mwh21 at cam.ac.uk  Fri Sep  3 13:24:49 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Sep 1999 18:24:49 +0100
Subject: Newbies: Re: Returning none
References: <000001bef265$fe64f340$192d153f@tim>  <37CF5499.BA24C33E@prescod.net>   
Message-ID: 

"Evan Simpson"  writes:

> Michael Hudson  wrote in message
> news:Pine.LNX.4.10.9909031629210.21765-100000 at atrus.jesus.cam.ac.uk...
> > Got bored.
> [and implemented the whole damn thing!]
> 
> Wow.  That looks ... remarkably painless.

That's why I did it. Hacking the source to Python can be almost as
much fun as hacking Python. If only all C code in the world was
written by Guido.

>  Why don't you think it's a good idea?

Because it would annoy me. It adds a complication to Python's simple
conceptual model that, as John Skaller eloquently explained, is one of
it's main attractions.

Another problem is that given a "callable object" `f', there is no way
to check whether it returns Nothing other than by try:-ing it; and that
sucks, IMHO. You can't even compare it; 

if f() == Nothing:

raises TypeError. 

Perhaps the test at the end of the CALL_FUNCTION should be extended to
allow COMPARE_OP. But that's inconsistent; there's no satisfactory
resolution IMHO, so I vote for the status quo.

I-feel-this-proposal-is-unlikely-to-be-accepted-ly y'rs
Michael



From campbeld2 at my-deja.com  Fri Sep 17 10:20:36 1999
From: campbeld2 at my-deja.com (campbeld2 at my-deja.com)
Date: Fri, 17 Sep 1999 14:20:36 GMT
Subject: os.popen results diff when executed in browser vs DOS
Message-ID: <7rtin6$9oa$1@nnrp1.deja.com>

This sample program illustrates my dilema.

I am trying to execute a command ( could be
any command ) within Python and capture the
output.  It works fine when I run it at command
level, however when I invoke the script in the
browser the code does not return any values from
the os.popen(...)readlines.  I tried try: and
except: but nothing is trapped.

Any ideas?

------

"""
 test.py :
 returns correct results at command level (great)
 returns nothing through Browser !!!
"""
import os

print 'Content-Type: text/html\n\n',
for line in os.popen( 'cd','r' ).readlines():
	print '...within for loop \n\n'
	cd = line

print '
cd = ', cd, '\n\n'

#end of test.py


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From davidopp at megsinet.net  Sun Sep  5 00:27:33 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Sun, 05 Sep 1999 00:27:33 -0400
Subject: V - GUIDO !!!  was Re: Pirate -> RAD
References: <7qno38$6uk$1@news.is.co.za> <37D17998.5CF4@warande.uu.nl>
Message-ID: <37D1F135.6F242B93@megsinet.net>

I have an idea for creating a  WYSIWYG  GUI builder for Python that will allow you
to create slick looking GUI's just by dragging and dropping elements into a visual
design space that will output code into wxPython.  After a user was satisfied with
the appearance of the GUI, the program would output the code into wxPython code
which would be seamlessly integrated with the code that had already been written
inside of a design space (make it so that it would work flawlessly as either a
component of the major IDE's including PythonWin, IDLE, Python in a DOS box or
have it callable from a command line).  It will also be able to take written code
and show a preview of what your GUI would look like without having to first save
the program and run it thru the Python interpreter.  This would allow you to
instantly see the effect of changes made to your code. It would be awesome to
integrate it seamlessly with the PythonWin IDE and also be callable from the
command line in IDLE or Python in a DOS box.

If you would like to join with me in this creative endeavor (code name V - GUIDO
which stands for Visual Graphical User Interface Development Optimizer), I would
be happy to share the details with you.

Sincerely,

David Oppenheimer

P.S. This is still just in the brainstorming stage right now.  I am doing
conceptual planning right now; features, how it should look, how it should act,
etc.  I would ask anyone who is serious about doing such a project to join forces
with me and see if we can make something happen.


Martin Oti wrote:

> Jacques Oosthuizen wrote:
> >
> > Hi All
> >
> > I am busy working on a tool I named Pirate ("Python RAD") it has a component
> > library similar to Delphi and shall in the future work with win32,gtk,qt. It
> > will have a IDE.
> >
> > Before I do to much and work to hard on it, will anyone be interested in
> > using a
> > tool like this.
>
> Absolutely! I'm new to Python and the more I find out about it, the more
> I like it. :) One major aspect which I miss is a good RAD GUI-building
> tool.
>
> I use Visual Basic at work, mainly for within-company scripting and
> automation tasks, and I often need to whip stuff up pretty fast. I've
> already started to use Python as much as possible and with a RAD
> GUI-building tool I will be able to ditch VB for good. :) There are some
> lovely Python GUI toolkits out there, but I'm no expert programmer and I
> need to whip GUIs up pretty fast.
>
> Considering the masses of Visual basic programmers out there, I think a
> visual RAD GUI-builder/IDE for Python will have a large audience. It
> would be particularly handy for those (like me ;) who would also like
> VB-like functionality on Linux. Indeed, considering that KOffice will
> use Python as it's extension language, Python could then become the
> scripting-language-for-the-masses on Linux the way VB is on MS Windows.
>
> Martin
>
> PS: A zillion thanks to Mark Hammond for the PyWin IDE, Robin Dunn for
> wxPython (which I only just discovered yesterday), and the entire Python
> community for a great programming/scripting language/tool. :)
>
> PS2: I don't know what this would involve, but it would be handy if the
> GUI-Builder could also interface with (soon to be) existing Python GUI
> toolkits (like TkInter, wxPython and Parrot). I have no clue how
> difficult it would be to separate the front-end from the back-end.
>
> PS3: Pirate and Parrot. How ironic. ;) Is this a coincidence?
>
> ------------
> Trying is the first step towards failure.
>        -- Homer Simpson




From cmedcoff at sprynet.com  Sat Sep 18 21:02:06 1999
From: cmedcoff at sprynet.com (Charles Medcoff)
Date: Sat, 18 Sep 1999 21:02:06 -0400
Subject: Seeking GUI advice
Message-ID: <7s1bs8$7jp$1@nntp2.atl.mindspring.net>

I don't want to start a flame war but I'm looking for advise on what GUI to
use with Python.

I'm pretty much a newbie still at this point.  I've only written a few
simple scripts for source control, builds, etc. to help out my C++
development.  I'm getting interested in using the GUI aspects of python.
I've used Tk a little bit.  There isn't any really great documentation for
Tkinter IMHO although I do appreciate the efforts of Fredrik Lundh.  I
understand that he may be publishing a Tkinter book this fall.  Is this
true?

I do most all of my work on the Windows platform and usually use Pythonwin
for my development but I'd like to create portable scripts if possible.
Beside there isn't any good documentation for Pythonwin's GUI either IMHO.
(Thanks for a cool development environment though;)  I guess I'm looking for
some documentation that does a fair amout of hand holding alla "Learning
Python".

I kind of thought that Tkinter was being swamped out by Pythonwin, and
wxPython. Is this true?  Is there something else available?  Is here
something else Python developers are going to standardize on?  Any advise
appreciated.

Chuck





From moun at usenix.org  Fri Sep 10 20:20:07 1999
From: moun at usenix.org (Moun Chau)
Date: Sat, 11 Sep 1999 00:20:07 GMT
Subject: DSL Conference Hotel & Pre-Reg Savings Deadline - Sept 13, 1999
Message-ID: 

2ND CONFERENCE ON DOMAIN SPECIFIC LANGUAGES (DSL '99)
October 3-5, 1999
Omni Hotel, Austin, Texas, USA
Sponsored by USENIX, The Advanced Computing Systems Association
In cooperation with ACM SIGPLAN and ACM SIGSOFT

---------------------------------------------------------
Hotel and Registration Savings Rate Ends
Monday, September 13, 1999
---------------------------------------------------------

The Omni Hotel is offering the USENIX Conference
special rate through Monday only, and after that the rate will
increase.  Please make sure to say that you are attending the
USENIX Conference to get the discount rate.
Single/Double Occupancy - $125

Omni Austin Hotel - Downtown
700 San Jacinto at 8th Street
Austin, TX  78701
1-800-843-6664
Local telephone: 1-512-476-3700

DSL Conference Topics include:
* Different Approaches To DSL Construction
* New DSLs For Problem Domains such as Specifying Hardware Circuits
  and Robot Control Protocols
* Creating Data-Intensive Web Sites and Collaborative Applications
* Domain-Specific Languages for Programming and Security in Active
  Networks, by Carl A. Gunter, University of Pennsylvania
* A Methodology for Designing Domain-Specific Languages Using Program
  Specialization, by Charles Consel, IRISA/University of Rennes
* Language Technology for Performance and Security
* The Next 700 Markup Languages

---------------------------------------------------------
See USENIX website for complete conference details
and Online Registration
http://www.usenix.org/events/dsl99
---------------------------------------------------------




From mavip5 at inet.polyu.edu.hk  Fri Sep 10 01:33:33 1999
From: mavip5 at inet.polyu.edu.hk (Li Dongfeng)
Date: Fri, 10 Sep 1999 13:33:33 +0800
Subject: Reading plain text file database tables
References: <37D7508A.B02789A7@inet.polyu.edu.hk>
Message-ID: <37D8982D.59CD3D15@inet.polyu.edu.hk>

I have come up with a rarther ugly, slow, but
working RE solution for the problem. The two
functions can read comma delimited tables and
space delimited tables. Quotation, backslash escape
are considered. Python source below.

-----------------------------------------------------------
def cut_comma(lines):
    """This function is used to read comma delimited files(.csv).

    Input: the rows of the data file in a list 'lines'.
    Output: a list of list x, x[i] is the i'th row of the input, 
            x[i][j] is the j'th data item on the i'th row.

    The scanning respects quoting, e.g., the row
          "Bond, James", 13, 45
    will be recognized as three column ["Bond, James", "13", "45"]
successfully. 
    Escape characters like 
          'That\'s absolutely possible.', 1999, 8, 30
    are treated correctly.

    Shortcoming: very slow when reading large datasets. On a SUN Sparc
2, reading
    10000 lines of input took 20 seconds. Suggestion is welcome.

    Author: Li Dongfeng, ldf at statms.stat.pku.edu.cn
    Last modified: 1999.9.10
    """
     
    r = re.compile(r"""
	\s*     # data item can start with any number of spaces
	(?P        # start of data item we need to extract
          '' | ""        # empty string
         |
	  '.*?[^\\]'   # matches any thing enclosed in '...'(can have comma)
		         #  but the last "'" have no '\' before it. Find first ending
'
	 |
	  ".*?[^\\]"   # matches any thing enclosed in "..."(can have comma) 
		       # but the last '"' have no '\' before it
	 |
	  [^"',][^,]*?  # matches anything that do not contain a comma, 
		       #and not starting with "'" or '"' or comma
	 |
	  "[^,]*?(?!.*?[^\\]")   # anything starting with '"' and have no
comma,
                               # no matching '"' behind
	 |
	  '[^,]*(?!.*?[^\\]')   # anything starting with "'" and have no comma,
                               # no matching "'" behind
         |           # can be empty between two commas
	)            
	(?:\s*
	 ,  #  end with a comman(can have spaces before it)
	)    
       """, re.VERBOSE)
    r2=re.compile(r"""^(['"]).*\1$""", re.MULTILINE)  # anything quoted
    def quote(match):
        return eval(match.group())
    if type(lines) is type(()) or type(lines) is type([]):
        # sub comma with newline, but respect string quoting
        x = map(lambda s, r=r: r.sub("\\g\n", s), lines)
	# unquote all quoted
	x = map(lambda s, r=r2, f=quote: r.sub(f, s), x)  
	# split the rows
        x = map(lambda s: string.split(s, "\n"), x)
    else:
	x = r.sub("\\g\n", lines)
	x = r2.sub(quote, x)
        x = string.split(x, "\n")
    return x

def cut_space(lines):
    """This function is used to read space delimited files.

    Input: the rows of the data file in a list 'lines'.
    Output: a list of list x, x[i] is the i'th row of the input, 
            x[i][j] is the j'th data item on the i'th row.

    Any number of spaces(space, tab) can be used between two data items.
    The scanning respects quoting, e.g., the row
          "James Bond"  13  45
    will be recognized as three column ["James Bond", "13", "45"]
successfully. 
    Escape characters like 
          'That\'s absolutely possible.'  1999  8  30
    are treated correctly.

    Shortcoming: very slow when reading large datasets. On a SUN Sparc
2, reading
    10000 lines of input took 20 seconds. Suggestion is welcome.

    Author: Li Dongfeng, ldf at statms.stat.pku.edu.cn
    Last modified: 1999.9.10
    """
     
    r = re.compile(r"""
	(?P        # start of data item we need to extract
          '' | ""        # empty string
         |
	  '.*?[^\\]'   # matches any thing enclosed in '...'(can have comma)
		         #  but the last "'" have no '\' before it. Find first ending
'
	 |
	  ".*?[^\\]"   # matches any thing enclosed in "..."(can have comma) 
		       # but the last '"' have no '\' before it
	 |
	  [^"'\s][^\s]*?  # matches anything that do not contain a space, 
		       #and not starting with "'" or '"' or space
	 |
	  "[^\s]*?(?!.*?[^\\]")   # anything starting with '"' and have no
space,
                               # no matching '"' behind
	 |
	  '[^\s]*(?!.*?[^\\]')   # anything starting with "'" and have no
space,
                               # no matching "'" behind
	) 
	(?:\s+)  #  end with a space
       """, re.VERBOSE)
    r2=re.compile(r"""^(['"]).*\1$""", re.MULTILINE)  # anything quoted
    def quote(match):
        return eval(match.group())
    if type(lines) is type(()) or type(lines) is type([]):
        # sub comma with newline, but respect string quoting
        x = map(lambda s, r=r: r.sub("\\g\n", s), lines)
	print x[0]
	# unquote all quoted
	x = map(lambda s, r=r2, f=quote: r.sub(f, s), x)  
	print x[0]
	# split the rows
        x = map(lambda s: string.split(s, "\n"), x)
	print x[0]
    else:
	x = r.sub("\\g\n", lines)
	x = r2.sub(quote, x)
        x = string.split(x, "\n")
    return x

-------------------------------------------------------------------------

Li Dongfeng wrote:
> 
> Do we have a module to read plain text file
> database tables?
> 
> All the data management software, e.g. excel,
> dBase, SAS, etc., support input/output a table
> from/to a plain text file, fields can be separated
> by their column position, by spaces, by tabs,
> by commas, etc.
> 
> How can we read this kind of file into a matrix like
> structure(list of lists)? I have written one reading
> files with fixed-width fields. For delimited files,
> simply using string.split work most of the time, but
> fails reading lines like
> 
>   "Peter Thomson"  25  36
> 
> or even
> 
>   "Peter\" Thomson" 25 36
> 
> I think this is a common task, so maybe someone has
> already given a very good solution.
> 
> Li Dongfeng



From MHammond at skippinet.com.au  Fri Sep  3 19:50:16 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 4 Sep 1999 09:50:16 +1000
Subject: Announce: CVS Access to source code for Python for Windows Extensions
Message-ID: <7qpmso$4jj$1@m2.c2.telstra-mm.net.au>

All source code for the Python for Windows extensions are now being managed
by CVS.  Anyone can use anonymous access to ensure they always have the
most recent sources available.  There is also a "checkins" mailing list
provided, so you can be informed of all CVS checkins made into this
repository as they happen (or daily if you use the digest mode!)

All the gory details can be found via
http://starship.python.net/crew/mhammond/cvs.html

Many thanks to the Python Pros (http://www.pythonpros.com) for assisting in
providing this service.

Mark.





From nhead at houbits.com  Wed Sep 15 03:09:27 1999
From: nhead at houbits.com (Nigel Head)
Date: Wed, 15 Sep 1999 09:09:27 +0200
Subject: [Newbie] test_signal hangs
References: <37DEDFF7.1DC7C4B3@siprelle.com>
Message-ID: <37DF4626.B0E86158@houbits.com>

Lynn:

>  And now I'm trying to run the test set, but it hangs on
> test_signal--I let it run for more than an hour, and it just sat there.
>
> Should I even care? Or is this a real problem?

I had exactly the same problems (manual edits to configure - just doing
configure.h by hand didn't work for me - and hanging of test_signal). I also
had sporadic failures of other tests, which would all run when called
individually -- this I put to down to the draconian limitations on virtual
address space that the iServer boxes appear to apply!

I ended up just deleting test_signal to get the rest of the test suite to
run and ignoring the problem. It seems to be a non-issue at least as far as
running Zope is concerned (saying this will probably get it to come back and
bite me now though :->).

Regards, and apologies for the slightly off topic Zope parts of the reply,
        Nigel.

PS: Make sure to read the docs in doc/FAQ.txt before you try installing Zope
- you'll have trouble if you don't. The relevant point is attached below --
note however that the makefile.pre.in that needs patching is NOT either of
the ones in the Zope tree but the one from the python installation which
gets copied across during the Zope build.

PPS: I also found it necessary to build python with the full prefix
(/home/usr/.....) and run Zope in non-virtual mode for things to work
smoothly - if you find a way to build so that it'll run in the chroot'ed
environment I'd love to hear about it.

PPS: After all this I shall be moving away from Iserver -- I've found
another bunch who offer completely naked RedHat boxes, with unrestricted
root account for the same price -- this way I'll be able to run Zserver on
port 80, instead of hiding behind Apache. If you're interested let me know
in private email and I'll pass on the details - don't want to be flamed for
introducing commercial stuff here.

------------------

Zope Installation Frequently Asked Questions
--------------------------------------------

  Installing Zope

    1. I am trying to build Zope on my BSD based system, but I
       keep getting "symbol not found" errors and the build
       scripts stop. What do I do?

       On some BSD based systems (BSDI, BSD/OS), you need to edit
       the Makefile.pre.in files in your Zope installation to add
       -rdynamic to the LDSHARED makefile variable. On some systems
       it may also be necessary to change the file named 'do.py'
       located in the 'inst' directory of your Zope installation,
       changing the line: do('make')  to: do('make -k').




From psd2300 at saintjoe.edu  Thu Sep 16 13:56:51 1999
From: psd2300 at saintjoe.edu (Phil Schroeder)
Date: Thu, 16 Sep 1999 12:56:51 -0500
Subject: high school programming & python
References:  <37DF9080.21A8D285@calibre-dd.com>  <37DFED78.6B258366@bioreason.com> 
Message-ID: <37E12F63.68737737@saintjoe.edu>

James Whitley Moore wrote:
> 
> Why not end this thread in the usual way?
> 
> I never wanted to be a high-level language programmer.
> 
> I always wanted to be. . .

a race car driver.  heck, maybe now i can write a python script to do it
for me.

-phil schroeder

 measly college student



From drushby at email.com  Thu Sep 30 23:56:40 1999
From: drushby at email.com (David Rushby)
Date: Thu, 30 Sep 1999 23:56:40 -0400
Subject: PC Speaker Sound
Message-ID: 

I know this is an offbeat question, but how does one go about generating PC
speaker tones from Python?
Is there a function like GWBasic's SOUND?
Thanks.





From jblake at stamp-coin.com  Thu Sep  9 20:34:51 1999
From: jblake at stamp-coin.com (jonathon)
Date: Fri, 10 Sep 1999 00:34:51 +0000 (UTC)
Subject: Python Error Messages
In-Reply-To: <14295.35617.292624.957926@gargle.gargle.HOWL>
Message-ID: 

On Thu, 9 Sep 1999, Charles G Waldman wrote:

>jonathon writes:
> > 
> > 	The current exasperating error message:
>It seems pretty likely to me (but this is just a guess!) that the
>error is on the line:
>
>add_keywords = ( pre_keywords + country_name + country_section_name +
> scott_data + michel_data + y_t_data + s_g_data + minkus_data +
> end_of_line )

	That wasn't the error.  Some of the errors turned out to 
	be the result of my not setting "return return_object"
	in a couple of defs.  Python very nicelly returned NULL.
	Still haven't managed to track the rest down.  :-(
	
	For the curious the other obnoxious error is:
	"IndexError: string index out of range"
	
	Obnoxious, because:
	A) It is supposed to be looking at a list, not a string.
	B) The actual error is several hundred lines from where
	       the message occurs.

>What seems most annoying to me, and I really don't have a clue about,
>is why you're getting an off-by-one error on the line number.  Is
>there something strange about this file perhaps?

	The file it is reading is roughly 4 MB in size.  Each line is
	roughly 420 characters long.  

>By the way,  I'd write the line in question as:
>
>add_keywords = "%s%s%s%s%s%s%s%s%s" % (pre_keywords, country_name,
>	     country_section_name, scott_data, michel_data, y_t_data,
>	     s_g_data, minkus_data, end_of_line )

	Did that, and replaced most, if not all of the similar lines.
	Didn't fix the error, since it wasn't the error, but it did 
	speed up the script by a considerable margin.  :-)

	xan
	
	jonathon		




From clgonsal at keeshah.penguinpowered.com  Fri Sep  3 12:54:37 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Fri, 03 Sep 1999 16:54:37 GMT
Subject: Spanish Inquisition?!
References: <37CFC38C.20A9C267@megsinet.net>
Message-ID: 

On Fri, 03 Sep 1999 08:48:12 -0400, David Oppenheimer
 wrote:
> David O.
> 
> "Any Sufficiently Advanced Technology Is Indistinguishable From Magic"
> - Arthur C. Clarke

That looks awfully familiar...

> P.S. Creating compiled code from Python  that didn't require the
> interpreter to be distributed with it would be high magic indeed.

Yeah, especially compiling those eval and exec functions...

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From jetman3 at netzero.net  Fri Sep 24 07:15:39 1999
From: jetman3 at netzero.net (The Jetman)
Date: Fri, 24 Sep 1999 07:15:39 -0400
Subject: [PyWin] Does Win32RAS Work ?
References:  <7scon8$p5a$1@nnrp1.deja.com>
Message-ID: 

G. Lewis wrote in message <7scon8$p5a$1 at nnrp1.deja.com>...
>
>
>>     Have spent a couple of hours trying to narrow down a
>> prob w/ Win32RAS and I'm at the point where I suspect it simply
>> doesn't work.  Have trimmed down the rastest script to simply
>
>Works for me, though i don't use it for much. I just use it in a script to
>check if i have a connection like this...
>
>    import win32ras
>    cons = win32ras.EnumConnections()
>    if not cons:
>    print 'No Connection...'
>
>Hehe, that's all i use it for! But it works for that at least....
>
>
>
>
>Sent via Deja.com http://www.deja.com/
>Share what you know. Learn what you don't.

    It's been hit and miss for me and I wanted to use it for something
paying   I'm posting this msg via a RAS connection that I
established w/ win32ras, but PyWin itself expired just as the connection
was made, apparently due to a fault in the VC R/T lib.  As originally
coded, the rastest script fails catastrophically on my Win95 system.  But
I'm
going to keep trying for a while longer....Jet

--

================================================================
  In Paris, they simply stared when I spoke to them in French;
  I never did succeed in making those idiots understand their
  own language.         - Mark Twain, The Innocent Abroad,1869
=======================================  jetman3 at netzero.net  ==







From nolan at ethereal.dhis.org  Thu Sep 30 00:39:50 1999
From: nolan at ethereal.dhis.org (Nolan Darilek)
Date: 29 Sep 1999 23:39:50 -0500
Subject: Nolan's question of the day: distributed servers?
Message-ID: <87btal3td5.fsf@ethereal.dhis.org>

I've kept up a good tradition (Maybe not, depending on your point of
view :) of asking one question per day. Here's today's. :)

My current big project is a MUD server in Python. I have some
interesting ideas, only one of which I'll bore you all with. :) I
honestly don't know how Python will perform in this area. I'm not
aiming my code at lower-end processors though, so I'm not terribly
worried, but in order to alleviate some of this, I'm trying to either
use an existing object distribution scheme such as CORBA/ILU, or
*shudder* write my own. :)

Let me start by saying that I am not intimately familiar with CORBA or
ILU. I know what they were meant to accomplish, and I could probably
write a simpler program using them, but my needs are slightly
different, and I don't know if these schemes will handle them.

Basically, persistent objects are loaded into Namespaces. A Namespace
is a class which stores all objects in a specified location (Location
meaning a database) as local variables. So, if you had an object
called Ball in a database which is stored in a file called
"objects.db", you could do:

# Open the database and load all objects.
root = Namespace("objects.db")
# Access our Ball.
ball = root.Ball
# And then bounce it.
ball.bounce()

Since a namespace itself is a persistent object, databases can store
other namespaces. However, namespaces don't just access databases;
they can also access servers. So, you could do:

root.server1 = Namespace("thematrix.mud.org", "4000)
and then all objects which are available to the server
thematrix.mud.org 8000 would be available. Initially this scheme will
require a trusted environment, and security should be implemented
later.

So, a network of namespaces will exist, each accessing either a
database or a server. My current concern, though, is that the server
shouldn't have to know how every single object on every single server
works before it starts. From what I've seen of CORBA (And correct me
if I'm wrong.) class interfaces need to be specified before hand and
compiled to skeletons. This wouldn't work for me, since if someone on
server x decides to code, say, a flying car, making a skeleton
wouldn't be practical.

So my current questions are, is it possible to somehow export a
dynamic interface in realtime, as opposed to carefully specifying an
interface beforehand and loading it into the server? Another idea
which I have been considering is breaking the object into two
components, an exportable version and a server-resident version. So,
if someone gets a ball from another server, the user's server is only
sent a ball with its description and a list of commands/methods. If
the user chooses to bounce the ball, the user's server passes the
bounce request to the server which originated the ball, the ball
processes the method and returns its results to the user. This method
seems somewhat unclean though, since the person's ball wouldn't bounce
if the server which originated it was down. If, instead, the user's
server knew how to bounce the ball, the server would be able to work
independently. Another option would be to pickle requested objects and
dump them over the network to the originating server, possibly using
zlib for speed. So, if someone requests a copy of root.server1.ball,
the ball object would be pickled at the destination and sent across
the network. Perhaps this would be the best solution, since both sides
of the connection would have a copy of the object, though changes in
one would still need to effect the other, which may prove difficult.

Has anyone done something like this? If so, do you have any
recommendations? Thanks for listening/reading the somewhat confusing
and confused rants of a newbie. :)



From m.faassen at vet.uu.nl  Mon Sep 27 14:43:55 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Sep 1999 18:43:55 GMT
Subject: Python compilers?
References: <87aeq86am8.fsf@ethereal.dhis.org>
Message-ID: <7sodtb$oau$1@newshost.accu.uu.nl>

Nolan Darilek  wrote:
[snip]
> So, are there any native Python compilers? Someone on IRC mentioned a project
> called Viper, and my searches seem to indicate that it is under
> development, but is there any information on the web about its status?

John Skaller is actively working on it, last I heard. He is first working on
viperi, however, a Python interpreter, after that he plans to move on to
viperc, a Python compiler. viperi seems to actually run Python code, but
I don't think viperc is out of the design phase yet.
John wants to do all kinds of grand things with viper, such as type
inference and other advanced code analysis.

> Also, What is swallow? I tried doing a websearch for swallow and
> Python, and ended up with tons of Holy Grail references, but nothing
> to answer my question. :) I ask because I've seen swallow coupled with
> viper in a few messages; is swallow another python/pseudopython
> interpreter/compiler?

Swallows was an earlier vaporware project that I came up with. The idea
behind Swallow is to implement a compiler for a limited subset of Python
(along with nonintrusive type annotations). This could be used to create
Python based fast extension modules.

Currently I lack the time to do a lot about it, though. Maybe later, but
currently it is complete vaporware! No website. :)

Cool that people ask, though.. What some posts and a cool name can do. :)

There *are* actual Python to C compilers out there; others can point you
to the relevant web pages better than I could. The disappointment here is
that they tend to create executables that aren't significantly faster
than Python. That's why for Swallow I think the subset part + type 
annotations were essential (as concepts). My hope is that this could
lead to translation to actually *fast* C (or native) code. And the
restrictions also allow us to skip the difficult parts of type 
inferencing and dealing with lots of Python's dynamism.
 
> One final question. How compatible/cross-platform are Python
> bytecodes? If I wanted to distribute applications in Python while
> hiding the sources from the user, would it be possible to distribute
> .pyc files, or the bytecodes specific to a specific version/platform?

As far as I'm aware Python's bytecodes are fully portable to other 
platforms.

Regards,

Martijn

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



From sitric at home.com  Fri Sep 10 09:37:10 1999
From: sitric at home.com (Eric B.)
Date: 10 Sep 1999 06:37:10 PDT
Subject: How do I connect to Postgress from Windows version of Python 1.5.2 ??
Message-ID: <37D90926.ED4DAE31@home.com>

We have Postgress running on a Linux box and I am trying to connect to
it from my Windows 98 machine using Python 1.5.2 (Windows package).  I
looked into "PyGress" but that seemed to be set up for a non-Windows
flavor of Python.  Can anyone help me?
Thanx!!!

-Eric B.  (sitric at home.com)




From hniksic at srce.hr  Sun Sep 19 18:29:02 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 20 Sep 1999 00:29:02 +0200
Subject: Marking translatable strings
References:   
Message-ID: <87vh964jtd.fsf@pc-hrvoje.srce.hr>

Fran?ois Pinard  writes:

> Python should be simple as well.  Yet, after the XEmacs
> internationalisation failure, one good lesson, that we would be
> foolish to not retain, is that we should take time to think things
> out, before starting to think too big.

I'm curious -- what "XEmacs internationalization failure" are you
talking about?



From junkster at nospam.rochester.rr.com  Tue Sep 28 22:40:30 1999
From: junkster at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Wed, 29 Sep 1999 02:40:30 GMT
Subject: Python TimeServer (RFC 868)
Message-ID: 

Hi Folks!

	I've been messing around with a python timeserver that I've 
written.... 

	There's only one problem, I've got the time standardized, and seems
	to work correctly.

	But time clients don't seem to want to communicate with it.

	I'm not exactly certain of the problem, but I suspect something.

	I'm dumping the "seconds" value, in otherwords, the seconds are being
	output to the socket, as a text value.

	I believe that I need to send it as a binary value (a 4 byte 
integer), instead of a 
	string... But can't figure out a way to do this...?

	Any suggestions?

	Or examples of a working timeserver in python >g<

			- Benjamin

================================
Please feel free to copy any and or
all of this sig.
A little something for spam bots:

root at localhost postmaster at localhost admin at localhost
abuse at localhost postmaster at 127.0.0.1

Chairman William Kennard: bkennard at fcc.gov 
Commissioner Susan Ness: sness at fcc.gov
Commissioner Harold Furchtgott-Roth: hfurchtg at fcc.gov
Commissioner Michael Powell: mpowell at fcc.gov
Commissioner Gloria Tristani: gtristan at fcc.gov
consumerline at ftc.gov
fccinfo at fcc.gov
ssegal at fcc.gov




From mal at lemburg.com  Fri Sep  3 07:53:40 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 03 Sep 1999 13:53:40 +0200
Subject: Python speed and `pcre'
References:    <00af01bef543$0affe8f0$f29b12c2@secret.pythonware.com> 
Message-ID: <37CFB6C4.35578D33@lemburg.com>

[substrings]

Note that Python 1.5.2 already support such beasts... albeit
rather well hidden (for obvious reasons ;-) and not fully
interoperatable yet:

>>> s = '12345678901234567890'
>>> snippet1 = buffer(s,8,3)
>>> print snippet1
901
>>> snippet2 = buffer(s,14,4)
>>> print snippet2
5678
>>> import string
>>> string.join(snippet1,snippet2)
'95678056781'
>>> string.split(snippet2,'6')
['5', '78']
>>> string.join([snippet1,snippet2])
Traceback (innermost last):
  File "", line 1, in ?
TypeError: first argument must be sequence of strings

You can always turn a buffer back into a string using the str()
function.

Perhaps the string module should be extended to use "s#"
more often...

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   123 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From jmrober1 at ingr.com  Tue Sep  7 14:26:45 1999
From: jmrober1 at ingr.com (Joseph Robertson)
Date: Tue, 07 Sep 1999 13:26:45 -0500
Subject: case for open lang on win32 - fill in the table
References: <7r0049$bfh$1@nnrp1.deja.com> <7r0d3t$chk$1@newshost.accu.uu.nl> <7r1hab$9bd$1@m2.c2.telstra-mm.net.au> <7r3fhd$oqb$1@nnrp1.deja.com>
Message-ID: <37D558E5.3D1D7BDF@ingr.com>


nielsenjf at my-deja.com wrote:

> power and cross-platform (Where I work we're going 100% microsoft and
> 100% visual basic). The main argument being visual basic/vbscript is so
> easy to use and the only other viable option is C++ which is too hard
> (in their opinion).

Good grief, man get out of there FAST!  100%VB  yuk.  Also sounds like a
group that doesn't know how to use the "right tool for the job".

> IMHO, python has a feel similar vb/vbscript except it doesn't have
> the vb world's limitations.

Python is many orders of magnitude better than VB...

> For example, you can do everything COMwise in C++.
> Choose any other language and you start to make compromises.
> What is there that you can do in C++ that you can't do in python?
> In what cases is visual basic(a probably weaker language) better than
> python to use?

Real world example:  I have several VB apps that I "inherited" over the
years.  One is fairly large, and was using DAO simply to handle the large
text files (via ISAM) and provide sorting, searching, etc.  Using Mark
Hammonds' Python-COM support I have ripped the DAO nonsense, made a mess
into a logical class heirarchy, and provided common base functionality
available to other supporting apps.

So I have 4 major Python-COM objects for this app and VB is just the GUI to
tie it all together.  The app is faster, handles very large (90K+lines)
flatfiles, reusable, smaller, and cleaner(more logical).  Once I get a
descent grid done in Tk even VB can be removed.  Did I mention portable,
the COM code are merely wrappers, so the base code is there for any
platform to use.  And Python runs on about 28 more platfoms than
VB.

Point is, use VB for its strength...GUI frontends, and Python for its
strength...strong, fast core functionality.
6 versions and VB string handling still suks roks...

As an added bonus, I also recreated 2 of the COM objects on VC++6.0, and it
took about 3 times as long to do it, and I started with a working Python
object.  Needless to say I am a firm Python-COM believer.



Thanks, Mark!


Joe Robertson
jmrober1 at ingr.com




From hniksic at srce.hr  Fri Sep 10 14:11:13 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 10 Sep 1999 20:11:13 +0200
Subject: Python example: possible speedup?
References: <000b01befb24$05ca6fe0$f0a0143f@tim>
Message-ID: <87g10my6v2.fsf@pc-hrvoje.srce.hr>

"Tim Peters"  writes:

> Was the Perl code to which you're comparing it as general as your
> Python code?

The Perl code was almost equivalent to the first posted version of
Python code -- the only important difference is that it used a
functional approach rather than OOP.  I can mail it to you if you are
curious.

Generality is important to me because another day I might wish to do
something else with those files, like draw a tree of dependencies, or
something.  In that case, I'd surely want to reuse the existing
reader/parser.  Speed is good, but not so good to justify loss of code
reuse.

Robert's solution is ideologically similar to the `grep' solution
proposed below.



From robin at jessikat.demon.co.uk  Sun Sep 12 00:02:38 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 12 Sep 1999 05:02:38 +0100
Subject: Matching strings with regexps
References: <37D8982D.59CD3D15@inet.polyu.edu.hk>
 <000001befcb7$83e9b9a0$4c2d153f@tim>
Message-ID: 

In article <000001befcb7$83e9b9a0$4c2d153f at tim>, Tim Peters
 writes
>
...
>
>give-a-man-a-meal-and-you've-fed-him-for-a-day-
>    give-a-man-a-regexp-and-you-may-as-well-kill-him-ly y'rs  - tim
>
>
...
so to turn a dead man into a zombie we pose the timbot the problem of
recognizing those regular expressions which he so kindly quoted for us
to immediately forget and then whine later that we lost the ref.
-mind numbedly yrs- 
Robin Becker



From da at ski.org  Tue Sep 28 17:41:34 1999
From: da at ski.org (David Ascher)
Date: Tue, 28 Sep 1999 14:41:34 -0700 (Pacific Daylight Time)
Subject: Global confusion
In-Reply-To: <816010E2456BD111A48700805FBBE2EEFDF634@ex-quebec-u1.baan.com>
Message-ID: 

On Tue, 28 Sep 1999, Gaetan Corneau wrote:

> # Function to add a paragraph
> def AddParagraph(style, text):
>     Srv.ActiveDocument.Paragraphs[nbPar].Range.InsertParagraphAfter()
>     Srv.ActiveDocument.Paragraphs[nbPar].Range.Style =
> Srv.ActiveDocument.Styles(style)
>     Srv.ActiveDocument.Paragraphs[nbPar].Range.InsertAfter(text)
>     nbPar = nbPar + 1

> NameError: nbPar
> 
> I looked in my "Learning Python" book and told my friend to insert  "global
> nbPar" as the first line of the "AddParagraph" function.
> It works, but I don't understand why he has to do that...
> 
> Anybody who can exlain?

The problem is that because nbPar is *assigned to* in the function body,
Python tags all references to 'nbPar' within that function as references
to a local variable (which fails, because nbPar doesn't have a 'local'
value inside the function when the nbPar + 1 is evaluated. The general
rule about Local-Global-Builtin namespaces applies to variables which are
not otherwise tagged.  The global declaration is one such (explicit)
tagging, which is needed to counter the effect of the (implicit) tagging
done by Python at compilation time.

Side note 1: in the next version of Python, a different error message
             UnboundLocalError (or such) will be raised instead.

Side note 2: JPython does something else altogether.  

Side note 3: Since you do want to modify the global variable, adding the
             global declaration is the right thing to do.

Hope this helps.

--david ascher






From yopen at my-Deja.com  Mon Sep  6 22:55:51 1999
From: yopen at my-Deja.com (Bill Wilkinson)
Date: Mon, 6 Sep 1999 21:55:51 -0500
Subject: Printing under Pythonwin
References: <37D4FC28@pcmgw.phm.gov.au>
Message-ID: 

This kind of printing from Pythonwin is not implemented yet....  Perhaps
soon though...  In the mean time, you can open up your py files in word or
wordpad and print them from there.... Or do what I do and write a script
that uses py2html to convert them to html and then print them out....  This
way you can get color printouts....(if that is important to you).

fwiw  Pythonwin is incredibly feature rich.  Source printing is sure to
follow soon... :)


John Hirsch  wrote in message
news:37D4FC28 at pcmgw.phm.gov.au...
>
> I have a problem printing from the pythonwin editor.
> The Print, Print Preview and Page Setup are greyed out.
> I am printing on a network printer and have no problem printing from
> other programs.
> Thanks for any help.
>
>





From Doug at news.vex.net  Mon Sep 13 01:26:06 1999
From: Doug at news.vex.net (Lewis)
Date: 13 Sep 1999 05:26:06 GMT
Subject: Examples using urllib, htmllib
References: <7rgon4$4a0$1@nnrp1.deja.com>
Message-ID: <7ri1te$2ki7$1@hub.org>

On Sun, 12 Sep 1999 17:43:03 GMT, nolanl at my-deja.com  wrote:
>Can someone point me to some example Python
>scripts that I can download that show accessing
>the web using the urllib, and htmllib.  I'm

I have a couple of scripts with extremely sloppy code at
http://www.vex.net/~x/python_stuff.html which might have 
a few hints for you. One is a simple web page 'crawler' 
which seeks and downloads background images, using urllib.
There is also a file upload CGI which is not what you want,
but demonstrates a bit at least how web uploads work somewhat.
Good luck.





From 71351.356 at compuserve.com  Sat Sep 25 08:39:54 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Sat, 25 Sep 1999 08:39:54 -0400
Subject: Source line number for log purpose (newbie)
References: <7shiqh$d96$1@ssauraaa-i-1.production.compuserve.com> <19990925130431.B11328@optiplex.palga.uucp>
Message-ID: <7sig2t$oar$1@ssauraaa-i-1.production.compuserve.com>

Thanks, I'll look at the Exception module again.

Gerrit Holl  wrote in message
news:19990925130431.B11328 at optiplex.palga.uucp...
> T. C. Mits:
> > In c there was a way to get the current source line number via a macro,
> > __LINE__.
> >
> > I need to be able to do the same thing.  I created a logger class that
uses
> > a level setting.  At high levels I want to
> > log the source line number where the inline log.prtl( level, msg) method
> > call appears.
> >
> > I'm sure this is easy but am getting lost digging through the Python
source
> > and docs.
> >
>
> That's possible via the 'Exception' module.
>
> regards,
> Gerrit.
> --
> RTFM: Read That FreshMeat
>
>





From jblake at stamp-coin.com  Fri Sep 24 17:38:26 1999
From: jblake at stamp-coin.com (jonathon)
Date: Fri, 24 Sep 1999 21:38:26 +0000 (UTC)
Subject: A more straightforward solution wanted
Message-ID: 

	
	I'm bashing my head trying to think of a more straightforward
	way to do the following:

		Read a file.
		Validate that the data in the file is correct.
		If it is correct, add several fields to each record,
			content of the added fields depends upon the
			content in the fields allready present. 

	Here is _one_ short example of what the functions currently
	look like.  It also happens to be one of the more straight
	forward ones.  In most of the others, the values are vastly
	different.  

	Is there a more straightforward method of doing this, other
	than using yet another database, for the lookup functions, etc?
	<< I hesitate to create a database with the data to be added,
	because I'm not sure how many more fields will eventually
	be requested for the database that the output of the script
	currently goes to.  The other question is what to use for the
	lookup database. >> 

	
def vryburg ( record_list ):
   return_list = record_list
   return_list[2] = "Vryburg"
   return_list[30] = 216501
   temp_number = return_list[31]
   scott_number = return_list[6]
   scott_suffix = return_list[7]
   if temp_number == 0 :
       return_list[38] = "regular issues"
       return_list = valid_scott_number_range( return_list, 1, 5)
       return_list[91] = "Cape Of Good Hope"
       return_list[92] = "Vryburg: Boer Occupation"
       return_list = valid_scott_prefix( return_list, "N", "Issued in Vryburg Under Boer Occupation")
       if return_list[26] in [ "sp-", "sp+" ]:
           print "Invalid Scott Number"
       else:
           return_list[50] = "SACC "
           if scott_number = 1:
               return_list[44] = " "
               if scott_suffix = " ":
                   return_list[45] = 1
                   return_list[46] = "Type A"
                   return_list[51] = " "
                   return_list[52] = 1
                   return_list[53] = " "
               elif scott_suffix = "a":
                   return_list[45] = 1
                   return_list[46] = "Type B"
                   return_list[51] = " "
                   return_list[52] = 1
                   return_list[53] = "b"
               else:
                  return_list[45] = 1
                  return_list[46] = "Type A/Italic Z"
                  return_list[51] = " "
                  return_list[52] = 1
                  return_list[53] = "a"
           elif scott_number = 2:
               return_list[44] = " "
               if scott_suffix = " ":
                   return_list[45] = 2
                   return_list[46] = "Type A"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = " "
               elif scott_prefix = "a"
                   return_list[45] = 2
                   return_list[46] = "l for 1"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = "c"
               elif scott_suffix = "b":
                   return_list[45] = 2
                   return_list[46] = "Type B1"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = "b"
               else:
                   return_list[45] = 2
                   return_list[46] = "Type A/Italic Z"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = "a"
           elif scott_number = 3:
               return_list[44] = " "
               if scott_suffix = " ":
                   return_list[45] = 3
                   return_list[46] = "Type B"
                   return_list[51] = " "
                   return_list[52] = 3
                   return_list[53] = " "
               else
                   return_list[45] = 3
                   return_list[46] = "Type B/Italic Z"
                   return_list[51] = " "
                   return_list[52] = 3
                   return_list[53] = "b"
           elif scott_number = 4
               if scott_suffix = " ":
                   return_list[45] = 4
                   return_list[46] = "Type A"
                   return_list[51] = " "
                   return_list[52] = 4
                   return_list[53] = " "
               elif scott_suffix = "a":
                   return_list[45] = 4
                   return_list[46] = "Type B"
                   return_list[51] = " "
                   return_list[52] = 4
                   return_list[53] = "b"
               else
                   return_list[45] = 4
                   return_list[46] = "Type A/Italic Z"
                   return_list[51] = " "
                   return_list[52] = 4
                   return_list[53] = "a"
   elif temp_number == 501 :
       return_list[38] = "Vryburg Occupation Issues"
       return_list = valid_scott_number_range( return_list, 1, 4)
       return_list[91] = "Cape Of Good Hope"
       return_list[92] = "Vryburg: Boer Occupation"
       return_list = valid_scott_prefix( return_list, "N", "Issued in Vryburg Under Boer Occupation")
       if return_list[26] in [ "sp-", "sp+" ]:
           print "Invalid Scott Number"
       else:
           return_list[50] = "SACC "
           if scott_number = 1:
               return_list[44] = " "
               if scott_suffix = " ":
                   return_list[45] = 1
                   return_list[46] = "Type A"
                   return_list[51] = " "
                   return_list[52] = 1
                   return_list[53] = " "
               elif scott_suffix = "a":
                   return_list[45] = 1
                   return_list[46] = "Type B"
                   return_list[51] = " "
                   return_list[52] = 1
                   return_list[53] = "b"
               else:
                  return_list[45] = 1
                  return_list[46] = "Type A/Italic Z"
                  return_list[51] = " "
                  return_list[52] = 1
                  return_list[53] = "a"
           elif scott_number = 2:
               return_list[44] = " "
               if scott_suffix = " ":
                   return_list[45] = 2
                   return_list[46] = "Type A"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = " "
               elif scott_prefix = "a"
                   return_list[45] = 2
                   return_list[46] = "l for 1"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = "c"
               elif scott_suffix = "b":
                   return_list[45] = 2
                   return_list[46] = "Type B1"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = "b"
               else:
                   return_list[45] = 2
                   return_list[46] = "Type A/Italic Z"
                   return_list[51] = " "
                   return_list[52] = 2
                   return_list[53] = "a"
           elif scott_number = 3:
               return_list[44] = " "
               if scott_suffix = " ":
                   return_list[45] = 3
                   return_list[46] = "Type B"
                   return_list[51] = " "
                   return_list[52] = 3
                   return_list[53] = " "
               else
                   return_list[45] = 3
                   return_list[46] = "Type B/Italic Z"
                   return_list[51] = " "
                   return_list[52] = 3
                   return_list[53] = "b"
           elif scott_number = 4
               if scott_suffix = " ":
                   return_list[45] = 4
                   return_list[46] = "Type A"
                   return_list[51] = " "
                   return_list[52] = 4
                   return_list[53] = " "
               elif scott_suffix = "a":
                   return_list[45] = 4
                   return_list[46] = "Type B"
                   return_list[51] = " "
                   return_list[52] = 4
                   return_list[53] = "b"
               else
                   return_list[45] = 4
                   return_list[46] = "Type A/Italic Z"
                   return_list[51] = " "
                   return_list[52] = 4
                   return_list[53] = "a"
           elif scott_number = 5:
               return_list[5]  = "Vryburg: British Occupation"
               return_list[92] = "Vryburg: British Reoccupation"
               return_list[45] = 1
               return_list[46] = " "
               return_list[51] = " "
               return_list[52] = 2
               return_list[53] = "a" 
   else:
       return_list[38] = "Miscellaneous: Other"
       return_list[31] = 999
       return_list[28] = "Sec"
   return return_list

#    This top listing of the tupple is the authoratative one.
#
#    record_list
#    #    This is a list
#    0    This is a null --- i forgot that lists start with zero, not one.
#    1    string_record_id
#    2    temp_country_text : var_country_name
#    3    string_country_number
#    4    string_country_section
#    5    temp_scott_prefix
#    6    string_scott_number
#    7    temp_scott_suffix
#    8    temp_michel_prefix
#    9    string_michel_number
#    10   temp_michel_suffix
#    11   temp_grade : var_grade
#    12   temp_mint_used : var_mint
#    13   string_quantity
#    14   string_price
#    15   temp_description
#    16   temp_certificate
#    17   temp_date
#    18   string_cost
#    19   temp_additional_description
#    20   temp_list_field
#    21   temp_stamp_finder
#    22   temp_photo
#    23   temp_image
#    24   temp_deleted
#    25   temp_del_date
#    26   var_field_list
#    27   date_today
#    the following are for internal program use only
#    28   temp_field_list
#    29   temp_record_id    [ This is slated for elimination ]
#    30   temp_country_number : var_country_number
#    31   temp_country_section : var_country_section : this is a number
#    32   temp_scott_number
#    33   temp_quantity
#    34   temp_price
#    35   temp_cost
#    36   var_graphic_url
#    37   var_html_url
#    38   name_country_section
#    39   ultimate_file_destination
#    40   standard_computer_stamp_number
#    #    Fields between 40 & 70 are for parsing stamp catalog numbers
#    41   Y_T_prefix
#    42   Y_T_number
#    43   Y_T_suffix
#    44   stanley_gibbons_prefix
#    45   stanley_gibbons_number
#    46   stanley_gibbons_suffix
#    47   minkus_prefix
#    48   minukus_number
#    49   minkus_suffix
#    50   specialized_catalog_1_name
#    51   specialized_catalog_1_prefix
#    52   specialized_catalog_1_number
#    53   spcialized_catalog_1_suffix
#    54   specialized_catalog_2_name
#    55   specialized_catalog_2_prefix
#    56   specialized_catalog_2_number
#    57   spcialized_catalog_2_suffix
#    58   specialized_catalog_3_name
#    59   specialized_catalog_3_prefix
#    60   specialized_catalog_3_number
#    61   spcialized_catalog_3_suffix
#    62   specialized_catalog_4_name
#    63   specialized_catalog_4_prefix
#    64   specialized_catalog_4_number
#    65   spcialized_catalog_4_suffix
#    66   specialized_catalog_5_name
#    67   specialized_catalog_5_prefix
#    68   specialized_catalog_5_number
#    69   spcialized_catalog_5_suffix
#    #    The following are for internal file manipulation
#    70   expiration_date          # today's date + 10 days
#    71   read_bad_data_from_here
#    72   the_c_base_is
#    73   the_scratch_file_is
#    74   the_graphic_prefix_is
#    75   the_error_file_is
#    79   the_test_file_is
#    77   number_columns
#    78   " "
#    79   scott_prefix_name
#    80   validate_line
#    The following are background details about the stamp issue
#    81   date_of_issue
#    82   denomination_of_issue
#    83   currency_of_issue
#    84   background_colour
#    85   foreground_color
#    86   watermark
#    87   type_of_printing
#    88   type_of_paper
#    89   luminesence
#    90   specialized_catalog_1_country_name
#    91   specialized_catalog_2_country_name
#    92   specialized_catalog_3_country_name
#    93   specialized_catalog_4_country_name
#    94   specialized_catalog_5_country_name
#    95   stanley_gibbons_country_name
#    96   michel_country_name
#    97   y_t_country_name
#    98   minkus_country_name
#    99   scott_country_name
#   100   topical_category_1
#   101   topical_category_2
#   102   topical_category_3
#   103   topical_category_4
#   104   topical_category_5
#   105   topical_category_6
#   106   topical_category_7
#   107   topical_category_8
#   109   topical_category_9
#   110   topical_category_10





From kernr at mail.ncifcrf.gov  Mon Sep 13 01:45:55 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Mon, 13 Sep 1999 05:45:55 GMT
Subject: ANN: PyFT 0.5 - Python Wrapper for FreeType
Message-ID: <37dc8d9e.43792124@news.erols.com>

I haved used SWIG to wrap the FreeType library, a TrueType font
rendering engine.  It can be used to render text onto a 2-color bitmap
or anti-aliased text onto a 256-grayscale pixmap.  These readily
convert to PIL Images.

Information on FreeType is on their site http://www.freetype.org
Documentation on PyFT is available on
http://starship.python.net/crew/kernr/PyFT.html

PyFT is located here:
http://starship.python.net/crew/kernr/Projects.html

Robert Kern

PyFT
0.5 - Wrappers for the FreeType library for rendering TrueType
fonts (13-Sep-99)

--
Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From vapicel at tin.it  Mon Sep  6 22:55:46 1999
From: vapicel at tin.it (vapicel at tin.it)
Date: Mon, 06 Sep 1999 19:55:46 PDT
Subject: red and white WINE STRICTLY KOSHER
Message-ID: <7r0uoa$das$5876@nslave1.tin.it>

Dear friend,

This is a special offer to give you the opportunity to taste  a special wine "strictly Kosher " product in Tuscany in Pitigliano (GR).

Pitigliano is a small town on the tuscan hill, where a flourishing Jewish community had lived and grown since the Middle Ages and there Jews found a quiet refuge. For this reason Pitigliano was called "Little Jerusalem". In the ancient Jewish area, there is the Synagogue and its entrance door is shown on the
label of the bottle.

Only untill 15 September 1999 will be possible to buy two bottle of this wine (1 red and 1 white) at
special price of  U.S.$ 24.00. transport not includet.

Booking and payment:
Will be necessary to book by e-mail, so will be possible for us to calculate the cost of transport.
After we have receiving the booking's e-mail with the exact address, in 24 hours, you will receive
by e-mail the exact complete cost and the form of payment.

                                                                                                   best regards  
                                                                                              Vincenzo Apicella





From greg.ewing at compaq.com  Mon Sep  6 18:07:44 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 07 Sep 1999 10:07:44 +1200
Subject: how often does re.compile compile?
References: <7r1451$3n7$1@nnrp1.deja.com> 
Message-ID: <37D43B30.F31F6E0D@compaq.com>

"A.M. Kuchling" wrote:
> 
> wade at lightlink.com  wrote:
> >If I say re.compile() inside the function
> >definition, does the pattern get compiled afresh every time the function
> >is called?

The re.compile call certainly gets executed on every call.

> There's a cache containing the most recent 20-odd
> patterns, so if you're using fewer patterns, the re.compile() will ...
> find the precompiled cached pattern

If you don't want to rely on that, you can use the standard
hack for creating local precomputed things:

def f(..., pat = re.compile(my_expr)):
  ...

Just make sure you don't accidentally call f with one
too many parameters :-)

Greg



From tim_one at email.msn.com  Sat Sep 25 21:21:56 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 25 Sep 1999 21:21:56 -0400
Subject: Dialect, A Python derived environment (and language ?)
In-Reply-To: <1273949950-4000037@hypernet.com>
Message-ID: <000401bf07bd$86475b60$602d153f@tim>

[Tim, notes that the Dialect IDE
    http://www.aristar.com/DeveloperTools.html
 makes a 685Kb executable out of the 21-byte Dialect program
    print "Hello world"
]

[Gordon McMillan]
> Gee, just ran your app thru Installer. For the exe that's
> Run+CArchive([HelloWorld.py]) it comes out to 24K. For the
> one that's got python15.dll etc. built-in, it's still only 277K.

Bloating size by a mere factor of 13,507 is a killer advantage for sure
.  OTOH, all of Dialect's std libraries (primarily GUI, files, sockets
and some Windowsisms) are included in the Dialect exe, so e.g. there's no
significant additional size penalty for a "Hello world" that instead pops up
a dialog box with a button or two.  If they bother to compress anything
(doesn't look like they do now), the difference between 685 and 277 won't
last.

Now if a Dialect *fan* were to post here, I'd say they'd say this misses the
point entirely.  The point for them is that *making* the .exe is a
no-brainer:  select File -> Make EXE in the IDE, then click "Make Now" after
confirming that the IDE guessed reasonable answers to the four questions it
asks in the box that pops up (name of main module, which directory it lives
in, path to Dialect interpreter, and output exe name).  It was so easy, even
I managed to do it the first time.

every-language-on-earth-does-*something*-better-than-python-ly
    y'rs  - tim





From amir_e_a at netvision.net.il  Sun Sep 19 12:55:45 1999
From: amir_e_a at netvision.net.il (Amir E. Aharoni)
Date: Sun, 19 Sep 1999 18:55:45 +0200
Subject: indentation in XEmacs
Message-ID: <37E51590.DB041B1D@netvision.net.il>

Hello

Can anyone please tell me how do i configure XEmacs to auto-indent with
tabs instead of spaces?

I have

(custom-set-variables
 '(tab-width 4)
 '(indent-tabs-mode t))

in my ~/.emacs, but it still uses four spaces for indentation.

My configuration:

RH Linux 6.0 on a Pentium / XEmacs 21.1 May 1999 / python-mode 3.105

-- 
mailto:amir_e_a at netvision.net.il
Amir Elisha Aharoni, the Original Israelite
"Barkeep! Put all my meals in one plate.
 Don't ask me what kind of music I play.
 I play the good kind." - Oliver Lake



From thomas at bibsyst.no  Tue Sep 28 02:59:30 1999
From: thomas at bibsyst.no (Thomas Weholt)
Date: Tue, 28 Sep 1999 08:59:30 +0200
Subject: Python Berkley DB and speed issues
Message-ID: <37F06752.7D98775@bibsyst.no>

Hi,

Was just wondering about how Python is when it comes to speed using it`s
built-in dictionaries. I`m converting a perl-project where a Berkley DB
hash-table stored several million entries about files stored on cd-roms.
The speed in the perl-project was ok, and it seems like Python has a
very similar approach to this method, using pickle- and shelve-modules (
or Berkley DB). My question is : will the speed drop as the data gets
larger, using pickle and shelve, or do I have to use Berkely DB? We`re
talking millions of entries here. The whole thing will be indexed, so
that you look up a seperate table for words appearing in file-entries
and get a list of keys back. It`s not a problem to use Berkley DB, but
if I don`t have to ... I`ll prefer to use the built-in stuff.

People advised me to use a bigger databasesystem instead of the built-in
stuff and Berkley DB, but it worked fine. Berkley DB is a standard
core-module with Python 1.5 isn`t it? Any experience with similar
problems?

If anybody has any input on how to solve/optimize this problem, I`ll
appreciate it. The whole thing is going to be a cd-indexing project
released under the GPL-licence, with advanced search, report and
scanning-features ( Allready got a working Perl-version, but want to
move it to Python as soon as possible :-> ).

Thomas Weholt



From sblakey at u.washington.edu  Tue Sep  7 20:05:48 1999
From: sblakey at u.washington.edu (S. Blakey)
Date: Tue, 7 Sep 1999 17:05:48 -0700 (PDT)
Subject: Python interface to Groupwise mailoxes?
Message-ID: 

Is there an existing python interface to Novell'sGroupwise mailboxes?
The excellent library reference mentions the mailbox module (section
12.18) which provides access to many standard mbox formats, but I see no
reference to this one.
Failing an existing python module, are there ANY open-source programs that
interface with Groupwise that I can use as a starting point to build such
a module?  Even some good documentation of Novell's proprietary crap would
appreciated.




From emile at fenx.com  Wed Sep  1 18:51:00 1999
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 1 Sep 1999 15:51:00 -0700
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com><37CCE10D.D07D023B@strs.co.uk><7qjgq8$k1i$1@newshost.accu.uu.nl> <14285.38262.884171.416195@dolphin.mojam.com>
Message-ID: 

Skip Montanaro  wrote in message
news:14285.38262.884171.416195 at dolphin.mojam.com...

> So, aside from the differences in how you get from source to running
> program, depending on your definitions, they can all be called
compiled or
> interpreted languages.
>

>From my perspective, <25+years business basic> it's interpretive if it
politely stops when it hits a bug, and waits for you to fix the program
and continue.

If-you-can't-change-a-running-program-then-it's-compiled-ly yrs

--

Emile van Sebille
emile at fenx.com
-------------------







From souvienne at my-deja.com  Sat Sep  4 03:42:50 1999
From: souvienne at my-deja.com (souvienne at my-deja.com)
Date: Sat, 04 Sep 1999 07:42:50 GMT
Subject: Chinese Interpreters needed!!!
Message-ID: <7qqihq$qt3$1@nnrp1.deja.com>



Language Line Services is the world's largest provider of over-the-phone
interpretation (OPI), helping people communicate regardless of the
language they speak.  Language Line Services provides OPI from English
into 140 languages, 24 hours a day, seven days a week.  Our clients are
hospitals, law enforcement agencies, insurance companies, credit card
companies, court system, immigration, and 911 centers throughout the
United States and Canada.

Language Line customers call the Monterey Communications Center where an
English-speaking Answer Point takes their language request and connects
the client to the appropriate interpreter.  Once the interpreter is
connected, a three-way telephone conversation among the client, their
non-English speaking customer and the Interpreter ensues.

All of our interpreters work from their home offices throughout the
United States and Canada. Language Line Interpreters gain experience in
a wide range of subjects while helping Language Line customers and
non-English speakers communicate.

We are currently looking for interpreters for Mandarin Chinese, and
Cantonese.  The person must be fluent in English and Mandarin Chinese or
Cantonese, and has good customer service skills.  Interpretation
experience is a plus.

We pay $12.50/hr, and we need someone who can work 40 hours per week
during the daytime.  Currently there are no openings for part-time
position.

Please e-mail your resume to cathylo at languageline.com
Or fax it to 1-800-496-5330 Attn: Cathy Lo
Thank you.




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From " HPeter.Friedrichs" at alliedsignal.com  Tue Sep 21 18:28:00 1999
From: " HPeter.Friedrichs" at alliedsignal.com (H. P. Friedrichs)
Date: Tue, 21 Sep 1999 15:28:00 -0700
Subject: Linux Dependencies! Grrr...What now?
Message-ID: <37E8066F.1C13BF33@alliedsignal.com>

While I have having successfully written some scripts for python 1.4 (on
my rh5.0 linux box), and successfully installing mysql, I have been
trying for a week to figure out how to get the mysqldb module to work
(to allow python to talk to mysql).

First, I found out that mysqldb will only work with python 1.5.2...which
necessitated an upgrade. Dependencies! Dependencies! I ended up
installing python, python-devel, blt, tix _tkinter and tkinter. (I've
been warned that as the result of my upgrade, I have now "broken"
netcfg, but I'll deal with that on another day.)

Now I go to install mysqldb, and it complains, among other things, of
missing files and directories. Looks like I need mysql-devel.
Dependencies! Dependencies!

I go to the web and find it. I try to install the rpm, and it won't. It
seems I need glibc6 (2.0 and 2.1). But what about everything compiled
with the existing library? What happens if I "upgrade" to glibc6? What
do I risk breaking by doing that?



--

-----------------------------------------------
| The opinions expressed here are my own,     |
| and not necessarily those of my employer.   |
-----------------------------------------------





From tim_one at email.msn.com  Tue Sep 14 22:55:58 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 14 Sep 1999 22:55:58 -0400
Subject: Python Error Messages
In-Reply-To: 
Message-ID: <000101beff25$d65dd8e0$612d153f@tim>

[Tim, explains that source file lines longer than 1000 chars screw
 up Python tracebacks]

[Fran?ois Pinard]
> Seriously, couldn't we consider this as an implementation bug?  If
> fgets did not return a whole line, the line count should not get
> incremented.

I'm afraid I gave such a dry account of it that it came off looking like a
bit of conventional wisdom Python programmers "are just supposed to know".
Sorry about that -- it was a surprise to me too, and my explanation was
derived from having tracked down the cause just minutes before posting.

That is, I agree it's a bug, and already entered a bug report and submitted
a patch to repair it.  A fix should appear in the CVS tree "soon".  Since
nobody reported this before, and tens of thousands of Python programmers
must have stared at hundreds of tracebacks each, I believe we can safely
conclude that the danger of running without the fix would be hard to
underestimate .

in-the-meantime-just-set-word-wrap-in-.py-files-to-999-ly y'rs  - tim





From skaller at maxtal.com.au  Tue Sep 28 15:26:38 1999
From: skaller at maxtal.com.au (skaller)
Date: Wed, 29 Sep 1999 05:26:38 +1000
Subject: May I loop over a changing list?
References:  <14312.58596.157032.657963@gargle.gargle.HOWL> <14312.60400.366044.933684@dolphin.mojam.com> 
Message-ID: <37F1166E.A2A4AF2C@maxtal.com.au>

Olaf Delgado wrote:
> Of course, the interpreter
> might do some clever optimization for builtin types, or there might be
> alternative ways to implement for loops in future versions of python.
> 
> So, the _real_ question (tm) here is: is the passage in the tutorial a
> "Don't do it, it's undefined." or a "Oh no, kiddy, this will confuse your
> little mind."?

This is a long standing problem with python. The 'reference manual'
documents what the current version of the interpreter does,
rather than providing a language specification. However, the
manual is now in a reasonable state, and a more formal partition
between language semantics and implementation details could
be implemented, if only the two could be disentangled :-)

-- 
John Skaller, mailto:skaller at maxtal.com.au
1/10 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
downloads: http://www.triode.net.au/~skaller



From hniksic at srce.hr  Fri Sep 10 14:06:22 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 10 Sep 1999 20:06:22 +0200
Subject: Python example: possible speedup?
References: <87puztv2gt.fsf@pc-hrvoje.srce.hr>  <87aeqw2vrc.fsf@pc-hrvoje.srce.hr> <37D84D5A.B2562946@home.com>
Message-ID: <87k8pyy735.fsf@pc-hrvoje.srce.hr>

Doug Hellmann  writes:

> > Thanks for the suggestion.  However, I do install new packages
> > often, so that's not really an option.
> 
> Probably not every time you run your app, though.

I use the application to see which packages waste too much space, so
that I know which ones to remove.  Not to mention that I run nightly
updates of my Debian system.  It is practically *guaranteed* that the
files will have changed between two runs of the program.

Sorry.  I like the idea of pickling/marshalling, and I'd love to be
able to use it here.  But it can't work that way.

> You could parse the data, pickle the results, and keep up with the
> mod times on the two files.  If the source data is newer than the
> pickle database you have to update the pickle data.

The problem is that to "update" the pickle data (which I'd have to do
every time, as explained above), I have to parse the text files.  And
it's the parsing that's slow, not the post-processing.  If I need to
parse them every time, then I don't need pickling.



From I.Clarke at strs.co.uk  Thu Sep 16 09:35:58 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Thu, 16 Sep 1999 14:35:58 +0100
Subject: high school programming & python
References:    <937300939snz@vision25.demon.co.uk>
Message-ID: <37E0F23E.B0199331@strs.co.uk>

> Machine language! You young people today don't know what hardship
> is... when I were a lad, we had to manually patch the ALU using
> plug boards to change the logic.

Plug boards!  You young people today don't know what hardship is... when
I were a lad, we had to manually train rats the two (we had only
discovered two) logic gates so that they could operate the strings and
levers to perform even the most basic calculation.  I can recall waiting
three hours to find out what 13 squared was.  Worse - the smell was
unbelievable (so much so that the rats demanded that we programmers
washed once in a while).

Ian.



From henk_p at my-deja.com  Mon Sep 27 10:05:41 1999
From: henk_p at my-deja.com (henk_p at my-deja.com)
Date: Mon, 27 Sep 1999 14:05:41 GMT
Subject: How does a CGI script get complete request header text ?
References: <7sju0v$pc2$1@nnrp1.deja.com>
Message-ID: <7sntje$b32$1@nnrp1.deja.com>

In article <7sju0v$pc2$1 at nnrp1.deja.com>,
  rturpin at my-deja.com wrote:
> Is there a way for a CGI script to get the complete header text,
rather
> than just the preparsed contents that are provided in the environment
> variables?.

This is not possible. The apache mod_cgi module does not pass these
headers down to the script. It uses the original header to build the
environment for the script.

If you really need the headers (and the info you need cannot be
reconstructed from the env. vars, query string or POST data) then you
would need to write your own cgi like module for apache.

At the moment I am working on a module for apache called mod_psi.
(Python Script Interface). with this module requests for url with
extension .psi are routed to a pythons script listening on some port.
(This way the script is persistent over multiple requests, so I can keep
my db connections open).
At the moment this passes the normal cgi headers to the python script.
It would be very easy to also pass the original client request headers.

I you are interested, mail me at csg6031 at wing.rug.nl, I will mail you
the psi_mod source (C) and hello world example (python).


Henk.


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



From tratt at dcs.kcl.ac.uk  Wed Sep  1 05:16:19 1999
From: tratt at dcs.kcl.ac.uk (Laurence Tratt)
Date: Wed, 01 Sep 1999 10:16:19 +0100
Subject: CGI speedup strategies
References: <37CCE7C9.D340BBF5@strs.co.uk>
Message-ID: 

In message <37CCE7C9.D340BBF5 at strs.co.uk>
          Ian Clarke  wrote:

> Has anyone tried actually implementing a Python webserver which just
> starts a new thread and runs a piece of Python code in it every time
> someone makes a HTTP request?

When I had a need for such a thing, I messed around with the *HTTPServer
stuff in the standard distribution to get what I wanted; because of, er,
moving targets it ended up pretty messy but it's one way of doing things
quickly if you have esoteric needs. Making it run threaded was trivial,
although you still have to be aware of the standard threading problems
.


Laurie
-- 
http://eh.org/~laurie/comp/python/



From srenner at lycosmail.com  Sun Sep 12 15:31:52 1999
From: srenner at lycosmail.com (srenner at lycosmail.com)
Date: Sun, 12 Sep 1999 19:31:52 GMT
Subject: ANN: Tk Slashdot Headline Viewer
References: <7rgmd0$2htg$1@hub.org>
Message-ID: <7rgv38$8m8$1@nnrp1.deja.com>

It works! Although I'm not sure what it's for -- why not just read
Slashdot directly? But it definitely works. Also I am trying the folding
editor. Would you consider modifying IDLE's class browser into a folding
code editor? Because somebody should. Keep up the good work, sr


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From aprasad at magix.com.sg  Sun Sep 12 05:45:00 1999
From: aprasad at magix.com.sg (Ajith Prasad)
Date: Sun, 12 Sep 1999 17:45:00 +0800
Subject: Sun's StarOffice & Python
Message-ID: <7rgk53$dbm$1@coco.singnet.com.sg>

With Sun making the StarOffice application suite available for a free
download and making its source-code available, I presume that there would
now be a viable alternative to MS Office with a growing user base. Is anyone
working on Python connectivity to StarOffice? Thanks in advance.





From vnesek at maja.zesoi.fer.hr  Wed Sep  1 03:09:28 1999
From: vnesek at maja.zesoi.fer.hr (Vjekoslav NESEK)
Date: 1 Sep 1999 07:09:28 GMT
Subject: Pythom in embedded system?
Message-ID: 

Hi,

  I would like to ask all you python source code hackers:

1) Is it possible to remove all os dependant (just ansii c) code
   from distributions, leaving only core intepreter?

2) Is it possible to run it on 16bit machine

3) What is a minimum RAM requirements for such trimmed down Python?

4) How slow would it be on lets say 16Mhz 80x86?

  I am asking because I would like to use python for glue code in
embedded system. It is not a real time embedded system. Anyway if
not python, can you point me to some other free interpeter
suitable for embedding in such a limited system?

Thanks,

-- 
// Vjeko (vnesek at zesoi.fer.hr)

import java.advertising.*;
new URL("http://diana.zesoi.fer.hr/~vnesek/");



From pj at sgi.com  Wed Sep 29 17:12:14 1999
From: pj at sgi.com (Paul Jackson)
Date: 29 Sep 1999 21:12:14 GMT
Subject: More Python books coming...
References: <5lpuz483p4.fsf@eric.cnri.reston.va.us> <37F23C23.1DD36623@megsinet.net>
Message-ID: <7stvbe$bk3hq@fido.engr.sgi.com>

|> I found a free e-book ... I will e-mail this book ...

Perhaps you could briefly describe this e-book?  Say title,
author, date of writing and a few word summary?
-- 

=======================================================================
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj



From tim_one at email.msn.com  Wed Sep  8 20:57:46 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 8 Sep 1999 20:57:46 -0400
Subject: Python example: possible speedup?
In-Reply-To: <19990908230036.A3530@krikkit.vollmert.home>
Message-ID: <000001befa5e$5493e140$362d153f@tim>

[Hrvoje Niksic, writes a pretty text-crunching program in Python
 that's-- horrors! --slower than a Perl workalike]

[Robert Vollmert, among others, does various irksome things to
 speed it up, and remarks ...]
> The main speed advantage seems to be that the files aren't
> processed line by line, which is of course very memory consuming.

This probably bears repetition about once a month :  Python does
line-at-a-time input on top of the platform's stdio package.  On the popular
platforms, Perl does line-at-a-time input by breaking into the platform's
stdio package, peeking and poking C FILE structs directly in
platform-specific ways.  This can give Perl a huge speed advantage on
line-at-a-time input; indeed, on Windows, Perl does line-at-a-time input
faster than pure optimized C/C++ code using Microsoft's fgets, and about 3x
faster than Python under Win95.  Perl's speed advantage here is directly
proportional to the stupidity of the platform's stdio implementation ("read
a line from a file?  who on earth would want to do that?" ).

eschew-files-embrace-3d-buttons-ly y'rs  - tim





From nathan at islanddata.com  Mon Sep 13 12:13:49 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Mon, 13 Sep 1999 09:13:49 -0700 (PDT)
Subject: Queue not releasing memory
In-Reply-To: <000401befd03$3813bf20$492d153f@tim>
Message-ID: 

On 12-Sep-99 Tim Peters wrote:
> Python has no control over when your OS decides that released memory is
> available to something else.  If you name your OS, perhaps someone will
> know
> how to influence it reliably.

The results I described were under Linux, glibc 2.1.1.  The target
platform for this application, however, is Solaris 2.7, which is worse. 
Under Solaris, the memory is not reclaimed when the queue is deleted.  The
entire python process must terminated.  Not excited about this.  I just
did a simple malloc/free test with the exact same results.  Bad Solaris.

> By the way, look at that code carefully:  the default Queue
> implementation
> wasn't meant to handle memory-busting amounts of data!  That "del
> self.queue[0]" line makes a physical copy of the entire data remaining
> every
> time.

I do need the thread-safety of the Queue class.  The large size
requirement is only an extreme situation, which is why it is important
that it backs down afterward.  Perhaps I need to take some pressure off
Queue and move more quickly to my other lines of defense.

Is there no way to remove an element from a list without making a copy? 
append() doesn't need a copy.  I was under the impression that lists were
internally represented by a more dynamic form.

Thanks for the information.  I will reevaluate my plan of attack.

----------------------------------
Nathan Clegg
 nathan at islanddata.com





From etsang at my-deja.com  Tue Sep 28 14:29:28 1999
From: etsang at my-deja.com (etsang at my-deja.com)
Date: Tue, 28 Sep 1999 18:29:28 GMT
Subject: ExpectPy Installation problem --- problem restated.
Message-ID: <7sr1dn$iou$1@nnrp1.deja.com>

I used this during installation of ExpectPy:
./configure --with-python=/s1/sqa/edwardt/Python/Python-1.5.2--with-
expect=/tools/2.5.1/local/lib

But it says cannot compile with the python library.....
Any heads up? Is there something I need to download before this?? I
have got the
python from www.python.org. The final release one...
But for some reason, the installation program is  still trying to look
for the network installed Python!!!!

Attached below is the config.log:
Please help experts!!!
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.

configure:568: checking host system type
configure:659: checking for gcc
configure:736: checking whether the C compiler (gcc  ) works
configure:750: gcc -o conftest    conftest.c  1>&5
configure:770: checking whether the C compiler (gcc  ) is a cross-
compiler
configure:775: checking whether we are using GNU C
configure:784: gcc -E conftest.c
configure:799: checking whether gcc accepts -g
configure:828: checking whether gcc and cc understand -c and -o together
configure:843: gcc -c conftest.c -o conftest.o 1>&5
configure:844: gcc -c conftest.c -o conftest.o 1>&5
configure:849: cc -c conftest.c 1>&5
configure:851: cc -c conftest.c -o conftest.o 1>&5
configure:852: cc -c conftest.c -o conftest.o 1>&5
configure:879: checking how to run the C preprocessor
configure:900: gcc -E  conftest.c >/dev/null 2>conftest.out
configure:940: checking for AIX
configure:974: checking for a BSD compatible install
configure:1026: checking for ranlib
configure:1053: checking whether ln -s works
configure:1076: checking for rm
configure:1116: checking for python
configure:1155: checking for expect
configure:1202: checking SO
configure:1224: checking LDSHARED
configure:1256: checking LDEXE
configure:1279: checking CCSHARED
configure:1341: checking Python version
configure:1370: checking Expect version
configure:1390: checking Expect library location
configure:1435: checking Tcl version (for testing)
configure:1473: checking use of Python Setup
configure:1548: checking for openpty
configure:1576: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c  1>&5
/var/tmp/ccafaaHJ1.o: In function `main':
/s1/sqa/edwardt/Python/ExpectPy-1.8/configure:1570: undefined reference
to
`openpty'
configure: failed program was:
#line 1553 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
    which can conflict with char openpty(); below.  */
#include 
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char openpty();

int main() {

/* The GNU C library defines this for functions which it implements
    to always fail with ENOSYS.  Some functions are actually named
    something starting with __ and the normal name is an alias.  */
#if defined (__stub_openpty) || defined (__stub___openpty)
choke me
#else
openpty();
#endif

; return 0; }
configure:1598: checking for openpty in -lutil
configure:1617: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c
-lutil   1>&5
/tools/2.5.1/local/sparc-sun-solaris2.5.1/bin/ld: cannot open -lutil:
No such
file or directory
configure: failed program was:
#line 1606 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char openpty();

int main() {
openpty()
; return 0; }
configure:1646: checking for cos in -lm
configure:1665: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c -lm
1>&5
configure:1658: warning: conflicting types for built-in function `cos'
configure:1693: checking for dlopen in -ldl
configure:1712: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c -ldl
-lm  1>&5
configure:1741: checking for pthread_create in -lpthread
configure:1760: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c
-lpthread  -ldl -lm  1>&5
configure:1877: checking for exp_expectv in -lexpect5.26
configure:1896: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c
-lexpect5.26  -ldl -lm  1>&5
/tools/2.5.1/local/lib/libexpect5.26.a(exp_clib.o): In function
`exp_spawnv':
/net/wrk/expect-5.26/exp_clib.c:282: undefined reference to
`Tcl_ErrnoMsg'
/net/wrk/expect-5.26/exp_clib.c:296: undefined reference to
`Tcl_ErrnoMsg'
/net/wrk/expect-5.26/exp_clib.c:369: undefined reference to
`Tcl_ErrnoMsg'
/net/wrk/expect-5.26/exp_clib.c:497: undefined reference to
`Tcl_ErrnoMsg'
/net/wrk/expect-5.26/exp_clib.c:509: undefined reference to
`Tcl_ErrnoMsg'
/tools/2.5.1/local/lib/libexpect5.26.a(exp_clib.o):/net/wrk/expect-
5.26/exp_clib
.c:532: more undefined references to `Tcl_ErrnoMsg' follow
/tools/2.5.1/local/lib/libexpect5.26.a(exp_clib.o): In function
`expectv':
/net/wrk/expect-5.26/exp_clib.c:771: undefined reference to
`TclRegError'
/net/wrk/expect-5.26/exp_clib.c:772: undefined reference to `TclRegComp'
/net/wrk/expect-5.26/exp_clib.c:773: undefined reference to
`TclGetRegError'
/net/wrk/expect-5.26/exp_clib.c:947: undefined reference to
`TclRegError'
/net/wrk/expect-5.26/exp_clib.c:948: undefined reference to `TclRegExec'
/net/wrk/expect-5.26/exp_clib.c:952: undefined reference to
`TclGetRegError'
/net/wrk/expect-5.26/exp_clib.c:953: undefined reference to
`TclGetRegError'
/tools/2.5.1/local/lib/libexpect5.26.a(pty_termios.o): In function
`exp_getptyslave':
/net/wrk/expect-5.26/pty_termios.c:623: undefined reference to
`Tcl_ErrnoMsg'
/net/wrk/expect-5.26/pty_termios.c:626: undefined reference to
`Tcl_ErrnoMsg'
/net/wrk/expect-5.26/pty_termios.c:629: undefined reference to
`Tcl_ErrnoMsg'
configure: failed program was:
#line 1885 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char exp_expectv();

int main() {
exp_expectv()
; return 0; }
configure:1920: checking for exp_expectv in -lexpect5.26
configure:1939: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c
-lexpect5.26 -ltcl -ldl -lm  1>&5
configure:2024: checking for PyObject_GetAttr in -lpython1.5
configure:2043: gcc -o conftest -g -O2
-I/s1/sqa/edwardt/Python/Python-1.5.2/Include -I/tools/2.5.1/local/lib
-L/s1/sqa/edwardt/Python/Python-1.5.2 -L/tools/2.5.1/local/lib
conftest.c
-lpython1.5  -ldl -lm  1>&5
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`setipaddr':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:404:
undefined
reference to `gethostbyname'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_accept':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:612:
undefined
reference to `accept'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_setsockopt':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:719:
undefined
reference to `setsockopt'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_getsockopt':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:759:
undefined
reference to `getsockopt'
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:773:
undefined
reference to `getsockopt'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_bind':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:803:
undefined
reference to `bind'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_connect':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:854:
undefined
reference to `connect'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_connect_ex':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:880:
undefined
reference to `connect'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_getsockname':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:953:
undefined
reference to `getsockname'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_getpeername':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:980:
undefined
reference to `getpeername'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_listen':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1008:
undefined
reference to `listen'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_recv':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1083:
undefined
reference to `recv'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_recvfrom':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1122:
undefined
reference to `recvfrom'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_send':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1167:
undefined
reference to `send'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_sendto':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1199:
undefined
reference to `sendto'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocketSock_shutdown':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1223:
undefined
reference to `shutdown'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocket_gethostbyname_ex':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1483:
undefined
reference to `gethostbyname'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocket_gethostbyaddr':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1551:
undefined
reference to `gethostbyaddr'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocket_getservbyname':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1583:
undefined
reference to `getservbyname'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocket_getprotobyname':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1617:
undefined
reference to `getprotobyname'
/s1/sqa/edwardt/Python/Python-1.5.2/libpython1.5.a(socketmodule.o): In
function
`PySocket_socket':
/s1/sqa/poonam/python/Python-1.5.2/Modules/./socketmodule.c:1651:
undefined
reference to `socket'
configure: failed program was:
#line 2032 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error.  */
/* We use char because int might match the return type of a gcc2
    builtin and then its argument prototype would still apply.  */
char PyObject_GetAttr();

int main() {
PyObject_GetAttr()
; return 0; }
configure:2066: checking for PyObject_GetAttr in -lpython1.5


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



From pinard at iro.umontreal.ca  Sat Sep 18 09:04:30 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 18 Sep 1999 09:04:30 -0400
Subject: Reaching subtrees
In-Reply-To: "Darrell"'s message of "Fri, 17 Sep 1999 22:13:40 -0400"
References:  <012e01bf017b$6ce64a20$0100a8c0@rochester.rr.com>
Message-ID: 

"Darrell"  ?crit:

> Not understanding why you like this [recursive tuple unpacking] so much.
> But happy that you do.

Thanks.  I like it because I have code that explore subtrees according to
literally hundreds of structural patterns (if not thousands when you put all
the programs together), and I even wrote a language to help me at this.  So,
it is a common problem for me, and I like expressing constraints fluently.

If Python allows me to easily name subtrees while picturing the skeleton
model, it means that I could use unadorned Python to write unusually legible
code for such things without having to recourse to special languages.
A Python application would have to be small, as implementing my language
revealed me how necessary it is to precondition how huge rule sets are
handled -- an important part of my effort was devoted to optimisation.

However, having contemplated Python in some more detail, there is still a few
important cases of structural patterns that my language implements, and that
do not seem to be easily representable with mere recursive tuple unpacking.
But I would guess that for many simple cases, Python would just suffice.

By the way, and quite unrelated to my own tool language, if recursive
tuple unpacking was accepting an immutable (instead of a variable) in
the skeleton as forcing an exact match instead of an assignment, and if
the same variable could be repeated in the skeleton and then, force an
identity check instead of an assignment, maybe that would make Python not
very far from Prolog unification :-).  Just a random thought, of course.

> Yes, this looks like standard python stuff.

I'll try to find a precise statement of this in Guido's reference manual.

> As you know if the tuples don't unpack correctly they throw an exception.
> Exceptions are slow so doing this on a large scale might be a problem.

I guess they might be slow because they have to somewhat scan the stack
back to find an exception catcher, but otherwise, if exceptions are caught
quite early on the stack, do they still have to be slow?  That would be
a little unfortunate, at least for my case.  I thought that `try:' could
become an important part of Python writing style.  It should rather not?

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From pinard at iro.umontreal.ca  Mon Sep 13 23:01:35 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 13 Sep 1999 23:01:35 -0400
Subject: French pun on Python! :-)
Message-ID: 

Hi, people.

While introducing Python to a few friends, I noticed they tend to naturally
use `je Pythonne', `nous Pythonnons', etc. as an easy pun on the word Python.

I think that English people pronounce `pie-ton', while French people say
`pea-to[n]' (said faster, with the `n' silent).  This has the same sounds as
for the word `piton'.  A `piton' is a familiar word to design any mechanical
device on which someone pushes, like an electrical button, or often, one
of the keys of a computer keyboard.  So, `je pitonne', `nous pitonnons',
etc. already became a familiar way to express any soft computer hacking.

I thought this might amuse some of you.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From Doug at news.vex.net  Mon Sep 13 01:19:05 1999
From: Doug at news.vex.net (Lewis)
Date: 13 Sep 1999 05:19:05 GMT
Subject: ANN: Tk Slashdot Headline Viewer
References: <7rgmd0$2htg$1@hub.org> <7rgv38$8m8$1@nnrp1.deja.com>
Message-ID: <7ri1g9$2k5p$1@hub.org>

On Sun, 12 Sep 1999 19:31:52 GMT, srenner at lycosmail.com wrote:

>It works! Although I'm not sure what it's for -- why not just read
>Slashdot directly?

Admittedly it's probably of little practical value. It was really a project for
me to play with while learning Tkinter. Although on the other hand, it can be
nice to have a little window open (which is not a browser window) which
monitors the slashdot headlines and regularly checks for updates... just in
case soemthing interesting comes along. Also my little monitor is much more
concise in its listing than the web page which you have to actually scroll down
(gasp!) to see all the stories. Heh.

>Would you consider modifying IDLE's class browser into a folding
>code editor? 

Don't ask me! I still can't figure out how the heck to even change 
the normal text background in IDLE! Hehe. Which is why i never use it... 
it blinds me with its black and whiteness... alas.

I have heard newer versions have a config file now, which may or may
not have such settings... however it's only available on the CSV... a
trick i haven't learned to access yet.

I'm just a small time hobby hacker who doesn't really have any sort
of clue what he's doing most of the time after all. (-:





From akuchlin at mems-exchange.org  Wed Sep  8 15:00:24 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 08 Sep 1999 15:00:24 -0400
Subject: Programmers Editor
References: <37D5916B.606E@zeusedit.com>  <00da01bef9e0$6e1b5c70$f29b12c2@secret.pythonware.com>
Message-ID: <3dg10p1b5j.fsf@amarok.cnri.reston.va.us>

"Fredrik Lundh"  writes:
> but to help them, it's probably better to put up
> a site somewhere comparing different tools (or
> just pointing to them), than to follow up on every
> announcement with a plug for your own favourite
> tool.  with dozens of development tools out there,
> it doesn't really scale if everyone would do that ;-)

Andres Corrada-Emmanuel has written an editor HOWTO that hasn't been
put on the Web yet; currently it covers configuring Emacs, Alpha and
CodeWright for convenient editing of Python code.  (Hmm... I vaguely
recall sections on jed and vi; maybe my copy at work is out of date --
I'll have to look on my home machine.) Anyway, if someone wanted to
contribute info on editing Python with Zeus, KDevelop, or whatever, it
would be very helpful.  Just follow up to this posting with notes on
configuring your editor of choice.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
His first victim was a priest. Roger's senses were blessed with furtive
longings and sweet altar wine.
    -- The Head's backstory, in ENIGMA #1: "The Lizard, The Head, The Enigma"





From fredrik at pythonware.com  Thu Sep  9 09:04:29 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 9 Sep 1999 13:04:29 GMT
Subject: Inspecting long running programs
References: <19990909111034.12856@nms.otc.telstra.com.au> <7r7356$3b4@dfw-ixnews8.ix.netcom.com> <5lr9k84777.fsf@eric.cnri.reston.va.us>
Message-ID: <00c101befac3$dac08790$f29b12c2@secret.pythonware.com>

Guido van Rossum  wrote:
> > asyncore (which Medusa is built on) includes this functionality.  It can
> > be kind of trippy to just telnet to a blind socket and get a Python
> > interpreter session.
> 
> Really?  I don't see a trace of this in the version of asyncore I have
> in the standard Python library!  I'd love to have this feature!

it's a medusa thingie.  look for monitor.py and
monitor_client.py.






From g.gruber at xsoft.co.at  Thu Sep  2 05:40:54 1999
From: g.gruber at xsoft.co.at (Gerhard W. Gruber)
Date: Thu, 02 Sep 1999 11:40:54 +0200
Subject: adding to dictionaries
Message-ID: <37CE4626.8FCEA15B@xsoft.co.at>

Is it possible to add an entry to an already existing dictioniary?
-- 
bye,
   Gerhard
email: sparhawk at eunet.at
       g.gruber at xsoft.co.at



From tre17 at student.canterbury.ac.nz  Tue Sep 21 20:46:35 1999
From: tre17 at student.canterbury.ac.nz (tre17 at student.canterbury.ac.nz)
Date: 22 Sep 1999 12:46:35 +1200
Subject: Print/Input statements
References: <000601befe4c$3e585ae0$54a2143f@tim> <87u2oxeass.fsf@pc-hrvoje.srce.hr> <87n1ug5t0y.fsf_-_@compsoc.dur.ac.uk> <000b01bf043b$39f2fab0$f29b12c2@secret.pythonware.com> <19990921151648.B5974@compsoc.dur.ac.uk>
Message-ID: 

Andrew Stribblehill  writes:

[slightly misunderstood version snipped]
> 
> OK, a more obvious example (sorry!):
> 
> #!/usr/bin/python
> import sys
> print "I didn't expect a Spanish Inquisition\n>",
> x=sys.stdin.readline()
> print "Nobody expects a Spanish Inquisition!"
> 
> The "Nobody" has a space in front of it.
> 
> When I replace the print statement with:
>   sys.stdout.write("doodah\n>")
> the Nobody loses its annoying space.
> 
> Cheerio,
> 
> Andrew.

Well, you could always use raw_input to print your prompt, instead of
print:

#!/usr/bin/python
import sys
print "I didn't expect a Spanish Inquisition"
x = raw_input(">")
print "Nobody expects a Spanish Inquisition!"

If you also add `import readline' to the script (if readline is
available) then you get command editing and history.

--
Tim Evans



From ivanlan at callware.com  Wed Sep  8 11:20:20 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 8 Sep 1999 15:20:20 GMT
Subject: Feast or famine! - Anyone read or dealt with Wrox Press?
References: 
		<37d60e03.1299428@news.iol.ie> <9VtB3.14$114.1985@news.clarkson.edu>
Message-ID: <37D67EB4.22612A9E@callware.com>

Hi All--

Brad Clements wrote:
> 
> Not quite related but the Wrox folks seem interested in a Python book. Has
> anyone actually read a Wrox press book? Whats your opinion? Has anyone dealt
> with them as a publisher?
> 

Their books present a nice appearance; usually, they cover their areas
in exhaustive detail.  However, I do suspect them of rushing their
authors, as I've found many errors in example code.  I've got several
Wrox books, and the code in none of them is perfect.  I hesitate to say
that the code is not tested, as it may just be version creep.

-ly y'rs,
Ivan;-)
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From laheadle at boguscs.uchicago.edu  Tue Sep 14 04:20:43 1999
From: laheadle at boguscs.uchicago.edu (Lyn A Headley)
Date: Tue, 14 Sep 1999 08:20:43 GMT
Subject: embedded Python server - desperate for help
References: <37DD589D.5B0347E6@ncsa.uiuc.edu>
Message-ID: 

    Randy>   Py_Initialize(); // import some libs // wait for incoming
    Randy> strings to execute...

    Randy> then client #1 might do:
    server-> execute("s1 = Square(4)");

    Randy> what if client #2 also creates an object 's1':
    server-> execute("s1 = Square(6)");

    Randy> Aren't the object names in some global namespace and
    Randy> therefore problematic?

Well yes, that would be problematic if you want each client to have
its own namespace.  Are you sure you want clients to run arbitrary
code on the server side?

I'd suggest you define a client/server protocol, based on something
like xml-rpc or corba with fnorb.

-- 
Lyn Headley
remove the word "bogus" from my address for the real one.



From aahz at netcom.com  Wed Sep  1 17:35:43 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Sep 1999 21:35:43 GMT
Subject: Returning none
References: <000001bef265$fe64f340$192d153f@tim> <37CC79C5.1BFDC710@compaq.com> <7qidi6$nbb$1@towncrier.cc.monash.edu.au>
Message-ID: <7qk67f$9oc@dfw-ixnews15.ix.netcom.com>

In article <7qidi6$nbb$1 at towncrier.cc.monash.edu.au>,
Jonathan Giddy  wrote:
>
>This requires that we always use the result of a call to a real function,
>even if we just assign it to a dummy variable.  And that means someone
>will one day spend time trying to find where the dummy variable is used,
>instead of instantly being aware that the function is only called for its
>side effects.
>
>For example, FileObject.close needs to be a function to return the exit
>status of a popen process.  Hence, every file close will need to be
>assigned (or used in some way).

Your example does not illustrate your point IMO, because there *is* a
potentially useful value returned.  If a function returns a value that
is truly never used, there is Something Wrong.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From daniels at mindspring.com  Sun Sep 12 22:55:02 1999
From: daniels at mindspring.com (Alan Daniels)
Date: 13 Sep 1999 02:55:02 GMT
Subject: Programmers Editor
References: <37D5916B.606E@zeusedit.com>  <00da01bef9e0$6e1b5c70$f29b12c2@secret.pythonware.com> <3dg10p1b5j.fsf@amarok.cnri.reston.va.us> 
Message-ID: 

On 9 Sep 1999 18:42:51 GMT, the infinitely wise Bernhard Reiter
(bernhard at alpha1.csd.uwm.edu) spoke forth to us, saying...

>Someone should elaborate on the tab use in vim a bit.

I set up the following in my .vimrc file:
set list
set listchars=tab:#^
set noexpandtab

"list" says to show otherwise non-viewable characters, such as tabs,
newlines, or trailing spaces. "listchars" sets what those should show
up as -- in my case, I'm just interested in seeing all tabs as
something like "#^^^^^". The "noexpandtab" tells VIM to insert spaces
rather than an actual tab character.

The overall effect is that tab characters never make it into my source
code. Tabs are bad, bad, bad. And, with "list" turned on, I can
immediately see if any snuck in there, or (horrors), I have the source
code of others that has tabs in them. A simple...

:%s/\t/    /gc

...turns all of your tabs into spaces.

Of course, this means I usually trip up when handling Makefiles
(which, if you'll remember, really DO need actual tab chars), but the
benefit is worth the occasional headache. :=)

Also, if you're brave enough, you can compile VIM with a
--python-enabled option under "./configure", which will let you run
python code without leaving the editor (Hey! Almost like Emacs!). This
may sound complicated, but if you're using VIM to begin with you
obviously have a pathological lack of any fear of complexity to begin
with. :=) The interface will let you type stuff like...

:py print "Hello world!"

...and Python is fired up for you. The downside is, it's only good for
one-liners, so you can't do anything too complicated. I can only
surmise that this feature was added in the same spirit as VIM's ":pl"
command for Perl. :=)

-- 
=======================
Alan Daniels
daniels at mindspring.com
daniels at cc.gatech.edu



From englim at pc.jaring.my  Wed Sep  8 12:20:03 1999
From: englim at pc.jaring.my (lec)
Date: Thu, 09 Sep 1999 00:20:03 +0800
Subject: Bug in time module?
References: <1275332084-98066501@hypernet.com>
Message-ID: <37D68CB2.BB5F9CC9@pc.jaring.my>

Gordon McMillan wrote:

>
> The only editing Python's time module does is to ensure that you've
> passed the right number and kind of args. Everything else is the
> result of your platform's implementation of strptime. If you want
> something _good_, you should probably use MA Lemburg's mxDateTime.
>
> - Gordon

Consider the following exception raised:

Python 1.5.2 (#1, Apr 18 1999, 16:03:16)  [GCC pgcc-2.91.60 19981201
(egcs-1.1.1  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import time
>>> time.strptime('31/13/1999', "%d/%m/%Y")
Traceback (innermost last):
  File "", line 1, in ?
ValueError: format mismatch

Shouldn't the same exception be raised for time.strptime('31/02/1999',
"%d/%m/%Y")  ?




From mlh at idt.ntnu.no  Tue Sep 21 19:02:38 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 22 Sep 1999 01:02:38 +0200
Subject: cgi authentication
References: 
Message-ID: 

"Dan Grassi"  writes:

> I need to do authentication from Python, for various reasons using Apache 
> and .htaccess is not a workable solution.  I can get the request to show up
> with the following code but I can not figure out how to access the returned
> name/password pair.  Yes, I have looked hard for the info on the web and in
> the books. "-)

Well - I have had the same problem, and AFAIK, Apache refuses to give
you the password (ostensibly for security reasons...) There is a
patched version available somewhere...

> 
> Can anyone help?
> 

--

  Magnus
  Lie                                     Please forward all spam to
  Hetland                                                uce at ftc.org



From victor at idaccr.org  Wed Sep 22 16:51:24 1999
From: victor at idaccr.org (Victor S. Miller)
Date: 22 Sep 1999 16:51:24 -0400
Subject: PIL: decode jpeg not available
Message-ID: 

I have Python 1.5.2 build on Solaris 2.6 with gcc 2.7.2.2.  I also
have tcl8.1 and tk8.1, IJG jpeg-6b and zlib-1.1.3.  I built PIL 1.0
with all these, and uncommented and built Python with the WITH_PIL
stuff for tkinter.  However, when I try to use the viewer.py script
that came with PIL with a jpeg file, I get the following error:

Traceback (innermost last):
  File "viewer.py", line 45, in ?
    UI(root, im).pack()
  File "viewer.py", line 24, in __init__
    self.image = ImageTk.PhotoImage(im)
  File "/usr/local/coop/lib/python1.5/site-packages/PIL/ImageTk.py", line 85, in __init__
    self.paste(image)
  File "/usr/local/coop/lib/python1.5/site-packages/PIL/ImageTk.py", line 107, in paste
    im.load()
  File "/usr/local/coop/lib/python1.5/site-packages/PIL/ImageFile.py", line 133, in load
    d = Image._getdecoder(self.mode, d, a, self.decoderconfig)
  File "/usr/local/coop/lib/python1.5/site-packages/PIL/Image.py", line 229, in _getdecoder
    raise IOError, "decoder %s not available" % decoder_name
IOError: decoder jpeg not available

Is there something that I've forgotten?  Any suggestions as to how to
track down the problem?
-- 
Victor S. Miller     | " ... Meanwhile, those of us who can compute can hardly
victor at idaccr.org    | be expected to keep writing papers saying 'I can do the
CCR, Princeton, NJ   | following useless calculation in 2 seconds', and indeed
    08540 USA        | what editor would publish them?"  -- Oliver Atkin



From dnlbfx at aol.com  Thu Sep 23 22:28:04 1999
From: dnlbfx at aol.com (dnlbfx at aol.com)
Date: Fri, 24 Sep 1999 02:28:04 GMT
Subject: This Is The Best The Internet Has To Offer  9098
Message-ID: 

Hi,
My name is Ellie and I just stumbled across what has to be the BEST source on the internet!  Its totally free and easy to have the most fun ever!

http://www.amylinks.com/

Visit Amys Links, its totally worth it!
oibrpmeqtxifgghnvbkfcwmgdffikymfbsjhwcxvodrbykyktus




From os at webinspace.de  Thu Sep 30 02:05:17 1999
From: os at webinspace.de (Oliver Scheel)
Date: Thu, 30 Sep 1999 08:05:17 +0200
Subject: More Python books coming...
References: <5lpuz483p4.fsf@eric.cnri.reston.va.us> <37F0855B.461C6DC3@webinspace.de> <5lk8pb8ex5.fsf@eric.cnri.reston.va.us>
Message-ID: <37F2FD9D.EF7D475C@webinspace.de>

Guido van Rossum wrote:
> 
> If you don't mind the different cover art, David Beazley's "Python;
> Essential Reference" is pretty much what you'd expect from a nutshell
> book; packed with no-nonse reference information about language and
> library.  There's even an appending with C API info.

Visited Amazon to get a description. Sounds very interesting! It seems
that it is the book that I was looking for ;-) Thanks for that hint.

Oliver



From dave at see.sig.for.address  Sun Sep  5 17:03:29 1999
From: dave at see.sig.for.address (Dave)
Date: Sun, 05 Sep 1999 21:03:29 GMT
Subject: jpython and java apps
References: <7qsh7i$5nu$1@msunews.cl.msu.edu>
Message-ID: <37d4c8a5.17734643@read.news.global.net.uk>

On Sat, 4 Sep 1999 21:35:36 -0400, "Richard Hensh"
 wrote:

>Can one "compile" a Python script into a Java class (?) using JPython (a la
>NetRexx)?
>
>thanks
>ricky
>

Yes, though it is not very well documented. 

Essentially you have to do the following:

1) write a python class and have it inherit from a Java class or
interface. The python class must have the same name as the file it is
in, just like Java classes. 

2) add a doc string to each python method that specifies the Java
signature, starting with "@sig". This bit is not documented, but is
done in the example that comes with jpython. 

3) compile the python class using jpythonc. This will create the
.class files that can be used just like any other Java class. There
are numerous options to jpythonc for things like creating jar files
and packages, which are documented to some extent.

Make sure you have the latest version of jpython (1.1a2), especially
if you want to use Java packages.  I would also suggest joining the
jpython mailing list, which is where all the real experts hang out.

	Dave K
--------------------------------------------------
All great ideas start as heresy and end as dogma.

dkirby@   <-figure this out, spambots!
bigfoot.   My opinions are my own,
com        but I'm willing to share.



From nathan at islanddata.com  Tue Sep 28 12:07:20 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Tue, 28 Sep 1999 09:07:20 -0700 (PDT)
Subject: keys and values lining up
In-Reply-To: <000601bf094d$43e55ce0$18a0143f@tim>
Message-ID: 

d.items() is my backup plan, of course, but I'd rather not.  My purpose is
to insert into a database based on the contents of a dictionary.  SQL
requires a list of keys and a separate list of values:

INSERT INTO table (key1, key2, key3) VALUES (val1, val2, val3)

If I were to use d.items(), I would them have to go back and use the list
of two-tuples to generate a two-tuple of lists :)  No point in going
through the back door if you can work it right.  If I'm guaranteed to get
keys and values in the same order from the two function calls, that's far
easier for my purposes.



On 28-Sep-99 Tim Peters wrote:
> [Nathan Clegg]
>> If d is a dict, am I guaranteed that d.keys() and d.values() will
>> line up, in proper order, if called successively with no modification
>> in between?  That is, is the following safe?
>>
>> k,v = d.keys(), d.values()
> 
> Yes, under the mild assumption that you don't have another thread
> running
> that manages to sneak in and mutate d between the calls.  If that's a
> potential problem, do
> 
>     kv = d.items()  # list of (key, value) tuples
> 
> If you want both keys and values, it's clearer to use the method
> designed
> for that purpose anyway.
> 
> something-fishy-about-keys-that-don't-unlock-anything-ly y'rs  - tim
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



----------------------------------
Nathan Clegg
 nathan at islanddata.com





From holger at phoenix-edv.netzservice.de  Wed Sep 15 09:29:42 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Wed, 15 Sep 1999 13:29:42 GMT
Subject: timesync with python?
References: <37D66AF3.6A3E96F7@phoenix-edv.netzservice.de> <002701befabb$1a884880$f29b12c2@secret.pythonware.com> 
Message-ID: <37DF9F46.85E7A332@phoenix-edv.netzservice.de>

Thank you all for answers,

I'll try it this week...

greetings,
Holger
(a little bit in a hurry)



From gerrit.holl at pobox.com  Wed Sep  8 15:01:14 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 8 Sep 1999 21:01:14 +0200
Subject: string.zfill()
Message-ID: <19990908210113.A3754@optiplex.palga.uucp>

Hello,

Why doesn't zfill takes as third argument the character (or string) to fill up
the string with, instead of '0's?

Has it a reason, or is it a bug/mistake?

regards,
Gerrit.
-- 
Unix is easy: type 'man man' and the rest comes of itself.



From wware-nospam at world.std.com  Wed Sep 15 12:45:10 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Wed, 15 Sep 1999 16:45:10 GMT
Subject: high school programming & python
References:  <37DF9080.21A8D285@calibre-dd.com>
Message-ID: 

Stephen J King (king at calibre-dd.com) wrote:
: Luxury! When we was kids all we 'ad was old wires which we 'ad to strip
: with our teeth and wrap around dirty binding posts with our bare
: fingers.

Least your school could afford electrons. All we had was neutrons.
They were slow and heavy, and you couldn't just impose an ambient
electric field and expect them to move. Back then, computing took
more skill and finesse.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From paul at prescod.net  Sat Sep  4 18:40:32 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 4 Sep 1999 22:40:32 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <1275835455-67788915@hypernet.com>
		<37CE9F55.88DFC9FC@megsinet.net>
		<37CEC4A9.C2CE2A91@prescod.net> 
Message-ID: <37D19FE0.C87DA906@prescod.net>

Blake Winton wrote:
> 
> Which points out, IMHO, Java's biggest problem.
> There is no distinction between Java *the language*, Java *the VM*, Java
> *the OS*, etc...  

This is a problem from a purist point of view but from a political point
of view it is brilliant. Sun owns all of those things and it is much
harder for anyone else to take control as long as they are bound
together. What if Microsoft "bought" JPython and started promoting it?
"That isn't Java" says Scott McNealy. What about when Microsoft makes a
really great native compiler for Java. "That isn't Java either."

 Paul Prescod



From watterss at pilot.msu.edu  Sun Sep 12 23:51:02 1999
From: watterss at pilot.msu.edu (Shannon Watters)
Date: Sun, 12 Sep 1999 23:51:02 -0400
Subject: [NEWBIE]:  Tk icon
Message-ID: <7rhsk2$gbi$1@msunews.cl.msu.edu>

How can I get rid of the Tk logo in the top left corner of a Tk(inter)
window?  It's 'tacky'.  I don't use that word much but it's the best one
that came to mind when I was staring at it up there.  It's making my highly
complex "Hello world" button look unprofessional (<-spelled?).

Shannon





From Cobra__77 at hotmail.com  Fri Sep 17 20:29:50 1999
From: Cobra__77 at hotmail.com (coby)
Date: Sat, 18 Sep 1999 10:29:50 +1000
Subject: ftp files based on start and finish byte
Message-ID: <37E2DCFE.6FE5BE87@hotmail.com>

Here is a little problem...

I want to transfer a 15MB file from one ftp sever to another using the
passive mode...but I can't transfer 15MB at once...I can only transfer
5MB at a time...so I need a little script or ftp client (???) that will
allow me to transfer files based on start and finish bytes..ie. I want
to specify when to start and finish transfer...so basically I will get 3

five mega byte files..which I can merge together with a merge utility
and end up with the 15MB file...

any suggestions on how to do this....if you know any scripting or know
any ftp client that does the above then please let me  know...

thanks






From aahz at netcom.com  Wed Sep  8 19:24:49 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Sep 1999 23:24:49 GMT
Subject: Looking for someone to program Python with
References: <37D6EAF4.E3EEC55A@hotmail.com>
Message-ID: <7r6r81$87p@dfw-ixnews5.ix.netcom.com>

In article <37D6EAF4.E3EEC55A at hotmail.com>,
Matt Lauer   wrote:
>
>I am new in programming python and i am just looking for someone that
>could help me with programming and write a prog. if possible.

Try the Tutor mailing list, http://www.python.org/mailman/listinfo/tutor
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From skip at mojam.com  Mon Sep 27 14:12:24 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 27 Sep 1999 13:12:24 -0500 (CDT)
Subject: Circular refs
In-Reply-To: <000701bf0909$46bcb0b0$3acbd9c2@peridot.optichrome.com>
References: 
	<000701bf0909$46bcb0b0$3acbd9c2@peridot.optichrome.com>
Message-ID: <14319.45960.133595.441653@dolphin.mojam.com>

    Adrian> I can dream up a million ways to kludge around this bug, but I
    Adrian> was referring to a 'fix' to the Python source code. How hard can
    Adrian> it be?

You can take a look at Neil Schemenauer's recent garbage collection patch.
It's only used to reclaim circular garbage.

    http://www.acs.ucalgary.ca/~nascheme/python/gc.html

There are other GC patches for Python floating around.  This just happens to
be the most recent one I've seen posted.  

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098   | Python: Programming the way Guido indented...



From pinard at iro.umontreal.ca  Thu Sep  2 10:23:07 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: Thu, 2 Sep 1999 14:23:07 GMT
Subject: Python speed and `pcre'
In-Reply-To: "Fredrik Lundh"'s message of "Thu, 2 Sep 1999 15:54:00 +0200"
References:    <00af01bef543$0affe8f0$f29b12c2@secret.pythonware.com>  <002101bef54a$9c7682b0$f29b12c2@secret.pythonware.com>
Message-ID: 

"Fredrik Lundh"  ?crit:

> Fran?ois Pinard  wrote:
> > Yes, of course.  Given strings are immutable, an implementation could have:
> >
> >         ref=str
> >         str=str+str1
> >
> > and have `ref' points to a substring of `str', without copy most of
> > the time, when `str1' is small.

> what do you do when you have to reallocate str?

Then, you have copies.  The substring is still referenced.  How to
implement that?  I do not know and would have to study the matter to find
possible answers, but it really is an implementation issue, that should
not bother users.

The key point of the whole argument is that people should program in Python
in the spirit of its simplicity, while the implementor should worry about
how to make the implementation speedy.  It is not healthy that users need
to know how things are implemented, then alter their programming style.
It does not fit well in what I understand from Python, as a language.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From wware-nospam at world.std.com  Fri Sep 24 12:23:46 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Fri, 24 Sep 1999 16:23:46 GMT
Subject: Read stdout from shell command, was: Becoming root
References:   <19990924165114.A753@krikkit.vollmert.home>
Message-ID: 

Robert Vollmert (rvollmert at gmx.net) wrote>
> On Fri, Sep 24, 1999 at 04:14:15AM +0000, Will Ware wrote:
> > if os.popen('whoami').readline()[:-1] == 'root':
> > Dumb question for those wiser than myself:
> > The os.popen().readline()[:-1] construction is pretty ugly.
> > Is there a better way to do it?
> How about 'if os.getuid() == 0:' ?

Thanks, a few others have mentioned this and slight variations.
The thing I'd really like to know is how to do the Python
equivalent of a shell command of this form:

variable = `some shell command`

where "some shell command" prints to standard output.
I find I do this quite frequently, usually using something like
os.popen('some shell command').readline()[:-1] and sure, it works,
but it's really ugly, and I'm always left with the nagging suspicion
that since I didn't explicitly close the os.popen(), it might still
be floating around.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From aahz at netcom.com  Wed Sep  1 12:10:54 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Sep 1999 16:10:54 GMT
Subject: SyntaxError exception 1.5.1 vs 1.5.2
References: <7q7822$ba6$1@nnrp1.deja.com>
Message-ID: <7qjj6e$4eu@dfw-ixnews21.ix.netcom.com>

In article <7q7822$ba6$1 at nnrp1.deja.com>,   wrote:
>
>i've noticed differences between 1.5.1 and 1.5.2 wrt: handling a
>SyntaxError exception. Which behavior is correct?
>
>Here is the scenario:
>
>bad_syntax.py:
>hello"
>
>build_py.py:
>import sys
>
>try:
>	py_compile.compile('bad_syntax.py')
>	print "after compile"
>except:
>	print "compile error"
>
>now the output:
>
>python1.5.1:
>compile error
>
>python1.5.2:
>File "bad_syntax.py", line 1
>    hello"
>         ^
>SyntaxError: invalid token
>after compile

I dunno.  Tim? Fred? Fredrik?  Yoo-hoo!
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From fredrik at pythonware.com  Mon Sep  6 03:25:51 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 6 Sep 1999 07:25:51 GMT
Subject: Localizing strings?
References: <7qv8c8$qtm$1@nnrp1.deja.com>
Message-ID: <004701bef839$0e303fe0$f29b12c2@secret.pythonware.com>

Preston Landers  wrote:
> Is there an easy way to use translatable strings in your Python program
> a la GNU gettext?  The os.localize module doesn't seem to do what I need
> to do.  can anyone point me to some documentation or examples?

how about:

    # Translate.py

    messages = {}

    # ... check environment variable to figure out what file
    # to load, and load it using marshal, fp.readlines, pickle, ...

    # do it by hand, for now...
    message["hello, world"] = "tjaba"

    def T(message):
        return table.get(messages, messages)

and:

    # main

    from Translate import T

    print T("hello, world")






From donn at oz.net  Mon Sep  6 12:43:45 1999
From: donn at oz.net (Donn Cave)
Date: 6 Sep 1999 16:43:45 GMT
Subject: ANN: BeOS system API modules
Message-ID: <7r0r01$qqg$0@216.39.151.169>

First release of some stuff I've been fiddling with for a long time.
BeOS is, of course, the new computer operating system that has Microsoft
quaking in their boots, as we learned during recent judicial proceedings.
It has a C++ API, which has been some fun to integrate with Python.

The result is a more or less idiomatic native BeOS Python interface.
One module per C++ class, modules import each other liberally.
Only a fairly minimal complement of modules at this point, but
they're generated from templates and fairly easy to write.  I have
included a USENET news reader as a sample application.

Lots of stuff remains to be done, haven't even thought up a name for it.
http://www.oz.net/~donn/

	Donn Cave, donn at oz.net, donn at u.washington.edu



From gmcm at hypernet.com  Thu Sep  9 17:19:33 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 9 Sep 1999 16:19:33 -0500
Subject: Turing Compliant?
In-Reply-To: <7r935e$u0f$1@nnrp1.deja.com>
Message-ID: <1275214276-477894@hypernet.com>

Jim Hefferon wrote:

[snip]

> Somebody else in this thread mentioned quantum computation.  I've
> tried to read up on it, but I'm afraid I didn't get it.  Is it the
> case that anything quantum computable is Turing computable, that is,
> quantum computability may go faster but doesn't add anything
> actually new?

Well there's biological computation, in which each fork() grows it's 
own "CPU", thus making some formerly non-computable problems 
computable.

No, I didn't make that up, and yes, I probably got something subtly 
wrong, thus ensuring the non-boundedness of this thread...

- Gordon



From gmcm at hypernet.com  Thu Sep  9 08:26:32 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 9 Sep 1999 07:26:32 -0500
Subject: Reading plain text file database tables
In-Reply-To: <37D781B8.FE238AC8@inet.polyu.edu.hk>
Message-ID: <1275246255-103229193@hypernet.com>

I think you'll find a regex that does this on Hans Nowaks snippets 
page, whereever that is at the moment (he just posted a notice that 
it had moved in the last few days).

Just found it at http://www.hvision.nl/~ivnowa/snippets/source/67.py

Li Dongfeng wrote:

> Thanks for the direction. I'm also
> trying to solve this with RE. But in order
> to extract "ab \" cd" like structure,
> your [^"\\] seems puzzling, because anything
> inside [] represent only one character, not
> the '\"' two character sequence. Any real
> worked out result?
> 
> Stephan Houben wrote:
> > 
> > On Thu, 09 Sep 1999 14:15:38 +0800, Li Dongfeng
> >  wrote:
> > 
> > >
> > >Do we have a module to read plain text file
> > >database tables?
> > >
> > >All the data management software, e.g. excel,
> > >dBase, SAS, etc., support input/output a table
> > >from/to a plain text file, fields can be separated
> > >by their column position, by spaces, by tabs,
> > >by commas, etc.
> > >
> > >How can we read this kind of file into a matrix like
> > >structure(list of lists)? I have written one reading
> > >files with fixed-width fields. For delimited files,
> > >simply using string.split work most of the time, but
> > >fails reading lines like
> > >
> > >  "Peter Thomson"  25  36
> > >
> > >or even
> > >
> > >  "Peter\" Thomson" 25 36
> > >
> > >I think this is a common task, so maybe someone has
> > >already given a very good solution.
> > 
> > Use regular expressions.
> > A regular expression that matches strings within "..",
> > while escaping " with \, is:
> > 
> > r = re.compile(r'"([^"\\]|(\\.))*"')
> > 
> > Extend this to get what you want.
> > If you're very worried about speed, create
> > a lexer using (f)lex, and then call the generated
> > C code from python.
> > 
> > Greetings,
> > 
> > Stephan
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

- Gordon



From guido at cnri.reston.va.us  Wed Sep 29 08:36:11 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 29 Sep 1999 08:36:11 -0400
Subject: Freezing script gives error message:
References: <7sdlht$e0f$1@nnrp1.deja.com> <37f103a0.70349727@freenews.remarq.com>
Message-ID: <5l3dvx99ok.fsf@eric.cnri.reston.va.us>

jiwils at stopspam.acxiom.com (Jimmy Wilson) writes:

> On Thu, 23 Sep 1999 16:46:58 GMT, swgeorge at my-deja.com wrote:
> 
> >  Does anybody know why I am getting the following error message:
> >  Error:  needed directory D:\Program Files\Python\lib\python1.5\config
> >  not found.
> 
> I have received the same error message, and I am puzzled as well.  I
> also install the binary distribution of Python as well as downloading
> the source.  I suspect that some registry entries are part of the
> problem as I removed the binary installation of Python and I the
> freeze.py script still knew where that old installation was.

The freeze script is confused -- it seems to be thinking it is running
on a Unix system.  How did you invoke it?  What Python version are you
using?    You *could* try running freeze.py -w...

--Guido van Rossum (home page: http://www.python.org/~guido/)



From m.faassen at vet.uu.nl  Sun Sep  5 14:12:34 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 1999 18:12:34 GMT
Subject: Spanish Inquisition?!
References: <37CFC38C.20A9C267@megsinet.net>
Message-ID: <7qubqi$q02$3@newshost.accu.uu.nl>

David Oppenheimer  wrote:
> In a previous thread (about languages interpreted versus languages
> compiled) there was talk of the Spanish Inquisition.  I am guessing that
> this is 1) a joke or 2) the code name for a "secret" project to create
> compiled code from python code and thereby permanently blur the line for
> Python.  Anyone know anything about this?

> I know no one ever expects the Spanish Inquisition, but if its coming
> I'd like to know about it.

Yes, this is currently vaporware starting to lose its vapor. It's been
called 'Swallow' by me (don't know if it's African or European). More
recently, John Skaller has been working on Viper (which apparently actually
has some source code, so even less vapor). Before my time, there have
been other attempts I don't know the name of. :)

The Spanish Inquisition is definitely a cool name for such a project, 
though. :)

Regards,

Martijn




From geek+ at cmu.edu  Mon Sep 13 09:30:42 1999
From: geek+ at cmu.edu (Brian E Gallew)
Date: 13 Sep 1999 09:30:42 -0400
Subject: ANN: Tk Slashdot Headline Viewer
References: <7rgmd0$2htg$1@hub.org>
Message-ID: 

From: Doug at news.vex.net (Lewis)
> Inspired by the Slashdot headline viewer demo in the wxPython
> package, i decided to implement my own in Tkinter.
> 
> My implementation, while simpler, is much cuter. (-:
> 
> Should anyone care to try it out, you can snag it from....
> 
> http://www.vex.net/~x/python_stuff.html#slashdot

Dude, if you want this to run under unix, change the spawnBrowser
function to do the following:
os.system(browser % url)
And browser should look like:
browser = "/usr/local/bin/netscape -remote 'openURL(%s)'"


-- 
=====================================================================
| JAVA must have been developed in the wilds of West Virginia.      |
| After all, why else would it support only single inheritance??    |
=====================================================================
| Finger geek at cmu.edu for my public key.                            |
=====================================================================
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 266 bytes
Desc: not available
URL: 

From jrush at summit-research.com  Tue Sep 21 22:52:01 1999
From: jrush at summit-research.com (Jeff Rush)
Date: 21 Sep 1999 21:52:01 -0500
Subject: Module to open URLs that set cookies?
References: <15d84bec.887c3642@usw-ex0107-051.remarq.com>
Message-ID: 

On Thu, 9 Sep 1999 22:26:15, Marshall T. 
 wrote:

> I would like to know how to write an http client that will 
> accept cookies when opening webpages.  I'm attempting to 
> write a module that open a page and note the latency in 
> doing so.  I've been successful in doing this with the 
> sockets and urllib library modules, but have been foiled by 
> pages that set cookies.  Is there a canned library for 
> doing this in Python?

I do exactly that in my webbot module at:

    http://starship.python.net/crew/jrush/Webbot/

-Jeff Rush




From news at dorb.com  Thu Sep  2 22:14:17 1999
From: news at dorb.com (Darrell)
Date: Thu, 2 Sep 1999 22:14:17 -0400
Subject: Bring Out Yer Dead
References: <37CE2648.D4254FCC@megsinet.net> <37ceeef4.5698550@news.erols.com>
Message-ID: 

As a check.
Down loaded PIL_win32.tar.gz from this link and extracted it with WinZip 7.0

--
--Darrell
Robert Kern  wrote in message
news:37ceeef4.5698550 at news.erols.com...
> On Thu, 02 Sep 1999 03:24:59 -0400, David Oppenheimer
>  wrote:
>
> >I am going to give up on trying to get the PIL to work with Pythonwin
> >1.5.2 unless someone who has it working with their setup can give me
> >explicit instructions on how to do this.  I am unable to implement
> >Robert Kern's instructions due to the fact that I cannot get my Win Zip
> >program to unzip the tar.gz file format.
>
> This is probably a browser problem.  Clear your cache first.  Be sure
> that the .gz and .tar extensions are associated with Winzip in your
> browser.  Their MIME types should be "application/x-gzip" and
> "application/x-tar", respectively.  At least, that's how I have them.
>
> Then try downloading again (use a shift-click from
> http://starship.python.net:9673/crew/da/Code/PIL rather than a
> straight left-click).
>
> >So ... time for me to put PIL for Pythonwin 1.5.2 on the pile of
> >software corpses and set it ablaze.  Can you feel the heat?!  Anyone
> >have any marshmallows?
>
> You'll be fine.  BTW, _Pythonwin_ is the name of Mark Hammond's IDE
> for Python on Windows, not the general name for Python on Windows.
>
> >David O.
>
> Robert Kern           |
> ----------------------|"In the fields of Hell where the grass grows high
> This space            | Are the graves of dreams allowed to die."
> intentionally         |           - Richard Harter
> left blank.           |





From wtanksle at dolphin.openprojects.net  Thu Sep  2 02:15:30 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 02 Sep 1999 06:15:30 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com><37CCE10D.D07D023B@strs.co.uk><7qjgq8$k1i$1@newshost.accu.uu.nl> <14285.38262.884171.416195@dolphin.mojam.com> 
Message-ID: 

On Wed, 1 Sep 1999 15:51:00 -0700, Emile van Sebille wrote:
>Skip Montanaro  wrote in message

>> So, aside from the differences in how you get from source to running
>> program, depending on your definitions, they can all be called
>> compiled or interpreted languages.

>From my perspective, <25+years business basic> it's interpretive if it
>politely stops when it hits a bug, and waits for you to fix the program
>and continue.

You see, that's what I call an "interactive" language.

I don't have much use for the word "interpreted".  It used to mean a lot,
back when only a few mmethods of compilation and interpretation were
commonly used.  Its meaning and use are now completely gone.

>If-you-can't-change-a-running-program-then-it's-compiled-ly yrs

>Emile van Sebille

-- 
-William "Billy" Tanksley



From rvollmert at gmx.net  Wed Sep  8 17:00:36 1999
From: rvollmert at gmx.net (Robert Vollmert)
Date: Wed, 8 Sep 1999 23:00:36 +0200
Subject: Python example: possible speedup?
In-Reply-To: <87puztv2gt.fsf@pc-hrvoje.srce.hr>; from Hrvoje Niksic on Wed, Sep 08, 1999 at 05:37:38PM +0200
References: <87puztv2gt.fsf@pc-hrvoje.srce.hr>
Message-ID: <19990908230036.A3530@krikkit.vollmert.home>

On Wed, Sep 08, 1999 at 05:37:38PM +0200, Hrvoje Niksic wrote:
> As a Python exercise, I wrote a simple program to "scratch an itch",
> i.e. do something useful.  However, I found that Python's lack of
> speed really bytes me here, so I'd like to hear suggestions for
> speedup.  People who don't like that kind of topic, please skip to the 
> following article.  Others, read on.

... 

> The program was quite easy to write, and easy to read afterwards.  The
> problem is that it is also quite slow.  On my system, it takes about
> 27 CPU seconds (as reported by `time' shell builtin) to do the work,
> which can extend to more than a minute of real time, depending on the
> system load.
> 
> As a comparison, the equivalent Perl program does the same thing in 9
> CPU seconds.  I tried everything I knew to make the Python version
> fast.  I tried to use `re' to avoid returning headers other than the
> ones we're interested in.  I tried changing self.__current to just
> current to avoid a dictionary lookup.  I tried to make self.__current
> a list, to avoid the expensive `current = current + line' operation.
> All of these things made the program measure slower.
> 
> I would really appreciate some suggestions.  The code is not large,
> and is (I hope) rather elegant.  I am a Python beginner, so I'd also
> appreciate tips on Python style and OO technique.  I'll post/mail the
> Perl equivalent on demand.

Hi,

the following code is about five times faster here. That means it's
faster than perl, I suppose :-). It is of course not quite as general
as your's, but it seems to fit the job nicely. For this kind of
problem, I usually don't bother with OO.

The main speed advantage seems to be that the files aren't processed
line by line, which is of course very memory consuming.

HTH,
Robert

#! /usr/bin/env python

import string

def get_installed():
    fstatus = open('/var/lib/dpkg/status', 'r')
    status = fstatus.read()
    fstatus.close()
    status = string.split(status, '\n\n')
    installed = []
    for package in status:
        fields = string.split(package, '\n')
        name = fields[0][9:]
        for line in fields:
            if line[:7] == 'Status:':
                if string.split(line[8:])[-1] == 'installed':
                    installed.append(name)
                break
    return installed

def get_sizes(packages):
    favailable = open('/var/lib/dpkg/available', 'r')
    available = favailable.read()
    favailable.close()
    available = string.split(available, '\n\n')
    results = []
    for package in available:
        fields = string.split(package, '\n')
        name = fields[0][9:]
        if name in packages:
            for line in fields:
                if line[:15] == 'Installed-Size:':
                    results.append(name, int(line[16:]))
                    break
    return results

def main():
    results = get_sizes(get_installed())
    results.sort(lambda a, b: cmp(b[1], a[1]))
    for r in results:
        print '%s: %d' % r

if __name__ == '__main__':
    main()

-- 
Robert Vollmert                                      rvollmert at gmx.net



From pinard at iro.umontreal.ca  Thu Sep  2 08:30:45 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 02 Sep 1999 08:30:45 -0400
Subject: Python speed and `pcre'
In-Reply-To: "Darrell"'s message of "Tue, 31 Aug 1999 13:25:09 -0400"
References:  
Message-ID: 

"Darrell"  ?crit:

> I've found a common mistake is splitting and combining strings.  Use
> string.join(), don't do this str=str+str1 in a loop.  Strings are immutable
> so you end up allocating memory and coping every time.  Very expensive.

Hi, Darrell, and gang.

I do not fully understand the cost of splitting.  If strings are immutable,
there is really no need to copy a slice contents.  The only reason I
really see to copy would be to simplify the garbage collector, which is
more related to a particular implementation than the language design itself.

Somewhere, it is said that some extra-space is allocated at end of sequences,
to allow for some peaceful growth.  Does `str=str+str1' at least attempt
to use of this extra-space?  It should, in my opinion.  Of course, you
might tell me that the implementation does not notice the form `str=str+X'
as really meaning `str+=X', but then, again, the inefficiency might be
related to the implementation.

The trend, for a lot of years now, is to improve the implementation
for speed, rather than adapt the programming style of all users to a
given implementation.  If Python is meant to be a clean, clear language,
one should not have to resort to convolutions to give it reasonable speed.
Or else, things look a bit artificial, and the gain in clarity is a bit lost.
Oh, I'm sure we get used to idioms after a while, then simplicity is lost.
I'm not saying that `str+=X' should make it into the language, I quite
guess this has been debated to death already, and I agree that the writing
`str=str+X' has the virtue of clarity.  But making the language more clear
should not necessarily be accompanied by making it slower.  Avoiding a
syntactic construct does not free the implementation from recognising the
intent behind most usual alternate writings for it (in our case, here,
extending a given string) and acting best, accordingly.

Once again, I'm fairly new to Python, and ask for your kind forgiveness if
I err too much.  Maybe that `str=str+X' is already recognised and acted
upon as it should, usually without copy when X is small, and maybe that
slicing a string just knows, already, how to not copy the slice contents.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From jh at cddk.dk  Thu Sep  9 10:44:02 1999
From: jh at cddk.dk (Jesper Hertel)
Date: Thu, 9 Sep 1999 16:44:02 +0200
Subject: Query system date
References: <7r8fjh$dno$1@nnrp1.deja.com>
Message-ID: <7r8h3k$pqo$1@news1.tele.dk>

Hello Mark,

Try this:

def getCurrentTimeString():
    from time import *
    return strftime ("%d-%m-%Y", localtime(time()))

print getCurrentTimeString()


Jesper Hertel

 wrote in message
news:7r8fjh$dno$1 at nnrp1.deja.com...
> I am new to python and I am trying to figure out how to return the
> computers system date and time and then format it as MM-DD-YYYY.  I have
> searched the documentation up and down and have not found anything about
> how to accomplish this.  Any help would be greatly appeciated.
>
> Thanks,
>
> Mark.
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.





From davidopp at megsinet.net  Wed Sep 22 16:57:10 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 22 Sep 1999 16:57:10 -0400
Subject: PTUI
Message-ID: <37E942A6.969CB5C4@megsinet.net>

Does anyone know if PTUI (python tkinter user interface) is still being
actively maintained?  I tried to send an e-mail to Zachary Roadhouse and
it was returned undeliverable.

Thanks,

David O.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From hniksic at srce.hr  Wed Sep  8 11:37:38 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 08 Sep 1999 17:37:38 +0200
Subject: Python example: possible speedup?
Message-ID: <87puztv2gt.fsf@pc-hrvoje.srce.hr>

As a Python exercise, I wrote a simple program to "scratch an itch",
i.e. do something useful.  However, I found that Python's lack of
speed really bytes me here, so I'd like to hear suggestions for
speedup.  People who don't like that kind of topic, please skip to the 
following article.  Others, read on.

When browsing my Debian packages, I found I often wanted to know the
size of the installed packages.  To achieve this, I wrote a simple
program that prints the installed packages and their sizes, all sorted
by size.  Output looks like this:

tetex-base: 27652
tetex-extra: 27118
cmucl-safe: 22174
mozilla: 13007
timidity-patches: 11662
locales: 11489
quake2: 10800
[...]

To achieve this, the program needs to parse two files:
/var/lib/dpkg/status, to find which packages are installed, and
/var/lib/dpkg/available, to find the size of the installed packages.
Both files are textual, in the form of headers.  Like this:

Package: telnet
Status: install ok installed
Priority: standard
Section: net
Installed-Size: 130
Maintainer: Herbert Xu 
Source: netkit-telnet
Version: 0.14-4
Replaces: netstd
Depends: libc6 (>= 2.1), libncurses4 (>= 4.2-3.1)
Description: The telnet client.
 The telnet command is used for interactive communication with another host
 using the TELNET protocol.

Package: packaging-manual
Status: purge ok not-installed
Priority: optional
Section: doc
[...]

The program was quite easy to write, and easy to read afterwards.  The
problem is that it is also quite slow.  On my system, it takes about
27 CPU seconds (as reported by `time' shell builtin) to do the work,
which can extend to more than a minute of real time, depending on the
system load.

As a comparison, the equivalent Perl program does the same thing in 9
CPU seconds.  I tried everything I knew to make the Python version
fast.  I tried to use `re' to avoid returning headers other than the
ones we're interested in.  I tried changing self.__current to just
current to avoid a dictionary lookup.  I tried to make self.__current
a list, to avoid the expensive `current = current + line' operation.
All of these things made the program measure slower.

I would really appreciate some suggestions.  The code is not large,
and is (I hope) rather elegant.  I am a Python beginner, so I'd also
appreciate tips on Python style and OO technique.  I'll post/mail the
Perl equivalent on demand.

The code follows:

#!/usr/bin/python

import string

class Dpkg_Reader:
    def __init__(self, file):
        self.__current = ''
        self.__fp = open(file)

    def next_header(self):
        while 1:
            line = self.__fp.readline()
            if line == '':
                return (None, None)     # EOF
            if self.__current:
                if line == "\n" or (line[0] != ' ' and line[0] != "\t"):
                    #print "{%s}" % (self.__current)
                    try:
                        name, value = string.split(self.__current, ': ', 1)
                    except:
                        name, value = string.split(self.__current, ':', 1)
                    value = value[:-1]
                    if line == "\n":
                        self.__current = ''
                    else:
                        self.__current = line
                    return (name, value)
            self.__current = self.__current + line

    def close(self):
        self.__fp.close()

def process_status(file):
    reader = Dpkg_Reader(file)
    package = None
    installed = {}
    while 1:
        name, value = reader.next_header()
        if name is None:
            break
        if name == 'Package':
            package = value
        elif name == 'Status':
            status = string.split(value, ' ')
            if status[2] == 'installed':
                installed[package] = 1
    reader.close()
    return installed

def process_available(file, installed):
    reader = Dpkg_Reader(file)
    sizes = {}
    while 1:
        name, value = reader.next_header()
        if name is None:
            break
        elif name == 'Package':
            package = value
        elif name == 'Installed-Size':
            if installed.has_key(package):
                sizes[package] = string.atoi(value)
    reader.close()
    return sizes

def main():
    installed = process_status('/var/lib/dpkg/status')
    sizes = process_available('/var/lib/dpkg/available', installed)
    lst = sizes.keys()
    lst.sort(lambda a, b, sizes=sizes: cmp(sizes[b], sizes[a]))
    for pack in lst:
        print "%s: %d" % (pack, sizes[pack])

if __name__ == '__main__':
    main()



From greg.ewing at compaq.com  Mon Sep  6 17:17:15 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 07 Sep 1999 09:17:15 +1200
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com> <37CCE10D.D07D023B@strs.co.uk> <7qjgq8$k1i$1@newshost.accu.uu.nl>  <7qlj3b$p7u$1@newshost.accu.uu.nl>  <7qrce6$dgh$1@news.bayarea.net> 
Message-ID: <37D42F5B.F8D00F97@compaq.com>

Jean Jordaan wrote:
> 
> Z-code! That's the bytecode of the Infocom text adventure engine. Infocom
> wrote Z-code engines for all their target machines, and released one version
> of the bytecode of the games for all of them. It's been implemented as Inform
> by Graham Nelson, and ported all over the place. Does this qualify :?

It doesn't qualify for the "published" part. Infocom never publicised
the
specification of the Z-machine; it was reverse-engineered later by I.F.
enthusiasts.

The Amsterdam Compiler Kit uses an abstract machine code called EM with
a
published specification, although it's intended more as an intermediate
form
between the front and back ends of a compiler than as a medium for
distributing
programs.

Greg



From Bill at nospam.com  Fri Sep 24 12:10:18 1999
From: Bill at nospam.com (Bill Rodgers)
Date: Fri, 24 Sep 1999 10:10:18 -0600
Subject: NT Services In Python?
Message-ID: 

Has anyone done this?  I don't see any mention of it
in the latest release.  Any info greatly appreciated.
Thanks,
--
Bill Rodgers
Enterprise Management Consultant
Specialty: BMC/Patrol
brodg1 at ix.netcom.com





From jean at lnx2.sun.ac.za  Sun Sep  5 08:50:20 1999
From: jean at lnx2.sun.ac.za (Jean Jordaan )
Date: 5 Sep 1999 12:50:20 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com> <37CCE10D.D07D023B@strs.co.uk> <7qjgq8$k1i$1@newshost.accu.uu.nl>  <7qlj3b$p7u$1@newshost.accu.uu.nl>  <7qrce6$dgh$1@news.bayarea.net>
Message-ID: 

On Sat, 4 Sep 1999 08:09:29 -0700, Frank Mitchell  wrote:
>Stephan Houben wrote in message ...
>>m.faassen at vet.uu.nl (Martijn Faassen) writes:
>>Of course, Sun marketed the fact that Java used bytecode as if it was
>>something revolutionary. "No, it is not interpreted, it is compiled to
>>Byte Code!" Of course, all those other languages which had been called
>>"interpreted" where actually compiled to byte code since the
>>beginning of time. Byte-code compilation is really as old as the Sun. ;-)
>
>Right, but as the guy who hired me into Sun pointed out, it's the first time
>(AFAIK) that a byte code format has been fully specified and published, with
>the intent of transferring bytecode among different implementations.
>
>If someone could point me to a previous bytecode format that worked in more
>implementations than the one that issued it, and stayed stable from revision
>to revision, I will stand corrected. 

Z-code! That's the bytecode of the Infocom text adventure engine. Infocom
wrote Z-code engines for all their target machines, and released one version
of the bytecode of the games for all of them. It's been implemented as Inform 
by Graham Nelson, and ported all over the place. Does this qualify :?

--jean



From gambino at news.mondenet.com  Tue Sep  7 19:13:42 1999
From: gambino at news.mondenet.com (Jack)
Date: Tue, 07 Sep 1999 23:13:42 GMT
Subject: Formatting numbers
Message-ID: 

I'm learning Python and the documentation is great,
but I haven't found an easy way to do the following.
Given an array (2-D) of numbers, I'd like to print 
them out nicely formatted. The numbers can differ
greatly in magnitude, so I'd like to use fewer
decimal places for big numbers than for large ones.

The output should look something like this (or better):

 123456.78   -0.000123   -54231.321
-4321567.1    9876.543    etc.

Someone somewhere must have written a module to do
this. Any pointers?

P.S. I want to use the Python interpreter only at
this time -- no tk or gtk.

Thanks.



From philh at vision25.demon.co.uk  Tue Sep 28 17:22:07 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 28 Sep 99 21:22:07 GMT
Subject: Please Help Convert This Code
References: <37E96EE5.95270EA9@megsinet.net>
Message-ID: <938553727snz@vision25.demon.co.uk>

In article <37E96EE5.95270EA9 at megsinet.net>
           davidopp at megsinet.net "David Oppenheimer" writes:
> Dear Group,
> 
> I would really appreciate it if someone could help convert this Tkinter
> code into equivalent wxPython code.  Need this to be done to create a
> drag and drop GUI maker that will have Tkinter AND wxPython as output
> choices.  Thanks in advance.
> 
> David O.
> 
> [Tkinter code deleted]

Unfortunately, I don't know wxPython, so I can't help you with your 
query.

But I couldn't help feeling that what you're doing is very similar
to what I'm doing with Parrot. 

(Parrot is a text-based GUI builder; it converts a text file into 
GUI code -- currently it produces Python/Tkinter code, but it is
designed to be able to output to other formats too (different
python GUI libraries, and different languages. For details,
see http://www.vision25.demon.co.uk/prog.parrot.html ).

If you are going to be writing a wxPython code generator, and so 
am I, perhaps we could work together on this one?

-- 
Phil Hunt - - - - - - - - -  philh at vision25.demon.co.uk
   - Linux was 8 years old on 17th September! See: -
http://www.vision25.demon.co.uk/prog/linuxbirthday.html




From emile at fenx.com  Sat Sep  4 11:20:32 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 4 Sep 1999 08:20:32 -0700
Subject: "if" syntax error
References: <37D138C1.3EABC811@dialnet.net>
Message-ID: 

Hi Andrew,

It may not be the if... statement.  When I get these *obviously correct*
syntax errors, it's typically that I've left an incomplete statement
unterminated.

HTH

--

Emile van Sebille
emile at fenx.com
-------------------


Andrew Kesterson  wrote in message
news:37D138C1.3EABC811 at dialnet.net...
> I've got a CGI script (python) set to check the value of form element
> "Listing_Type", and it does it as thus:
>
> CGIHeaders = cgi.FieldStorage()
> if CGIHeaders["Listing_Type"].value == "Residential":
>
> when I compile it, I get this error:
>
>   File "C:\Program Files\Python\Source\Skydev\addlisting.py", line 82
>      if CGIHeaders["Listing_Type"].value == "Residential":
>       ^
>  SyntaxError: invalid syntax
>
> There is NO WAY that this is invalid syntax! I've used if comparisons
> like that one hundreds of times and compiled fine!
> What's going on? Or is it just that my CGI *really* doesn't want to
run
> from command line?
>
> --
> "I never came across anyone in whom the moral sense
> was dominant who was not heartless, cruel, vindictive,
> log-stupid and entirely lacking in the smallest sense
> of humanity. Moral people, as they are termed, are
> simply beasts. I would sooner have fifty unnatural
> vices than one unnatural virtue."
> --Oscar Wilde





From davidopp at megsinet.net  Wed Sep  8 01:27:10 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 08 Sep 1999 01:27:10 -0400
Subject: Stackless Python
Message-ID: <37D5F3AE.D187DB3C@megsinet.net>

Found this link to a paper by Christian Tismer called Stackless Python.
I still don't understand everything he's talking about but I found it
very interesting.

David O.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From wtbridgman at radix.net  Sat Sep 18 09:39:47 1999
From: wtbridgman at radix.net (T. Bridgman)
Date: Sat, 18 Sep 1999 9:39:47 -0400
Subject: [Mac] Python mode for Alpha editor
References: 
Message-ID: <01HW.B4090E63001D90290D2F1590@news1.radix.net>

Thanks Howard.

There is in fact a Python mode for Alpha but it doesn't seem to work in release 
7.0 or later.  Does this version work with 7.+?

Tom

On Fri, 17 Sep 1999 15:05:07 -0400, Howard Oakley wrote
(in message ):
> 
> Gentlepersons, 
> 
> I looked in vain for a Python mode for the wonderful (if Tcl-based!) text 
> editor, Alpha. I could not find any. Here, below, is an Alpha mode file for 
> Python, which makes the Alpha text editor much more usable for editing 
> Python code. Please feel free to post it anywhere that you feel appropriate. 
> The code works fine here, but is kludgy and largely borrowed from other Tcl 
> mode files - I could hardly claim any sort of copyright anyway! 
> 
> I am also sending it to Pete Keleher, the author of Alpha, and Jack Jansen. 
> 
> Howard. 




From fredrik at pythonware.com  Wed Sep 22 06:13:44 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 22 Sep 1999 10:13:44 GMT
Subject: lists and append, and loop iteration
References: <37E8860B.C024C4F8@radiata.com> <14312.40896.797841.421283@gargle.gargle.HOWL> <87ln9ztgml.fsf@ifm.uni-kiel.de>
Message-ID: <001501bf04e3$27634e10$f29b12c2@secret.pythonware.com>

Janko Hauser  wrote:
> I could not write
> >>> dict={}
> >>> dict['new_key'] = dict.get('new_key',[]).append(['new_value'])
> >>> dict
> {'new_key': None}
> 
> To this refered my last note about the reference of a. Although I do
> not understand this. If I get not an empty list, why can I do an
> append?

"get" returns an empty list, "append" returns None.






From mal at lemburg.com  Thu Sep  9 18:56:27 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 10 Sep 1999 00:56:27 +0200
Subject: mxODBC: File DSNs?
References: <37D5E859.78BB8DF0@compaq.com> <37D6B32E.40846F82@lemburg.com> <37D82A0A.F7F84305@compaq.com>
Message-ID: <37D83B1B.1B0F4388@lemburg.com>

Greg Ewing wrote:
> 
> "M.-A. Lemburg" wrote:
> >
> > Just add the File DSN to your ODBC Manager setup and then point
> > mxODBC at it using ODBC.Windows.DriverConnect().
> 
> This doesn't work.
> 
> Under the "File DSN" tab in the ODBC Data Source Administrator
> I have an entry called "Py2kTest2.dsn" which refers to a
> Microsoft Access database. But "Py2kTest2" or "Py2kTest2.dsn"
> does not seem to be a valid data source name:
> 
>    >>> db = ODBC.Windows.DriverConnect("DSN=Py2kTest2")
>    Traceback (innermost last):
>      File "", line 1, in ?
>        db = OW.DriverConnect("DSN=Py2kTest2")
>    OperationalError: ('IM002', 0, '[Microsoft][ODBC Driver Manager] Data
> source name not found and no    default driver specified', 4316)
> 
>    >>> db2 = ODBC.Windows.DriverConnect("DSN=Py2kTest2.dsn")
>    Traceback (innermost last):
>      File "", line 1, in ?
>        db2 = ODBC.Windows.DriverConnect("DSN=Py2kTest2.dsn")
>    OperationalError: ('IM002', 0, '[Microsoft][ODBC Driver Manager] Data
> source name not found and no
>    default driver specified', 4316)
> 
> If I create a User DSN with the same parameters, everything
> is fine.
> 
> What magic string do I need to give to DriverConnect to
> specify a File DSN?

Try "FILEDSN=" as parameter. The ODBC docs for
SQLDriverConnect have more details on the dsn string to be
used.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   113 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From skip at mojam.com  Fri Sep  3 14:56:09 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri,  3 Sep 1999 13:56:09 -0500 (CDT)
Subject: smtplib: subject? to?
In-Reply-To: <19990903202313.A23408@optiplex.palga.uucp>
References: <19990903142456.A2124@optiplex.palga.uucp>
	
	<19990903202313.A23408@optiplex.palga.uucp>
Message-ID: <14288.6544.705239.464758@dolphin.mojam.com>

    Gerrit> Ah.  But if I have to put a To: in the message, why do I have to
    Gerrit> give recipients?

SMTP has "mail from" and "rcpt to" commands.  Those are what the to and from
args are for I believe.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From wware-nospam at world.std.com  Wed Sep  8 11:56:03 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Wed, 8 Sep 1999 15:56:03 GMT
Subject: Stackless Python
References: <37D5F3AE.D187DB3C@megsinet.net>
Message-ID: 

David Oppenheimer (davidopp at megsinet.net) wrote:
: Found this link to a paper by Christian Tismer called Stackless Python.
: I still don't understand everything he's talking about but I found it
: very interesting.

Python normally uses the C stack for some of its house-keeping when it's
running code. Functions, and I believe also blocks within functions, require
data structures called frames to keep track of variable scope and some other
things. These are usually maintained as local variables of eval_code2, a
function in Python/ceval.c. eval_code2 calls itself recursively whenever one
Python function calls another (except for functions which are coded directly
in C), so the eval_code2 call representing the outer function is still on
the C stack, including any frames for the outer function.

What Christian is doing is to set up frame stacks as independent Python
objects, distinct from the C stack. This means that you could interrupt a
Python process and store or study the frozen snapshot. You could continue
the process some time later. You could step many threads, each a few opcodes
at a time, to implement a system of multiple threads. There are probably
other applications that I'm not aware of.

The reason I know this is because I implemented my own multiple-thread package
a few weeks back, having paid insufficient attention when Stackless Python
was first mentioned on the newsgroup. People then pointed out to me that
almost all of what I was doing was already done in Stackless Python.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From MHammond at skippinet.com.au  Mon Sep  6 19:04:08 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 7 Sep 1999 09:04:08 +1000
Subject: case for open lang on win32 - fill in the table
References: <7r0049$bfh$1@nnrp1.deja.com> <7r0d3t$chk$1@newshost.accu.uu.nl>
Message-ID: <7r1hab$9bd$1@m2.c2.telstra-mm.net.au>

Martijn Faassen wrote in message <7r0d3t$chk$1 at newshost.accu.uu.nl>...
>[followups set to comp.lang.python only]

Good move :-)  I recently cross-posted a Python question to the Perl groups
'cos I didnt look hard enough - was then honoured with an autoresponder
message from an tchrist. (:-)

>> multi thread        none              recent-ok    ?
>
>Python can do multithreading, though there's a global interpreter lock.
>Ask someone who knows more about multithreading than I do what this means.

Python is fully multi-threaded.  It wont perform as well as you may expect
on an SMP machine, but otherwise is fully thread safe.

>> file system access  no                yes          yes
>> basic locking       no (obviously)    flock        flock?
>
>Python has the fcntl module -- again, ask others for details as I'm
>not familiar with this. Seems limited to Unix.

Indeed it is.  There are no portable locking semantics, but there is full
support for locking (via the "msvcrt" module, or the win32file module.

>> idispatch only      yes               ?            ?

IDispatch plus about 100 extra interfaces.  We can not use raw interfaces if
there is no C++ support, but many useful interfaces already have that
support (and it is trivial to add new ones).  From a marketting POV, this
should definately read "yes" :-)

>> COM threading       STA only          STA only?    STA only?

No - full support for all COM threading models.  STA is the default.

>> project size        small to medium   big          big
>> (lines of code)
>
>What does that mean, lines of code? for COM, or in general?

Agreed = I would have to argue "small", especially compared to VB.  For
_anything_ other than GUI code (which VBScript doesnt do anyway) the LOC for
a Python solution will always be smaller than for VB.

Hope this helps...  Let us know if you complete this and post it to the web.

Mark.





From skip at mojam.com  Wed Sep 15 15:46:15 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 15 Sep 1999 14:46:15 -0500 (CDT)
Subject: ANN: Biopython.org (Call for Participation)
In-Reply-To: <7rorsc$tf1$1@nnrp1.deja.com>
References: 
	<7rorsc$tf1$1@nnrp1.deja.com>
Message-ID: <14303.63367.306074.152475@dolphin.mojam.com>

    Peter> I am very interested in becoming a part of the biopython project
    Peter> but I am new to these types of forums. Could someone please let
    Peter> me know how I can get involved with writing code(how do you get a
    Peter> topic for code developement) and the scientific discussions.

Peter,

Best bet would be to trot on over the the biopython web site and sign up for 
the mailing list.  If you ask there, I'm sure someone will be able to direct 
you to some meaningful information.  You might also want to browse the
archives of the biopython lists (they just got started so this should only
take a few minutes).

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From kuncej at mail.conservation.state.mo.us  Tue Sep  7 11:13:49 1999
From: kuncej at mail.conservation.state.mo.us (Jeffrey Kunce)
Date: Tue, 07 Sep 1999 10:13:49 -0500
Subject: Snippets site temporarily (?) down
Message-ID: 

>OK, the title says it all... Well, in fact it's not just my snippets 
>...
>I have been thinking about moving the site elsewhere, since this 
>...

Why not on starship.python.net ?
Free accounts for PSA members.

  --Jeff





From henk_p at my-deja.com  Wed Sep 15 08:45:32 1999
From: henk_p at my-deja.com (henk_p at my-deja.com)
Date: Wed, 15 Sep 1999 12:45:32 GMT
Subject: Python, FastCGI and FreeBSD
Message-ID: <7ro4d9$apg$1@nnrp1.deja.com>

Hi,

I am experiencing this strange problem with a Python script running as a
fastCGI application under FreeBSD 3.2.
For some reason when the fastcgi process manager in Apache starts the
script, it (or the Python interpreter) immediately crashes. It does not
matter what the script contains, it always crashes upon initialization.
('Uncaught signal 11, Segmentation Fault' as it says in the apache error
log)

The strange thing is that the same script runs perfectly OK on a Redhat
linux box (5.2).

Are there any known bugs that could cause Python interpreter to crash on
initialization?, maybe due to the user privilege level it is running, or
the layout of the python libs on the platform (FreeBSD3.2).

The script runs perfectly fine when invoked from a shell. It just seems
like the python interpreter does not like to be forked by the FastCGI
process manager...

Well anyway, thanx in advance,


Henk.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jblaine at shell2.shore.net  Sat Sep  4 10:49:04 1999
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Sat, 04 Sep 1999 14:49:04 GMT
Subject: raw_input() and readline
References: 
Message-ID: 

Did you 'import readline' ?

>...but if I use 
>
>python t.py
>
>I don't have any readline functionality.
>I would appreciate any help.



From etsang at my-deja.com  Fri Sep 17 16:01:52 1999
From: etsang at my-deja.com (etsang at my-deja.com)
Date: Fri, 17 Sep 1999 20:01:52 GMT
Subject: NEWBIE: Try to intialise a big C structure easily from Python...HELP!!!!
Message-ID: <7ru6n9$pf1$1@nnrp1.deja.com>

Hey, I am trying to initialise a C structure from
Python.
My Aim is to init a C structure without passing
each indivual elements through the interface, but
rather package the whole thing and pass that it
one time. Then the C may do somethign about it,
initialise its won struct and pass back a struct
with the same kind of elements but with different
parameters. I am stuck with the intialising the C
struct part, and have not clue how to convert the
C struct back to a PyObject appropriate to pass
back to the Python script. Please help . Below is
the code.

Error: segmentation fault occurres at
PyArg_ParseTuple
(args,"s:mymath",string_from_Python))
PyString_Size(string_from_Python)calls inside
function pyfunc_mystruct

Here is what I tried:

>From the python script I have:
import struct
import mymath_wrapper
fmt = "120sf"
a=255
b='knights who say ni'
c=3.14159011841
data = struct.pack(fmt,a,b,c)
mymath_wrapper.mystruct(data)

in my mymath_wrapper.c file I have:
#include 
#include "mymath.h"

PyObject*
pyfunc_mystruct(self, args)
 PyObject *self, *args;
 {
 PyObject* string_from_Python;
 if(!PyArg_ParseTuple
(args,"s:mymath",string_from_Python))
 	return NULL;
 if(PyString_Size(string_from_Python)!= sizeof
(struct diffstruct))
 {
 	PyErr_SetString(PyExc_AssertionError,
"Given strgin not a good size");
 	return NULL;
 }


 globaldiffstruct = (diffstruct *)
PyString_AsString(string_from_Python);

 printf("globaldiffstruct calling within C:%d  %s
%f\n",globaldiffstruct->a, globaldiffstruct-
>b,globaldiffstruct->c);
 /* python side will print the last res value*/
 return string_from_Python; /* not correct, deal
with this later */
 }

 static PyMethodDef mymathMethods[] =
 {
 	{"mystruct", pyfunc_mystruct, 1},
 	{NULL,NULL}
 };

 void initmymath_wrapper()
 {
 	(void) Py_InitModule("mymath_wrapper",
mymathMethods);
 }






Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From wtanksle at dolphin.openprojects.net  Sat Sep 11 01:16:19 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sat, 11 Sep 1999 05:16:19 GMT
Subject: Text UI
References: <65118AEEFF5AD3118E8300508B124877073BEE@webmail.altiris.com> 
Message-ID: 

On Fri, 10 Sep 1999 21:10:16 -0700, Robin Dunn wrote:

>Mike Steed  wrote in message

>> One possibility worth investigating is wxWindows, which includes a curses
>> port.  I have not used the curses port, nor I do not know whether the
>> Python interface will work with it....

>> wxWindows: http://www.wxwindows.org/
>> wxPython: http://alldunn.com/wxPython/

>It doesn't.  The curses port was only for wxWindows 1.x, and if I remember
>correctly it was never quite finished.  The current wxPython is just for
>wxWindows 2.x.

The COOL thing is that according to the "supported platforms" page, Curses
is being ported back into 2.0.  Who knows what that means, but there's
some hope...

>In the Python sources, in the Modules directory there is a module for
>ncurses that can be built.

That's cool.  It would be nice to have a standard Python text mode
interface (wxPython would be the best imaginable ;-).

>Robin Dunn

-- 
-William "Billy" Tanksley



From kuncej at mail.conservation.state.mo.us  Wed Sep  8 12:02:50 1999
From: kuncej at mail.conservation.state.mo.us (Jeffrey Kunce)
Date: Wed, 08 Sep 1999 11:02:50 -0500
Subject: Python interface to Groupwise mailoxes?
Message-ID: 

>Is there an existing python interface to Novell'sGroupwise mailboxes?
>Failing an existing python module, are there ANY open-source programs that
>interface with Groupwise that I can use as a starting point...

Yep. Groupwise is wide open to python on windows using COM.
In fact, my "project of the week" has been to read the groupwise
calendars of a group of employees, and to generate html pages
that display everyone's appointments on a particular day.

You need to get the "Groupwise 5.x Object API" from Novell.
(Go to developer.novell.com, sign up for a free online membership to 
the Novell Developer Network, find the file and download it.)

To make it work, you need these things installed:
  -Python 
  -Python Win32 Extensions
  -GroupWise client

I've attached some simple/rough demo code to get you started.

  --Jeff


-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: gwcomdemo.py
URL: 

From skip at mojam.com  Wed Sep 15 09:43:29 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 15 Sep 1999 08:43:29 -0500 (CDT)
Subject: Boiler plate method for Module
In-Reply-To: <7rn9am$87t$1@news.is.co.za>
References: <7rn9am$87t$1@news.is.co.za>
Message-ID: <14303.41601.531563.688130@dolphin.mojam.com>

    Jacques> Hi am looking for a good but simple example of how to write my
    Jacques> own Python Object in C

    Jacques> 1) Must show me how to write the constructor/destructor
    Jacques> 2) Sow me how to do getattr/setattr pair

    Jacques> PLEASE DO NOT TELL ME TO LOOK AT TO SOURCE OF PYTHON, I HAVE
    Jacques> DONE THAT I JUST WANT TO SEE SIMPLE LAYOUT.

What source did you look at?  There are hardly any better examples of what
you are looking for than the actual, functioning source code.  Here are the
places I recommend you look:

1. The Extending and Embedding Manual - find a pointer at
   http://www.python.org/doc/. 

2. Modules/xxmodule.c in the source distribution - a skeleton module that
   outlines all the necessary bits and pieces.  There's even a stub xxmodule 
   line in Modules/Setup.

3. Any of a number of actual modules in the Modules directory:

   * dbmmodule.c is a pretty simple example of a module that creates an
     object and has to have allocation/deallocation methods, and to define
     both module methods and object methods.

    * soundex.c is a very simple example of a module that doesn't need to
      create any objects.  It just wraps a simple algorithm.

If you can't fathom the contents of the documentation or the source code to
simple modules, you're probably better off looking for a consultant to hire.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From wware-nospam at world.std.com  Thu Sep  2 10:02:40 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Thu, 2 Sep 1999 14:02:40 GMT
Subject: Turing Compliant?
References: <37CD7CBC.54AD295F@megsinet.net>
Message-ID: 

David Oppenheimer (davidopp at megsinet.net) wrote:
: What the heck does Turing Compliant mean?  I've heard discussion that
: Python is not Turing Compliant.  Is this true and why would this be an
: important consideration for someone who is programming in Python?

Many people have already discussed "Turing Complete". The meaning of
"Turing Compliant" is that Python will remain Turing Complete after
January 1st, 2000. "Turing Delectable" means that Python will also
remain Turing Complete after September 9th, 9999. The Posix module
should break by then, but otherwise, I believe Python is Turing
Delectable.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From emile at fenx.com  Sat Sep  4 10:40:50 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 4 Sep 1999 07:40:50 -0700
Subject: Spread/Worksheet
References: <37CA8EAB.5DD1314@uniserve.com> <7qeks8$pd5$1@news4.isdnet.net> <37CEC2AB.B72137AB@uniserve.com>
Message-ID: 

http://alldunn.com/wxPython/

--

Emile van Sebille
emile at fenx.com
-------------------





From nielsenjf at my-deja.com  Tue Sep  7 12:46:41 1999
From: nielsenjf at my-deja.com (nielsenjf at my-deja.com)
Date: Tue, 07 Sep 1999 16:46:41 GMT
Subject: case for open lang on win32 - fill in the table
References: <7r0049$bfh$1@nnrp1.deja.com> <7r0d3t$chk$1@newshost.accu.uu.nl> <7r1hab$9bd$1@m2.c2.telstra-mm.net.au>
Message-ID: <7r3fhd$oqb$1@nnrp1.deja.com>

I've decided to expand the table with comments and code examples
including compiled languages.
(though I'll probably need help with the vb examples :-)

I hope to show that you can use open languages rather than
vbscript or visual basic and have just as easy or easier time
doing it that using proprietary stuff with the benefits of more
power and cross-platform (Where I work we're going 100% microsoft and
100% visual basic). The main argument being visual basic/vbscript is so
easy to use and the only other viable option is C++ which is too hard
(in their opinion).

IMHO, python has a feel similar vb/vbscript except it doesn't have
the vb world's limitations.

Now of course, reality may not match my ideas that (maybe you really
have to use vbscript and visual basic to function the best in their
closed world). However, on first inspection it doesn't look that way.
There is actually very little supporting vbscript.

For example, you can do everything COMwise in C++.
Choose any other language and you start to make compromises.
What is there that you can do in C++ that you can't do in python?
In what cases is visual basic(a probably weaker language) better than
python to use?

> IDispatch plus about 100 extra interfaces.  We can not use raw
interfaces if
> there is no C++ support, but many useful interfaces already have that
> support (and it is trivial to add new ones).  From a marketting POV,
this
> should definately read "yes" :-)

When you say C++ support, I assume you mean there is code written than
allows vtable access for python for some interfaces?

Is there c++ support for ADSI or MTS or OLEDB (which look to be the big
things for windows 2000)? Or can we only access it via Idispatch(I know
perl can do some stuff w/ADSI via Idispatch).

OLDEDB access would be cool since ADO is so slow.
Perl has dbi to fix that problem.

> >> COM threading       STA only          STA only?    STA only?
>
> No - full support for all COM threading models.  STA is the default.

Great. I think even visual basic is stuck in the STA.

> >> project size        small to medium   big          big
> >> (lines of code)
> >
> >What does that mean, lines of code? for COM, or in general?
>
> Agreed = I would have to argue "small", especially compared to VB.
For
> _anything_ other than GUI code (which VBScript doesnt do anyway) the
LOC for
> a Python solution will always be smaller than for VB.

I guess that was ambigous. It should be broken up into 2 things.

Maximum size of project you can use it with:
It looks like python projects can get huge.

Amount of code needed to solve a problem:
(for example in the recent Practice of Programming book, perl, java,C,
c++ and others were compared, with perl being the fastest except for
raw C code, but it only needed a fraction of the code. Java was slowest
and needed a lot of code).


> Hope this helps...  Let us know if you complete this and post it to
the web.

Will do, thanks a lot for your help.

john


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kernr at mail.ncifcrf.gov  Thu Sep  9 14:16:53 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Thu, 09 Sep 1999 18:16:53 GMT
Subject: [Newbie:]Importing modules from different files
References: <7r8dl0$cao$1@nnrp1.deja.com>
Message-ID: <37d7f6b7.1849692@news.erols.com>

On Thu, 09 Sep 1999 13:45:08 GMT, gtnorton at my-deja.com wrote:

>                           Hello folks,
>                                      Up until now, I've been putting
>all of my modules in my main Python folder.But now it grows too large to
>find quickly.When I create a folder(still in the main Python file) for
>the scripts and try to import them with: >>>import spam    I get back
>"no module named spam".Please clue me in.
>                                Thanks in advance
>                                              gtnorton
>(P.S.)Running Windows98

I the long standing tradition of this newgroup to post as many
possible, different solutions to a problem (who says there's "only one
way"?), here is what I have done:

Say I have a subdirectory 'Stuff' that has most of my non-package
scripts (like 'compile.py', 'ipp.py', 'codeplot.py', etc.).  I place a
text file in the main Python folder ('C:\Python' for me) called
'Stuff.pth'.  The name is irrelevant, but the '.pth' extension is
necessary.  The content of this file is the single word 'Stuff'.

When the interpreter searches the PYTHONPATH and it comes across such
a file, it will add the directory specified in the file to the search
path.  In general, the content of a '.pth' file can be any directory
specification either relative or absolute.  Therefore, if you have
scripts in a place like 'C:\Even\More\Python\Scripts', the '.pth' file
would contain '/Even/More/Python/Scripts' (following the UNIX path
separator convention because it's easier.  It works on Windows fine).

I know this is documented somewhere, but I don't know where.

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From gmcm at hypernet.com  Wed Sep  8 14:06:08 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 8 Sep 1999 13:06:08 -0500
Subject: Bug in time module?
In-Reply-To: <37D68CB2.BB5F9CC9@pc.jaring.my>
Message-ID: <1275312392-99250951@hypernet.com>

lec wrote:

> Gordon McMillan wrote:
> 
> >
> > The only editing Python's time module does is to ensure that you've
> > passed the right number and kind of args. Everything else is the
> > result of your platform's implementation of strptime. If you want
> > something _good_, you should probably use MA Lemburg's mxDateTime.
> >
> > - Gordon
> 
> Consider the following exception raised:
> 
> Python 1.5.2 (#1, Apr 18 1999, 16:03:16)  [GCC pgcc-2.91.60 19981201
> (egcs-1.1.1  on linux2 Copyright 1991-1995 Stichting Mathematisch
> Centrum, Amsterdam
> >>> import time
> >>> time.strptime('31/13/1999', "%d/%m/%Y")
> Traceback (innermost last):
>   File "", line 1, in ?
> ValueError: format mismatch
> 
> Shouldn't the same exception be raised for
> time.strptime('31/02/1999', "%d/%m/%Y")  ?

"Should / shouldn't" is a very different question from "why".

Consider the following:

static PyObject *
time_strptime(self, args)
 PyObject *self;
 PyObject *args;
{
 struct tm tm;
 char *fmt = "%a %b %d %H:%M:%S %Y";
 char *buf;
 char *s;

 if (!PyArg_ParseTuple(args, "s|s", &buf, &fmt)) {
  PyErr_SetString(PyExc_ValueError, "invalid argument");
  return NULL;
 }
 memset((ANY *) &tm, '\0', sizeof(tm));
 s = strptime(buf, fmt, &tm);
 if (s == NULL) {
  PyErr_SetString(PyExc_ValueError, "format mismatch");
  return NULL;
 }
 while (*s && isspace(*s))
  s++;
 if (*s) {
  PyErr_Format(PyExc_ValueError,
        "unconverted data remains: '%.400s'", s);
  return NULL;
 }
 return tmtotuple(&tm);

Do you see any Python edits of values there? 

Consider the following:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam >>>
import time >>> time.strptime('31/13/1999', '%d/%m/%Y') Traceback
(innermost last):
  File "", line 1, in ?
AttributeError: strptime

No strptime in Windows c runtime.

I leave the "should" question untouched.

- Gordon



From evan at tokenexchange.com  Tue Sep 21 16:30:34 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Tue, 21 Sep 1999 15:30:34 -0500
Subject: global in nested functions (viper)
References: <37E7AF7B.8DEC7CC8@maxtal.com.au> <5l67149gwc.fsf@eric.cnri.reston.va.us>
Message-ID: <9XRF3.2$Ae3.15469@news-dal.corridex.com>

This sounds relevent to one of my many proposals to make work for other
folks .  Perhaps instead of causing a reference to the
(grand-)*parent's locals, this could simply copy the object into a local
variable at definition time?

Example:

>>> def f1(x):
...  y = x + 2
...  def f2(x):
...    global y
...    print x+y
...  y = 0
...  return f2
...
>>> f = f1(3)
>>> del f1
>>> f(5) # the behavior below would print 5, and force f1's entire locals()
to stay alive.
10


Guido van Rossum  wrote in message
news:5l67149gwc.fsf at eric.cnri.reston.va.us...
> skaller  writes:
>
> > The following file and viperi run demonstrate the current handling
> > of the globals statement in viper.  What happens is: global causes
> > a variable to refer to the containing context, often a module,
> > but in this demo, there is also a nested function, global then refers
> > to the locals() of the containing function, unless the variable
> > is _also_ declared 'global' in that function.
>
> Hm, this looks elegant, although I also see the elegance of
> disallowing assignments into the parent function altogether (i.e., you
> can *read* your parent's locals, but you can't *write* them).  So I'm
> undecided.
>
> --Guido van Rossum (home page: http://www.python.org/~guido/)





From fredrik at pythonware.com  Tue Sep 28 08:23:06 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 28 Sep 1999 12:23:06 GMT
Subject: File recursion, info extraction
References: <37F0AD76.911900AE@bibsyst.no>
Message-ID: <007801bf09ac$496dc9e0$f29b12c2@secret.pythonware.com>

Thomas Weholt  wrote:
> I need a way of handling ( get size, possibly type, modification time,
> path, filename, possible extension ) all files in a directory and all
> sub-folders. 
> 
> Furtermore, I want to extract info on images, like size, resolution, and
> id-tag-info from mp3 files. I`ve seen a module for that could at least
> set id-tag info on mp3-files. Can it read them too? The 
> PIL-module/project, can that take care of the "getting the info from the
> images"-thing?

the following script implements everything
except the MP3 ID tag stuff.  it requires PIL
and PST, available from:

    http://www.pythonware.com/products/pil/
    http://www.pythonware.com/products/pst/

it's probably fairly easy to add ID support to
PST's MpegSoundPlugin module, but I haven't
gotten around to do that yet (please send me
the patches if you implement that).

see the library docs for more info on what
exactly os.stat returns.



...

import os
import time

import Image # PIL
import Sound # PST

def check(userdata, directory, files):

    for file in files:

        filename = os.path.join(directory, file)

        # check if image
        try:
            im = Image.open(filename)
        except IOError:
            pass
        else:
            print filename, im.format, im.size, im.mode, 
            return

        # check if sound
        try:
            sn = Sound.open(filename)
        except IOError:
            pass
        else:
            print filename, sn.format, sn.mode, sn.rate, sn.channels
            return

        return

        # ordinary file
        try:
            st = os.stat(filename)
        except os.error:
            pass
        else:
            print filename, st # see docs for details
            return

        print "cannot process", filename

os.path.walk("/", check, None)

...




From uche.ogbuji at fourthought.com  Sun Sep 26 09:51:52 1999
From: uche.ogbuji at fourthought.com (Uche Ogbuji)
Date: Sun, 26 Sep 1999 09:51:52 -0400
Subject: Turing Compliant?
References: <1275048323-10459679@hypernet.com>
Message-ID: <37EE24F8.245CE626@fourthought.com>

Gordon McMillan wrote:
> 
> Bryan VanDeVen wrote:
> 
> > Gordon McMillan wrote:
> ...
> > > Well there's biological computation, in which each fork() grows it's
> > > own "CPU", thus making some formerly non-computable problems
> > > computable.
> >
> > I have heard of some experiments with biological computing - most
> > taking advantange of the much higher information density that can be
> > achieved using DNA, for instance, but I am very skeptical of the claim
> > that this adds something fundamentally new.  Any pointers to a proof?
> 
> No experiments, just theory. Grow a new "cell" for each
> decomposition of the problem. With a bit of overhead, you're getting
> close to solving problems of exponential complexity in linear time.

I'm pretty sure I read about some actual experiments in bio-computing in
The Economist.  True, this is not exactly Nature, but they are pretty
reliable for combing the major peer-reviewed journals and exctracting
interesting and accurate material.

I don't keep back-issues of The Economist, but I do remember that there
was some twist to the sorts of DNA with which they'd had success,
Mitochondrial, or perhaps messenger-RNA?

But then again, I have only had a sip of the Pierian, so I'm dangerous.

-- 
Uche Ogbuji
FourThought LLC, IT Consultants
uche.ogbuji at fourthought.com	(970)481-0805
Software engineering, project management, Intranets and Extranets
http://FourThought.com		http://OpenTechnology.org



From hoel at GermanLloyd.org  Thu Sep 16 08:40:15 1999
From: hoel at GermanLloyd.org (Berthold Hoellmann)
Date: Thu, 16 Sep 1999 14:40:15 +0200
Subject: pyfort2 question
References: <7rqnf5$kot$1@bignews.shef.ac.uk>
Message-ID: <37E0E52F.88CBB29A@GermanLloyd.org>

Jose L Gomez Dans wrote:
> 
> Hi!
>         I've just downloaded pyfort2, and I'm trying to convert a
> subroutine. Everything seems to work fine, and I get my module.so in
> the end. When I try "from  import *", python complains that there is
> an udefined symbol G77_rand_0. I suppose it is related to some random
> routines I use in my fortran code.
> 
>         What can I do to get over this problem? I'm running linux, with g77
> and ld, and python 1.5.2, if that's of any help.

In your Setup file just add -lg2c to your compile flags. In general the
documentation should mention, that you have to link the Fortran runtime
library with any compiler.

Cheers

Berthold
-- 
email: hoel at GermanLloyd.org
   )
  (
C[_]  These opinions might be mine, but never those of my employer.



From dragondm at polar.integral.org  Wed Sep 29 08:17:08 1999
From: dragondm at polar.integral.org (The Dragon De Monsyne)
Date: 29 Sep 1999 12:17:08 GMT
Subject: Intelligent datastructure?
References: 
Message-ID: <7st004$r9s$1@utah.laird.net>

In article ,
	mlh at idt.ntnu.no (Magnus L. Hetland) writes:
> 
> Hi!
> 
> I'm making a stripped-down language for teaching algorithms (and
> datastructures) without requiring much knowledge of a specific
> language - (it might end up being a stripped-down version of
> Python...)
> 
> I only have two datatypes - integers and sequences. The thing I have
> been thinking about lately is the implementation of the sequences. I
> want them to be usable both as simple lists or arrays, and as general
> hash-tables (with numeric keys) without the user having to care about
> the difference... Thus, you could make yourself a sequence like this:
> 
>   s = [1,2,3]
> 
> and then do something like
> 
>   s[100] = 4
> 
> without getting any complaints. (I think I will only allow positive
> integers as indices, but I'm not sure...)

[ --- 8< --- ]

	Basically what you are looking for is a 'sparse array'
(basically an array (list) that allows unfilled 'holes' in it, so you can 
add list[x] = 42 without needing to actually store x elements. 

	My suggestion is this, if you implement this in Python, just use a
dictionary. Python dictionaries are hashtables, which are pretty fast and 
efficient. 

I timed adding 10k integers to a list, and to an integer-keyed 
dictionary. The list took 2.25 sec, the dict took 0.5 sec. 
(caveat benchmark, YMMV)

--
#############
# I 
# Really,   
# Really,   
# Really,   
# Really,   
# Really,   
# Really,   
# Really,   
# Really,   
# __Really__  hate newservers that have bogus limits on quoted text. :P
############# 

-- 
	-The Dragon De Monsyne



From balance_is_healthy at my-deja.com  Tue Sep 28 06:03:35 1999
From: balance_is_healthy at my-deja.com (balance_is_healthy at my-deja.com)
Date: Tue, 28 Sep 1999 10:03:35 GMT
Subject: where did www.python.org go?
Message-ID: <7sq3pk$sbg$1@nnrp1.deja.com>

Hi I?m a real newcommer (haven?t posted before)
But I?m really insterested in programming.
Can anyone please tell me where www.python.org
went? Have they moved ?

All advices on pythonresources are appreciated.
Though please don?t send any files. If I posted on
the wrong place please have mercey.

Thanks wizzes!

Henrik


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



From Denys.Duchier at ps.uni-sb.de  Fri Sep 10 17:39:08 1999
From: Denys.Duchier at ps.uni-sb.de (Denys Duchier)
Date: 10 Sep 1999 23:39:08 +0200
Subject: Using indentation for blocking
References: <7raqm7$5ag$1@nnrp1.deja.com> <002f01befb88$5b930b80$f29b12c2@secret.pythonware.com>
Message-ID: 

"Fredrik Lundh"  writes:

>   "We will perhaps eventually be writing only small
>   modules which are identified by name as they are
>   used to build larger ones, so that devices like
>   indentation, rather than delimiters, might become
>   feasible for expressing local structure in the source
>   language."
> 
>   -- Donald E. Knuth, "Structured Programming with goto
>   Statements", Computing Surveys, Vol 6 No 4, Dec. 1974 

If memory serves, this (the offside-rule) was one of the many
visionary things proposed by Landin in his 1966 CACM paper "The next
700 programming languages".

While that idea is quite convenient for authoring programs manually,
it is a nuisance for automated processing, e.g. for macros (i.e. the
combination of textual program fragments).  In Haskell, you can use
indifferently a mix of indentation or braces and semicolons to convey
layout structure.  This is, in my opinion, a more reasonable
compromise.

-- 
Dr. Denys Duchier			Denys.Duchier at ps.uni-sb.de
Forschungsbereich Programmiersysteme	(Programming Systems Lab)
Universitaet des Saarlandes, Geb. 45	http://www.ps.uni-sb.de/~duchier
Postfach 15 11 50			Phone: +49 681 302 5618
66041 Saarbruecken, Germany		Fax:   +49 681 302 5615



From jacques at integra.co.za  Wed Sep 15 01:05:10 1999
From: jacques at integra.co.za (Jacques Oosthuizen)
Date: Wed, 15 Sep 1999 07:05:10 +0200
Subject: Boiler plate method for Module
Message-ID: <7rn9am$87t$1@news.is.co.za>

Hi am looking for a good but simple example of how to write my own Python
Object in C

1) Must show me how to write the constructor/destructor
2) Sow me how to do getattr/setattr pair


PLEASE DO NOT TELL ME TO LOOK AT TO SOURCE OF PYTHON, I HAVE DONE THAT
I JUST WANT TO SEE SIMPLE LAYOUT.





From bobklin at idt.net  Wed Sep  8 19:54:30 1999
From: bobklin at idt.net (Bob Kline)
Date: Wed, 8 Sep 1999 19:54:30 -0400
Subject: Choosing a MySQL interface
References: 
Message-ID: <7r6t12$b3t@nnrp3.farm.idt.net>

jims  wrote in message
news:rtcnbgef1iv21 at corp.supernews.com...

> I'm a bit of a Python newbie, and am building an app
> that uses a MySQL database (the latest version) as the back
> end.

> I wonder if anyone can tell me whuich is the currently "correct"
> way to connect to a MySQL database using Python. There
> appear to be several methods available (like here and here)
> but I can't tell which one is the right one to use at this time.

The advantage of the first package to which you refer
(MySQLmodule) is that it appears to build properly on the
two platforms I've tried it on.  The advantage of the second
package (MySQLdb) is that it conforms to the new standard
API for database access using Python.  Unfortunately,
although I've succeded in getting the second package
working on Linux, the build fails on NT.  I posted a
message last week reporting the error messages I got
and asking what components are missing from the win32
Python distribution that are needed to make the error
messages go away, but no one seems to know the
answer.

I hope you have better luck finding knowledgable folks on
this list.  Of course, if your db client is on UNIX or Linux,
you may not have any problems.

Bob






From jetman3 at netzero.net  Thu Sep 23 05:34:32 1999
From: jetman3 at netzero.net (The Jetman)
Date: Thu, 23 Sep 1999 05:34:32 -0400
Subject: [PyWin] Does Win32RAS Work ?
References: 
Message-ID: 

    Neglected to mention that the problem is happening under
Win95....Jet

--

================================================================
  In Paris, they simply stared when I spoke to them in French;
  I never did succeed in making those idiots understand their
  own language.         - Mark Twain, The Innocent Abroad,1869
=======================================  jetman3 at netzero.net  ==


The Jetman wrote in message ...
>
>    Have spent a couple of hours trying to narrow down a
>prob w/ Win32RAS and I'm at the point where I suspect it simply
>doesn't work.  Have trimmed down the rastest script to simply
>call win32ras.Dial().  It starts dialing, then hangs the
>PyWin process (starts beeping incessantly, until I do a
>three-finger salute.)  Any ideas ?
>
>--
>
>================================================================
>  In Paris, they simply stared when I spoke to them in French;
>  I never did succeed in making those idiots understand their
>  own language.         - Mark Twain, The Innocent Abroad,1869
>=======================================  jetman3 at netzero.net  ==
>
>
>
>





From warren at nightwares.com  Thu Sep  9 12:46:14 1999
From: warren at nightwares.com (Warren 'The Howdy Man' Ockrassa)
Date: Thu, 09 Sep 1999 11:46:14 -0500
Subject: And now for something completely boneheaded...
References: <37D623A2.DC09F0CC@nightwares.com>
Message-ID: <37D7E456.728C0BC1@nightwares.com>

OK, first a big thank you to those who answered my question in a patient
manner, and a big razzberry to those who did not. :)

So it seems clear to me now that what I need, more than anything else,
is a good language ref. I don't think I've seen one online, and the
_Programming Python_ book by Mark Lutz, while it contains some teriffic
content, is not, as Mark Lutz himself points out, a reference manual. :\

So. Anyone have suggestions? A language ref is pretty much precisely
what I need, I think; and while I am aware of other O'Reilly books on
the subject, I don't know for sure which would be more along the lines
of what I'm looking for.

Thanks once more, all!

--WthmO



From JACarson at addcoinc.com  Wed Sep 29 11:10:00 1999
From: JACarson at addcoinc.com (Carson, Jon A.)
Date: Wed, 29 Sep 1999 10:10:00 -0500
Subject: Newbie Stuff
Message-ID: <290CDF914950D3118D79006008BD51AF0B4F8E@rc.addcoinc.com>

Hi,

I have a "newbie" question for the list.

I would like to write a small aplication with python/tkinter that deals with
information to and from the serial port. this application would potentially
run on serveral different platforms...WinDoze, Mac, *nix...etc.

Question: How do I deal with ports in Python, and is the same across
platforms.

Thanks in Advance,
Jon c.

-----------------------------------------------
jacarson at addcoinc.com
                      ___..---'~~~`---..___
                  .-=========================-
  _______________/ :.::..-~--..___..---~~~'         
 (___________(_||_)____/                          
  /____/___:..::.:::. /      NCC-1701 D            
           \_________/        "Engage"             


-----------------------------------------------




From evb0 at gte.com  Fri Sep 24 11:12:13 1999
From: evb0 at gte.com (Evgeni V. Belin)
Date: Fri, 24 Sep 1999 11:12:13 -0400
Subject: color JPEGs
Message-ID: <199909241512.LAA01413@kuai.gte.com>

Hi Paul,

=> IOError: cannot write mode P as JPEG

I'd suggest to look at pilconvert.py that comes in the Scripts subdir.
Basically, you can use *drafts* to to convert images before saving in
different format:

>>> import Image
>>> im = Image.open('foo.gif')
>>> im.save('foo.jpg')
...
IOError: cannot write mode P as JPEG
>>> im.draft("RGB", im.size)
>>> im = im.convert("RGB")
>>> im.save('foo.jpg')
>>> 

Hope that helps, 
-Evgeni
------- Forwarded Message

Return-Path: python-list-admin at python.org
Delivery-Date: Fri Sep 24 10:47:55 1999
Received: from newman.gte.com (newman.gte.com [132.197.8.26])
	by kuai.gte.com (8.8.8/8.8.8) with ESMTP id KAA01363
	for ; Fri, 24 Sep 1999 10:47:51 -0400 (EDT)
Received: from cliff2.gte.com (cliff2.gte.com [132.197.8.43])
	by newman.gte.com (8.9.1/8.9.1) with ESMTP id KAA11552
	for ; Fri, 24 Sep 1999 10:47:50 -0400 (EDT)
Received: from relay.gte.com (relay.gte.com [132.197.248.10])
	by cliff2.gte.com (8.9.1/8.9.1) with ESMTP id KAA25962
	for ; Fri, 24 Sep 1999 10:47:50 -0400 (EDT)
Received: from python.org ([132.151.1.90])
	by relay.gte.com (8.9.1/8.9.1) with ESMTP id KAA09793
	for ; Fri, 24 Sep 1999 10:47:49 -0400 (EDT)
Received: from python.org (localhost [127.0.0.1])
	by python.org (8.9.1a/8.9.1) with ESMTP id KAA13674;
	Fri, 24 Sep 1999 10:45:53 -0400 (EDT)
From: Paul Barrett 
Newsgroups: comp.lang.python
Subject: color JPEGs
Date: Fri, 24 Sep 1999 10:17:23 -0400
Organization: Space Telescope Science Institute
Lines: 20
Message-ID: <37EB87F2.87A9A19D at STScI.Edu>
NNTP-Posting-Host: dhcp-042.stsci.edu
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.6 [en] (X11; I; Linux 2.2.12 i586)
X-Accept-Language: en
Path: news!uunet!ffx.uu.net!pulsar.dimensional.com!dimensional.com!coop.net!ncar!noao!stsci.edu!not-for-mail
Xref: news comp.lang.python:72687
To: python-list at python.org
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-Mailman-Version: 1.0.2
Precedence: bulk
List-Id: General discussion list for the Python programming language 
X-BeenThere: python-list at python.org

I'm trying to create a color JPEG from Numeric data.  I used Rob Hooft's
message from earlier this year to get me started.  All of his code
appears to work fine.  The problem appears when I try to save the image,
i.e. img.save('filename.jpg'). I get  the following error message:

IOError: cannot write mode P as JPEG

Can a PIL guru tell me what I'm doing wrong and how to proceed? I've
attached the relevant python code to help.

img = Image.Palette('P', (data.shape[1], data.shape[0]),
data.tostring(), 'raw', 'P')
img.palette = ImagePalette.raw('RGB', string.join(palette, ''))
img.save('filename')

- -- Paul





- -- 
http://www.python.org/mailman/listinfo/python-list

------- End of Forwarded Message




From jam at quark.emich.edu  Thu Sep 30 00:15:47 1999
From: jam at quark.emich.edu (Jeff)
Date: Thu, 30 Sep 1999 00:15:47 -0400
Subject: Python TimeServer (RFC 868)
In-Reply-To: 
References:  <19990929000705.A21607@quark.emich.edu> 
Message-ID: <19990930001547.B22550@quark.emich.edu>

On Thu, Sep 30, 1999 at 01:57:13AM +0000, Benjamin Schollnick wrote:
> Jeff,
>  
[..snipped..]
> 
> 		- Benjamin

greetings.

I poked around a little, and came up with some code, which I have attached.
please let me know if it works for you-- I haven't tested it other than to
telnet to the port, and it seems to work.. I've got no idea if it will work
under NT or not (though I can't think of any particular reason why it
wouldn't), but use it for what it's worth.

-- begin code --
#!/usr/bin/env python

import socket
import struct

def timeserver_calculation():
	import time
	
	return time.time()

def server(host="", port=3737, backlog=5):
	sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
	sock.bind(host, port)
	sock.listen(backlog)
	print "listening on port %s (%s, %d)" % (port, host, backlog)
	while 1:
		conn, addr = sock.accept()
		print "connected by %s at %s" % (conn, addr)
		ft = timeserver_calculation()
		conn.send(struct.pack("!i", ft))
		conn.close()

if __name__ == "__main__":
	server("localhost", 3737)
-- end code --

-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From bartek69 at kki.net.pl  Sun Sep 26 03:49:46 1999
From: bartek69 at kki.net.pl (Bartek Nicka)
Date: Sun, 26 Sep 1999 09:49:46 +0200
Subject: wxPython problems
Message-ID: <7sjl5n$4gc$1@sunsite.icm.edu.pl>

    I got strange problem with wxPython on RH6.0. I've written the small
application and each time I run it the some texts ( eg. buttons' captions
and labels )  look  in other way. I mean that the font's size and,style
changes each time and I have no idea why it happens.

Thanks in advance for help and sorry for my bad English.

bartekn at jota.ps.pl





From madq968 at djeksta.comNOSPAM  Sun Sep 26 14:01:40 1999
From: madq968 at djeksta.comNOSPAM (madQ)
Date: 26 Sep 1999 18:01:40 GMT
Subject: Download Ia.n.i.!!! It's free!
Message-ID: <7sln24$7l9$501@nslave1.tin.it>

Download Ia.n.i. RemoteControlSystem 1.2 beta. It's free!!!
New site: http://jump.to/IaniProject




From aahz at netcom.com  Wed Sep 15 16:20:43 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Sep 1999 20:20:43 GMT
Subject: Checking for 'new' POP3 mail
References: 
Message-ID: <7rov2r$h6r@dfw-ixnews16.ix.netcom.com>

In article ,
Python Addict  wrote:
>
>Messages come and go so just purely checking on the number of messages in
>the spool isn't going to work. Our POP server supports the optional UIDL
>command which returns the message number, a space, and then a unique
>identifier for each message.

Bingo.

>Am I making things complex for no reason or is there some built in
>functionality I'm missing from the POP spec? Is there a Python-ish way of
>comparing two lists that I might not be aware of? I was going to start at
>the beginning of the second list, waddle through it, and search for each
>element in the first list.

Use a dict instead of a list.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From mjackson at wc.eso.mc.xerox.com  Tue Sep  7 13:21:15 1999
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 7 Sep 1999 17:21:15 GMT
Subject: 1.5.2 NT install question(s)
Message-ID: <7r3hib$24f$1@news.wrc.xerox.com>

I've done this twice recently and encountered problems both times.

On a fresh system (never had Python or Tcl/Tk) things (for example,
PySol) seem to work, but whenever I use Tkinter (specifically, when
_tkinter is imported) I get a warning box about not being able to find
tcl80.dll in the specified path.  (tcl80.dll is present, in Program
Files\Tcl\bin.) What's wrong, and how do I put it right?

On a system where 1.5.1 (with Tcl/Tk) had previously been installed, I
carefully uninstalled both before installing 1.5.2.  Here both "import
_tkinter" and "inport Tkinter" are OK, but Tkinter._test() yields:

    TclError: Can't find a usable init.tcl in the following directories:
	{} ./lib/tcl8.0 D:/tc18.0/library {D:/Program Files/library}
    
    This probably means that Tcl wasn't installed properly.

(init.tcl is present, in Program Files\Tcl\lib\tcl8.0.)  What's wrong,
and how do I put it right?

Thanks in advance.  Sorry to be asking such trivial-seeming questions,
but I found nothing helpful through python.org.  And I haven't dealt
with the innards of NT enough for its design coherence to become
manifest (yeah, right).

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	Spend the optimal amount of time on each decision
	and pretty soon you run out of life.
				- Steven Waldman





From laheadle at boguscs.uchicago.edu  Thu Sep 30 04:18:35 1999
From: laheadle at boguscs.uchicago.edu (Lyn A Headley)
Date: 30 Sep 1999 03:18:35 -0500
Subject: Nolan's question of the day: distributed servers?
References: <87btal3td5.fsf@ethereal.dhis.org>
Message-ID: 

>>>>> "Nolan" == Nolan Darilek  writes:

    Nolan> I've kept up a good tradition (Maybe not, depending on your
    Nolan> point of view :) of asking one question per day. Here's
    Nolan> today's. :)

whew!  this one's a doozy.

    Nolan> So, a network of namespaces will exist, each accessing
    Nolan> either a database or a server. My current concern, though,

So do I get this right?  A main architectural goal for your mud server
framework will be to allow servers access to objects on other servers,
along with a protocol for discovering and invoking services offered by
those objects.  

The hairy part (as you've mentioned below) is where do the _effects_
of those services take place?  If you are hoping to invoke an object
on one server and have it change the state of another server, good luck.
That sounds pretty tough to me.  They will need to share knowledge
about the state of each other in a very intimate way, which would be
especially hard if the servers have different coders.

I do believe corba will do what you want (not that I've DONE it or
anything ;-) They call it dynamic method invocation or something
equally obvious (also look at interface repositories).

However, if you're serious about this I'd advise you to look at Erlang.

www.erlang.org

-- 
Lyn Headley
remove the word "bogus" from my address for the real one.



From mnot at mnot.net  Wed Sep  8 19:07:57 1999
From: mnot at mnot.net (Mark Nottingham)
Date: Thu, 9 Sep 1999 09:07:57 +1000
Subject: Python example: possible speedup?
References: <87puztv2gt.fsf@pc-hrvoje.srce.hr>
Message-ID: <00a801befa4e$fd89c450$0301a8c0@mnot.net>

Have you tried profiling it? I.e.,

[...]

if __name__ == '__main__':
    import profile
    profile.run('main()')






From robin at jessikat.demon.co.uk  Wed Sep 15 05:00:44 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 15 Sep 1999 09:00:44 GMT
Subject: high school programming & python
In-Reply-To: 
References: 
 
  <937300939snz@vision25.demon.co.uk>
 <$JL+lCAFvt33Ewr8@jessikat.demon.co.uk>
 
Message-ID: 

In message , Bill Janssen
 writes
>Excerpts from ext.python: 14-Sep-99 Re: high school programming.. Robin
>Becker at jessikat.de (951)
>
>> You were lucky to have plugboards. We had to key things in with a bunch
>> of switches.
>
>Hard little metal toggle switches, too -- none of these posh molded
>plastic rocker switches!
>
>Bill
>
yes indeed
-- 
Robin Becker



From phil at geog.ubc.ca  Mon Sep 27 14:54:56 1999
From: phil at geog.ubc.ca (Phil Austin)
Date: 27 Sep 1999 11:54:56 -0700
Subject: Moving from Perl to Python
References: 
Message-ID: 

jonathon  writes:

> On 25 Sep 1999, Harry George wrote:
> 
> 	Other interesting/useful language translations might be
> 	
> 	_Numerical Fortran_/_Numerical C_/_Numerical Pascal_  	
> 			<< Or has this been done?  >>

Thanks to Paul Dubois and collaborators, my guess is that the
difficulty threshold for wrapping Fortran code using either CXX or
pyfort (http://xfiles.llnl.gov/python.htm) has become so low that
reimplementation probably isn't worth it (expecially since netlib
usually contains a free Fortran/C/C++ free alternative that could be
substituted).  For those who haven't seen the new technology, here's a
small example: the complete Python/CXX wrapper for the NR routines fasper.f
(Section 13.8 of the 2nd edition) and gasdev.f:

#include "Python.h"
#include "CXX_Objects.h"
#include "CXX_Extensions.h"
#include "CXX_Array.h"
#include 
#include 
#include 

using namespace std;

extern "C" {
  void fasper_(float* x,float* y,int& n,float& ofac,float& hifac,float* wk1,
	       float* wk2,int& nwk,int& nout,int& jmax,float& prob);
  float gasdev_(int& idum);
}

static PyObject *
ex_fasper(PyObject* self, PyObject* args)
{
  Py_Initialize();
  import_array();
  PyImport_ImportModule("Numeric");
  Py::Tuple inargs(args);  
  Py::Array x(inargs[0]);
  assert(x.species()==PyArray_FLOAT);
  Py::Array y(inargs[1]);
  assert(y.species()==PyArray_FLOAT);
  int jint = Py::Int(inargs[2]);
  float ofacfloat = Py::Float(inargs[3]);
  float hifacfloat = Py::Float(inargs[4]);
  int MPint = Py::Int(inargs[5]);
  Py::Array wk1(MPint,PyArray_FLOAT);
  Py::Array wk2(MPint,PyArray_FLOAT);
  int nout;
  int jmax;
  float prob;
  fasper_( (float*) x.to_C(), (float*) y.to_C(), jint, ofacfloat,  
              hifacfloat ,(float*) wk1.to_C(),
	      (float*) wk2.to_C(), MPint ,nout,jmax,prob);
  Py::Dict output;
  output["wk1"]=wk1;
  output["wk2"]=wk2;
  output["nout"]=Py::Int(nout);
  output["jmax"]=Py::Int(jmax);
  output["prob"]=Py::Float(prob);
  return Py::new_reference_to(output);
}


static PyObject *
ex_gasdev(PyObject* self, PyObject* args)
{
  Py_Initialize();
  import_array();
  PyImport_ImportModule("Numeric");
  Py::Tuple inargs(args);  
  int idumint = Py::Int(inargs[0]);
  float ranNum=gasdev_(idumint);
  Py::Tuple output(2);
  output[0]=Py::Int(idumint);
  output[1]=Py::Float(ranNum);
  return Py::new_reference_to(output);
}


static Py::ExtensionModule* lombper;

extern "C" void initlombper();

void initlombper()
{
    // experimental initialization stuff
    lombper = new Py::ExtensionModule("lombper");
    lombper->add("fasper", ex_fasper, "calculate the Lomb periodogram");
    lombper->add("gasdev", ex_gasdev, "return a gaussian deviate");
    Py::Dict d = lombper->initialize();
}




From fredrik at pythonware.com  Thu Sep 23 04:55:38 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 23 Sep 1999 08:55:38 GMT
Subject: Postscript printing in Python
References: <37E98032.FA3B77F5@ans.com.au>
Message-ID: <014c01bf05a1$6b6507c0$f29b12c2@secret.pythonware.com>

Clement  wrote:
> I checked the libraries and other Python documentation and failed to
> find any help on printing to a postscript printer!

a few options:

there's PIDDLE:
http://www.strout.net/python/piddle/intro.html

if you have PIL installed, there's a very basic PostScript
driver in there:
http://www.pythonware.com/library/pil/handbook/psdraw.htm

and yes, Tkinter can print the contents of a canvas to a
PostScript printer, and Grail (http://grail.cnri.reston.va.us/grail/)
comes with an HTML to PostScript converter...






From davidopp at megsinet.net  Sat Sep 11 14:42:52 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Sat, 11 Sep 1999 14:42:52 -0400
Subject: Python or Java or "X" from me and my little job
References: <1275048881-10426114@hypernet.com>
Message-ID: <37DAA2AB.1ABFA8E9@megsinet.net>


Gordon McMillan wrote:

>
>
> > 2)Should I go with JAVA instead?
>
> Not unless you shave with a belt sander and use a gasoline-based
> aftershave.
>

Q:  Are trying to point out that Java isn't the right tool for the job or
are you implying that Java is for insane people or both?

David O.

"Any Sufficiently Advanced Technology Is Indistinguishable From Magic"

-Arthur C. Clarke




From gerrit.holl at pobox.com  Sat Sep 25 07:04:31 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 25 Sep 1999 13:04:31 +0200
Subject: Source line number for log purpose (newbie)
In-Reply-To: <7shiqh$d96$1@ssauraaa-i-1.production.compuserve.com>
References: <7shiqh$d96$1@ssauraaa-i-1.production.compuserve.com>
Message-ID: <19990925130431.B11328@optiplex.palga.uucp>

T. C. Mits:
> In c there was a way to get the current source line number via a macro,
> __LINE__.
> 
> I need to be able to do the same thing.  I created a logger class that uses
> a level setting.  At high levels I want to
> log the source line number where the inline log.prtl( level, msg) method
> call appears.
> 
> I'm sure this is easy but am getting lost digging through the Python source
> and docs.
> 

That's possible via the 'Exception' module.

regards,
Gerrit.
-- 
RTFM: Read That FreshMeat



From amiramir at my-deja.com  Mon Sep 27 02:31:10 1999
From: amiramir at my-deja.com (amiramir at my-deja.com)
Date: Mon, 27 Sep 1999 06:31:10 GMT
Subject: smtp and pop servers?
Message-ID: <7sn2vf$p8m$1@nnrp1.deja.com>

I can't seem to find any code for smtp and pop servers. Plenty of
client stuff, but no server.

Any pointers?

Thanks,
-A-



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



From nikolai at micon.no  Sun Sep  5 17:50:32 1999
From: nikolai at micon.no (Nikolai Kirsebom)
Date: Sun, 5 Sep 1999 23:50:32 +0200
Subject: Using Python in ASP scripts
References:  
Message-ID: 

I had a look at it (Aaron's description), but it appears to be information
intended for CGI-scripts.  I'm running (trying to) ASP scripts with Python
as the scripting language.  As I state in my first message - the pages runs
ok on one of my machines (actually only a backup server).  I have checked
(manually) the registries (for Python issues) and they are identical.  I'm
running Python 1.5.2b2 and PythonWin build 125.

Even the simplest python-code does not work.  As I wrote in my previous
message, I suspect there must be something wrong with a 'reference' to the
scripting language.  When I select the ASP page in the browser - it waits
endlessly (even if I in the server has configured a 90 second for ASP script
timeout).  I cannot see anything peculiar in the task-manager (processes
heavily using the cpu).

Does anyone know how I could 'trace' what happends when IIS receives the URL
referencing a ASP file.

The last resort will probably be to re-install everything on the machine and
upgrade to SP5.

Hope someone can help.

Nikolai



Bill Wilkinson wrote in message ...
>Have you read this bit?
>http://starship.python.net/crew/aaron_watters/pws.html
>
>It might help.
>






From da at ski.org  Fri Sep 24 16:51:54 1999
From: da at ski.org (David Ascher)
Date: Fri, 24 Sep 1999 13:51:54 -0700 (Pacific Daylight Time)
Subject: replacing the class of an instance
In-Reply-To: <99Sep24.123307pdt."3473"@watson.parc.xerox.com>
Message-ID: 

On Fri, 24 Sep 1999, Bill Janssen wrote:

> I'd like to be able to replace the class of previously created
> instances when I create a new version of the class.  Is it sufficient
> to keep track of all the instances, then go through and change the
> __class__ member of each instance when I change the class definition?
> 
> Or is there an even easier way?

Depending on your needs, you may be able to rip out the guts (aka the
methods and class attributes) of the class that the instances all share
and squeeze in the new class as a base class of the original class.  That
would avoid having to deal with circular references.

Something like (untested):

  orig_class.__dict__.clear()
  orig_class.__bases__ = (new_class, )

--david ascher




From bob at horvath.com  Wed Sep 22 15:55:20 1999
From: bob at horvath.com (Bob Horvath)
Date: Wed, 22 Sep 1999 19:55:20 +0000
Subject: wxPython on Debian Linux
References:  <37E8C20A.F4051E56@horvath.com> <87u2omg1yc.fsf@baal.winnegan.fake>
Message-ID: <37E93428.F1AAEC8C@horvath.com>

Siggy Brentrup wrote:

> Bob Horvath  writes:
>
> [...]
>
> > I haven't tried it myself, but itsn't it at
> >
> > http://www.debian.org/Packages/unstable/interpreters/python-wxwin.html
>
> This won't help if the original poster is running stable aka slink aka
> Debian 2.1 as I am doing. But you can always get the source package
> and build you own local .deb.
>

True, but if you aren't living on the edge, you're taking up way too much
space!




From m.faassen at vet.uu.nl  Sun Sep  5 13:40:27 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 1999 17:40:27 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com> <37CCE10D.D07D023B@strs.co.uk> <7qjgq8$k1i$1@newshost.accu.uu.nl>  <7qlj3b$p7u$1@newshost.accu.uu.nl>  <7qrce6$dgh$1@news.bayarea.net>
Message-ID: <7qu9ub$q02$1@newshost.accu.uu.nl>

Frank Mitchell  wrote:
> Stephan Houben wrote in message ...
>>m.faassen at vet.uu.nl (Martijn Faassen) writes:

Hm, confusing misattribution, I didn't write what's coming. You should've
snipped this. :) Stephan wrote it.

>>Of course, Sun marketed the fact that Java used bytecode as if it was
>>something revolutionary. "No, it is not interpreted, it is compiled to
>>Byte Code!" Of course, all those other languages which had been called
>>"interpreted" where actually compiled to byte code since the
>>beginning of time. Byte-code compilation is really as old as the Sun. ;-)

Regards,

Martijn




From ngps at madcap.dyn.ml.org  Tue Sep 14 09:07:53 1999
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 14 Sep 99 13:07:53 GMT
Subject: [ANNOUNCE] M2Crypto 0.02 - now does SSL
Message-ID: <37de48a9.0@news.cyberway.com.sg>

(Apologies if you've seen this. I didn't see the posting appear.
Apparently Usenet is not my ISP's long suit.)


I am pleased to announce the release of M2Crypto 0.02.
The package is available at http://www.post1.com/home/ngps

This release incorporates support for SSL, specifically,

	For building servers, blocking i/o:
		- An SSLServer modeled after SocketServer.
		- A ForkingSSLServer that seems to work well.
		- A ThreadingSSLServer that runs one thread at a time. (!) ;-)
		
	For building servers, nonblocking i/o:
		- An ssl_dispatcher modeled after asyncore.dispatcher.

	A HTTPS server based on Medusa.
	
	For client-side web programming:
		- httpslib
		- urllib2

Your feedback is appreciated.

Cheers.
-- 
Ng Pheng Siong 


-- 
Ng Pheng Siong 




From jpersson1 at yahoo.com  Tue Sep  7 09:48:18 1999
From: jpersson1 at yahoo.com (jpersson1 at yahoo.com)
Date: Tue, 07 Sep 1999 13:48:18 GMT
Subject: JPython and serialization
Message-ID: <7r352v$ggh$1@nnrp1.deja.com>

Hello

Is there anyone out there who has experience with using
JPython together with the Java serialization API.

I want to subclass a native Java class in JPython
and then serialize an instance of this class to
a file, but when I try to do it I get the following error:

Traceback (innermost last):
  File "ser/test.py", line 11, in ?
java.io.NotSerializableException: org.python.core.PyFile$TextWrapper
        at java.io.ObjectOutputStream.outputObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.writeObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.defaultWriteObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.outputObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.writeObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.defaultWriteObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.outputObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.writeObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.defaultWriteObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.outputObject
(ObjectOutputStream.java)
        at java.io.ObjectOutputStream.writeObject
(ObjectOutputStream.java)
        at org.python.core.PyReflectedFunction.__call__
(PyReflectedFunction.java:156)
        at org.python.core.PyMethod.__call__(PyMethod.java:40)
        at org.python.core.PyObject.__call__(PyObject.java:268)
        at org.python.core.PyInstance.invoke(PyInstance.java:272)
        at org.python.pycode._pyx0.f$0(test.py)
        at org.python.pycode._pyx0.call_function(test.py)
        at org.python.core.PyTableCode.call(PyTableCode.java:75)
        at org.python.core.Py.runCode(Py.java:935)
        at org.python.core.__builtin__.execfile(__builtin__.java:244)
        at org.python.core.__builtin__.execfile(__builtin__.java:248)
        at org.python.util.PythonInterpreter.execfile
(PythonInterpreter.java:132)
        at org.python.util.jpython.main(jpython.java:123)

java.io.NotSerializableException: java.io.NotSerializableException:
org.python.core.PyFile$TextWrapper

It seems that the JPython proxy class, that are created
on the fly when I am subclassing, contains references to
non serializable objects.

Can it be done or must I rethink my design?

--
Regards
  //Jan Persson (Syren Software, Sweden)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From battery841 at usa.net  Wed Sep 15 22:50:46 1999
From: battery841 at usa.net (battery841 at usa.net)
Date: Thu, 16 Sep 1999 02:50:46 GMT
Subject: Python Programmer needed
Message-ID: <7rplu3$gag$1@nnrp1.deja.com>

Hey,
	I am part of a new Linux site called Linuxfeed.  You can check us out
at www.linuxfeed.com.  Anyways, someone is programming a language which
is based off of corba called BLADE.  BLADE will be used eventually in
Linuxfeed.  We need someone to work help make bindings for BLADE.  If
you can help, please contact bob at thestuff.net .
Thanks
Kevin
PS. check us out at www.linuxfeed.com


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mhuster_AT at hotmail_DOT.com  Thu Sep 30 20:49:19 1999
From: mhuster_AT at hotmail_DOT.com (Michel Huster)
Date: Fri, 01 Oct 1999 00:49:19 GMT
Subject: How Can I do vi-like substitutions?
Message-ID: <7t10nc$97k$1@thoth.cts.com>

I want to do some complex replace inside python. I am used to vi power like:

s/\([0-9]*\) at \([0-9]*\)/\2 AT \1/

which will change
1234 at 5678
into
5678 AT 1234

I can only find re.sub(expr, repl, count) but this does not 'remember' the
saves matches between arguments.

Help!
?Ayudame!

Michael Huster
mhuster_AT at hotmail_DOT.com

(remove _AT and _DOT to reply)



From hniksic at srce.hr  Tue Sep 14 18:11:18 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 15 Sep 1999 00:11:18 +0200
Subject: Python and MySQL
Message-ID: <87d7vlcfeh.fsf@pc-hrvoje.srce.hr>

Do people use Python connected to MySQL?  If so, what client do you
use?  I've seen two, the MySQLdb, and the generic mxODBC.  The latter
seems better, but I was unable to compile the required infrastructure, 
i.e. the myodbc libraries.  What are the experiences of others?

I ask this because I will soon start working for a company that
connects Perl with Microsoft SQL server extensively.  (They connect
through, I think, Sybase ODBC drivers.  I could be wrong.)  I would
like to gradually convert their infrastructure to more maintainable
Python, so I'm using MySQL for practice.

If there are useful experiences or cautionary tales regarding mxODBC
and MSSQL/Sybase, I'd like to hear them too.



From randall_burns_hp at my-deja.com  Thu Sep  2 20:53:31 1999
From: randall_burns_hp at my-deja.com (randall_burns_hp at my-deja.com)
Date: Fri, 03 Sep 1999 00:53:31 GMT
Subject: How to sell Python to your boss?
References: 
Message-ID: <7qn667$e30$1@nnrp1.deja.com>

In article ,
  Petri Mikael Kuittinen 
wrote:
> I have lots of programming experience in C, and
>much less in Python. Despite this I am confident
>that I would be able to implement
> the prototype using Python many times faster
>than using C.
Well, I agree that you are representative of a
big constituency out there. We've been making
substantial use of Python at HP/Cupertino and our
experience so far indicates that Python can be
used to develop programs rapidly that can be
maintained easily.

So far, one of our biggest issues has been finding
developers enthusiastic about developing in
Python.  I've seen lots of folks who are rather
pessimistic about Python's future(my current boss
is not among them).

If anyone would like to talk to me about Python
opportunities at HP/Cupertino, I'd be happy
to speak with them(we are doing some substantial
e-commerce applications in Python).

> I can understand my bosses opinion. He wants to
protect the interest
> of this firm. If I would leave the firm, then
someone else would have
> tp spend time learning Python in order to
modify my program code.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
There is a lot to be said for sticking with
widely used languages. The problem is that
the mostly widely used languages may not be suited
for all areas.

>
> For several months I have tried to get my
>colleagues excited to learn Python, but so far
>nobody has been interested to learn new
>languages, except for some mild interest towards
>Java (which is IMHO an overly-hyped language).
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bsb at baal.infodrom.north.de  Thu Sep 23 20:28:01 1999
From: bsb at baal.infodrom.north.de (Siggy Brentrup)
Date: 24 Sep 1999 02:28:01 +0200
Subject: Becoming root within a Python script
In-Reply-To: =?ISO-8859-1?Q?Fran=E7ois?= Pinard's message of "23 Sep 1999
 18:31:53 -0400"
References: 
Message-ID: <87zoydrw4u.fsf@baal.winnegan.fake>

Fran?ois Pinard  writes:

> Hi, people.  I have a small problem, for which I see no elegant solution yet.
> Is there a common idiom I could use?
> 
> I would like, within a executing Python started as non-root, ask for the root
> password, and then, continue executing part of the script while being root,
> becoming the previous again after some sub-job is done.
> 
> Of course, I can kludge something in, but I would prefer something nice. :-)

What you are asking for requires suid root scripts, a really bad thing 
modern Unices don't allow any more.

You have to hack up a wrapper written a compiled language, which
starts the interpreter with root priveledge.

A while ago I did some work on such a beast using an md5 registry for
trusted scripts, but after rethinking I threw it away - just to big a
security risk.

CU
  Siggy

-- 
noch nichts Aufregendes:

Siggy Brentrup - bsb at baal.infodrom.north.de - voice: +49-441-6990134




From stephan at pcrm.win.tue.nl  Fri Sep 10 03:55:16 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 10 Sep 1999 07:55:16 GMT
Subject: Programmers Editor
References: <37D5916B.606E@zeusedit.com>  <00da01bef9e0$6e1b5c70$f29b12c2@secret.pythonware.com>  <3dg10p1b5j.fsf@amarok.cnri.reston.va.us>
Message-ID: 

On 08 Sep 1999 15:00:24 -0400, Andrew M. Kuchling 
 wrote:
>Anyway, if someone wanted to
>contribute info on editing Python with Zeus, KDevelop, or whatever, it
>would be very helpful.  Just follow up to this posting with notes on
>configuring your editor of choice.

JED is a very nice editor. It comes with Python support right out
of the box. I actually didn't need anything to configure to work
comfortably with it. (OK I chose another colour scheme in the menu.)

But make sure you get the latest version (0.98).
0.97 comes with some older Linux distributions (in my case, RH 6.0),
but is much less nice.

It is also much more light-weight then Emacs. But then again, there's
no kitchen-sink built-in.

Greetings,

Stephan



From JohnH at PHM.GOV.AU  Sun Sep 19 07:44:42 1999
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Sun, 19 Sep 1999 21:44:42 +1000
Subject: CygWin version
Message-ID: 

Sorry to appear so stupid, But if I open an MS-DOS window then install
Doskey it installs OK and works. But if I then start Python in the same DOS
window, Doskey does not seem to work within Python what is my problem.?

> 
> If your on 95 running doskey will give you command history.
> NT has the equivalent of doskey always on.
> 
> Bill
> 
> > 
> > Hi all
> > In the Python tutorial it says "The interactive editing and history
> > described here are optionally available in the Unix and 
> > CygWin versions of
> > the interpreter."
> > Is there a CygWin version of the interpreter and if so where 
> > can I get it.
> > If not is there another way to get interactive editing and 
> > history in the
> > command line interpreter.
> > Thanks in advance
> > 
> > Jonathan Hirsch
> > johnh at phm.gov.au
> > 
> > 
> > 
> 



From mkrisch at radiata.com  Wed Sep 22 03:32:27 1999
From: mkrisch at radiata.com (Mark Krischer)
Date: Wed, 22 Sep 1999 17:32:27 +1000
Subject: lists and append, and loop iteration
Message-ID: <37E8860B.C024C4F8@radiata.com>

this might be a newbie a question, but is there a reason why append
doesn't create a list if the list doesn't exist?

it seems to add a bit of complexity to have to do something like:

dict = {}

if 'new_key' not in dict.keys():
	dict['new_key'] = [new_value]
else:
	dict['new_key'].append(new_value)


it seems like i should just be able to say "list.append(new_value)" and
if list doesn't exist, 
i get "list[0] = [new_value]

am i missing something fundamental that this would completely break?


and while i'm bending your ear, let me ask a second question.

if i do:
	for element in list:
		do something with element

is there any to find out what my current index is? i've basically been
doing something like:
	list.index(element)

this happens to work fine for what i'm doing now, but only because i
don't care about doubles in the list, but should that ever be a problem,
how will i know which element i'm dealing with.

i figure i can always do:
	for i in range(len(list)):
		do something with list[i]

but it just feels like i should be able to things the other way as well.

thoughts?

thanks in advance.

--mk



From culliton at clark.net  Wed Sep  1 09:51:40 1999
From: culliton at clark.net (Tom Culliton)
Date: Wed, 01 Sep 1999 13:51:40 GMT
Subject: Pythom in embedded system?
References:  <37ccfed3.3577714@news.omnilink.de>
Message-ID: 

In article <37ccfed3.3577714 at news.omnilink.de>,
Stefan Franke  wrote:
>On 1 Sep 1999 07:09:28 GMT, vnesek at maja.zesoi.fer.hr (Vjekoslav NESEK) wrote:
>
>>Anyway if not python, can you point me to some other free 
>>interpeter suitable for embedding in such a limited system?

Forth is a common choice in embedded environments which require a
small footprint and good speed, but while the concepts are elegant the
reality is icky IMO.  TCL also has a very small footprint but is much
slower and thinks everything is string.  You may also want to look at
slang which has a forth like core and a more procedural syntax above
that, any of several tiny lisp/scheme interepters if you can handle
the parenthesis, an LPC derivative called Pike which looks promising
but doesn't say much about embedding, ...

>Have a look at Lua:
>  http://www.tecgraf.puc-rio.br/lua/
>
>Stefan

Lua seems to be small enough and designed to be embedded...



From jnc at ecs.soton.ac.uk  Wed Sep 29 06:00:11 1999
From: jnc at ecs.soton.ac.uk (John Carter)
Date: Wed, 29 Sep 1999 10:00:11 GMT
Subject: ImageTk.PhotoImage vs. Tkinter.PhotoImage
References: 
Message-ID: <37f1e251.6515442@news.ecs.soton.ac.uk>

On Mon, 27 Sep 1999 21:42:15 GMT, Les Schaffer 
wrote:

>i need to zoom in on a user specified subset of an image. 
>
>but i am confussed about ImageTk.PhotoImage and
>Tkinter.PhotoImage. the former creates an instance of the latter in
>its self.__photo, used for pasting. but i need access to
>Tkinter.Photoimage's copy and zoom commands. i dont understand why
>ImageTk.PhotoImage doesnt just extend Tkinter.PhotoImage so the copy
>and zoom methods are available, or delegate those calls to __photo
>methods.
>
>also, i notice that in Tkinter.PhotoImage the -from and -to keywords
>for the image copy methods are not explicitly defined. wonder why....
>
>this is my first real foray into using tkinter/PIL for some astro
>work, so please excuse if my questions are stupid.
>
>
>so, with code like this:
>
>	...
>        self.photo = ImageTk.PhotoImage(image)
>        self.canvas.create_image(0, 0, image=self.photo, anchor=NW)
>	...
>
>i want to create a new PhotoImage consisting of a zoom in on a subset
>of self.photo (i have the x,y coords of the desired zoom area in
>self.photo).

I've simply used the crop method, so I would do something like

image1 = image,crop((x1,y1,x2,y2))
self.photo = ImageTk.PhotoImage(image1)
self.canvas.create_image(0, 0, image=self.photo, anchor=NW)

Do all the image manipulation before I create an instance of Image.TK.

This worked well, except when I called a C extension, there was
clearly something I dodnt usnder stand about PIL's internal data
structures, as my C code accted on the whole origional image. This was
easily solved by using a copy after the crop.

John Carter

jnc at ecs.soton.ac.uk




>
>-- 
>____        Les Schaffer              ___| --->> Engineering R&D <<---
>Theoretical & Applied Mechanics          |  Designspring, Inc. 
>Center for Radiophysics & Space Research |  http://www.designspring.com/
>Cornell Univ.  schaffer at tam.cornell.edu  |  les at designspring.com




From paul at prescod.net  Sat Sep  4 19:11:03 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 4 Sep 1999 23:11:03 GMT
Subject: Turing Compliant?
References: <37CD7CBC.54AD295F@megsinet.net>
		 
Message-ID: <37D1A707.6CFD5159@prescod.net>

C.Laurence Gonsalves wrote:
> 
> Actually, there are programming languages that are not Turing complete.

I claim that Turing completeness is the only useful (or at least
quasi-useful) definition of programming language that we have. If we
accept that a programming language could be non-complete then how do we
fend off the Barbarians that call HTML a programming language?

 Paul Prescod



From gduzan at gte.com  Fri Sep  3 15:20:54 1999
From: gduzan at gte.com (Gary Duzan)
Date: Fri, 03 Sep 1999 15:20:54 -0400
Subject: Problems with threads
References: <936212879.1129421157@news.nikoma.de> <7qkh06$r2$1@towncrier.cc.monash.edu.au>
Message-ID: <37D01F96.DB020567@gte.com>

Jonathan Giddy wrote:
> 
> Dominic Hillenbrand  writes:
> >This code produces this type of error message:
> 
> >currentThread(): no current thread for 1025
> >currentThread(): no current thread for 2050
> >(.....)
> 
> >The program still seems to run correct!!!
> >But what is the reason for the error message?
> 
> It is an interaction between the builtin thread module and the threading
> module.  If you start your threads using the threading.Thread class, the
> message will not appear.
> 
> Essentially, this is a warning from threading.py that there is no Thread
> instance representing this thread.

   However, depending on your code, this may be a perfectly valid
state, and threading handles it reasonably well, but there is no
way to turn off the message short of editing threading.py, which
we did here. This should really be fixed in a future release.

					Gary Duzan
					GTE Laboratories



From tim_one at email.msn.com  Mon Sep 13 20:55:15 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 13 Sep 1999 20:55:15 -0400
Subject: error in module re?
In-Reply-To: <937225378.3989@mhs0096x.mchh.siemens.de>
Message-ID: <000501befe4b$cef02700$54a2143f@tim>

[Chris...]
>   Is this an implementation or documentation error?

Neither.

> >>> import re
> >>> re.sub('(a+)\s(b+)', '\2-\1', 'aaa bbb')
> '\002-\001'
> instead of
> 'bbb-aaa'
> The documentation says, that you can use backreferences in replacement
> string, but only \g works.

Read the cautions about using regular strings with regexps, then get in the
habit of using r-strings.  This does what you want:

>>> re.sub(r'(a+)\s(b+)', r'\2-\1', 'aaa bbb')
'bbb-aaa'
>>>

assuming-anybody-could-want-that-ly y'rs  - tim





From pinard at iro.umontreal.ca  Thu Sep  2 09:49:30 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 02 Sep 1999 09:49:30 -0400
Subject: Turing Compliant?
In-Reply-To: Stephan Houben's message of "02 Sep 1999 10:05:38 +0200"
References: <37CD7CBC.54AD295F@megsinet.net> 
Message-ID: 

Stephan Houben  writes:

> [...] it was a lame joke, and it confused people.

I might have missed the original message, maybe sent before I came in.
OK, I'll try to read any more article on this topic as extensions of the
original joke, which begins to be a real, heavy good one :-).

> # Lambda calculus interpreter in Python.

Thanks.  Who knows, I'll play with it one of these days! :-)

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From tre17 at student.canterbury.ac.nz  Thu Sep  9 02:47:24 1999
From: tre17 at student.canterbury.ac.nz (tre17 at student.canterbury.ac.nz)
Date: 09 Sep 1999 18:47:24 +1200
Subject: tkinter: identifying callback button?
References: <7r7anm$21a7$1@hub.org>
Message-ID: 

Doug at news.vex.net (Lewis) writes:

> I have a bunch of (named) buttons which call the same callback.
> I've been trying to figure out how (or if) I can get the callback
> function to be able to identify which one.
> 
> For example....
> 
>     for x in range(0,5):
>         b = Button(frame,text='#%s' % x,command=self.xxx, name='b%s' % x)
>         b.pack()
> 
> So one of the five buttons gets pressed and calls App.xxx(self) ...
> but can App.xxx figure out which button triggered the command? 
> 
> I could save the button references in a list, but don't know if that
> would help.

The easiest way of doing this is the create a new class that inherits
from `Button'.

==============================
class MyButton(Button):
    def __init__(self, name, *args, **kw):
        self.name = name
        apply(Button.__init__, (self,)+args, kw)
        self.configure(command=self.clicked)

    def clicked(self):
        some_command(self.name)
==============================

--
Tim Evans



From hannah at schlund.de  Mon Sep 13 09:32:47 1999
From: hannah at schlund.de (Hannah Schroeter)
Date: 13 Sep 1999 13:32:47 GMT
Subject: List Comprehensions Enhancement
References: <37D9E932.66F2B0D3@cosc.canterbury.ac.nz>
Message-ID: <7riudv$24g$1@c3po.schlund.de>

Hello!

In article <37D9E932.66F2B0D3 at cosc.canterbury.ac.nz>,
Greg Ewing   wrote:
>[...]

>print [3 * x for x in nums]
>print [x for x in nums if x > 2]
>print [(i, s) for i in nums for s in strs]
>print [(i, s) for i in nums for s in [f for f in strs if "n" in f]]

The syntax looks a bit verbose, compared to e.g. Haskell's list
comprehension syntax which looks quite similar to the mathematical set
comprehensions:

print [3 * x | x <- nums]
print [x | x <- nums, x > 2]
print [(i,s) | i <- nums, s <- strs]
print [(i,s) | i <- nums, s <- [f | f <- strs, 'n' `elem` f]]

Also, I think there should be a clear separator between the
generators/predicates in your syntax too:

print [(i,s) for i in nums, for s in strs]

or similar.

Nice thing, though :-)

Regards, Hannah.



From donn at u.washington.edu  Thu Sep 23 14:11:03 1999
From: donn at u.washington.edu (Donn Cave)
Date: 23 Sep 1999 18:11:03 GMT
Subject: SIGCHLD, fork, listen
References: <7sd6q4$gqq$1@salome.rfa.org>
Message-ID: <7sdqfn$m0o$1@nntp6.u.washington.edu>

Quoth bill :
| I'm trying to set up a fairly "simple"
| server that listens for a socket connection,
| forks off a process and goes back to listening.
| This works fine for multiple clients, except
| I get zombies.  I try redefining SIGCHLD as
| SIG_IGN, but this doesn't work (as apparently
| is what's expected under POSIX).  I then
| try redefining it with my own function that
| includes a WAIT.  Fine, except "listen"
| doesn't get signalled when the child process
| dies, only when the next socket connection
| occurs.  (which could be hours later, not
| good).

This sounds like your platform "restarts" the accept() system call
after an interrupt.  That's kind of a feature for C programmers,
because otherwise they have to check for EINTR after this kind of
system call and restart it themselves.  Python programmers do too,
for that matter, but your Python signal handlers don't execute in
a kind of parallel Python thread, they wait for the next chance at
the interpreter.  That chance is going to come when the EINTR hit -
unless the system call restarts on its own.

I don't like SIGCHLD in C programs, and it's even less useful in
Python.  In your shoes, I would probably just run waitpid() when
the opportunity arises in the accept/fork loop, and live with the
occasional zombie processes.

If you can't stand them, you can get rid of them with a double fork().
The forked child immediately forks again while its parent calls waitpid().
The first forked process exits, and the parent's waitpid() returns, right
away.  That leaves an orphaned process whose parent reverts to 1 ("init").

In C, on at least some UNIX platforms that support X/Open standard signal
operations, you can use sigaction(2) with SIGCHLD and a SA_NOCLDWAIT flag
to just dispense with the zombie system.  Haven't actually used this
myself, and I don't know how widely it's supported.  It's not an option
in the standard Python signal module.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From cjc26 at nospam.cornell.edu  Sat Sep 11 12:50:04 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 11 Sep 1999 16:50:04 GMT
Subject: Python-based IRC bots..and a #python channel
References: <37D879A1.EE4D10DE@op.net>
Message-ID: 

Pada Fri, 10 Sep 1999 03:24:05 GMT, Eron bilang:
| Hello. I'm looking to really get into python hacking, and I want to
| write a IRC bot in it. Does anyone know of any already existing IRC
| libraries and bots (well-documented ones, with LOTS of in-line comments
| to help me understand),  that I can use and extend, and then eventually
| use it as a model to build one from scratch. Also, is there an
| "official" #python channel somewhere? If there isn't, we should
| establish one. Any info would be greatly appreciated.

Hmm, I was going to wait to announce this, but...
I'm also working on a Python IRC framework, called "Perk"; see
 for details.  I
haven't actually written any bots in it yet, but it should be especially
suited for doing so; the core of the program is a file containing a list
of regular expressions, each with an associated block of Python code.
Each line of input from the server is compared to each regex, and when a
match is found, the corresponding code is executed.  So to make a bot
all you would need to do is modify/extend this file to watch for kick
messages, commands in privmsg's, etc.
The only caveats are, this version is PRE-alpha :), and there's not too
much documentation yet.


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From bthomas at trey-industries.com  Sat Sep 11 19:32:20 1999
From: bthomas at trey-industries.com (Ben Thomas)
Date: Sat, 11 Sep 1999 19:32:20 -0400
Subject: Python or Java or "X" from me and my little job
References: <37DA8D01.3D6616A@trey-industries.com>
Message-ID: <37DAE684.FE7A05F7@trey-industries.com>

Thanks to you all. Python it is. Using command call to other programs is
major coll idea. I did not know python could do that. I wll go forth and
"PYTHON" my jobs.

Is there a great place to get the availible functions list? I think I
need another Python book. I just read Learning Python, Lutz and Ascher,
I would like another book or download I could print that would list all
the functions I can call. Anyone know one?



expect me back soon for tech support :)
Ben





Ben wrote:
> 
> I am new to Linux. I have Python 1.5.2 up and working. I want to have
> Python grab some files, ZIP them, and FTP them offsite. I will let CRON
> start the job every night. Since I am just getting to look at Python, I
> wanted to ask to make sure I don't pick the wrong tool for the job.
> 
> 1)Am I selelcting a good tool to get light weight sys admin jobs done?
> 
> 2)Should I go with JAVA instead?
> 
> 3)Can Python ZIP files easily?
> 
> thankful for any comments;
> Ben



From ivnowa at hvision.nl  Fri Sep  3 02:37:23 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Fri, 3 Sep 1999 06:37:23 GMT
Subject: OH, I'm A Lumberjack And I'm OK...
In-Reply-To: <37CF55DC.816F38F5@megsinet.net>
References: <37CF55DC.816F38F5@megsinet.net>
Message-ID: <199909030637.IAA23513@axil.hvision.nl>

On 3 Sep 99, David Oppenheimer wrote:

> ... I play with Python all night and I sleep all day... lol
> 
> I received an e-mail message which told me that WIN ZIP sometimes deletes part
> of the extension and to try adding ".tar" to the file or placing a "." before
> the tar extension if it was missing.  This allowed me to extract the files
> Robert Kern was recommending, but still the PIL doesn't work.
> 
> Again I will relate some hopefully relevant info in the hope (getting less
> hopeful about this all the time) that someone will know what the $%@# is going
> on and how to fix it so I can run files from the PIL.
> 
> Computer: Packard Hell P-75
> OS: Windows 98
> Python Version: Python for Windows 1.5.2 with all of the win32all - 125
> extensions
> 
> PATHS     (yes, I took the suggestion to place my Python & related files in
> pathways that have no spaces)
> 
> Path to Python executable: C:\Python
> Path to Tcl executable: C:\Tcl\bin
> Path to PIL: C:\Python\PIL
> Path to Tkinter: C:\Python\Lib\lib-tk
> Path to Painter (a program that I am more than curious to see work, I am almost
> obsessed now): C:\Python\scripts\painter.py
> 
> My Traceback (error message):
> 
> >>> Traceback (innermost last):
>   File "C:\Python\Pythonwin\pywin\framework\scriptutils.py", line 237, in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\Python\scripts\painter.py", line 12, in ?
>     import Image, ImageTk
> ImportError: No module named Image
> 
> My Autoexec.bat settings:
> 
> PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\WINDOWS;
> set PATH=%PATH%C:\mingw32\bin;
> set PATH=%PATH%C:\PYTHON;C:\PYTHON\PIL;C:\TCL\BIN;C:\PYTHON\LIB\LIB-TK;

Did you set the PYTHONPATH environment variable? If so, what are its 
contents?

Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
"A smile, a song, and a compiler"



From rainer2207 at my-deja.com  Wed Sep 22 07:53:53 1999
From: rainer2207 at my-deja.com (rainer2207 at my-deja.com)
Date: Wed, 22 Sep 1999 11:53:53 GMT
Subject: Toplevel window size
Message-ID: <7sag0b$3io$1@nnrp1.deja.com>

How can I define a maximum toplevel window size? Thanks.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From davidopp at megsinet.net  Thu Sep  9 20:37:44 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 09 Sep 1999 20:37:44 -0400
Subject: Newbie needs help
References: <37D851C2.DEB53659@hotmail.com>
Message-ID: <37D852D7.772D981E@megsinet.net>

 http://www.python.org/doc/current/

Matt Lauer wrote:

> I am very new at this program stuff and i don't have a clue so if there
> is someone that can help me email me, (wpvnetx at hotmail.com)




From jam at quark.emich.edu  Wed Sep 29 18:17:16 1999
From: jam at quark.emich.edu (Jeff)
Date: Wed, 29 Sep 1999 18:17:16 -0400
Subject: Article on IBM developerWorks
In-Reply-To: 
References: 
Message-ID: <19990929181716.B22321@quark.emich.edu>

On Wed, Sep 29, 1999 at 05:38:28PM +0200, Henning Schroeder wrote:
> Hi!
> 
> Even IBM knows about Python :-) Check out
>    http://www-4.ibm.com/software/developer/library/python101.html
> on their "developerWorks Linux Library". 
> 
>   Henning

I read this article, and it is an even toned comparison between not only the
'other language that starts with "p"', but tcl and java as well. it should
be recommended reading for every phb. it is fairly short, and all around
is a good article.

regards,
J
-- 
|| visit gfd  
|| psa member #293  
|| New Image Systems & Services, Inc. 



From nathan at islanddata.com  Wed Sep 29 14:12:04 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Wed, 29 Sep 1999 11:12:04 -0700 (PDT)
Subject: keys and values lining up
In-Reply-To: <5l905p9a4c.fsf@eric.cnri.reston.va.us>
Message-ID: 

Hmmmm...not sure this worked quite the way you thought it would :)  Output
below.

>>> d = {'key1' : 'val1', 'key2' : 'val2', 'key3' : 'val3'}
>>> keys, values = transpose(d.items())
Traceback (innermost last):
  File "", line 1, in ?
ValueError: unpack sequence of wrong size
>>> transpose(d.items())
array([[[k, k, k],
        [v, v, v]],
       [[e, e, e],
        [a, a, a]],
       [[y, y, y],
        [l, l, l]],
       [[1, 2, 3],
        [1, 2, 3]]],'c')
>>> 

It does work when all of the keys and values are numbers, and actually
raises an exception with string keys and numeric values.  Eek!



On 29-Sep-99 Guido van Rossum wrote:
>>     >>> d = {'key1' : 'val1', 'key2' : 'val2', 'key3' : 'val3'}
> Of course, if you have NumPy, there's always
> 
>       keys, values = transpose(d.items())
> 
> which probably beats everything else.
> 
> --Guido van Rossum (home page: http://www.python.org/~guido/)


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From M.Oti at warande.uu.nl  Sat Sep  4 15:57:12 1999
From: M.Oti at warande.uu.nl (Martin Oti)
Date: Sat, 04 Sep 1999 21:57:12 +0200
Subject: Pirate -> RAD
References: <7qno38$6uk$1@news.is.co.za>
Message-ID: <37D17998.5CF4@warande.uu.nl>

Jacques Oosthuizen wrote:
> 
> Hi All
> 
> I am busy working on a tool I named Pirate ("Python RAD") it has a component
> library similar to Delphi and shall in the future work with win32,gtk,qt. It
> will have a IDE.
> 
> Before I do to much and work to hard on it, will anyone be interested in
> using a
> tool like this.

Absolutely! I'm new to Python and the more I find out about it, the more
I like it. :) One major aspect which I miss is a good RAD GUI-building
tool. 

I use Visual Basic at work, mainly for within-company scripting and
automation tasks, and I often need to whip stuff up pretty fast. I've
already started to use Python as much as possible and with a RAD
GUI-building tool I will be able to ditch VB for good. :) There are some
lovely Python GUI toolkits out there, but I'm no expert programmer and I
need to whip GUIs up pretty fast.

Considering the masses of Visual basic programmers out there, I think a
visual RAD GUI-builder/IDE for Python will have a large audience. It
would be particularly handy for those (like me ;) who would also like
VB-like functionality on Linux. Indeed, considering that KOffice will
use Python as it's extension language, Python could then become the
scripting-language-for-the-masses on Linux the way VB is on MS Windows.


Martin

PS: A zillion thanks to Mark Hammond for the PyWin IDE, Robin Dunn for
wxPython (which I only just discovered yesterday), and the entire Python
community for a great programming/scripting language/tool. :)

PS2: I don't know what this would involve, but it would be handy if the
GUI-Builder could also interface with (soon to be) existing Python GUI
toolkits (like TkInter, wxPython and Parrot). I have no clue how
difficult it would be to separate the front-end from the back-end.

PS3: Pirate and Parrot. How ironic. ;) Is this a coincidence? 

------------
Trying is the first step towards failure.
       -- Homer Simpson



From skip at mojam.com  Sun Sep 26 14:44:53 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 26 Sep 1999 13:44:53 -0500 (CDT)
Subject: re vs. sgmllib (was Re: Moving from Perl to Python)
In-Reply-To: <7skbhk$b06@dfw-ixnews16.ix.netcom.com>
References: <000601bf07cb$5e1ede20$602d153f@tim>
	<7skbhk$b06@dfw-ixnews16.ix.netcom.com>
Message-ID: <14318.27045.968552.30026@dolphin.mojam.com>

    Aahz> Tim Peters  wrote:
    >> Regexps are not Python's forte, and all of speed, clarity and
    >> maintainability will increase in proportion to your zeal in purging
    >> them.

    Aahz> That sounds like a Call To Action for rewriting sgmllib to use
    Aahz> something other than re.  Has anyone started any work in this
    Aahz> area?

I think a more profitable direction is to look at the mad scientist's sgmlop
extension module:

    http://www.pythonware.com/madscientist/

It comes with suitably modified versions of the xmllib and sgmllib modules. 

I'd like to see them incorporated into the Python distribution if possible,
depending upon what the XML sig is doing.  (They have their own
distribution.  I'm not sure what's included in it.)

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098   | Python: Programming the way Guido indented...



From fleck at informatik.uni-bonn.de  Thu Sep 16 17:38:00 1999
From: fleck at informatik.uni-bonn.de (Markus Fleck)
Date: 16 Sep 1999 21:38:00 GMT
Subject: Splitting c.l.py
References:  <7rr088$74b@dfw-ixnews8.ix.netcom.com>
Message-ID: <7rrnvo$p26@news.rhrz.uni-bonn.de>

Aahz Maruch  wrote:
> c.l.py.announce could be better used than it currently is (for example,
> the Weekly Python URL doesn't go to c.l.py.announce).

I'm currently starting to process a backlog of about 50-60 postings for
c.l.py.a; this includes three issues of "Python-URL". So the current
lack of "Python-URL" there appears to be entirely my fault...

splitting-comp.lang.python.announce-ly,
Markus.

-- 
////////////////////////////////////////////////////////////////////////////
   Markus B Fleck - University of Bonn - CS Department IV - WHOIS MF5079
          UNIX Administrator - comp.lang.python.announce Moderator
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\



From gherman at darwin.in-berlin.de  Thu Sep 30 11:41:56 1999
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Thu, 30 Sep 1999 17:41:56 +0200
Subject: lzw.py
Message-ID: <37F384C4.9486B017@darwin.in-berlin.de>

Hi,

I had some fun writing Python functions to encode and decode
data using a variant of the Lempel-Ziv coding algorithm family,
namely LZW, Lempel-Ziv-Welch. 

This is something like a "taboo algorithm" if I understand the 
weird logic of companies and their mind-boggling interpretation
of intellectual property rights. To me this sounds like saying
some of Newton's algorithms were illegal to use without paying
for it.

In any case I don't want to mess with Unisys and let them im-
pose their view of the world on me, but I also don't want to
end, somewhat early, in "what-happens-to-be-just-a-regular" 
car accident...

So, is there a way to make my rather limited brain understand,
what I and you can do with such a code without going to all 
patent offices and legal departments of companies? Am I, my-
self, allowed to use it? Is there a difference between using 
the coding and decoding part? Can I pass anything of this to 
somebody else? Is it illegal to send it, receive it, read it, 
download it, use it? -- All of it? 

Of course, there is zlib, but who says there won't be another 
deep-pocket company popping up tomorrow just to say they have 
invented this "before" already? 

Regards,

Dinu

-- 
Dinu C. Gherman

................................................................
"An average of more than 15 % of adults in 12 industrialized 
countries are functionally illiterate; in Ireland, the United 
Kingdom and the United States, the rates are over 20 %."

  (The State of the World's Children 1999,
   UNICEF, http://www.unicef.org/sowc99)



From boud at rempt.xs4all.nl  Thu Sep 30 09:25:14 1999
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 30 Sep 1999 13:25:14 GMT
Subject: DP-Api 2 and autoincrement
Message-ID: <7svobq$ef$1@essle.valdyas.org>

I'm trying to build a sort of object framework for a database application,
using the db-api II specification.  It's a bit like:

                          db-api
                            |
               thin connection object (dbObj)
                            |
                SQL abstraction layer (dbSQL)
                            |
                  record object (dbRecord)
                            |
                    table object (dbTable)

I also have a set of objects that define the tables, records and
relations, like a repository, so in theory the sql layer could generate
all necessary sql for inserting, updating, selecting, deleting and
working with child tables and so on. 

Now, most databases have some sort of sequence generator or autoincrement
datatypes, especially for surrogate keys. When my SQL layer inserts a
record, one field is left blank, and the database fills the number in.
I'd like to have the SQL layer return that number, so it can be entered
in the dbRecord object. Is there a way to get that number back from the
database using db-api II? I'd really hate to build a sequence generator
myself, or to try and find the number using selects before or after
inserts.

-- 

Boudewijn Rempt  | http://ww



From mal at lemburg.com  Fri Sep  3 07:56:40 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 03 Sep 1999 13:56:40 +0200
Subject: How to pass options to an extension
References: <37CE7CB0.979ADF87@onera.fr>
Message-ID: <37CFB778.4EC31926@lemburg.com>

Marc POINOT wrote:
> 
> Our extension module has a getopt()-like class (C++).
> It is used to drive some of the toolkit options. How can
> I pass my option through the python interpreter without
> making it stop on my -- ?

Could you be a bit more verbose about what exactly you want ?
Perhaps you are looking for PySys_SetArgv (int argc, char **argv) ?

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   123 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From fredrik at pythonware.com  Mon Sep  6 10:40:00 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 6 Sep 1999 14:40:00 GMT
Subject: Python speed and `pcre'
References: <000301bef83b$9d7fa620$da2d153f@tim> <7r0ivu$95j@dfw-ixnews21.ix.netcom.com>
Message-ID: <002301bef875$b3b18410$f29b12c2@secret.pythonware.com>

Aahz Maruch  wrote:
> >The design issues are more complicated than that, alas.  Here's one you
> >weren't thinking of :  for easy chatting with C, under the covers a
> >Python string is always terminated with a null byte.  You can't see that
> >from Python code (it's not counted in the string length, and you can't index
> >to it), but your C code can rely on it.
> 
> Hmmm....  From what I've seen, Python strings work perfectly well with
> null bytes in the middle.  Are there any circumstances under which that
> does not hold true?

from Include/stringobject.h:

    Type PyStringObject represents a character string.  An extra zero
    byte is reserved at the end to ensure it is zero-terminated, but a
    size is present so strings with null bytes in them can be represented.

in other words, any Python string can be passed directly to any
C function that expects a null terminated string, without having
to copy it to a separate buffer.






From damien_morton at my-deja.com  Wed Sep 15 19:20:28 1999
From: damien_morton at my-deja.com (damien_morton at my-deja.com)
Date: Wed, 15 Sep 1999 23:20:28 GMT
Subject: mxODBC and NT services
Message-ID: <7rp9jg$7sh$1@nnrp1.deja.com>

hi people,

   Im about to embark on something on an adventure in Python
programming - the creation of an SMTP server that uses a database to
store incoming and outgoing emails. All of the SMTP severs I have
examined store incoming and outgoing emails as files in spool
directories. For extremely large numbers of emails (50k+), this will
bring an OS to its knees. A database, on the other hand, will cope just
fine (at least, thats my theory).

    Before I embarked on this project, I thought Id post a note to see
if anyone had any comments, suggestions, or warnings about the use of
the mxODBC drivers, and about my project in general.

    I intend using the mxODBC drivers in a fairly heavily multithreaded
environment under Windows NT - i.e. in an NT service. Most of the
outgoing emails will be generated on demand as lists are processed, so,
in truth, there wont be massive volumes of data being transferred over
ODBC. This thing will have to be pretty damned reliable though.

    I have read that ODBC has certain limitations when running as an NT
service of with threads. Do anyone think this will be a problem at all?
Does anyone have any experince with this that they might share?


Ive listed the relevant MS KB artiles below. Ive read them and think
that the problems is probably much more minor than I imagine.

Limitations of DAO, DAO SDK in NT Service or with Threads
http://support.microsoft.com/support/kb/articles/Q156/1/38.asp

Running ODBC Applications as Windows NT Services
http://support.microsoft.com/support/kb/articles/Q136/4/81.ASP



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bill at hk.rfa.org  Thu Sep 30 07:08:39 1999
From: bill at hk.rfa.org (Bill Eldridge)
Date: Thu, 30 Sep 1999 19:08:39 +0800
Subject: Help - Timezone conversion problem
Message-ID: <37F344B7.3663458F@hk.rfa.org>


I had a Python script running on a Washington
machine to convert Eastern time to Hong Kong
time.  No problem - convert local Eastern time
to UTC and add 8 hours (no Daylight Savings Time
in Asia).

Now I'm trying to do something similar in Hong Kong
to convert Eastern time to Hong Kong, but I can't
seem to find an easy way to do this, since the
machines have TZ=CST (Hong Kong time) and there's
no obvious (to me) way to temporarily make the
script pretend it's in Eastern time to figure
out if it's EDT or EST.  (even if I set time.altzone
time.timezone and time.tz, the time module looks
to use the system values for these - i.e. the
variables reflect the system but aren't used for
calculations).

So, basically, I have a timestamp 1999-10-01-1030
Eastern time and need to figure out if it's daylight (+1)
or standard  (+0) and then add the offset of +13.

I could hardcode this to work for a few years, but
that's not very elegant.

Ideas?

Thanks,
Bill
--
Bill Eldridge
Radio Free Asia
bill at hk.rfa.org



From aaron_watters at my-deja.com  Mon Sep 13 09:25:51 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Mon, 13 Sep 1999 13:25:51 GMT
Subject: Announcing: A Python Curses-Based Telnet Client
References:  <37DCDC05.8E25CCBE@sdm.de>
Message-ID: <7riu0h$jdf$1@nnrp1.deja.com>

In article <37DCDC05.8E25CCBE at sdm.de>,
  Peter Stoehr  wrote:
> Hi Moshe,
>
> it would be great if you can send me a copy of your "Python Cursed-
Based
> Telnet Client"....


Yes, sounds interesting.  Also, fwiw, there are many free
sites where you can put files, I think.  I think geocities is
one of them, also yahoo, lycos, egroups...
   --Aaron Watters

===
Call the Dr.
I think I'm gonna crash
The doctor says he's coming
but you gotta pay him cash
Life in the fast lane
Everything.
All the time.
Life in the fast lane
Uh huh


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From emile at fenx.com  Wed Sep  8 10:21:39 1999
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 8 Sep 1999 07:21:39 -0700
Subject: Python interface to Groupwise mailoxes?
References:  <37D5B211.E9C7C9F3@nightwares.com>
Message-ID: <37d64753@news-server.direcpc.com>

When I checked into delivering content through to a GroupWise mailbox, I
found out the easiest way was to bounce it off the net through the
user at company.com internet mail address.

--

Emile van Sebille
emile at fenx.com
-------------------


Warren 'The Howdy Man' Ockrassa  wrote in message
news:37D5B211.E9C7C9F3 at nightwares.com...
> "S. Blakey" wrote:
>
> > Even some good documentation of Novell's proprietary crap would
> > appreciated.
>
> That's the problem. Novell's way of doing things is, well, Novell's.
> I've tried to hack into my GroupWise box at work using non-Novell
tools,
> in non-MS IDEs, and had nothing but grief.
>
> However note that Novell *does* market a POP server which *in theory*
> makes their G'Wise mailboxen available to any POP3 client. It's spendy
> -- about US$1300 for a 50-seat license as I recall -- but it opens the
> gateway, so to speak, and in large companies or companies with many
> users on many OSen or many users travelling amortizes well.
>
> I am still pushing hard at work for that implementation, as the
current
> scheme is intolerable (G'Wise client hoses NT4's networking,
surprise.)
>
> Don't know if they have a gateway for SMTP for sending stuff. Sigh.
> Their site used to have lots of info like that available, but recently
> they seem to have buried the relevant docs under a lot of noise, so
it's
> darn near impossible to find even the POP server, which in fact they
> released under the aegis of being something like "Netscape mail server
> for Novell", when you search their site. I found the POP server before
> AOL sucked up Mozilla though, and that might be a factor too.
>
> Short answer: If anyone has figured out Novell's server scheme for
mail
> I'd sure love to know about it, but I fear it has not been done yet.
(Or
> did I see a Red Hat install go by a while back which included a G'Wise
> client?)
>
> --
>     warren ockrassa | nightwares | mailto:warren at nightwares.com
>                     http://www.nightwares.com/





From wtanksle at dolphin.openprojects.net  Thu Sep  2 22:38:49 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 03 Sep 1999 02:38:49 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <7qi3q3$n1d$1@nnrp1.deja.com>
Message-ID: 

On Wed, 01 Sep 1999 02:42:13 GMT, Mr Kelvin Lee wrote:
>Hi,

>The main weakness I hear about language interpreters/ interpreted
>languages (btw.. is that the same thing?) is that they are slow(?)..

You'll need to define your criterion for 'slow'.  That sounds Clintonian,
but it's true -- different projects have different needs.  For heavy
numeric work, Python might wind up being the fastest language available to
you because of NumPy (although I can't imagine the circumstances).

The usual meaning of "slow" is that they do common tasks, such as function
calls, much more slowly than the machine could do them.  Python certainly
has that particular weakness.

For the rest of your post, I'm going to assume that when you say
"interpreted" you mean "dynamic".

>is this because of the way they handle (data) variables?

No, there are many reasons -- the main one is that they have so much more
to do at runtime.  This makes things you do naturally in a static language
slow -- but it makes many more things _possible_, where before they would
have taken a LONG time to develop and be very buggy.

>Are there any other major weaknesses of interpreted languages over
>conventional languages such as C, C++, Ada etc.?

Yes, and vice versa.

Dynamic languages are capable of much less compiler optimization, so
they're slower.  They're also capable of doing much more useful work per
word of code, so they wind up doing the job faster.

>What are the most popular interpreted languages around?

Perl, Python, Common Lisp, CAML, Forth...

>Thanks,
>
>--
>Regards,
>
>KL.
>
>
>Sent via Deja.com http://www.deja.com/
>Share what you know. Learn what you don't.


-- 
-William "Billy" Tanksley



From 101.139915 at germanynet.de  Tue Sep  7 11:43:16 1999
From: 101.139915 at germanynet.de (Stefan Knappmann)
Date: 7 Sep 1999 15:43:16 GMT
Subject: How to get the python sourcecode line, when debuging the Python C sources?
Message-ID: <7r3bqk$hbv$1@newshost.germany.net>

Hi,

I extended Python with a self written C Module.
Unluckily my program crashes with a segmentation fault. 
I recompiled Python with debugger information and reprodced the
error. 
In the C debugger I can see that the program is in an initialisation
function of a datastructure in my C module.
So far so good, but this initialisation function is only
called at the start of the python script and the crash commes 
aproximately after two hours of running.

I would like to know in which line of python code is the program,
when it calls my function.

Does anybody know in which C variable I can find the Python Stack?
Or how to get the activ Python source code line from within the C
debugger?


Thanks in advance

Stefan






From I.Clarke at strs.co.uk  Wed Sep 22 10:44:31 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Wed, 22 Sep 1999 15:44:31 +0100
Subject: Matching templates against a tree - any ideas?
References: <37E7A272.51E5C4E1@strs.co.uk> <37E7F8C0.AC421C83@maxtal.com.au>
Message-ID: <37E8EB4F.BC42825E@strs.co.uk>

> you have a template and you want to match them against EVERY subtree 
> of a given tree, efficiently. 

Correct, but there will be multiple trees and multiple templates (I
mention this just in case some preprocessing or building of lookup
tables might be useful).

> Question: what happens if there is more than one overlapping match?

One of the possibilities is selected and returned as a match - it
doesn't matter [at this stage] which one [although that may change in
the future].
 
[...snip interesting parser idea...]

I would rather implement a solution from scratch instead of attempting
to reformulate the problem into that of parsing something.  Do you think
that you could provide an explanation of what one of the parsers you
describe would do - but in terms of the problem I described, rather than
in terms of parsing.

Cheers,

Ian.

-- 
"Don't throw your PC out the window, throw Windows out of
 your PC, and install Linux"
Ian Clarke                    http://www.gnu.demon.co.uk/



From hnowak at cuci.nl  Sat Sep  4 04:09:10 1999
From: hnowak at cuci.nl (Hans Nowak)
Date: Sat, 4 Sep 1999 08:09:10 GMT
Subject: termios disappears during object deletion
In-Reply-To: 
References: 
Message-ID: <199909040810.KAA18642@dionysus.fw.cuci.nl>


On 3 Sep 99, at 22:58, Brien Barton wrote:

> Can anyone explain this behavior??  For some reason, the termios and
> TERMIOS modules get wiped out when python is running the __del__ method
> during program termination.  I've tried this with Python 1.5.1 on Linux
> and with Python 1.5.2 on SGI IRIX.  If you explicitly delete the object,
> you don't have this problem.
> 
> ~> cat test.py
> import math, termios, TERMIOS
> 
> print 'math=', math
> print 'termios=', termios
> print 'TERMIOS=', TERMIOS
> 
> class Test:
>     def __del__(self):
>         print 'deleting Test object ...'
>         print 'math=', math
>         print 'termios=', termios
>         print 'TERMIOS=', TERMIOS
> 
> t = Test()

I'm not sure, but I have a vague idea why this happens... an instance 
of Test is created, then the program is finished. Python starts 
cleaning up some stuff, freeing memory, etc. *Maybe* (as said, I 
don't know this fer shure) it first gets rid of termios and TERMIOS, 
then it takes on the Test instance t, calling its __del__ method. 
This method then correctly sees that termios and TERMIOS are already 
deleted.

Just an idea... I don't know much about Python's internals, but it 
would make sense to me...

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/



From nathan at islanddata.com  Wed Sep  1 16:23:50 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Wed, 01 Sep 1999 13:23:50 -0700 (PDT)
Subject: Queue class deadlocking application
In-Reply-To: <1275908198-63413490@hypernet.com>
Message-ID: 

On 01-Sep-99 Gordon McMillan wrote:
> How do you know the queue size if you only ever use put_nowait and 
> get_nowait ?

Okay, okay...I do a qsize() now and then, but the code is not dependent on
it, nor do I ever use a blocking get or put.  I've also found that the
problem is totally non-existent if I remove the consumer or decrease the
number of producers to one.  I should try two to see if that's safe or
not...

Thanks for putting some time into it, though.  It's quite important that
this application run robustly enough to be left unmonitored.  The worst
part of this is that I can't even check for it, as it seems to freeze up
all the threads...I'm going to also enter a new thread that does nothing
with the queue but check its size and see if it survives.


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From nathan at islanddata.com  Sat Sep 11 23:26:23 1999
From: nathan at islanddata.com (nathan at islanddata.com)
Date: Sat, 11 Sep 1999 20:26:23 -0700 (PDT)
Subject: Queue not releasing memory
Message-ID: <19990912032623.13533.qmail@islanddata.com>

After filling a Queue class instance with large amounts of data, the
memory footprint of python grows dramatically.  I would expect, then,
that this footprint would decrease in size when the queue is emptied,
but it does not.  The memory use drops instantly when the empty queue
is delete, however.  I've examined the Queue.py code and cannot see
where this data may be cached, etc.  Executing:

sys.getrefcount(q.queue[0])

for example, yields 2: the temporary reference and a single one for
the queue entry.  It stands to reason that after get'ing an item from
the queue that it should be lost forever.  Test program:

q = Queue.Queue(20000)
for x in string.uppercase:
 for y in string.uppercase:
  for z in string.uppercase:
   m = (x + y + z) * 1024
   q.put(m)

for x in string.uppercase:
 for y in string.uppercase:
  for z in string.uppercase:
   q.get()

The memory taken up by this program does not decrease until q is
deleted.  Can anyone tell me why this happens and how I can fix it
*without* deleting the queue and creating a new one?  Thanks...



From jussij at zeusedit.com  Tue Sep  7 18:27:55 1999
From: jussij at zeusedit.com (Jussi Jumppanen)
Date: Wed, 08 Sep 1999 08:27:55 +1000
Subject: Programmers Editor
Message-ID: <37D5916B.606E@zeusedit.com>

Zeus for Windows Version 3.00 Now Released
------------------------------------------

The Zeus for Windows programmer's editor has been totally 
re-developed and the results of this full re-write are now
ready for use. The  Programmer's is designed to run on the
Windows 95, Windows 98 or NT platforms.

So what is new:
    
    1) New syntax highlighting engine should now support almost
          any language.

    2) Comes with pre-configured syntax highlighting for:

        C/C++, Java, Perl, HTML, Python, Pascal

       and is easily to almost any other language.

    3) New plugable scripting modules allows you to write Zeus
          macro scripts using the Python, LUA or Small C macro
          languages.

    4) New configuration model now makes it even easier for you
          to configure Zeus to do just what you want.

    5) Enhanced compiler, project and tool support.

    6) Improved keyboard mapping.

You can download the shareware version from:

  http://www.zeusedit.com/z300/download.htm

For some screen shots of Zeus in action goto:

  http://www.zeusedit.com/z300/look.htm

For more details visit the Zeus web page:

  http://www.zeusedit.com

Jussi Jumppanen
Author of: Zeus for Windows, Win32 (Brief, WordStar, Emacs) Text Editor
"The C/C++, Java, HTML, Pascal, Cobol, Fortran programmers text editor" 
http://www.zeusedit.com



From valb at vtek.com  Thu Sep 23 20:04:56 1999
From: valb at vtek.com (Val Bykoski)
Date: Thu, 23 Sep 1999 20:04:56 -0400
Subject: PST - Python Sound Tool
Message-ID: <7sefdn$8au$1@winter.news.rcn.net>

Hello,dear Pythoneers:

    I was playing with PST-0.1, trying to play
any wav file.  No luck.  And for now there is
no sample,or usage doc.  Any help would be great.
Thanx.  val

Val Bykoski
valb at vtek.com





From mlh at idt.ntnu.no  Sun Sep  5 13:15:45 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 05 Sep 1999 19:15:45 +0200
Subject: cgi post
References: <37D179CC.2DDFD7A7@netscape.com>
Message-ID: 

vijay Mallela  writes:

> how do i do something like this in a python cgi script:
> 
> form = cgi.fieldstorage()
> if form['foo'] == 'bar':
>     loadhttp('spam.com')
> please let me know if there is a way to do this..
> 
> Right now, I am opening the file and spam.com and then printing it..
> what if the file is some where else and I would like to direct the user
> to that site..

Just print a header like this:

Content-type: text/html
Location: spam.com

The user will then be redirected to spam.com without any problems.
(Remember the empty line after spam.com, though... Even if you aren't
printing anything else)

> 
> Thanks
> v
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From gerrit.holl at pobox.com  Fri Sep 10 16:59:17 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 10 Sep 1999 22:59:17 +0200
Subject: [Python-checkins] CVS: python/dist/src/Lib pdb.py
In-Reply-To: <199909091632.MAA24868@anthem.cnri.reston.va.us>; from Barry A. Warsaw on Thu, Sep 09, 1999 at 12:32:42PM -0400
References: <199909091632.MAA24868@anthem.cnri.reston.va.us>
Message-ID: <19990910225917.A3356@optiplex.palga.uucp>

Hello,

I have a question because of this email on the python-checkins mailinglist.

On Thu, Sep 09, 1999 at 12:32:42PM -0400, Barry A. Warsaw wrote:
> Update of /projects/cvsroot/python/dist/src/Lib
> In directory anthem:/projects/python/develop/bwarsaw/python/dist/src/Lib
> 
> Modified Files:
> 	pdb.py 
> Log Message:
> Pdb.lineinfo(): Don't use os.popen('egrep ...') to find the line in
> the file that a function is defined on.  Non-portable to Windows and
> JPython.

I understand that it isn't portable to Windows, but why isn't it portable to
JPython? Doesn't the os.popen function exist in JPython? If so, why not?

[cut]

regards,
Gerrit.
-- 
Unix is easy: type 'man man' and the rest comes of itself.



From Elie.Naulleau at westgroup.com  Mon Sep 13 16:16:11 1999
From: Elie.Naulleau at westgroup.com (Elie Naulleau)
Date: Mon, 13 Sep 1999 15:16:11 -0500
Subject: Perf. Python/Tcl comparison ?
Message-ID: <37DD5B8B.9402908F@westgroup.com>

I wondering if some specs have been done about comparing
Tcl and Python performances, on strings integers and floats for
instance.
Thks,

Elie.Naulleau at westgroup.com




From bthomas at trey-industries.com  Tue Sep 14 11:48:20 1999
From: bthomas at trey-industries.com (Ben Thomas)
Date: Tue, 14 Sep 1999 15:48:20 GMT
Subject: python call external program (WHAT THE?!)
References: <37DE6156.FD4384C6@trey-industries.com> <37DE68B7.49D0C5A5@mach.uni-karlsruhe.de> <37DE6980.2FC85A6F@trey-industries.com>
Message-ID: <37DE6D8B.FD6AC6E0@trey-industries.com>

Hey guys the below is true with the addition of
system('d:\\data\\vacman\\mybat.bat') = ok

Have I found a bug? Do I get a t-shirt? :) Could a py-xpert explain to me my
results?

Ben



Ben Thomas wrote:

> I just tried but putting the batch file in three places on a drive. The I tried
> them all.
>
> system('d:\mybat.bat') = ok
> system('d:\data\mybat.bat') = ok
> system('d:\data\vacman\mybat.bat') = fails, it works fine from DOS prompt!
>
> what gives?
> Ben
>
> Andreas Cardeneo wrote:
>
> > Ben Thomas wrote:
> > >
> > > very simple question:
> > >
> > > What would I do to have Python execute something from the DOS command
> > > line. I am very new to python but I bet it is in a moudule but don't
> > > know the command or the modules to look in. I want to
> > >
> > > dir >%myvar.txt
> > >
> > > #this code would run dir from the command line and pass the var "myvar"
> > > , dir would dump a listing of the files into a text file called xxxx.txt
> > >
> > > pointers?
> > >
> > > Ben
> >
> > Hi Ben,
> >
> > you should use the system() function from the "os" module.
> >
> > os.system("dir *.txt") shows all files in the current directory matchin
> > *.txt.
> > The function returns the exit code.
> >
> > Bye,
> > Andreas
> >
> > --
> > Andreas Cardeneo                Universit?t Karlsruhe (TH)
> > Tel: +49 721 608-4576           Institut f?r F?rdertechnik und Logistiksysteme
> > Fax: +49 721 75 83 78           Hertzstra?e 16 - 76187 Karlsruhe
> > eMail: Andreas.Cardeneo at mach.uni-karlsruhe.de




From news at dorb.com  Thu Sep 30 22:31:02 1999
From: news at dorb.com (Darrell)
Date: Thu, 30 Sep 1999 22:31:02 -0400
Subject: Last question of the day (This one's easy!)
References: <877ll88031.fsf@ethereal.dhis.org>
Message-ID: 

Tried your example in C, Java and Python. "C" ~2sec, Java was 1.5min and
Python was 4.5 min.

This isn't as significant as it sounds though. The algorithm is often the
largest impact on performance. Making significant algorithm changes to gain
performance is vastly easier in Python.

Drop into Java or "C" for that inner loop. Builtins like string.split and
string.join are examples of some stuff better done in "C".

Notice how much faster "C" is than Java.


--
--Darrell





From fredrik at pythonware.com  Thu Sep 30 07:43:31 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 30 Sep 1999 11:43:31 GMT
Subject: !! Non blocking IO !!
References: <7st1ra$6vb$1@oceanite.cybercable.fr>  <7svfn1$ce2$1@oceanite.cybercable.fr>
Message-ID: <001e01bf0b39$06733c40$f29b12c2@secret.pythonware.com>

Pierre Saloni wrote:
> In fact what I want is an autoflush when I print, i do not want print to
> bufferize at all.

the easiest way to do this is to use the -u option
to the interpreter.  this also enables binary stdio
where necessary:

$ python -?
Unknown option: -?
usage: python [option] ... [-c cmd | file | -] [arg] ...
Options and arguments (and corresponding environment variables):
/.../
-u     : unbuffered binary stdout and stderr (also PYTHONUNBUFFERED=x)
/.../

...

an alternative solution is to use an autoflushing
wrapper:

class AutoFlush:
    def __init__(self, stream):
        self.stream = stream
    def write(self, text):
        self.stream.write(text)
        self.stream.flush()

import sys

sys.stdout = AutoFlush(sys.stdout)
sys.stderr = AutoFlush(sys.stderr)

print "hello"

...

(maybe the CGI module should contain a
helper function that does exactly this?)






From amitp at Xenon.Stanford.EDU  Tue Sep 28 16:32:29 1999
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 28 Sep 1999 20:32:29 GMT
Subject: Matching templates against a tree - any idea?
References:  <000901bf089c$911731c0$032292cb@akha>
Message-ID: <7sr8kt$3c8$1@nntp.Stanford.EDU>

 Jon Fernquest  wrote:
| Fernando Pereira wrote:
| 
| Thanks for making the posting on term rewriting systems.
| 
| 1. Do you think it's possible to make a rewriting system extension 
| to an existing language like Python or Perl?
| 
| Everything seems to be in ML and ML extensions such as
| Isabelle right now. It's so easy to cook up (or integrate 
| components into) whole systems in Python 
| (or Perl) and since so many people use them in their
| day to day work modules have a greater chance of being used.

ML already has some pattern-matching constructs that can be used to
match subtrees.  For example, this tree:

         Mul
         / \
       Add  B
       / \
      A   0

can be matched like this:

  case(tree) of
     Mul(Add(A,Num(0)), B) =>   ... do something with A and B ...

My guess is that complicated tree matching becomes easier when you
have these simpler tree matching constructs built into the language.

ObPython:  Pattern matching is the one ML thing I miss when I'm in Python..



	   - Amit



From skip at mojam.com  Thu Sep 30 10:42:03 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 30 Sep 1999 09:42:03 -0500 (CDT)
Subject: When Good Regular Expressions Go Bad
In-Reply-To: <000101bf0a47$98084da0$ce2d153f@tim>
References: 
	<000101bf0a47$98084da0$ce2d153f@tim>
Message-ID: <14323.30395.394477.753318@dolphin.mojam.com>

    Tim> [Douglas Alan]
    >> It seems to me that even when a regular expression fails to match a
    >> string, you might want to know just how far it was able to get before
    >> getting stuck. 

    Tim> I believe this would be easy to add to any regexp engine I've ever
    Tim> seen, and also believe I've never seen one that keeps track of it.
    Tim> I confess I'm at a loss to think of a compelling use for it,
    Tim> though.

Take a look at

    http://www.musi-cal.com/fast-itineraries.shtml

I want to let people set up schedule submissions and give them some help
when they format their schedules incorrectly.  Currently I use three regular 
rexpressions per selected format, the one that will match a correctly
formatted line and two shorter ones.  When the real thing fails I use the
shorter ones to try and give the user some idea of where they might have
gone astray.  Of course, users aren't told that regular expressions underly
the pattern matcher.

compellingness-is-in-the-eyes-of-the-programmer-ly y'rs,

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098   | Python: Programming the way Guido indented...



From paul at prescod.net  Thu Sep  2 14:40:41 1999
From: paul at prescod.net (Paul Prescod)
Date: Thu, 2 Sep 1999 18:40:41 GMT
Subject: language interpreters/ interpreted languages weaknesses?
References: <1275835455-67788915@hypernet.com> <37CE9F55.88DFC9FC@megsinet.net>
Message-ID: <37CEC4A9.C2CE2A91@prescod.net>

David Oppenheimer wrote:
> 
> I am sure that many of you can
> relate to having run into a page where a Java applet is loading and been quite
> irritated that while this Java code is loading (being interpreted or the JVM
> is firing up or whatever) that your browser (and your cursor too) will lock-up
> until the applet is ready to run.  

There are by now dozens of impementations of Java and your complaint is
with a single one (or maybe two, if you have had the same problem with
Internet Explorer and Netscape). Seriously this has nothing to do with
Java *the language*. A bad implementation of Python could do the same
thing.

 Paul Prescod



From brianr at dnr.state.ak.us  Thu Sep  9 17:16:41 1999
From: brianr at dnr.state.ak.us (Brian Raynes)
Date: Thu, 09 Sep 1999 13:16:41 -0800
Subject: question about the safety of os.popen()
Message-ID: <37D823B9.9D4337EE@dnr.state.ak.us>

I'm writing a simple module that, in part, sends an email.
>From various examples, I use the following line to start
sendmail, then I use the write method of the resulting mail
object to send the header and body text, then mail.close().

mail = os.popen("/usr/sbin/sendmail -t", "w")

I believe that this method of using os.popen() is relatively
safe since it does not pass any user input directly to the
shell.  All the examples I can find concerning security are
written in perl, but this statement seems to differ from the
dangerous examples because it does not include any user
input on the command line.  If I understand what os.popen()
is doing, this is opening a pipe directly to sendmail, so
that stuff sent via the .write() method does not go to the
shell, but to sendmail.  It is my understanding that this is
safer - most warnings about CGI security warn against
allowing user input to be sent to the shell.  

Am I doing the safe thing in regards to not sending input to
the shell?
Is there a better way to accomplish this?

I realize there are also security concerns with stuff sent
to sendmail, but the information on what to look for there
is more available and more appropriate in a sendmail
discussion than a Python one.

I'm still pretty new, but I've gone through the vast
majority of "Learning Python" and have read about popen() in
the Library Reference.  I don't know C, or else I would look
and understand exactly what's happening from the source
code.


Brian Raynes



From watterss at pilot.msu.edu  Wed Sep  8 22:33:27 1999
From: watterss at pilot.msu.edu (Shannon Watters)
Date: Wed, 8 Sep 1999 22:33:27 -0400
Subject: NEWBIE:  M. Lutz's Programming Python
Message-ID: <7r76hh$cqv$1@msunews.cl.msu.edu>

I'm reading Programming Python and I've come across a problem with running
one of the examples.  In Chapter 9 there is a module to generalize list and
dictionary menus using functions.  When I test the functions I get an
unexpected result (my module looks just like Mr. Lutz's, or I'm just bleary
eyed).

Here is the module (menu0.py):

from string import upper, lower

def interactDict(menu):
     while 1:
          for name in menu.keys():
               print '\t' + name
          tool = raw_input('?')
          try:
               menu[tool]()
          except KeyError:
               print 'what? - try again'

def interactList(menu):
     while 1:
          for name, func in menu:
               print '\t' + upper(name[0]) + ')' + name[1:]
          tool = lower(raw_input('?'))
          for name, func in menu:
               if tool == name[0] or tool == name:
                    exitflag = func()
                    break
               else:
                    print 'what? - try again'
                    continue
          if exitflag: break

def interact(menu):
     try:
          if type(menu) == type([]):
               interactList(menu)
          elif type(menu) == type({}):
               interactDict(menu)
          else:
               print "bad menu: must be a list or dictionary"
     except EOFError: pass

-------------------------------------------------------------

We (me and Mr. Lutz :-) use another module of simple menus to test the above
module(testmenu.py):

from sys import stdout, exit

dmenu = {'spam' : lambda:stdout.write('SPAM\n'), 'stop' : exit }

lmenu = [('eggs', lambda:stdout.write('EGGS\n')), ('stop', lambda:1)]

-----------------------------------------------------------------------

When I go to test menu0 in the interaction window I get this:

>>> from testmenu import *
>>> from menu0 import interact
>>> interact(lmenu)
         E)ggs
         S)top
?E
EGGS
         E)ggs
         S)top
?S
what? - try again
>>> interact(dmenu)
         spam
         stop
?spam
SPAM
         spam
         stop
?stop
Traceback (innermost last):
  File "", line 1, in ?
    interact(dmenu)
  File "menu0.py", line 34, in interact
    interactDict(menu)
  File "menu0.py", line 11, in interactDict
    menu[tool]()
SystemExit:
>>>

What I don't understand is why it prints 'what? - try again' since my
raw-input was a valid choice.  I'm not exactly sure why I get the traceback
when I stop using dmenu either.

I would appreciate any reply and I hope that the answer isn't one that makes
me say 'damn, I should have seen that!'.

Thanks,
Shannon













From skip at mojam.com  Thu Sep 16 13:03:18 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 16 Sep 1999 12:03:18 -0500 (CDT)
Subject: Marking translatable strings
In-Reply-To: <14305.3887.816696.344596@dolphin.mojam.com>
References: 
	<14305.3887.816696.344596@dolphin.mojam.com>
Message-ID: <14305.8918.763630.110275@dolphin.mojam.com>

    Skip> 2. Could a loadable module be defined that would put Python into
    Skip>    "translate" mode, thus forcing all raw strings in the code to
    Skip>    be mapped into translatable strings at runtime (whose semantics
    Skip>    would then be definable as appropriate).

Ack!  Should have been "... thus forcing all string literals in the code..."

S



From gmcm at hypernet.com  Sat Sep 11 14:25:24 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 11 Sep 1999 14:25:24 -0400
Subject: yacc/lex for Python
In-Reply-To: <37DA861F.57212BC3@atl.mediaone.net>
Message-ID: <1275048324-10459585@hypernet.com>

Kristopher Johnson asks:

> Are there any lex- or yacc-like tools available for Python?

See:
http://starship.python.net/crew/amk/python/string.html
(towards the bottom of the page) for a (probably out of date) list of 
various parsing packages.

- Gordon



From mal at lemburg.com  Mon Sep 27 17:24:52 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 27 Sep 1999 23:24:52 +0200
Subject: Bug report : fcntlmodule.so (fonction ioctl) python 1.5.2
References: <37EB7EAE.E2D5A7A@lirmm.fr>
Message-ID: <37EFE0A4.639A3218@lemburg.com>

David Berthelot wrote:
> 
> Hi,
> 
> i think, i found a bug (and a fix).
> 
> In the "fcntlmodule.so", the "ioctl" function does not work with
> an optional int parameter but currently it works only with an
> optional string parameter.
> 
> So i think that the line number 76 of fcntlmodule.c :
> 76: ret = ioctl(fd, code, arg);
> 
> should be replaced by:
> 76: ret = ioctl(fd, code, &arg);
> 
> I think it does it, doesn't it ?
> Yes, unlike fcntl, ioctl uses a pointer for its 3rd arg.

Not always... it depends on the fd and code.

Note that Python 1.5.2 (well at least the CVS version)
has code for handling strings *and* integers. The struct
module will help you build pointers to integers if needed.
If you need a read/write memory area, you can use the new
buffer objects for this task.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                    98 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From holger at phoenix-edv.netzservice.de  Thu Sep 16 03:43:19 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Thu, 16 Sep 1999 07:43:19 GMT
Subject: timesync with python?
References: <37D66AF3.6A3E96F7@phoenix-edv.netzservice.de> <002701befabb$1a884880$f29b12c2@secret.pythonware.com> 
Message-ID: <37E09F97.F546A143@phoenix-edv.netzservice.de>

...works great, thanx!!!;-)))

Ciao,
Holger



From tim_one at email.msn.com  Sat Sep 11 16:29:51 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 11 Sep 1999 16:29:51 -0400
Subject: whrandom.randint(): no LONGs?
In-Reply-To: <19990911163530.A11476@optiplex.palga.uucp>
Message-ID: <000601befc94$6692ab00$6aa2143f@tim>

[Gerrit Holl]
> whrandom.randint() always try to convert the given arguments to a
> integer. Why?
> I'm not able to use Longs now. I don't require it directly, but
> it might be a useful thing to know the "why". I can learn from it...

It's a good question!  The "deep" reason is that the method used by whrandom
to create the illusion of randomness has limited precision.  Using this
method to ask for a random choice from among 2**15 things works pretty well
for most purposes; but using it to ask for a random choice from among
2L**150 things would work terribly for many purposes.  Creating the illusion
of random longs requires a different approach altogether, so rather than
give you garbage results Python raises an error.

If you get to the point where you need pseudo-random longs, I posted a
LongRan.py module about a year ago that can do it.

whrandom-knows-its-limitations-so-it's-smarter-than-us-ly y'rs  - tim





From robert.yodlowski at intel.com  Mon Sep 27 16:19:29 1999
From: robert.yodlowski at intel.com (Robert Yodlowski)
Date: Mon, 27 Sep 1999 16:19:29 -0400
Subject: Python interface to Spice or other circuit simulator?
Message-ID: <37EFD151.39411C5C@intel.com>

An HTML attachment was scrubbed...
URL: 

From actgal8 at aol.com  Thu Sep  2 18:24:26 1999
From: actgal8 at aol.com (actgal8 at aol.com)
Date: Thu, 02 Sep 1999 16:24:26 -0600
Subject: 18 AND OVER 65769
Message-ID: <020999162426@aol.com>

free xxx pics:

http://207.240.225.250


:Q$$RawN



From fredrik at pythonware.com  Thu Sep  2 09:54:00 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 2 Sep 1999 13:54:00 GMT
Subject: Python speed and `pcre'
References:    <00af01bef543$0affe8f0$f29b12c2@secret.pythonware.com> 
Message-ID: <002101bef54a$9c7682b0$f29b12c2@secret.pythonware.com>

Fran?ois Pinard  wrote:
> Yes, of course.  Given strings are immutable, an implementation could have:
>
>         ref=str
>         str=str+str1
>
> and have `ref' points to a substring of `str', without copy most of
> the time, when `str1' is small.

what do you do when you have to reallocate str?






From arne at leithe.no  Sun Sep 12 16:50:43 1999
From: arne at leithe.no (Arne Leithe)
Date: Sun, 12 Sep 1999 20:50:43 GMT
Subject: ASP with Python question
References: <37d87ed7@news.interq.or.jp> <7rf5ke$3c1$1@m2.c2.telstra-mm.net.au> <37dbdf82.1605161@news.online.no>
Message-ID: <37dc0fa6.13931092@news.online.no>

On Sun, 12 Sep 1999 17:17:10 GMT, arne at leithe.no (Arne Leithe) wrote:

>It does :-)

However, this syntax doesn't seem to work with COM-objects; for
example:

<%@language = Python%>
<%
fso = Server.CreateObject("Scripting.FileSystemObject")
Session.SetValue('Key', fso)
%>

yields:

Python ActiveX Scripting Engine error '80020009' 

Traceback (innermost last): File "